05-2861. Aviation Data Modernization

AGENCY:

Office of the Secretary, Department of Transportation.

ACTION:

Notice of proposed rulemaking (NPRM).

SUMMARY:

The Department of Transportation (the Department) is proposing to revise the rules governing the nature, scope, source, and means for collecting and processing aviation traffic data. Those reporting requirements are known as the: Origin—Destination Survey of Airline Passenger Traffic (O&D Survey); and Form 41, Schedule T-100—U.S. Air Carrier Traffic and Capacity Data by Nonstop Segment and On-flight Market and Form 41, Schedule T-100(f)—Foreign Air Carrier Traffic Data by Nonstop Segment and On-flight Market (collectively, the T-100/T-100(f)). Current traffic statistics no longer adequately measure the size, scope and strength of the air travel industry. This NPRM proposes to simplify the requirements placed upon Carriers reporting the O&D Survey. The proposed O&D Survey will eliminate the ambiguity in the identification of the Participating Carrier and eliminate the need for manual data collection by designating the Issuing Carrier as the Participating Carrier. It will also increase accuracy by expanding the volume of data to 100 percent of Ticketed Itineraries, and make the data more useful to Department, airport, and industry planners by collecting broader information about the Ticketed Itinerary sale and the scheduled itinerary details. The proposed T-100/T-100(f) will improve the quality of the data by maximizing the congruence of the O&D Survey and the T-100/T-100(f).

DATES:

Comments must be submitted by April 18, 2005.

FOR FURTHER INFORMATION CONTACT:

Richard Pittaway, Office of Aviation Analysis, 400 Seventh St. SW., Room 6401, Washington, DC 20590, [phone].

SUPPLEMENTARY INFORMATION:

Electronic Access

You can view and download this document by going to the Web site of the Department's Docket Management System (http://dms.dot.gov/​). On that page, click on “simple search.” On the next page, type in the last four digits of the docket number shown on the first page of this document, 4043. Then click on “search.” An electronic copy of this document also may be downloaded from http://regulations.gov and from the Government Printing Office's Electronic Bulletin Board Service at [phone]. Internet users may reach the Office of the Federal Register's home page at: http://www.archives.gov/​federal_​register/​index.html and the Government Printing Office's database at: http://www.gpoaccess.gov/​fr/​index.html.

Anyone is able to search the electronic form of all comments received into any of our dockets by the name of the individual submitting the comment (or signing the comment, if submitted on behalf of an association, business, labor union, etc.). You may review the Department's complete Privacy Act Statement in the Federal Register published on April 11, 2000 (65 FR 19477-78) or you may visit http://dms.dot.gov.

Public Meeting

Based on the significant proposed changes to the O&D reporting system, the Department is considering holding a public meeting. If necessary, the public meeting would allow the Department to gather additional input from the Air Carriers and other stakeholders. Any meeting would be open to the public and a record of the meeting would be placed in the rulemaking docket. If the Department decides a public meeting is necessary, the Department will publish a notice announcing the meeting in the Federal Register.

Table of Contents

A. Authority

B. Background

1. Current Method of Collecting O&D Survey Data

2. Current Method of Collecting T-100/T-100(f)

3. Office of Inspector General's Report

4. Advanced Notice of Proposed Rulemaking

C. Need for Data Modernization

1. Background

2. Review of Deficiencies in the Current O&D Survey

D. O&D Survey Data Usage

1. The Department

2. Other Government Agencies

3. Other Stakeholders

E. Limitations of O&D Survey and T-100/T-100(f)

F. Need for Regulatory Action

G. Development of the Record in this Rulemaking

H. Scope of this Rulemaking

I. O&D Survey Redesign

1. Summary of the Proposed O&D Survey

2. Discussion of the Proposed O&D Survey

3. Reporting Requirements

4. Significant Issues Related to the Data to Be Collected

5. Transition Period

J. T-100/T-100(f) Considerations

1. Background

2. T-100/T-100(f) Changes to be Considered

K. Data Dissemination

1. Dissemination of Data by Month

2. Proposed Construction of One-way Trips

3. Proposed Proration Method

4. Proposed Changes to Confidentiality

L. Rulemaking Analyses and Notices

M. Glossary

N. Proposed Rule

A. Authority

The Civil Aeronautics Board Sunset Act of 1984 (Pub. L. 98-443) requires the Department of Transportation (the Department), under the authority of the Secretary for Transportation (49 U.S.C. 329(b)(1)), to collect and disseminate information on civil aeronautics and aviation transportation in the U.S., other than that collected and disseminated by the National Transportation Safety Board. The Department must, at minimum, collect information on the origin and destination of passengers and information on the number of passengers traveling by air between any two points in air transportation. Additionally, the Department must be responsive to the needs of the public and disseminate information to make it easier to adapt the air transportation system to the present and future needs of the commerce of the U.S. (49 U.S.C. 40101(a)(7)). In meeting this responsibility, the Department collects data submitted under 14 CFR Part 217 (Reporting Traffic Statistics by Foreign Air Carriers in Civilian Scheduled, Charter, and Nonscheduled Services), 14 CFR Part 241 (Uniform System of Accounts and Reports for Large Certificated Air Carriers) and 14 CFR Part 298 (Exemptions for Air Taxi and Commuter Air Carriers).

Under 14 CFR Part 217, Foreign Air Carriers that are authorized by the Department to provide scheduled passenger services to or from the U.S. must file Form 41 Schedule T-100(f) “Foreign Air Carrier Traffic Data by Nonstop Segment and On-flight Market,” accumulated in accordance with the data elements prescribed in Section 217.5 (14 CFR Part 217 section 217.3). These requirements reflect changes made to international data submissions by large Air Carriers (Docket No. OST-1996-1049, RIN 2105-AC34, 62 FR 6715; Docket No. OST-1998-4043, RIN 2139-AA08, 67 FR 49217).

Under 14 CFR Part 241, all U.S. certificated and commuter U.S. Air Carriers must report their traffic movements in the T-100. Under 14 CFR Part 217, all Foreign Air Carriers that operate to the U.S. must report their traffic movements involving a U.S. point in the T-100(f). Participation in the O&D Survey is required by 14 CFR Part 241 Section19-7. The source documents are airline tickets ending in double-zero (major domestic markets) or zero (all other markets), reported only by the first honoring and Operating Air Carrier, which shall report the required data for the entire Ticketed Itinerary.

B. Background

This NPRM is part of an effort by the Department to conduct a broad-based review of the requirements for aviation data and to modernize the way the Department collects, processes and disseminates aviation data. Specifically, it addresses the collection and processing of traffic reporting requirements described in the O&D Survey and T-100/T-100(f). It reflects prior analyses of the aviation data collected and processed by the Department and the effective use of that data by the government, the airline industry, consumers, and other stakeholders, which indicate a need to revise and update the O&D Survey and T-100/T-100(f).

1. Current Method of Collecting O&D Survey Data

The O&D Survey collects a sample of itineraries quarterly from large certificated U.S. Air Carriers. Foreign Air Carriers granted antitrust immunity as part of code-share agreements with U.S. Air Carriers contribute O&D Survey data under a similar but separate program. The current method of gathering data for the O&D Survey requires large certificated Air Carriers that transport passengers (i.e. “Participating Carriers”) to examine each flight coupon to determine whether the ticket, or Ticketed Itinerary, is reportable. Reportable tickets are those with a ticket number ending in a double-zero (major domestic markets) or zero (all other markets). In practice, tickets ending in zero are reported, presumably representing ten percent of all Ticketed Itineraries. The ticket must be reported unless it is apparent that another Participating Carrier has already reported it. If it is not apparent, then the Participating Carrier must report the ticket. Data are reported quarterly.

If the Participating Carrier issued the ticket, it will likely have saved the itinerary data for use in reporting the ticket to the Department's O&D Survey. If the Participating Carrier did not issue the ticket, the Carrier must either receive the necessary data from the Carrier that issued the ticket or employ staff to examine the physical passenger document and transcribe as much of the Ticketed Itinerary as possible from a used flight coupon.

2. Current Method of Collecting T-100/T-100(f)

The current method of gathering data for the T-100/T-100(f) requires Reporting Carriers (e.g. all Carriers required by 14 CFR Part 217, 14 CFR Part 241, and 14 CFR Part 298 to report operating statistics) to report the movement of traffic in accordance with the uniform classifications prescribed. They are compiled by Flight-Stage as actually performed and represent 100 percent of operations. The requirements reflect revisions made to T-100/T-100(f) reporting requirements for both Foreign and Domestic Air Carriers (Docket No. OST-1996-1049, RIN 2105-AC34, 62 FR 6715; Docket No. OST-1998-4043, RIN 2139-AA08, 67 FR 49217). Data are submitted monthly.

3. Office of Inspector General's Report

At the request of The Bureau of Transportation Statistics (BTS), the Office of the Inspector General (OIG) audited the Passenger Origin-Destination Survey (O&D Survey) data submitted by the Air Carriers to the Department. The OIG report, released in February 1998, acknowledged that passenger data was critical for basic departmental responsibilities and for making sound policy decisions. It declared the O&D Survey to be insufficiently reliable for use in supporting these decisions. Specifically, the OIG report concluded that “[a]lthough O&D data are used by Department analysts to provide quantitative support for key policy and funding decisions, we found that O&D data are unreliable for use in making these important decisions.'' (Office of Inspector General Audit Report Number AV-1998-086 Feb. 24, 1998 p.iii).

4. Advanced Notice of Proposed Rulemaking

In July 1998, the Office of the Assistant Secretary for Aviation and International Affairs and BTS jointly issued an advance notice of proposed rulemaking (ANPRM) (July 15, 1998, 63 FR 28128) as a first step in reviewing aviation data collected by the Department (Docket OST-1998-4043-1). The Department solicited comments about (1) whether the existing airline traffic and financial data should be amended, supplemented or replaced; (2) whether selected forms and reports should be retained, modified, or eliminated; (3) whether aviation data should be filed electronically; and (4) how the aviation data system should be reengineered to enhance efficiency and reduce costs for both the Department and the airline industry. The ANPRM explored not only the scope of traffic and financial information, but also the sources of data, the timing of the reporting of data, the methods of processing data, and the release of data to the public. The Department subsequently conducted additional outreach and research activities to further assess data requirements and potential improvements to the reporting and processing systems. In the ANPRM, the Department stated its goal that the aviation data systems should be reviewed and modernized to adapt to the present and future needs of commerce.

As a result of the ANPRM, the Department issued an NPRM on August 28, 2001, to assessment changes to the T-100/T-100(f) Traffic Reporting System (Docket No. OST-1998-4043, RIN 2139-AA08, 66 FR 45201). On July 30, 2002, the Department issued a final rule modifying the T-100/T-100(f) Traffic Reporting System (Docket No. OST-1998-4043, RIN 2139-AA08, 67 FR 49217). This NPRM proposes additional data modernization changes that were not previously addressed in prior rulemakings.

C. Need for Data Modernization

In 1947, the U.S. Government under the Civil Aeronautics Board (CAB) began keeping information about the origin and destination of passenger air travel based on passenger reservations. In 1968, the O&D Survey was overhauled and the basis of counting passengers was changed to the present system of counting sold tickets reported after first use. With the exception of a few added data elements to record code-share ticketing, the O&D Survey collected today has changed little since 1968, although some changes were made to the T-100/T-100(f) (Docket OST-1996-1049, RIN 2105-AC34, 62 FR 6715; Docket OST-1998-4043, RIN 2139-AA08, 67 FR 49217). The industry, however, has changed a great deal since then.

1. Background

Worldwide, the scheduled air transportation industry is divided into those Carriers that share passengers with one another on the same Air Travel Ticket, a practice called interlining, and those Carriers that operate independently without interline agreements. For both types of Carriers, only one Carrier serves as the Issuing Carrier, but for interlining Carriers, the Issuing Carrier plays a coordinating role for all other Carriers included in the Ticketed Itinerary. The Issuing Carrier is responsible for holding the ticket purchaser's funds until they are earned, paying taxes due to government agencies, and paying the travel agent commission, if any. The Issuing Carrier is also known as the plating Carrier because, in the age when flight coupons had red carbon paper backing, the Issuing Carrier's three-digit identifier was stamped on a metal plate that travel agents and airline ticket agents used to imprint the first three positions of a 13-digit ticket number of an Air Travel Ticket.

The Issuing Carrier holds the ticket purchaser's funds until they have been earned by providing transportation to the passenger. When the passenger's travel plans include travel on multiple Carriers on the same Ticketed Itinerary, the Carrier that transports the passenger provides evidence to the Issuing Carrier that the passenger has been transported in order to receive its share of the funds. This process is called “interline settlement” or “interline billing.” When presented with evidence that the passenger has been transported, the Issuing Carrier credits the billing Carrier with its prorated share of the passenger's fare. Since sharing passengers internationally is common, the interline billing process is standardized worldwide across all Carriers that choose to interline passengers. Because travel agencies all over the world sell tickets on Carriers located in many countries, and because passenger travel plans often involved multiple Carriers, interlining Carriers and travel agents worldwide created the standard agent ticket, which is used universally by interlining Carriers. These Carriers use identical, or near identical, billing processes to facilitate the handling of shared tickets. Even when travel is scheduled on a single Carrier, extenuating circumstances due to weather, mechanical, or other operational difficulties can result in passengers being transported on multiple Carriers. After accommodating a displaced passenger, the Carriers use standard interline billing processes to transfer funds from the Issuing Carrier to the Carrier that transported the passenger. Carriers that do not choose to interline passengers and that do not rely on travel agents to distribute their travel products are not bound by these standard procedures and agreements, but most Carriers choose to use industry standard procedures nonetheless.

Tax authorities generally require the Issuing Carrier to remit all taxes and fees associated with the Air Travel Ticket on behalf of all Carriers that appear on the Ticketed Itinerary. The Issuing Carrier, regardless of the identity of the Carrier that will operate each Flight Coupon Stage, will remit the tax tied to each Flight Coupon Stage. A case in point is the Aviation and Transportation Security Act (ATSA), Public law 107-71. Under the ATSA, the Issuing Carrier remits the September 11th Security fee. Even though the fee is calculated based upon the number of Flight Coupon Stages in the Air Travel Ticket, carriers that transport the passengers have no responsibility for collecting and remitting this fee.

For example, a passenger purchasing non-stop service transportation from Washington to St. Louis and back will be assessed the September 11th Security Fee one time for each One-way Trip. The Issuing Carrier will remit the September 11th Security Fee within 60 days of the purchase of the ticket, regardless of the scheduled travel date. Here, if U.S. Airways, Inc. (US Airways) issues a Ticketed Itinerary with outbound travel on US Airways and return travel scheduled several months later on United Air Lines (United), it is the responsibility of US Airways, as the Issuing Carrier, to remit the September 11th Security fees for travel on both outbound and return travel. Passengers pay the September 11th Security fee based on the number of enplanements described in the Ticketed Itinerary, not on the number of actual enplanements that the exigencies of travel actually require the passenger to make. If, on the day the passenger leaves Washington, a problem arises that results in the passenger traveling to another city (and, perhaps, on another Carrier) to change planes before continuing on to St. Louis, the passenger is not assessed a second September 11th Security Fee because the assessment of the September 11th Security Fee was made by the Issuing Carrier when the itinerary was issued.

It is a misnomer to say that travel agents issue tickets. Travel agents distribute (sell or issue for free) Ticketed Itineraries on behalf of an Issuing Carrier, and send the pertinent information about the sale, and the proceeds of the sale, to the Issuing Carrier. Originally, travel agents remitted funds directly to Issuing Carriers. With growing numbers of airlines, the international nature of air travel, and growing numbers of travel agencies, Carriers and travel agencies throughout the world formed clearing houses, which came to be known as Bank Settlement Plans (BSPs), to provide a central location for handling Air Travel Tickets distributed (sold) by travel agents. There is a BSP for each country or, sometimes, clusters of countries. Travel agencies in North America remit sales to the Airlines Reporting Corporation (ARC), organized in the early 1980s, which operates in much the same way that BSPs operate in other parts of the world.

When the current O&D Survey was established in the 1960s, the most common accounting system was a lift-based system. The airline industry used flown flight coupons, also known as lifts, as the primary source of accounting and marketing data. It was customary to make a reservation, and then ticket the reservation at a later time. The ticket consisted of one flight coupon for each enplanement and a summary or auditor's coupon. Every flight coupon contained all the information about the itinerary.

Moving all evidence of the ticket sale to each airline's accounting center was time-consuming and laborious. In the years prior to the widespread use of computers, tickets sold in the U.S. took weeks to reach the Carrier; tickets sold in foreign countries would typically take months. Some ticket sales were processed within a week or two, but very often sales took so long that the passenger had completed the journey before the Issuing Carrier processed the sale of the Air Travel Ticket. In contrast, after each flight departure, the airport personnel sent a flight envelope containing all the flight coupons to the Operating Air Carrier's accounting offices for processing. The flown flight coupons came to the accounting center organized in flight envelopes for flights departed mostly in the prior week. By virtue of the ubiquitous red carbon paper, every flight coupon included a copy of the entire itinerary. Therefore, in a pre-computer environment, a lift-based accounting system organized around the lifted flight coupons made sense. Taxes and commissions had to wait until the sale records reached the Issuing Carrier, but in a lift-based accounting system, a Carrier's accounting and market data needs were met with the information on the lifted flight coupon.

In 1968, the CAB designed the O&D Survey around the lifted flight coupon to reflect the standard procedures that were in use in the airline industry. Collecting the ticket sale data after one coupon had been used was not only in line with Carrier accounting practices of the time but also had two other advantages. First, this collection method grouped the reported tickets together in a date close to the passenger's use of a flight coupon rather than the ticket issue date. Second, it kept fully refunded and fully exchanged tickets from being included in the O&D Survey.

The CAB also recognized that manual procedures are labor intensive and expensive. In keeping with the desire to minimize the burden of collection, the CAB specified very few elements from the ticket for collection, required only 10 percent of the tickets to be examined, and limited the number of surveys to four a year.

The Carriers were early adopters of computer systems. The first of the customer interactions to be automated was the reservation process. The major Carriers built large reservation systems to match passengers to departing aircraft. The reservations system computers had an operating system that was designed specifically for the requirements of Carrier reservation systems. Passengers and travel agents worldwide called Carriers to make a reservation and the airline employees entered the passenger information. Several of the Carriers eventually packaged their systems as a product, called a Computer Reservation System (CRS). They sold the ability to access the reservations system to the travel agents. Marketed as Sabre, PARS, Apollo, and System One, the CRS owners gained revenue from others' access to the system, and Carriers lowered their costs because travel agents, rather than airline employees, were now entering the passenger information into the reservations system.

When the reservations systems began to issue automated tickets, the travel agent and the airline ticket counters achieved higher efficiency and productivity. Automated ticketing lowered costs by copying data already in the reservations system onto a paper ticket. However, since the reservations computer operating system was incompatible with the Carrier accounting computers, the information from the ticketing record had to be copied again onto an electronic record that was transmitted to the Carrier's accounting computer systems. Since the accounting system received a copy of the ticket data but not a direct link to the reservations system, the accounting system had no direct way of recording changes made in the reservation system.^[1] Changes to the passenger's reservation that were important enough to cause an agent to re-issue the ticket would, in turn, generate a new ticket record that would be forwarded to the accounting system. Changes to the passenger's reservation that did not cause an agent to re-issue the ticket would not be communicated to the accounting system. Nevertheless, whereas moving manual ticket data from the ticket sellers to the Carriers had been laborious, slow, and costly, the automated computerized ticketing process opened up new possibilities to move ticket information quickly, efficiently, and at low cost to Carriers.

Automated ticket processing opened up cost saving opportunities in passenger revenue accounting. The huge cost of rewriting an accounting system from lift-based to sales-based was justified, in part, because the lift-based accounting system required hundreds of employees trained to process the lifted flight coupons. Because a sales-based accounting system makes use of information already stored in the computer, Carriers gradually shifted away from reliance on information from lifted flight coupons and toward reliance on information stored from the ticket sale. By 2004, Carriers use sales-based accounting systems almost exclusively.

Regardless of the accounting system, there remained a gap in data when the itinerary included multiple Carriers. Only the Carrier that issued the ticket had a complete computer record of it. A Carrier that transported a passenger on a ticket that it did not issue had to employ staff to enter the itinerary into its computer system. In the 1980s, American Airlines initiated agreements to share ticket information about shared passengers with Trans World Airlines, United Air Lines and Eastern Airlines to avoid the cost of manually re-typing each other's tickets. In 1990, the system of sharing ticket information was formalized with an industry standard record structure for all Carriers called Transmission Control Number (TCN) record. Whenever a Carrier needed to share information about a ticket with the other Carriers in the itinerary, a TCN record could be sent between Carriers. Responsibility to oversee the data sharing was given to the Airline Tariff Publishing Company (ATPCO). ATPCO would forward TCN records to the operating Carriers in the itinerary on behalf of the Issuing Carrier. The ATPCO TCN exchange service was offered to all Carriers, although not all Carriers decided to participate.

The TCN data sharing was created as an optional service to facilitate more efficient information exchange among interlining Carriers electing to use the service, not as a compulsory system. Tickets continued to be created without a corresponding TCN record. Conversely, multiple TCNs were sometimes created to describe a single sale. Sometimes this happened because TCN records were generated for tickets for customers who failed to complete the purchase. Other times, customers demanded a change that resulted in a second TCN being created while the first could not reliably be nullified. Testing can generate a TCN or, sometimes, TCNs by the thousands, for which there was no ticket sale. Carriers' passenger revenue accounting systems were designed to find the TCNs they needed for accounting purposes, ignore the extraneous TCNs, and still be able to accept manual data on tickets for which no TCN exists. Not all Carriers used TCN records in the course of business. Of those that did, some created TCNs for their own internally-issued tickets, while other Carriers did not.

After the CRSs became known as Global Distribution Systems (GDSs) in the 1990s, they inherited the responsibility to create the TCN records for travel agency tickets. With this development, TCNs became the vehicle to send information about the ticket from the travel agencies to the Issuing Carrier as well as to any other Carrier that participated in the itinerary. The GDSs sell the TCN information to the Carriers for a small fee. The GDSs also sell the travel agent's reservation information. The product, called marketing information data tapes (MIDT), contains no information about the price of the travel except the selling class codes and is limited to segments booked through travel agencies. The MIDT data are marketed to Carriers for use in business planning activities.

While increasing computerization simplified many of the carriers' data collection, processing, and exchange activities, manual collection of the O&D Survey information became more difficult for the Participating Carriers. With reliance on computerized ticketing and the shift to sales-based accounting systems, there was little interest or need to continue the practice of using carbon paper to print the whole itinerary on all of the ticket's flight coupons. Examination of coupons, standard procedure in the old lift-based system, is not necessary in the normal course of business when using a sales-based accounting system. Since the Department's O&D Survey continued to require the Operating Air Carrier to provide information from the lifted flight coupons, it became increasingly vital for the Operating Air Carrier to receive information about the issuance of the ticket from the Issuing Carrier. If the first Participating Carrier is not the Issuing Carrier or did not receive that sale information from the Issuing Carrier, then the Participating Carrier is required to employ staff to locate that lifted flight coupon. This is an intensely manual process, and it is a significant burden on limited human and financial resources of the Operating Air Carrier. In the pre-computer era, Carriers could draw on accounting department employees trained in obtaining information from lifted flight coupons, but increasing reliance on computer records and sales-based accounting systems left Carriers with only a small number of employees with sufficient training to glean the O&D Survey information from a lifted flight coupon. Sales processing by computer has become so reliable that as of May 2004, the GDSs no longer print a paper version of the auditor's coupon. Employees with the skills needed to extract the necessary information from visual examination of a lifted flight coupon have become increasingly scarce.

The level of effort that the current O&D Survey imposes on an Operating Air Carrier to identify whether it is the first Participating Carrier in the itinerary is compounded by the number of Carriers the Department exempts from reporting to the O&D Survey. Tens of thousands of passengers fly each day on commuter Carriers and Foreign Air Carriers operating under code-share agreements. As a result of code-share ticketing procedures, the identity of the Operating Air Carrier is often hidden from an outside observer. When the Issuing Carrier does not provide the itinerary details to the Operating Air Carrier, via a TCN record or other means, then it is difficult for the Operating Air Carrier to determine whether any of the other Carriers whose Airline Designator appears on the ticket as the Marketing Carrier is scheduled to operate the flight. A Participating Carrier may not be aware that a Code-Share partner is scheduled to operate a flight. The CFR specifically absolves the Participating Carrier from the burden of determining the scheduled Operating Air Carrier if the Issuing Carrier did not notify it and it is not a Carrier involved in the code-share agreement.

If the reporting carrier does not know the operating carrier on a downline code-share segment, it would use the ticketed carrier's code for both the operating and the ticketed carriers. The reporting carrier is not responsible for knowing the operating carrier of a downline code-share where it is not a party to the code-share segment.

—14 CFR Sec 19-7 V. Selection of Sample and Recording of Data (D)(2)(b)

In addition to the higher cost, examination of a printed paper coupon to obtain information that is usually transferred by computer yields less information than it did in the 1960s, when manual processing was the norm. Electronic ticketing has become the standard practice for most U.S. Air Carriers. However, when authorization to board a plane must be communicated between Carriers, and electronic means are for any number of reasons unavailable, issuing a paper flight coupon remains the standard practice of the industry.

The O&D Survey requires Participating Carriers to report information about an entire ticket based on the knowledge of the flight coupon they have in hand. Paper coupons today generally only contain the information for a single flight segment. The itinerary must be deciphered by examining the pricing area of the ticket. Unfortunately, the pricing area lists city codes instead of airport codes. For cities with only one airport, the limitation poses no problem, but for cities such as New York, the pricing area will list the price to NYC. The use of NYC obscures whether the passenger is scheduled to arrive at LaGuardia (LGA) or Kennedy (JFK) or, for that matter, at Newark (EWR) or Newburgh (SWF) airports.

The passengers' purchased itinerary has always been limited to four segments per ticket because only four could be printed plainly on carbon paper copies. If a passenger's itinerary required more than four flight coupons, the Carriers used two or more tickets in conjunction with each other. When the itinerary was long enough to require spanning two tickets, the information from the second ticket was never available to the Participating Carrier. Recognizing this, the Department exempted the Participating Carrier from reporting the second and subsequent conjuncted tickets from the O&D Survey. However, even when some portions of the Ticketed Itinerary go unreported, the total amount collected for the ticket is still reported in full. Reported flight coupons are artificially over-valued when the full ticket value, but only the partial itinerary, is reported. The number of partially reported itineraries currently being reported in the O&D Survey is assumed to be low, but since they are not detectable, there is no ability to quantify them, and, therefore, the impact of exempting long itineraries on the current O&D Survey is unknown.

Reliance on the ability of the Operating Air Carrier to examine the lifted flight coupons no longer provides the best reasonably obtainable economic information about the purchase of air travel on scheduled Carriers. The Department acknowledges that the current O&D Survey burdens Participating Carriers with obligations to examine the details of lifted flight coupons that they would not ordinarily do in the course of their business.

Significant among these burdens is the obligation to determine first Participating Carrier. Under the requirements of the current O&D Survey, the only way to meet the obligation of determining whether an Operating Air Carrier is the first Participating Carrier is for each Operating Air Carrier to examine the complete routing of every Ticketed Itinerary that was used to transport passengers in the quarter. There is no other way for Operating Air Carriers to determine whether or not it is apparent that another Participating Carrier has already reported the ticket.

The Survey data are taken from the coupon that is lifted by a participating carrier, unless it is apparent from the lifted coupon that another participating carrier has already recorded and reported the data, in which instance the ticket coupon is non-reportable for the second honoring/participating carrier.

—14 CFR Sec 19-7 Appendix A (I.) General Description of O&D Survey (B) Narrative Description

The “unless it is apparent” standard for determining whether an Operating Air Carrier is responsible for reporting a Ticketed Itinerary is a difficult standard to meet. Every Operating Air Carrier must diligently examine every Ticketed Itinerary to find out whether it has a ticket number ending in zero. For ticket numbers ending in zero, when the Operating Air Carrier is the initial Carrier in the routing, then clearly it should report the Ticketed Itinerary. When the Operating Air Carrier is the second or third Carrier in the routing, it must compare the identifiers of the previous Carriers in the routing to the list of Participating Carriers provided by the Department's Office of Airline Information (OAI). Under the current regulation, even the most diligent Participating Carrier will not report all O&D Survey tickets correctly if there is an unrecognized code-share flight present in the itinerary, the itinerary spans multiple physical tickets (known as conjuncted tickets), or the itinerary includes cities with multiple airports.

2. Review of Deficiencies in the Current O&D Survey

Respondents to Docket OST-1998-4043-1 (ANPRM, July 15, 1998; 63 FR 28128) agreed that the O&D Survey, as it exists, exempts too many passengers from the report, is cumbersome and expensive to compile, and fails to collect key elements of information. In addition, the results of the O&D Survey published by the Department are unwieldy to use. The Department wishes to address problems such as those identified in the 1998 OIG report, which concluded that O&D data were unreliable for use in key policy and funding decisions.^[2] For example, the Inspector General determined that of 8,894 city pairs, the O&D Survey report on 6,661 city pairs (69 percent) did not meet the Department's accuracy criteria when using enplanement statistics as a benchmark. The Inspector General (IG) used the enplanement statistics as a reliable comparison because they are also used by the Carriers for aircraft operational purposes. The IG cited several reasons for the inaccuracies, most of which were attributed to the fact that the basic reporting requirements of the O&D Survey have not been aligned with current industry practices.

a. Reporting Exemptions

Exemptions from reporting, granted in the 1960s, have become a major problem in today's O&D Survey. For example, Carriers flying planes with 60 or fewer seats are exempt from reporting. As such, passengers whose entire itineraries are flown on smaller Carriers will not be reported, yet their participation in the air transportation system is critical. Similarly, code-share agreements between large and small Carriers were non-existent when the current O&D Survey was designed. Today, Carriers of all sizes are connected to a global air transportation system through global alliances and international ticket agreements. This intertwining of service adds complexity and increases the potential for error when reporting Ticketed Itineraries.

For example, the IG pointed out that a Participating Carrier is exempt from proper reporting of the code-share relationship if it has no knowledge of that relationship. In a code-share situation, the Carrier that transports the passenger (Operating Air Carrier) is not the Carrier printed on the itinerary (Marketing Carrier). The Carrier that issues the ticket is responsible for knowing when this is occurring and notifying the passenger of the code-share situation. However, when the Participating Carrier is not the Issuing Carrier, the Participating Carrier cannot always report the code-share portions of the Ticketed Itinerary properly.

Code-sharing with regional Carrier partners has created a situation wherein customers can begin travel on a regional Carrier that does not report the O&D Survey because of size exemptions. In that case, the second Carrier in an itinerary should report the ticket. However, the second Carrier may not be a code-share partner with the regional Carrier that first transported the passenger. The second Carrier will believe the ticket to have been reported by the first Carrier when, in fact, it has not been reported. This causes the entire itinerary to go unreported.

Exceptions for Foreign Air Carriers also impact the accuracy of the O&D Survey, and the IG cited this exception as a prominent problem. Excluding those Foreign Air Carriers granted antitrust immunity for alliances with U.S. carriers, Foreign Air Carriers may transport passengers without reporting their Origin and Destination traffic to the Department. In consequence, some travelers bound for foreign countries are counted in the Department's statistics, and some are not. Excluding these passengers introduces a bias into the statistics that is difficult to evaluate. As the code-share and marketing alliances between U.S. and Foreign Air Carriers developed throughout the 1990s, this reporting gap became even more significant.

b. Sample Size

The IG pointed out that having Participating Carriers report only those tickets ending in zero or double-zero is not an appropriate sample design. It is not certain that those tickets will be randomly distributed across all Ticketed Itineraries. A survey must be based on a random sample of the population if the results of the survey are to be generalized to the entire population. Unfortunately, there are indications that the sample used in the existing O&D Survey is not entirely random, although it is not always clear how this non-randomness occurs.

When the O&D Survey was established, ticket numbers were preprinted sequentially on paper ticket stock. As each customer appeared, each had an equal chance of receiving a ticket number ending in zero. Since ticket numbers are now assigned by a computer program, the possibility that ticket numbers are assigned for reasons other than randomness arises. For example, a tour operator might use its block of ticket numbers to issue all the ticket numbers that end in the same digit to members of a particular tour, resulting in all those tickets being selected for the sample or excluded from the sample depending on which tour was assigned ticket numbers ending in zero. One Carrier has analyzed its ticket numbers and found that 11 percent end in zero, which would not occur if the numbers were entirely random. While the sample is intended to be 10 percent of all tickets, analysis by BTS' Office of Statistical Quality in 2001 concluded that the actual sample size ranged from 10.1 percent in 1999 to 9.6 percent in 2000. This is a larger variation than one would expect purely from normal sampling error, suggesting some non-randomness in the creation or selection of ticket numbers.

c. Definition of Origin and Destination

The common understanding of a True O&D is a passenger who is traveling from the origin of the trip to arrive at the destination of the trip where the individual intends to conduct business or engage in leisure activity. Passengers generally prefer to arrive at the True O&D destination in the fewest possible Flight-Stages, but often a passenger travels over many Flight-Stages, many Flight Coupon Stages, and, sometimes, many modes of transportation to reach the True O&D destination, and in the case of a very remote destination, the journey might take several days. The Department's intent has always been to track, to the greatest extent possible, the passenger's intended True O&D.

Carriers, airports, the Department, and other stakeholders use various methodologies to approximate the passenger's True O&D. The standard approximation is known as a One-way Trip. The principal determination of One-way Trip is based on the time spent on the ground between sequential Flight-Coupon Stages. A short time between sequential Flight-Coupon Stages implies a connection in a continuing One-way Trip. A long time on the ground between sequential Flight-Coupon Stages implies an end of the prior One-way Trip and a beginning of the next One-way Trip. Flight Number and Fare Basis Code are sometimes used, in addition to time on the ground, to calculate a One-way Trip. The One-way Trip is usually completed in a single day, although the definition of One-way Trip encompasses the possibility that travel continues overnight and into the following day(s).

However, the information Carriers currently supply in the Department's O&D Survey is devoid of flight number, travel date, departure time and arrival time, so the data collected by the Department has left it without the ability to use time spent on the ground to establish a One-way Trip. As a result, since the beginning of the O&D Survey, the Department has used continuous direction of travel as its approximation of True O&D. This methodology is known as Directional Passenger construction. In a regulated airline environment, determining passenger trips by measure of least circuity was an adequate measure of passenger travel. In that environment, passengers had no incentive to travel in any direction other than toward their destination as efficiently as possible. However, following the extensive development of hub-and-spoke systems following deregulation, passengers are often motivated by price or incentivized by Carrier loyalty programs that reward taking circuitous connecting flights even when a non-stop flight is offered.

The Department's Directional Passenger concept considers a passenger to be on a continuous trip so long as the passenger continues in the same direction regardless of the number of days the journey takes, subject to certain circuity rules that allow some backtracking. For example, the Department's circuity based rules consider an itinerary of Albuquerque to Denver to Reno to be a single Directional Passenger trip. However, an itinerary of Albuquerque to Denver to Las Vegas will never be considered as a single directional trip because the location of Las Vegas airport in relation to Albuquerque causes the circuity check to break the trip into two directional passenger trips. Because the Department does not collect flight date or flight time, the O&D Survey always identifies Albuquerque to Denver to Reno as a single Directional Passenger trip, regardless of the number of days the passenger stays in Denver. On the other hand, regardless of the short number of hours spent in Denver, the O&D Survey always identifies Albuquerque to Denver to Las Vegas as one Albuquerque to Denver Directional Passenger trip and counts the Denver to Las Vegas stage as a separate Directional Passenger trip.

Itinerary construction and circuity rules together determine Directional Passengers. When an Albuquerque-Las Vegas passenger purchases a round trip ticket traveling through Denver on both the outbound and the return trip, then the directional passenger rules will recognize the pattern, and determine that the outbound journey should be considered a single Albuquerque-Las Vegas trip and the return trip to be a single Las Vegas-Albuquerque trip. However, when an Albuquerque-Las Vegas passenger purchases a round trip ticket with the outbound journey changing planes in Denver and a return trip changing planes in San Francisco, then the directional passenger rules will interpret the outbound journey to be an Albuquerque-Denver trip, the return trip will be a San Francisco-Albuquerque trip with a separate Denver-San Francisco trip sandwiched between them. In this situation, the Directional Passenger construction views Las Vegas as a connecting city and does not recognize the passenger's true intention to visit Las Vegas. Itineraries like Albuquerque to Denver to Las Vegas have increased as a result of the development of extensive hub-and-spoke operations by incumbent carriers. Clearly, approximating True O&D using the Directional Passenger method is less accurate in the current environment than it was when it was instituted.

The Department cannot approximate True O&D consistently across all itineraries using the O&D Survey as it is currently collected. Furthermore, the Department cannot determine Directional Passengers on a consistent basis because travel that is part of a stand alone Directional Passenger trip is treated differently than if that travel is part of a round trip, and round trips are treated differently depending on the airport in which a passenger might choose to change planes.

In authorizing Passenger Facility Charges (PFCs), the Congress recognized the concept of One-way Trip in civil aviation law. No PFC on any passenger may be imposed for more than two boardings on a One-way Trip (14 CFR 158.9(a)(1)). The concept of One-way Trip was further ensconced in Federal law on November 19, 2001, when Congress established the September 11th Security Fee. Section 44940(b) and (c) of ATSA provides that the fee may not exceed $2.50 per enplanement or $5.00 per One-way Trip. Congress did not specify the definition of One-way Trip, but it is commonly understood that it was to be a journey from the passenger's point of view, concomitant with common practice.

The Carriers assess PFCs and September 11th Security Fees using time in hub as the principal determinant of a One-way Trip. The Department believes that the Carrier's method of determination for the One-way Trips is an acceptable methodology. However, because the Department uses directional travel as the determinant of its passenger counts, it cannot effectively monitor the enforcement of these Federal laws. Since the Department's Directional Passenger methodology for determining passenger counts does not match the One-way Trip methodology for determining passenger counts being used by the Air Carriers to assess the fees, the Department's counts can, at best, predict only the approximate value of the fees due to government agencies.

The Department's inability to measure One-way Trips consistent with industry standards leaves it without an adequate measure of passenger demand for air travel in the U.S. The OIG issues reports on airline metrics ^[3] that use the number of air travelers enplaned as the measure of air traffic demand. While the number of enplanements can be an accurate measure of passenger demand at individual airports, it has unfortunate implications when used as a measure of nationwide air traffic demand. When Carriers discontinue non-stop service between two airports, leaving connecting service as the sole option of passengers traveling between these airports, the number of enplanements doubles since passengers must now enplane a second aircraft. When enplanements are used as the sole measure of nationwide air travel demand, discontinuing direct service has the perverse effect of making it appear as if air travel demand is increasing. Thus the reduction in the true number of persons traveling after September 11, 2001 likely would be underestimated when using enplanements as a measure of demand, because the airlines' reduction in the number of non-stop flights caused the travelers to enplane more times to reach their destination. The Department believes that some of the perceived lack of accuracy in the O&D Survey is a result of measuring passenger traffic in terms of the Directional Passenger in an era when airlines are providing incentives for passengers to use circuitous connecting services.

d. Fares, Taxes, and Fees

Taxation of scheduled passenger aviation today is a combination of percentage of fare, ticket tax, itinerary-specific taxes such as international departure tax, and enplanement fees such as September 11th Security Fees, subject to limitations on the number of charges and fees that can be assessed on a One-way Trip. Because the O&D Survey commingles taxes and fees with the fare amount, exact measurement of the portion of the ticket price that represents tax has been an educated guess even when taxes were based on a percentage of the fare.

e. Passengers Versus Passenger Trips

It is generally believed that all the passenger counts reported in a quarter represent passengers scheduled to fly in that quarter. Rather, the current O&D Survey bundles all the travel on a Ticketed Itinerary in a single quarter. The complete itinerary is reported as if it took place entirely within the quarter in which travel commences. Therefore, a misunderstanding often exists between passengers reported and passenger trips. For example, all passengers who travel to a destination in December and return in January have all their travel reported in the December quarter; none of the passengers' journeys are reported in the first quarter of the next year.

f. Reporting Consistency

Different Carriers report data elements in different ways. For example, some Carriers with single-service cabins report all service as first-class, while others with single service cabins report all service as coach. Additional reliability problems occur because the Issuing Carrier sometimes provides the Participating Carrier with the information saved when the Ticketed Itinerary was issued, and sometimes it does not. When the Issuing Carrier does not provide information to the Participating Carrier, the Participating Carrier can only know what is printed on the lifted flight coupon and may find it difficult to report an itinerary correctly. Lack of correct knowledge is explicitly excused in the CFR.

When the Participating Carrier attempts to decipher the city codes for the complete itinerary using the pricing area of the ticket, inaccuracies can result. The designated city codes—not the airport codes—are present in the pricing section of the ticket. When the Carrier serves multiple airports in a metropolitan area, such as Dulles and Reagan National Airports in Washington, the pricing area displays WAS instead of the airport code. The segment's actual airport in that circumstance is unknown to the Participating Carrier. This is also the case with bulk tickets. Participating Carriers that are also Issuing Carriers can report the ticket price accurately, while Participating Carriers that did not issue the ticket, and did not receive a TCN, cannot report the actual amount paid. If the ticket value is not printed on the paper document, the Participating Carrier cannot know how to report it correctly.

The majority of users of the government's O&D Survey data purchase the data from third-party providers, which use internal decision rules to interpret the data. These independent companies obtain the data from the Department and reprocess it for sale. These companies make assumptions about the distortions that are inherent therein. For example, the third party providers perform extensive analysis on the data to separate the amount that was likely paid as fare from the amount that was likely paid as tax. Because the decision rules are specific to third-party providers, different interpretations of the same original data exist.

D. O&D Survey Data Usage

A diverse group of stakeholders including the Executive Branch and Congress use traffic data to help them in making decisions that affect the national air transportation system and the U.S. economy. Most responses to the ANPRM, including airports, labor unions, equipment manufacturers and industry consultants, identified the Department's aviation data as their most important source of data. These stakeholders depend upon the Department to provide accurate, timely, and comprehensive aviation data.

1. The Department

Air transportation is a significant sector of the nation's economy. Despite wars and economic downturns, the nation continues to experience long-term increases in demand for air travel. Through its efforts to measure economic activity, the Department affirms its role in fostering opportunities for transportation providers to create and maintain the best transportation system in the world and to enhance the quality of life of the American people, today and into the future. The Department uses aviation data to carry out its mandates, among them (1) improving international air services by seeking market liberalization, (2) ensuring the benefits of a deregulated, competitive domestic airline industry, and (3) developing policies to improve air service and/or access to the commercial aviation system for small and rural communities.

In particular, the Department uses O&D Survey information and the T-100/T-100(f):

• To exercise the Department's responsibilities for economic oversight of the airline industry as mandated under 49 U.S.C. 40101, including, but not limited to:
• (7A) “Developing and maintaining a sound regulatory system that is responsive to the needs of the public and in which decisions are reached promptly to make it easier to adapt the air transportation system to the present and future needs of the commerce of the United States”;
• (9) “Preventing unfair, deceptive, predatory, or anticompetitive practices in air transportation”;
• (10) “Avoiding unreasonable industry concentration, excessive market domination, monopoly powers, and other conditions that would tend to allow at least one air carrier * * * unreasonably to increase prices, reduce services, or exclude competition in air transportation”;
• (12A) “Encouraging, developing, and maintaining an air transportation system relying on actual and potential competition to provide efficiency, innovation, and low prices”;
• (13) “Encouraging entry into air transportation markets by new and existing air carriers and the continued strengthening of small air carriers to ensure a more effective and competitive airline industry”; and
• (16) “Ensuring that consumers in all regions of the United States, including those in small communities and rural and remote areas, have access to affordable, regularly scheduled air service”;
• As a base of information to assess, maintain, and preserve competition in the airline industry and in specific aviation markets, under various federal laws and programs, such as:
• To investigate allegations of unfair and deceptive practices and unfair methods of competition, under 49 U.S.C. 41712;
• To review proposed mergers and acquisitions to assess their competitive effect;
• To review code-share and marketing agreements between domestic major Air Carriers, under 49 U.S.C. 41720; and
• To review applications for antitrust immunity between U.S. and Foreign Air Carriers, under 49 U.S.C. 41308;
• To administer the Essential Air Services program assessing the air service needs of small communities (49 U.S.C. 41743);
• To administer the Small Community Air Service Development Program;
• To administer funds under the Aviation Investment and Reform Act for the 21st Century;
• To administer the Air Transportation Safety and System Stabilization Act;
• To monitor the trends and developments in the operating and competitive structures to ensure that Department policies remain consistent with commercial developments;
• To determine an Air Carrier's initial fitness to provide air transportation and review an Air Carrier's continuing fitness to provide air transportation (49 U.S.C. 41102);
• To evaluate certificate transfer applications (49 U.S.C. 41105);
• To grant or deny permits for Foreign Air Carriers to provide transportation as a Foreign Air Carrier to the U.S. by determining whether the public interest is being served in granting the permit (49 U.S.C. 41302) and to approve the transfer of such permit to another Foreign Air Carrier by determining whether the public interest is served (49 U.S.C. 41303); and
• To assemble information and prepare reports required and requested by the President and the Congress.

The O&D Survey and T-100/T-100(f), as currently collected, particularly impact the Department's evaluation of Air Carrier service to smaller communities. The Essential Air Services program (EAS) and the Small Community Air Service Development Program are directed towards smaller markets and require evaluation of service and fares. For example, under EAS, the Department determines the minimum level of service required at each eligible community by specifying a hub through which the community is linked to the national network, and specifying a minimum service level in terms of flights and available seats. Where necessary, the Department pays a subsidy to an Air Carrier to ensure that the specified level of service is provided. Similarly, research activities such as The Rural Air Fare Study,^[4] which was conducted pursuant to Section 1213 of the Federal Aviation Administration Reauthorization Act of 1996, require data on all passenger air travel, including many smaller markets served exclusively by airlines operating only aircraft having fewer than 60 seats.

The Federal Aviation Administration's (FAA) mandates include (1) regulating civil aviation to promote safety, (2) encouraging and developing civil aeronautics, including new aviation technology, (3) developing and operating a system of air traffic control and navigation for both civil and military aircraft, (4) researching and developing the National Airspace System and civil aeronautics, (5) developing and carrying out programs to control aircraft noise and other environmental effects of civil aviation, and (6) regulating U.S. commercial space transportation.

The FAA also administers the Airport Improvement Program (AIP) (authorized by 49 U.S.C. Chapter 471), which has the broad objective of assisting in the development of a nationwide system of public-use airports adequate to meet the currently projected growth of civil aviation. It also provides funding for airport planning and development projects. In addition, medium and large airports where one or two Carriers control more than 50 percent of passenger boardings must submit a written competition plan to receive approval to impose a Passenger Facility Charge (PFC) or to receive a grant under the AIP. All aspects of qualifying, planning, allocating, and monitoring of AIP funds rely on the integrity of the data that the Department collects.

The FAA uses O&D data for forecasting long-term growth in air travel demand and for determining corresponding needs for airport development and airspace system improvements. FAA also uses O&D data for conducting cost-benefit analyses of proposed safety rulemakings, infrastructure investments, and air traffic control improvements.

Within the Department, BTS has specific statutory responsibilities (49 U.S.C. 111(c)) to measure traffic flows, travel times, travel costs, and variables influencing traveling behavior and to collect data relating to the performance of transportation systems. BTS is specifically required to collect data that are suitable for conducting cost-benefit analyses.

BTS uses O&D data, together with other sources of passenger travel data (such as its National Household Travel Survey), to analyze passenger travel by all modes of transportation. Since passengers periodically shift the modes of transportation that they use (as they did after the terrorist attacks of September 11, 2001), passenger travel patterns by air are of great importance not only to airlines and airports, but also to transportation planners in other modes as well, such as highways and rail. BTS uses the O&D data to better understand what factors influence passengers' choices about which mode of transportation to use, so that transportation planners can plan appropriately.

The O&D data are used to measure the prices that passengers pay for air travel. These travel cost data are the basis of the Air Travel Price Index (ATPI), the price index developed for measuring airline prices.

Finally, the Department's Research and Special Programs Administration (RSPA) administers the Civil Reserve Air Fleet (CRAF) program, which provides civilian aircraft to the Federal government for use in war or other emergency situations. RSPA uses the T-100 to determine which Carriers can make what aircraft available, while minimizing the adverse effect that these commitments make to the airlines' normal civilian operations. Estimating these adverse effects requires data on the revenue that would be affected by the cancellation of any particular flight.

2. Other Government Agencies

a. The Department of Justice

The Department of Justice (DOJ) uses aviation statistics to assist in the prevention of anti-competitive conduct that is subject to criminal and civil action under the Sherman and Clayton Acts. The Department's aviation statistics have been one of the Justice Department's most important tools used to enforce various criminal statutes related to Sherman Act violations. DOJ also uses them to review mergers and acquisitions.

b. The Department of Homeland Security

The Department of Homeland Security (DHS) uses the Department's aviation data to help predict revenues from the collection of September 11th Security Fees. Because the Department's system bases its determination of passenger trips on least circuity, and the passengers are paying these fees on the basis of the industry standard One-way Trip, the Department's data provide poor predictions of these revenues. The current O&D Survey concept of Directional Passenger, which does not consistently predict the number of passengers arriving at the airport to change planes, which hampers DHS’ airport security manpower forecast. The ability to discern the difference between connecting passengers at a given airport versus passengers beginning their journey at that airport is critical to effectively managing security staffing and other resources at the airport. In addition, the O&D Survey cannot currently provide the critical time-of-day and day-of-week passenger volume data required by DHS to plan and forecast the manpower requirements of airport screeners.

Furthermore, the Air Transportation Safety and System Stabilization Act (Pub. L. 107-42) assigns the responsibility to remit the September 11th Security Fees for all travel described on the Air Travel Ticket to the Carrier that issues the ticket. Since the Department's O&D Survey information does not identify the Carrier that issued the ticket, the Department's data provide insufficient information for DHS to monitor the Carriers responsible for remitting the fees. Since the Federal government does not collect statistics about Carriers issuing tickets, the DHS uses the tickets reported in the O&D Survey as the best available substitute.

c. The Department of Commerce

The Department of Commerce's (DOC) ability to carry out its mandate to promote tourism is hindered by the Department's inability to know with certainty the beginning and ending of One-way Trips. Significant numbers of tourists travel by scheduled air transportation, and the Department's data collection policies leave DOC using only guesses about origins and destinations based on the Department's directional passenger counts.

The DOC's Bureau of Economic Analysis is also responsible for producing the official U.S. Government estimate of the Gross Domestic Product (GDP), and to adjust these estimates for inflation using the GDP Deflator. The GDP Deflator is a price index, similar to the Bureau of Labor Statistics' Consumer Price Index (CPI) that covers a broad range of prices, including prices not paid directly by consumers. The accuracy of the GDP Deflator would benefit from more accurate price data and more timely data. The reporting process proposed in this rulemaking would allow DOT to provide data that are more accurate to DOC. By the time the current quarterly O&D Survey data become available, it is no longer current, and, therefore, cannot be used in the GDP Deflator.

d. The Bureau of Labor Statistics

The Bureau of Labor Statistics (BLS) has a critical need for passenger O&D pricing information on a monthly basis, available promptly, so that it can achieve a more accurate index of air travel prices for incorporation into the monthly CPI. The proposed rule would provide these more accurate price data on a timely monthly basis. BLS’ ability to evaluate the cost of air travel and incorporate those evaluations into the consumer price index and the producer price index is compromised by the Department's current statistical techniques. Furthermore, the policy of reporting all travel in the quarter when travel commences compromises the attempt to allocate the cost of air travel to the proper travel month. The Producer Price Index (PPI) is supposed to be calculated net of taxes, but the Department's statistical data does not collect information to enable BLS to separate fares and taxes. Because BLS computes separate price indexes for purchases by consumers (the CPI) and purchases by producers (the PPI), it is important for BLS to be able to separate the purpose for which an airline trip is taken—whether business or leisure. The existing O&D data do not provide such information. The proposed rule would collect information that would enable better analysis of the purpose of travel.

BLS would like to adjust its monthly international price program for Exports by the amount paid by U.S. resident travelers to the Foreign Air Carriers on all routes. Because of the reporting exemptions granted to Foreign Air Carriers flying to the U.S., some U.S. citizens traveling to foreign destinations on Foreign Air Carriers are counted in the O&D Survey and some U.S. citizens are not. Lack of consistent Foreign Air Carrier statistics hinders BLS’ ability to keep its published statistics accurate and effective.

e. The Department of State

The Department of State (DOS) uses the Department's aviation data to provide the information base for policy decisions in international aviation negotiations.

f. The Government Accountability Office

The U.S. Government Accountability Office (GAO) uses O&D data to conduct special studies of the airline industry at the request of Congress. The quality of the analysis that GAO provides to Congress would be substantially improved by the additional and higher quality data collected under the proposed rule.

3. Other Stakeholders

Other stakeholders, such as public and private sector individuals, organizations, and agencies, rely on aviation data.

a. Existing and Potential Carriers

Carriers use the Department's data for traffic forecasting and evaluation of new routes. Evaluation of new market opportunities by Carriers is dependent on the O&D Survey. Even with their access to many internal sources of data, Air Carriers still report that they depend on the O&D Survey data. Almost all Carriers rely on the Department's data as the fundamental, and least expensive, source of industry demand data. For new Carriers, as well as smaller and low cost Air Carriers for which MIDT data is prohibitively expensive, the O&D Survey is the only viable source of traffic data. Third-party providers have developed new tools that enable smaller Carriers to participate in sophisticated route and strategic planning at a much lower cost. The success of such planning exercises is dependent, in part, upon the quantity and quality of data available to the Carriers. In addition, evaluation of traffic and routes is an essential component of aircraft acquisition planning.

b. Airports

Department traffic data provide the basis for analysis by the nation's airports. The O&D Survey, with its fare information, is the only source of information for airports to study price elasticity. In addition, the O&D Survey is the airports' primary source of data for evaluating new routes. The proposed O&D Survey would provide information about passengers originating at an airport and passengers transiting through an airport, an important distinction when planning for services that the passengers demand. Route evaluations are used to encourage new service from Carriers, and thereby improve their service to the consumer.

Smaller airports have a particular need for information about the destinations of passengers. Airports that do not have passenger volumes high enough to substantiate service to multiple cities need to establish service to cities in the region where the passengers using that airport want to go. When the airport can establish service only to a large city in one direction and most of the potential travelers in the area tend to travel in another direction, then the small airport that might have been viable on its own merits if it had service to the city in the appropriate direction may find that it must rely on the Federal government's small airport subsidy to remain viable. The O&D Survey is the primary source of destination information available to small airports.

Airports and state aeronautical agencies use the data to understand their customers and the airport's role in its regional transportation market. Airports must ensure that Air Carriers have reasonable access to essential airport facilities, so statistical forecasting of passengers is essential. Airport local and regional planning functions use, in part, Department O&D Survey and T-100/T-100(f) data to plan buildings and runways that are vital to expanding the nation's air transportation system into the future. Smaller airports, served primarily by Carriers that are exempt from current O&D Survey reporting requirements, are particularly hampered by the lack of relevant aviation data.

c. Consumers and the General Public

Consumers benefit from the availability and analyses of accurate and complete aviation data. In the past, the Department received numerous inquiries from the public regarding domestic airline fares. In response, the Department began issuing a quarterly report called The Domestic Airline Fares Consumer Report based on the Department's traffic data. It provides information about average prices being paid by consumers in the top 1,000 domestic city pair markets in the continental U.S. Similarly, Carriers have a vested interest in True O&D to effectively conduct route and other strategic planning. If Carriers are better able to accurately plan their services, consumers will be better served.

In addition, manufacturers, industry associations, consultants, academics, researchers, financial analysts, investors, and the general public use the Department's aviation data as the statistical base for a variety of studies on topics related to aviation.

d. Labor Unions

Labor unions consider the Department's data as a vital component of their negotiation strategies. Accurate and timely data are also crucial during times of economic downturn, particularly when Air Carriers request concessions from their unions.

e. Equipment Manufacturers

Because demand and traffic patterns reflect utilization of aircraft, demand and traffic data in the O&D Survey provide fundamental information on air transport markets that are vital in planning future products. Consequently, aircraft manufacturers are a prime user of the Department's traffic statistics.

E. Limitations of the O&D Survey and T-100/T-100(f)

The deficiencies of the O&D Survey and the T-100/T-100(f) have been known for some time. While changes were made to the T-100 and T-100(f) on July 30, 2002, the O&D Survey has not been substantially updated to reflect changes in the industry. It has become apparent that the cost of inadequate passenger and traffic information is significant for both the government and private sector aviation communities who rely on this data to fulfill their responsibilities and grow their businesses. Furthermore, recent changes in information technology and Carrier reservation and accounting systems have significantly reduced the cost of revising the Department's data collection requirements such that the benefits to all stakeholders of updating the system to provide more timely, accurate, and useful data far exceed the costs.

The current aviation era is characterized by rapid change. Carrier pricing can change multiple times a day. Carrier strategies sometimes change from month to month and require increasingly sophisticated analysis to support and evaluate business decisions and cases. The growth in the number of third-party providers of airline analytical software to evaluate the viability of new routes and other strategic decisions has made sophisticated Carrier analysis commonplace at even the smallest of Carriers. These software models, used by Carriers, consulting firms, and government agencies, require more detailed, timely, and comprehensive passenger demand data to optimize analyses of a dynamic industry and plan for its future. The Department's responsibility to identify and evaluate emerging trends in commercial aviation is constrained by traffic statistics that are only collected by month and by quarter and that are insufficiently comprehensive and detailed. The continuation of collecting insufficient, quarterly data to measure the transportation industry will severely hamper the ability of Federal, state, and local governments to provide the infrastructure to allow the airline industry to contribute to economic growth. Decisions on aviation infrastructure worth billions of dollars increasingly require more sophisticated analysis for which more accurate, timely, and comprehensive data are critical.

The nation is becoming more dependent on fast, efficient air travel. The nation's economy functions with the understanding that any person or any shipment of goods can be delivered across the nation within hours. Adequate quantitative data about the movement of passengers will help the Department prepare for the future needs of the transportation system.

Prior to September 11, 2001, delays associated with the capacity constraints of the air transportation system were undermining the efficiency of the system. These capacity constraints are now beginning to reemerge as demand recovers. Furthermore, the events of September 11, 2001, and the subsequent effects of those events on the aviation industry, further support the need for additional data modernization. Not only was the collection of data elements inadequate to measure important aspects of the aviation industry, vital information was not available in a timely fashion to interpret the short and medium term impacts of these events. It was also impossible to observe the recovery of the air transportation system in those crucial days after the system was restored.

More specifically, the data was inadequate for the following reasons: first, neither T-100/T-100(f) data (reported monthly) nor O&D Survey data (reported quarterly for ten percent, or less, of completed tickets) revealed daily changes in traffic and fares following 9/11. Without the ability to assess daily traffic levels, the Department could not fully assess the return of passengers to the nation's air transportation system and the extent to which the recovery was progressing differently in various regions of the country. Second, without any information about the sale of the Ticketed Itineraries, it was impossible to differentiate between the post September 11th passengers who purchased non-refundable tickets prior to September 11th and those travelers that purchased their Ticketed Itineraries after September 11th and thereby gauge the level of passenger confidence. Third, quarterly data submissions resulted in a significant delay in the Department's analysis of the impact of September 11th. The third quarter of 2001 O&D Survey data showed the 20 days most directly impacted by the events of September 11th mixed with the 71 days prior. The next data available in the O&D Survey could not be released until the end of the following quarter. Fourth, in implementing the provisions of the Air Transportation Safety and System Stabilization Act (Public Law 107-42), Congress and the Department exclusively relied on T-100 in providing assistance to Air Carriers and other industry participants. Even though the O&D Survey information is more useful in measuring some aspects of the nation's aviation economy, data collected only quarterly made it unusable for purposes of fulfilling the Air Transportation Safety and System Stabilization Act or for adequately monitoring the recovery of the industry following the terrorist attacks.

Although the events of September 11, 2001 were unprecedented, the need for more detailed, and more time-specific traffic data to monitor the impact of significant events on the industry and its recovery from them is not unique to that situation. Since the terrorist attacks, the industry has experienced the SARS outbreak, the Iraq war, and various elevated code orange alerts. In order to monitor the impact of these extraordinary events on the industry, the Department had to issue requests for supplemental data from the Carriers. Not only do these supplemental requests burden the industry with additional reporting requirements, they also highlight the fundamental need for the Department to routinely collect more detailed, time-specific data to fulfill its statutory obligations to monitor the health of the airline industry and respond to requests from Congress and other government agencies about the impact of such events on an industry that is vital to the U.S. economy. The current data collection systems are inadequate for providing timely answers to any question with more precision than a month for the T-100/T-100(f) and more precision than a quarter for the O&D Survey. Reliance on data that is only available quarterly for purposes of measuring the dynamics of airline prices is a critical shortcoming of the O&D Survey. The ATPI, for example, is severely handicapped by the limits of quarterly data. Flight date is an important element of the value of a flight and therefore an important factor in the computation of the ATPI.

The Transportation Security Administration (TSA) requires information about passenger travel by time-of-day and by day-of-week to plan airport security screener staffing requirements. The current T-100/T-100(f) averages data across a month and the O&D Survey averages data across an entire calendar quarter, so that variability over time within the calendar quarter cannot be measured. Variability over time and dates can only be measured if the Department begins collecting data about time and date of travel. The volume of passenger traffic varies by time-of-day and day-of-week and lack of information about passenger volumes can result in passenger delays due to too few screeners or in a useless expenditure of Federal dollars due to overstaffing at certain times.

TSA requires some method of forecasting the collection of revenue from the Air Carriers. The September 11th Security fee is remitted by the ticket's Issuing Carrier, but Issuing Carrier is not one of the data elements collected in either the O&D Survey or the T-100/T-100(f), making it difficult for TSA to forecast or monitor the proper remittance of tax dollars.

Neither the O&D Survey nor the T-100/T-100(f) provide any information about the sale of new tickets (e.g., changes in passenger booking windows), a key measure of traveler confidence in the air transportation system. Such information is critical to evaluating the likely financial impact of exogenous events, such as September 11th or SARS, on Carriers. In addition, these data limitations preclude the Department from precisely evaluating the impacts of even endogenous industry events such as potential strikes or Carrier shutdowns.

The problem resulting from the reporting exemption given to Air Carriers so long as they do not operate aircraft with more than 59 seats is illustrated by the emergence of Air Carriers flying substantial fleets of regional jets. For example, the commencement of operations by Independence Air in June of 2004 caused a profound adjustment of fares in small, medium and large markets in the Eastern half of the U.S. However, because Independence Air did not operate aircraft with more than 59 seats, it did not have to report O&D Survey data, thereby resulting in an incomplete picture of the effects of this Air Carrier's start of operations. When a major realignment of fares can result from the actions of an Air Carrier that qualifies for the small aircraft size exemption, then the small aircraft size exemption must be reevaluated.

The FAA acknowledged these and other issues at its 2001 Commercial Aviation Forecast Conference.^[5] Accurate and detailed data on the flow of passengers through the air transportation system is critical to addressing congestion and developing ways to make the system more efficient. The FAA requires data on the number of passengers flying at specific times of day and specific days of the week, allowing it to calculate more accurately the costs and benefits of safety regulations, infrastructure investments, operational changes, and other FAA actions.

Lack of information about catchment areas impacts the Department's ability to assess the effects of competitive services and alternative airports. A number of government agencies are charged with monitoring the airline industry and providing sufficient infrastructure to accommodate its growth. The use of secondary airports increasingly shapes the operating and competitive structures of the airline industry. These agencies increasingly require information that allows them to identify and analyze changes in the catchment areas of various airports, thereby understanding how such changes impact industry structure and airport and airway infrastructure planning and development. For the same reasons, such information would also be enormously useful to other users of the data, including airports, airlines, and aviation consulting firms.

BTS is specifically directed to gather data that are relevant to cost-benefit analysis. One requirement of cost-benefit analysis is estimating the number of people that are affected by a particular proposed regulation or infrastructure improvement or technology investment. A major weakness of the existing O&D Survey is that it does not provide flight-specific data, so it is not possible to estimate how many people are flying on any particular day of the week or at any particular time. Since infrastructure and air traffic control investments are most likely to produce benefits at times when the airspace system is congested, it is important to be able to measure how many people are flying at these times to measure of the number of people affected by proposed infrastructure and air traffic control improvements.

BTS’ current On-Time Data Base allows analysis of the particular flights that are affected by delays, but does not have the ability to know the number of passengers affected by delays. Since the number of passengers affected is likely to be greatest when congestion and delays are highest, current data are likely to understate the impact of delays on the traveling public. Information about the number of people traveling by time-of-day is vital to understanding the dynamics of the air transportation system.

The 10 percent sample is inadequate for fulfilling the Department's mandates and hampers the data quality of the O&D Survey. These data quality issues have a strong effect on programs that include measurements of air service to small communities. The EAS program is particularly impacted. Other programs affected include BTS' quarterly research series (ATPI), an experimental measure currently under development. The ATPI uses O&D Survey data and is dependent upon accurate data for all markets.

The Department's inability to measure True O&D according to the industry standards using One-way Trips hinders its ability to accurately measure nationwide air travel demand. Nationwide measures of air travel demand, airport improvements financed by PFC revenue, and improved airport security financed by the September 11th Security fees all depend on the Department's ability to identify One-way Trips. However, the Department's T-100/T-100(f) statistics count enplanements, while the O&D Survey statistics count Directional Passengers. Consequently, the government is without any method of properly forecasting tax revenue and without means to monitor the effects of tax policy.

F. Need for Regulatory Action

The Department is obligated to collect and disseminate information about civil aeronautics including, at a minimum, information on (1) the origin and destination of passengers in interstate air transportation, and (2) the number of passengers traveling by air between any two points in interstate air transportation (49 U.S.C. 329 (b)). In addition, the Department allocates airport improvement funds, provides essential air service subsidies and allocates funds to the air traffic control system. The requirement that the Department judge the need for, and consequences of, a regulation based on accurate statistical information presupposes that sound economic information exists.

The Department has a unique role in collecting transportation industry information. The need for a statutory mandate to collect traffic statistics is underscored by the extensive differences between the various airline business models and the level of technical sophistication that make the task of gathering comprehensive industry-wide data on air transportation a very formidable task for private industry or an industry trade group to undertake. The only other government entities in a position to gather traffic statistics are the nation's airports. Airports are operated by a variety of State, Municipal, County and Regional authorities. Collectively, they do not have the resources to process statistics on all of the passengers flowing through them on a daily basis, and it would be cost prohibitive for each of the major airports to develop parallel statistical systems. It would be a burden on the Air Carriers to require reporting to more than four hundred airports, and a burden on the airports to reassemble the data into a nationwide view of passenger air travel. Although third-party providers offer “enhanced” aviation data, the original sources of third-party provider data remain the T-100/T-100(f) and O&D Survey. The underlying need for traffic information cannot be satisfied anywhere else because there are no other sources of comprehensive traffic data available in the aviation industry. We therefore conclude that the changes proposed in this NPRM are required to provide accurate statistical information.

Respondents to the Department's ANPRM overwhelmingly agreed that the O&D Survey and T-100/T-100(f) segment data are essential. Most named the T-100/T-100(f) and the O&D Survey as the basis for all analytical work done in their organizations. Those that have access to other sources of data reported that they generally crosschecked those sources with information from either the T-100/T-100(f) or the O&D Survey. The Department's traffic data provides the press and consumer groups with the ability to monitor prices and advise the public about low price alternatives to high fares, which fosters a more competitive industry that benefits all consumers. The traffic data and the press and consumer group analysis of the data strengthen American companies by allowing companies to negotiate with airlines on fares. The traffic data benefits consumers by providing new entrant Air Carriers with the ability to demonstrate the strength of their business plan to investors.

The O&D Survey, however, was singled out most often in responses to the ANPRM as the data source most in need of improvement. The abundance of complaints about the deficiencies that exist in the O&D Survey has caused the public and the aviation industry to be cautious about any conclusions that can be drawn from this data, yet a wide range of stakeholders use it because it is the only available source of economic information that describes key aspects of scheduled air passenger transportation. Data inaccuracies have doubtlessly led to sub-optimal decisions by stakeholders that are as impossible to quantify as they are essential to correct. We therefore conclude that the changes proposed in this NPRM are made necessary by compelling need to improve the safety and economic well being of the American people.

Furthermore, OMB has published guidelines for ensuring that Federal agencies establish practices for ensuring and maximizing the quality, objectivity, utility and integrity of information disseminated by Federal agencies. Disseminated information must be accurate, clear, complete, and presented in an unbiased manner. Where appropriate, data should have full, accurate, transparent documentation and error sources affecting data quality should be identified and disclosed to users. The IG has declared that the Department's O&D Survey does not meet the Department's standard of acceptability of 95 percent accuracy. Since the O&D Survey and T-100/T-100(f) remain the key measure of the economics of the passenger air travel industry, the Department is under obligation to provide the most accurate statistical information that it can reasonably provide. The 1998 OIG report, the 1998 ANPRM, and subsequent outreach activities support the necessity of aviation data modernization. The IG found that to compensate for the unreliable O&D data, Department aviation analysts often requested Air Carriers to provide supplemental data, but they sometimes simply used their experience to apply adjustment factors to the unreliable data. Lack of consistent data collection over time decreases the utility of that data, while every request for supplemental information increases the Air Carriers' and the Department's costs. We therefore conclude that the changes proposed in this NPRM are necessary to implement Congress' intent for the law.

Because the Executive Branch and Congress utilize this data to form and implement public policies to foster a safe, healthy, efficient, and competitive air transportation system that contributes to aviation safety, national security, and the U.S. economy, agency investment in aviation information is critical. The private markets and other government and quasi-governmental agencies agree that this information is also critical for their needs, but private markets are unable to provide adequate statistical information to address this need. The unreliability of the data undermines the Department's ability to perform its statutory mandate to disseminate information that enables the transportation system to adapt to the present and future needs of commerce and to ensure that public policy remains consistent with changing commercial reality.

G. Development of the Record in This Rulemaking

The Department received 48 comments in Docket OST-1998-4043 in response to its ANPRM (July 15, 1998, 63 FR 28128) from Air Carriers, Foreign Air Carriers, airports, industry consultants, trade associations, and unions. Typical of the responses was that of American Airlines, which, as both a supplier and a user of data, expressed full support of the Department's effort to simplify the data submissions and ensure the accuracy and integrity of the data disseminated to the public. The Regional Airline Association pointed out that it had long advocated modernizing the data. Delta Air Lines supported the initiative so long as it did not require the incursion of unreasonable computer programming costs. The Air Line Pilots Association and the Association of Flight Attendants favored any change that would improve data quality and integrity over the current data.

Comments received about the O&D Survey under the ANPRM indicate that there is significant concern about the data. Even while emphasizing the importance of having access to the Department's traffic data statistics, the respondents stressed that the O&D Survey has serious weaknesses. Respondents repeatedly mentioned that the data elements collected were insufficient to meet the data needs of the public and the aviation industry. There was consensus that the reporting exemptions granted to some Carriers significantly affected the reliability and completeness of the data. There was near universal agreement that the data collected by the Department suffer from a lack of both quality and consistency. Specific comments point to the O&D Survey's outdated design, which affects the quality and accuracy of data gathered. This is amply demonstrated by the list of improvements that were put forth in the ANPRM. The suggested modifications to make the O&D Survey more reliable include:

• Change the source of data;
• Decrease the data reporting exemptions;
• Improve data validation;
• Improve definitions of data elements to enhance uniformity;
• Improve enforcement of timely receipt of data to guarantee timely release of data;
• Expand the number of elements collected to increase the usefulness in measuring the industry;
• Increase the accuracy of the data to make it more reliable; and
• Decrease the complexity of the form of the published data to make it more useful for decision making.

Stakeholders agree that the collection, processing, and dissemination of aviation data, particularly through the O&D Survey and T-100/T-100(f), are critical to the continued function and well being of the U.S. airline industry. There was general affirmation that the suggestions the Department proposed in the ANPRM were acceptable. Furthermore, Executive Order 12866 obligates the Department to collect, process, and disseminate accurate, timely, and relevant aviation data. The Department's data is insufficient to accurately determine a consistent measure of passenger travel using its same general direction of travel passenger counting methodology. Therefore, it is unable to fulfill its mandate to provide the most relevant aviation data within the current reporting requirements.

The air travel industry has grown rapidly since deregulation. Deregulated markets, code-share and other cooperative marketing agreements, new airline business models, and the adoption of the hub-and-spoke model and the rolling hub variation of that model have changed the fundamental economics of the airline industry. These changes have left the Department attempting to measure an aviation economy that is not the economy that the existing data were designed to measure. As such, 14 CFR Part 241, Section 19-7 (“Passenger origin-destination survey”) has outlived the economic model for which it was designed. Despite some adjustments (specifically, Docket No. OST-1996-1049, RIN 2105-AC34, 62 FR 6715; Docket No. OST-1998-4043, RIN 2139-AA08, 67 FR 49217), these metrics have not kept pace with changes in the industry, nor do they measure essential features of aviation economics as we know them today. Therefore, the Department is issuing this NPRM.

H. Scope of This Rulemaking

The purpose of this rulemaking is to (1) reduce the reporting burden on the Participating Carriers, (2) make the O&D Survey more relevant and useful, (3) reduce the time it takes to disseminate the information and (4) achieve maximum congruence between the O&D Survey and the T-100/T-100(f). In so doing, the rulemaking will aid industry and government users by collecting the most accurate and consistently obtainable economic information about the purchase of air travel on scheduled Carriers to or from, or within, the U.S. This rulemaking will address the identification of the responsible reporting entity, the identification of the data elements required to measure economic activity in the scheduled passenger air transportation industry, and the identification of exemptions that shall be allowed in the reporting process.

The Department seeks to achieve these goals by making the O&D Survey more relevant and useful to all stakeholders. Specific concerns associated with the current O&D Survey reporting requirements include (1) minimizing the number of reporting exemptions, (2) increasing the level of detail, (3) increasing the quantity and quality of information collected, (4) eliminating the need for data providers to resort to manual data collection, thereby reducing reporting costs, (5) establishing more uniform reporting by updating guidelines and instructions to the Carriers, (6) achieving maximum congruence between the O&D Survey and the T-100/T-100(f), and (7) updating the means of submission to enhance the timeliness of data release.

I. O&D Survey Redesign

The Department believes that an accurate O&D Survey based on Revenue Passenger tickets is now both desirable and possible in light of recent changes in airline information technology.

1. Summary of the Proposed O&D Survey

a. Who Shall Report

The Department proposes that all U.S. Air Carriers, and Foreign Air Carriers reporting data under antitrust immunity granted under 49 U.S.C. 41308, that are operating at least one aircraft with 15 or more seats and issuing tickets for travel on scheduled interstate passenger services to or from, or within, the U.S. participate in the O&D Survey. By this change, the Department proposes to abandon the concept of first Participating Carrier reporting a portion of Ticketed Itineraries in favor of the Issuing Carrier reporting all eligible Ticketed Itineraries. In light of substantial changes in airline ticketing and revenue accounting practices, this alternative is the most efficient and cost effective, allowing for the broadest possible data availability with a minimum of ongoing reporting effort.

b. Data To Be Collected

The Department believes that a fundamental restructuring of the data collected under the O&D Survey is necessary for the Department to fulfill its Congressional mandate to ensure a healthy, safe, efficient, accessible, and competitive transportation system that meets our vital national interests and enhances the quality of life of the American people. The Department acknowledges that this mandate includes meeting the needs of the aviation community that relies on this data, and we have endeavored to incorporate as many of its suggestions as possible in this proposal. The Department recognizes its obligation to measure passenger travel utilizing techniques that Congress, the industry, and the public recognize as valid, current, and reasonable industry measurements. In order to do this, the Department proposes to collect information about the issuance of the Ticketed Itinerary and to collect additional information about the travel described in the itinerary. With these changes, the Department proposes to abandon the concept of Directional Passenger in favor of One-way Trips to define True O&D.

The Department proposes to expand the scope of data that, currently, results in an insufficient volume of data to meet basic tests of validity and reliability. Therefore, the Department is abandoning the reliance on a 10 percent sample and is proposing 100 percent reporting of eligible Ticketed Itineraries. The Department intends to eliminate the limitations imposed on the scope of data that resulted in an overabundance of exceptions that compromised data quality. Therefore, the Department is removing the various exceptions for reporting long itineraries and non-standard itineraries and eliminating alternative data sample collection techniques for travel in major markets.

The Department proposes to expand the scope of data in order to gather data elements required to understand and disseminate useful information about passenger travel and thereby proposes to eliminate the bundling of ticket taxes and fees with the ticketed fare.

The current O&D Survey includes the following data elements: (1) Point of origin, (2) Carrier on each flight-coupon stage, (3) fare-basis code for each flight-coupon stage, (4) points of stopover or connection (interline and intraline), (5) point of destination, (6) number of passengers, and (7) total dollar value of ticket. The proposed revision of the O&D Survey includes additional traffic elements that occur for each Flight-Stage and sale elements that occur only once for an individual itinerary.

c. Proposed Traffic Elements

1. Flight-Stage Sequence Number. A two-character ordinal sequence number beginning with 01 that the Participating Carriers will assign to each Flight-Stage of a Ticketed Itinerary.

2. Airport Codes. a. Flight-Stage Origin Airport. The airport's IATA location identifier from which a Flight-Stage departs. The Department proposes to accept a city code in lieu of airport code only when the Flight-Stage flight number is OPEN, the itinerary uses a City Code instead of an airport code, and the scheduled Carrier serves multiple airports within the city making the origin airport unknowable.

b. Flight-Stage Destination Airport. The airport's IATA location identifier at which a Flight-Stage arrives. The Department proposes to accept a city code in lieu of airport code only when the Flight-Stage flight number is OPEN, the itinerary uses a City Code instead of an airport code, and the scheduled Carrier serves multiple airports within the city making the destination airport unknowable.

3. Carrier Codes. a. Operating Carrier. The IATA issued Airline Designator code of the U.S. Air Carrier or Foreign Air Carrier operating the equipment used on the Flight-Stage.

b. Marketing Carrier. The IATA issued Airline Designator code of the U.S. Air Carrier or Foreign Air Carrier marketing the Flight-Stage.

4. Scheduled Flight Date. The date on which the Flight-Stage is scheduled to depart.

5. Scheduled Departure Time. The scheduled local flight departure time of the Flight-Stage.

6. Master Flight Number. The Airline Designator code and flight number under which the flight inventory is managed.

7. Scheduled Arrival Date. The date on which the Flight-Stage is scheduled to arrive.

8. Scheduled Arrival Time. The scheduled local arrival time of the Flight-Stage.

9. Fare Basis Code/Ticket Designator. The alphanumeric code identifying the fare by class, qualification, and restriction associated with the Flight-Stage.

10. Ticketing Class of Service. A one-character code indicating the service cabin within the aircraft in which the passenger is scheduled to be seated under the fare rules stated for each Flight-Stage of the Ticketed Itinerary.

d. Proposed Sale Elements

1. Issuing Carrier Identifier. The Issuing Carrier's assigned IATA recognized airline numeric code.

2. Ticketed Itinerary Identifier. The alphanumeric identifier for the Ticketed Itinerary. This identifier identifies a unique itinerary for each Issuing Carrier Identifier and Date of Issue.

3. Date of Issue. The local date on which the Ticketed Itinerary was issued.

4. Fare Amount. The Fare Amount is the monetary amount the Issuing Carrier receives from the ticket purchaser on behalf of all the U.S. Air Carriers or Foreign Air Carriers included in the itinerary. The Fare Amount includes the Carrier-imposed fees and surcharges, such as fuel surcharges, for the carriage of a passenger and allowable free baggage on the passenger's complete itinerary, denominated in U.S. dollars, and accurate to two decimal places, rounded. The Fare Amount excludes taxes and fees imposed by Federal, state, local and foreign governments and excess baggage fees.

5. Government Taxes and Fees. a. Government Imposed Tax/Fee Identifier. The government tax or fee identifier. The Department's codes will be listed in the Passenger Origin-Destination Survey Directives issued by the Department.

b. Government Imposed Taxes/Fee Amount. This field will contain the value of the tax or fee specified by the identifier that precedes it, denominated in U.S. dollars and accurate to two decimal places, rounded.

6. Ticketing Entity Outlet Type. The identifying code of the distribution channel through which the Ticketed Itinerary was issued. The Department's codes will be listed in the Passenger Origin-Destination Survey Directives issued by the Department.

7. Customer Loyalty Program Identifier. The program identification code assigned to the airline customer loyalty program or alliance customer loyalty program under which the passenger accrues benefits.

8. Customer Loyalty Program Award Ticket Indicator. The one-character identifying code to indicate that customer loyalty program credits were expended in obtaining the Ticketed Itinerary.

9. Number of Passengers. The numeric value representing the number of passengers traveling on the Ticketed Itinerary. If multiple passengers have flown on a ticketed itinerary, we are considering requiring carriers to report separate records, with separate fares, for any groups of passengers on the itinerary that have flown under differing fare basis codes or under special discount fares. For example, if lower fares are paid for children within a tour group, the children's fares should be reported in a separate data record with a separate fare. When the projected number of passengers on a group ticket differs from the actual number, we are considering requiring carriers to report the actual number of passengers who flew on the group ticket as of the reporting event. BTS believes that these disaggregations are necessary to calculate its air travel price index. We seek comment on carrier practices and handling of group tickets and on the feasibility of the methodology we are considering.

10. Itinerary Copy Date. The date that the Participating Carrier copied the Ticketed Itinerary data for submission to the Department.

2. Discussion of the Proposed O&D Survey

a. Traffic Elements

In its comments to the Department's ANPRM, the Regional Airline Association (Docket OST-1998-4043-11) stated that the measure of passenger traffic used in the O&D Survey fails to satisfy the industry's need for timely and relevant information. Unisys Corporation (Docket OST-1998-4043-22) and Delta Air Lines (Docket OST-1998-4043-21) stated that the O&D Survey should adopt the True O&D concept. The Port of Portland (Docket OST-1998-4043-19) urged the recognition of multi-carrier O&Ds. In requesting that the Department begin using “relevant information,” “True O&D,” and “multi-carrier O&D” to measure passenger traffic, these respondents made clear that, for the aviation industry, the Directional Passenger is no longer an acceptable measure of True O&D. The Department agrees with the Regional Airline Association that, if we are to provide relevant information about the scheduled air transportation industry, we must change the basic calculation of the True O&D used in the O&D Survey to the calculation of One-way Trip commonly used in the air travel industry.

Scheduled Air Carriers in the U.S. use a variety of methodologies to construct One-way Trips in order to comply with the provisions of collecting September 11th Security Fees. The most widely accepted is a methodology based on “time in hub.” Here, the number of hours spent in an airport is the gauge by which it is determined whether the passenger (1) intended to continue the trip by changing planes, or (2) intended to remain in that city for other purposes. It is sometimes known as “the four hour rule” methodology because four hours is the most common maximum domestic connection time allowed with this method. In this methodology, certain other decision criteria are applied to supplement the time in hub determination, such as special rules for itineraries in which there are no stops that exceed the time allowance, itineraries with “void” and “OPEN” coupons, and itineraries that backtrack over the same set of airports.

The Department proposes to define a One-way Trip in terms of time spent in transit, subject to certain other rules. All other methodologies that are in use at Carriers require proprietary knowledge or were uniquely adapted to the needs of a particular Carrier, and would not apply industry-wide to all Carriers. These characteristics make the other methodologies unsuitable for use by the Department on a universal basis. The Department seeks comments from the industry and the public regarding the optimal method for constructing a One-way Trip. We will consider all the suggestions for appropriate determination of a One-way Trip, and establish a consensus of the guidelines provided by the industry to use in processing data in the O&D Survey for dissemination. We propose to require the following data elements for each segment of the Ticketed Itinerary as input for the One-way Trip determination: (1) Flight-Stage Sequence Number, (2) Airport Codes, (3) Carrier Codes, (4) Scheduled Flight Date, (5) Master Flight Number, (6) Scheduled Departure Time, (7) Scheduled Arrival Date, (8) Scheduled Arrival Time, (9) Fare Basis Code/Ticket Designator, and (10) Ticketing Class of Service.

1. Flight-Stage Sequence Number. Every Flight-Stage of an itinerary must have a sequence number assigned to it by the Issuing Carrier. Should problems arise, a positive identifier, assigned by the provider of the data, will help facilitate communication and resolution. Flight-Stage Sequence Number will begin each itinerary with Flight-Stage 01 and continue with sequential Flight-Stages. Surface Flight Coupon Stages (known within the industry as surface segments, including those provided by designated surface carriers such as railroads) that are included in the itinerary will be included in the numbering sequence. Voids (also known as arrival unknown segments, or ARNK segments) and OPEN segments are to be included in the numbering sequence.

2. Airport Code. Airport code for both Flight-Stage Origin Airport and Flight-Stage Destination Airport will be identified by the IATA location identifier that uniquely identifies that airport. American Airlines (Docket OST-1998-4043-5) and others commented that the presence of City Codes in the itinerary in lieu of airport codes resulted in data inconsistency. In the current O&D Survey, Participating Carriers from time to time had to attempt to decipher the itinerary using the pricing area of the ticket. The Department believes that our proposed change, which designates the Issuing Carrier as the Participating Carrier, will eliminate the problem caused by manual examination of the pricing area. However, the Department recognizes that when a Carrier sells an itinerary known as an “OPEN” itinerary, where (1) the itinerary is purchased but not booked, (2) the purchased itinerary includes a City Code instead of an airport code, and (3) the scheduled Carrier provides service to multiple airports at that city, then the airport code is unknowable. In this case, the Air Carrier must issue a ticket where the appropriate value is a City Code and the Department proposes to accept in the O&D Survey the reporting of City Codes in the itinerary only under this circumstance.

3. Carrier Code. Where once Carrier Code would have been described simply as the Airline Designator of the U.S. Air Carrier or Foreign Air Carrier that transported the passenger, the onset of code-sharing has introduced multiple Carriers into the ticketing process. The Marketing Carrier Code is the Carrier identifier that the passenger sees when examining the Ticketed Itinerary. The Operating Carrier is the Carrier that operates the aircraft that transports the passenger. Marketing Carrier and Operating Carrier will be identified by the IATA Airline Designator assigned to them. If the Carrier has no IATA Airline Designator code, then the Department will assign a reporting code. When a Carrier markets surface transportation as an extension of its air transportation service, and the transportation is (1) provided by a common carrier that is not an Air Carrier or Foreign Air Carrier, and (2) described on the Ticketed Itinerary and included in the total fare, then the surface carrier's IATA Airline Designator will serve as the Operating Carrier and the Carrier's IATA Airline Designator will serve as the Marketing Carrier.

4. Scheduled Flight Date. The Department's ability to determine One-way Trips from the O&D Survey information is crippled by a lack of information about Scheduled Flight Date. The lack of information about Scheduled Flight Date makes it impossible to know which passengers pass through a location on their itinerary to stay only long enough to change planes, and which passengers remain multiple days at a location.

In its comments, Data Base Products, Inc. (Docket OST-1998-4043-36) cited another inaccuracy, mentioning that the O&D Survey passengers are counted in the quarter in which the first departure took place regardless of the flight date scheduled in the itinerary. It pointed out that this inaccuracy is most noticeable in the transition from fourth quarter to first quarter where all trips are reported in the fourth quarter despite a large number of people departing in December who are ticketed to return in January. The scheduled air transportation industry does not always fluctuate in orderly monthly cyclic patterns. Holidays such as Thanksgiving and Easter have a great effect on air travel patterns and thereby require daily data.

Monthly data are problematic in other ways. From time to time, including times of emergency such as September 11th, the Department has found it necessary to request flight data at the weekly or daily level. Complying with these ad hoc data requests imposes a burden on Air Carriers. By routinely collecting data by flight date instead of by flight month, the Department will be able to avoid the need for special reporting requests by flight date. The ability to analyze air travel by day-of-week and in seven day moving averages will enable the Department to facilitate more robust economic measurement and analysis and be prepared to analyze the effects on air transportation when significant economic, weather and security related shocks to the nation occur. Because the determination of One-way Trips is critical to the Department's assessment of the air transportation industry, the Department proposes to collect information by Scheduled Flight Date.

5. Scheduled Departure Time. The Department's ability to determine One-way Trips from the O&D Survey information is also crippled by a lack of information about Scheduled Departure Time. The lack of information about Scheduled Departure Time makes it impossible to know which passengers pass through a location on their itinerary to stay only long enough to change planes, and which passengers remain for an extended period at a location.

Knowledge of the scheduled time of departure helps the Department understand the economics of the air travel industry. The FAA oversees the development of the nation's air travel infrastructure, and knowledge of Scheduled Departure Time allows it to calculate the costs and benefits of safety regulations and infrastructure improvements. Similarly, departure time will assist the TSA in meeting the needs of airports and Air Carriers with the appropriate staff levels for airport security. Flight-Coupon Stages where the travel plans are OPEN will be assigned an early morning departure time to be determined later, and the results of that determination will be published in the Passenger Origin-Destination Survey Directives issued by the Department.

6. Master Flight Number. Master Flight Number shall consist of the two-character Airline Designator of the Carrier that manages the inventory and the flight number under which that Carrier manages the flight. In flights that are not involved in a code-share and not involved in starburst or funnel flight operations, the Master Flight Number will be the same as the Marketing Flight Number. When code-shares, funnel flights and starburst flights are involved, this data element will be used to identify the Airline Designator and true flight number under which the flight inventory is controlled. The Department proposes to collect this data element to fill in the gap between the data the industry uses to track flights and the data the Department collects.

The term “code-share” is not sufficiently precise to describe what has become two distinct concepts. For purposes of this rulemaking, the term Alliance Code-Share will be used to describe the code-share relationship wherein each Carrier keeps its identity and livery distinct from one another and wherein each Carrier has the opportunity to market the other's flights. The term Franchise Code-Share will be used to describe the code-share relationship wherein the Franchise Code-Share Partner never appears as the Marketing Carrier and generally, although not necessarily, paints its aircraft in the livery of the Mainline Partner.

At the inception of code-sharing, the scheduled air passenger industry coined the term Marketing Carrier to distinguish it from the Operating Air Carrier that transported the passenger. According to the ATPCO TCN Ticket Exchange Service Specifications Guide instructions for populating the data element “Coupon/Segment Marketing Carrier” (glossary reference MCAR), the Marketing Carrier is:

The carrier that appears as the Carrier for a segment on the ticket. In a code-sharing arrangement, if a CRS knows the Servicing Carrier (CARR) and the Marketing Carrier (MCAR) both elements CARR and MCAR should be populated. If the CRS only knows the Marketing Carrier (MCAR), Marketing Carrier should be populated and Servicing Carrier should be blank.

According to the ATPCO TCN Ticket Exchange Service Specifications Guide instructions for populating data element “Coupon/Segment Carrier Code” (glossary reference CARR), the Carrier is:

The Carrier that carried the passenger. A CRS will populate this element with the same code as the Marketing Carrier (MCAR) unless the CRS knows of a code-sharing arrangement. If the CRS knows of a code-sharing arrangement, the CRS will code the Carrier that appears on the ticket as the Marketing Carrier (MCAR) and the Carrier that carries the passenger as the Carrier Code (CARR).

The Department, recognizing the importance of keeping track of code-share relationships on Ticketed Itineraries, amended the O&D Survey to provide for code-share ticketing practices. The Department defined the term “Ticketed air carrier”, which functions as the equivalent of the industry term Marketing Carrier. The definition of Ticketed Air Carrier in 14 CFR Part 241 Section 19-7 Appendix A, X. Glossary of Terms is:

Under a code-share arrangement, the air carrier whose two-character air carrier code is used for a flight segment, whether or not it actually operates the flight segment.

However, the Department diverged from standard industry practice when we defined Operating Air Carrier in a way that is slightly different than the industry term Coupon/Segment Carrier Code. Operating Air Carrier 14 CFR Part 241 Section 19-7 Appendix A, X. Glossary of Terms is:

Under a code-share arrangement, the air carrier whose aircraft and flight crew are used to perform a flight segment.

In an Alliance Code-Share, the industry's definition of Marketing Carrier is the equivalent of the Department's Ticketed Air Carrier, and the industry's definition of Coupon/Segment Carrier is the equivalent of the Department's Operating Air Carrier. However, in a Franchise Code-Share, the industry data is populated as if the relationship is a wet-lease and, therefore, the Airline Designator of the Mainline Partner serves as both the Marketing Carrier and the Coupon/Segment Carrier. Although the Department rules require the Issuing Carrier (or Issuing Carrier's agent) to notify the passenger of the identity of each Operating Air Carrier in the routing, standard industry practice does not list the Franchise Code-Share Partner's Airline Designator on the Ticketed Itinerary. Nevertheless, the O&D Survey rules require the Participating Carrier to report the Airline Designator of the Franchise Code-Share Partner Carrier as the Operating Air Carrier, and report the Airline Designator of the Mainline Partner as the Marketing Carrier.

The difference in the treatment of data between the industry and the Department's O&D Survey is most clear when examining an itinerary that includes both an Alliance Code Share and a Franchise Code-Share. For example, if Lufthansa German Airlines (Lufthansa) had authority to sell a code-share itinerary from Frankfurt (FRA) to Dulles (IAD) to Norfolk (ORF), and the IAD to ORF portion is on an aircraft operated by Mesa Airlines (Mesa), then the O&D Survey submission would show two flights. The FRA to IAD portion would be reported as Ticketing Air Carrier of Lufthansa and Operating Air Carrier of Lufthansa. The IAD to ORF portion of the travel would be reported as Ticketing Air Carrier of Lufthansa and the Operating Air Carrier of Mesa. The Department does not know the identity of the Mainline Partner Air Carrier. Logically, in this case, a user would assume Mesa is operating as United Express but there is nothing to preclude Mesa from flying IAD to ORF as US Airways Express, so such assumptions are not to be relied on. The Department's data is used for time series analysis over many years and no user of the data can logically deduce an Air Carrier's livery and operations over many years of service.

The Department has a statutory responsibility to monitor airline code-share relationships. As regional Carriers have increasingly taken multiple Mainline Partner Carriers into code-share arrangements, Franchise Code-Shares have become increasingly difficult for the Department to monitor. When an Air Carrier takes on a Franchise Code-Share relationship with two Mainline Partners that, in turn, have Alliance code-share relationships with each other, the need for a new data element in the O&D Survey is clear. When a Carrier operates as a Franchise Code-Share Partner for both US Airways and United Air Lines (United), the O&D Survey data records cannot distinguish between (1) flying in the livery of United, ticketed as a US Airways flight and (2) flying in the livery of US Airways, ticketed as a US Airways flight. In situation (1), the identity of the Mainline Partner (United, in this case) is lost. In situation (2), the identity of the Mainline Partner (US Airways, in this case) is not lost, but there is no way for the user of the data to know that. Since the user is provided no ability to distinguish between a record reported in situation (1) and a record reported in situation (2), the value of the data in assessing code-share travel partnerships is greatly diminished.

To further illustrate how Carriers would report the Marketing Carrier, Operating Carrier, and Master Flight Number data elements under this proposed system, consider the following hypothetical examples of itineraries involving a single US Airways Express flight operated by Mesa. Under this scenario, US Airways contracts with Mesa (IATA Airline Designator YV) to operate regional jet service between Charlotte (CLT) and Charleston, SC (CHS) on a fee per departure basis. Mesa operates the aircraft but the aircraft is painted in US Airways' livery. US Airways is wholly responsible for managing the inventory on the flight and bears all of the revenue risk associated with the flight. US Airways markets this flight to its customers as US Airways Express flight 2808. Mesa does not market this flight to the public under its own designator code and has no responsibility for managing the inventory. US Airways' alliance partners United and Lufthansa market US Airways Express flight 2808 as United 7808 and Lufthansa 8808, respectively. Although United and Lufthansa sell seats on US Airways flight 2808 under their respective designators, neither Carrier has any responsibility for managing the inventory on this flight. The following itinerary examples illustrate how the proposed system would work in practice.

Itinerary 1: Lufthansa marketed Munich-Charleston One-way Trip with connection over Charlotte to US Airways Express flight 2808. Under this scenario, the passenger buys a ticket from Munich to Charlotte on LH100, a Lufthansa operated flight. In Charlotte, the passenger will connect to Charleston on LH8808, which is the Lufthansa marketing flight number for US Airways Express flight US2808 operated by Mesa. For the LH8808 Flight-Stage, the Participating Carrier would populate the Marketing Carrier, Operating Carrier, and Master Flight Number data elements as follows:

Marketing Carrier: LH.

Operating Carrier: YV.

Master Flight Number: US2808.

Itinerary 2: United marketed Chicago-Charleston One-way Trip with connection over Charlotte to US Airways Express flight 2808. Under this scenario, the passenger buys a ticket from Chicago to Charlotte on UA200, a United operated flight. In Charlotte, the passenger will connect to Charleston on UA7808, which is the United marketing flight number for US Airways Express flight US2808 operated by Mesa. For the UA7808 Flight-Stage, the Participating Carrier would populate the Marketing Carrier, Operating Carrier, and Master Flight Number data elements as follows:

Marketing Carrier: UA.

Operating Carrier: YV.

Master Flight Number: US2808.

Itinerary 3: US Airways marketed Charlotte-Charleston One-way Trip, Non-stop on US Airways Express flight 2808. Under this scenario, the passenger buys a ticket from Charlotte to Charleston on US2808. For the US2808 Flight-Stage, the Participating Carrier would populate the Marketing Carrier, Operating Carrier, and Master Flight Number data elements as follows:

Marketing Carrier: US.

Operating Carrier: YV.

Master Flight Number: US2808.

In all three of the situations described above, if the US Airways flight from Charlotte to Charleston were to be operated by US Airways itself (i.e. with mainline equipment rather than by one of its regional affiliates) as hypothetical flight US Airways 518, the Operating Carrier field in all of the above examples would instead reflect “US.” The Master Flight Number field would reflect “US518.”

It is also important to know the Master Flight Number when Carriers use funnel flights and starburst flights to market their product to consumers. Correlations between the T-100/T-100(f) would be very difficult if the O&D Survey is only reported under the various flight numbers that are assigned in funnel flights and starburst flights. Knowing the Master Flight Number will provide the common element needed for accurate correlation.

The Department must require this data element to fulfill its mandate to protect consumers by monitoring code-share ticketing and other marketing practices. Therefore, the Department proposes to collect the Master Flight Number, which will consist of the Airline Designator and true flight number of the Mainline Partner that manages the inventory of the flight. The Department invites comment on this topic and on the efficacy and difficulty of populating this data element.

7. Scheduled Arrival Date. The Department's ability to determine One-way Trips is dependent on knowing when a scheduled flight arrives in an airport. Scheduled Arrival Time is meaningless without Scheduled Arrival Date.

8. Scheduled Arrival Time. The Department's ability to determine One-way Trips from the O&D Survey information is further crippled by a lack of information about Scheduled Arrival Time. The lack of information about Scheduled Arrival Time makes it impossible to know which passengers pass through a location on their itinerary to stay only long enough to change planes, and which passengers remain for an extended period at a location.

Flight-Coupon Stages where the travel plans are OPEN will be assigned an arrival time to be determined later and the results of that determination will be published in the Passenger Origin-Destination Survey Directives issued by the Department.

9. Fare Basis Code/Ticket Designator. The Department requires fare basis code and ticket designator to understand the restrictions placed on the purchase of travel and the economics of the air travel industry. Several respondents to the ANPRM requested that the Department collect information that will enable it to provide a classification of fares. The Fare Basis Code is the alphanumeric code identifying the fare by class, qualification, and restrictions associated with the travel segment. The Ticket Designator is the code indicating that the fare basis code is modified by rules associated with the ticket designator code. Ticket Designator is specified in the ATPCO TCN Ticket Exchange Service Specifications Guide instructions for populating data element “Coupon/Segment Fare Basis/Tkt Designator” (glossary reference FBTD) as the code that appears in the same field as the Fare Basis Code separated by an oblique “/”.

10. Ticketing Class of Service. In order to understand service demand and to understand the quality of services to communities, the Department proposes to continue the practice of collecting information about class of service, also known as cabin class. In response to the ANPRM, American Airlines (Docket OST-1998-4043-5) and others noted that non-standard reporting of class of service degrades the usefulness of the published data. The most expensive class of service, generally provided in the cabin located nearest the nose of the plane, is typically referred to as the first class cabin. The least expensive class of service (coach/economy/main) cabin is typically located in the aft-most section of the aircraft. Sometimes a Carrier will avoid offering a class of service marketed as first class, and choose to market the front cabin as business class instead. To further complicate matters, more than one Carrier markets the front cabin of its narrowbody aircraft flying on a domestic route and the front cabin of its widebody aircraft flying on an international route with the same “first class” designation. Today, certain Carriers offer “premium coach” seating and in the future, Carriers may offer an “ultra-premium” (i.e. more expensive than first class) cabin. We are unaware of an objective class of service definition maintained anywhere in the industry that distinguishes between these classes of service. Indeed, currently there is no objective class of service definition that would prohibit a Carrier providing only a single class of service from calling it first class, even if that single class of service was comparable to coach class at a Carrier that offers multiple classes of service.

The Department desires to change the class of service designations to make them as objective and as meaningful as possible. However, we believe the marketplace is the best arbiter of a Carrier's claim to offer first class service. We do not wish to codify a particular standard of service or seat pitch as the point that differentiates a first class accommodation from a business class accommodation. The Department seeks consistent class of service designations but there are no objectively defined designations in the industry. Therefore, the Department proposes to provide a framework in which each airline will assign a number to the service cabins in its fleet from the least expensive to the most expensive, such that the least expensive cabin (usually the aft-most cabin) is designated as “1” and each defined cabin class above cabin 1 (i.e. those that the Carrier markets at higher price points and that are generally physically located toward the front of the aircraft) will be designated with the next highest ordinal number. The number “2” will generally designate what has heretofore been described as premium coach. The number “3” will generally designate what has heretofore been described as business class or first class of a two cabin aircraft. The number “4” will generally designate what has been described as first class of a three cabin aircraft. The number “5” will designate ultra-premium first class. The Carriers would provide the Department with up to date definitions of its 5 class of service designations and would use their own internal class of service codes to classify their passengers. When a Carrier operates a fleet of aircraft with a class of service that is arguably similar to the class of service offered by competing Carriers, and if the Department believes a compelling public interest is served by re-designating the passengers as having been transported in a different class of service, the Department reserves the right to re-designate passengers on such an airline into a different class of service. The Department seeks comment from Carriers and the public on the efficacy of this proposal.

b. Sale Elements

1. Issuing Carrier Identifier. Every Carrier that issues Ticketed Itineraries must have a unique three-digit numeric identifier. The Issuing Carrier is responsible for the ticket stock on which the itinerary is issued. The Department proposes to identify the Issuing Carrier with the Carrier's assigned IATA three-digit code. This code also serves as the first three digits of the 13-digit ticket number on a standard agent ticket.

2. Ticketed Itinerary Identifier. Carriers assign a ticket number or Passenger Name Record (PNR) identifier to every Ticketed Itinerary that is unique when used in conjunction with an Issuing Carrier Identifier and the Date of Issue. This data element will contain the value of that identifier. The Department requires a unique identifier to facilitate communication with the Participating Carriers in the Department's effort to monitor the data and the Participating Carrier requires a unique identifier to facilitate communication with the Department when data must be corrected and resubmitted. The Ticketed Itinerary Identifier is necessary for effective resolution of problems.

3. Date of Issue. The Department proposes to require Date of Issue because it is part of the unique identifier of the Ticketed Itinerary. In the past, the Department has often had to require Air Carriers to provide supplemental information about travel because it lacked information about ticket sales dates. DOJ and DOC both require knowledge of the date of sale in the course of carrying out their mandates. The date the Ticketed Itinerary is issued is an important component of understanding the economics of the airline industry. Falling passenger counts or rising passenger counts have traditionally been the measure of the economic engine that travel provides to the economy. However, for some purposes, the rising and falling volume of daily ticket sales over time is a better measure of industry economics. Another key element of air transportation economics is the measurement of the number of days between ticket sale and first use of the Ticketed Itinerary. Known as the booking window, or advance purchase window, the increase or decrease of the booking window year over year is an important measure of consumer confidence. To understand the dynamics of rising and falling volume of itineraries sold and the size of the booking window, the Department must collect the Date of Issue.

4. Fare Amount. The Department's ability to measure fare information independent of taxes collected is vital to the understanding of aviation commerce. Carriers shall convert fares paid in currencies other than U.S. Dollars into U.S. Dollars using whatever currency conversion methods the Carrier customarily uses in its normal course of business. The current O&D Survey requires Participating Carriers to truncate the cents from the reported total amount. This practice artificially lowers the Department's estimate of total amount collected because an unknown number of cents have been dropped from millions of tickets. Rounding to the nearest cent will allow some imprecision to remain, but the Department believes that losing fractions of one half cent is an acceptable degree of imprecision. Fare amounts have customarily not been whole dollar amounts even when they do not require currency conversion to U.S. dollars. Therefore, the Department proposes to collect fare information independent of tax information, and further proposes to collect fare information accurate to two decimal places rounded.

5. Government Imposed Taxes/Fees. The ability to identify each and every tax, passenger facility charge, and fee that the consumer must pay is central to the Department's understanding of the economics of travel. Disaggregating taxes and government-imposed fees from the fare will enable the Department to more accurately monitor changes in airfares and separately monitor the changes in taxes and fees paid, both of which have substantial policy considerations.

On January 9, 2003, Captain Duane Woerth, President of the Air Line Pilots Association International, testified before the Senate Committee on Commerce, Science, and Transportation that airline taxes were choking the industry.^[6] He testified that, according to the Air Transport Association (ATA), taxes on a $100 domestic ticket could be as high as 44 percent of the amount collected. Without improvements to the O&D Survey, it is impossible to use Department data to precisely determine whether the testimony was based on an example of a typical ticket or an extreme case, and whether it is indicative of an industry-wide trend affecting a substantial number of passengers.

The Department proposes to adopt the industry's standard Government imposed tax/fee identifiers as documented in the ATPCO TCN Ticket Exchange Service Specifications Guide instructions for populating data element “Tax/Miscellaneous Fee Type” (glossary reference TMFT). Carriers shall convert amounts paid in currencies other than U.S. Dollars into U.S. Dollars using whatever currency conversion methods the Carrier customarily uses in its normal course of business. The Department proposes to require the reporting of taxes and fees collected by the Carriers on behalf of government entities and further proposes to collect tax and fee information accurate to two decimal places rounded.

6. Ticketing Entity Outlet Type. BLS (Docket OST-1998-4043-54), American Airlines (Docket OST-1998-4043-5), and Northwest Airlines (Docket OST-1998-4043-49), among others, specifically requested that the O&D Survey include a distribution channel component. The Department has conducted studies of airline marketing and distribution practices and how they affect the cost structure of Carriers as well as the associated impact on consumers. Knowledge of the distribution channel used to deliver the ticket to the passenger has become an important part of aviation analysis.

The Department has lacked the data to sufficiently examine such changes precisely at a time when they have become an important part of the Carrier's efforts to reduce costs. The Department proposes to collect an indicator that identifies the type of location responsible for issuing the Ticketed Itinerary. The Department seeks comment regarding the efficacy of using codes based on those already in use in the industry as listed in ATPCO's TCN Ticket Exchange Service Specifications Guide instructions for populating data element “Ticketing Entity Outlet Type” (glossary reference TIOT).

• A = Airline office
• B = Business corporate account
• C = Consolidator
• D = Direct dial in locations (Consumers, PC Users)
• E = End user access via third party (Internet, Minitel, etc)
• G = General sales office
• I = Internal CRS locations
• M = Multi-access
• N = Non-IATA agents
• P = Pending agents
• S = Self service machine
• T = IATA travel agent
• U = Unknown
• V = Vendor (car, hotel)
• W = Wholesaler or tour operator

The codes will be listed in the Passenger Origin-Destination Survey Directives issued by the Department.

7. Customer Loyalty Program Identifier. Some users of the O&D Survey data have requested a data element to record the program name when a passenger has declared a membership in a loyalty program. The need to monitor domestic and international alliances and the causes and consequences of share shift associated with the alliance have become critical in understanding industry trends and discerning their competitive impact. The Department proposes to collect the name of the program in which the passenger is earning credit. We are unaware of any industry standard loyalty program identifiers. The ATPCO TCN Ticket Exchange Service Specifications Guide instructions for populating data element “Coupon/Segment Frequent Flyer Reference” (glossary reference FFRF) indicate that the reference include the “Airline Designator of the airline that assigned the Frequent Flyer Number” which presupposes that loyalty programs belong to an airline rather than an alliance of airlines.

We propose to use the industry standard loyalty program identifiers if a consensus exists, otherwise, the Department will maintain and publish a list of loyalty programs and appropriate identifying codes for those programs. We are aware that not all ticket purchasers declare their membership in a loyalty program at the time the itinerary is ticketed. Passengers that identify themselves as members of a program after the Ticketed Itinerary has been submitted to the O&D Survey will remain unrecognized in the Department's statistics. The Department seeks comment from the industry and the public regarding the ability of the Carriers to reliably populate this element.

8. Customer Loyalty Program Award Ticket Indicator. The Department believes that, to carry out its mandate, it must know when a passenger has expended mileage points or award credits to obtain a Ticketed Itinerary. The Department proposes the values of “A” when the customer paid no fare at all, “P” when the customer pays partially with award credits, and “U” when the passenger paid the appropriate fare for passage, but used award credit to upgrade to a more exclusive class of service. The Department seeks comment from the industry and the public regarding the ability of the Carriers to reliably populate this element.

9. Number of Passengers. The majority of Ticketed Itineraries are issued to one passenger, but some Ticketed Itineraries describe the travel of multiple passengers traveling together on the same itinerary. The Department must collect the count of passengers included in the Ticketed Itinerary. Without knowledge of this value, the data from several of the other elements, particularly the Fare Amount, become invalid.

10. Itinerary Copy Date. Since Ticketed Itinerary databases are operational databases for the Carriers, and since operational systems are by their nature constantly updating data, and since the Department is requiring a copy of the Participating Carrier's Ticketed Itinerary data to be taken at a given point in time, it is important to have that point in time recorded. The copy date will also facilitate the correction of data. Participating Carriers wishing to replace previously submitted data can do so more easily if the Department can identify old and new copies of records using the copy date.

We explored the possibility of omitting this data element on the assumption that the Department could record the date that the data was received. However, this option would record the date of successful data transmission rather than record the date the Participating Carrier's operational data was copied. To best facilitate communication, the Date of Submission must be set by the Participating Carrier at the date the data is copied, not by the Department at the date the data is received. Knowledge of the Itinerary Copy Date will help alleviate questions and concerns about data quality.

c. Other Suggested Elements

Various members of the air transportation community have suggested the following as elements the Department should collect. The Department does not propose to collect these data elements, but we seek further comment advocating the inclusion of these suggested elements, and we will consider including any one or all of these elements in the list of mandatory elements collected under this rule.

1. Passenger Type. The airline industry has an established passenger type code that is used as an indicator of the characteristics of the passenger based on a pricing decision. ATPCO's TCN Ticket Exchange Service Specifications Guide instructions for populating data element “Passenger Type” (glossary reference PAST) describes this as a three-digit code indicating the type of passenger (e.g., ADT for adult fares, CHD for child fares, MIL for military fares and GOV for government fares.) Several Carriers and airports that responded to the ANPRM requested some kind of information about the type of passenger traveling on the Ticketed Itinerary. The Department would also benefit from having passenger data type in planning for air transportation needs of the future. From time to time, the Department is required to conduct reviews of government fares. For example, on at least one occasion, the Department has been asked to supply information on the number of children that fly on commercial Carriers.

The National Transportation Safety Board has recommended that BTS improve the quality of exposure data available for safety analysis (See National Transportation Safety Board, Transportation Safety Databases, Report No. SR-02-02, September 11, 2002, p. 38). Exposure data (i.e., the number of passengers exposed to the risk of an accident in any particular type of transportation) are essential for measuring the accident rate for different types of transportation and measuring the benefits of safety improvements. Aviation safety analysts are particularly interested in certain data that would be collected under the proposed rule on characteristics of airline passengers (e.g., whether the passengers are adults, children, or infants), so that they can estimate the likelihood that passengers would take an alternative mode of transportation if safety regulations increased the cost of flying. BTS believes that information about passenger type will help it calculate a more meaningful ATPI. The Department is considering collecting passenger type as a data element and, therefore, we seek comment on the availability of passenger information, the consistency with which it is populated in airline systems proposed as the source for O&D Survey data in this rulemaking, and the reliability of the Carriers maintaining a uniform understanding about what each value signifies.

2. Fare Basis Category. The Department currently collects class of service information and rudimentary fare classification information in a dual-use field called fare basis code. The current classification has seven possible values: C (Unrestricted Business Class), D (Restricted Business Class), F (Unrestricted First Class), G (Restricted First Class), X (Restricted Coach/Economy Class), and Y (Unrestricted Coach/Economy Class), plus U (Unknown). The dual-use codes indicate (1) the class of service (also known as cabin class) appropriate to the fare basis the passenger purchased and (2) whether or not the passenger's fare basis was issued under one or more restrictions, such as the fare's minimum advance purchase requirement or the fare's eligibility to be refunded. We continue to believe that Ticketing Class of Service is an important element to collect, and we have proposed collecting it as explained under section I.(2)(a)(10) of this document. In addition, we are considering collecting information about fare basis restrictions. We believe that policy makers and the aviation industry as a whole would benefit from information about the purpose for which the passenger is traveling.

Several Air Carriers requested fare categorization in their ANPRM comments. The most often mentioned classification was a business or leisure dichotomy classification. The Department believes that the business—leisure dichotomy is a useful but very subjective evaluation, which is very difficult to categorize in a standardized manner industry-wide, given the data currently available. Our understanding of the difficulties faced by the Air Transport Association in its attempt to build criteria for categorizing business and leisure fares based on existing data elements in Carrier reservations and accounting systems tends to verify that belief.

We believe that classification based on objective and verifiable criteria would provide a more useful classification methodology. The current classification has only a single aspect, “restricted” or “unrestricted.” This, though verifiable, is so broad that it provides very little understanding of passenger fares in the current aviation environment. We are, therefore, considering and requesting comment on, classifications based on a combination of three criteria (1) travel eligibility date, (2) purchase eligibility restrictions, and (3) refundability/exchangeability. We believe that knowledge of these three aspects of a fare would enable a comparison of fares across Carriers and provide useful “passenger type” data while relying on common information stored in carrier accounting and reservations systems.

The Department believes that categorization of fares would be extremely useful to the government and industry users alike, but we recognize that there are substantial difficulties in collecting, categorizing, and validating the data given current data in Carrier reservation and accounting systems. First, the Department would necessarily rely on Carriers' classification designations. The Department cannot independently edit or validate the Carriers' classifications beyond issuing guidelines, which would be as specific as possible, but would necessarily be fairly general in nature. Second, the complexity and diversity of fare basis codes is enormous. Some fare basis codes are designated for single markets. Some are designated for a group of markets. Some are designated for all markets, but carry restrictions that apply only in some markets. Third, the volume of fare basis codes on file for many Carriers is huge. It is not uncommon for an individual carrier to have thousands of fare basis codes and combinations of codes. The volume of fare basis codes in combination with their complexity and diversity make classification of fares a very challenging task. Fourth, fare basis codes do not have a universal meaning across all Carriers in the industry. Pricing structures are unique within each Carrier. A given set of fare basis codes reflects the pricing structure only within the context of the given Carrier.

One approach to a classification plan would be for each carrier to submit its list of fare hierarchies to the Department. The list or lists would include the fare basis codes and the attendant rules for these fare basis codes as expressed in terms of the Department's three classification criteria or some other set of classification criteria suggested by members of the industry. With an understanding of the fares included in each category across multiple Carriers, the Department could publish a map of fares that would serve as the industry fare basis category for purposes of classifying the value of fares across all carriers.

There appear to be two options for collecting this type of data, (1) retain the existing system of classification of “restricted” or “unrestricted”, or (2) use the fare basis codes as a means for establishing more accurate comparisons across carriers. Given the inconsistency of fare basis code application from Carrier to Carrier, some method of mapping by the Department would be required. Whenever possible, the Department prefers data elements that can be objectively collected and consistently validated industry-wide.

The Department seeks further comment on the utility and efficacy of collecting Fare Basis Category based on an aggregated fare basis classifications as well as any other data element that could prove useful to users of the O&D Survey in understanding the nature and purpose of passenger air travel in the U.S. Comments should address (1) the usefulness and efficacy of the continuation of collecting the current “restricted” or “unrestricted” fare designation only, and (2) the usefulness and efficacy of establishing a new system based on some form of mapping fare basis codes according to similar values assigned to different codes by various Carriers by periodically collecting and publishing comprehensive fare hierarchies from each Participating Carrier. We request that comments be as specific as possible in outlining any proposed methodologies and that they address issues involved in making industry-wide comparisons accurate and meaningful.

3. Commission Amount. This data element represents the amount paid by the Issuing Carrier to the travel agent for selling a Ticketed Itinerary on its behalf. The Department recognizes that, in general, the role of sales commissions paid to the travel agents on the issuance of a Ticketed Itinerary have diminished in the U.S. However, commission payments have not disappeared from the air travel industry. In light of this, the Department seeks comment regarding the efficacy of collecting this information.

4. Form of Payment Type. As shifts occur between payment by cash, credit card, or one of the new forms of Internet payment, collection of this data may provide insight into ticket purchasing behavior. The Department seeks comment on the efficacy of collecting this information.

5. Electronic Ticket Indicator. This element, used in conjunction with ticketing entity outlet type, could help isolate information about selling and distribution channels. The Department seeks comment on the collection of an indicator to determine information about electronic ticketing. The proposed values would be the ones used in ATPCO's TCN Ticket Exchange Service Specifications Guide instructions for populating data element “Electronic Ticket Indicator” (glossary reference ETKI).

6. Passenger Citizenship Nation. BLS requested citizenship information to determine whether a trip constitutes an export transaction or an import transaction. DOC's International Trade Administration (ITA) requested citizenship information to help in its mandate to facilitate trade and tourism. DOS, which negotiates air treaties with foreign governments, would benefit from citizenship data. The Department seeks comment regarding the efficacy of collecting statistical information about passenger citizenship.

7. Country Code and Area Code of the Passenger's Phone Number. US Airways, United Air Lines, Southwest Airlines, the Sabre Group, Northwest Airlines, Continental Airlines, and American Airlines all included in their ANPRM comments their desire that the Department obtain information about the passenger's point of origin. BLS needs citizenship information to determine whether a trip constitutes an export transaction or an import transaction. The passenger's phone number area code, in conjunction with passenger's phone number country code, is one indication of passenger point of origin. In light of the increasing use of cell phones and the increasing disassociation between the area in which a passenger resides and the geographical area of the cell phone's area code, the Department seeks comment regarding the efficacy and the cost/benefit proposition of collecting this information as an indication of passenger residence in general, and in light of announced DHS requirements.

8. Passenger Zip Code/Postal Code. Sabre, US Airways, American, Continental, Northwest, and Southwest commented in the ANPRM that they would like to have some measure of the passenger's place of origin. Carriers, such as US Airways and Northwest, identified this need as generic point of sale information. Academics, consultants and Carriers alike want to study point of origin demographics.

United, Airports Council International—North America, and airports that supplied ANPRM comments specifically requested passenger zip code as a point of sale identification to identify the geography of the area served by the airport. Several comments from airports declared that this element would be a vital component of their ability to serve their communities. The Department believes that this element is the best indicator of passenger point of origin, and, perhaps, the single most important data element needed for prudent infrastructure planning and investment. The Department's mandate to ensure that the transportation system is healthy, efficient, and competitive cannot be fully realized until we know where the users of the system reside. The Department's ability to study the region in which an airport's customers reside, or catchment area analysis, is not currently possible.

The passenger is not currently required to declare a Zip Code/Postal Code as a precondition of purchasing a Ticketed Itinerary from a Carrier, and, therefore, this data element is not available. DHS may seek specific individual identification data on airline passengers that would require the Carriers to collect and store passenger residential Zip Code, among other elements, for a system designed to use passenger information to increase homeland security. If it could be collected without impinging on individual privacy rights, Zip Code/Postal Code would make important point of origin information available for statistical purposes for the first time.

If the Carriers develop the capability to collect and store Zip Code/Postal Code, then the cost of collecting it for statistical purposes will not be significant. In light of the many benefits to the industry, the Department would consider collecting this data element. However, since it is not a data element that is routinely collected by the Carriers we are not proposing to collect this data element at this time. We seek comment regarding the continued interest in collecting this information for statistical purposes.

3. Reporting Requirements

a. Data Source Criteria

One of the most critical questions asked in the ANPRM was whether the Department should change its source of data for the O&D Survey. Heretofore, the Department has required the Operating Air Carrier to use a data stream created specifically for reporting the O&D Survey. The Department has three objectives for the data provided by Carriers. First, the data available to the Department must meet the OMB quality objectives of accuracy, reliability, completeness and non-bias to the extent that it is practical. Second, the source of data must be selected in a way that minimizes the burden of collection on the Participating Carriers. Third, the Department must minimize the effects of changes to itineraries over time, because changes that take place following the reporting event are invisible to the O&D Survey. All sources of data, including alternative data sources proposed in responses to the ANPRM, must be evaluated on these three criteria.

b. Discussion of Interactions Between the Carriers and Their Customers

The source of data is inextricably linked to the event that triggers the creation of that data. Each source of data suggested in the ANPRM comments represents data captured at a point in time where an interaction between the passenger and the Carrier, or one of its agents, takes place. Adopting a new source of data necessarily means that we accept the state of the data as it existed when that data source was created or introduce a procedure to report subsequent changes to the itinerary.

For an electronic ticket sold over the Internet at the Carrier's website, the creation of a reservation, the creation of the ticket, the financial payment transaction, and the recording of the itinerary by the revenue accounting system of the Issuing Carrier all occur simultaneously when the customer agrees to purchase the itinerary. However, for itineraries sold through other outlets or provided gratis by the Carrier, some of the events occur simultaneously and some occur serially. In a handwritten ticket, all of these events are separate and distinct. It is important to be aware of these distinctions because the Department must establish its procedures and data sources to be equally valid when collecting information about all passengers from all Carriers with the least amount of procedural or statistical bias.

c. Problems in the Current Source of Data

The Department created the current source of O&D data for the express purpose of collecting the O&D Survey. The problems that result in designating an Operating Air Carrier as the Participating Carrier have already been discussed. Since the Operating Air Carrier does not always know enough about the Ticketed Itinerary to report it correctly, unless it is also the Issuing Carrier or the Issuing Carrier provides the necessary information, the Department has been forced to code a large number of reporting exemptions in the current O&D Survey methodology that we now seek to eliminate.

The current CFR grants reporting exemptions for itineraries that are flown entirely on some Carriers. Every Participating Carrier transporting the passenger must examine the itinerary to determine whether it is the first Carrier in the itinerary that is listed by the Department as a Participating Carrier. The Air Carrier is exempted from reporting a Ticketed Itinerary if another designated Participating Carrier precedes it in the scheduled itinerary. A Ticketed Itinerary is, in effect, exempted from reporting when the code-share ticketing situation makes it appear as if the itinerary has already been reported when, in fact, the itinerary has not been reported. The current system also grants exemptions for reporting all of the travel on reportable Ticketed Itineraries if the Participating Carrier is unable to obtain information about the entire itinerary from the Issuing Carrier and is unable to obtain the information from looking at the passenger's documents. Roberts Roach and Associates, Inc. (Docket OST-1998-4043-4) summed up the frustration of most users of Department data when, in its comments to the ANPRM, it advocated that the Department allow no exceptions whatsoever.

Exemptions are not the only problems associated with the O&D Survey's source data. Under the current rule, the Department requires the full amount collected for the Ticketed Itinerary to be reported even when the full itinerary was not, which causes the reported portions of the itinerary to be overvalued. For example, conjuncted tickets consist of more than four Flight-Stages and require multiple ticket documents. If the first reporting Carrier is not the Issuing Carrier, and can view a partial list of airports but a full fare amount, the identified portion of the Ticketed Itinerary will be overvalued.

Equally troublesome, the Department requires the full itinerary reported, even if the full amount collected for the Ticketed Itinerary is not known. For example, when the Ticketed Itinerary is issued as a bulk ticket, the amount collected is either not shown or appears as zero amount collected. Usually, a bulk ticket is reported by the Issuing Carrier, in which case the fare amount would be known. However, in some circumstances, a passenger who possesses a bulk ticket may be diverted or transferred to another carrier. Under the current rule, should this situation occur, the Participating Carrier will not know the amount of fare collected and will report the amount collected as zero dollars.

The Department recognizes that designating an Operating Air Carrier as the Participating Carrier necessitates that the Department grant reporting exemptions for conditions that exist when the Operating Air Carrier does not and cannot know some of the data elements. Therefore, the Department believes that the currently designated reporting entity, the Operating Air Carrier, does not have sufficient information to reliably produce a source of data for the Department's O&D Survey.

d. Discussion of the Sources of Data Proposed by ANPRM Respondents

In the ANPRM, the Department solicited input on alternative data sources for the O&D Survey. The following data sources were proposed: (1) The computer reservation systems', or GDS', marketing information data tapes (MIDT) data triggered by the creation of a reservation, (2) Airlines Reporting Corporation's (ARC) sales tapes triggered by the sale of a ticket by a travel agency, (3) ATPCO's TCN records triggered by the creation of a ticket, (4) a new data stream from Carriers that issue electronic tickets triggered by the recording of the ticket in the Carrier's accounting system, and (5) a new data stream of passenger boardings triggered by the Operating Carrier's records from each flight segment.

1. MIDT. Metropolitan Washington Airports Authority and the Airports Council International—North America (Docket OST-1998-4043-68) suggested using the GDS systems' MIDT data as a source of data. The GDS MIDT records include customers' travel schedule information and obtaining it from these systems would impose a relatively small burden on industry. However, MIDT data represent only those bookings made through the reservations systems. Tickets purchased directly from the Carriers and through other outlets not connected to the MIDT would be excluded. This, in effect, would create an exemption for the reporting of tickets that were not created through the GDS distribution channel, and would deflate travel statistics. There is no reliable method of measuring the number of Ticketed Itineraries created through non-GDS distribution channels in order to gain a sense of the total number of Ticketed Itineraries issued. The reliability of the O&D Survey would suffer because the proportion of under-reported travel to actual travel would be unmeasurable. Even if the Department made an estimation of that proportion, the proportion of MIDT reservations as a percentage of the universe of tickets would fluctuate over time, which would invalidate the estimates.

As airlines encourage more bookings made directly with the Carrier, the number of tickets captured by MIDT is declining. Moreover, some Carriers' bookings are not represented in the MIDT data due to almost total reliance on direct sales. These situations would cause this source of data to under-report travel in an unmeasurable degree. Conversely, American Airlines (Docket OST-1998-4043-5) stated that many reservations are never ticketed. The IG estimated the number of unticketed reservations at 15 percent of CRS-based travel reservations.^[7] These unused reservations that inflate the passenger travel statistics would cause the O&D Survey to over-report travel. The proportion of this over-reported travel to actual travel would be as unmeasurable as the under-reported travel. It has been argued that the over-reporting of travel might balance out the under-reporting of travel, but the extent to which that would happen is unmeasurable, leaving the ratio of reservations to tickets sold in a constant state of statistical instability. In addition, the level of over-or under-reporting may disproportionately affect different types of markets (e.g., predominantly leisure versus predominantly business markets) further reducing the validity of the survey for the analytical purpose it was intended to serve. In addition, MIDT data do not include information about fare or about taxes charged. Therefore, MIDT data cannot meet the content, validity, and reliability needs of the O&D Survey.

2. ARC Travel Agent Sales Data. Some respondents to the ANPRM suggested that the ARC sales tapes be used as a source of data. ARC is a clearinghouse that receives ticket sales data from travel agency sales reports, processes those sales on behalf of Carriers, and re-combines all the agency ticket data into a comprehensive set of ticket data for each Carrier. The ARC ticket data is limited to tickets sold in North America. The proponents of this method suggested that ARC sales could be supplemented with travel agent data from other countries and regions, known as BSPs, but tickets issued through any other outlets would, in effect, be excluded from reporting. As with the MIDT data, even if the proper proportion of agency issued tickets to all valid tickets could be calculated, this plan would presume that the character of agency sold tickets would exactly mirror the character of tickets purchased through other outlets. For the extrapolation to be valid, tickets purchased directly from the Carriers or through direct links via third parties such as Orbitz' Supplier Link tickets and those purchased from other overseas outlets would have to statistically mirror agency-sold tickets for all markets for all Carriers.

Even if a valid extrapolation could be made with extensive testing, the proportion of agency issued tickets as a percentage of all issued tickets has continuously fluctuated and has been steadily declining as Carriers cut costs by providing incentives to passengers to book directly with the Carrier. Calculating the constantly fluctuating sample size, (i.e. the proportion of tickets issued through travel agencies as a percentage of all tickets issued each month) when the count of all tickets is unknown, would be impossible.

It should be noted that, in 2004, ARC and several Carriers began testing a product called the “AIA First & Final Interline Billing Service” based on ARC's Compass data warehouse. This product might assist some Carriers who elect to use it to provide some O&D Survey data to the Department. This is a fundamentally different proposition than using ARC travel agent sales as the sole source of data for the O&D Survey.

3. Transmission Control Number (TCN) records. Most of the Air Carriers that responded to the ANPRM either endorsed or acknowledged the possibility of using GDS TCN records combined with TCN records generated by the Carriers. A TCN is a supplementary record created to carry information about a Ticketed Itinerary between interested parties. The information on a TCN record is a copy of the information used to create a Ticketed Itinerary, but the presence of a TCN record does not necessarily guarantee that a Ticketed Itinerary was issued. This distinction is important. An issued Ticketed Itinerary is a legal contract for carriage. Whereas each Ticketed Itinerary will generate exactly one sale record in the Issuing Carrier's accounting system, some Ticketed Itineraries will have generated multiple TCNs and some Ticketed Itineraries will have generated no TCN at all.

The Carriers' passenger revenue accounting systems record the issuance of a Ticketed Itinerary when the company itself issues a Ticketed Itinerary or when it is notified by a travel agent that a Ticketed Itinerary was sold on their ticket stock. The sharing of TCN records in the industry is based on the concept that the TCN is supplementary information about a Ticketed Itinerary, and it is not, itself, a Ticketed Itinerary. The Carrier accounting systems are built to anticipate that there will be missing TCN records and duplicate TCN records in the TCN exchanges between Carriers. Accounting systems are designed to handle these contingencies with a variety of supporting subsystems. Using TCNs as a surrogate for actual Ticketed Itineraries in these situations would over-report travel when duplicate TCNs are present. Ticketed Itineraries that are issued for which there is no corresponding TCN compound the problem. As with the unreported reservations in the MIDT data, Ticketed Itineraries created under circumstances in which a TCN is not generated result in under-reporting of travel. Like the MIDT, the proportion of over-reported travel and the proportion of under-reported travel are both unmeasurable and, again like the MIDT, we cannot assume that the over- and under-reported tickets are equivalent.

Proponents of this method advocate that the Department require Carriers to manufacture TCNs for tickets for which a TCN does not already exist. Mandating participation of all Carriers in what is now a voluntary TCN exchange could constitute a significant cost for Carriers, particularly those Carriers with a business model that does not benefit them to participate in the TCN system in their usual course of business. A less burdensome alternative for Carriers that do not now participate in the TCN exchange system would be for these Carriers to format an alternate simpler record structure rather than require the Carrier to format the TCN record. The simpler record would be designed specifically for submitting data to the O&D Survey and would be less burdensome to create than the more complex TCN record, which supports the needs of the Carriers' passenger revenue accounting.

TCNs contain sensitive personal identification and financial information that, while an important component of the Carriers' accounting needs, is unwanted by the Department. The Carriers would have to purge the personal information from records prior to transmission to the Department. Purging this data makes the TCN unfit for the use it was designed to serve. Several respondents to the ANPRM endorsed the concept of employing a third party to perform this task on behalf of the Carriers. However, Continental Airlines (Docket OST-1998-4043-44), supported by Wayne County Detroit Metropolitan Airport (Docket OST-1998-4043-23), pointed out that the ultimate burden to accurately report a ticket is on the Carrier. Proposing that a third party cleanse TCNs does not absolve the Carrier of its ultimate responsibility to properly report to the Department. The third party processor would have to be the agent of the Carriers not an agent of the Department because the Department holds the Carriers responsible for the integrity of the data. Thus, introducing a third party to purge personal data would complicate the Carriers' administrative burden because of the added responsibility to select and to monitor a third party processor.

The GDSs create the TCNs for Ticketed Itineraries distributed by travel agents, but the Department holds the Carriers responsible for accurate O&D Survey reporting. In order to improve the accuracy of its O&D Survey data, the Department may have to require Carriers to accept TCNs from the GDSs and match them to their internal list of tickets to verify that a TCN and a Ticketed Itinerary had been created before reporting the itineraries to the Department. Introducing the additional verification step would be an added burden. Carriers that rely on travel agencies to distribute their Ticketed Itineraries would likely find that it would be less burdensome to create original records for its Ticketed Itineraries, and submit them directly to the Department, rather than sort through the GDS generated TCNs from travel agencies to determine whether any TCN records were missing and whether any TCN records did not have a corresponding Ticketed Itinerary. Thus, should the Department use TCN exchange records, Carriers even that now participate in the TCN exchange system might find it less burdensome to simply generate O&D reporting records from their accounting system.

A TCN record contains data that are a copy of itinerary data that was valid as of the date the record was created. Passengers often change plans after the ticket purchase, necessitating the passenger initiate changes to the Ticketed Itinerary. Some changes are considered minor and Carriers, typically, do not perform the exchange transaction for minor changes. Conversely, some subsequent changes to the passenger's itinerary prompt the generation of a new Ticketed Itinerary in exchange for the existing one. Each Carrier makes that determination based on its own needs and performs the exchange transaction according to its own business practices. If the Department uses TCN records as its reporting mechanism, the Department's data needs would necessitate that the Carriers notify the Department of the intended change in travel plans. The need for standardized reporting would, in turn, necessitate standardization of Ticketed Itinerary exchange policies in the industry. Carriers that exchange Ticketed Itineraries would necessarily have to follow the same set of decision criteria in order to standardize the collection of passenger statistics. Carriers with business practices that do not now require the exchange of Ticketed Itineraries when passengers make significant itinerary changes would have to create a process to simulate such a Ticketed Itinerary exchange.

The TCN system that the Carriers use to share data among themselves efficiently serves its intended purpose. Imposing a requirement to mold the Carriers' TCN data exchange system to the Department's purpose would impose a significant cost and administrative burden to the Carriers, and the increased volume could possibly degrade some of the efficiency of the existing TCN system. As modified, the Carriers' TCN exchange system would be less useful for its original intent yet be less robust than the Department requires. The expense of forcing a functioning system to adapt to a new use would be unwarranted when other sources of data are available.

United (Docket OST-1998-4043-15) acknowledged the problem of over counting passengers due to changed routings, and refunded tickets and stated that the data inaccuracies could easily be addressed. “Air carriers' internal use of TCN reports has shown that relatively simple adjustment factors can be employed to obtain an accurate measure of actual traffic lift.” The Department acknowledges that individual Carriers could and do use the information from the TCN exchange system as a substitute for actual Ticketed Itinerary sales for decision support functions. When a Carrier can use its other internal data for validation and its unique experience with TCNs arriving from various sources, it could find information from TCNs quite useful. However, the knowledge and experience of each Carrier within its route structure and within its operating experience is a fundamental requirement of making TCN data a useful source of information. Furthermore, the Carriers have the ability to use information from their accounting systems to edit, supplement, or purge the TCN records they use as the input to their decision support systems. The Department cannot duplicate that ability nor can we duplicate each Carrier's experience and knowledge of the mathematical relationship between the numbers of TCN records to the numbers of actual passengers. If the Department does not require TCN records to be verified by a sale record by the carrier prior to being submitted to the O&D Survey, then using TCN records that are unverified by an actual sale would require that the Carriers maintain a complex set of adjustment factors. Each Carrier's experience with TCN adjustments would have to be submitted so that it can be included in the Departmental adjustment factor. Since the flow and composition of TCN's changes from month to month and season to season, each Participating Carrier would have to calculate and provide to the Department an accurate adjustment on a monthly basis.

We believe that using unverified TCN's with adjustment factors would be a significant burden on the Participating Carriers without providing the accuracy the Department requires. We believe that using TCN's verified by actual sales would cause a significant burden on the Carrier's existing TCN exchange system, and would also necessitate standardization of exchange ticketing practices that would enable the Department to set up a system to remove exchanged tickets and refunded tickets from the database. Neither of these two options is as compelling as the simple requirement to report tickets verified by a sale and first use of the ticket for travel, and therefore, we are not advocating the use of TCN records as the basis of reporting the O&D Survey.

Nevertheless, the Department recognizes the key role of the Carrier's TCN project in standardizing data elements regarded as important to the Ticketed Itinerary and the industry wide agreement on the definitions of those elements. The Department seeks comment to incorporate this standardized consensus to the extent possible in its proposal to revise the O&D Survey in accordance with established industry practice.

4. Electronic Tickets. Continental Airlines (Docket OST-1998-4043-44) proposed that a survey consisting exclusively of electronic tickets would be sufficient data for the O&D Survey. Electronic tickets are widespread in the aviation industry and would include the majority of Ticketed Itineraries sold in the U.S. and used on U.S. Air Carriers. However, not all Ticketed Itineraries are electronic. Non-electronic Ticketed Itineraries would, in effect, be exempt from reporting. In addition, electronic tickets only contain information about Ticketed Itineraries issued through a particular set of circumstances. Even if the proper proportion of electronic tickets to all valid Ticketed Itineraries could be calculated, this plan would presume that the character of electronic tickets would exactly mirror the character of Ticketed Itineraries purchased through other means. Interline itineraries and Ticketed Itineraries issued in foreign countries would be disproportionately represented in the non-electronic Ticketed Itineraries. Since these populations are likely to have travel patterns that differ from the travel patterns of electronic ticket holders, it is very unlikely that the character of non-electronic Ticketed Itineraries would be mirrored in electronic tickets. The level of over-reporting or under-reporting could disproportionately affect different types of markets (e.g., predominantly leisure versus predominantly business markets), further reducing the validity of the survey for the analytical purposes it was intended to serve. Even if we could validate the extrapolation of known electronic ticket data to unknown non-electronic Ticketed Itinerary data, the proportion of electronic tickets as a percentage of all issued Ticketed Itineraries would continuously fluctuate. Calculating the constantly fluctuating proportion when the count of all Ticketed Itineraries is unknown would be impossible.

5. Actual Passenger Transportation. Many of the airports that responded to the ANPRM advocated that the Carrier that operates the passenger's flight perform the O&D Survey reporting as each flight takes place. However, the Carrier that transports the passenger does not always have the itinerary information that would make it possible to determine the True O&D of the One-way Trip from any given passenger flight segment. Even if it did, operational problems, weather problems, and an uncountable variety of human errors or situations involving airport security or even city traffic beyond the passenger's control can affect the way a passenger completes scheduled travel. The supporters of this technique did not suggest a method to reassemble the various segments of a single passenger's journey, reported at various times by multiple Carriers, into a coherent One-way Trip. Diverted flights, delayed flights, and lost flight envelopes would make it impossible to decipher the intended One-way Trip without a lift/sale match system. Carriers that have built lift/sale match database systems have found it to be a long and expensive undertaking. United Air Lines (Docket OST-1998-4043-15) commented that it firmly believed that reconciling to actual lift was both difficult and unnecessary.

The Department believes that construction of a lift/sale match system on an industry-wide basis would be a significant burden for both the Department and the Carriers, which would not be offset by the benefits. Moreover, for purposes of analyzing traffic flows and understanding market size and characteristics (the primary uses of the O&D Survey), the Department believes that it is more valuable to know the itinerary the customer purchased than to know all of the exigencies of air travel that have interfered with the passenger's stated travel intention. American Airlines (Docket OST-1998-4043-5) and US Airways (Docket OST-1998-4043-7) commented that there would likely be an undesirable time lag incurred in obtaining, reassembling, and processing acceptable accurate Flight-Coupon Stage information. The Department believes that the potential problems of gathering the data from multiple sources, the expense of building the database for re-assembling the itinerary data from the multiple sources, and the potential undesirable time lag associated with such a system render the use of this data source for the O&D Survey impractical.

e. Review of Existing Data Sources

By far the least intrusive way of obtaining aviation data from the industry is through the use of existing sources of industry data. Each of the existing sources of data the respondents suggested as a source of data for the O&D Survey provides information at minimal cost to the Carriers. However, none is a comprehensive source of information and therefore fails the test of accuracy, reliability, and completeness. In investigating each proposed data source, the Department has considered the possibility of supplementing each data stream. However, the effort required of the Carriers to supplement the data to enhance the quality adds complexity and cost. In every case, the data still fall short of OMB guidelines for ensuring quality of information disseminated by Federal agencies.

Furthermore, Carrier participation in these sources of data is not universal. The Department's use of any of those data sources would, effectively, mandate Carrier participation in processes in which they have chosen not to participate to date, or have participated at a very low level. Moreover, a Carrier's level of participation in the selected data source might result in varying levels of representation of its passengers in the data reported to the O&D Survey. This would disproportionately disadvantage a particular Carrier, or group of Carriers. The Department seeks comment as to whether the O&D Survey can be satisfactorily revised by reusing another collection of industry data compiled for a purpose other than the O&D Survey (e.g. TCN, MIDT, ARC, etc.). Comments should specify the extent to which the existing industry source of data will (1) maximize accuracy, reliability, completeness, and non-bias, (2) minimize the burden of collection on the Participating Carriers, and (3) minimize the effects of changes to itineraries over time, as well as the specific modifications required of that data source. Comments should also specify the costs and benefits of using an existing source of industry data, including the costs and benefits of modifications to the existing data source to meet the three criteria described above.

f. Designating the Issuing Carrier as the Participating Carrier

The Port of Portland (Docket OST-1998-4043-19) recommended that the selling Carrier be incorporated into the O&D Survey. The Department prefers the term “Issuing Carrier” to “selling Carrier”, since some Revenue Passengers travel on Ticketed Itineraries for which no funds change hands. Nevertheless, we believe this suggestion has merit. This suggestion would require creating a dedicated source of data such as the current one the Department requires from the Operating Air Carrier. It has several advantages, notably the simplicity of gathering information from the Issuing Carrier. A data source created by the Issuing Carrier easily meets two of the three criteria for selection of an appropriate data source for the O&D Survey (See Section I.3—O&D Survey Redesign: Reporting Requirements). The data quality concerns, criterion number one, are minimized because the Issuing Carrier has the most accurate and reliable knowledge of the passenger itinerary. The burden on the Carriers, criterion number two, is less than the burden heretofore placed on the Operating Air Carrier because it removes the burden of requiring the Operating Air Carrier to obtain information from the Issuing Carrier before reporting the itinerary. The changes that take place in an itinerary over time, criterion number three, remain a concern, depending on when the data is copied for submission to the O&D Survey. In all sources of data, a change that takes place after triggering the reporting event is invisible to the O&D Survey.

g. Issuing Carrier's Ticketed Itineraries at the Time of Sale

We considered an O&D Survey design that requires the Issuing Carrier to report the Ticketed Itinerary triggered at the time when the Ticketed Itinerary is entered into its passenger revenue accounting system. Depending on the Carrier, from zero to five percent of Ticketed Itineraries issued are refunded, and between five percent and 20 percent of Ticketed Itineraries are changed after the itinerary is issued. The Department considered ignoring refunds and changes subsequent to the issue date, but determined that doing so would introduce unacceptable unreliability. The number of refunded tickets is small, but five percent of issued tickets are not inconsequential. The itinerary changes pose a more significant problem.

Carrier systems handle passenger-initiated changes to a Ticketed Itinerary in two ways. In some cases the change will be noted in the existing itinerary record, and in some cases the change will cause a new Ticketed Itinerary to be issued in lieu of the previous one.

When the existing itinerary is changed after it has been reported, then the changes will not be reported to the O&D Survey. Once the O&D reporting criteria are encountered, the Participating Carrier copies the information to a submission record and subsequent changes are invisible to the O&D Survey. In some cases, however, a new Ticketed Itinerary is issued in exchange for the previous one, and the Department would have to formulate a policy to address these cases. Unless the original Ticketed Itinerary is removed when the newly issued itinerary is added, the passenger is counted twice when the reissued Ticketed Itinerary is reported to the O&D Survey. There is inconsistent handling of data between Carriers that issue new tickets in exchange for the previous tickets and Carriers that alter tickets in place. If the reissued ticket is ignored, then it becomes, in effect, an exempted ticket.

The Department considered requiring that Carriers provide the Department with the identifiers of refunded Ticketed Itineraries and identifiers of Ticketed Itineraries that were replaced in an itinerary reissue transaction so that these could be removed from the data and the new Ticketed Itinerary entered instead. The undertaking would be the equivalent of a nation-wide ticket database matching system, and would involve the Department in the accounting details of the revenue accounting peculiarities of each of the Participating Carriers. The diversity of the Carrier business models is reflected in the diversity of their passenger revenue accounting procedures, which would necessitate that correspondingly complex procedures be in place at the Department to handle the various situations that arise from each airline passenger revenue accounting system.

The Department believes processing itinerary changes after the reporting event would greatly compound the complexity and substantially increase the expense of the O&D Survey reporting system to both industry and government. Recording all of these changes would appear to increase the accuracy of the statistics, but would require considerably more effort and expense from the Carriers and impose dramatically more effort, complexity, and expense on the Department. The Department must consider the possibility that the increase in complexity may increase the incidence of errors that would, in reality, decrease accuracy. Finding and removing previously issued Ticketed Itinerary from the data would be similar in complexity to matching lifted flight coupons to Ticketed Itinerary records. The ANPRM comments by American Airlines (Docket OST-1993-4043-5) and US Airways (Docket OST-1998-4043-7) indicate that the attempt to match the sale and actual use would be time consuming as well as complex. Therefore, the Department believes that maintaining multiple reporting events for the same Ticketed Itinerary would interfere with the Department's goal of processing and disseminating data in a timely fashion. In light of the significant complexity, significant cost, the risk of introducing reporting errors, and the risk of introducing timing delays, the Department is not proposing to undertake a nation-wide ticket database matching system to track changed itineraries. However, we seek comment from industry and the public on the merits of these issues.

h. Issuing Carrier's Ticketed Itineraries at the Time of First Use

An alternative to tracking multiple changes to a Ticketed Itinerary is to delay the reporting of the itinerary until the last acceptable point at which a reliable trigger for a reporting event can be designated. The last unambiguous event that can reasonably be used as a reliable trigger for reporting is the first use of the Ticketed Itinerary. The final use of an itinerary is not acceptable as a reporting event trigger because many months can separate the first use of a Ticketed Itinerary from the final use. If the Department collects data at first use, we can hold the information about subsequent flights until the appropriate month for the Flight-Stage of travel to be disseminated. If the Department collects data at final use, we would be confronted with knowledge of Flight-Stages that occurred from one to 11 months earlier. It is not a reasonable alternative to hold the reporting of all data for 11 months in order to collect data from Ticketed Itineraries with widely spaced travel, it is not reasonable to be constantly updating data that has already been released and it is not reasonable to ignore all data from the outbound portions of Ticketed Itineraries that describe travel that is spaced more than one month apart. Therefore the first use of a Ticketed Itinerary is the last reasonable and unambiguous event that can be used as a reporting event.

The first use of the Ticketed Itinerary for travel is the triggering event for reporting in the current O&D Survey. Refunds and reissued tickets that occur subsequent to the reporting event are currently ignored. Fortunately, the numbers of refunds and exchanges that take place after a passenger has already begun the journey are extremely low. Whereas we have accumulated ample evidence that naming the Operating Carrier as the Participating Carrier is the root of many of the reporting problems found in the O&D Survey, we have no accumulation of evidence that indicates that the first use of the Ticketed Itinerary for transportation is unsuitable as the trigger for the reporting event. The Carriers have confirmed that the preponderance of refunds and exchanges take place prior to the first flight, and the Department deems the small number of missed itinerary changes due to subsequent refunds and travel changes to be marginal. The Air Carriers have indicated that the most common change request that occurs after the commencement of travel is for a different return flight that is within a few hours of the original. Therefore, the Department has concluded that the designation of first use of the ticket for travel should continue to serve as the trigger for the reporting event.

i. Proposed Source of Data for the O&D Survey

The Department agrees with the Port of Portland (Docket OST-1998-4043-19) that the selling/issuing Carrier should be incorporated into the O&D Survey. Standard industry accounting practices require that the Issuing Carrier hold the passenger's funds in an unearned revenue account until the passenger flies, or exchanges or seeks a refund, of one or more of the Flight-Coupon Stages. The Operating Carrier notifies the Issuing Carrier when the Operating Carrier transports the passenger on a Flight-Stage. When the Operating Carrier is the Issuing Carrier, the notification is an internal transaction; when the Operating Carrier and the Issuing Carrier are different companies, the notification is an external transaction. In either case, the Issuing Carrier is notified that an Operating Carrier has transported a passenger on a Flight-Coupon Stage. The Issuing Carrier will have knowledge of the triggering event—the first use of the Ticketed Itinerary for travel—because worldwide industry accounting practices already dictate that the Operating Carrier notify the Issuing Carrier that passenger travel has taken place. Moreover, the Issuing Carrier is the only Carrier that has full knowledge of the Ticketed Itinerary, fare, and taxes. Therefore, the Department proposes that the O&D Survey (1) continue to require a dedicated reporting file format, (2) continue to use the Ticketed Itinerary as the source of data, (3) continue to use the first use of the ticket to travel as the trigger for the reportable event, but (4) designate the Issuing Carrier as the reporting entity.

The change in designated reporting entity from Operating Carrier to Issuing Carrier, while keeping the same reporting event trigger, has significant advantages. For Carriers that operate only as Franchise Code-Share Partners on behalf of larger Mainline Partners and do not issue tickets on their own ticket stock, the task of reporting the Code-Share passengers will shift to the respective Mainline Partners. For Carriers that do not interline passengers with other Carriers, the Department anticipates that the change in reporting entity will require very little change in current procedure beyond gathering the additional data elements. This change is a significant improvement for carriers that maintain interline agreements because tickets from re-accommodating passengers as a result of irregular operations represent a large portion of the most troublesome and time consuming itineraries to report. Under this proposal, responsibility for reporting the itineraries of those re-accommodated passengers will go to the Issuing Carrier.

The most significant advantage of the change in reporting responsibility for interlining Carriers is that the identification of the Carrier with the responsibility to report data is no longer ambiguous. The current system requires each itinerary to be scanned to determine whether it is apparent that another Participating Carrier has already reported the Ticketed Itinerary. This is a complex task that requires examination of the itinerary for the presence of other Participating Carriers scheduled earlier in the itinerary. The task requires knowing whether the other Carriers present are Participating Carriers and whether there are any code-share relationships to be considered.

The current O&D System discourages early reporting because Issuing Carriers must have sufficient time to send data to the Participating Carrier. This proposed O&D Survey encourages early reporting because the Participating Carrier is the Issuing Carrier. The most cost efficient method of reporting is to enable the sale/lift match procedure to copy the requisite data as soon as the Issuing Carrier realizes that the lifted Flight-Stage coupon is the first use of the Ticketed Itinerary for travel. This is a single, clearly identifiable reporting event.

Usually, the knowledge that a Ticketed Itinerary has been issued precedes the first evidence of use of the Ticketed Itinerary in a Carrier's passenger revenue accounting system. However, the Department recognizes that information about the Ticketed Itinerary's issuance sometimes arrives after the evidence of first use. This happens most frequently in itineraries sold in foreign countries. Although the reporting event trigger remains the passenger's use of the ticket, the Department's intent is to obtain the best possible data. Therefore, we propose that the Participating Carrier match the first evidence of flown use with the information from the Ticketed Itinerary's issuance by whatever means the Issuing Carrier creates the match in its normal course of business. The itinerary must be reported when the Issuing Carrier's accounting system resolves the problem. Monitoring for first use includes interline billing notification that a Flight-Stage coupon was used for transportation on another Carrier, including those that were flown on other Carriers as a re-accommodation.

The Department believes that ignoring itinerary changes after the commencement of travel is an acceptable trade off for the simplicity and lower cost of reporting. Continuing the practice of ignoring itinerary changes subsequent to the commencement of travel is consistent with the current O&D Survey. This will minimize disarticulation that will occur in the transition from the old O&D Survey data to the proposed O&D Survey data. The Department seeks comment from the industry and the public as to the advantages or disadvantages of changing the reporting source or changing the reporting event. We request that recommendations of alternative reporting sources or alternative reporting events discuss the explicit and implicit reporting exemptions inherent in the recommended source of data, and the efficacy of processing itinerary changes that may take place after the triggering of the recommended reporting event.

4. Significant Issues Related to the Data To Be Collected

a. Proposed End to Sampling

There are several factors that support the redesign of the current sample selection procedures. There are concerns with bias related to the current sample. The current rule requires a Ticketed Itinerary to be selected when the Ticketed Itinerary number ends in zero. This methodology assumes that all Carriers use ticket numbers, and it assumes that ticket numbers are randomly distributed (i.e., that each passenger has an equal chance of obtaining a Ticketed Itinerary number ending in a specific digit). When the O&D Survey was established, these assumptions were, in all likelihood, valid. All Participating Carriers used carefully controlled and guarded ticket stock that was pre-printed with ticket numbers. There was little incentive to deviate from the simplicity of taking each ticket sequentially from the box for each new customer. Thus, drawing a sample of tickets ending in zero lent itself to obtaining a random 10 percent sample of the passengers.

At least one Participating Carrier that uses ticket numbers on standard agent tickets is aware that ticket numbers ending in a zero constitute 11 percent of their total Ticketed Itineraries, but does not know the cause of the variance from the expected 10 percent. Ticket numbers are assigned to travel agencies and Carriers in blocks of assigned numbers. When a ticket distributor (a ticket agency or Carrier itself) uses preprinted ticket number stock, then the actual paper tickets are physically delivered to the entity that distributes the Ticketed Itineraries. In the air travel industry today, the use of preprinted paper ticket stock is very low. The ticket distributors are assigned a set of numbers that are applied to Automated Ticket and Boarding Pass (ATB) ticket stock and a set of numbers that are applied to electronic tickets. The basis of sampling every Ticketed Itinerary with a number ending in zero assumes that ticket numbers continue to be assigned sequentially to passengers, but there is no guarantee that this assignment process is followed by all ticketing systems.

Members of a travel group, such as an inclusive tour group, might be assigned ticket numbers in some systematic way, such as grouping them according to the final digit of their ticket numbers. Such use would invalidate the Department's assumption that each passenger has an equal chance of being assigned a ticket number ending in zero. We are unaware of any practice of systematic group assignment of ticket numbers to Ticketed Itineraries other than random assignment, but we are also unaware of a prohibition on such assignment of numbers.

However, currently, three Participating Carriers have requested permission to use non-standard sampling under the current O&D Survey rules because these Carriers do not assign traditional ticket numbers to their Ticketed Itineraries. Because some Carriers do not use ticket numbers, and because there is no longer a compelling reason to believe that ticket numbers are assigned sequentially, or assigned randomly, the Department proposes to discontinue the use of ticket number as a determinant of a 10 percent sample of Ticketed Itineraries.

Even if it were possible to draw an unbiased 10 percent sample, a 10 percent sample is inadequate for fulfilling the Department's mandates, particularly with respect to programs designed to foster air service to small communities. The IG (AV-1998-086, page 26) stated “in these ‘thin’ markets, the number of passengers, and therefore sample tickets, is relatively small. As a result, errors from a 10% sample are likely to be significant so that the sampling results are unreliable.” The Department has calculated that using a valid, random, 10 percent sample, the smallest market in which a 10 percent change in the market could be detected with 95 percent confidence is a market of approximately 29,000 passengers. The fourth quarter 2003 O&D Survey measured 94,347,000 Directional O&D passengers accommodated on 31,385 routes in the 48 contiguous states in that quarter. Of the 31,385 routes, 754 (2.4 percent) had 29,000 or more passengers in the quarter. This means that the Department can measure a 10 percent change in passengers with 95 percent confidence from quarter to quarter on only 2.4 percent of the total number of routes in the 48 contiguous states.

When researching a market with multiple airlines, the minimum number of passengers must exceed 29,000 on each airline in order for the research to attain this level of validity. There are considerably fewer than 754 routes wherein all the Carriers are transporting 29,000 passengers. These 754 routes accounted for more than half the total passengers traveling between the 48 states in that quarter, but the Department's mandate to adapt the air transportation system to the present and future needs of commerce requires the study of many of the remaining 97.6 percent of routes. Of the remaining 97.6 percent of markets, those that suffer the most distortion are ones where the passenger count is low, such as small city markets. Increasing the sample size would enable more precise measurement of smaller markets. However, detecting a 10 percent change with 95 percent confidence in a study of a market with an estimated total of 10,000 passengers would require a 24.4 percent sample.

The Essential Air Services program (EAS) and the Small Community Air Service Development Program are the two primary examples illustrating the Department's need for more comprehensive data. These programs are focused on smaller markets and require evaluation of service and fares. Under EAS, the Department determines the minimum level of service required at each eligible community, by specifying (1) a hub through which the community is linked to the national network and (2) a minimum service level in terms of flights and available seats. Where necessary, the Department pays a subsidy to a U.S. Air Carrier to ensure that the specified level of service is provided. The Federal government budget for EAS exceeds $100 million each year.

All but a handful of the EAS markets are less than 20,000 passengers annually, and the majority of EAS markets are less than 10,000 passengers annually. While decisions about EAS markets could be made at confidence levels much lower than 95 percent, the Department has long acknowledged that the 10 percent sample is not sufficiently valid for use in monitoring the EAS program. The candidate Carriers provide fare and destination information to the Department as part of the application process. The O&D Survey is not generally used to validate or refute the Carriers' assertions because the sample size of 10 percent is not sufficiently accurate. Aggregating data to an annual basis from a quarterly basis increases the validity of the O&D Survey data. However, even on an annual basis, for most EAS decisions, increasing the sample size to 24.4 percent is still insufficient to validate the Carriers' assertions with a high level of confidence.

While EAS and the Small Community Air Service Development Program specifically focus on markets served by smaller carriers, the Department's statutory responsibility to adapt the air transportation system to the present and future needs of commerce is much more extensive than the needs of the EAS program. Because these markets are inadequately represented in the current O&D Survey, the Department's mandate requires a disproportionately high amount of time and resources in studying markets with lower than average traffic volume.

The Department considered the possibility of reducing the cost of the O&D Survey by creating a sample that would collect less data overall and still fulfill the data needs of the users of the O&D Survey. Ideally, the Department could reduce the cost of collection by obtaining samples of varying sizes depending on the markets to be studied. To achieve that efficiency, a system of assigning various sample sizes to corresponding market sizes would need to be established. The Department could develop an algorithm where samples larger than 10 percent could be drawn for those markets where the 10 percent sample is inadequate. The process of increasing the sampling rates disproportionately for relatively rarer subgroups, in order to have adequate sample sizes for estimation, is called oversampling.

In order to oversample specific itineraries based on selected characteristics, the Carriers will have to know those characteristics for every individual itinerary. A collection of all the eligible units that have a known probability of sampling, along with the characteristics that will be used to draw the sample, is known as a sampling frame. Thus, a sampling frame of all itineraries with the relevant sampling variables (characteristics that would determine the oversample such as arrival and departure airports and date of travel) must be assembled. Once this was done, each Carrier would have to apply the different sampling rates for the different subgroups and draw the sample.

Finding a reasonable way to oversample subgroups to obtain estimates for all affected markets would be difficult. The Carriers submit data in the form of Ticketed Itineraries to the O&D Survey. Airport pairs of varying sizes and combinations appear on a single Ticketed Itinerary. Collecting the portion of the Ticketed Itinerary that corresponds to the specific sample size for that market is a complicated task. In April 1986, Department regulations began allowing a stratified sample, but continued to collect data by collecting whole itineraries instead of portions of itineraries appropriate to the stratified sample. The rule stated that large markets were to be sampled at one percent when the Ticketed Itinerary consisted of travel only within that large market, and all itineraries that included travel to any other destination, or combination of destinations, were to be sampled at 10 percent. All Participating Carriers decided that the simplicity of using a single reporting selection criterion outweighed any savings that might accrue from sending the smaller volume of data. This illustrates the Department's position that due to the technical complexities and additional burden for the Issuing Carriers associated with differential sampling rates, it is less burdensome for Participating Carriers to apply a single sampling rate. Given the need for details on all smaller markets, the only sampling rate that will lead to the fulfillment of both the Department's and industry's needs is a census or 100 percent sample.

Furthermore, as market sizes change over time, the designated sample size for a market would have to be adjusted. Determining market size is not a simple operation. In effect, Ticketed Itineraries have multiple components. In Ticketed Itineraries that include outbound and return travel that are scheduled to be at least 30 days apart, the return portion of travel is reported at least 30 days in advance. Ticketed Itineraries would be sampled at the rate that was in effect when that Ticketed Itinerary was reported. When the designated sample size for one component of the itinerary is adjusted based on changes in that market, Ticketed Itineraries reported before the change would be sampled at the rate in effect before the change, but the Ticketed Itineraries that were reported after the change would be sampled at the rate that was in effect after the change. The sampling at differential rates would occur for up to 11 months, which is the number of months a Ticketed Itinerary can be sold in advance of travel.

Users of the data in those changing markets would have to find a way to properly account for varying sample sizes for Ticketed Itineraries submitted before and after the market sample size was adjusted. Therefore, even if a way could be found for the Participating Carriers to report portions of Ticketed Itineraries appropriate to the stratified sample, the changes in market size over time could make the data very difficult to use. Even if the Carriers were able to implement such a sample design, the complexities associated with weighting make a sample less attractive for Carriers, the Department, and other stakeholders. The Participating Carriers would have to provide data about the entire sampling frame in order for the Department to create correct sampling weights. These sampling weights are necessary when a sample of itineraries is selected instead of all itineraries. Sampling weights would be necessary to ensure that the O&D Survey provides accurate estimates of the total number of itineraries nationally and for each market. In comparison, we believe that sending the entire census of itineraries will be simpler and much less burdensome than stratified sampling for Participating Carriers.

The Department has considered conducting a census for small markets and a sample for the remaining larger markets. Any parallel system of differential sampling, whether it is in one single survey or multiple related surveys, will lead to a greater burden on Carriers due to the need for a sampling frame with all the necessary sampling variables. However, the cost to Participating Carriers would increase considerably because two systems would be required. Participating Carriers declined use of multiple sample rates in 1986, citing the relatively low expense of transmitting additional records compared to the relatively high expense of additional computer programming work. Since the relative cost of storage and transmission of data has continued to decline, especially compared to the increasing cost of programmers, we believe that the increased complexity of applying multiple sampling rates would be far more burdensome to Participating Carriers than keeping a single O&D reporting system.

The sampling process must be changed in order to draw an unbiased sample. Yet, there is evidence that a 10 percent sample provides insufficient accuracy for the needs of the Department and other users of the O&D Survey data. Using multiple sampling rates adds undue burden upon Participating Carriers. Because the airline ticketing and accounting systems are all computerized, the Department feels that a census would be the most efficient and least burdensome solution for the Participating Carriers and the Department. We therefore propose to end the sampling process and begin the collection of 100 percent of Ticketed Itineraries.

The Department is willing to reconsider sampling, subject to comments from the industry and the public regarding the suitability of continuing to use a sample. The Department's data collection guidelines state that data collection of 100 percent of the population of inferences is the most accurate approach, but that the cost of collection and other resource restrictions should be considered when making this decision. If the cost of collection and transmission of 100 percent of Ticketed Itineraries is unacceptably high, then a sample design based on sampling theory, making use of a methodology other than ticket number for selection, will be needed to address the goals of efficiency and accuracy. The sample design should ensure that there are enough sample cases for reliable information about small markets. The Department seeks comment regarding the continuation of a sampling methodology, and requests that these comments make detailed proposals on methods of revising the sampling. Proposals should suggest a probability sample based on established sampling theory, including methods of estimating the variance and taking into account the nature of the missing data. The proposed methodology must give all members of the target group a known non-zero probability of being represented in the sample taking into consideration the tremendous variations in relevant Carrier business models and practices, geographic markets, and sales distribution outlets.

b. Effect of Proposed Changes on Small Entities

The development of hub-and-spoke networks increased the demand for small- and medium-sized aircraft to feed the hubs, which, in turn, over time fostered the growth of the Carriers specializing in the operation of these aircraft. Regional Carriers have substantially changed their business model to one heavily based on the “fee for departure” service in which a larger Mainline Partner pays the regional Carrier for operating flights under a long term contract using the Mainline Partner branded livery. The Mainline Partner typically assumes all responsibility for pricing, selling, marketing and inventory management for its regional partner's flights. However, most importantly, the Mainline Partners have assumed the role of Issuing Carrier for the Ticketed Itineraries issued to passengers for travel on their regional partners. The passengers on these smaller Carriers represent a significant portion of the passengers worldwide although, historically, most have not been obligated to report passengers to the O&D Survey.

It is common now for a regional Carrier, operating as a Franchise Code-Share Partner, to acquire jet aircraft having 60 or more seats on behalf of one of its Mainline Partners and thereby acquire O&D Survey reporting status for all its flights for all its Mainline Partners. More often than not, however, the Franchise Code-Share Partner is not in a position to report passengers because the “fee for departure” arrangements leave the necessary passenger data in the hands of its Mainline Partners. Currently, the larger Mainline Partner typically prepares the O&D Survey report on behalf of the Franchise Code-Share Partner and sends it to the Franchise Code-Share Partner, which in turn forwards it to the Department. The Department's designation of the Operating Air Carrier as the Participating Carrier requires the Mainline Partner and the Franchise Code-Share Partner to take these additional steps to get the appropriate data transmitted by the Participating Carrier, adding cost and complexity while providing no added value.

When a regional Carrier negotiates code-share arrangements with two or more Mainline Partners, the Franchise Code-Share Partner may qualify for reporting because of the acquisition of an aircraft operated on behalf of one of its Mainline Partners. Once qualified as a Participating Carrier, however, it must begin reporting all passengers for all Mainline Partners. This causes added complexity to be placed on all Mainline Partners, even if the regional Carrier does not fly 60 seat aircraft for all its Mainline Partners. Even worse, relinquishing its aircraft of more than 60 seats returns a regional Carrier to non-Participating status for all its Mainline Partners. In the past, the increase and decrease in the volume of Ticketed Itineraries being reported as a result of acquisitions and divestitures of larger or smaller aircraft have created significant problems for users of the O&D Survey data.

The responses to the ANPRM expressed the unanimous opinion that the exemption for small Carriers requires significant revision. Northwest Airlines (Docket OST-1998-4043-49) stated that smaller aircraft are serving meaningful markets. The City of Chicago (Docket OST-1998-4043-27) pointed out that the 60-seat limit is irrelevant and outmoded. Los Angeles World Airports (Docket OST-1998-4043-28) noted that some Carriers are important to an airport regardless of whether they meet current reporting criteria. The Regional Airline Association (Docket OST-1998-4043-11) in its ANPRM comments objected to the 60 seat rule stating, “It is clear that for the U.S. regional airline industry, the current data collection process is both inappropriate and inconsistent. The current structure of reporting rules and regulations offer what the Association considers to be an approach to information gathering that is out of step with the current operating environment for regional airlines.” It further stated, “A vestige of a bygone era, the 60-seat distinction is ill-suited to the regional airline industry of today, but perhaps more importantly, that envisioned for the future.” The entire aviation community has noted that, to understand passenger flows, it is crucial to include in the O&D Survey passengers traveling on Carriers that operate aircraft with fewer than 60 seats.

The opinions provided in the responses to the ANPRM varied widely regarding the point at which a regional Carrier's passengers are no longer significant enough to be counted. The Regional Airline Association (Docket OST-1998-4043-11) stated that any Carrier with annual revenues of $20 million should report its tickets. ALPA (Docket OST-1998-4043-18) recommended a $10 million cutoff. The Port Authority of New York and New Jersey (Docket OST-1998-4043-25) would set the revenue cutoff at $1 million so long as the Carrier did not operate any aircraft with more than ten seats. The Allied Pilots Association (Docket OST-1998-4043-16) recommended defining the threshold as any carrier operating aircraft having at least 30 seats and transporting at least 100,000 annual passengers. Delta Air Lines (Docket OST-1998-4043-21) and US Airways (Docket OST-1998-4043-7) both recommended that any passenger carried on a jet aircraft should be reported. Los Angeles World Airports (Docket OST-1998-4043-28) recommended using a revenue threshold or a given number of flights in lieu of the size of aircraft the Carrier operates, but left the calculation of the specific threshold to the Department.

Metropolitan Washington Airports Authority (Docket OST-1998-4043-38) recommended reporting by Carriers that operate aircraft with 25 or more seats or that are owned by Participating Carriers. Oakland International Airport (Docket OST-1998-4043-14) and R.W. Mann & Company (Docket OST-1998-4043-13) both recommended a proposal similar to the Metropolitan Washington Airports Authority proposal, but both used 30 seats as the cutoff, and both believed that code-share Carriers should report regardless of their Mainline Partner's position. Daniel Kasper (Docket OST-1998-4043-62), an industry analyst who filed a response, echoed the 30-seat cutoff, but recommended that operators of 30-seat aircraft would only have to report if they transported 100,000 annual passengers. Wayne County and Detroit Metropolitan Airport (Docket OST-1998-4043-23) was even more stringent, recommending that Carriers transporting 100,000 annual passengers, operating under a code-share agreement with a Mainline Partner, or operating aircraft with 15 or more seats should report. American Airlines (Docket OST-1998-4043-5), the City of Chicago (Docket OST-1998-4043-27), John Brown Company (Docket OST-1998-4043-33), Norfolk Airport Authority (Docket OST-1998-4043-31), Northwest Airlines (Docket OST-1998-4043-49), The Port Authority of New York and New Jersey (Docket OST-1998-4043-25) (the latter in conjunction with the $1,000,000 cutoff mentioned above) endorsed 10-seat aircraft as the criterion for reporting. The National Transportation Safety Board (Docket OST-1998-4043-48) provided the most rigid recommendation. It recommended that every U.S. certificated Air Carrier should report regardless of size, even air taxis.

The Department believes that moving the threshold of reporting from operators of 60-seat aircraft to operators of 15-seat aircraft will not be a significant reporting burden on small Carriers if the reporting responsibility is shifted to the Issuing Carrier. Since the majority of small Carriers are not Issuing Carriers, under the proposed system they would not be required to report the O&D Survey. Nonetheless, small Carriers, such as non-scheduled air taxis and other similarly small operations, represent a significantly different transportation market. The Department acknowledges that passengers in this market must be measured differently than the passengers in the global scheduled air transportation market. We do not wish to burden the truly small airline operations serving local needs. Rather, the Department wishes to reduce the ambiguity in the definition and classification of a Participating Carrier. Moving into and out of the Participating Carrier classification over time is problematic for both the Carrier concerned and the users of the O&D Survey. Therefore, we propose that (1) Carriers flying strictly intra-state service, (2) Carriers flying no aircraft with 15 or more seats, (3) non-scheduled air taxi service, and (4) non-scheduled helicopter service will continue to be exempt from reporting the O&D Survey.

c. Timeliness of Reporting

Respondents representing all constituencies indicated that the erratic publication schedule maintained by the Department was a problem. The Allied Pilots Association (Docket OST-1998-4043-16), Back Associates, Inc. (Docket OST-1998-4043-3), the City of Chicago (Docket OST-1998-4043-27), and United Air Lines (Docket OST-1998-4043-15), among others, noted the delays in the release of data. United Air Lines cited the timeliness of the data release as the most important factor the Department could address to make the data more useful. Both Carrier and non-Carrier respondents indicated that the data should be released on a monthly schedule instead of a quarterly schedule.

The Department is aware that each Participating Carrier must verify its Issued Ticketed Itineraries that were first used for travel during a reporting month. It is our understanding that the majority of Participating Carriers will require some period of time, following the end of a month, for this verification process. However, the erratic receipt of data from Participating Carriers affects the Department's release of data to all stakeholders. For example, BLS produces the all-items CPI, an important economic indicator which includes an airfare index. BTS has begun publishing a quarterly experimental research air travel price index (ATPI) that uses O&D Survey data. When monthly O&D Survey data become available, BTS intends to forward its ATPI to BLS for possible inclusion in the CPI. Because BLS requires all index components to be submitted no later than the fifth day of the month following the reference month, we are considering requiring each Participating Carrier to submit O&D Survey data for each month no later than the 5th day of the following month so that BTS can submit its ATPI within the time constraints of the CPI production schedule. Under this option, we would permit daily, weekly, and/or monthly data submissions by Participating Carriers. We are aware that weekly reporting cycle for travel agents would cause some passengers who purchase air travel near the end of the month and fly within the month to remain unreportable on the fifth day of the month due to missing information about the sale of the Ticketed Itinerary. We seek comment on the costs and benefits of requiring Participating Carriers to submit O&D Survey data for a particular month by the 5th day of the following month. Comments advocating alternative reporting due dates should include information addressing both the alternative due date's influence on the timeliness and on the accuracy of the data.

The Department proposes that Participating Carriers will provide the name and contact information for a Designated Carrier Liaison to act on behalf of the Participating Carrier in operational matters pertaining to the company's collection and submission of the O&D Survey. In order to maintain its own data dissemination schedule, the Department will monitor the receipt of Participating Carrier data very closely, and contact the Designated Carrier Liaison promptly when problems arise. Exact deadlines for reporting will be published in Passenger Origin-Destination Survey Directives issued by the Department.

d. Data Monitoring

Guidelines in the Paperwork Reduction Act of 1995 direct agencies to develop information resource management procedures for reviewing and substantiating the quality of information before it is disseminated. The IG (AV-1998-086) found that a lack of quality control by Carriers was responsible for chronic inaccuracies in the O&D Survey. In the responses to the ANPRM, the most common request after removal of the 60-seat Carrier exemption and reporting exemption for Foreign Air Carriers was to improve the Department's monitoring of the data that is received. The Port of Portland (Docket OST-1998-4043-19) stated this succinctly: “Enforce data quality standards by filing carriers”. The Department will, therefore, initiate a rigorous process of monitoring and enforcement to maximize the quality of the data submitted to the Department.

It is too early in the planning process to discuss specific data quality monitoring. However, the Department proposes to establish mechanisms to monitor (1) the timeliness of Carrier submissions and (2) the composition of submitted Ticketed Itineraries to ascertain the reasonableness of a Carrier's reporting. The Department will adopt a basic standard of quality and take appropriate steps to enforce the quality criteria subject to an acceptable degree of imprecision. Some late reporting of itineraries will be expected, and, therefore, the degree of promptness and precision that is tolerated may be reduced or increased depending on the circumstances. Established guidelines and methods will be made publicly available and uniformly enforced. The Department will use these guidelines to determine the expected number of late reported itineraries and initiate an investigation when we detect Carriers to be outside those guidelines.

e. Certification of Accuracy

In accordance with OMB guidelines, the Department proposes to establish administrative mechanisms allowing affected stakeholders to seek and obtain correction of information disseminated in the O&D Survey. Since the public relies on accurate Carrier data, we propose to maintain a mechanism of ongoing communications with Participating Carriers through designated representatives. Therefore, each Participating Carrier will provide the name and contact information for its Designated Company Official, who will certify the accuracy of the data submissions. The Participating Carrier will also supply the name and contact information for its Designated Carrier Liaison, who will have the responsibility for resolving day to day operational issues with the Participating Carrier's submitted data.

The Department proposes to collect and record information from Carriers from time to time that the Department deems necessary to adequately monitor the Carrier's data submissions. The requirements will be published in the Passenger Origin-Destination Survey Directives issued by the Department, although this Carrier-provided information will be kept confidential. The information retained in this manner includes, but is not limited to: (1) The Carrier's IATA Issuing Carrier numeric code, also known in the industry as the Carrier's three-digit code; (2) The Carrier's Airline Designator, also known in the industry as the Carrier's two character code; (3) The name and contact information of the Designated Company Officer who certifies the accuracy of the data; (4) The name and contact information of the Designated Carrier Liaison who resolves operational submission issues; (5) The means, method, and timing the Carrier has selected for data submission; (6) The source and accuracy statement that discloses the Participating Carrier's (a) data source, (b) data collection methodology, and (c) measures to assure data quality; and (7) The methodology the Carrier uses to convert foreign currencies into U.S. Dollars.

f. Licensed Foreign Air Carrier Participation

While foreign ownership restrictions have led the world's Carriers to share the task of transporting passengers across international boundaries, making international aviation one of the most global of industries, tremendous changes in both regulatory and business practices have dramatically reconfigured the operating and competitive structure of global aviation. Open Skies agreements, now in place between the U.S. and growing numbers of countries, are producing enormous benefits for consumers. Liberalization of air service agreements has enabled Carriers around the world to deepen their cooperative agreements with their foreign counterparts. International operations are becoming an increasingly important component of network Carrier operations. The distinctions between domestic and international networks are increasingly blurred as the interline partnerships provide seamless services through code-sharing, marketing, and strategic alliance agreements.

As a result, policy makers, international airlines, and consumers would all benefit from the capability to better understand and map global traffic flows that would promote sound public policy and business decisions. Not surprisingly, the ANPRM responses from U.S. Air Carriers advocated that their foreign-based counterparts be included in contributing data to the O&D Survey. Similarly, comments received from the nation's airports and airport consultants were unified in requesting that Foreign Air Carriers' exemption from reporting be ended. The enthusiasm with which they endorsed Foreign Air Carrier reporting is all the more pronounced because the airports, as a group, refrained from offering opinions on ANPRM topics on which they did not feel that they had sufficient expertise or that did not directly affect their needs. The Norfolk Airport Authority (Docket OST-1998-4043-31) fully endorsed a change of policy to require Foreign Air Carriers to report. Operators of larger international gateway airports made commensurately stronger statements. The City of Chicago (Docket OST-1993-4043-27) wrote, “The City strongly supports including the O&D data of Foreign Air Carriers * * *. The lack of foreign airline O&D data is arguably the greatest gap in our knowledge of the market”. When asked to list everything that would make the O&D Survey data more functional, Los Angeles World Airports (Docket OST-1998-4043-28) responded with only a single item: “collect information from all domestic and international carriers”. John Brown Company (Docket OST-1998-4043-33), an airport management consultant, wrote, “given the open-skies posture of the U.S. government toward international air service, it would be appropriate and not unreasonable to require the same standards of traffic reporting by Foreign Air Carriers operating air service at U.S. airports as for U.S. Air Carriers. U.S. airports need a complete picture of their existing air traffic flows in order to identify opportunities and develop proposals for new routes”.

Advocates of the collection of more international aviation data were not limited to Air Carriers and airports. The DOC (Docket OST-1998-4043-37) commented that, “to provide comprehensive, quality data to DOT and the industry, both U.S. flag and foreign flag carriers should be providing traffic data. Without the foreign flag data, DOT cannot truly assess the market”. ALPA (Docket OST-1998-4043-18) wrote, “In ALPA's view, one of the significant gaps in DOT's data collection system is that Foreign Air Carriers are not, as a general rule, required to file O&D data”. Comments to the ANPRM reveal that all the users of the O&D Survey data, including unions, airports, consultants, carriers, and other government agencies, agreed that the lack of Foreign Air Carrier data is a significant flaw in the usefulness of the data and that this flaw should not be underestimated. In addition to the ANPRM comments, the IG (Office of Inspector General Audit Report Number AV-1998-086) noted in its 1998 report on the O&D Survey that, “the Department is at a disadvantage in reviewing and negotiating international air route awards to ensure U.S. carriers retain competitive parity with Foreign Air Carriers”.

In the past, the Department has declined to impose the same burden of direct reporting of the O&D Survey on Foreign Air Carriers given the manual processes involved. The Department issues licenses to Foreign Air Carriers to authorize them to sell Ticketed Itineraries for travel to the U.S. as specified in 49 U.S.C. 41301, but the license does not include a responsibility to report information about the Ticketed Itineraries they issue. The Department decided to forgo knowledge about the U.S citizens that Foreign Air Carriers transport from U.S. gateway cities when the passenger does not interline on a U.S. Air Carrier. There is a special provision for reporting O&D information imposed on Foreign Air Carriers that operate under antitrust immunity granted under 49 U.S.C. Sections 41308 and 41309, but the provision only requires a Foreign Air Carrier to report the Ticketed Itineraries it issues, thus avoiding the more complicated requirements imposed on U.S. Air Carriers to report interline tickets. The data from those reporting Foreign Air Carriers, in combination with the O&D Survey reports from U.S. Air Carriers, give the Department only limited insight into the global airline industry. Furthermore, Foreign Air Carrier data are kept highly confidential and are restricted to internal Department analysis related to the monitoring of these alliances.

Instead of burdening the Licensed Foreign Air Carriers, the Department requires that U.S. Air Carriers assume the burden of obtaining the passenger information from the Foreign Air Carrier when the U.S. Air Carrier transports an interline passenger on Ticketed Itineraries issued by a Licensed Foreign Air Carrier. For example, the Department does not require Licensed Foreign Air Carriers, such as British Airways, to report the Ticketed Itineraries of its passengers transported from U.S. gateway airports, such as those in Washington or New York. However, we do require U.S. Air Carriers, such as US Airways, to report the Ticketed Itineraries of passengers that they bring from interior airports, such as those in Knoxville or Harrisburg, to the gateway airports where passengers connect to British Airways flights. Since the Carrier that transports the passenger on the international Flight-Stage is customarily the Issuing Carrier on tickets with connecting passengers, in this example British Airways, the current regulation burdens the U.S. Air Carriers with the task of obtaining O&D Survey information from these Foreign Air Carriers. By requiring the U.S. Air Carriers to report tickets issued by Foreign Air Carriers, the current regulation has been able to fully account for domestic passengers and international passengers that begin their journey at interior airports. Even so, passengers that begin their travel at U.S. gateway airports traveling on Foreign Air Carriers are missing from the current O&D Survey.

Similarly, when Foreign Air Carriers issue Ticketed Itineraries for travel to the U.S. to residents of other countries, the current regulation burdens the U.S. Air Carriers with the task of reporting those Ticketed Itineraries. For example, when SN Brussels issues Ticketed Itineraries on its ticket stock to passengers traveling to the U.S. on its ticket stock, it does so under its license to issue Ticketed Itineraries granted under the authority of 49 U.S.C. 41301. If a U.S. Air Carrier, such as American Airlines, participates in the itinerary, then the current regulation requires American Airlines to obtain a copy of the Ticketed Itinerary from SN Brussels and report it. If all of the transportation is on a non-reporting Foreign Air Carrier, such as Aer Lingus, then information about that passenger will go unreported in the O&D Survey.

Additional complexity in the current system is created because U.S. Air Carriers report Ticketed Itineraries directly to the O&D Survey while Foreign Air Carriers reporting Ticketed Itineraries under 49 U.S.C. Sections 41308 and 41309 participate in a similar, but different, program. When a reporting Foreign Air Carrier issues a Ticketed Itinerary that includes a U.S. Air Carrier in the itinerary, the current regulation requires the Foreign Air Carrier to report the Ticketed Itinerary to the alternative O&D Survey created for non-U.S. Carriers. It also requires the U.S. Air Carrier to report the same Ticketed Itinerary to the O&D Survey. Because of the dual reporting system established for the Ticketed Itineraries flown on Foreign Air Carriers, the Department must, when monitoring alliance activity, weed out the duplicates before compiling combined statistics.

If a Foreign Air Carrier, such as SN Brussels in the previous example, issues a Ticketed Itinerary to be flown on a Foreign Air Carrier required to report by agreement under 49 U.S.C. Sections 41308 and 41309, such as KLM, the passenger would go unreported because KLM is only required to report the Ticketed Itineraries for which it is the Issuing Carrier. Continuing this example, if the itinerary includes a connection to a U.S. Air Carrier, such as Northwest, at the gateway, then the Ticketed Itinerary will be reported to the O&D Survey by Northwest. If, however, a U.S. Air Carrier is not in the itinerary, then the Department will not receive this itinerary in its O&D reports. The current O&D Survey does not require SN Brussels to report the Ticketed Itinerary because SN Brussels did not transport the passenger to the U.S. Similarly, the current O&D Survey does not require KLM to report the Ticketed Itinerary because KLM did not issue that itinerary. Ticketed Itineraries are not reported with specific identifiers, and thus the Department can only presume that Ticketed Itineraries issued by Foreign Air Carriers are (1) reported twice when they are supposed to be reported twice, (2) reported once when they are supposed to be reported once, and (3) not reported when they are not supposed to be reported. Since Ticketed Itineraries are reported in aggregate, without unique identifiers, it is very difficult for the Department to verify the presumption that the Carriers are properly reporting the Ticketed Itineraries. Our presumptive dropping of duplicate itineraries on the assumption that they were reported twice adds to the uncertainty surrounding the statistics reported from the current system.

Licensed Foreign Air Carriers indirectly contribute itinerary data about their passengers. While U.S. Air Carriers use the O&D Survey in planning and marketing their services to and from the U.S., Foreign Air Carriers are at a distinct disadvantage in not being able to use this information. Confidentiality rules ban the sharing of data with non-U.S. entities. If all Licensed Foreign Air Carriers contributed data to the O&D Survey, then the confidentiality rule banning dissemination of information to Foreign Air Carriers could be lifted. This would benefit foreign entities, including Foreign Air Carriers. The anticipated further liberalization of aviation markets intensifies the need of governments and airlines for accurate traffic data as they seek to understand commercial developments and accommodate growth in international air travel. As alliances further develop and integrate, understanding their impact on non-aligned Carriers and on the industry's operating and competitive structures is increasingly more challenging. The effect of such developments as strategic alliances between U.S. and Foreign Air Carriers having antitrust immunity cannot be adequately evaluated without more complete and accurate traffic data for all Carriers. It is difficult to determine the impact of a subset of the market without an accurate picture of the whole market.

The competitive effects of these dynamic international alliances and their impact on competition, traffic flows, and aviation infrastructure cannot be effectively evaluated in isolation. Monitoring and planning both business and public policy decisions in a global network industry requires more complete data on international traffic flows between, behind, and beyond U.S. and foreign gateway airports. The global air transportation marketplace represents an important component of air transportation for U.S citizens and the U.S. economy. Having properly imposed the burden of reporting the O&D Survey on the Issuing Carrier, we are reluctant to re-impose an undue burden on U.S. Air Carriers by (1) continuing the practice of requiring them to report the O&D Survey in the current manner for Foreign Air Carrier issued itineraries and (2) requiring to report in the new manner as Issuing Carriers for their own Ticketed Itineraries. Imposing a dual reporting burden on U.S. Air Carriers would be particularly onerous because it would require continuation of all the antiquated current reporting processes in addition to instituting the new reporting processes. This scenario would further require the Participating Carrier to examine each Ticketed Itinerary to identify the appropriate reporting process for that itinerary. Even worse, it is these itineraries, issued on the ticket stock of Foreign Air Carriers, that are responsible for most of the reporting problems that occur in the current O&D Survey system. However, by not imposing the dual reporting burden, the Department would continue to miss O&D Survey information about travelers to gateway airports as well as begin to miss O&D Survey information about passengers traveling on domestic routes on itineraries issued by Licensed Foreign Air Carriers.

The Department is therefore considering requiring Foreign Air Carriers licensed under 49 U.S.C. Section 41301 to report O&D Survey data. There does not appear to be an alternative workaround that is more efficient than the simple requirement for all Issuing Carriers to report the tickets they issue for travel to and from, and within, the U.S. The Foreign Air Carriers required to report their issued Ticketed Itineraries as a condition of immunity with a U.S. Air Carrier partner have complied with this requirement and managed to adapt accordingly. The new system, designed specifically to interface with the common industry information technology infrastructure, should reduce the reporting burden for the currently reporting Foreign Air Carriers.

In addition, recent developments in the interline settlement processes would further assist Foreign Air Carriers in reporting the O&D Survey data. An alliance of Carrier-owned industry organizations—ATPCO, International Air Transport Association (IATA) and ARC—in October 2003 launched a comprehensive, global solution for financial settlement of interline travel to streamline inter-airline accounting. The interline accounting settlement service offers the possibility that Foreign Air Carriers can create a cost effective vehicle to provide the necessary data, and thus enable Foreign Air Carriers to minimize the cost of complying with the Department's reporting requirement. It is possible that combining the reporting processes with interline settlement processes will reduce the reporting burden to such a level that the cost would be far less than the benefits derived from having access to the information.

With full participation of the affected Carriers, the Department could provide access to the international data to all Participating Carriers and all stakeholders. As the largest aviation market, the U.S. is a key component in global aviation traffic flows. Complete O&D data to and from the U.S. would be an extremely valuable resource for global Carriers in planning their services. This is especially true as MIDT data, the current industry standard, decreases in utility as more bookings circumvent the GDSs. The Department seeks comment on the efficacy of requiring O&D Survey reports from Licensed Foreign Air Carriers in terms of costs and benefits and we seek comment on alternatives that would enable the Department to obtain the information it needs from Ticketed Itineraries issued by Licensed Foreign Air Carriers.

g. Charter Flights

In their responses to the ANPRM, the airports noted that passengers on non-scheduled flights merit inclusion in the O&D Survey. They observe that there are extensive public charter operations that operate on such a regular basis that differentiating a regularly scheduled charter from regularly scheduled passenger service is difficult. Even if they are a relatively small component of the national air transportation system, some charter Carriers transport a significant number of passengers to certain destinations.

Respondents have requested that these categories of passengers be counted in the O&D Survey in order to supply a complete picture of domestic and international aviation.

The Department believes that including charter Carriers would represent a considerable expansion of the scope of the O&D Survey. We further believe that doing so would most certainly impose a significant burden on small entities since charter operations generally qualify as small businesses. In addition, the advancing coverage of low cost Carriers into the markets that traditionally were most attractive to charter Carriers could potentially reduce the number of passengers charter services transport, further reducing the impact of charter services on the national transportation system. In light of this, we do not propose to expand the scope of the O&D Survey to include charter services, but we invite further comment on this issue.

h. Reporting by Flight-Stage

Several respondents to the ANPRM commented on inconsistencies that are allowed to exist in the O&D Survey because of funnel flight and starburst flight situations. American Airlines (Docket OST-1998-4043-5) noted that the root of the inconsistency is the generally accepted, albeit little known, practice of reporting single flight segments with multiple Flight-Stages as if they were a single flight segment with one Flight-Stage. For example, a passenger traveling from Washington Dulles (IAD) to Los Angeles International (LAX) might travel on a non-stop flight, represented as IAD-LAX. However, another passenger might travel from Washington to Los Angeles on a direct one-stop by way of St. Louis under a single flight number and a single flight coupon. Since the passenger in the second example does not deplane in St. Louis, both example itineraries will be reported as IAD-LAX in the O&D Survey.

The Department believes that checking the congruency of the O&D Survey with the T-100/T-100(f) is the best method of verifying the accuracy of both sets of data. Since the Ticketed Itineraries that describe nonstop travel are indistinguishable from Ticketed Itineraries that describe one-stop or two-stop travel, checking the O&D Survey against the statistics in the T-100/T-100(f) is very difficult. For example, passengers can be routed from Washington Dulles to Los Angeles International by way of any of a dozen or more airports. Each Ticketed Itinerary will describe that one-stop travel as IAD-LAX to the O&D Survey but as the actual route in the T-100. In this same way, one-stop and two-stop travel is available in practically all of the airports in the U.S. and in foreign countries. The Department must collect O&D Survey data on a stage-by-stage basis (wheels up to wheels down) rather than the current coupon-by-coupon basis (passenger enplanement to passenger deplanement) in order to attain the desired congruency with the T-100/T-100(f).

This change in reporting requirements will have minor impact on those Carriers that store information about the intermediate stops that exist in the passengers' Ticketed Itineraries. Carriers that do not store information about the intermediate stops that their customers are making will have to either retain that information from the passengers' flight reservations or re-acquire the information from a source of flight schedule data such as that provided by the Official Airline Guide (OAG). In its ANPRM comments, the OAG (Docket OST-1998-4043-43) offered its services in determining the identity of Franchise Code-share Partner Carriers and we believe that their services or those of other organizations could be similarly utilized to determine information about intermediate stops.

To obtain the highest level of accuracy when knowledge of hidden intermediate stops must be re-acquired, that process should take place in a time frame commensurate with the creation of the Ticketed Itinerary. Flight schedules change over time, and the shortest possible time lag between the creation date of the Ticketed Itinerary and the time when knowledge of intermediate stopping is re-acquired will provide the fewest possible instances of flights not found in the schedule data.

The missing Flight-Stage information has significant effect on the quality and reliability of the information required and disseminated by the Department. Therefore, we propose to collect data on a Flight-Stage basis rather than the current Flight-Coupon Stage basis. We seek comments from the industry and the public regarding how the Flight-Stage Origin Airport and Flight-Stage Destination Airport should be determined.

i. Data Retention

The Department's policy on data quality recognizes that no data system is free of data errors. The Department must have the means of redressing a problem found in the data quality. The data submitted under the provisions of the proposed O&D Survey and the T-100/T-100(f) will be subject to regulations under 14 CFR Part 249—Preservation of Air Carrier Records. The Department's procedure concerning the requests for correction of information gives stakeholders the right to request correction of information disseminated by the Department.

5. Transition Period

The Department proposes to establish a transition period, also known as concurrent processing, between initialization of the proposed O&D Survey and discontinuation of the current O&D Survey. During the transition period, the Department will begin collecting data under the rules of the new O&D Survey. The transition period will consist of a test phase for initial testing, sometimes called unit testing, and a test phase for large volume testing, sometimes called system testing. The current survey must continue to be produced during both phases of the transition to the new system.

There are two primary objectives for the transition period. The first is to ensure that the data being reported under the new system are accurate, complete, and comparable across Carriers using different internal accounting systems. The second objective is to ensure, to the extent possible, the relative comparability between data submitted under the current O&D Survey and data submitted under the proposed O&D Survey. Many stakeholders rely on the Department's aviation traffic data to discern broad trends in services, fares, and capacity. The modernization of aviation data must therefore ensure that the ability to use the data to perform such critical time series analyses is preserved both in terms of the databases maintained by the Department as well as in the traffic data products it disseminates. Time series analyses are required for critical government and business decisions, which are predicated on identifying and understanding trend changes. We believe we can preserve time-series continuity by disseminating the same data in both formats, helping the users assess the full impact of the change in the O&D Survey and, thereby, mitigating the need for a long transition period collecting data under dual systems. Because continued integrity in data collected in the current system is crucial to the testing of the new system, reduced attentiveness to reporting accuracy on the part of the current Participating Carriers may lengthen the transition period.

The need for concurrent processing is self-evident. Statistics must continue in the current format while the new statistical system is being tested and validated. During the test phase of the transition period, the Department will begin accumulating data from all Participating Carriers and correlate that data with data from the enhanced T-100/T-100(f). Meanwhile, data continuity will be preserved with continued O&D Survey submissions under the current rule. The Department will be accepting data from a variety of systems and we anticipate that it will take some time to establish communications and data validity checks appropriate for each Carrier.

In addition to testing the quality of the data received from each Carrier, the Department will use the time in the test phase to accumulate data that will be necessary for the commencement of the large volume testing phase. Since Ticketed Itineraries are purchased in advance of travel date, data must necessarily be collected over the length of time each Carrier allows for advance purchase. For example, Carriers with a four-month advance purchase availability, or booking window, would provide full test data for the four months to accumulate a full set of passenger data for the Department to test. Carriers with an 11-month booking window, however, would send the appropriate data for 11 months. The Department cannot begin conducting meaningful overall comparisons between the data from the current O&D Survey and the proposed system until it has accumulated data over the length of the advance booking windows.

Once the Department is satisfied that 100 percent of the data from each Participating Carrier has been collected and processed, the second phase of the transition can begin. During this full-volume testing phase, the Department will evaluate the new stream of data over time to ensure that the methodology and technology are robust, after which the old system can be shut down.

Users of O&D Survey data will require a period in which they can understand the impact of the change in data and data processes by comparing the results of the new O&D Survey with the existing O&D Survey. This continuity is equally important for Participating Carriers since Carriers are users as well as suppliers of data. The Department is aware of the advantages of a long full-volume testing phase, but we are also aware that these advantages come at the cost of running two data collection systems in parallel. We acknowledge that requiring the Carriers to supply data for two systems simultaneously will require extraordinary efforts on their part. Recognizing the burden to file data under both reporting systems, the Department wishes to minimize the length of the second transition phase. However, we acknowledge that data suppliers have many constraints and data users have many data testing needs of which we are unaware. Therefore, the Department seeks comment regarding the proposed length of the second transition phase.

J. T-100/T-100(f) Considerations

The T-100/T-100(f), consisting of Form 41, Schedule T-100—U.S. Air Carrier Traffic and Capacity Data by Nonstop Segment and On-flight Market and Schedule T-100(f)—Foreign Air Carrier Traffic Data by Nonstop Segment and On-flight Market, contains monthly segment and market traffic data (Part 217). The proposed changes to the O&D Survey will provide the Department with information about the numbers of passengers scheduled to use the air transportation system by flight and by day, but the proposed NPRM does not provide any capability, except when aggregated to the month of travel, to cross check the scheduled passengers with the actual passengers carried on the aircraft. The Department is considering modifying the T-100/T-100(f) to enable us to validate the data that will be collected under the O&D Survey to ensure the data's accuracy.

1. Background

The T-100/T-100(f) collects summarized flight stage data and on-flight market data. The Reporting Carriers collect these traffic statistics for each revenue Flight-Stage as actually performed and compile them for reporting to the Department. Since the statistics are collected by counting the people who board an aircraft, nothing can be known about other flights the passenger may have taken prior to boarding that aircraft and nothing can be known about flights the passenger may be taking as part of the same itinerary subsequent to disembarking from that aircraft. Significantly, nothing can be known about what the passenger paid for the transportation on the current aircraft. The Carriers collect this information on each Flight-Stage departure each day, and at the end of the month, they summarize it by (1) equipment type, (2) class of service, and (3) airport pair, all without regard to individual flight number for the month.

2. T-100/T-100(f) Changes To Be Considered

The O&D Survey, in contrast to the T-100/T-100(f) report of actual passengers boarded, collects copies of the passenger's scheduled itinerary. O&D Survey passenger reports are copied and reported after the passenger's initial departure on that Ticketed Itinerary. Since the bulk of the passenger's itinerary has not yet been flown at the time of initial departure, the O&D Survey collects information about itineraries as they are scheduled to be performed, not as they are actually performed. As has been previously described in this rulemaking, two significant features of the O&D Survey are (1) the information about the passenger's connecting flights that enable users to obtain a sense of the passenger's true origin and true destination and (2) the information about the fare that the passenger paid that enable users to assign a value to air transportation. The contrasting differences, between the narrow source of information about passengers that are actually transported and the robust source of information about passengers that are scheduled to be transported, make the T-100/T-100(f) and the O&D Survey ideal companion data products that the Department makes available to the industry and the public.

Making the changes to the O&D Survey proposed in this rulemaking without making commensurate changes in the T-100/T-100(f) would leave the two data collection systems focused on two different levels of aggregation and would severely limit the advantages now enjoyed by having companion data products. The current O&D Survey is validated by knowledge of the established relationships between passengers scheduled to fly between a set of airport pairs and passengers actually on board flights between those airport pairs. The proposed revisions allow the users of the O&D Survey to have knowledge of passengers scheduled to fly between airports by time-of-day and day-of-week, which is a level of detail that the T-100/T-100(f) does not possess. Without commensurate changes in the T-100/T-100(f), the desired match between the O&D Survey and the T-100/T-100(f) data will be limited to highly aggregated monthly comparisons. The Department is concerned about its inability to validate the receipt of flight date and flight number elements into the O&D Survey as proposed in this rulemaking. For example, one of the most important new features of the O&D Survey is the ability to disseminate data by One-way Trips. The Department's ability to validate the data that goes into deriving the One-way Trips is dependent on getting commensurate robust T-100/T-100(f) information by flight and by date.

In addition to the need to keep the data congruent for validation purposes, knowing the on-board count of passengers by flight and by date on the T-100/T-100(f) would be helpful for the Department in planning airport capacity expansion. The usefulness of knowing the passengers flying between airports for an entire month is limited to long range planning functions. For example, the FAA would use the T-100/T-100(f) in long-range planning where trends measured to the nearest month are useful. The data would be more useful if it included details that could help with facility planning by time-of-day and by day-of-week.

The Department has provided information about the costs and benefits of collecting and disseminating the T-100/T-100(f) data by flight and by day (See section L(3)). Preserving data validity and accuracy by flight and by day by coordinating the O&D Survey with the T-100/T-100(f) to the highest degree practicable will benefit the Department and the public. To this end, the Department is considering the collection of T-100/T-100(f) data by Master Flight Number and by flight date. We seek comments on the efficacy of this possible course of action.

K. Data Dissemination

The Department proposes to continue to disseminate O&D Survey products from the data collected under this rulemaking to serve the needs of various stakeholders in the aviation community. If the significant enhancements proposed in this rulemaking were adopted, these products would be substantially richer in content, more timely, and more accurate than the products disseminated under the current system. While it would be premature to identify the precise nature and format of such products, they would certainly not be less detailed than the data products disseminated under the current system, including dissemination of data by itinerary, within the constraints of Vision 100 regarding flight-specific data. We have spent considerable effort to understand the data needs of various user groups and recognize that different users have diverse requirements in terms of the level of data granularity most suitable to their needs. The Department therefore seeks detailed comments and suggestions on aviation data products, based on our proposed changes, that would satisfy the various needs of different types of users.

We recognize that, in order to be able to comment effectively, interested parties require further information on key methods that will be applied to the data, particularly those which will be used to determine a passenger's True O&D using the industry standard One-way Trip methodology. Among these important methods are: (1) Dissemination of data by month according to the scheduled flight date, (2) grouping of flights by One-Way Trip instead of by Directional Passenger trip, and (3) reporting the fare obtained by the Carrier(s) using an industry standard proration methodology rather than relying on the current practice of reporting the total fare amount collected with the total itinerary. The processes by which data are collected and disseminated affect the accuracy of those data. Since such methods define the utility of the fundamental data elements, we outline our proposals in each of these areas in detail. We seek comment on our proposed methodology, the resulting aviation data products, and the composition of these disseminated products.

1. Dissemination of Data by Month

The Department has heretofore disseminated all data about travel in the quarter in which it was reported. Although the Department proposes to continue to collect Ticketed Itinerary data on a ticket basis in the month it is first used for travel, we propose to disseminate the data on the basis of the month in which travel is scheduled to take place. This dissemination is made possible because the proposed rule expands the data collected for each Ticketed Itinerary. At this time, we are considering disseminating data by month in at least two formats: (1) The Ticketed Itinerary (similar to the DB1B Ticket file) and (2) the One-way trip (similar to the DB1B Market file) aggregations, subject to Vision 100 constraints on the dissemination of flight-specific data. To create a market file, the Department proposes to separate the Ticketed Itinerary into One-way Trips, allocate the itinerary fare to the One-way Trips, and store the One-way Trips for dissemination at the appropriate time. The Scheduled Flight Date of the first Flight-Stage in a One-way Trip will serve as the flight date for that One-way Trip. We seek comment about the construction and dissemination of these data products.

2. Proposed Construction of One-Way Trips

As explained in the proposed data elements discussion (Section I.2.a.—O&D Survey Redesign: Discussion of the Proposed O&D Survey) for the One-way trip format, each Ticketed Itinerary will be divided into a series of one or more One-way Trips according to the guidelines published in the final rule. We anticipate basing these guidelines on industry consensus and seek comment about methods of constructing One-way Trips.

The Department proposes to use four hours in an airport as the maximum amount of time to consider that airport as a connecting airport in a domestic U.S. airport to U.S. airport itinerary, or between a U.S. airport and an airport in either Canada or Mexico. The Department proposes to use 24 hours in an airport as the maximum amount of time to consider that airport as a connecting airport in a Ticketed Itinerary for international travel.

3. Proposed Proration Method

The current O&D Survey is published on a Ticketed Itinerary basis. The amount collected is summed for the itinerary. In the proposed One-way trip format, the Department will divide the Ticketed Itinerary into One-way Trips. To perform meaningful analysis, the fare amount must be allocated to the One-way Trips in an equitable manner. The industry term for the process of allocating the fare to the One-way Trips is proration.

Four proration techniques are widely used in the industry: (1) Straight rate prorate, (2) international prorate factors, (3) mileage, and (4) square root of the miles. Each has advantages and disadvantages. Straight rate prorate methodology compares, for each itinerary, the Carrier's unrestricted fares, for each local Flight-Coupon Stage, that are in effect when the Ticketed Itinerary is issued to the total fare collected. A ratio is established between all the Flight-Coupon Stages using the unrestricted local fares and the resulting ratios are applied to the fare that was actually collected for the itinerary being processed. In international prorate factors, instead of looking up the fares to establish a ratio, the ratios are already established and they are referenced and applied. In mileage prorate, the ratio is obtained by using the number of miles distant between airports. In the square root of the miles methodology, the ratio used for dividing the fares is established by using the square root of the number of miles distant between cities.

Unlike a Carrier that can chose a proration method that is most advantageous to its own situation and needs, the Department is constrained by its requirement to be able to apply one technique with equanimity for all Carriers across all conceivable itineraries. Further, the Department is constrained by a requirement that its processes be repeatable (i.e., a Ticketed Itinerary processed through the system today must provide the same result as it will if processed again several months later). Since straight rate prorate and international prorate factors require inputs from outside systems that change over time, the Department would have to keep copies of all possible permutations of those inputs by day in order to meet the repeatable standard. This would clearly be costly, and in light of other available proration methods, excludes these methods from further consideration.

The mileage and square root of the miles methodologies have a distinct advantage, because the miles between airports change very rarely. In the previous decade, only the opening of a new airport in Denver and the relocation of the terminal in Pittsburgh have had an effect on the number of miles between airports in the U.S. The Department considers this to be an acceptable level of variance inherent in these two proration techniques. Of the two, the Department prefers the square root of the miles methodology over a mileage proration methodology. When there are two Flight-Stages in a trip, and the Flight-Stages are of equal distance, both techniques will allocate half the money to each leg. When there are two Flight-Stages of a trip, and one stage length is significantly longer than the other, mileage allocates the short stage length a miniscule amount of the fare while square root of the miles allocates a bit more and tends to be more consistent with prorate agreements between Carriers.

For example, in a hypothetical 850-mile trip with two Flight-Stages that are 425 miles distant, both techniques will give each 425-mile stage one half of the fare amount. In another hypothetical 850-mile trip with one flight stage of 729 miles and one of 121 miles, the mileage prorate gives 85.8 percent of the fare amount to the longer leg and 14.2 percent to the shorter stage. The square root of the miles on that same itinerary gives the longer stage 71 percent of the fare amount while the shorter stage gets 29 percent. The square root of the miles prorate calculation mimics typical Carrier revenue allocations more closely than does the mileage prorate.

The Department seeks comment on the best practices in the application of proration methodology in the scheduled air transportation industry. Respondents that advocate a methodology other than the one proposed by the Department, the square root of miles, must consider in their recommendation the Department's constraints: (1) The methodology must treat all carriers with equanimity and (2) the methodology must be repeatable.

4. Proposed Changes to Confidentiality

One of the most critical elements of the Department's proposed changes to the O&D Survey involves addressing data confidentiality. The current O&D Survey data confidentiality rules (14 CFR Sec 19-7(d)) exist to preclude international data from being disclosed since Foreign Air Carriers were excluded from reporting. Domestic data in the current O&D Survey are released in full after a certain period of time elapses.

In its response to the ANPRM, the Allied Pilots Association (Docket OST-1998-4043-16) pointed out that the time lags under the current O&D Survey reduce the usefulness of the data. There was a divergence of opinion on how long the data should remain confidential, but most advocated a short confidentiality period for all data. No respondent registered strong disapproval of a short confidentiality period. Short confidentiality periods were endorsed by Airports Council International—North America (Docket OST-1998-4043-6), American Airlines (Docket OST-1998-4043-5), Continental Airlines (Docket OST-1998-4043-26), and Metropolitan Washington Airports Authority (Docket OST-1998-4043-38). The Air Line Pilots Association (Docket OST-1998-4043-18) said the data should be released no later than 6 months after the report date. Respondents that went on record to say that the confidentiality period should not be greater than six months are Delta Air Lines (Docket OST-1998-4043-21), Oakland International Airport (Docket OST-1998-4043-14), BACK Associates, Inc. (Docket OST-1998-4043-3), John Brown and Company (Docket OST-1998-4043-33), Los Angeles World Airports (Docket OST-1998-4043-28), Port Authority of New York and New Jersey (Docket OST-1998-4043-25), Port of Portland (Docket OST-1998-4043-19), and Wayne County and Detroit Metropolitan Airport (Docket OST-1998-4043-23).

Any changes to the present reporting system must satisfy the statutory requirements of Section 805 of Vision 100—Century of Aviation Reauthorization Act (Pub. L. 108-176; 117 Stat. 2490). Section 805 mandates that, if the Secretary requires Carriers to provide flight-specific information, the Department will not: (1) Make public the flight-specific fare information until at least nine months after the flight date and (2) issue a rule requiring public dissemination of flight-specific fare information without giving due consideration to and addressing the Carriers' confidentiality concerns.

The Department recognizes that Carriers will view flight-specific fare information as “sensitive,” in that a competitor could potentially exploit this information in the marketplace. The Department also recognizes that, when combined with other data elements, the combined data elements could raise certain competitive confidentiality concerns. The Department believes there exists a wide range of opinion about data elements that should be withheld from public dissemination and the appropriate holding period. The Department's initial position is that, while it may be appropriate to withhold some of the new data elements from public dissemination for a time, all data should eventually be released into the public domain. We seek comment regarding the timing of the release of flight-specific fare information.

The Department is cognizant of the sensitive nature of any data element that could be used to identify any specific individual passenger. No data requested in this rulemaking will include any personal information on a specific passenger that would enable the identification of a specific individual. We have declined to propose collection of any of the elements that were suggested in ANPRM comments as point of sale identifiers (these are Passenger Citizenship, Phone Number, and Zip Code/Postal Code.) Furthermore, if the Department were to collect the any of these elements, it would never release any data that could be used to identify an individual passenger. The Department will only use such data for statistical purposes. These passenger data will be protected under the Confidential Information Protection and Statistical Efficiency Act of 2002 (CIPSEA), which appears as Title V of the E-Government Act of 2002. We invite comment from the industry and public on issues of confidentiality of passenger information.

The expanded amount of information that the Department proposes to collect is required to fulfill the Department's statutory mandates. However, the O&D Survey information to be disseminated to the public has not yet been fully determined. We anticipate releasing data that are of immediate economic value, but do not disclose competitive positions, as soon as the data are received and processed for dissemination, subject to the constraints mandated by law. The Department seeks comment on a proposal to release aggregated data on a monthly basis in the shortest possible time needed to process the data. We are also requesting public comments on whether certain, and if so which, data elements should be withheld from public dissemination and the appropriate holding period. We invite comment from the industry regarding public dissemination of flight-specific fare information according to the provisions of Vision 100—Century of Aviation Reauthorization Act.

L. Rulemaking Analyses and Notices

In order to increase efficiency and effectiveness; improve the integrity, quality, and utility of the information available; and reduce information collection costs to the Carriers; the Department proposes to modernize its data collection products. The legal authority for the proposed rule is provided by the Civil Aeronautics Board Sunset Act of 1984 (Pub. L. 98-443), which requires the Department, under the authority of the Secretary (49 U.S.C. 329(b)(1)), to collect and disseminate information on civil aeronautics and aviation transportation in the U.S., other than that collected and disseminated by the National Transportation Safety Board. The Department must, at minimum, collect information on the origin and destination of passengers and information on the number of passengers traveling by air between any two points in air transportation. Additionally, the Department must be responsive to the needs of the public and disseminate information to make it easier to adapt the air transportation system to the present and future needs of commerce of the U.S. (49 U.S.C. 40101(a)(7)). In meeting this responsibility, the Department collects data submitted under 14 CFR Part 217 (Reporting Traffic Statistics by Foreign Air Carriers in Civilian Scheduled, Charter, and Nonscheduled Services), 14 CFR Part 241 (Uniform System of Accounts and Reports for Large Certificated Air Carriers) and 14 CFR Part 298 (Exemptions for Air Taxi and Commuter Air Carriers).

The purpose of the proposed rule is to improve the accuracy and utility of reported traffic data while reducing the burden on the Carriers. For the O&D Survey, this objective is achieved by replacing 14 CFR Part 241 Section 19-7 with Section 26, which modifies the set of existing data elements, revises reporting time frames, and redefines the set of Carriers that report the O&D Survey in accordance with industry standards and practice. We are considering changes to the T-100/T-100(f) to enhance congruency between the O&D Survey and the T-100. The changes we are considering would amend 14 CFR Part 241 Section 25, thus modifying the set of existing data elements reported on the T-100 and amend 14 CFR Part 217 Section 5, thus modifying the set of existing data elements reported on the T-100(f).

The proposed modernization of the Department's aviation data will bring the data gathering process into alignment with current airline industry accounting practices. It will provide more accurate, more timely, and more complete data for all stakeholders. Furthermore, it is the least intrusive informational alternative sufficient to accomplish the statutory objective of gathering accurate information about air travel. The proposed rule has been evaluated under the following Acts, Executive Orders, and Departmental Policies. We seek comment from interested parties about the rulemaking analyses contained in this section.

1. Affected Carriers

The Carriers that would, under the proposed changes to the O&D Survey, be required to report the O&D Survey are those defined in Section I.3. (O&D Survey Redesign: Reporting Requirements) as Participating Carriers. These Participating Carriers are (1) U.S. Air Carriers that issue tickets for travel on scheduled interstate passenger services to or from, or within, the U.S. and operate aircraft with 15 seats or more for scheduled service and (2) Foreign Air Carriers that operate under 49 U.S.C. Sections 41308 and 41309 and are required, under the grant of antitrust immunity, to report itineraries involving a U.S. point. The group of Participating Carriers consists of Currently Participating Carriers and Newly Participating Carriers. Because the proposed rule changes the criteria defining which Carriers shall report the O&D Survey, there will be 38 Participating Carriers (25 U.S. Air Carriers, versus the 34 U.S. Air Carriers that submitted the O&D Survey in Third Quarter 2003, and 13 Foreign Air Carriers) under the proposed rule, compared to 47 Carriers under the current rule.

Currently Participating Carriers are those U.S. Air Carriers and Foreign Air Carriers that report the O&D Survey under the current rule and would continue to report the O&D Survey under the proposed rule. Newly Participating Carriers are (1) those U.S. Air Carriers that do not currently report the O&D Survey but would begin to report under the proposed rule and (2) those Foreign Air Carriers that would report the O&D Survey if they operate under antitrust immunity pursuant to 49 U.S.C. Sections 41308 and 41309 for alliance(s) with U.S. Air Carrier(s). In addition, under the proposed rule, 13 U.S. Air Carriers that currently report the O&D Survey would no longer be required to report. These carriers are identified as Formerly Participating Carriers.

The Department is considering modifying the data elements reported by U.S. Air Carriers on the T-100 and by Foreign Air Carriers on the T-100(f). The additional data elements being considered would, in combination with the proposed changes to the O&D Survey, enhance the validity and reliability of the Department's aviation data and benefit all stakeholders. We have included the regulatory impact of the potential changes to the T-100/T-100(f) in this section, although we note that these changes have not been specifically proposed within this NPRM.

The Department is also considering requiring Foreign Air Carriers that: (1) Are licensed to hold out service to the U.S. under 49 U.S.C. Section 41301; (2) do not have antitrust immunity for an alliance with a U.S. Air Carrier; and (3) operate aircraft with 15 seats or more for scheduled service to or from, or within, the U.S. to report all itineraries involving a U.S. point to the O&D Survey. At this time, we have not included these Foreign Air Carriers in the Regulatory Analyses contained in Section L. We seek comment on the costs and benefits of including in, or excluding from, the O&D Survey data from these Foreign Air Carriers.

Table 1.—Carriers Affected by Proposed Changes to the O&D Survey

	Continue to report (currently participating carriers)	Begin to report (newly participating carriers)	No longer required to report (formerly participating carriers)
U.S. Air Carriers	21	4	13
Foreign Air Carriers	13	0	0
Total Carriers	34	4	13

The Carriers that would, under the changes we are considering to the T-100/T-100(f), be required to report the T-100/T-100(f) are those defined in Section J.1. (T-100/T-100(f):—Background) as Reporting Carriers. Because the proposed rule does not alter the definition of Reporting Carrier, no Carriers would be added as Reporting Carriers based solely on the possible changes to the T-100/T-100(f). There were 282 Reporting Carriers in Third Quarter 2003. However, 52 of those Carriers are all-cargo Carriers. Because the additional data elements being considered for the T-100/T-100(f) are flight-specific and would be used, in part, to match the O&D Survey to the T-100/T-100(f), all-cargo Carriers would not have to report these elements. The changes that we are considering making to the T-100/T-100(f) would, therefore, affect the remaining 230 Reporting Carriers (121 U.S. Air Carriers and 109 Foreign Air Carriers) that are not all-cargo Carriers.

Table 2.—Carriers That Would Be Affected by Changes Being Considered for T-100/T-100(f)

	Continue to report (currently reporting carriers)	Begin to report (newly reporting carriers)	No longer required to report (formerly reporting carriers)
U.S. Air Carriers	121	0	0
Foreign Air Carriers	109	0	0
Total Carriers	230	0	0

2. The Unfunded Mandates Reform Act of 1995

The Unfunded Mandates Reform Act of 1995, codified at 2 U.S.C. 1531-1538, requires Federal agencies to prepare a written assessment of the costs, benefits, and other effects of proposed or final rules that include a Federal mandate likely to result in expenditures by State, local, or tribal governments, in the aggregate, or by the private sector, of more than $100 million annually.

The proposed changes to the O&D Survey and the changes we are considering making to the T-100 would not result in expenditures by State, local, or tribal governments because no such government operates a Carrier subject to the proposed regulation. While the proposed changes to the O&D Survey and the changes we are considering making to the T-100(f) will affect Foreign Air Carriers, some of which are operated (in whole or in part) by foreign governments, the Unfunded Mandates Reform Act of 1995 does not apply to foreign governments.

3. Regulatory Evaluation

a. Executive Order 12866, Regulatory Planning and Review

Executive Order 12866, Regulatory Planning and Review (58 FR 51735; September 30, 1993) defines a significant regulatory action as one that is likely to result in a rule that may have an annual effect on the economy of $100 million or more or adversely affect, in a material way, the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities. Regulatory actions are also considered significant if they are likely to create a serious inconsistency or interfere with the actions taken or planned by another agency or if they materially alter the budgetary impact of entitlements, grants, user fees, or loan programs or the rights and obligations of the recipients of such programs.

The proposed changes to the O&D Survey are estimated to collectively cost U.S. Air Carriers approximately $1.3 million in the first year, including initial costs and annual reporting costs, and approximately $281,000 each year thereafter. If these changes are not made, the collective reporting costs to U.S. Air Carriers are estimated to be approximately $509,000 each year. When Foreign Air Carriers that operate under 49 U.S.C. Sections 41308 and 41309 and are required, under grant of antitrust immunity, to report itineraries involving a U.S. point are included, the proposed changes to the O&D Survey are estimated to collectively cost the world airline industry approximately $1.9 million in the first year, including initial costs and annual reporting costs, and approximately $427,000 each year thereafter. If these changes are not made, the collective reporting costs to the world airline industry are estimated to be approximately $704,000. Thus, if we make no changes to the current O&D Survey, we will continue to collect data under that rule. The collective annual costs to U.S. carriers will continue to be approximately $509,000 per year and the collective annual costs to the world airline industry will continue to be approximately $704,000. Table 3 compares the annual costs of the proposed changes to the O&D Survey to the annual costs of continuing the current O&D Survey collection. These costs are further detailed in Tables 8 and 9.

Table 3.—Collective Costs for U.S. Air Carriers and World Airline Industry Proposed Changes Versus Current Rule O&D Survey

	First year collective costs (including initial costs)	Subsequent year collective costs
Proposed O&D:
U.S. Air Carriers	$1,273,110	$280,800
World Airline Industry	1,915,336	426,816
Current O&D:
U.S. Air Carriers	509,184	509,184
World Airline Industry	703,872	703,872

The changes that we are considering making to the T-100/T-100(f) are estimated to collectively cost U.S. Air Carriers approximately $1 million in the first year, including initial costs and annual reporting costs, and approximately $204,000 each year thereafter. If these changes are not made, the collective reporting costs to U.S. Air Carriers are estimated to be approximately $159,000 each year. When Foreign Air Carriers are included, the changes that we are considering making to the T-100/T-100(f) are estimated to collectively cost the world airline industry approximately $1.9 million in the first year, including initial costs and annual reporting costs, and approximately $387,000 each year thereafter. If these changes are not made, the collective reporting costs to the world airline industry are estimated to be approximately $301,000. Thus, if we do not make the changes to the T-100/T-100(f) that we are considering, we will continue to collect data under the existing rule. The collective annual costs to U.S. carriers will continue to be approximately $159,000 per year and the collective annual costs to the world airline industry will continue to be approximately $301,000. Table 4 compares the annual costs of the changes to the T-100/T-100(f) that we are considering making to the annual costs of continuing the current T-100/T-100(f) collection. These costs are further detailed in Tables 10 and 11.

Table 4.—Collective Costs for U.S. Air Carriers and World Airline Industry Considered Changes versus Current Rule T-100/T-100(f)

	First year collective costs (including initial costs)	Subsequent year collective costs
Proposed:
U.S. Air Carriers	$1,002,460	$203,860
World Airline Industry	1,905,503	387,503
Current:
U.S. Air Carriers	158,559	158,559
World Airline Industry	301,392	301,392

Because the proposed changes to the O&D Survey and the changes we are considering making to the T-100/T-100(f) will not collectively cost members of the private sector more than $100 million in the first year of effectiveness under the proposed rule, the Department finds that the changes would not, collectively or separately, place a significant burden on the world-wide airline industry. The Department also finds that the benefits of the proposed changes outweigh the potential costs. Therefore, the proposed rule should not be considered an economically significant regulatory action under Executive Order 12866. However, regulatory actions that raise novel legal or policy issues can be considered significant. Because the proposed changes to the O&D Survey, as well as the changes we are considering for the T-100/T-100(f), change the collection procedures of influential aviation data, this NPRM is considered a significant regulatory action under Executive Order 12866 and was reviewed by the Office of Management and Budget.

Net Present Value Analysis. The current rule is expected to cost approximately $1 million each year. The cost of the current O&D Survey is estimated by multiplying the average annual reporting burden of 960 hours per reporting Carrier by an estimated hourly wage of $15.60. The total burden, for the 47 Carriers that report the O&D Survey under the current rule, is $703,872. The cost of the current T-100/T-100(f) is estimated by multiplying the average annual reporting burden of 84 hours per reporting Carrier by an estimated hourly wage of $15.60. The total burden for the 230 Carriers that report the T-100/T-100(f) under the current rule is $301,392.

As shown in Tables 8, 9, 10, and 11, the proposed changes to the O&D Survey and the changes we are considering making to the T-100/T-100(f) are expected to cost the affected Carriers approximately $3.82 million in the first year and $814,320 in each subsequent year. That is, while the proposed changes to the O&D Survey and the changes we are considering making to the T-100/T-100(f) will require a one-time investment of about $3.82 million, annual reporting costs for the initial and subsequent years would decrease, collectively by about $71,000 per year and individually by about 240 hours per Carrier.

Table 5, below, shows the present value costs, using a 7 percent discount rate, under (1) the current rule, (2) the proposed rule, and (3) the proposed rule if Carriers engage in one year of concurrent processing. As discussed in Section I.5. (O&D Survey Redesign: Transition Period), a transition period may be required. During that time, both Formerly Participating Carriers and Currently Participating Carriers would report under the current rule, while Currently Participating Carriers and Newly Participating Carriers would also report under the proposed rule. For the purposes of present value cost analyses, we estimate a concurrent test period of one year.

Table 5.—Estimated Present Value Costs

[Including changes being considered for the T-100/T-100(f).]

Elapsed time	Current rule	Proposed rule	Proposed rule (with 1 year concurrent processing)
5 Years:
Total Present Value Cost	$4,121,781	$6,148,705	$7,088,204
(incremental cost over current rules)		+ 2,026,924	+2,966,423
10 Years:
Total Present Value Cost	7,060,544	8,529,275	9,468,774
(incremental cost over current rules)		+ 1,468,731	+ 2,408,230
15 Years:
Total Present Value Cost	9,155,858	10,226,588	11,116,087
(incremental cost over current rules)		+ 1,070,730	+ 1,960,229
20 Years:
Total Present Value Cost	10,649,781	11,436,749	12,376,249
(incremental cost over current rules)		+ 786,968	+ 1,726,468

The initial reporting burden associated with the proposed changes to the O&D Survey and the changes considered for the T-100/T-100(f) results in higher present value costs. However, the benefits to Participating Carriers and Reporting Carriers, as well as to the Department, Federal agencies, airports, consultants, academics, State and local transportation planners, other State and local agencies, consumers, and other stakeholders, are significant and immediately available (See Sections L.3.d.2. and L.3.e.2.). Because these benefits are less readily quantifiable, Table 6 contains the present value benefits, using a 7% discount rate, under three possible scenarios, for the proposed rule.

The first scenario assumes a total annual benefit, as a result of the proposed and considered changes, of $250,000 per year. If the Participating Carriers were assumed to be the sole beneficiaries, each would, under this very conservative scenario, receive annual benefits of about $6,600 a year. We believe that information about 100 percent of Ticketed Itineraries issued for travel to or from, or within, the U.S. by U.S. Air Carriers operating aircraft with 15 seats or more is likely worth much more than approximately $7,000 per year. In fact, we are certain that the cost to purchase this degree of information, for a 12-month period and from a GDS or other source not based on the O&D Survey, would be considerably more expensive. Again, if we assume the only beneficiaries to be the Participating Carriers, the second scenario would attribute annual benefits to those 38 Participating Carriers of about $13,200 per year. Based on our knowledge of non-Departmental data sources, we find this estimated benefit to be conservative.

We find the third scenario, total annual benefits of $1,000,000 for all stakeholders, to be more realistic. This estimate is the equivalent of about $27,000 of annual benefits per stakeholder if only the 38 Participating Carriers are considered. Furthermore, submission of 100 percent of Ticketed Itineraries by Participating Carriers significantly reduces the likelihood that the Department will need to request supplemental data about markets not represented in the O&D Survey. Participating Carriers will be able to apply resources previously dedicated to supplemental data request to other internal priorities. Assigning an estimated total annual benefit of $1,000,000 per year only to Participating Carriers, however, ignores the benefits to the Department's regular analyses of competition in the aviation industry and its EAS and Small Community Air Service Development Program. In addition, we have not enumerated the annual benefit, to the FAA, DOJ, DOS, DOC, DHS, BLS, and other Federal agencies and programs, of having 100 percent of Ticketed Itineraries issued by Participating Carriers.

Therefore, we base our assessment of the costs and benefits of the proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) on the moderate estimate of $1,000,000 of total annual benefits for all stakeholders. We seek comment about the estimated benefits, for individual stakeholders as well as collectively, used in this regulatory evaluation.

Table 6.—Estimated Present Value Benefits Under Proposed O&D Survey

[Including changes being considered for T-100/T-100(f).]

Time period	Estimated total annual benefits for all stakeholders
	Very conservative $250,000 per year ($)	Conservative $500,000 per year ($)	Moderate $1,000,000 per year ($)
5 Years Total Present Value Benefits	1,025,049	2,050,099	4,100,197
10 Years Total Present Value Benefits	1,755,895	3,511,791	7,023,582
15 Years Total Present Value Benefits	2,276,979	4,553,957	9,107,914
20 Years Total Present Value Benefits	2,648,504	5,297,007	10,594,014

As shown in Table 7, the net present value of the proposed rule is positive in the majority of estimated scenarios. For example, the proposed rule alone yields a positive net present value within five years for two of the three benefit estimates and under all benefit estimates within 10 years. Using the moderate estimate of $1,000,000 total annual benefits for all stakeholders, the net present value of the proposed changes to the O&D Survey and changes being considered for the T-100/T-100(f) is positive within five years—even when including one year of concurrent processing.

Table 7.—Net Present Value Proposed Changes to the O&D Survey and Estimated Benefits

[Including changes being considered for T-100/T-100(f).]

Elapsed time	Total Net Present Value
	Very conservative $250,000 total annual benefits		Conservative—$500,000 total annual benefits		Moderate—$1,000,000 total annual benefits
	Proposed rule ($)	Proposed rule + 1 year concurrent ($)	Proposed rule ($)	Proposed rule + 1 year concurrent ($)	Proposed rule ($)	Proposed rule + 1 year concurrent ($)
5 Years	−1,001,874	−1,941,373	23,175	−916,324	2,073,274	1,133,775
10 Years	287,174	−652,325	2,043,070	1,103,571	5,554,861	4,615,361
15 Years	1,206,248	266,749	3,484,227	2,543,728	8,037,184	7,097,685
20 Years	1,861,535	922,036	4,510,039	3,570,540	9,807,046	8,867,547

It is our conclusion that the benefits of the proposed rule will significantly outweigh the costs. We also conclude that, because the present value costs for the proposed rule clearly do not exceed $100 million, for total or incremental costs and even when including one year of concurrent processing, the proposed rule should not be considered an economically significant regulatory action under Executive Order 12866.

b. Vision 100—Century of Aviation Reauthorization Act

Vision 100—Century of Aviation Reauthorization Act (Pub. L. 108-176) recognizes the need for the U.S. to increase its investment in research and development to revitalize the aviation industry as well as to improve aviation information collection. Section 805(a) states that, if the Secretary requires Carriers to provide flight-specific information, the Department will not: (1) Make public the flight-specific fare information until at least nine months after the flight date and (2) issue a rule requiring public dissemination of flight-specific fare information without giving due consideration to and addressing the Carriers' confidentiality concerns. Moreover, Section 805(b)—Effective Date stipulates that the amendment to 49 U.S.C. Section 329(b)(1), stated in Section 805(a), shall become effective on the date of the issuance of a final rule to modernize the O&D Survey. The final rule, pursuant to the ANPRM (RIN 2105-AC71; 63 FR 28128, July 15, 1998), must propose change that “reduces the reporting burden for air carriers through electronic filing of the survey data collected under Section 329(b)(1) of Title 49, U.S.C.” The calculations for burden reduction are shown in Sections L.3.d.1. (Regulatory Analysis—O&D Survey: Regulatory Assessment—Costs) and L.3.e.1. (Regulatory Analysis—T-100/T-100(f): Regulatory Assessment—Costs), below.

The proposed changes to the O&D Survey support electronic filing and reduce manual activity and paperwork. The Issuing Carrier possesses, within its internal systems, the data elements required by the proposed rule. By designating the Issuing Carrier as the Participating Carrier, the proposed rule eliminates the need for the Participating Carrier to manually examine, and obtain information from other carriers about Ticketed Itineraries that were not issued by the Participating Carrier.

We find that the proposed changes to the O&D Survey and the changes considered for the T-100/T-100(f) meet the requirements of Vision 100, specifically Section 805(b), in that the changes “reduce the reporting burden for air carriers through electronic filing of the survey data collected under Section 329(b)(1) of Title 49, U.S.C.” There are three tests of “reduction of reporting burden for air carriers through electronic filing of the survey data”: (1) Net present costs, (2) net present value, and (3) change in annual reporting burden. We base our conclusion on the third test—the change in annual reporting burden for affected carriers. We seek comment about our definition of “reduction of reporting burden for air carriers through electronic filing of the survey data” and our conclusion that the proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) meet the requirement of Vision 100, Section 805(b).

i. Annual Collective Industry Reporting Burden. We believe that the proposed rule reduces the collective reporting burden for the airline industry, for both U.S. Air Carriers and Foreign Air Carriers, even if we include the reporting burden associated with the T-100/T-100(f) changes we are considering. Under the current rule, 47 Carriers (U.S. Air Carriers and Foreign Air Carriers) report the O&D Survey and 230 Carriers report the T-100/T-100(f). Collectively, the industry faces a total annual reporting burden under the current rule of 64,440 hours. Under the proposed changes to the O&D Survey, 38 Carriers would report the O&D Survey. Under the changes to the T-100/T-100(f) that we are considering, 230 Carriers would report the T-100/T-100(f). Under both the proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f), the industry would face a total annual reporting burden of 52,200 hours. The proposed rule, including the changes being considered for the T-100/T-100(f) decreases the industry's collective annual reporting burden by 12,240 hours, or about 18 percent.

The collective annual reporting burden for affected U.S. Air Carriers alone also decreases. Under the current rule, the total annual reporting burden for 34 Carriers reporting the O&D Survey and 121 Carriers reporting the T-100 is 42,804 hours. Under the proposed rule, including the changes being considered for the T-100/T-100(f), the total annual reporting burden for the 25 Carriers reporting the O&D Survey and the 121 Carriers reporting the T-100 would 31,068 hours. This is a collective decrease of 11,736 hours, or about 27 percent.

ii. Annual Individual Carrier Reporting Burden. The proposed changes to the O&D Survey result in substantial decreases for U.S. Air Carriers and Foreign Air Carriers that will continue to report, or cease to report, the O&D Survey. First, the total number of Participating Carriers is reduced from 47 to 38. Second, by designating the Issuing Carrier as the Participating Carrier, the proposed rule reduces the manual processing and intervention inherent in the current rule, thereby simplifying electronic filing.

For informational purposes, we have calculated the annual reporting burden for the changes being considered for the T-100/T-100(f). While these changes would, if adopted, increase the annual reporting burden for each U.S. Air Carrier and each Foreign Air Carrier that will report only the T-100/T-100(f) from 84 hours to 108 hours, they would maximize congruence with the proposed O&D Survey.

The average annual reporting burden of each U.S. Air Carrier or Foreign Air Carrier that currently reports both the O&D Survey and the T-100/T-100(f) will decrease by 216 hours, or about 20 percent, (from 1,044 hours under the current rule to 828 hours under the proposed rule, even when the changes being considered for the T-100/T-100(f) are included. Similarly, under the proposed changes to the O&D Survey and the changes being considered for the T-100, the average annual reporting burden of each of the 13 U.S. Air Carriers that will cease to report the O&D Survey, but continue to report the T-100, will decrease from 1,044 hours to 108 hours, or about 89 percent. Excluding the changes being considered for the T-100, these 13 U.S. Air Carriers would see their annual reporting burden decrease by 91 percent.

c. Departmental Regulatory Policies and Procedures

The Department's Regulatory Policies and Procedures (initially issued February 26, 1979, 44 FR 11034; restated May 22, 1980, DOT Order 2100.5) establish objectives to be pursued in reviewing existing regulations and in issuing new regulations. The objectives include the identification of a regulation as a (1) significant regulation, (2) emergency regulation, or (3) non-significant regulation. One key issue in the determination of a significant rulemaking is the extent to which the affected information is influential. Influential information will have or does have a clear and substantial impact on important public policies or important private sector decisions. The aviation data collected by the O&D Survey and the T-100/T-100(f) are critical for policy makers, Carriers, airports, and other stakeholders (See Section D—O&D Survey Data Usage and Section J—T-100/T-100(f)). Because the proposed changes to the O&D Survey, as well as the changes we are considering for the T-100/T-100(f), change the collection procedures of influential aviation data, this NPRM is considered a significant regulatory action under the Department's Regulatory Policies and Procedures.

d. Regulatory Analysis—O&D Survey

The proposed rule defines a Participating Carrier for the O&D Survey as (1) a U.S. Air Carrier that issues Ticketed Itineraries for travel on scheduled interstate passenger services to or from, or within, the U.S. and operates aircraft with 15 seats or more for scheduled service and (2) a Foreign Air Carrier that has an alliance with a U.S. Air Carrier (pursuant to 49 U.S.C. 41308 and 41309) and is required to report itineraries involving a U.S. point. Under the proposed rule, the total number of Participating Carriers would decrease by about 19 percent, from 47 to 38. The specific costs and benefits of the proposed changes to the O&D Survey are discussed in the following sections.

i. Regulatory Assessment—Costs. For Currently Participating Carriers, we estimated (1) the initial costs of revising the reporting systems to include the proposed new data items and enable monthly reporting of the full universe of issued tickets and (2) the annual costs of monthly submissions of the proposed O&D Survey for 100 percent of Ticketed Itineraries for travel to or from, or within, the U.S. For Newly Participating Carriers, we estimated (1) the initial costs of obtaining systems to include all data elements and enable monthly reporting of the full universe of issued tickets containing a U.S. point and (2) the annual costs of monthly submissions of the proposed O&D Survey for 100 percent of Ticketed Itineraries for travel to or from, or within, the U.S. The initial and annual reporting costs for Formerly Participating Carriers are, of course, zero.

We estimate the total initial reporting costs for the O&D Survey for all Participating Carriers to be approximately $1.49 million, of which approximately $993,000 would be expended by Participating U.S. Air Carriers. We estimate the annual reporting costs for the proposed O&D Survey for all Participating Carriers to be approximately $427,000, of which approximately $281,000 would be expended by Participating U.S. Air Carriers.

We recognize that the initial and annual reporting costs of individual Participating Carriers are likely to differ and, for some Participating Carriers, may be smaller than our estimates. Nevertheless, we have applied a single cost estimate in our regulatory assessment. We recognize that some Participating Carriers may choose to utilize third-party providers, for the initial systems development and/or for monthly data submission, but we do not include estimates of third-party provider costs in this regulatory assessment. However, we are aware that third-party providers already serve the airline industry with systems that collect, bundle, process, and transfer data between Carriers and between Carriers and the Department. Thus, third-party providers may choose to customize or adjust existing data systems, already used by Participating Carriers, to meet the submission requirements of the proposed rule. We assume Participating Carriers would select this option only if its costs were lower; as such, it is possible that Participating Carriers that decide to use third parties would incur lower costs than those we have estimated. We seek comment about the costs and benefits of the use of third-party providers under the proposed O&D Survey.

Initial Reporting Burden. Currently Participating Carriers would incur an initial reporting burden, based on the systems changes required to expand one and add seven ticketed itinerary-level data elements and to expand three and add six Flight Stage-level data elements (See Section I.2.—O&D Survey: Discussion of the Proposed O&D Survey). The proposed data elements are available within the Currently Participating Carriers' internal systems and, therefore, we anticipate that Currently Participating Carriers will be able to access the data elements.

We anticipate that the Currently Participating Carriers will create new automated processes to produce the proposed O&D Survey rather than simply modify the current processes. This is because the proposed procedures will no longer require continual information updates from sources outside the Participating Carrier's control, such as ticketing information from Issuing Carriers, and because the proposed procedures are simpler. In its response to the ANPRM, United Air Lines (Docket OST-1998-4043-15) estimated that “there would be a moderate one time development effort to create and implement the software which would create a TCN-like file each day containing internal [carrier] * * * sales and non-automated agency sales”. We agree, and estimate a “moderate effort” to be the equivalent of two and one-half work months ^[8] of internal development and testing and one and one-half work months ^[9] of external testing and coordination with the Department, for a total of four work months, or 694 staff-hours. We do not estimate the costs of materials or other resources.

Newly Participating Carriers will incur an initial reporting burden based on the O&D Survey data collection and reporting requirements. As with Currently Participating Carriers, Newly Participating Carriers are expected to have the majority of this data present within their internal sales-based systems and TCN records. Furthermore, in 1997, as part of the Rural Airfare Study (Federal Aviation Administration Reauthorization Act of 1996, Section 1213; Pub. L. 104-264), the Department began to collect 100 percent of Ticketed Itineraries for domestic passengers from all certificated and commuter carriers providing scheduled passenger service to communities in the continental U.S. (Docket OST-1997-2767; Order 97-7-27, July 28, 1997). We note that two of the four Newly Participating Carriers were affected by this order and, therefore, are familiar with data submission requirements that are similar to those requested in the proposed rule.

In its response to the ANPRM, United Air Lines (Docket OST-1998-4043-15) stated that a non-CRS participating Carrier could create similar files from its own revenue accounting-type data, “which should not be a major difficulty. Indeed, it should be no more difficult than complying with the present O&D Sampling requirements.” We also note that, when conducting its Rural Airfare Study, the Department solicited comments about the costs of compliance—that is, the cost to submit 100 percent of domestic continental U.S. Ticketed Itineraries. No comments about the costs of complying with this data request were received (Collectively, Docket OST-97-2767).

We agree that Newly Participating Carriers should not find the task of obtaining systems to report the proposed O&D Survey more onerous than obtaining systems for the current O&D Survey. However, we recognize that Newly Participating Carriers will face some development and testing challenges that Currently Participating Carriers will not. We therefore estimate the equivalent of three work months ^[10] of internal development and testing and two work months of external testing and coordination with the Department, for a total of five work months, or 867 staff-hours. We do not estimate the costs of materials or other resources.

Under the Service Contract Act of 1965 (as amended), the U.S. Department of Labor sets the minimum hourly rate, excluding benefits, for Federal Contracts. In 2004, DOL estimated an hourly rate of $27.62 per hour for the positions of Computer Programmer IV and Computer Systems Analyst III.^[11] We recognize that the carriers' hourly costs are likely to be higher, particularly for skilled employees with specialized knowledge of aviation data and reporting. Thus, we estimate an industry hourly cost for a computer programmer/analyst of $55.00 per hour.

Given these assumptions, we estimate the initial reporting costs for the proposed O&D Survey to be $38,170, or 694 hours, per Currently Participating Carrier. For Newly Participating Carriers, we estimate the initial reporting costs to be $47,685, or 867 hours, per Newly Participating Carrier. These estimated costs are based on staff hours only and do not include estimates for materials or other resources. We seek comment about the methods used to determine these initial reporting costs under the proposed rule.

Annual Reporting Burden. The proposed changes to the O&D Survey would require Participating Carriers to report additional data elements for each reported Ticketed Itinerary. The proposed rule would also require Participating Carriers to report 100 percent of all Ticketed Itineraries for travel involving a U.S. point, compared to the 10 percent sample required by the current rule, and to report those itineraries monthly rather than quarterly. However, even though the reporting frequency and total volume of reported data for a Participating Carrier would increase under the proposed rule, we believe that the total annual reporting burden for individual Carriers will decrease.

For example, in 1997, as part of the Rural Airfare Study (Federal Aviation Administration Reauthorization Act of 1996, Section 1213; Pub. L. 104-264), the Department estimated the average annual cost to carriers to comply with data submissions of the Rural Airfare Study ^[12] at approximately 113 hours per carrier (Docket OST-1997-2767-1; Order 97-7027, July 28, 1997). We recognize that the costs of submitting 100 percent of Ticketed Itineraries and incorporating the proposed additional data items would be higher than the costs of submitting monthly Rural Airfare study itineraries. However, we also believe that costs to Participating Carriers under the proposed rule would be lower than those costs under the current rule.

We estimate that the total annual reporting burden for individual Participating Carriers would decrease from 960 hours (current rule) to 720 hours (proposed rule), a total decrease of 240 hours per year per Participating Carrier compared to our 2003 OMB estimate. While this estimation seems counter-intuitive, we believe that such savings are possible. We attribute the 240 hour per year reduction in annual reporting burden for an individual Participating Carrier to (1) the designation of Issuing Carrier, rather than Operating Carrier, as Participating Carrier (192 hours) and (2) the more efficient process by which Issuing Carriers will report 100 percent of Ticketed Itineraries in monthly, rather than quarterly, submissions (48 hours).

As discussed in Section C.1. (Need for Data Modernization: Background), under the current rule, the level of effort required by an Operating Air Carrier to identify whether it is the first Participating Carrier in the itinerary is complex and time-consuming. If the first Participating Carrier is not the Issuing Carrier and did not receive that sale information, then the Participating Carrier is required to employ staff to locate that lifted flight coupon. This is an intensely manual process, and it is a significant burden on limited human and financial resources of the Operating Carrier. Employees with the skills needed to extract information from visual examination of a lifted flight coupon have become increasingly scarce.

On any given day, tens of thousands of passengers fly on commuter carriers and foreign air carriers operating under code-share agreements. As a result of code-share ticketing procedures, the identity of the Operating Air Carrier is often hidden from an outside observer. When the Issuing Carrier does not provide the itinerary details to the Operating Air Carrier, via a TCN record or other means, then it is difficult for the Operating Air Carrier to determine whether any of the other Carriers whose Airline Designator appears on the ticket as the Marketing Carrier is scheduled to operate the flight. The industry has evolved into Code-Share agreements between Franchise Code-Share Partners and Mainline Partners, wherein the Mainline Partner holds the itinerary information yet the current rule holds the Franchise Code-Share Partner responsible for reporting the Ticketed Itinerary. The current rule, in effect, requires a Mainline Partner to prepare multiple O&D Survey reports because it must prepare one for itself and one for each Franchise Code-Share Partner.

We believe that the proposed designation of the Issuing Carrier as the Participating Carrier will result in significantly less manual intervention, matching, and processing than is required under the current rule. Participating Carriers will report those Ticketed Itineraries that they themselves issued and for which they have full information present in their internal systems. Removing the need for Mainline Partners to prepare O&D Survey reports for their Franchise Code-Share Partners is the reason why data can be gathered from 13 fewer Carriers without loss of information from missing Ticketed Itineraries. We therefore estimate that each Currently Participating Carrier will see a reduction in its annual reporting burden of 192 hours per year. Under the proposed reporting frequency, this equates to a reduction of 16 hours per month. Similarly, we estimate a Newly Participating Carrier's annual reporting burden to be equal to that of a Currently Participating Carrier.

We further anticipate that the costs of incorporating the proposed additional data elements are included in the initial reporting costs associated with the configuration of the reporting system. In addition, under the current rule, Participating Carriers are required to submit a 10 percent sample of Ticketed Itineraries using specific sampling methods (49 U.S.C. Part 241 Section 19-7, Appendix A). We believe that the burden to a Carrier of extracting the prescribed 10 percent sample, particularly for Carriers that do not use ticket numbers, is greater than that of generating a dataset containing the full universe of tickets. We therefore expect that the incremental costs of reporting 100 percent of Ticketed Itineraries, rather than a specified 10 percent sample of Ticketed Itineraries, will be extremely small and that the total costs of electronically submitting 12 monthly reports should be very similar to the total costs of electronically submitting 4 quarterly reports.

Identifying the specific cost savings or cost increases associated with each of these issues is complex. However, we note that changes within the industry itself, as well as changes in Carriers' internal data processing systems, often yield considerable savings in the annual reporting burden. In its 2000 submission to OMB (65 FR 19961; April 13, 2000), the Department estimated a 200-hour per year per carrier, or 17 percent, reduction in annual reporting burden, from 1,152 hours to 952 hours. This estimated burden reduction was based on conversations with several large U.S. Air Carriers.

As part of our outreach activities, we spoke with the majority of U.S. Air Carriers about their current internal sales, accounting, and reservations systems and about their system requirements. These discussions were based, in part, on the comments we received in response to the ANPRM. As a result of these conversations, we estimate that these proposed changes—more data elements reported more frequently for all Ticketed Itineraries—to the O&D Survey, when combined with the processing changes inherent in the new reporting systems, are unlikely to result in cost increases and are more likely to yield relatively small savings. We estimate these savings to be 48 hours per year, or 4 hours per month, per Participating Carrier.

In its most recent submission to OMB (68 FR 49543; August 13, 2003), the Department estimated an average annual hourly burden of 960 hours per Participating Carrier. This is an increase of 8 hours per year over the estimate submitted to OMB in 2000 and was based on the changed mix of reporting carriers (several smaller Carriers ceased reporting, thus increasing the average reporting burden for all Carriers). We make no adjustments to the average burden based on the mix of Participating Carriers because, although four small carriers are Newly Participating Carriers under the proposed rule, four of the Formerly Participating Carriers are also small Carriers. We define a small Carrier as a entity employing 1,500 or fewer employees (Air Passenger Carriers, Scheduled; NAICS Code 481111; SAIC Code 4512), as specified by the Small Business Administration's Table of Small Business Size Standards.

We therefore anticipate that the annual reporting burden for Participating Carriers, under the proposed rule, of preparing and submitting monthly O&D Survey data sets containing the proposed data elements and 100 percent of Ticketed Itineraries would not exceed 720 hours on an annual basis for each Participating Carrier. The resulting annual reporting cost per Participating Carrier would be approximately $11,232 (based on an estimated industry salary rate of about $15.60 per hour ^[13] ). These estimated costs are based on staff hours only and do not include estimates for materials or other resources. We seek comment about the methods used to determine these annual reporting costs under the proposed rule.

Reporting Burdens for Participating Carriers. Under the proposed O&D Survey, we estimate a total initial reporting burden for all 38 Participating Carriers of $1,488,520, or 27,064 hours. We estimate a total annual reporting burden for all 38 Participating Carriers of $426,016, or 27,360 hours. Tables 8 and 9 (below) break out the reporting costs for Participating U.S. Air Carriers and Participating Foreign Air Carriers.

Table 8.—Estimated Reporting Costs for Proposed O&D Survey U.S. Air Carriers

	Initial reporting costs			Annual reporting costs
	Hours per carrier	Total hours	Total cost	Hours per carrier	Total hours	Total cost
21 Currently Participating U.S. Air Carriers	694	14,574	$801,570	720	15,120	$235,872
4 Newly Participating U.S Air Carriers	867	3,468	190,740	720	2,880	44,928
25 Participating U.S. Air Carriers		18,042	992,310	720	18,000	280,800

Table 9.—Estimated Reporting Costs for Proposed O&D Survey Foreign Air Carriers

	Initial reporting costs			Annual reporting costs
	Hours per carrier	Total hours	Total cost	Hours per carrier	Total hours	Total cost
13 Currently Participating Foreign Air Carriers	694	9,022	$496,210	720	9,360	$146,016
0 Newly Participating Foreign Air Carriers	867	0	0	720	0	0
13 Participating Foreign Air Carriers		9,022	$496,210	720	9,260	146,016

ii. Regulatory Assessment—Benefits. The proposed rule (1) Expands the number of data elements reported on the O&D Survey, (2) expands the number of annual data submissions of the O&D Survey from four (quarterly) to twelve (monthly), and (3) expands the number of O&D Survey records reported by an individual carrier from a ten percent sample to the full universe of Ticketed Itineraries involving a U.S. point. Our initial regulatory assessment indicates that the benefits of the expanded information that would be collected under the proposed rule would accrue to the Department, other Federal government agencies and offices, Carriers, airports, and other stakeholders. These benefits substantially outweigh the additional costs associated with the initial reporting burden of reconfiguring existing, or obtaining new, systems to report the proposed O&D Survey.

The first benefit is associated with a reduction in annual hourly reporting burden. Under the proposed rule, a Currently Participating Carrier will see a 240-hour per year reduction in its annual hourly reporting burden, from 960 hours to 720 hours (See Section L.3.d.1.). The second benefit is the reduction in the set of Participating Carriers. Because the proposed rule designates the Carrier that issued the Ticketed Itinerary as the Participating Carrier, nine, or approximately 19 percent, fewer Carriers will submit the O&D Survey under the proposed rule. That is, under the proposed rule, fewer Participating Carriers with reduced annual burdens would provide more detailed information than is available under the current rule. Other benefits are likely as well.

The change in reporting time frame will benefit reporting carriers by providing key industry data in a more timely fashion. We are proposing that data be disseminated as discussed in Section K.—Data Dissemination. Furthermore, data will be available by month of travel, rather than quarter of first travel, enabling a more fine-grained assessment of travel demand.

As discussed in Sections D.1. and D.2., a number of agencies within the Department, other Federal agencies, and other stakeholders rely on timely and accurate aviation data when making a variety of policy and business decisions. Monthly data releases will enhance both the usefulness and quality of the O&D Survey. That is, users will be able to assess travel at the monthly level, facilitating more precise analyses. Monthly data further clarify the changes in traffic flows due to seasonality, Carrier route changes, and preferred Carrier. O&D Survey data used in international negotiations would be more timely (i.e., at most three months old) and aid the U.S.' position in these sensitive negotiations.

Monthly O&D submissions will enable the Department to respond more quickly to errors, late submissions, and other data quality concerns. In addition, because of the changes that are being considered for the T-100/T-100(f), monthly O&D submissions could be validated against monthly T-100/T-100(f) submissions. Carriers utilize these data to plan their businesses, accurately forecast potential new services, and, for new entrants, devise more accurate business plans based on real industry demand data. Moving to monthly O&D Survey reporting and dissemination enhances the air carriers' access to this critical information. Furthermore, in their responses to the ANPRM, a number of Carriers recommended more timely reporting and more frequent availability of the data.

Carriers rely not only on timely data but also on detailed information to create more efficient and competitive markets, as well as to estimate the impact of new services at alternative airports. We believe that the proposed new data elements will provide valuable additional data for Carriers as they evaluate market entry and exit. Other stakeholders, discussed in Section D.3., also rely on these data.

The Department has been reporting Directional Passenger trips in the O&D Survey as the best substitute for True O&D since the inception of the O&D Survey. The additional data elements will enable the department to report True O&D according to the One-way Trip methodology widely used in the industry. This provides a much closer approximation to the True O&D than did the Directional Passenger trip methodology.

Flight arrival and departure times will provide a more accurate and useful view of passenger air travel. Similarly, the proposed change from a Directional Passenger to a One-way Passenger (See Section K.2.—Data Dissemination: Proposed Construction of One-way Trips) will enable the FAA and TSA to more effectively plan airport staffing requirements. The identification of passengers as traveling through an airport versus deplaning and remaining will support airport facility planning. State and local transportation planners could also use this information for planning purposes.

Periodically, the Department has requested special data submissions from Carriers because national economic interests are at stake, but the O&D Survey and T-100/T-100(f) do not provide the requisite information. The 2003 SARS outbreak was one such instance. The war in Iraq is another example of a time when the Department has requested more detailed data. Responses to special requests for data, such as the previous examples, are costly in terms of time and other resources. The more robust data gathered by the O&D Survey and the T-100 under the proposed rule would largely eliminate the need for such requests.

The increase in the volume of data to be reported under the proposed rule will result in substantial benefits to Carriers as well as other stakeholders. Carriers currently must generate samples meeting the specific requirements of 14 CFR Section 19-7, Appendix A. The complex sampling methodology introduces the likelihood of sample errors. Furthermore, Carriers themselves have chosen more simplistic reporting processes when available. For example, although the Department permitted alternative sampling methodologies beginning in April 1986, such as sampling at least one percent of Ticketed Itineraries in major domestic markets, all Carriers reporting the O&D Survey have decided that the simplicity of using a single reporting selection criterion outweighs any savings that might accrue from sending the smaller volume of data. Similarly, we expect the process of submitting 100 percent of Ticketed Itineraries will be simpler and more efficient than the creation of more complex sampling techniques, such as stratified sampling or oversampling, intended to capture more representative samples of all markets, despite the larger volume of data.

The proposed changes will also reduce the burden of reporting for Participating Carriers by bringing the responsibility to report a Ticketed Itinerary into alignment with standard global Carrier accounting practices. These practices are based on the presumption that the Issuing Carrier has all the necessary information to report a Ticketed Itinerary; therefore, the Participating Carriers will generally be self-sufficient and able to report the itinerary.

Many Carriers can appear as Operating Carriers on a Ticketed Itinerary, but only one Carrier is the Issuing Carrier. When there are multiple Operating Carriers in an itinerary, the second and subsequent Operating Carriers cannot know with certainty whether the first Operating Carrier reported the itinerary. There is a considerable burden placed on Operating Carriers in the current methodology to determine whether or not they have a responsibility to report any given multiple-Carrier itinerary. The proposed change in Participating Carrier dramatically lowers the burden to report by shifting the reporting responsibility to the Carrier that issued the Ticketed Itinerary and away from the Carrier that transported the passenger. This change will reduce the burden of reporting for Participating Carriers because it eliminates ambiguity.

Currently, if Carriers operate no aircraft with more than 60 seats, they are exempt from reporting. Since 1993, at least one carrier has gone from non-reporting (operating no aircraft with more than 60 seats) to reporting (operating some aircraft with more than 60 seats) to non-reporting (ceasing operation of all aircraft with more than 60 seats). As Carriers reconfigure existing equipment or increase their use of smaller aircraft, the threshold of 60 seats excludes Ticketed Itineraries that provide critical information about passenger air travel and fares. For example, the commencement of operations by Independence Air in June 2004 caused a profound adjustment of fares in small, medium and large markets in the Eastern half of the U.S. However, because Independence Air does not currently operate aircraft with more than 60 seats, it does not have to report O&D Survey data, thereby resulting in an incomplete picture of the effects of this Carrier's start of operations. When a major realignment of fares can result from the actions of a Carrier that qualifies for the small aircraft size exemption, then the small aircraft size exemption must be reevaluated.

When passengers fly their entire itineraries on smaller Carriers that are not required to report the O&D Survey, their travel will not be included under the existing system. Yet, their participation in the air transportation system is significant. By requiring all U.S. Air Carriers issuing tickets for travel to or from, or within, the U.S. operating aircraft with 15 or more seats to report O&D Survey data, the resulting passenger traffic database will contain the majority of Ticketed Itineraries issued by U.S. Air Carriers. The resulting data will capture the increasing role played by regional jets and regional Carriers in the domestic air transportation system.

EAS and the Small Community Air Service Development Program are directed towards smaller markets and require evaluation of service and fares. The Department's statutory responsibility to adapt the air transportation system to the present and future needs of commerce is much more extensive than the needs of the EAS program. Because these markets are inadequately represented in the current O&D Survey, the Department's mandate requires a disproportionately high amount of time and interest in studying markets with lower than average traffic. By requiring Participating Carriers to submit 100 percent of Ticketed Itineraries, the Department will have more accurate and reliable data for small markets impacted by Federal programs. The Department will also be able to compare data for small markets served by EAS or the Small Community Air Service Development Program with similar small markets that are not direct beneficiaries of these programs.

We seek to capture the complex interrelationships between Operating Carrier, Marketing Carrier, and Issuing Carrier. The reduced ambiguity obtained by requiring the Issuing Carrier to report the Ticketed Itinerary should eliminate the possibility that an itinerary will not be reported. In addition, the Issuing Carrier will have all of the necessary data present in its internal systems, maximizing the efficiency and accuracy of data reporting. The increasing role played by code-share agreements will be represented in the O&D Survey.

The benefits to all Carriers and all other stakeholders accrue from the first dissemination of data. Participating Carriers will have access to aggregated monthly data (See Section K—Data Dissemination) for the full universe of Ticketed Itineraries issued by Participating Carriers. Other stakeholders would also have access to more timely and more complete data.

The overall annual reporting burden for the 34 currently Participating Carriers decreases by 8,160 hours and $127,296. Although we are asking four U.S. Air Carriers to begin reporting the O&D Survey, the proposed rule will no longer require 13 U.S. carriers to report. The annual savings for those 13 carriers are estimated to be 12,480 hours and $194,688. These savings are 433 percent greater than the total estimated annual reporting cost for the four newly Participating U.S. Air Carriers.

Although the initial reporting burden for the 38 Participating Carriers is approximately $1.49 million, the number of Participating Carriers will decrease. Under the current rule, 47 Participating Carriers have a collective annual reporting burden of 45,120 hours. The 38 Participating Carriers would, under the proposed rule, have a collective annual reporting burden of 27,360 hours. The proposed rule, therefore, decreases the annual reporting burden by approximately 39%. That is, collectively, the 38 Participating Carriers would expend 17,760 hours per year less under the proposed rule. In the first year, these Participating Carriers face a one-time initial reporting burden of 27,260 hours.

We seek comment about these, and other, benefits that would accrue to any or all stakeholders as a result of the proposed rule.

e. Regulatory Analysis—T-100/T-100(f)

We are considering changes to the set of data elements reported under the T-100/T-100(f). These changes would not affect the definition of Reporting Carrier in 14 CFR Part 217 Section 217.3 and 14 CFR Part 241 Section 19-1. However, because the data elements being considered are flight-specific and are associated with scheduled passenger air transportation, all-cargo Carriers would not be affected by the proposed rule. Should we adopt the changes to the T-100/T-100(f) discussed in this NPRM, the remaining 230 Currently Reporting Carriers would be affected. Accordingly, although we are only considering, and not proposing, the additional data items for the T-100/T-100(f), we include estimates of the cost to Reporting Carriers (U.S. Air Carriers and Foreign Air Carriers) of including the data elements in their T-100/T-100(f) submissions.

i. Regulatory Assessment—Costs. For the 230 Currently Reporting Carriers, we estimated (1) the initial costs of revising the reporting systems to include the new data items being considered and (2) the annual costs of submitting the additional data elements that are being considered. The changes being considered do not change the reporting requirements and do not expand the set of Reporting Carriers; therefore, no estimates are made for Newly Reporting Carriers.

We estimate the total initial reporting costs for the changes being considered for the T-100/T-100(f) for all Currently Reporting Carriers to be approximately $1.52 million, of which approximately $799,000 would be expended by Currently Reporting U.S. Air Carriers. We estimate the annual reporting costs for the changes being considered for the T-100/T-100(f) for all Currently Reporting Carriers to be approximately $387,504, of which approximately $203,861 would be expended by Currently Reporting U.S. Air Carriers. The incremental cost of the changes being considered for the T-100/T-100(f) is approximately $86,000 for all Currently Reporting Carriers.

We recognize that the initial and annual reporting costs of individual Reporting Carriers are likely to differ and, for some Reporting Carriers, may be smaller than our estimates. Nevertheless, we have applied a single cost estimate in our regulatory assessment. In the past, the Department has provided to Reporting Carriers software to enable reporting of the T-100/T-100(f). Because the Department has not yet determined whether the modifications necessary under the proposed rule would be made to Department-provided T-100/T-100(f) reporting software, we do not assume that modified software would be made available to Reporting Carriers.

We recognize that some Reporting Carriers may choose to utilize third-party providers, for the initial system reconfiguration or for monthly data submission but we do not include estimates of third-party provider costs in this regulatory assessment. We are aware that third-party providers already serve the airline industry with systems that collect, bundle, process, and transfer data between Carriers and between Carriers and the Department. Thus, third-party providers may choose to customize or adjust existing data systems, already used by Reporting Carriers, to meet T-100/T-100(f) submission requirements if the changes being considered are adopted. We assume Reporting Carriers would select this option only if its costs were lower; as such, it is possible that Reporting Carriers that decide to use third parties would incur lower costs than those we have estimated. We seek comment about the costs and benefits of the use of third-party providers for submission of the T-100/T-100(f) should the changes we are considering be adopted.

Initial Reporting Burden. Currently Reporting Carriers will incur an initial reporting burden, based on the system changes that would be required to add the two data elements we are considering adding to the current T-100/T-100(f). However, should we adopt the changes being considered, Currently Reporting Carriers are expected to have these data elements within their internal systems and, therefore, we anticipate that Reporting Carriers would be able to access the data elements.

We anticipate that, if the changes we are considering are adopted, the Currently Reporting Carriers would create supplemental automated processes to produce the expanded T-100/T-100(f) to access the additional data elements. The Department had previously (Docket OST-1996-1049-2) estimated that the addition of two capacity data items, available seats and available payload capacity, would not be an unreasonable burden because the data elements were not difficult to calculate or determine and were readily available to all air carriers through computer access. We believe the data elements that we are considering, Master Flight Number and flight date, should also be readily available to Carriers.

The cost to link the sources of Master Flight Number and flight date to Currently Reporting Carriers' existing T-100/T-100(f) reporting systems will be based on a number of factors, including the current level of integration of individual Carriers' systems. We believe that this cost would be significantly less than the cost estimated for the one-time changes to the O&D Survey reporting systems. We therefore estimate that Reporting Carriers would require, should the changes we are considering be adopted, the equivalent of two work weeks ^[14] of internal development and testing and one work week of external testing and coordination with the Department, for a total of three work weeks, or 120 staff hours, to incorporate the changes into their systems.

Under the Service Contract Act of 1965 (as amended), the U.S. Department of Labor establishes the minimum hourly rate, excluding benefits, for Federal Contracts. In 2004, DOL estimated an hourly rate of $27.62 per hour for the positions of Computer Programmer IV and Computer Systems Analyst III.^[15] We recognize that the carriers' hourly costs are likely to be higher, particularly for skilled employees with specialized knowledge of aviation data and reporting. Thus, we estimate an industry hourly cost for a computer programmer/analyst of $55.00 per hour.

Given these assumptions, we estimate that, should the changes we are considering making to the T-100/T-100(f) be adopted, the initial reporting cost for the revised T-100/T-100(f) would be $6,600, or 120 hours, per Currently Reporting Carrier. This estimated cost is based on staff hours only, and does not include estimates for materials or other resources. We seek comment about the methods used to determine the initial reporting cost under the changes being considered for the T-100/T-100(f).

Annual Reporting Burden. The current structure of the T-100/T-100(f) Market file groups traffic data by carrier, entity, Origin, Destination, and service class. The current structure of the T-100/T-100(f) Segment file further groups traffic data by aircraft type. The total number of records reported for each file type is dependent upon the extent to which traffic data can be grouped during the reporting period.

Hypothetically, in a given 31-day month, a Carrier operates one daily flight with one service class between a particular Origin Airport and Destination Airport. Under the current T-100/T-100(f) it would report one Market record summarizing the traffic data for that Carrier, entity, Origin, Destination, and service class for the entire month. It would report the number of Segment records corresponding to the different numbers of aircraft types used to service that route in that month. If the Carrier used only one aircraft type, it would report one Segment record. If it used two different aircraft types, it would report two Segment records, and so forth, for a maximum of 31 Segment records.

In the final rule adopting the T-100/T-100(f) reporting system (53 FR 46294, November 16, 1988; Referenced in Docket OST-96-1049-13), the Department estimated that the reporting burden for the entire T-100/T-100(f) system would vary between one hour and 20 hours per month per Reporting Carrier, with an average of seven hours per monthly response. Therefore, submitting Segment records and Market records, grouped as described above, takes an average of seven hours per month, or 84 hours per year, per Reporting Carrier.

The changes that we are considering making to the T-100/T-100(f) would group Market records and Segment records by Master Flight Number and flight date, expanding the total number of records reported. As in the previous example, for a 31-day month, a hypothetical Carrier operates one daily flight, with a single Master Flight Number, with one service class, between a particular Origin Airport and Destination Airport. For that month, because there are 31 flight dates for that Master Flight Number, the Carrier would report 31 Market records (grouped by carrier, entity, Origin, Destination, service class, Master Flight Number, and flight date). It would report 31 Segment records (grouped by carrier, entity, Origin, Destination, service class, aircraft type, Master Flight Number, and flight date).

The estimated increase in annual reporting costs, for Currently Reporting Carriers, associated with the changes we are considering making to the T-100/T-100(f) is based on the increased costs to identify, store, and transmit records that are specific by Master Flight Number and flight date. We anticipate that these costs would be reduced through efficient reporting systems. We incorporate that assumption into our estimates of the initial reporting costs that Currently Reporting Carriers would incur if the changes we are considering are adopted. We therefore estimate that the monthly reporting would increase by 2 hours per month, or 24 hours per year, for a total of 9 hours per month, or 108 hours per year.

Given these assumptions, we estimate the annual reporting cost for the T-100/T-100(f) would increase by $375, or 24 hours, per Currently Reporting Carrier if the changes we are considering are adopted. This estimated cost is based on staff hours only and does not include estimates for materials or other resources. We therefore anticipate that the annual reporting burden for Reporting Carriers, should the changes we are considering be adopted, of preparing and submitting monthly T-100/T-100(f) data sets containing the additional data elements would average 108 hours, or approximately $1,685 (based on an estimated industry salary rate of about $15.60 per hour ^[16] ), per Currently Reporting Carrier. These estimated costs are based on staff hours only and do not include estimates for materials or other resources. We seek comment about the methods used to determine these annual reporting costs given the changes we are considering making to the T-100/T-100(f).

Reporting Burden for Reporting Carriers. We are considering the addition of two data elements to the T-100/T-100(f). Should those changes be adopted, we estimate a total initial reporting burden for the 230 Currently Reporting Carriers of $1,518,000, or 27,600 hours. We further estimate that adoption of the changes being considered would result in an annual reporting burden for all 230 Reporting Carriers of 24,840 hours, or $387,504. This is an increase of 5,520 hours, or approximately $86,000. In Tables 10 and 11, below, we break out the initial reporting costs and annual reporting costs for U.S. Air Carriers and Foreign Air Carriers.

TABLE 10.—Estimated Reporting Costs for Changes Being Considered for the T-100

[U.S. Air Carriers]

	Initial reporting costs			Annual reporting costs
	Hours per carrier	Total hours	Total cost	Hours per carrier	Total hours	Total cost
121 Currently Reporting U.S. Air Carriers	120	14,520	$798,600	108	13,068	$203,860

TABLE 11.—Estimated Reporting Costs for Changes Being Considered for the T-100(f)

[Foreign Air Carriers]

	Initial reporting costs			Annual reporting costs
	Hours per carrier	Total hours	Total cost	Hours per carrier	Total hours	Total cost
109 Currently Reporting Foreign Air Carriers	120	13,080	$719,400	108	11,772	$183,643

iii. Regulatory Assessment—Benefits. We recognize that, by considering the collection of T-100/T-100(f) data by Master Flight Number and flight date, we would increase the total number of records to be submitted by Current Reporting Carriers. However, the addition of Master Flight Number and flight date would enable the T-100/T-100(f) to continue to be used to verify the O&D Survey. The proposed data elements will improve the quality and use of traffic data in decision making by enabling a maximum congruence between the T-100/T-100(f) and the O&D Survey. As such, it supports the benefits associated with the proposed changes to the O&D Survey (Section L.3.d.2). The changes being considered for the T-100/T-100(f) would, through data specific to Master Flight Number and flight date, provide additional information for airport and air traffic control planning. Stakeholders would have information about aircraft size, number of passengers, and flow of passengers and aircraft by time of day.

4. Regulatory Flexibility Act of 1980, Small Business Regulatory Enforcement Fairness Act of 1996, Executive Order 13272

The Regulatory Flexibility Act (RFA) of 1980 (Pub. L. 96-354; 94 Stat. 1164; codified at 5 U.S.C. 601) requires agencies to consider the impact of their regulatory proposals on small entities, analyze effective alternatives that minimize the impact on small entities, and make their analyses available for public comment. It does not, however, seek preferential treatment for small entities, require agencies to adopt regulations that impose the least burden on small entities, or mandate exemptions for small entities.

The Small Business Regulatory Enforcement Fairness Act (SBREFA) of 1996 amended the Regulatory Flexibility Act of 1980. The Department has established a Guidance Manual on SBREFA.

Executive Order 13272 (67 FR 52462, August 16, 2002) requires each agency to establish written procedures and policies to promote compliance with the Regulatory Flexibility Act and to ensure that potential impacts of draft rules on small entities will be properly considered. The Department has established Policies and Procedures for Implementing Executive Order 13272. We define a small Carrier as an entity employing 1,500 or fewer employees (Air Passenger Carriers, Scheduled; NAICS Code 481111; SAIC Code 4512), as specified by the Small Business Administration's Table of Small Business Size Standards.

a. Affected Businesses

The changes we are considering making to the T-100/T-100(f) would affect all Air Carriers that are required to report the T-100/T-100(f) under the current rule. The definition of Reporting Carrier is not affected by the possible changes. Previous changes to the T-100/T-100(f) were expected to affect approximately 100 small entities, and were certified as not having a significant economic impact on a substantial number of small entities (Docket OST-1998-4043; 67 FR 49217, July 30, 2002). Therefore, we believe that, if the changes we are considering making to the T-100/T-100(f) are adopted, there will likely be no significant economic impact on a substantial number of small entities.

The proposed changes to the O&D Survey would affect all Carriers operating aircraft with 15 or more seats and issuing tickets for travel on scheduled interstate passenger services to or from, or within, the U.S. Four small entities would cease to report the O&D Survey, while four different small entities would begin to report the O&D Survey. Small entities represent 9.5 percent of Participating Carriers under the proposed rule, and 100 percent of Newly Participating Carriers under the proposed rule. Our proposed rules do contain direct reporting, recordkeeping, or other compliance requirements that would affect small entities. However, the Department cannot exempt all small carriers from reporting the passengers they carry without jeopardizing the completeness and accuracy of the traffic statistics. Small entities are integrated into the fabric of the global aviation industry. Many passengers carried by large U.S. Air Carriers begin their journeys on small Carriers. Exemption of that category of Ticketed Itineraries from reporting affects the integrity of the statistical data and would affect some markets disproportionately, thereby introducing bias into the data. The Department believes that the best way to minimize the negative effects of regulation on small entities is to correct the Department's reliance on Directional Passengers, change the reporting responsibility to the Issuing Carrier, and obtain information about 100 percent of Ticketed Itineraries.

Small entities benefit from cost-effective access to better information that is critical to making sound business decisions. Small entities are more dependent on the Department's data than are larger competitors which can afford alternative data sources. However, all Carriers must be confident that the statistical and financial data disseminated by the Department measure the industry accurately. The Department must use the correct metrics to reflect the global airline industry and must disseminate industry statistics in ways that are useful and understandable for all stakeholders. The proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) will increase the efficiency of all Carriers. More complete data reduce the need for supplemental reports and specialized data processing, which are a greater burden to smaller Carriers. Our new rules would also benefit most Carriers because, within confidentiality constraints, all Carriers will have access to data that accurately and completely reflect the state of the airline industry, including traffic and operating data. More timely data submission (by carriers) and data dissemination (by the Department) will enhance the usefulness of the collected data. Furthermore, small entities will benefit from complete (e.g., 100 percent) data for the markets they themselves serve.

Section 213(a) of SBREFA requires the Department to assist small entities in understanding the proposed rule so that they can better evaluate its effects of them and participate in the rulemaking process. We encourage small entities to contact Richard Pittaway at the address listed under FOR FURTHER INFORMATION CONTACT with any questions about the proposed rule, its provisions, or options for compliance.

b. Initial Regulatory Flexibility Statement

We do not anticipate that the changes we are considering making to the T-100/T-100(f) will have a significant economic impact on a substantial number of small businesses. Although we anticipate that the proposed changes to the O&D Survey, and therefore the proposed rule, may have a significant economic impact on the four small entities that will become Newly Participating Carriers, we believe that the benefits gained by all small entities, including these four Carriers, offset the additional costs. Because four small entities will become Participating Carriers while four other small entities will no longer be required to report the O&D Survey, we believe that the net impact of the proposed rule is relatively small. Accordingly, I certify that the proposed rule will not have a significant economic impact on a substantial number of small entities. Interested persons may address our conclusions under the Regulatory Flexibility Act in their comments submitted in response to this notice of proposed rulemaking.

c. O&D Survey

Inherent in the RFA is a desire to remove barriers to competition. New entrant competitors are the lifeblood of the airline industry, bringing innovations and new business concepts to the marketplace. Within the aviation sector, small entities are disadvantaged relative to larger entities. Large carriers have the resources and longevity to research and develop markets using costly information independent of the statistical data disseminated by the Federal government.

Small and new entrant Carriers depend on the Department's traffic data to a greater degree in planning their businesses than do larger and incumbent Carriers. Inaccurate and incomplete data disseminated by the Department disproportionately hinders small and new entrant Carriers. The Regional Airline Association (Docket OST-1998-4043-11), an association of small and medium-sized Carriers, stated in its ANPRM comments that “it is clear that for the U.S. regional airline industry, the current data collection process is both inappropriate and inconsistent. The current structure of reporting rules and regulations offer what the Association considers to be an approach to information gathering that is out of step with the current operating environment for regional airlines.”

Smaller airports are also disadvantaged under the current reporting requirements. These airports are often predominantly served by smaller, non-reporting Franchise Code-Share Partners; trips taken on non-reporting Carriers are missing from the O&D Survey data. Small airports that are served from only one hub are more vulnerable to circuity factors inappropriately identifying a break in the direction of travel. Even if every part of a Ticketed Itinerary were reported correctly, small airports would still be disadvantaged because the 10 percent sample is less accurate and reliable for the small number of passengers traveling there. Without accurate and complete scheduled passenger traffic data, smaller airports are less able to schedule services, assess facilities demand, and identify growth opportunities.

As shown in Table 1, 38 U.S. Air Carriers will be affected by the proposed changes to the O&D Survey. Of the 13 formerly Participating Carriers (i.e., those Carriers that would no longer submit the O&D Survey under the proposed rule), four are considered small business entities under the Small Business Administration's Table of Small Business Size Standards. The remaining nine have more than 1,500 employees and/or are subsidiaries of parent companies where the total employees are more than 1,500 employees.

All four of the newly Participating Carriers are considered small business entities under the Small Business Administration's Table of Small Business Size Standards. Because four small entities will no longer be required to report, and four different small entities will become Participating Carriers, there is a net addition of zero small business entities as Participating Carriers for the O&D Survey.

We anticipate that the proposed changes to the O&D Survey may have a significant economic impact on the small businesses affected. Small entities represent 100 percent of the newly Participating Carriers and 9.5 percent of Participating Carriers under the proposed rule. We believe that the annual reporting burden will be less for smaller entities because they generate, process, store, and submit fewer Ticketed Itineraries than larger entities. However, we recognize that the initial reporting burden will be proportionately greater for both the currently participating small entities and newly participating small entities.

The Department believes that the most significant reporting burden on small Carriers will be removed by shifting the reporting responsibility to the Issuing Carrier. The vast majority of small carriers, under the proposed system, would not be required to report the O&D Survey at all. Nonetheless, Carriers that issue Ticketed Itineraries on their own ticket stock remain a concern under SBREFA.

The Department recognizes that the markets served by Air Taxis and other similarly small operations represent a significantly different transportation market. The Department acknowledges that passengers in this market must be measured differently than the passengers in the global air transportation market. We do not wish to burden these truly small carriers serving local needs and have therefore not proposed to require them to report data. The Department wishes to reduce the ambiguity in a Carrier's classification as a Participating Carrier. Moving into and out of the Participating Carrier classification from time to time is problematic for both the Carrier concerned and for the community of users of the O&D Survey. This ambiguity in the current system has had a disproportionately negative impact on smaller entities since they are more likely to be affected by the current reporting threshold. Therefore, we propose that (1) carriers only flying planes within a single state, (2) carriers flying no aircraft with 15 or more seats, (3) non-scheduled air taxi services, and (4) non-scheduled helicopter carriers will continue to be exempt from reporting the O&D Survey.

Because small Carriers provide service to smaller markets, they will benefit from the additional traffic data that will be available under the proposed rule. EAS and the Small Community Air Service Development Program are directed towards smaller markets and require evaluation of service and fares. Under EAS, the Department determines the minimum level of service required at each eligible community by specifying a hub through which the community is linked to the global air transportation system, and specifying a minimum service level in terms of flights and available seats. Where necessary, the Department pays a subsidy to a Carrier to ensure that the specified level of service is provided. More detailed data will assist the Department in its allocation of funds to these programs and to eligible Carriers participating in them.

d. T-100/T-100(f)

As shown in Table 2, 121 U.S. Air Carriers would be affected by the changes we are considering making to the T-100. Because the proposed rule makes no change in the criteria for Reporting Carrier, we conclude that the number of small entities that would be impacted if the changes we are considering making are adopted is not affected by the content of those potential changes. Eighty-nine of the 121 U.S. Air Carriers that would be affected if the changes were adopted are small entities under the Small Business Administration's Table of Small Business Size Standards. Nine of the 121 entities are subsidiaries of larger airlines and the total employee base is greater than 1,500. Twenty-nine of the 121 entities have 1,500 or more employees. Of the remaining 89, 24 have been confirmed as having fewer than 1,500 employees and 59 are presumed to have fewer than 1,500 employees based on the total number of aircraft operated by the individual Carrier. Sources include internal departmental counts of Carriers' employees, the Regional Airline Association (http://www.raa.org/​members/​AirlineDirectory.htm) and Reference USA (http://www.referenceusa.com).

As with the proposed O&D Survey, we believe that the annual reporting burden associated with the changes we are considering making for the T-100/T-100(f) will be less for smaller entities because they operate fewer flights and, therefore, generate, process, store, and submit fewer records than larger entities. The estimated initial reporting burden, assuming adoption of the changes being considered, would be approximately 120 hours, or $6,600 per carrier. However, we note that BTS has, in the past, provided T-100 reporting software to Carriers upon request. Small entities that have, in the past, relied upon BTS software to reduce or even eliminate the initial reporting burden associated with past changes to the T-100/T-100(f) may be able to continue to do so.

Furthermore, we note that when approximately 100 small entities first began to report the T-100, in place of Form 298-C, Schedule T-1, we found that change would not have a significant economic impact on a substantial number of small entities (67 FR 49217, July 30, 2002). Therefore, we conclude that the changes we are considering making to the T-100/T-100(f) would not, if adopted, have a significant economic impact on the small businesses affected.

5. Paperwork Reduction Act of 1995

The Paperwork Reduction Act of 1995 (Pub. L. 104-113; 5 CFR 1320.0; 44 U.S.C. 3501 et seq.) requires each Federal agency to (1) Establish a process, independent of program responsibility, to evaluate proposed collections of information; (2) manage information resources to reduce information collection burdens on the public; and (3) ensure that the public has timely and equitable access to information products and services. Its purposes include (1) The minimization of the paperwork burden resulting from the collection of information by or for the Federal government; (2) ensuring the greatest possible public benefit from and maximization of the utility of information created, collected, maintained, used, shared and disseminated for or by the Federal government; (3) improving the quality and use of Federal information to strengthen decision making, accountability, and openness in government and society; (4) minimization of the cost to the Federal government of the creation, collection, maintenance, use, dissemination, and disposition of information; and (5) providing for the dissemination of public information on a timely basis, on equitable terms, and in a manner that promotes the utility of the information to the public and makes effective use of information technology.

The proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) contain collection-of-information requirements subject to the Paperwork Reduction Act. Under the Paperwork Reduction Act, a person is not required to respond to a collection of information by a Federal agency unless the collection displays a valid OMB control number. The reporting and recordkeeping requirement associated with this proposed rule is being sent to OMB for approval in accordance with the Paperwork Reduction Act, under OMB NO: 2139-0001 (for the O&D Survey) and OMB NO. 2138-0040 (for the T-100/T-100(f)).

The proposed changes to the O&D Survey are estimated to reduce the annual reporting for U.S. Air Carriers from 960 hours per year (240 hours per submission, with data reported quarterly) to 720 hours per year (60 hours per submission, with data reported monthly). In addition, by designating the Issuing Carrier as the Participating Carrier, the proposed changes to the O&D Survey are estimated to reduce the number of Participating U.S. Air Carriers by nine (13 U.S. Air Carriers would cease to report while four U.S. Air Carriers would begin to report). In sum, under the proposed changes to the O&D Survey, the collective annual reporting burden for U.S. Air Carriers is estimated at 18,000 hours. When Foreign Air Carriers that operate under 49 U.S.C. 41308 and 41309 and are required, under grant of antitrust immunity, to report itineraries involving a U.S. point are included, the proposed changes to the O&D Survey are estimated to result in a collective annual reporting burden for the world airline industry of 27,360 hours. These data are detailed in Tables 8 and 9. If these changes are not made, the collective annual reporting burden for U.S. Air Carriers is estimated to be 32,640 hours and the collective annual reporting burden for the world airline industry is estimated to be 45,120.

The changes that we are considering making to the T-100/T-100(f) are estimated to increase the annual reporting burden for Reporting Carriers by 2 hours per month, or a total of 24 hours per year. If we were to make the changes to the T-100/T-100(f) that we are considering, the collective annual reporting burden for U.S. Air Carriers would be 13,068 hours and the collective annual reporting burden for the world airline industry would be 24,840. These data are detailed in Tables 10 and 11. If we do not make the changes we are considering, the collective annual reporting burden under the T-100/T-100(f) would be 10,164 hours for U.S. Air Carriers and 19,320 for the world airline industry.

Table 12, below, compares the collective annual reporting burden for the proposed O&D Survey changes to the collective annual reporting burden under the current rule. Table 13, below, compares the collective annual reporting burden for the changes we are considering making to the T-100/T-100(f) to the collective annual reporting burden under the current rule.

Table 12.—Collective Annual Reporting Burden for U.S. Air Carriers and World Airline Industry Proposed Changes Versus Current Rule O&D Survey

	Proposed changes to O&D survey collective annual reporting burden (hours)	Current O&D survey collective annual reporting burden (hours)
U.S. Air Carriers	18,000	32,640
World Airline Industry	27,360	45,120

Table 13.—Collective Annual Reporting Burden for U.S. Air Carriers and World Airline Industry Considered Changes Versus Current Rule T-100/T-100(f)

	Proposed changes to O&D survey collective annual reporting burden (hours)	Current O&D survey collective annual reporting burden (hours)
U.S. Air Carriers	13,068	10,164
World Airline Industry	24,840	19,320

a. O&D Survey

Agency: Office of the Secretary.

Title: Passenger Origin-Destination Survey Report.

Type of Request: Revision of a currently approved collection.

Affected Public: Businesses.

OMB Clearance Number (Current): 2139-0001 (expires 12/31/06).

OMB Clearance Number (Proposed): To be determined.

Requested Expiration Date of Approval: Three years from the date of approval.

Proposed Use of Information: Electronic Dissemination to Transportation Planners and Analysts.

Frequency: Monthly.

Summary of the Collection of Information: We are proposing that Issuing Carriers operating aircraft with at least 15 seats report 100 percent of the ticketed itineraries that they issue involving at least one Origin and/or Destination in the U.S. and to do so monthly. Data from the O&D Survey are used by the Department to fulfill its aviation mission.

Description of the Need for the Information and Proposed Use of the Information: To capture the proliferation of code-sharing and increased use of regional carriers, we will collect information on the Issuing Carrier, Marketing Carrier, and Operating Carrier as well as flight-specific data and information about passenger catchment areas.

Description of the Likely Respondents: Respondents are U.S. Air Carriers issuing tickets for service to, from, or within the U.S. and operating aircraft with 15 or more seats and Foreign Air Carriers that operate service involving a U.S. Point under 49 U.S.C. Sections 41308 and 41309.

Estimate of the Total Annual Reporting and Recordkeeping Burden Resulting from the Collection of Information: We estimate the total annual burden to 25 U.S. Air Carriers and 13 Foreign Air Carriers resulting from the proposed rule to be 27,260 hours and $426,816. For Carriers reporting under the current rule, the proposed rule results in a net decrease of 240 hours per year per Carrier.

b. T-100/T-100(f)

Agency: Office of the Secretary.

Title: Passenger Report of Traffic and Capacity Statistics—The T-100/T-100(f) System.

Type of Request: Revision of a currently approved collection.

Affected Public: Businesses.

OMB Clearance Number (Current): 2138-0040 (expires 7/31/05).

OMB Clearance Number (Proposed): To be determined.

Requested Expiration Date of Approval: Three years from the date of approval.

Proposed Use of Information: Electronic Dissemination to Transportation Planners and Analysts.

Frequency: Monthly.

Summary of the Collection of Information: We are considering requiring Carriers subject to T-100/T-100(f) reporting submit expanded T-100/T-100(f) reports containing two additional data elements. Data from the T-100/T-100(f) are used by the Department to fulfill its aviation mission.

Description of the Need for the Information and Proposed Use of the Information: The T-100/T-100(f) provides information about the movement of aircraft and passengers through the national air space system. The additional data elements will allow a more detailed view of this traffic and enable the continuation of validating the enhanced O&D Survey with the T-100/T-100(f) reports.

Description of the Likely Respondents: Respondents are those U.S. Air Carriers subject to reporting under 14 CFR Part 241 and Foreign Air Carriers subject to reporting under 14 CFR Part 217.

Estimate of the Total Annual Reporting and Recordkeeping Burden Resulting from the Collection of Information: We estimate that, should the changes we are considering to the T-100/T-100(f) be adopted, the total annual burden would increase by 5,520 hours and $86,112.

6. The National Environmental Protection Act of 1969

The Department has analyzed the proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) for the purpose of the National Environmental Protection Act (Pub. L. 91-190 as amended; 42 U.S.C. 4321-4347). The proposed amendments will not have any impact on the quality of the human environment.

7. Executive Order 13132

Executive Order 13132, Federalism (64 FR 43255, August 10, 1999), requires Federal agencies to adhere to the fundamental federalism principles set out in Section 2 as well as to adhere to the criteria specified in Section 3.

The proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) have been analyzed in accordance with the principles and criteria contained in Executive Order 13132. We have determined that the proposed rule will have no substantial direct effects on the States, on the relationship between the national government and the States, or on the distribution of power and responsibilities among the various levels of government. Therefore, we have determined that it does not have sufficient federalism implications to warrant the preparation of a Federalism Assessment or to warrant consultations with State and local governments.

8. Executive Order 12630

Executive Order 12630, Government Actions and Interference with Constitutionally Protected Property Rights (53 FR 8859, March 15, 1998; 3 CFR 1988 Comp., p.554), specifies that Federal Agencies should be sensitive to, anticipate, and account for, the obligations imposed the Just Compensation Clause of the Fifth Amendment in planning and carrying out governmental actions, among other purposes.

The proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) would not effect a taking of private property or otherwise have taking implications under Executive Order 12630.

9. Executive Order 12988

Executive Order 12988, Civil Justice Reform (61 FR 4729, February 7, 1996), seeks to improve legislative and regulatory drafting to enact legislation and promulgate regulations that do not unduly burden the Federal Court System, among other purposes.

The proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) meet applicable standards in Sections 3(a) and Section 3(b)(2), of Executive Order 12988, to minimize litigation, eliminate ambiguity, and reduce burden.

10. Executive Order 13045

We have analyzed the proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) under Executive Order 13045, Protection of Children From Environmental Health Risks and Safety Risks (62 FR 19883, April 23, 1997). The proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) do not concern an environmental risk to health or risk to safety that may disproportionately affect children.

11. Executive Order 13175

The proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) will not have tribal implications, will not impose substantial direct compliance costs on Indian tribal governments, and will not preempt tribal law. Therefore, they are exempt from the consultation requirements of Executive Order 13175, Consultation and Coordination With Indian Tribal Governments (65 FR 67249, November 9, 2000). If tribal implications are identified during the comment period, we will undertake appropriate consultations with the affected Indian tribal officials.

12. Executive Order 13211

We analyzed the proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) under Executive Order 13211, Actions Concerning Regulations that Significantly Affect Energy Supply, Distribution, or Use. We have determined that the proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) are not classified as a “significant energy action” under that order and would not have a significant adverse effect on the supply, distribution, or use of energy.

13. OMB Circular No. A-76 (Revised)

We have analyzed the proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) under Circular No. A-76 (revised), Performance of Commercial Activities. It is the policy of the Federal government to ensure that the American people receive maximum value for their tax dollars by subjecting certain activities of the government to competition. We find that the activity of collection of data under the proposed changes to the O&D Survey and the changes being considered for the T-100/T-100(f) may be deemed a commercial activity.

14. Regulation Identifier Number

A regulation identifier number (RIN) is assigned to each regulatory action listed in the Unified Agenda of Federal Regulations. The Regulatory Information Service Center publishes the Unified Agenda in April and October of each year. The RIN number 2105-AC71 contained in the heading of this document can be used to cross-reference this action with the Unified Agenda.

M. Glossary

1. Air Carrier. Any citizen of the United States who undertakes, whether directly or indirectly or by lease or any other arrangement to engage in air transportation.

2. Airline Designator. The two character airline identifier as listed in the IATA Airline Coding Directory.

3. ARC. Airlines Reporting Corporation (ARC) is a clearinghouse owned collectively by Carriers to collect ticket information and funds from individual travel agencies and distribute the information and funds to the appropriate Carriers.

4. ARNK. Arrival unknown.

5. Carrier. A U.S. Air Carrier or Foreign Air Carrier.

6. City Code. The IATA location identifier assigned to a city associated with multiple airports.

7. Currently Participating Carrier. An Air Carrier or Foreign Air Carrier that is required to report the O&D Survey under the current rule and would be required to report the O&D Survey proposed in this rulemaking.

8. Currently Reporting Carrier. An Air Carrier or Foreign Air Carrier that is required to report the T-100/T-100(f) under the current rule and would be required to report the T-100/T-100(f) under the rule proposed in this rulemaking.

9. Designated Carrier Liaison. An individual authorized to act on behalf of the Participating Carrier in operational matters pertaining to the Carrier's collection of data and subsequent submission of the data to the Department.

10. Directional Passenger. A passenger's continuous trip in the same direction regardless of the number of days the journey takes, but subject to certain circuity rules designed to approximate the passenger's True O&D.

11. Fare Category. A summary category of fare basis codes.

12. Franchise Code-Share. A code-share relationship wherein one Carrier markets air travel as a wet-lease on another Carrier's flights whether or not a wet-lease agreement per se actually exists, and wherein one of the Carrier's partners will never appear as the Marketing Carrier for the other.

13. Franchise Code-Share Partner. In a Franchise Code-Share, the Carrier that is reported in the O&D Survey as the Operating Carrier but not as the Marketing Carrier.

14. Flight-Coupon Stage. The portion of a Ticketed Itinerary that lies between two sequential points of a Ticketed Itinerary. A passenger's Flight-Coupon Stage may involve multiple takeoffs and landings. A Flight-Coupon Stage may be on any scheduled transportation held out and ticketed by the Issuing Carrier.

15. Flight-Stage. The operation of an aircraft from takeoff to landing. Technical stops are disregarded.

16. Flight-Stage Origin Airport. The airport identifier of the airport from which a Flight-Stage departs. For intermodal ticketed ground stations, such as a bus station or a train station, that station should be treated as an airport.

17. Flight-Stage Destination Airport. The airport identifier of the airport in which a Flight-Stage arrives. For intermodal ticketed ground stations, such as a bus station or a train station, that station should be treated as an airport.

18. Foreign Air Carrier. An airline that is not a U.S. Air Carrier.

19. Formerly Participating Carrier. An Air Carrier or Foreign Air Carrier that is required to report the O&D Survey under the current rule but would not be required to report the O&D Survey under the rule proposed in this rulemaking.

20. Formerly Reporting Carrier. An Air Carrier or Foreign Air Carrier that is required to report the T-100/T-100(f) under the current rule but would not be required to report the T-100/T-100(f) under the rule proposed in this rulemaking.

21. Issuing Carrier. The Air Carrier or Foreign Air Carrier that is responsible for the ticket stock on which the Ticketed Itinerary is issued and that is responsible for collecting the remuneration for the fare and the taxes and fees. Also known as plating carrier.

22. Issuing Carrier Identifier. The IATA assigned code that identifies the Carrier that issued a Ticketed Itinerary.

23. Licensed Foreign Air Carrier. A Foreign Air Carrier with a permit issued under the requirement described in 49 U.S.C. 41301.

24. Mainline Partner. In a Franchise Code-Share, the Mainline Partner is the Carrier that appears as the marketing carrier.

25. Marketing Carrier. The Carrier that appears as the Carrier for a Flight-Coupon Stage on a Ticketed Itinerary, whether or not it actually operates the flight.

26. MIDT. The Marketing Information Data Tape is information, sold by a GDS, about air travel reservations made through travel agents.

27. Newly Participating Carrier. An Air Carrier or Foreign Air Carrier that is not required to report the O&D Survey under the current rule but would be required to report the O&D Survey under the rule proposed in this rulemaking.

28. Newly Reporting Carrier. An Air Carrier or Foreign Air Carrier that is not required to report the T-100/T-100(f) under the current rule but would be required to report the T-100/T-100(f) under the rule proposed in this rulemaking.

29. One-way Trip. A collection of information about a journey of one or more Flight-Stages of a Ticketed Itinerary, which are associated with one another using a standard methodology that is designed to approximate the passenger's True O&D.

30. One-way Trip Origin. The first airport of a One-way Trip.

31. One-way Trip Destination. The final airport of a One-way Trip.

32. Operating Carrier. The Carrier whose aircraft and flight crew are used to perform a Flight-Coupon Stage.

33. Participating Carrier. An Air Carrier or Foreign Air Carrier that is required to report the O&D Survey.

34. Passenger, Nonrevenue. A person traveling free or under token charges, except those expressly named in the definition of Revenue Passenger; a person traveling at a fare or discount available only to employees or authorized persons of air carriers or their agents or only for travel on the business of the carriers; and an infant who does not occupy a seat. The definition includes, but is not limited to following examples of passengers when traveling free or pursuant to token charges:

a. Directors, officers, employees, and others authorized by the air carrier operating the aircraft;

b. Directors, officers, employees, and others authorized by the air carrier or another carrier traveling pursuant to a pass interchange agreement;

c. Travel agents being transported for the purpose of familiarizing themselves with the carrier's services;

d. Witnesses and attorneys attending any legal investigation in which such carrier is involved;

e. Persons injured in aircraft accidents, and physicians, nurses, and others attending such persons;

f. Any persons transported with the object of providing relief in cases of general epidemic, natural disaster, or other catastrophe;

g. Any law enforcement official, including any person who has the duty of guarding government officials who are traveling on official business or traveling to or from such duty;

h. Guests of an air carrier on an inaugural flight or delivery flights of newly-acquired or renovated aircraft;

i. Security guards who have been assigned the duty to guard such aircraft against unlawful seizure, sabotage, or other unlawful interference;

j. Safety inspectors of the National Transportation Safety Board or the FAA in their official duties or traveling to or from such duty;

k. Postal employees on duty in charge of the mails or traveling to or from such duty;

l. Technical representatives of companies that have been engaged in the manufacture, development or testing of a particular type of aircraft or aircraft equipment, when the transportation is provided for the purpose of in-flight observation and subject to applicable FAA regulations;

m. Persons engaged in promoting air transportation;

n. Air marshals and other Transportation Security officials acting in their official capacities and while traveling to and from their official duties; and

o. Other authorized persons, when such transportation is undertaken for promotional purpose.

35. Passenger, Revenue. A passenger for whose transportation an air carrier receives commercial remuneration. This includes, but is not limited to, the following examples:

a. Passengers traveling under publicly available tickets including promotional offers (for example two-for-one) or loyalty programs (for example, redemption of frequent flyer points);

b. Passengers traveling on vouchers or tickets issued as compensation for denied boarding or in response to consumer complaints or claims;

c. Passengers traveling at corporate discounts;

d. Passengers traveling on preferential fares (Government, seamen, military, youth, student, etc.);

e. Passengers traveling on barter tickets; and

f. Infants traveling on confirmed-space tickets.

36. Reporting event. The event that signals the Participating Carrier to report a Ticketed Itinerary.

37. Reporting Carrier. An Air Carrier or Foreign Air Carrier that is required to report the T-100/T-100(f).

38. TCN. The Transmission Control Number record is a record used to share information about a Ticketed Itinerary between a GDS and multiple Carriers or between one Carrier and multiple Carriers.

39. Ticketed Itinerary. The collection of information from an Air Travel Ticket, issued by an Air Carrier or Foreign Air Carrier to a Revenue Passenger. The collection of information about a journey shall be unique for the Issuing Carrier for the Date of Issue.

40. True O&D. A passenger's view of a purposeful trip of one or more Flight-Stages, one or more of which include travel by scheduled air transportation, measured from the beginning of the trip (origin) until the end of the trip (destination), where the individual intends to conduct business or engage in leisure activity.

41. United States. The States of the United States, the District of Columbia, and the territories and possessions of the United States, including the territorial sea and the overlying airspace.

List of Subjects

14 CFR Part 241

• Air carriers
• Reporting and recordkeeping requirements
• Uniform System of Accounts

14 CFR Part 249

• Air carriers
• Reporting and recordkeeping requirements
• Truth in lending
• Uniform System of Accounts

N. Proposed Rule

Accordingly, the Department proposes to amend 14 CFR chapter II as follows:

PART 241—UNIFORM SYSTEM OF ACCOUNTS AND REPORTS FOR AIR CARRIERS

1. The authority citation for part 241 continues to read as follows:

Authority: 49 U.S.C. 329 and chapters 401, 411, 417.

2. Sections 26-1 through 26-5 and an undesignated center heading are added to read as follows:

Passenger Origin—Destination Survey

Section 26-1 Applicability

(a) Participating Carriers shall provide data for the Passenger Origin-Destination Survey (O&D Survey). Participating Carriers shall prepare information from Ticketed Itineraries for submission as described in Appendix A to this section and as described in the Passenger Origin-Destination Survey Directives issued by the Department of Transportation.

(b) Participating Carriers with special operating characteristics may request a waiver and propose an alternative O&D Survey collection and reporting procedure to the Department. Such departures from the prescribed O&D Survey practices shall not be authorized unless approved in writing by the Department.

(c) A Participating Carrier in the O&D Survey shall include:

(1) All Air Carriers issuing Ticketed Itineraries for interstate or international scheduled passenger services and that operate aircraft with 15 or more seats, and

(2) Foreign air carriers licensed to hold out service to the U.S. under 49 U.S.C. 41301 and that have been granted antitrust immunity for an alliance with a U.S. Air Carrier partner under 49 U.S.C. 41308 and 41309 and operate aircraft with 15 or more seats when issuing Ticketed Itineraries that include an airport within the U.S.

(d) Carriers that qualify as a Participating Carrier after the effective date of this regulation will be required to:

(1) File O&D Survey data for testing purposes no more than 30 days after qualifying as a Participating Carrier and

(2) File O&D Survey data as of the first day of the month that begins more than 60 days and no more than 91 days after the month the carrier qualifies as a Participating Carrier.

Section 26-2 Submission of Reports to the O&D Survey

(a) Each Participating Carrier shall submit to the Department, in the manner specified in the Passenger Origin-Destination Survey Directives, information about Ticketed Itineraries it issues. The information about Ticketed Itineraries to be reported is found in Appendix A of this section.

Section 26-3 Certification and Authentication

(a) Certification. (1) Each Participating Carrier shall designate an elective officer, an executive or a director or such other person as may be authorized by the carrier to serve as the Designated Company Official. The Participating Carrier shall disclose the individual's name, title and such contact information as the Department specifies in the Passenger Origin-Destination Survey Directives.

(2) The Participating Carrier's Designated Company Official shall:

(a) Certify the authenticity and accuracy of the Participating Carrier's submission of O&D Survey data to the Department,

(b) Maintain the accuracy of the Participating Carrier's information on file with the Department,

(c) Provide the Department with a source and accuracy statement, and

(d) Authorize a Designated Carrier Liaison to act on behalf of the Participating Carrier in operational matters pertaining to the company's collection and submission of the O&D Survey.

(3) The certification of the reports, embodied in Schedule A thereof, shall read as follows: I, the undersigned (Title of certifying official) of the (Full name of the Participating Carrier) do certify that reports and supporting documents which are submitted for the O&D Survey are prepared under my direction: that I carefully examined them and that they correctly reflect the accounts and records of the company, and to the best of my knowledge and belief are a complete and accurate statement of the Ticketed Itineraries to be reported in the periods reported; that the various items herein reported were determined in accordance with the standard accounting practices of the company; and that the data contained herein are reported on a basis consistent with that of the preceding report except as specifically noted in explanations that preceded the submission of the Ticketed Itineraries.

(b) Source and Accuracy Statement. The Participating Carrier's Source and Accuracy Statement shall disclose the Participating Carrier's data source, data collection methodology, and measures to assure data quality.

(c) Designated Company Official. A Participating Carrier's Designated Company Official may authorize an agent to prepare and to file the O&D Survey information on behalf of the Participating Carrier. Such an arrangement does not alter the obligation of the Participating Carrier to deliver the information properly, deliver the information promptly, and certify the completeness and accuracy of the information.

(d) Designated Carrier Liaison. The Participating Carrier's Designated Carrier Liaison will serve as the point of contact between the Department and the Participating Carrier for the resolution of reporting issues.

Section 26-4 Retention and Accessibility of Data

Each Participating Carrier shall maintain its prescribed operating statistics in a manner and at such locations as will permit ready accessibility for examination by representatives of the Department. The record retention requirements are prescribed in part 249 of this chapter.

Section 26-5 Confidentiality of Data.

Data covering the operations of Air Carriers and Foreign Air Carriers will not be available to the public when the data would cause damaging competitive impact on the Air Carriers or Foreign Air Carriers and when adverse effects upon the public interest would result from disclosure of the data.

3. Appendix A to section 26 is added to read as follows:

Appendix A to Section 26—Instructions to Participating Carriers for Collecting and Reporting Passenger Origin-Destination Survey Statistics

I. Participating Carriers shall provide data for the O&D Survey. The authority for these instructions is found in 14 CFR part 241, section 26, and in the CAB Sunset Act of 1984 (Pub. L. 94-443).

(a) Submission of reportable itineraries.

(1) All Ticketed Itineraries issued by the Participating Carrier shall be submitted to the Department as described in the Passenger Origin-Destination Survey Directives issued by the Department of Transportation.

(2) The source of information for the O&D Survey data shall be the information recorded about a Ticketed Itinerary issued to a Revenue Passenger by a Participating Carrier. The Participating Carrier shall record the information about the complete routing of the Ticketed Itinerary by Flight-Stage the first time the Participating Carrier receives evidence that the passenger has used the Ticketed Itinerary for transportation. Evidence that the passenger has used the Ticketed Itinerary for transportation shall include notification from the Participating Carrier's own accounting function or flight boarding control function that the passenger has been transported or notification from another Air Carrier or Foreign Air Carrier that the Ticketed Itinerary has been used for transportation.

(b) Information about Ticketed Itineraries to be reported.

(1) The data to be recorded and reported from Participating Carriers shall include the following data elements for each Ticketed Itinerary:

a. Issuing Carrier Identifier: The Issuing Carrier's assigned IATA recognized three-character identification code.

b. Ticketed Itinerary Identifier: The alphanumeric identifier for the Ticketed Itinerary.

c. Date of Issue: The local date on which the Ticketed Itinerary was issued.

d. Fare Amount: The monetary amount the Issuing Carrier receives from the ticket purchaser, excluding government imposed taxes and fees, and including the carrier-imposed fees and surcharges, such as fuel surcharges, for the carriage of a passenger and allowable free baggage on the passenger's complete itinerary, denominated in U.S. dollars, and accurate to two decimal places, rounded.

e. Ticketing Entity Outlet Type: The location type code for the distribution channel that issued the Ticketed Itinerary. The Department's codes for use in this data element will be listed in the Passenger Origin-Destination Survey Directives issued by the Department and will be consistent with standard industry practice.

f. Customer Loyalty Program Identifier: The Carrier or alliance customer loyalty program identifying code under which the passenger accrues benefits. The Department's codes for use in this data element will be listed in the Passenger Origin-Destination Survey Directives issued by the Department.

g. Customer Loyalty Program Award Indicator: The one character identifying code to indicate that customer loyalty program credits were expended in obtaining the Ticketed Itinerary.

h. Number of Passengers: The count of passengers traveling on the Ticketed Itinerary.

i. Itinerary Copy Date: 02-14-05 the date that the Participating Carrier copied O&D Survey information from the Ticketed Itinerary.

(2) The data to be recorded and reported, as many times as necessary, from Participating Carriers shall include the following data elements repeated for each tax or fee imposed by local, state, and national government authorities in all countries that are applicable to the Ticketed Itinerary:

a. Government-imposed tax/fee identifier: The identification code of each government-imposed tax and government-imposed fee. The Department's codes for use in this data element will be listed in the Passenger Origin-Destination Survey Directives issued by the Department.

b. Government-imposed tax/fee amount: This field will contain the value of the tax or fee specified by the identifier that precedes it, denominated in U.S. dollars and accurate to two decimal places, rounded.

(3) The data to be recorded and reported, as many times as necessary, from Participating Carriers shall include the following data elements for each Flight-Stage in the order that they appear in the Ticketed Itinerary:

a. Flight-Stage Sequence Number: The two character ordinal sequence number beginning with 01 that uniquely identifies the Flight-Stage of a Ticketed Itinerary.

b. Flight-Stage Origin Airport: The IATA location identifier of the airport from which a Flight-Stage departs. For intermodal ticketed ground stations, such as a bus station or a train station, that station should be treated as an airport.

c. Flight-Stage Destination Airport: The IATA location identifier of the airport in which a Flight-Stage arrives. For intermodal ticketed ground stations, such as a bus station or a train station, that station should be treated as an airport.

d. Marketing Carrier Code: The IATA Airline Designator of the Air Carrier or Foreign Air Carrier marketing the Flight-Stage.

e. Operating Carrier Code: The IATA Airline Designator of the Air Carrier or Foreign Air Carrier operating the equipment used on the Flight-Stage.

f. Scheduled Flight Date: The date on which the Flight-Stage is scheduled to depart.

g. Master Flight Number: The scheduled Carrier Code and true flight number under which the flight inventory is managed.

h. Scheduled Departure Time: The local time the flight is scheduled to depart from the Flight-Stage Origin Airport.

i. Scheduled Arrival Time: The local time the flight is scheduled to arrive at the Flight-Stage Destination Airport.

j. Scheduled Arrival Date: The local date on which the flight is scheduled to arrive at the Flight-Stage Destination Airport.

k. Fare Basis Code/Ticket Designator: The carrier-assigned alphanumeric code identifying the fare by class, qualification, and restriction associated with the Flight-Stage.

l. Ticketing Class of Service: a one-character code indicating the service cabin within the aircraft in which the passenger is scheduled to be seated under the fare rules stated for each Flight-Stage of the Ticketed Itinerary.

(c) Means of reporting.

(1) Participating Carriers shall report data in an electronic Report Transmission according to the instructions in the Passenger Origin-Destination Survey Directives issued by the Department of Transportation.

(d) Corrections to reported information.

(1) When Participating Carriers discover that data have been incorrectly reported or improperly reported, the Participating Carrier shall immediately notify the Department of Transportation according to the instructions found in the Passenger Origin-Destination Survey Directives issued by the Department. The Participating Carrier shall correct the problem and resend the complete record of information about the incorrectly or improperly reported Ticketed Itineraries according to the procedures found in the Passenger Origin-Destination Survey Directives.

II. Glossary

Airline Designator means an airline's IATA identifier for the purpose of marketing flights and listing them in standard publications such as the OAG.

Air Travel Ticket means one or more paper or electronic documents or other evidence of contract issued by an Air Carrier or Foreign Air Carrier to record information about a passenger's complete itinerary of travel when air travel comprises at least one part of the journey.

Customer Loyalty Program Identifier means the identifying code of the Carrier or alliance customer loyalty program under which the passenger accrues benefits.

Date of Issue means the date an Air Carrier or Foreign Air Carrier issued the Ticketed Itinerary to a passenger.

Designated Carrier Liaison means the individual that will serve as the point of contact between the Department and the Participating Carrier for the resolution of operational submission issues.

Designated Company Official means an elective officer, an executive or a director or such other person as may be authorized by the carrier to certify the accuracy of information supplied to the Department and to specify a Designated Carrier Liaison.

Fare Amount means the monetary amount the Issuing Carrier receives from the ticket purchaser, excluding government-imposed taxes and fees, and including the Carrier-imposed fees and surcharges, such as fuel surcharges, for the carriage of a passenger and allowable free baggage on the passenger's complete itinerary denominated in U.S. dollars and accurate to two decimal places, rounded.

Fare Basis Code/Ticket Designator means the alphanumeric code identifying the fare by class, qualification, and restriction associated with the Flight-Stage.

Fare Category means a summary category of fare basis codes.

Flight-Coupon Stage means the portion of an itinerary that lies between two sequential points of a Ticketed Itinerary. A passenger's Flight-Coupon Stage may involve multiple takeoffs and landings. A Flight-Coupon Stage may be on any scheduled transportation held out and ticketed by the Issuing Carrier.

Flight-Stage Destination Airport means the airport identifier of the airport in which a Flight-Stage arrives. For intermodal ticketed ground stations, such as a bus station or a train station, that station should be treated as an airport.

Flight-Stage Origin Airport means the airport identifier of the airport from which a Flight-Stage departs. For intermodal ticketed ground stations, such as a bus station or a train station, that station should be treated as an airport.

Flight-Stage Sequence Number means the two character ordinal sequence number beginning with 01, followed by 02 etc. that uniquely identifies each Flight-Stage of a Ticketed Itinerary in the sequence to be traveled by the passenger. Government-Imposed Tax/Fee Amount means the monetary amount of the tax or fee associated with the corresponding Government-Imposed Tax/Fee Identifier, denominated in U.S. Dollars and accurate to two decimal places, rounded.

Government-Imposed Tax/Fee Identifier means the identification code of a tax or fee.

Issuing Carrier means the plating Air Carrier or Foreign Air Carrier that is responsible for the ticket stock on which the itinerary is issued. Also, the Air Carrier or Foreign Air Carrier that is responsible for collecting the remuneration for the fare and the taxes and fees from the purchaser of a Ticketed Itinerary.

Issuing Carrier Identifier means the IATA recognized identification code on file at the Department that uniquely identifies the carrier that issued the Ticketed Itinerary.

Itinerary Copy Date means the date that the Participating Carrier copied O&D Survey information from the Ticketed Itinerary.

Marketing Carrier Code means the IATA Airline Designator of the Air Carrier or Foreign Air Carrier that appears on a Ticketed Itinerary as if it will operate the Flight-Stage, whether or not it actually operates the Flight-Stage.

Marketing Flight Number means the number assigned by the Marketing Carrier to the Flight-Stage that appears in the Ticketed Itinerary.

Master Flight Number means the scheduled Carrier Code and true flight number under which the flight inventory is managed.

Number of Passengers means the count of passengers traveling on a Ticketed Itinerary.

One-way Trip means a journey taken by a Passenger, described on Ticketed Itinerary, from the One-way Trip Origin to the One-way Trip Destination.

One-way Trip Origin means the first airport of a One-way Trip.

One-way Trip Destination means the final airport of a One-way Trip.

Operating Carrier Code means the carrier code of the Air Carrier or Foreign Air Carrier operating the equipment used on the Flight-Stage.

Participating Carrier means an Air Carrier or Foreign Air Carrier that is required to report the O&D Survey.

Report Transmission means a regularly scheduled electronic transmission of information about a collection of Ticketed Itineraries including the transmission identification information specified in the Passenger Origin-Destination Survey Directives issued by the Department.

Scheduled Arrival Time means the local time at which the Flight-Stage is scheduled to arrive at the Flight-Stage Destination Airport.

Scheduled Departure Time means the local time at which the Flight-Stage is scheduled to depart from the Flight-Stage Origin Airport.

Scheduled Flight Date means the local date on which the Flight-Stage is scheduled to depart.

Source and Accuracy Statement means a disclosure of the Participating Carrier's data source, data collection methodology, and measures taken to assure the quality of the data submitted to the Department.

Ticketed Itinerary means the collection of information from an Air Travel Ticket, issued by an Air Carrier or Foreign Air Carrier to a Revenue Passenger.

Ticketed Itinerary Identifier means the primary identifier of a Ticketed Itinerary. The Ticketed Itinerary Identifier must be unique for the Air Carrier or Foreign Air Carrier for the Date of Issue. The Ticketed Itinerary Identifier may a combination of alphanumeric characters and blanks.

Ticketing Class of Service means a one-character code indicating the service cabin within the aircraft in which the passenger is scheduled to be seated for each Flight-Stage of the Ticketed Itinerary.

Ticketing Entity Outlet Type means the identifier of the distribution channel through which the Ticketed Itinerary was issued.

PART 249—PRESERVATION OF AIR CARRIER RECORDS

4. The authority citation for part 249 continues to read as follows:

Authority: 49 U.S.C. 329 and chapters 401, 411, 413, 417.

§ 249.20

[Amended]

5. Amend the table in § 249.20 by adding a new entry 11 to read as follows:

§ 249.20

Preservation of records by certificated air carriers.

* * * * *

Schedule of Records

Category of records	Retention period
*    *    *    *    *
11. All books, records, and other source and summary documentation that support the carrier's T-100 reports filed under Rural Service Improvement Act of 2002 (Pub. L. 107-206)	7 years
*    *    *    *    *

Issued in Washington, DC on: January 31, 2005.

Norman Y. Mineta,

Secretary.

Footnotes

1.  This was true at some carriers until the advent of electronic ticketing in the mid-1990s.

2.  Office of Inspector General Audit Report Number AV-1998-086 Feb. 24, 1998 p. iii.

3.  For example, Airline Industry Metrics, Trends on Demand and Capacity, Aviation System Performance, Airline Finances, and Service to Small Airports Number: CC-2004-006 (http://www.oig.dot.gov/​show_​pdf.php?​id=​1237).

4.  Summary may be found at http://ostpxweb.dot.gov/​aviation/​rural/​scexec.pdf).

5.  Information may be obtained from http://apo.faa.gov/​2001ConferenceProc/​proc2001/​procdoc.htm.

6.  Source: http://commerce.senate.gov/​pdf/​woerth010903.pdf.

7.  “Passenger Origin-Destination Data Submitted by Carriers. AV-1998-086 issued Feb. 24, 1998 pp. 33.

8.  One work month = 173.3 staff hours = ((40 hours per week * 52 weeks) divided by 12 months).

9.  One work month = 173.3 staff hours = ((40 hours per week * 52 weeks) divided by 12 months).

10.  One work month = 173.3 staff hours = ((40 hours per week * 52 weeks) divided by 12 months).

11.  Source: http://www.procurement.irs.treas.gov/​tirno04r00005/​amend04/​wage_​determination.txt. Although these rates are for West Virginia, they are the most recent wages established by the government and are comparable, in the past, to rates assigned to other states and districts. We believe that they represent an accurate estimate of wages for this set of positions, effective in 2004. Furthermore, we do adjust the wages for this employment category to reflect the specialized requirements of the airline industry.

12.  Federal Aviation Administration Reauthorization Act of 1996, Section 1213 (Pub. L. 104-264).

13.  The average hourly wage for the industry was estimated to be $10.40 in 1997 (See 62 FR 6718, February 13, 1997). While wages have, in general, increased over the past seven years, many employees in the airline industry have experienced wage reductions and concessions. We therefore estimate the average hourly wage for the airline industry today to be $15.60 (a 50% increase over the 1997 average hourly wage).

14.  One work week = 40 hours.

15.  Source: http://www.procurement.irs.treas.gov/​tirno04r00005/​amend04/​wage_​determination.txt. Although these rates are for West Virginia, they are the most recent wages established by the government and are comparable, in the past, to rates assigned to other states and districts. We believe that they represent an accurate estimate of wages for this set of positions, effective in 2004. Furthermore, we do adjust the wages for this employment category to reflect the specialized requirements of the airline industry.

16.  The average hourly wage for the industry was estimated to be $10.40 in 1997 (See 62 FR 6718, February 13, 1997). While wages have, in general, increased over the past seven years, many employees in the airline industry have experienced wage reductions and concessions. We therefore estimate the average hourly wage for the airline industry today to be $15.60 (a 50% increase over the 1997 average hourly wage).