E8-25544. Safe Harbor Procedures for Employers Who Receive a No-Match Letter: Clarification; Final Regulatory Flexibility Analysis

AGENCY:

U.S. Immigration and Customs Enforcement, DHS.

ACTION:

Supplemental final rule.

SUMMARY:

The Department of Homeland Security (DHS) is finalizing the Supplemental Proposed Rule published on March 26, 2008 and reaffirming regulations providing a “safe harbor” from liability under section 274A of the Immigration and Nationality Act for employers that follow certain procedures after receiving a notice—either a “no-match letter” from the Social Security Administration (SSA), or a “notice of suspect document” from DHS—that casts doubt on the employment eligibility of their employees. DHS is also correcting a typographical error in the rule text promulgated in August 2007.

DATES:

This final rule is effective as of October 28, 2008.

ADDRESSES:

The comments on the supplemental proposed rule and the proposed rule on docket DHS Docket No. ICEB-2006-0004, may be reviewed by one of the following methods:

• Federal eRulemaking Portal: http://www.regulations.gov.
• In person at U.S. Immigration and Customs Enforcement, 500 12th St., SW., 5th Floor, Washington DC 20024. Contact Joe Jeronimo, U.S. Immigration and Customs Enforcement, Telephone: [phone] (not a toll-free number) for an appointment.

FOR FURTHER INFORMATION CONTACT:

Joe Jeronimo, U.S. Immigration and Customs Enforcement, 500 12th St., SW., 5th Floor, Washington DC 20024. Telephone: [phone] (not a toll-free number).

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Docket

II. Background

A. History of the Rulemaking

B. Purpose of the Rulemaking

C. Supplemental Final Rule

1. Authority to Promulgate the Rule

2. “Reasoned Analysis” Supporting Perceived Change in Policy Reflected in the Final Rule

3. Anti-Discrimination Provisions of the INA

4. Regulatory Flexibility Analysis

III. Public Comments and Responses

A. Authority To Promulgate the Rule

B. “Reasoned Analysis” Supporting Perceived Change in Policy Reflected in the Final Rule

C. Anti-Discrimination Provisions of the INA

D. Regulatory Flexibility Analysis

1. Scope of Regulatory Flexibility Act Review

2. Direct and Indirect Impact

3. Baseline Costs, Unauthorized Alien Workers, and the Immigration Reform and Control Act of 1986

4. Variability of SSA Criteria for Issuing No-Match Letters

5. Base Assumptions Made in the IRFA and SEIA

6. Opportunity and Productivity Costs

7. Human Resources and Employee Tracking

8. Other Costs

9. Rehiring Seasonal Employees

10. Conclusions

E. Further Interpretation of the August 2007 Final Rule

F. Other Comments Received

IV. Changes Made in Republishing the Final Rule

V. Statutory and Regulatory Reviews

A. Administrative Procedure Act

B. Regulatory Flexibility Act

1. Need for, Objectives of, and Reasons Why the Rule is Being Considered

2. Significant Issues Raised in Public Comments

3. Description of and Estimate of the Numbers of Small Entities to Which the Rule Would Apply

4. Proposed Reporting, Recordkeeping, and Other Compliance Requirements

5. Significant Alternatives Considered

6. Minimization of Impact

C. Unfunded Mandates Reform Act of 1995

D. Small Business Regulatory Enforcement Fairness Act of 1996

E. Executive Order 12,866 (Regulatory Planning and Review)

F. Executive Order 13,132 (Federalism)

G. Executive Order 12,988 (Civil Justice Reform)

H. Paperwork Reduction Act

PART 274a—CONTROL OF EMPLOYMENT OF ALIENS

I. Docket

Comments on the supplemental proposed rule, the proposed rule, and the Small Entity Impact Analysis may be viewed online at http://www.regulations.gov (docket ICEB-2006-0004), or in person at U.S. Immigration and Customs Enforcement, Department of Homeland Security, 500 12th St., SW., 5th Floor, Washington, DC 20024, by appointment. To make an appointment to review the docket, call telephone number 202-732-3979 (not a toll-free number).

II. Background

A. History of the Rulemaking

DHS published a proposed rule in June 2006 that proposed a method for employers to limit the risk of being found to have knowingly employed unauthorized aliens after receiving a letter from the SSA—known as a “no-match letter”—notifying them of mismatches between names and social security numbers provided by their employees and the information in SSA's database, or after receiving a letter from DHS—called a “notice of suspect document”—that casts doubt on their employees' eligibility to work. 71 FR 34281 (June 14, 2006). A sixty-day public comment period ended on August 14, 2006.

DHS received approximately 5,000 comments on the proposed rule from a variety of sources, including labor unions, not-for-profit advocacy organizations, industry trade groups, private attorneys, businesses, and other interested organizations and individuals. The comments varied considerably; some commenters strongly supported the rule as proposed, and others were critical of the proposed rule and suggested changes. See http://www.regulations.gov,, docket number ICEB-2006-0004.

DHS published a final rule on August 15, 2007, setting out safe harbor procedures for employers that receive SSA no-match letters or DHS notices. 72 FR 45611 (Aug. 15, 2007). Each comment received was reviewed and considered in the preparation of the August 2007 Final Rule. The August 2007 Final Rule addressed the comments by issue rather than by referring to specific commenters or comments.

On August 29, 2007, the American Federation of Labor and Congress of Industrial Organizations and others filed suit seeking to enjoin implementation of the August 2007 Final Rule in the United States District Court for the Northern District of California. AFL-CIO v. Chertoff, No. 07-4472-CRB, D.E. 1 (N.D. Cal. Aug. 29, 2007). The district court granted plaintiffs' initial motion for a temporary restraining order, AFL-CIO v. Chertoff, D.E. 21 (N.D. Cal. Aug. 31, 2007) (order granting motion for temporary restraining order and setting schedule for briefing and hearing on preliminary injunction), and on October 10, 2007 granted plaintiffs' motion for preliminary injunction. AFL-CIO v. Chertoff, 552 F.Supp.2d 999 (N.D. Cal. 2007) (order granting motion for preliminary injunction).

The district court concluded that plaintiffs had raised serious questions about three aspects of the August 2007 Final Rule. Specifically, the court questioned whether DHS had: (1) Supplied a reasoned analysis to justify what the court viewed as a change in the Department's position—that a no-match letter may be sufficient, by itself, to put an employer on notice, and thus impart constructive knowledge, that employees referenced in the letter may not be work-authorized; (2) exceeded its authority (and encroached on the authority of the Department of Justice (DOJ)) by interpreting the anti-discrimination provisions of the Immigration Reform and Control Act of 1986 (IRCA), Public Law 99-603, 100 Stat. 3359 (1986), 8 U.S.C. 1324b; and (3) violated the Regulatory Flexibility Act, 5 U.S.C 601 et seq., by not conducting a regulatory flexibility analysis. 552 F.Supp.2d at 1006. Following its entry of the preliminary injunction, the district court stayed proceedings in the litigation. See AFL-CIO v. Chertoff, D.E. 149 (N.D. Cal. Dec. 14, 2007) (minute entry).

DHS published a supplemental notice of proposed rulemaking in March 2008 to address the specific issues raised by the court in the preliminary injunction order. 73 FR 15944, 45, 46-47 (March 26, 2008). In the supplemental proposed rulemaking, DHS reviewed past government communications about SSA no-match letters to clarify the history of the Department's policy on the significance of those letters, and supplied additional “reasoned analysis” in support of the policy set forth in the rule. 73 FR at 15947-50. DHS also clarified that the authority to interpret and enforce the anti-discrimination provisions of the IRCA rests with DOJ, 73 FR at 15950-51, and provided an initial regulatory flexibility analysis, 73 FR at 15951, 52-54, including a small entities analysis. Docket ICEB-2006-0004-0233.

The public comment period on the supplemental proposed rule ended on April 25, 2008. DHS received approximately 2,950 comments on the supplemental proposed rule from a variety of sources, including labor unions, not-for-profit advocacy organizations, industry trade groups, private attorneys, businesses, and other interested organizations and individuals.

A number of public comments were the product of mass-mailing campaigns, resulting in DHS receiving identical or nearly identical electronic filings during the comment period. Other comments included multiple-signature petition drives that presented a specific point of view. Many comments expressed opinions on immigration policy generally but provided little substantive information or supporting documentation that DHS could use to refine its judgment on the efficacy of the rulemaking or that was pertinent to the issues raised by the supplemental proposed rulemaking.

DHS viewed every comment received from a different source as a separate comment, notwithstanding similarities in wording. When multiple comments were received from the same source but via different media (e.g. electronic and mail), DHS attempted to identify and correlate the comments. DHS reviewed the substance of every comment and considered the substance of the comments in formulating this final rule. We summarize the substance of the comments received below.

During the public comment period, DHS received requests that the comment period be extended. DHS reviewed these requests and concluded that they presented no novel or difficult issues justifying an extension of the comment period, particularly in light of the rulemaking's extensive history, as well as the limited number of issues raised by the district court and addressed in the supplemental proposed rule. Accordingly, DHS declines to extend the comment period.

In developing this supplemental final rule, DHS has considered the entire administrative record of the August 2007 Final Rule, as well as the record of proceedings in the pending litigation, including arguments made in the various motions and briefs, and orders of the district court, that were relevant to the issues addressed in this action. AFL-CIO v. Chertoff, D.E. 129 (N.D. Cal. Oct. 1, 2007) (certified administrative record); D.E. 146-2 (N.D. Cal. Dec. 4, 2007 (errata)) (hereinafter “Administrative Record”). The docket of the United States District Court for the Northern District of California is a public record and the documents contained therein are available from the court clerk's office.

After considering the full record, including the comments received in response to the supplemental notice of proposed rulemaking, DHS has made adjustments to the cost calculations in the Initial Regulatory Flexibility Analysis (IRFA) and prepared a Final Regulatory Flexibility Analysis (FRFA), finalized the additional legal analysis set out in the supplemental notice of proposed rulemaking, and determined that the rule should issue without change. Therefore this final rule reaffirms the text of the August 2007 Final Rule without substantive change and makes one typographical correction.

B. Purpose of the Rulemaking

The Federal Government has been aware for many years that employment in the United States is a magnet for illegal immigration, and that a comparison of names and social security numbers submitted by employers against SSA's data provides an indicator of possible illegal employment. In 1997, the U.S. Commission on Immigration Reform found the following:

Reducing the employment magnet is the linchpin of a comprehensive strategy to deter unlawful immigration. Economic opportunity and the prospect of employment remain the most important draw[s] for illegal migration to this country. Strategies to deter unlawful entries and visa overstays require both a reliable process for verifying authorization to work and an enforcement capacity to ensure that employers adhere to all immigration-related labor standards.

* * * * *

The Commission concluded that the most promising option for verifying work authorization is a computerized registry based on the social security number; it unanimously recommended that such a system be tested not only for its effectiveness in deterring the employment of illegal aliens, but also for its protections against discrimination and infringements on civil liberties and privacy.

* * * * *

The federal government does not have the capacity to match social security numbers with [Immigration and Naturalization Service (INS)] work authorization data without some of the information captured on the I-9. Congress should provide sufficient time, resources, and authorities to permit development of this capability.

U.S. Comm'n on Immigration Reform, Becoming an American: Immigration and Immigrant Policy 113-14, 117 (1997) (emphasis in original); Administrative Record at 139-140, 143.

Similarly, the Federal Government has been long aware of the potential for abuse of social security numbers by aliens who are not authorized to work in the United States. Such abuse has been the subject of numerous public reports by the Government Accountability Office and the SSA's Inspector General, as well as congressional hearings. See, e.g., Administrative Record, at 35-661; Government Accountability Office, Report to the Subcommittee on Terrorism, Technology and Homeland Security, Committee on the Judiciary, U.S. Senate, Estimating the Undocumented Population: A “Grouped Answers” Approach to Surveying Foreign-Born Respondents (GAO Rept. No. GAO-06-775, Sept. 2006) (describes alternative means of gathering interview data from undocumented aliens to reduce the “question threat” to some respondents because they fear that a truthful answer could result in negative consequences); Subcommittee on Oversight and Subcommittee on Social Security, Committee on Ways and Means, U.S. House of Representatives, Social Security Number and Individual Taxpayers Identification Number Mismatches and Misuse, 108th Cong., 2nd Sess., No. 108-53 (March 10, 2004).

The illegal alien population in the United States and the number of unauthorized workers employed in the United States are both substantial. See, e.g., J. Passel, Pew Hispanic Center, The Size and Characteristics of the Unauthorized Migrant Population in the U.S. (March 2006), found at http://pewhispanic.org/​files/​factsheets/​17.pdf (estimating approximately 11.2 million illegal aliens in the United States; approximately 7.2 million illegal aliens in the workforce); M. Hoefer, N. Rytina & C. Campbell, Office of Immigration Statistics, Policy Directorate, U.S. Department of Homeland Security, Estimates of the Unauthorized Immigrant Population Residing in the United States: January 2006 (August 2007) found at http://www.dhs.gov/​xlibrary/​assets/​statistics/​publications/​ill_​pe_​2006.pdf (estimating unauthorized population of 11,550,000 as of January 2006).

The scale of the problem that this rule seeks to address—that is, the unlawful employment of aliens not authorized to work in the United States—has become more well-defined through the rulemaking and related litigation. The comments submitted in response to the initial proposed rule in 2006 by organizations such as Western Growers, and the public statements by representatives of such organizations, have been bracingly frank:

In the midst of the combustive debate over immigration reform, we in agriculture have been forthright about the elephant in America's living room: Much of our workforce is in the country illegally—as much as 70%.

T. Nassif, “Food for Thought,” The Wall Street Journal, Nov. 20, 2007, at A19. See also Docket ICEB-2006-0004-0145 (August 14, 2006), Administrative Record at 1306 (comments of the National Council of Agricultural Employers, suggesting over 76% of agricultural workers are not authorized to work in the United States). DHS recognizes this critical fact—that many employers are aware that a substantial portion of their workforce is unauthorized—and has therefore taken steps within the Department's existing authorities to assist employers in complying with the law.

Public and private studies in the administrative record of this rulemaking make clear that social security no-match letters identify some portion of unauthorized aliens who are illegally employed in the United States. One private study concluded that “most workers with unmatched SSNs are undocumented immigrants.” C. Mehta, N. Theodore & M. Hincapie, Social Security Administration's No-Match Letter Program: Implications for Immigration Enforcement and Workers' Rights (2003) at i; Administrative Record at 309, 313.

Based on the rulemaking record and the Department's law enforcement expertise, DHS finds that there is a substantial connection between social security no-match letters and the lack of work authorization by some employees whose SSNs are listed in those letters. While social security no-match letters do not, by themselves, conclusively establish that an employee is unauthorized, DHS's (and legacy INS's) interactions with employers that receive no-match letters have consistently shown that employers are also aware that an employee's appearance on a no-match letter may indicate the employee lacks work authorization. Nevertheless, as Mehta, Theodore & Hincapie found, SSA's no-match letters currently “do[] not substantially deter employers from retaining or hiring undocumented immigrants. Twenty-three percent of employers retained workers with unmatched SSNs who failed to correct their information with the SSA.” C. Mehta, N. Theodore & M. Hincapie, supra at ii; Administrative Record at 314.

Some employers may fail to respond to no-match letters because they have consciously made the illegal employment of unauthorized aliens a key part of their business model or because they conclude that the risk of an immigration enforcement action is outweighed by the cost of complying with the immigration laws by hiring only legal workers. See C. Mehta, N. Theodore & M. Hincapie, supra at 2, 20-30; Administrative Record at 314, 316, 334-44 (noting employer “complaints” over loss of illegal workforce when employees are asked to correct their SSN mismatches, as well as the practice by some employers of encouraging workers to procure new fraudulent documents to provide cover for their continued employment). DHS's interactions with employers have also shown that many law-abiding employers are unsure of their obligations under current immigration law after they receive a no-match letter, and that some employers fear allegations of anti-discrimination law violations if they react inappropriately to no-match letters.

In light of these facts, DHS has concluded that additional employer guidance on how to respond to SSA no-match letters will help law-abiding employers to comply with the immigration laws. Accordingly, in this final rule, DHS outlines specific steps that reasonable employers may take in response to SSA no-match letters, and offers employers that follow those steps a safe harbor from ICE's use of SSA no-match letters in any future enforcement action to demonstrate that an employer has knowingly employed unauthorized aliens in violation of section 274A of the Immigration and Nationality Act (INA), 8 U.S.C. 1324a.

C. Supplemental Final Rule

1. Authority to Promulgate the Rule

Congress has delegated to the Secretary of Homeland Security the authority to promulgate regulations that implement, interpret and fill in the administrative details of the immigration laws. INA section 103(a), 8 U.S.C. 1103(a); Homeland Security Act of 2002, Public Law 107-296, sections 102(a)(3), (b)(1), and (e), 110 Stat. 2135 (Nov. 25, 2002) (HSA), as amended, 6 U.S.C. 112(a)(3), (b)(1), and (e). Under Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-45 (1983), the courts afford due deference to agency interpretations of these laws as reflected in DHS's rules. The Executive Branch may, as appropriate, announce or change its policies and statutory interpretations through rulemaking actions, so long as the agency's decisions rest on a “rational connection between the facts found and the choice made.” Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins., 463 U.S. 29, 43 (1983).

DHS is authorized by the HSA and the INA to investigate and pursue sanctions against employers that knowingly hire or continue to employ unauthorized aliens or do not properly verify their employees' employment eligibility. HSA sections 102(a)(3), 202(3), 441, 442, 6 U.S.C. 112(a)(3), 251, 252; INA section 274A(e), 8 U.S.C. 1324a(e). All persons or entities that hire, recruit or refer persons for a fee for employment in the United States must verify the identity and employment eligibility of all employees hired to work in the United States. INA section 274A(a)(1)(B), (b)(1), (b)(2) 8 U.S.C. 1324a(a)(1)(B), (b)(1), (b)(2). Under the INA, this verification is performed by completing an Employment Eligibility Verification form (Form I-9) for all employees, including United States citizens. INA section 274A(b)(1), (b)(2), 8 U.S.C. 1324a (b)(1), (b)(2); 8 CFR 274a.2. An employer, or a recruiter or referrer for a fee, must retain the completed Form I-9 for three years after hiring, recruiting or referral, or, where the employment extends longer, for the life of the individual's employment and for one year following the employee's departure. INA section 274A(b)(3), 8 U.S.C. 1324a(b)(3). These forms are not routinely filed with any government agency; employers are responsible for maintaining these records, and they may be requested and reviewed by DHS Immigration and Customs Enforcement (ICE). See 71 FR 34510 (June 15, 2006).

DHS's authority to investigate and pursue sanctions against employers that knowingly hire or continue to employ unauthorized aliens necessarily includes the authority to decide the evidence on which it will rely in such enforcement efforts. It also includes the authority to decide the probative value of the available evidence, and the conditions under which DHS will commit not to rely on certain evidence. Under the prior regulations, an employer who had received an SSA no-match letter or DHS letter and was charged with knowing employment of unauthorized aliens could defend against an inference that the employer had constructive knowledge of the workers' illegal status by showing that the employer had concluded, after exercising reasonable care in response to the SSA no-match letter or DHS letter, that the workers were in fact work-authorized. 8 CFR 274a.1(l)(1) (2007). Those regulations, however, provided no detailed guidance on what would constitute “reasonable care.” In the August 2007 Final Rule—as supplemented by this final rule—DHS announces its interpretation of INA section 274A and limits its law enforcement discretion by committing not to use an employer's receipt of and response to an SSA no-match letter or DHS letter as evidence of constructive knowledge, if the employer follows the procedures outlined in the rule. This limitation on DHS's enforcement discretion—this safe harbor—is well within the rulemaking powers of the Secretary of Homeland Security. See, e.g., Lopez v. Davis, 531 U.S. 230, 240-41 (2001) (upholding categorical limitation of agency discretion through rulemaking). This rule does not affect the authority of SSA to issue no-match letters, or the authority of the Internal Revenue Service (IRS) to impose and collect taxes, or the authority of DOJ to enforce the anti-discrimination provisions of the INA or adjudicate notices of intent to fine employers.

The ongoing litigation involving the August 2007 Final Rule does not constrain DHS's authority to amend and reissue the rule. The Executive Branch's amendment of regulations in litigation is a natural evolution in the process of governance. As the United States Court of Appeals for the District of Columbia has noted:

It is both logical and precedented that an agency can engage in new rulemaking to correct a prior rule which a court has found defective. See Center for Science in the Public Interest v. Regan, 727 F.2d 1161, 1164-65 (D.C. Cir. 1984); Action on Smoking and Health v. CAB, 713 F.2d 795, 802 (D.C. Cir. 1983). Where an injunction is based on an interpretation of a prior regulation, the agency need not seek modification of that injunction before it initiates new rulemaking to change the regulation.

NAACP, Jefferson County Branch v. Donovan, 737 F.2d 67, 72 (D.C. Cir. 1984). See generally Thorpe v. Housing Auth. of Durham, 393 U.S. 268, 281-82 (1969).

As noted in the supplemental notice of proposed rulemaking, the district court enjoined implementation of the August 2007 Final Rule and the issuance of SSA no-match letters containing an insert drafted by DHS. AFL-CIO v. Chertoff, D.E. 137 (N.D. Cal. 2007) (preliminary injunction); 73 FR at 15947. The preliminary injunction did not prohibit further rulemaking by DHS. The district court subsequently stayed proceedings in the litigation to allow for further rulemaking. AFL-CIO v. Chertoff, D.E. 142 (stay motion); 144 (statement of non-opposition); 149 (minute order staying proceedings pending new rulemaking) (N.D. Cal. 2007). Accordingly, not only does DHS continue to have the authority to revise and finalize this rulemaking but the orders of the district court contemplate such rulemaking action.

2. “Reasoned Analysis” Supporting Perceived Change in Policy Reflected in the Final Rule

An agency action is arbitrary and capricious if the agency fails to examine relevant data and articulate a satisfactory explanation for its action including a “rational connection between the facts found and the choice made.” Motor Vehicle Mfrs. Ass'n v. State Farm Mutual Auto. Ins., 463 U.S. 29, 43 (1983). In its order granting the preliminary injunction, the district court found that “DHS has sufficiently articulated a rational connection between the facts found and the choice made.” 552 F.Supp.2d at 1010. The district court expressed concerns, however, that DHS had not sufficiently articulated a rationale for what the court saw as DHS's “change” in position on the significance of SSA no-match letters when promulgating that August 2007 Final Rule. While the district court acknowledged that the preamble to the August 2007 Final Rule remained consistent with DHS's and legacy INS's prior informal guidance by “assur[ing] employers that ‘an SSA no-match letter by itself does not impart knowledge that the identified employees are unauthorized aliens,’ ” 559 F.Supp.2d at 1009 (quoting 72 FR 45616), the court concluded that “DHS decided to change course” in the text of the August 2007 Final Rule by “provid[ing] that constructive knowledge may be inferred if an employer fails to take reasonable steps after receiving nothing more than a no-match letter.” Id. Having identified what it believed to be a change in DHS's position, the court concluded that “DHS may well have the authority to change its position, but because DHS did so without a reasoned analysis, there is at least a serious question whether the agency has ‘casually ignored’ prior precedent in violation of the APA.” 552 F.Supp.2d at 1010.

DHS provided in the supplemental proposed rule an extensive review of the non-precedential correspondence and public reports relating to the value of SSA no-match letters as an indicator that individuals listed in a letter may not be authorized to work in the United States and the obligations of employers to respond to such letters. 73 FR at 15947-48. That review showed that neither the former INS nor DHS had issued a formal or precedential statement of agency policy regarding the significance of SSA no-match letters, and that, therefore, there was no agency precedent that had been “casually ignored” in DHS's promulgation of the August 2007 Final Rule. It also showed that DHS's consistent, if informal, view of SSA no-match letters has been that (1) SSA no-match letters do not, by themselves, establish that an employee is unauthorized, (2) there are both innocent and non-innocent reasons for no-match letters, but (3) an employer may not safely ignore SSA no-match letters, and (4) an employer must be aware of and comply with the anti-discrimination provisions of the INA. The position reflected in the August 2007 Final Rule—that a no-match letter, and an employer's response to such a letter could, in the totality of the circumstances, constitute proof of an employer's constructive knowledge that an employee is not authorized to work in the United States—was consistent with the informal agency interpretations offered to employers over the past decade.

Nevertheless, in light of the court's concerns that DHS had changed its position on these issues in the August 2007 Final Rule, the supplemental notice of proposed rulemaking set forth the “reasoned analysis” sought by the court and identified four significant reasons for the issuance of this rule: (1) The need to resolve ambiguity and confusion among employers regarding their obligations under the INA following receipt of an SSA no-match letter; (2) the growing evidence and consensus within and outside government that SSA no-match letters are a legitimate indicator of possible illegal work by unauthorized aliens; (3) DHS's view that SSA's criteria for sending employee no-match letters helps to focus those letters on employers that have potentially significant problems with their employees' work authorization; and (4) the established legal principle that employers may be found to have knowingly employed unauthorized alien workers in violation of INA section 274A based on a constructive knowledge theory. 73 F.R. 15949-50.

a. Need for Clear Guidance Regarding No-Match Letters

As was noted in the supplemental notice of proposed rulemaking, one key justification for issuance of this rule is to eliminate ambiguity regarding an employer's responsibilities under the INA upon receipt of a no-match letter. As one business organization with nationwide membership commented in response to the initial publication of the proposed rule in 2006:

Disagreement and confusion [of an employer's obligations upon receipt of a no-match letter] are rampant and well-intended employers are left without a clear understanding of their compliance responsibilities. [Organization] members have had substantial concerns regarding whether mismatch letters put them on notice that they may be in violation of the employment authorization provisions of the immigration law, since the Social Security card is one of the most commonly used employment authorization documents.

Administrative Record at 1295 (comment from National Council of Agricultural Employers, Aug. 14, 2006). See also id. at 849 (comment by the National Federation of Independent Business: “Clarification of the employer's obligation on receiving a no-match letter and the safe harbor provided for in the proposed rule is critical.”).

As noted above, all previous agency guidance was in letters responding to individual queries from employers, members of Congress, or other interested parties—neither the INS nor DHS had ever released any formal statement of agency policy on the issue. In addition, agency correspondence over the years was heavily caveated, at times even equivocal, and although more recent letters from DHS had articulated more clearly employers' obligations upon receiving a no-match letter, those letters did not purport to supplant prior statements by legacy INS. In the absence of a clear, authoritative agency position on the significance of no-match letters, employers and labor organizations were left free to stake out positions that best served their parochial interests, by in some cases misconstruing language in the no-match letter aimed at preventing summary firings or discriminatory practices as instead commanding employers to turn a blind eye to the widely-known fact that unauthorized alien workers would often be listed in those letters. In the face of this ambiguity, well-meaning employers' responses to SSA no-match letters were also affected by concern about running afoul of the INA's antidiscrimination provisions. Thus, employers concluded that the risks of inaction in the face of no-match letters—with the possibility of being found to have knowingly employed unauthorized workers in violation of INA 274A—was outweighed by the risks of embarking on an investigation after receiving a no-match letter only to face charges of discrimination.

The August 2007 Final Rule was designed to remedy this confused situation by reminding employers of their obligation under the INA to conduct due diligence upon receipt of SSA no-match letters, and by formally announcing DHS's view that employers that fail to perform reasonable due diligence upon receipt of SSA no-match letters or DHS suspect document notices risk being found to have constructive knowledge of the illegal work status of employees whose names or SSNs are listed. Further, because the constructive knowledge standard applies a “totality of the circumstances” test to the facts of a particular case, and is therefore not reducible to bright-line rules, the August 2007 Final Rule sought to provide greater predictability through a clear set of recommended actions for employers to take, and assured employers that they would not face charges of constructive knowledge based on SSA no-match letters or DHS letters that had been handled according to DHS's guidelines.

b. No-Match Letters Are Legitimate Indicators of Possible Illegal Work by Unauthorized Aliens

DHS's reasoned analysis on the evidentiary value of SSA no-match letters in the August 2007 Final Rule, and in this supplemental rulemaking, also includes the growing evidence and consensus within and outside government that SSN no-matches are a legitimate indicator of possible illegal work by unauthorized aliens. The SSA Office of the Inspector General (SSA IG) has reported, after reviewing earnings suspense file data for tax years 1999-2001, that fraudulent use of SSNs ^[1] was widespread in the service, restaurant, and agriculture industries and that such fraud was a significant cause of SSA no-matches:

[OIG] identified various types of reporting irregularities, such as invalid, unassigned and duplicate SSNs and SSNs belonging to young children and deceased individuals. While we recognize there are legitimate reasons why a worker's name and SSN may not match SSA files, such as a legal name change, we believe the magnitude of incorrect wage reporting is indicative of SSN misuse. Employees and industry association representatives acknowledged that unauthorized noncitizens contribute to SSN misuse.

Office of the Inspector General, Social Security Administration, Social Security Number Misuse in the Service, Restaurant, and Agriculture Industries, Report A-08-05-25023, at 2 (April 2005), Administrative Record at 456. See generally Administrative Record at 35-661.

SSA no-match letters have also formed a basis for multiple criminal investigations by ICE and prosecutions on charges of harboring or knowingly hiring unauthorized aliens.^[2]

DHS's view—that no-match letters regularly identify unauthorized alien workers—was also overwhelmingly affirmed by those who submitted comments on the proposed rule in 2006. See, e.g., Administrative Record at 866 (comment by U.S. Chamber of Commerce: “It is estimated that annually 500,000 essential workers enter the U.S. to perform much needed labor without work authorization. * * * The proposed regulation will strip needed workers from employers without providing employers with an alternative legal channel by which to recruit to fill the gaps. * * *”); id. at 874 (comment by Essential Workers Immigration Coalition including same statement); id. at 850 (comment by National Federation of Independent Business: “a substantial number of workers identified by no-match letters are undocumented immigrants who are unable to provide legitimate social security numbers”); id. at 858 (comment by Western Growers opposing the rule on grounds that “it would have a most devastating effect on California and Arizona agriculture, where an estimated 50 to 80 percent of the workers who harvest fruit, vegetables and other crops are illegal immigrants”); id. at 887 (comment by American Immigration Lawyers Association: “[T]he proposed regulation admittedly will ‘smoke out’ many unauthorized workers.”); id. at 1306 (comment by National Council of Agricultural Employers suggesting that, as a conservative estimate, 76% of agricultural workers are not authorized to work in the United States, that “employers would likely lose a significant part of their workforces,” and that “a substantial number of workers would not return to work” when faced with the requirement to verify work authorization “because they would be unable to do so”). See also AFL-CIO v. Chertoff, 552 F.Supp.2d at 1008 (“th[e] Court cannot agree with plaintiffs” fundamental premise that a no-match letter can never trigger constructive knowledge, regardless of the circumstances”).

c. SSA's Procedures Better Target No-Match Letters to Employers With Potential Workforce Problems

SSA's criteria for sending employer no-match letters also inform DHS's position in the August 2007 Final Rule and in this supplementary rulemaking. SSA does not send employer no-match letters to every employer with a no-match. Instead, SSA sends letters only when an employer submits a wage report reflecting at least 11 workers with no-matches, and when the total number of no-matches in a given wage report represents more than 0.5% of the employer's total Forms W-2 in the report.

In addition, SSA has continued to refine the wage reporting process in ways that help to reduce administrative error resulting in a no-match letter. Employers filing more than 250 Forms W-2 are required to file electronically (see 42 U.S.C. 405(c)(2)(A); 20 CFR 422.114; 26 CFR 301.6011-2), and electronic filing of Forms W-2 has risen from 53% of all employee reports in FY2003 to over 80% in FY2007—a 51% increase.^[3] This direct electronic filing substantially reduces the likelihood that SSA errors—such as during data entry of the information submitted on a paper Form W-2—would result in discrepancies in the wage reports. Employers also have access to SSA's system for identifying name-SSN mismatches at the time they file the wage reports. That system can only be used to verify current or former employees and only for wage reporting (Form W-2) purposes. Employers who use SSA's system are able to eliminate most no-matches in their reports and thereby significantly reduce their likelihood of receiving a no-match letter.

DHS is also aware that SSA has developed a series of computerized error-checking routines to resolve certain common errors that result in unmatched name and SSN. These routines resolve name discrepancies caused by misspellings, typographical errors, first name and last name transpositions, and female surname changes (e.g. marriage or divorce). They can also resolve discrepancies from the use of a derivative nickname instead of a proper name or from scrambling compound or hyphenated surnames. The routines can also resolve SSN discrepancies such as numerical transpositions.

GAO has reported that approximately 60 percent of no-matches in recent tax years' wage reports are corrected by SSA's algorithms. See Government Accountability Office, Social Security: Better Coordination among Federal Agencies Could Reduce Unidentified Earnings Reports (GAO Report 05-154, 2005), Administrative Record at 400. See also Office of the Inspector General, Social Security Administration, Effectiveness of the Single Select Edit Routine (Audit Report A-03-07-17065, Sept. 2007). While these routines cannot resolve all discrepancies, they reduce the number of inadvertent no-matches that are reported to employers.

DHS believes that, taken together, these efforts better direct no-match letters to employers that have potentially significant problems with their employees' work authorization. Employers with stray mistakes or de minimis inaccuracies are much less likely to receive no-match letters.

d. The Longstanding Principle That Employers May Be Liable for INA Violations Based on Constructive Knowledge

Both pre-existing regulations and consistent case law demonstrate that an employer can be found to have violated INA section 274A(a)(2), 8 U.S.C. 1324a(a)(2), by having constructive rather than actual knowledge that an employee is unauthorized to work. The concept of constructive knowledge appeared in the first regulation that defined “knowing” for purposes of INA section 274a, 8 CFR 274A.1(l)(1) (1990); 55 FR 25928 (June 25, 1990). As noted in the preamble to the original regulation, that definition of knowledge is consistent with the Ninth Circuit's decision in Mester Mfg. Co. v. INS, 879 F.2d 561, 567 (9th Cir. 1989) (holding that, after receiving information that employees were suspected of having presented false documents to show work authorization, the employer had constructive knowledge of unauthorized status because the employer failed to make inquiries or take appropriate corrective action). See also New El Rey Sausage Co. v. INS, 925 F.2d 1153, 1158 (9th Cir. 1991).

The rulemaking record demonstrates that employers have continued to demand clear guidance on appropriately responding to SSA no-match letters, consistent with their obligations under the INA. It also demonstrates a well-established consensus that the appearance of employees' SSNs on an SSA no-match letter may indicate lack of work authorization. The record also shows that SSA's practices in generating no-match letters helps to focus those letters on employers that, in DHS's view, have non-trivial levels of employees with SSN mismatches in their workforce, and existing law clearly establishes that employers may be charged with constructive knowledge when they fail to conduct further inquiries in the face of information that would lead a person exercising reasonable care to learn of an employee's unauthorized status.

This reasoned analysis supports DHS's position in the August 2007 Final Rule—that an employer's failure to conduct reasonable due diligence upon receipt of an SSA no-match letter can, in the totality of the circumstances, establish constructive knowledge of an employee's unauthorized status. Assuming, as did the district court, that this position constituted a change from prior statements in informal agency correspondence, DHS has now provided additional—and sufficient—reasoned analysis to support that change.

3. Anti-Discrimination Provisions of the INA

The preamble to the August 2007 Final Rule said that employers that adopt the rule's safe harbor procedures to verify employees' identity and work authorization must apply them uniformly to all employees who appear on employer no-match letters. Failure to do so, the preamble warned, may violate the INA's anti-discrimination provisions. The preamble further noted that employers that follow the safe harbor procedures uniformly and without regard to perceived national origin or citizenship status will not be found to have engaged in unlawful discrimination. 72 FR 45613-14. The DHS insert prepared to accompany the no-match letter had similar language. AFL-CIO v. Chertoff, D.E. 7, Exh. C. (N.D. Cal. Aug. 29, 2007).

The district court questioned DHS's authority to offer what the court viewed as interpretations, rather than mere restatements, of settled anti-discrimination law, noting that DOJ, not DHS, has authority for interpretation and enforcement of the INA's anti-discrimination provisions. The court concluded that DHS appeared to have exceeded its authority. 552 F.Supp.2d at 1011.

DHS recognizes the jurisdiction of DOJ over enforcement of the anti-discrimination provisions in section 274B of the INA (8 U.S.C. 1324b). As stated in the preamble to the August 2007 Final Rule, “DOJ—through its Office of Special Counsel for Immigration-Related Unfair Employment Practices—is responsible for enforcing the anti-discrimination provisions of section 274B of the INA, 8 U.S.C. 1324b.” 72 FR 45,614. The August 2007 Final Rule also stated that DHS's rule “does not affect * * * the authority of DOJ to enforce the anti-discrimination provisions of the INA or adjudicate notices of intent to fine employers.” Id. DHS does not have the authority to obligate the DOJ or the Office of Special Counsel, and the August 2007 Final Rule did not purport to make any such obligation. Whether an employer has engaged in unlawful discrimination in violation of INA 274B is a determination that is made by DOJ through the Office of Special Counsel. A statement by one agency about the authority of another agency does not, in and of itself, encroach on the authority of that other agency, and DHS's statements in the August 2007 Final Rule were reviewed through an interagency process that was created to improve the internal management of the Executive Branch. Executive Order 12866, 58 FR 51735 (Oct. 4, 1993), as amended by Executive Order 13258, 67 FR 9385 (Feb. 28, 2002), as amended by Executive Order 13422, 72 FR 2763 (Jan. 23, 2007).

Nevertheless, in light of the district court's concerns, DHS rescinds the statements in the preamble of the August 2007 Final Rule discussing the potential for anti-discrimination liability faced by employers that follow the safe harbor procedures set forth in the August 2007 Final Rule.^[4] DHS has also revised the language in its insert letter that will accompany the SSA no-match letters. These changes do not alter existing law or require any change to the rule text.

DHS recognizes the concerns raised by commenters that discrimination litigation may be brought against them. As expressed by one commenter:

One of the greatest potential costs faced by employers as a result of this rulemaking is the increased likelihood of discrimination lawsuits brought about by the required termination of employees who cannot resolve “mismatches.” DHS' retraction of the assurances it attempted to provide in the proposed rule only increases the uncertainty that employers face. Moreover, even meritless claims brought by terminated employees will require significant expenses in legal fees and related costs to defend, and unless DHS can remove jurisdiction in all courts in which such actions might be brought, it cannot prevent these expenses. Our reality is that we will be “attacked” by numerous organizations * * * as we have been in the past.

ICEB-2006-0004-0498.1 at 1-2 (emphasis in original); see also ICEB-2006-0004-0571.1 at 2; ICEB-2006-0004-0679.1 at 2.

While DHS lacks the authority to announce interpretations of the anti-discrimination provisions of the INA, DOJ possesses such authority, and persons seeking guidance regarding employers' anti-discrimination obligations in following the safe harbor procedures in the August 2007 Final Rule, as modified by this supplemental rulemaking, should follow the direction provided by DOJ published in today's edition of the Federal Register, and available on the Web site of the Office of Special Counsel for Immigration-Related Unfair Employment Practices, at http://www.usdoj.gov/​crt/​osc/​htm/​Nomatch032008.htm. Employers may also seek advice on a case-by-case basis through OSC's toll-free employer hotline: [phone]. The Department continues to urge employers to apply the safe harbor procedures in this rule to all employees referenced in an SSA no-match letter or a DHS notice uniformly and without regard to perceived national origin or citizenship status.

4. Regulatory Flexibility Analysis

In its decision enjoining implementation of the August 2007 Final Rule, the district court construed the safe harbor in the rule as effectively creating compliance obligations for employers that received no-match letters. Doubting the voluntary nature of the safe harbor rule, the court found it likely that small businesses would incur significant costs to enter the safe harbor:

Because failure to comply subjects' employers to the threat of civil and criminal liability, the regulation is the practical equivalent of a rule that obliges an employer to comply or to suffer the consequences; the voluntary form of the rule is but a veil for the threat it obscures. The rule as good as mandates costly compliance with a new 90-day timeframe for resolving mismatches. Accordingly, there are serious questions whether DHS violated the RFA by refusing to conduct a final flexibility analysis.

552 F.Supp.2d at 1013 (internal quotations and citations omitted). In light of the district court's conclusion that a regulatory flexibility analysis would likely be required, DHS published an initial regulatory flexibility analysis (IRFA) in the supplemental proposed rule, 73 FR at 15952-54, and placed on the docket for public comment the Small Entity Impact Analysis, Supplemental Proposed Rule: Safe Harbor Procedures for Employers Who Receive a No-Match Letter, ICEB-2006-0004-0233 (hereinafter, the “SEIA”).

DHS continues to view the August 2007 Final Rule and this supplemental rule as interpretive, and does not believe that these rulemakings bear any of the hallmarks of a legislative rule. See Hemp Industries Ass'n v. Drug Enforcement Admin., 333 F.3d 1082, 1087 (9th Cir. 2003) (identifying three circumstances in which a rule is legislative); Syncore Int'l Corp. v. Shalala, 127 F.3d 90, 94 (D.C. Cir. 1997) (interpretive rule “typically reflects an agency's construction of a statute that has been entrusted to the agency to administer” and a statement of policy “represents an agency position with respect to how it will treat—typically enforce—the governing legal norm. By issuing a policy statement, an agency simply lets the public know its current enforcement or adjudicatory approach.”). DHS is not invoking its legislative rulemaking authority to mandate a specific action upon a certain event. Instead, this rulemaking informs the public of DHS's interpretation of Section 274A of the INA and describes how DHS will exercise its discretion in enforcing the INA's prohibition on knowing employment of unauthorized aliens. Although the district court questioned whether DHS has changed its position on the evidentiary force of no-match letters in enforcement proceedings against employers, neither the August 2007 Final Rule nor this supplemental rulemaking departs from any prior legislative rule. See Oregon v. Ashcroft, 368 F.3d 1118, 1134 (9th Cir. 2004). As noted above, the only record of the agency's previous position lies in correspondence between the agency and individuals and employers seeking advice on specific questions.

Thus, although DHS continues to believe that the Regulatory Flexibility Act does not mandate the analysis that has been undertaken here, see Central Texas Tel. Coop. v. FCC, 402 F.3d 205, 214 (D.C. Cir. 2005), the Department provided an IRFA and supporting economic analysis, and has now prepared a Final Regulatory Flexibility Analysis (FRFA) in response to the district court's concerns.

As the United States Court of Appeals for the Ninth Circuit has noted, the Regulatory Flexibility Act (RFA) “imposes no substantive requirements on an agency; rather, its requirements are `purely procedural' in nature. * * * To satisfy the RFA, an agency must only demonstrate a `reasonable, good-faith effort' to fulfill its requirements.” Ranchers Cattlemen Action Legal Fund v. USDA, 415 F.3d 1078, 1101 (9th Cir. 2005). See also Envtl. Def. Ctr. v. EPA, 344 F.3d 832, 879 (9th Cir. 2003) (“Like the Notice and Comment process required in administrative rulemaking by the APA, the analyses required by the RFA are essentially procedural hurdles; after considering the relevant impacts and alternatives, an administrative agency remains free to regulate as it sees fit.”).

The RFA, by definition, does not apply to individuals. Where it applies, the RFA requires agencies to analyze the impact of rulemaking on “small entities.” Small entities include small businesses, not-for-profit organizations that are independently owned and operated and are not dominant in their fields, and governmental jurisdictions with populations of less than 50,000. 5 U.S.C. 601(3), (5)-(6). Small businesses are defined in regulations promulgated by the Small Business Administration. 13 CFR 121.201.

The RFA provides that an initial regulatory flexibility analysis (IRFA) shall contain:

(1) A description of the reasons why action by the agency is being considered;

(2) A succinct statement of the objectives of, and legal basis for, the proposed rule;

(3) A description of and, where feasible, an estimate of the number of small entities to which the proposed rule will apply;

(4) A description of the projected reporting, recordkeeping and other compliance requirements of the proposed rule, including an estimate of the classes of small entities which will be subject to the requirement and the type of professional skills necessary for preparation of the report or record;

(5) An identification, to the extent practicable, of all relevant Federal rules which may duplicate, overlap or conflict with the proposed rule.

5 U.S.C. 603(b). Furthermore, an IRFA must also contain:

a description of any significant alternatives to the proposed rule which accomplish the stated objectives of applicable statutes and which minimize any significant economic impact of the proposed rule on small entities. Consistent with the stated objectives of applicable statutes, the analysis shall discuss significant alternatives such as—

(1) The establishment of differing compliance or reporting requirements or timetables that take into account the resources available to small entities;

(2) The clarification, consolidation, or simplification of compliance and reporting requirements under the rule for such small entities;

(3) The use of performance rather than design standards; and

(4) An exemption from coverage of the rule, or any part thereof, for such small entities.

5 U.S.C. 603(c). The RFA does not require that these elements be considered in a specific manner, following a prescribed formula or content. Given the nature of rulemaking, and its diversity, agencies develop IRFAs in a manner consistent with the statute and the rulemaking itself.^[5]

The IRFA provided with the supplemental notice of proposed rulemaking contained the elements listed in 5 U.S.C. 603(b) as well as the discussion of significant regulatory alternatives required by 5 U.S.C. 603(c). The supplemental proposed rule explicitly requested comments on the economic aspects of the analysis and on the discussion of regulatory alternatives. Publication of the supplemental proposed rule received significant media coverage. The U.S. Small Business Administration Office of Advocacy (Advocacy) hosted a small business roundtable shortly after publication of the supplemental proposed rule to collect comments from interested small businesses and submitted a public comment letter based on this input. The comments provided by Advocacy are addressed in the analysis below. As noted above, the supplemental proposed rule and accompanying IRFA received nearly 3,000 comments from the public, including a significant number of comments specifically addressing the IRFA and the underlying SEIA.

DHS has reviewed the comments received on the IRFA and has concluded that the IRFA complied with the statutory standards for such an analysis and provided the public sufficient information to submit informed comments regarding the possible impact of this rule.

In light of comments that identified plausible regulatory alternatives or areas needing further clarification or adjustments in the economic model underlying the SEIA, DHS has revised the analysis and assembled a FRFA. The RFA requires that a FRFA contain:

(1) A succinct statement of the need for, and objectives of, the rule;

(2) A summary of the significant issues raised by the public comments in response to the initial regulatory flexibility analysis, a summary of the assessment of the agency of such issues, and a statement of any changes made in the proposed rule as a result of such comments;

(3) A description of and an estimate of the number of small entities to which the rule will apply or an explanation of why no such estimate is available;

(4) A description of the projected reporting, recordkeeping and other compliance requirements of the rule, including an estimate of the classes of small entities which will be subject to the requirement and the type of professional skills necessary for preparation of the report or record; and

(5) A description of the steps the agency has taken to minimize the significant economic impact on small entities consistent with the stated objectives of applicable statutes, including a statement of the factual, policy, and legal reasons for selecting the alternative adopted in the final rule and why each one of the other significant alternatives to the rule considered by the agency which affect the impact on small entities was rejected.

5 U.S.C. 604(a). The discussion below and in the final SEIA on the docket addresses specific comments received on the IRFA and, together with the FRFA summarized in this supplemental final rule, provides the statutorily required agency assessment of comments received, projections of the number of affected small entities, description of the anticipated reporting and compliance burdens, and discussion of steps taken to limit any impact of the rule on small entities. In this way, DHS has “demonstrated a `reasonable, good-faith effort' to fulfill” the procedural and substantive requirements of the RFA.

III. Public Comments and Responses

A. Authority To Promulgate the Rule

A number of commenters challenged DHS's authority to promulgate this rule. DHS has reanalyzed its jurisdiction and authority in light of these comments, and concludes that it has the necessary authority to promulgate this final rule.

Several commenters suggested that the rule imposes an affirmative due diligence obligation on employers that does not exist in the INA once employers complete the Form I-9 process. As is explained in section II.C, supra, the INA's prohibition on “knowing” hiring or continued employment of unauthorized workers extends to employers that have constructive knowledge that an employee is unauthorized to work. The concept of constructive knowledge appeared in the first regulation that defined “knowing” for purposes of INA section 274a, 8 CFR 274A.1(l)(1) (1990); 55 FR 25,928. As noted in the preamble to that original regulation, that definition of knowledge is consistent with the Ninth Circuit's decision in Mester Mfg. Co. v. INS, 879 F.2d 561, 567 (9th Cir. 1989) (holding that when an employer who received information that some employees were suspected of having presented a false document to show work authorization, such employer had constructive knowledge of their unauthorized status when the employer failed to make any inquiries or take appropriate corrective action). See also New El Rey Sausage Co. v. INS, 925 F.2d 1153, 1158 (9th Cir. 1991). Contrary to the apparent view of some commenters, the INA does not absolve employers of any further responsibility once they have completed the initial Form I-9 verification process. The concept of constructive knowledge—and employers' responsibility to conduct reasonable due diligence in response to information that could lead to knowledge of their employees' illegal status—flows from the INA as interpreted in long-standing case law and federal regulations; it is not an invention of this rulemaking.

One commenter argued that the rule would undercut the good faith compliance defense available to employers that complete the Form I-9 employment eligibility verification process, and is therefore contrary to the INA. DHS disagrees. The affirmative defense the INA provides to employers that comply with the Form I-9 process in good faith remains available as protection against a charge of knowingly hiring unauthorized employees in violation of INA section 274A(a)(1)(A), but it has no force, by the statute's plain language, as a defense against an allegation of knowingly continuing to employ an unauthorized alien in violation of INA section 274A(a)(2). This rulemaking explains the evidentiary weight DHS may place on SSA no-match letters and DHS suspect document notices in identifying, investigating, and prosecuting employers suspected of continuing to employ unauthorized aliens in violation of section 274A(a)(2). The commenter's concern over the continuing viability of the good faith I-9 compliance defense is misplaced.

One comment also suggested that DHS could not promulgate this rule because it violates the congressional notification and review requirements of INA section 274A(d)(3), 8 U.S.C. 1324a(d)(3). That section provides that the President must notify Congress before he may make any “changes in (including additions to) the requirements of subsection (b)” of INA section 274A, which established the I-9 employment verification system. INA section 274A(d)(1)(B), 8 U.S.C. 1324a(d)(1)(B) (emphasis added).

The August 2007 Final Rule instructs employers that elect to follow the safe harbor procedures set out in the rule to confirm identity and work eligibility by filling out a new Form I-9 for any employees unable to resolve their mismatch through the 90-day process. This does not, however, constitute a change to “the requirements of subsection (b)” of INA section 274A. The procedures of the safe harbor rule are not a “requirement”; employers are encouraged to follow these procedures to limit their legal risk, but they are not compelled to do so. Moreover, while the I-9 reverification option in the safe harbor procedures is based on the I-9 process used at the time of hire, it is neither part of, nor an addendum to, the I-9 process that all employers must follow at the time of hire. Rather, the safe harbor rule helps employers to avoid violating the prohibition against knowingly continuing to employ unauthorized workers. INA section 274A(a)(2), 8 U.S.C. 1324a(a)(2).

B. “Reasoned Analysis” Supporting Perceived Change in Policy Reflected in the Final Rule

Many commenters argued that DHS had not provided an adequate “reasoned analysis” the district court suggested was necessary to support the perceived change in agency position. Several comments suggested that DHS must establish with certainty, or with some degree of confidence beyond a rational basis, that a Social Security no-match letter establishes that the indicated employee was an alien not authorized to work in the United States. Some argued that the rule would be arbitrary and capricious unless DHS could refute the claim “that the SSA database is not a certain indicator of one's right to work” in the United States. ICEB 2006-0004-0732.1 at 3.

The comments suggesting that DHS must base the rule on evidence that an SSA no-match is near-conclusive proof of a listed person's illegal status misunderstand the nature of this rulemaking action. DHS has consistently stated that an SSA no-match letter, standing alone, does not conclusively establish that any employee identified in the letter is an unauthorized alien. Nor does an employer's receipt of, and response to, an SSA no-match letter always prove that the employer had constructive knowledge that any listed employees were unauthorized to work in the United States. Rather, this rulemaking announces DHS's view that a no-match letter, and an employer's response to it, may be used as evidence, evaluated in light of “the totality of the circumstances,” of an employer's constructive knowledge. This rulemaking also announces DHS's commitment that an employer that follows the safe harbor procedures set forth in the rule will always be found to have responded reasonably to the no-match letter.

As the district court noted in the pending litigation, DHS does not claim, and need not prove, that a no-match letter will always be sufficient evidence to demonstrate constructive knowledge:

The flaw in plaintiffs' argument is their assumption that receipt of a no-match letter triggers a finding of constructive knowledge in every instance. In fact, the regulation is written such that whether an employer has constructive knowledge depends `on the totality of relevant circumstances.' Depending on the circumstances, a court may agree with plaintiffs that receipt of a no-match letter has not put an employer on notice that his employee is likely to be unauthorized. But this Court cannot agree with plaintiffs' fundamental premise that a no-match letter can never trigger constructive knowledge, regardless of the circumstances.

552 F.Supp.2d at 1008.

This safe harbor rule is a rational response to DHS's regulatory finding that a no-match letter can be evidence of such knowledge—a finding amply supported in record of this rulemaking and fairly conceded even by the rule's opponents.

Some commenters argued that the SSA database was fraught with errors, and that even if SSA no-match letters were an indicator of possible illegal employment, they are too unreliable to support the evidentiary weight DHS seeks to place on them. DHS disagrees with the commenters' suggestion that SSA's records are so substantially incorrect that DHS can not rely on no-match letters generated from those records. When attempting to post wages to its Master Earnings File, SSA compares the employee names and SSNs provided by employers on Forms W-2 to the names and SSNs recorded in the Agency's NUMIDENT file. “No-matches” may result from the number holder's failure to provide SSA updated information, such as a legal name change resulting from marriage. Other “errors” result from typographical mistakes annotated on the W-2s by employers. These types of errors are being reduced by a variety of programmatic efforts, and, with direct electronic reporting of over 80% of wage data, the potential for errors resulting from the government's handling of the information is reduced.^[6] As discussed in more detail below, the effective accuracy of the SSA data from which no-match letters are derived is estimated to be 99.5 percent. Moreover, as noted above, DHS views SSA's policy of limiting issuance of no-match letters to employers whose wage reports contain a certain level of mismatches as a useful means for separating employers whose reports contain a certain non-trivial number of errors that might reasonably indicate possible illegal employment or systematic problems in the employers' recordkeeping from employers with trivial errors in their wage reports.

Other commenters noted that the supplemental proposed rule did not explicitly limit the applicability of the safe harbor procedures to the SSA's “Employer Correction Request” or “EDCOR” letter. DHS is also aware that the rule text does not explicitly identify the “EDCOR” letter from SSA—addressed to employers and containing more than ten no-match social security numbers—as the notice from SSA to which the safe harbor procedures apply. The rule text is written in general terms to allow the safe harbor procedures to apply to notices that SSA may issue in the future. DHS has made it clear, however, that the SSA notice to which the safe harbor rule applies is the “EDCOR” letter listing multiple no-matches, rather than a “Request for Employee Information” or “DECOR” letter identifying a single employee with an SSN/name no-match. First, the text of the rule clearly states that the procedures may apply where an employer receives “written notice to the employer from the Social Security Administration reporting earnings on a Form W-2 that employees' names and corresponding social security account numbers fail to match Social Security Administration records.” The reference to plural no-matches and to W-2 reports distinguishes the “EDCOR” letters addressed to employers that list multiple no-matches from any notice unrelated to a W-2 report or from “DECOR” letters addressed to a single employee or to an employer regarding a single no-match. Second, DHS explained above and in the preamble to the supplemental proposed rule that the letter listing multiple employees with SSN and name no-matches is the notice to which the rule's safe harbor applies.

C. Anti-Discrimination Provisions of the INA

A significant number of commenters repeated concerns, previously summarized and addressed in the August 2007 Final Rule, that employers would engage in illegal discrimination in reaction to this rulemaking. Such comments regarding the consistency of this regulation with existing anti-discrimination law and regarding employers' continued anti-discrimination obligations were addressed in detail in the August 2007 Final Rule, 72 FR at 45620-21, and DHS declines to revisit those issues in this supplementary rulemaking.

Other commenters objected to DHS's rescission of the statements in the preamble to the August 2007 Final Rule explaining that employers will not be engaged in unlawful discrimination under the anti-discrimination provisions of the INA if they follow the safe harbor procedures uniformly for all employees, without regard to perceived national origin or citizenship. In their view, the removal of those assurances greatly reduced the value of the safe harbor being offered in this rule, and left employers exposed to potential litigation accusing them of illegal discrimination as a result of their efforts to follow the safe harbor procedures set forth in this rulemaking.

DHS agrees that guidance on anti-discrimination compliance is important to the successful implementation of the safe harbor procedures. As DHS noted in the August 2007 Final Rule, the Department of Justice is responsible for enforcing the anti-discrimination provisions of the INA. DHS believes that the commenters' concerns are addressed in the anti-discrimination guidance from the DOJ Office of Special Counsel published in today's edition of the Federal Register.

D. Regulatory Flexibility Analysis

Commenters were divided on whether an initial regulatory flexibility analysis, and by implication a final regulatory flexibility analysis, was required. In light of the district court's conclusion that a regulatory flexibility analysis would likely be required, DHS has conducted such an analysis, supported by the small entity impact analysis (SEIA) accompanying this rulemaking. Both are summarized in greater detail in Section V.B.

The bulk of the comments regarding the RFA argued that the analysis in the IRFA and in the SEIA was flawed. Commenters argued that the scope of the analysis conducted by DHS was too narrow, that the analysis incorrectly omitted certain costs from the equation, or that the analysis was based on inaccurate assumptions about the behavior of employers and employees that might be impacted by the rule. These comments regarding the SEIA and IRFA are addressed below.

1. Scope of Regulatory Flexibility Act Review

A number of commenters conflated the requirements of the Regulatory Flexibility Act, 5 U.S.C. 601 et seq. (RFA), with the requirements of other statutory and administrative reviews. For example, commenters suggested that the RFA analysis should include reviews called for by the Congressional Review Act, 5 U.S.C. 801, the federal data quality standards guidelines, Executive Order 12866, and other statutes and executive orders. No law requires that DHS combine all of the elements of these separate reviews, and DHS declines to do so.

One commenter conceded that these additional reviews are not required by the RFA:

The DHS Safe-Harbor Rule IRFA presents estimates of costs to employers associated with following the safe-harbor procedures set forth in the proposed rule. It excludes certain costs that are not cognizable under the Regulatory Flexibility Act but are crucial for estimating the full social impact of the rule—most notably, costs borne by employees. These costs are not exempt from being counted under Executive Order 12,866 or the Congressional Review Act.

ICEB-2006-0004-0637.1 at 4. Notwithstanding this admission, the commenter repeatedly drew from standards outside the RFA to criticize the content of the IRFA. The law is clear that no other analysis is bootstrapped into the RFA. It is the case that the RFA permits agencies to prepare IRFAs in conjunction with, or as a part of, other analyses required by law, so long as the RFA's requirements are satisfied. 5 U.S.C. 605(a) (“Any Federal agency may perform the analyses required by [the RFA] in conjunction with or as a part of any other agenda or analysis required by any other law if such other analysis satisfies the provisions of such sections.”) The fact that the RFA's requirements may be managed through other analyses, however, does not expand the requirements of the RFA or compel agencies to conduct such other analyses as part of an IRFA or a FRFA. These analyses are not required by the RFA, nor are they, for the reasons set forth below, mandated for this rule under any other provision of law.

a. Executive Order 12866 and OMB Circular A-4

Executive Order No. 12866, 58 FR 51735 (Oct. 4, 1993), as amended by Executive Order 13258, 67 FR 9385 (Feb. 28, 2002), as amended by Executive Order 13422, 72 FR 2763 (Jan. 23, 2007), directs agencies subordinate to the President to assess all costs and benefits of available regulatory alternatives and, when regulation is necessary, to select regulatory approaches that maximize net benefits (including potential economic, environmental, public health and safety, and other advantages, distributive impacts, and equity). In implementing Executive Order 12866, the Office of Management and Budget has provided further internal guidance to agencies through OMB Circular A-4 (Sept. 17, 2003), found at http://www.whitehouse.gov/​omb/​circulars/​a004/​a-4.pdf. OMB Circular A-4 states that it “is designed to assist analysts in the regulatory agencies by defining good regulatory analysis * * * and standardizing the way benefits and costs of Federal regulatory actions are measured and reported.” OMB Circular A-4, at 3.

Executive Order 12866 is an exercise of the President's authority to manage the Executive Branch of the United States under Article II of the Constitution. The implementation of the Executive Orders and OMB Circulars, and other internal guidance, is a matter of Executive Branch consideration and discretion. The Executive Branch may utilize its standards under Executive Order 12866 in analyzing regulations under the RFA because the standards of the RFA and Executive Order 12866 do not conflict, but the RFA does not require use of those standards internal to the Executive Branch. The comments invoking Executive Order 12866 and OMB Circular A-4 standards to identify alleged deficiencies in the IRFA are therefore misplaced.

The fact that preparation of a regulatory impact analysis (RIA) under Executive Order 12866 is a matter of Executive Branch discretion is underscored by the terms of Executive Order 12866, section 11:

Nothing in this Executive order shall affect any otherwise available judicial review of agency action. This Executive order is intended only to improve the internal management of the Federal Government and does not create any right or benefit, substantive or procedural, enforceable at law or equity by a party against the United States, its agencies or instrumentalities, its officers or employees, or any other person.

(emphasis added). The internal, managerial nature of this and other similarly-worded Executive Orders has been recognized by the courts, and actions taken by an agency to comply with the Executive Order are not subject to judicial review. Cal-Almond, Inc. v. USDA, 14 F.3d 429, 445 (9th Cir. 1993) (citing Michigan v. Thomas, 805 F.2d 176, 187 (6th Cir. 1986)).

b. Congressional Review Act

Some comments argued that this rule is a “major rule” for purposes of the Congressional Review Act, 5 U.S.C. 801 (CRA). The CRA delays implementation, and provides a mechanism for congressional disapproval, of regulations designated as “major rules” by the Administrator of the Office of Management and Budget. Such a designation is made where OMB finds the rule has resulted in or is likely to result in (a) An annual effect on the economy of $100,000,000 or more; (b) a major increase in costs or prices for consumers, individual industries, Federal, State, or local government agencies, or geographic regions; or (c) significant adverse effects on competition, employment, investment, productivity, innovation, or on the ability of United States-based enterprises to compete with foreign-based enterprises in domestic and export markets. 5 U.S.C. 804(2). Determinations by OMB under the CRA are not subject to judicial review. 5 U.S.C. 805. OMB has not determined that this rule is a major rule and, therefore, the CRA does not apply.

2. Direct and Indirect Impact

A number of comments on the supplemental proposed rule objected that the cost estimates presented in the IRFA did not include estimates for costs other than for direct compliance with the rule. Examples of costs commenters urged DHS to take into account included potential lost wages for individuals who take time away from work to visit an SSA office or another government office to resolve the no-match, travel expenses for employees attempting to resolve a no-match, and other costs incurred by employers, such as legal fees associated with lawsuits that could be filed by work-authorized employees terminated in response to a no-match letter.

In addition, many commenters suggested that DHS's RFA analysis should include a number of other general indirect costs that allegedly could be borne by society in general—higher cost of food resulting from the disruption of the agricultural labor force where illegal employment is common, depressed wages from employers shifting from direct employment to greater reliance on temporary employment agencies, social and economic cost of unauthorized workers becoming unemployed, general impact of the rule on the “macro economy,” economic impact of employers moving operations to Mexico or other foreign countries in search of reduced labor costs and less regulation, and possible growth in the underground economy and reduction in tax revenues.

DHS disagrees. All of these comments overstate the scope of the costs that are to be considered under the RFA. The RFA requires consideration only of the direct costs of a regulation on a small entity that is required to comply with the regulation. Mid-Tex Electric Coop. v. FERC, 773 F.2d 327, 340-343 (D.C. Cir. 1985) (holding indirect impact of a regulation on small entities that do business with or are otherwise dependent on the regulated entities not considered in RFA analyses). See also Cement Kiln Recycling Coalition v. EPA, 255 F.3d 855, 869 (D.C. Cir. 2001) (In passing the RFA, “Congress did not intend to require that every agency consider every indirect effect that any regulation might have on small businesses in any stratum of the national economy * * * [T]o require an agency to assess the impact on all of the nation's small businesses possibly affected by a rule would be to convert every rulemaking process into a massive exercise in economic modeling, an approach we have already rejected.”).

No judicial precedent supports the commenters' view that indirect economic or social impacts must be considered under the RFA. These costs can be considered under other analyses and reviews that DHS and other agencies may conduct in reaching decisions on regulatory matters, but they fall outside the RFA. See, e.g., Regulatory Flexibility Improvements Act, Hearing before the Subcommittee on Commercial and Administrative Law, Committee on the Judiciary, on H.R. 682, 109th Cong., 2nd Sess. (2006), at 13 (Statement of Thomas Sullivan, Chief Counsel for Advocacy, Small Business Administration, criticizing the RFA by noting that “the RFA * * * does not require agencies to analyze indirect impacts.”).

3. Baseline Costs, Unauthorized Alien Workers, and the Immigration Reform and Control Act of 1986

A number of commenters asserted that DHS should include in the IRFA and FRFA the cost of firing unauthorized alien workers and replacing those unauthorized alien workers who voluntarily resign or are terminated by employers when the workers are unable to confirm their identity and work authorization in accordance with the safe harbor procedures in this rule. In particular commenters criticized the exclusion from the IRFA of the costs of complying with section 274A(a)(2) of the INA. That section provides:

It is unlawful for a person or other entity * * * to continue to employ [an] alien in the United States knowing the alien is (or has become) an unauthorized alien with respect to such employment.

The commenters suggested that the cost of terminating and replacing workers who an employer learns are not authorized to work in the United States should be accounted for as a cost of the rule, since that knowledge (or constructive knowledge) results from the no-match letters, and the termination and replacement costs must be borne regardless of whether they are counted as a cost of the INA or of the rule. These comments fundamentally misunderstand the requirements of the RFA, as well as the INA's longstanding prohibition against employment of unauthorized aliens.

The RFA explicitly requires DHS to “describe the impact of the proposed rule on small entities” in an initial regulatory flexibility analysis. 5 U.S.C. 603(a) (emphasis added). The Act also states that a final regulatory flexibility analysis “shall contain * * * a description of the projected reporting, recordkeeping and other compliance requirements of the rule.” 5 U.S.C. 604(a)(4) (emphasis added). The RFA does not require that DHS analyze the impact of the underlying statutory provisions in either the initial or final regulatory flexibility analysis. And it would be particularly irrational to do so here, since termination and replacement costs are already being incurred by employers attempting to comply with the INA even before this safe harbor rule goes into effect. The comments themselves make this clear: such terminations have been documented since at least 2003—three years before this rule was first proposed. C. Mehta, N. Theodore & M. Hincapie, supra, at 13-14, Administrative Record at 327-8 (approximately 53.6 percent of surveyed employers terminated workers with listed no-matches). See also ICEB-2006-0004-0688.1 at 2 (“To date, the misuse of SSA's no-match letters by employers has already resulted in countless, unjust suspensions and/or firings of low-wage, immigrant workers”); ICEB-2006-0004-0652.1 at 8 (comment by NFIB, citing Mehta, Theodore & Hincapie, supra.).

As DHS explained in the supplemental notice of proposed rulemaking, the Immigration and Nationality Act expressly prohibits employers from knowingly hiring or knowingly continuing to employ an alien who is not authorized to work in the United States. INA section 274A(a)(1), (2), 8 U.S.C. 1324a(a)(1), (2). Employers that have actual or constructive knowledge of their employees' illegal work status are statutorily obligated to cease their employment, and any costs that result are attributable to the INA, not to this safe harbor rule.

While the cost of terminating or replacing unauthorized workers cannot properly be considered a cost of this rule, some turnover involving legal workers that are unable or unwilling to resolve their mismatches through the procedures outlined in this rule could be counted as a cost of the rule for any employer that elects to follow the safe harbor procedures. Such turnover costs for legal workers were estimated in the IRFA, and are discussed in more detail below.

Several comments also suggested that employers may summarily discharge workers rather than giving them an opportunity to correct records, and argued that the impact on work-authorized employees who leave their jobs or are terminated by their employers should be included in the RFA analysis as a cost of the rule. As mentioned above, the RFA instructs agencies to examine costs and impacts to “small entities”—defined by statute as “hav[ing] the same meaning as the terms ‘small business,' ‘small organization' and ‘small governmental jurisdiction' ”—and which does not include individuals. Therefore, the commenters misread the RFA. We also note that, if an employer were to summarily terminate legal workers, the impact on such workers would be caused not by the rule but by their employer's violation of the safe harbor procedures. Any legal workers who choose not to correct their records would effectively be voluntarily resigning, perhaps calculating that the opportunity cost of correcting their records was greater than the cost of finding alternate work.

4. Variability of SSA Criteria for Issuing No-Match Letters

A number of commenters suggested that the criteria used by SSA in determining whether to issue a no-match letter was subject to future change, and that increased costs could be incurred if SSA issues more no-match letters. DHS recognizes that the impact on small entities could vary if SSA alters its matching processes or changes its criteria for issuing no-match letters. But the RFA does not require DHS to speculate about every contingency that could have some impact on small entities, such as the potential for another agency to exercise its discretion differently. Since DHS is unaware of any plans to change SSA's policies for issuing “EDCOR” no-match letters, any attempt in the IRFA or FRFA to analyze hypothetical changes in SSA policy would be mere speculation.

Some commenters also suggested that the IRFA and FRFA must cover historical data to account for the existing variability in the number of no-match letters issued from year to year, even absent any change to SSA's policies on issuing no-match letters. While such variability exists, it is largely irrelevant to the calculation under the FRA of the “impact” that may result to an average “small entity” that chooses to follow the safe harbor procedures in the rule. Changes in the number of no-match letters sent to employers in a given year may change the aggregate costs incurred by all employers that choose to follow the safe harbor procedures, but DHS has no data (and commenters have provided none) that would lead DHS to conclude that such variations would alter either the share of all no-match letters in a given year that would be received by small entities or the impact felt by a specific small entity that receives a no-match letter and decides to follow the safe harbor procedures. DHS's reliance on 2007 statistics regarding employers whose reports would have generated no-match letters for the analysis in the IRFA and SEIA was reasonable.

5. Base Assumptions Made in the IRFA and SEIA

A number of commenters disagreed with assumptions made in the IRFA and SEIA regarding the impact of the rule on small entities. DHS sought to catalog all of the assumptions underlying the analysis to make the methodology, calculations, and findings of the SEIA transparent, reproducible, and accessible for public review and comment. One commenter catalogued over thirty assumptions underlying the economic analysis provided by DHS, and noted that even this list was a subset of the analytical assumptions openly disclosed by DHS. See ICEB-2006-0004-07321.1 at 23-25. Notwithstanding DHS's transparency about the analytical underpinnings of its analysis, commenters who objected to the substance of DHS's assumptions provided little information to call into question the reasonableness of those assumptions or even to assist DHS to evaluate the strength of the commenters' objections.

The analysis required by the Regulatory Flexibility Act need not produce statistical certainty; the law requires that the DHS “demonstrate a ‘reasonable, good-faith effort' to fulfill [the RFA's] requirements.” Ranchers Cattlemen Action Legal Fund, 415 F.3d at 1101. See also Associated Fisheries of Maine v. Daley, 127 F.3d 104, 114-15 (1st Cir. 1997). The IRFA and SEIA produced by DHS in this rulemaking meet that standard. The assumptions underlying the SEIA are reasonable, and DHS has utilized the best data available to produce the IRFA and the SEIA. Where data was unavailable, DHS consistently made analytically conservative assumptions regarding the cost to employers that choose to follow the safe harbor procedures in this rule. With one exception, the public comments did not provide better data or identify additional sources for empirical data within the scope of the RFA. In analyzing the comments received and in preparing the FRFA, DHS attempted once again to ensure that the best available data is used. Individual comments regarding specific assumptions in DHS's analysis are addressed in detail below.

a. Assumptions Regarding Impact on Legal Workers

i. Accuracy of SSA Records

A number of commenters suggested that the SSA data used to generate no-match letters (the Earnings Suspense File, or “ESF” database) is generated from an SSA database (the “NUMIDENT” database) that the commenters allege contains a large number of errors that will cause work-authorized employees to appear as no-matches, and to have to correct their discrepancies.^[7] Many of these comments cited a report by the SSA Office of the Inspector General regarding errors in SSA's NUMIDENT database,^[8] to argue that the data used for the no-match letters has an error rate of 4.1 percent. Some commenters suggested that DHS not use information derived from that database for immigration enforcement purposes until the database achieves a 99.5% accuracy level. Referring to the same SSA OIG report, another commenter alleged that SSA now maintains 17.8 million mismatched records that could result in no-match letters to employers.

DHS does not agree with the commenters' inference that the overall 4.1% data discrepancy rate estimated by SSA OIG is relevant to this rulemaking, or to SSA no-match letters generally, in the way suggested by the commenters. The SSA OIG's report reviewed the accuracy of four different data fields in SSA's system—“Name,” “Date of Birth,” “Death Indication,” and “Citizenship Status”—and the study's projected 4.1% data discrepancy rate was based on the cumulative data discrepancies in all four data fields sampled. But SSA no-match letters are generated only when an employee's name and SSN submitted by an employer cannot be matched to SSA records; discrepancies in the “Date of Birth,” “Death Indication,” and “Citizenship Status” fields do not cause an employee to be listed on a no-match letter because the Forms W-2 from which no-match letters are generated do not contain this information. The SSA OIG report showed that only 0.24% of native-born U.S. citizens had a name and number mismatch, while naturalized citizens and non-citizens had a 0.49% and 1.7% mismatch rate, respectively. This yields a projected overall name and SSN mismatch rate of 0.4% (weighted average) for all records in the NUMIDENT system. Based on the SSA OIG report cited by commenters, it appears that the database that generates no-match letters already exceeds the 99.5% accuracy level proposed in the comments.

ii. Turnover Rates

The SEIA assumed that employers that follow the safe harbor procedures may face increased turnover of employees authorized to work in the United States. To the extent that a work-authorized employee resigns or is terminated for failing to resolve the no-match, the employer could be reasonably expected to incur the cost of replacing that employee. For purposes of the SEIA, DHS estimated that 2% of authorized employees identified in no-match letters might resign or be terminated due to failure to resolve a no-match, and therefore the SEIA included those turnover costs as a cost of an employer's adoption of the safe harbor procedures in the rule.

It is important to note that this figure is not, as some commenters have incorrectly claimed, an estimate of the number of legal workers that “will be fired” as a result of this rule. Nothing in the August 2007 Final Rule or in this supplemental rulemaking requires an employer to terminate an employee at the end of the 93-day no-match resolution and reverification schedule if a no-match remains unresolved. Should an employer learn in the course of that process that an employee lacks work authorization, the INA requires—as it has for over 20 years—that the employment relationship be terminated. While the regulatory safe harbor is only available if the rule's procedures are completed with 93 days, an employer may still be seen to have acted reasonably if an employee has taken longer than 93 days to resolve a no-match, depending upon the particular circumstances.

Moreover, the SEIA's estimate includes turnover caused by voluntary departures of employees who decide to seek employment elsewhere rather than resolve the no-match with SSA. Neither the government nor employers can compel employees to correct no-matches, and DHS does not have sufficient data to conclude that 100% of all legal employees will correct their no-matches within the 93-day schedule set out in the rule. DHS recognizes that it will cost employers something to replace workers if (1) some of their employees decided to leave employment after day 90, and/or (2) some employees (a) attempted but failed to complete the process of resolving their no-matches in 90 days; (b) those employees would not or could not produce alternative documents to complete a new Employee Verification Form I-9; and (c) an employer took a strict approach to terminate every person with unresolved no-matches after 93 days. DHS has, therefore, included these turnover costs in the SEIA.

Several commenters suggested that this projected turnover rate of 2% for legal workers is too low. DHS disagrees. As section III.J of the SEIA explains, there are significant economic incentives for both the employer and employee to resolve a no-match. A work-authorized employee has an incentive to both keep his or her current employment and to ensure that his or her name and SSN properly match SSA's records so that he or she will receive full credit for contributions made into Social Security and maximize the amount of Social Security benefits he or she will receive in retirement or in case of disability. At the same time, an employer has an incentive to ensure that employees resolve their no-match issues to avoid turnover in the workforce, and the SEIA assumed that employers would pay for human resources staff to assist employees to resolve a no-match, given the cost to the employer of replacing those employees. In light of these incentives, DHS's estimate of 2% was reasonable.

Although the commenters did not provide a basis for changing this assumption, DHS has added an alternative scenario in an appendix to the SEIA to examine how these turnover costs could change if the legal worker replacement rate were doubled from 2% to 4%. That additional analysis did not result in a material change in the SEIA's estimate of the rule's impact on small entities or in the reasonable regulatory alternatives that DHS could consider in this rulemaking.

iii. No-Match Resolution Process

Some commenters also suggested that DHS should reconsider the SEIA's assumption that 66% of authorized employees will be able to resolve no-matches without visiting an SSA office. DHS continues to believe that this assumption is reasonable for purposes of the analysis required by the RFA.

The SEIA made specific assumptions regarding how the employer and employee would resolve a no-match in order to estimate the costs on a per employer basis. DHS believes the cost that an employer would bear to correct a no-match typically depends on the reason for the no-match. For example, if an employer were able to determine that the no-match resulted from an internal clerical error by the employer, the employer would likely be able to correct this discrepancy quickly and inexpensively. If the employer determined that there was no clerical error, the SEIA assumed that the employer would meet with the employee to verify that the employer's records show the correct name and social security number. If the employee then determined that the employer had submitted the correct name and social security number, the employee would need to visit SSA to resolve the no-match. If the employee needs to visit SSA, the employer may incur a lost productivity cost for the time the employee was away from work.

The SEIA stated that no specific data was available to show what percentage of no-match issues were clerical errors, incorrect information submitted by the employee to the employer, or an issue that required a visit to SSA. Accordingly, the SEIA assumed one-third of the authorized employee no-matches would be clerical errors, one-third of the authorized employee no-matches would be resolved when the employer identified an error in an employer's records, and one-third of authorized employees would visit SSA to attempt to correct the no-match. None of the comments provided data that could improve on the SEIA's estimates.^[9]

Even though DHS does not have hard data on how many mismatches may be resolved at each step of the safe harbor procedures, we can reasonably expect that a significant number of no-matches will be corrected internally by the employer without requiring the employee to visit SSA. For example, several comments suggested that work-authorized employees of Latin American and Asian descent appear on no-match letters because of compound naming conventions or inconsistent transliteration that sometimes results in inadvertent errors or discrepancies in employer records. Employers can easily resolve such inadvertent errors. In addition, electronic filing of W-2 reports limits SSA staff intervention in wage report data processing and increases the likelihood that mismatches originated with—and can be most readily resolved by—the employer.

Commenters did not provide information that would lead DHS to conclude its estimate was not reasonable. Nevertheless, as with the turnover rates discussed above, DHS has provided an alternative scenario in an appendix to the SEIA to model how the no-match resolution costs would change if the percentage of authorized employees that must visit a SSA office increases from 33% to 50%. We conclude that this alternative assumption does not materially change the SEIA's estimate of the impact on small entities or point to additional regulatory alternatives that DHS could consider in this rulemaking.

b. Percentage of No-Matches Relating to Unauthorized Aliens

One commenter suggested that the SEIA was inadequate because it assumed that the general employee turnover rate would be the same for authorized and unauthorized employees. The commenter believed that this is significant because the SEIA concludes that 57% of employees listed in no-match letters already have left their jobs by the time the employer receives the no-match letter. The commenter suggested that the turnover rate is likely to be much higher for unauthorized employees, meaning that authorized employees are more likely to be still employed when a no-match letter arrives and, thus, authorized employees are more likely to be impacted by the no-match letter and the safe harbor rule.

DHS is not aware of any Department of Labor, Bureau of Labor Statistics (BLS), or other data that presents separate turnover rates for authorized and unauthorized employees. Consequently, DHS is using the best data available for turnover rates. BLS provides turnover data for the non-farm sectors and is based on all employees on the payroll, without distinguishing between those authorized and unauthorized to work in the United States. Therefore, DHS believes the BLS industry turnover rates presented in the SEIA should be considered to be weighted averages of an authorized employee turnover rate and the unauthorized employee turnover rate.^[10] DHS has clarified the SEIA to address this point. DHS has not found, and the commenters have not provided, any empirical evidence that supports a specific turnover rate or range other than the weighted average in the BLS composite rate.

Another commenter suggested that the errors in the NUMIDENT data relating to United States citizens would be less likely to appear in no-match letters, and that few U.S. citizens would be affected by no-match letters or face the possibility of termination. Another commenter noted that the SEIA assumed it is possible that only 10% of employees appearing on no-match letters are not work-authorized, and suggested that any particular no-match letter identifying 11 employees would likely list only lawful employees.

These comments highlight that DHS estimated costs based over a broad range: assuming that between 10% to 80% of employees on no-match letters were unauthorized. DHS cannot determine with certainty the rate at which authorized and unauthorized employees appear in no-match letters. Even if DHS could, the percent of unauthorized workers on any given no-match letter would likely vary by employer and by industry. Consequently, using a broad range, such as the one in the SEIA, remains the best way to present the potential economic impact of the rule on small entities.

c. Specific Wage and Occupational Assumptions

i. Replacement Costs

One commenter noted that all employment decisions in small businesses are made by the principals, who must take time to search for, interview, hire, and train new employees. According to this commenter, those same principals must process the employment paperwork and resolve any no-matches, resulting in distraction from other managerial duties. The comment suggests that the SEIA's replacement costs estimate does not account for the possible effect on the principals' ability to manage, and is therefore too low.

DHS disagrees. The SEIA estimated that replacing an authorized employee would cost approximately $5,000. In arriving at this estimate, we reviewed studies that quantified turnover costs for businesses large and small, and we found that $5,000 was a reasonable estimate of the cost incurred by the employer to replace each legal employee. Several of the economic studies on which this estimate relies are discussed in section III.J. of the SEIA. DHS believes this estimate includes reasonable estimations of the costs of hiring, training new employees, and processing paperwork.

ii. Occupational Categories

Another commenter suggested that mismatch resolution requires time and effort from more than the five occupational categories stated in the analysis, and that the SEIA underestimated the response level of companies that receive no-match letters. The commenter suggested that the more serious consequences articulated by the no-match rule would likely cause employers to involve additional occupations in the process, including the Chief Operating Officer, Chief Financial Officer, Chief Executive Officer, as well as Company Compliance Officers, senior human resources managers, paralegals, secretaries, and other clerical employees.

The SEIA does not attempt to capture every occupational title that possibly could be involved with a specific Social Security no-match letter or DHS notice of suspect document or the implementation of steps to adopt a safe harbor procedure. Rather, the intent of the SEIA is to capture levels of effort for different activities and wage levels. Each listed occupation is representative of multiple occupations at the equivalent wage. For example, the activities listed for the human resources assistant may actually be carried out by a payroll assistant.

Nevertheless, the comments correctly noted that the SEIA assumed that the most senior person that would participate in responding to no-match letters would be a senior human resources manager, and that more senior management with broad company-wide oversight responsibilities would not be involved. DHS agrees that employers that appreciate the seriousness of no-match letters may choose to include very senior managers in planning for the appropriate response, and so the final SEIA adds additional hours for a senior manager with broad company-wide oversight responsibilities.

One commenter also suggested that union representatives and union attorneys might be involved because provisions in many collective bargaining agreements prevent the termination of employees without following prescribed steps. The RFA requires DHS to consider the direct costs of the supplemental final rule. There are no requirements within the rule for the employer to follow any additional steps that may be contained within a collective bargaining agreement. Consequently, to the extent any additional costs are incurred due to the existence of collective bargaining agreements, such costs are indirect and outside of the scope of the FRFA.

One comment also pointed out that the BLS wage data was based upon surveys almost five years old—surveys conducted in November 2003, 2004, 2005 and May 2004, 2005 and 2006. Additionally, the commenter pointed out that the May 2006 Occupational Employment Statistics (OES) Estimates Technical Notes indicate that the data was collected as a result of mailing forms to 200,000 establishments, and questioned whether the BIA survey contained enough samples of the five occupations whose wages were included in the SEIA's cost calculations to provide a reliable estimate of the prevailing wage for each of those five occupations.

DHS is not persuaded by these challenges to the reliability and relevance of the BLS data. As specified in the OES Technical Notes, the OES survey consists of six panels that are surveyed over a three-year period. Each panel includes 200,000 establishments, for a total of 1.2 million establishments surveyed. In addition, the wage data obtained from the five earliest panels are all adjusted for inflation to the current period, so that the average wage computed from the 1.2 million establishments represents a wage for the latest period that was surveyed.^[11] DHS continues to believe that the BLS data is the most reasonable data to use in the SEIA; the commenter did not suggest an alternative source of data for consideration.

d. Sources of Advice Other Than Legal Counsel

Some commenters, including an association of immigration attorneys, suggested DHS underestimated the share of employers that would seek legal services in implementing the safe harbor rule. DHS disagrees. DHS assumed that one-half of employers would seek professional legal advice in implementing the safe harbor rule, and that employers that did not seek legal counsel would rely on information available from trade associations or other advocacy groups. Trade associations, in particular, are a common source for small employers seeking guidance on best business practices, as an alternative to seeking formal legal advice. Even a cursory search of the Internet and review of trade publications unearths a number of professional human resource associations, publishers, law firms, and others providing advice on responding to no-match letters that is generally consistent with the steps outlined in the rule. Further, as the district court noted in the ongoing litigation involving this rule, business organizations “such as the Chamber of Commerce of the United States of America, already have begun to develop costly programs and systems for ensuring compliance with the safe harbor framework,” AFL-CIO v. Chertoff, 552 F.Supp.2d at 1014, and it is reasonable to assume that a significant number of small businesses will follow the advice available from such organizations instead of retaining legal counsel.

6. Opportunity and Productivity Costs

Several commenters suggested that DHS include the time away from work for hourly employees, most of whom may not be paid for time spent at a Social Security office or another agency's office. Similarly, some commenters suggested that travel costs to SSA offices should be included in the SEIA. As discussed above, the RFA requires federal agencies to consider the effects of regulatory action on small businesses and other “small entities,” and individual employees are not “small entities” as defined by the RFA. Costs to employees, such as lost wages from time away from work or travel expenses, are not properly included in the analysis for the purposes of the RFA.

A number of commenters suggested that DHS include lost productivity—both from the employee being away and from human resource personnel dealing with the no-match letter—as part of the SEIA. The SEIA did include an estimate of lost productivity due to the time an employee will spend meeting with human resource personnel to discuss the no-match. The SEIA also included an estimate of the lost productivity incurred by the employer when an employee visits SSA to resolve the no-match. And the SEIA included human resource labor costs as suggested by the commenter. See, e.g., sections III.C Wage Rates, III.G Cost of Employee Time, III.K Total Compliance Cost Estimates and Appendix I: Calculation of Human Resources Labor Cost.

Some commenters asserted that the rule will be costly to employees and the economy, suggesting that, because of the millions of inaccurate records in the SSA database, hundreds of thousands of employees will be required to take time off work to visit SSA field offices to correct the discrepancies. Commenters asserted that many of these employees will be required to make multiple visits, and specifically asserted that several lawful employees had contacted the SSA up to five times to correct no-matches.

As previously noted, employees are not small entities under the RFA and the RFA does not require agencies to measure indirect impact to the economy at large. Even so, some of the commenter's assertions warrant specific response. In analyzing potential lost productivity, the SEIA estimated the time an employee might be absent from work to travel to an SSA office to correct a no-match. The SEIA cited two publicly available Westat reports on which this time estimate was based.^[12] These reports contain closely analogous data—that is, the time required to visit an SSA office to address a “tentative non-confirmation” received from the E-Verify electronic employment verification system (formerly known as Basic Pilot).^[13] The reports suggested that on average, employees spend approximately five hours to visit SSA. For the purpose of the SEIA, DHS increased that estimate to a full eight hours of lost work time (a 60% increase over the reports' findings) to account for those employees that might need to make more than one visit to resolve their no-match.

The SEIA recognizes that there may be cases in which more than one trip to SSA is necessary, and consequently assumes that employees will spend an average of eight hours away from work to resolve the no-match with SSA. Because no supporting facts are provided, DHS cannot assess the validity of the assertion made by the commenter that some employees were required to contact SSA up to five times. Our consultations with SSA suggest that such an occurrence is highly unlikely.

Another commenter suggested that the SEIA estimates the opportunity cost to the employer of a no-match employee's time in visiting SSA is the equivalent of the average employee wage rate at $27.58. The commenter suggested that this estimate is wrong, since few employers pay an employee the full value of the labor provided, and the lost production of an individual employee may be several times greater than the employee's hourly wage. The commenter concluded that the SEIA underestimates the cost of lost production.

The SEIA did not use average wages to compute opportunity costs. As explained in the SEIA, DHS used “fully-loaded” wages to estimate lost productivity. A fully-loaded wage includes such benefits as retirement and savings, paid leave (vacations, holidays, sick leave, and other leave), insurance benefits (life, health, and disability), legally required benefits such as Social Security and Medicare, and supplemental pay (overtime and premium, shift differentials, and nonproduction bonuses). DHS used data from the Bureau of Labor Statistics, the government's source on such statistics, in order to estimate the fully-loaded wage.

DHS also assumed the employer would incur a lost productivity cost of 100% of the time an authorized employee needed to visit SSA to resolve the no-match. In practice, DHS believes that some employers frequently will incur no lost productivity or opportunity cost. If employees take paid leave time to visit SSA, they will have less leave time for other personal activities. The employer, however, incurs no additional productivity losses, because the employer had already counted on that employee taking that paid leave. Lost productivity would also be minimal in industries where workers' skills are largely interchangeable. For example, if a restaurant employee or retail clerk were away from work to resolve a no-match issue, the restaurant or store would normally attempt to schedule another employee to take that shift. Given the 90 days available under the safe harbor procedures to resolve the no-match, the employer has substantial flexibility to schedule around an employee's planned absence. Consequently, to the extent employers have the capability to plan around known absences and other employees are available, the productivity loss estimated in the SEIA is higher than what employers may see in practice.

DHS understands that some businesses cannot, through planning, mitigate productivity losses attributed to employee absences to resolve mismatches. No data is available that suggests how many businesses have the ability to schedule other employees to take the place of an absent employee, and therefore mitigate costs. For this reason, DHS estimated the highest possible impact, which is a 100% productivity loss.

In addition, DHS has attempted to estimate the cost of the rule on an “average cost per firm” basis. 73 FR at 15953. There may be cases in which the productivity loss to an employer of an employee's visit to SSA is greater than the “average cost per firm” estimate in the rule. However, given the fact that the SEIA estimated a lost productivity cost 100% of the time an authorized employee needed to visit SSA at the fully loaded wage rate for a full eight hour day, DHS does not believe that the “average cost per firm” estimate is unreasonable. In fact, DHS believes that, given the conservative assumptions underlying the analysis, the estimate of lost productivity due to an employee's trip to SSA likely overstates the impact to employers.

Other commenters took the view that DHS should consider the lost productivity or replacement costs resulting not only from the time employees spend resolving their mismatch, but also the lost productivity cost of employees terminated as a result of the employer following the no-match regulations. For instance, one commenter stated that when Swift & Co. was subject to a worksite enforcement action by ICE, the company lost 1,282 employees overnight, and Swift estimated that the lost production for one day was $20 million, or about $1,560 per employee per day.

The commenter did not detail how lost production costs of $1,560 per employee per day were calculated, other than it was Swift's estimate. Moreover, the workers lost by Swift were found to be unauthorized to work in the United States. These comments appear to be citing costs incurred by an employer that discovers—through the no-match letter or some other process—that large numbers of his workforce are unauthorized to work. But those costs are outside of the scope of the rulemaking and are attributable to the immigration laws of the United States.

7. Human Resources and Employee Tracking

a. Systems Costs

Some commenters suggested that if an employer does not possess a system that allows the employer to access an employee file based on a SSN, it could take substantial time to resolve large numbers of no-matches. The commenters were concerned that because the no-match letters only provide a list of SSNs without the corresponding employee names, the time and effort required of an employer to match the SSNs on the list with employees on the payroll. One commenter suggested that it would require a month to match 500+ SSNs to the correct employee names.

DHS disagrees with these estimates. The SEIA provided what DHS believes to be a reasonable estimate for the time and cost needed to match the SSNs listed on the no-match letter to current employees. The average number of mismatched SSNs per letter is approximately 65,^[14] well under the “500+” number referenced by the commenter. Moreover, the scenario posed by the commenter—in which an employer would need to identify over 500 employees with mismatched SSNs—is a logical impossibility for many small businesses, who have fewer than 500 total employees. The SEIA's estimate, and the resulting analysis in the IRFA and FRFA of the potential impact on “small entities,” provided a reasonable estimate of this cost.

DHS also reasonably assumed that the majority of social security numbers would be stored electronically, allowing for relatively rapid screening. As discussed above, employers that file more than 250 W-2s in a given year are required to do so electronically—so that only smaller employers, with correspondingly shorter lists of mismatched SSNs, could conceivably need to conduct this matching process manually—and more than 80 percent of the FY 2007 W-2 reports were filed electronically. DHS permits storage of Employment Eligibility Verification Form I-9 under the same standards as applied by the IRS to tax accounting documentation, 8 CFR 274a.2(e)-(i), 71 FR 34510 (June 15, 2006), and an employer's process for checking the accuracy of their internal records will be especially rapid for those that keep both sets of records electronically. DHS believes, based on the evidence and commercial availability of computer systems to comply with wage and tax reporting requirements, that employers that do not store their wage, tax and employment information electronically would be relatively small and, therefore, would have fewer social security numbers to match with names. The system costs estimated in the SEIA are reasonable.

b. Reverification Costs

Several comments addressed the time and cost of the Employment Eligibility Verification Form I-9 re-verification process. For example, one commenter suggested that re-completing Forms I-9 for every employee on a no-match letter will take a significant amount of time for employers and could be a massive undertaking, depending on the number of employees on the no-match list that are still current and will need to have Form I-9 reverified.

DHS disagrees and believes the commenters overstate the costs. The proposed rule, the August 2007 Final Rule, and the supplemental proposed rule provided a series of steps that DHS would find to be a reasonable response to the receipt of a no-match letter. As DHS explained in the original proposed rule, the steps are sequential and are designed to assist employers to confirm the work authorization of their employees while encouraging employees to correct their records with SSA. DHS's rule is designed to avoid interference with the basic purpose of SSA's No-Match Letter (EDCOR) program—which is to solicit corrections to SSA's records and reduce the Earnings Suspense File—and to provide employers and employees guidance on how DHS believes they can best comply with their existing obligations under the INA. Thus, the rule specifies that employers and employees should attempt to resolve the SSN mismatch with the SSA. Only when that process has not been completed within 90 days does the rule anticipate that an employer would choose to rely on the reverification process—i.e. completing parts of a new Form I-9 as set forth in the rule—to confirm the employee's work eligibility and obtain the safe harbor protection offered by the rule.

As noted above, see section 6.a.ii, the SEIA makes the reasonable assumption that only one-third of work-authorized employees still employed at the company and listed in a no-match letter would need to visit SSA to resolve the no-match.

DHS believes that only a small subset of these authorized employees will undergo the reverification process because most legal employees (citizens and aliens authorized to work) will resolve the no-match with SSA, in large part because it is in employees' personal financial interest to do so. Notwithstanding that financial incentive for employees to resolve their no-match and receive credit for retirement benefits, some employees that are referred to SSA to resolve their no-match may decide to complete a new Form I-9 instead of visiting the SSA. To the extent that employees might decline to visit an SSA office and instead choose to complete a new Form I-9, the SEIA overestimates the costs that would be incurred by employers. DHS estimates that completion of all sections of a new Form I-9 and preserving that form pursuant to the INA and regulations requires 12 minutes. 73 FR 18551 (April 4, 2008). The SEIA estimates an employee would be required to expend a full eight-hour day to visit SSA to resolve the no-match.

Given the assumption in the rule that the re-verification procedure will function as the last, fall-back step for employers to confirm an employee's work authorization, DHS assumed, for the purposes of the SEIA, that all employees who resort to the re-verification procedure will first have visited the SSA. DHS, therefore, will not lower the estimate of the number of employees expected to visit an SSA office. In order to allow for the possibility that a larger than anticipated number of legal employees may both visit SSA offices and use the I-9 reverification procedure, DHS will revise the SEIA to include additional re-verification costs for 3 percent of employees that might visit SSA and also complete a new Form I-9 reverification. Adding the reverification costs for this 3 percent without reducing the number of employees expected to visit SSA will likely result in a small overestimate of the actual costs, but due to limitations of available data, DHS believes that this approach is reasonable.

c. Outsourced Staffing Requirements

Several commenters suggested that many small businesses do not have an in-house human resources staff or payroll administrators and instead hire outside providers for this service. Some comments also criticized the wage rates used in the analysis because those rates do not take into account the difference between in-house wages and outsourced wages for the same services. A commenter pointed out, for example, that the wage rate of an in-house attorney cannot be equated with the cost charged to a client by outside counsel. These outsourced wage rates would include different and higher rates to recover overhead charges for rent, utilities, taxes, and other costs of doing business that might not be incurred by the employer. The commenter further suggested the cost of out-sourced wages are estimated to be two to three times the price of what an employer pays per hour in in-house wages.

DHS agrees that outsourced work may be more expensive than work conducted in-house as the commenter suggests. DHS also agrees to assume, for the purposes of the SEIA, that the cost of hiring services provided by an outside vendor or contractor is two to three times more expensive than the wages paid by the employer for that service produced by an in-house employee. The costs in the SEIA have been revised to take into account the higher costs that may be incurred when firms use outside service providers.

8. Other Costs

One commenter noted that while the SEIA included costs associated with replacing work-authorized employees who are terminated as a result of the rule, it did not include costs associated with payment of unemployment benefits to such employees. Unemployment benefit payments are a cost incurred by the federal and state governments, which are not “small entities” for purposes of the RFA. Moreover, such benefits are not paid by an employer as a result of that employer's adherence to the safe harbor procedures in this rule, and this cost is at best an indirect cost not covered by the RFA.

9. Rehiring Seasonal Employees

A number of commenters suggested that the employment of seasonal employees was not adequately considered in the IRFA. The two most common examples may be seasonal employment of farm employees and retailer seasonal employment of additional sales and support personnel during holiday seasons.

Some comments suggested that special systems would be needed to track seasonal employees no longer employed by the employer at the time the no-match letter is received. The rationale for such a tracking system would be to mitigate an employer's risk by ensuring that the employer can identify and appropriately examine the work authorization documents for returning job applicants who were previously listed on a no-match letter. The no-match rule does not address this scenario, and seasonal employers that hire returning employees could have had sufficient reason under INA section 274A, 8 U.S.C. 1324a, and the pre-existing regulations to compare past no-match letters against the identity information provided by all new and returning hires if employers believe such a comparison was needed. This rule provides a safe harbor after an employer has hired an employee, receives a no-match letter relating to that employee, and conducts due diligence to resolve the no-match letter. The rule does not address the initial hiring decision and employment eligibility verification. As with the costs that result from an employer's discovery of unauthorized workers on the payroll, the cost of any system that an employer may adopt to address knowledge acquired from previous no-match letters is attributable to the INA, not to this rule.

10. Conclusions

Several commenters noted that the thrust of the SEIA is that the proposed regulation will not affect a significant number of small entities and those small entities which are impacted will not incur significant expenses, and suggested that the IRFA and FRFA should contain an express statement to that effect.

The supplemental proposed rule did express the conclusion that “DHS does not believe that the direct costs incurred by employers that choose to adopt the safe harbor procedures set forth in this rule would create a significant economic impact when considered on an average cost per firm basis.” 73 FR at 15953. The SEIA, as revised in light of the comments received in the course of this rulemaking, continues to support the conclusion that the direct costs incurred by those small entities that avail themselves of the safe harbor are not expected to be significant on an average cost per small entity basis.

E. Further Interpretation of the August 2007 Final Rule

In this supplemental rulemaking DHS seeks to further clarify two aspects of the August 2007 Final Rule. First, the rule instructs employers seeking the safe harbor that they must “promptly” notify an affected employee after the employer has completed its internal records checks and has been unable to resolve the mismatch. After reviewing the history of the rulemaking, DHS believes that this obligation for prompt notice would ordinarily be satisfied if the employer contacts the employee within five business days after the employer has completed its internal records review. Some commenters suggested that this timeframe was inadequate, while others suggested that this guidance be made explicit in the text of the rule. DHS understands that too short a timeline for informing employees of their need to resolve a no-match may be unworkable for certain employers and employees, and so the Department declines to set a formal limit in the rule text on the time that an employer may take in providing “prompt” notice to affected employees. DHS emphasizes that an employer does not need to wait until after completing this internal review to advise affected employees that the employer has received the no-match letter and request that the employees seek to resolve the mismatch. Immediately notifying an employee of the mismatch upon receipt of the letter may be the most expeditious means of resolving the mismatch. Prompt notice to affected employees is important to enable them to take the steps necessary to resolve the mismatch, and an employer should not unreasonably delay such notice.

Second, plaintiffs in the litigation before the Northern District of California raised a question as to whether under the August 2007 Final Rule an employer could be found liable on a constructive knowledge theory for failing to conduct due diligence in response to the appearance of an employee hired before November 6, 1986 in an SSA no-match letter. When Congress enacted INA section 274A as part of the 1986 Immigration Reform and Control Act, it included a grandfather clause stating that employers' obligations created in that Act did not apply to the hiring, recruitment, or referral for employment for a fee, or to the continued employment, of workers hired before IRCA's date of enactment. See Public Law 99-603, section 101(a)(3), 100 Stat. 3359 (1986). Because those statutory bars against hiring or continuing to employ individuals without work authorization do not apply to workers within that grandfather clause, this rule does not apply to any such workers that may be listed in an SSA no-match letter. A number of commenters argued that this exclusion should be explicitly stated in the rule text. But employees hired before November 1986 are statutorily excluded from the operation of INA section 274A(a), and so no regulatory statement reiterating that effect is necessary.

F. Other Comments Received

The supplemental proposed rule made clear that DHS was addressing the three issues raised by the district court, 73 FR 15944, 45, and DHS did not reopen other aspects of the rulemaking. Several commenters understood the supplemental proposed rule as inviting comments generally, and they provided comments on a range of issues previously covered in the August 2007 Final Rule but not related to the three issues raised by the district court and addressed in the supplemental proposed rule. The August 2007 Final Rule addressed the substantive issues raised in these comments, and DHS declines to address those issues anew.

IV. Changes Made in Republishing the Final Rule

The final rule does not make any substantive changes from the August 2007 Final Rule or the Supplemental Proposed Rule. DHS has corrected a technical cross-reference in the text of the final rule and republishes the text of the regulation for the convenience of the reader.

V. Statutory and Regulatory Reviews

A. Administrative Procedure Act

DHS published the initial proposed rule and the supplemental proposed rule with requests for public comment in the Federal Register as a matter of agency discretion. This rule is not a legislative rule governed by the notice and comment provisions of 5 U.S.C. 553. DHS is publishing this supplemental final rule subject to the preliminary injunction entered by the district court. A delayed effective date is not required under the APA. 5 U.S.C. 553(d)(2).

B. Regulatory Flexibility Act

On the basis of the analysis in this preamble, DHS provides below its Final Regulatory Flexibility Analysis, as described under the Regulatory Flexibility Act, 5 U.S.C. 604. DHS published an initial regulatory flexibility analysis pursuant to 5 U.S.C. 603(b), (c), in response to the district court's injunction in the supplemental proposed rule. 73 FR at 15952-54. DHS published a small entity impact analysis in the docket of this rulemaking, ICEB-2006-0004-0233, and summarized that analysis in the supplemental proposed rule. DHS invited comments related to this Initial Regulatory Flexibility Analysis and the accompanying Small Entity Impact Analysis, including comments on the assumptions underlying that analysis.

1. Need for, Objectives of, and Reasons Why the Rule Is Being Considered

As discussed more fully in the supplemental proposed rule, DHS, as well as private employers in general, have become increasingly aware of the potential for abuse of social security numbers by aliens who are not authorized to work in the United States. DHS is responsible for the enforcement of the statutory prohibition against the hiring or continued employment of aliens who are not authorized to work in the United States. INA section 274A(a)(1), (2), 8 U.S.C. 1324a(a)(1), (2); HSA section 101, 6 U.S.C. 111. Given employers' evident confusion regarding how to respond to SSA no-match letters, DHS has concluded that it needs to clarify employers' duties under the immigration laws, and has set forth guidance for employers that seek to fulfill their obligation not to hire or employ aliens who are not authorized to work in the United States.

The objective of the proposed rule, the August 2007 Final Rule, the supplemental proposed rule, and this final rule is to provide clear guidance for employers on how to comply with the statutory bar against hiring or continuing employment of aliens who are not authorized to work in the United States. INA section 274A(a)(1), (2), 8 U.S.C. 1324a(a)(1), (2). The objective of this statute is to eliminate the “magnet” effect of employment opportunities that induces aliens to enter or remain in the United States illegally. DHS exercises investigative and prosecutorial discretion in enforcing this statute, and this interpretive rule explains how DHS will exercise that discretion, and provides guidance to employers that wish to limit their risk of liability under the immigration laws.

2. Significant Issues Raised in Public Comments

Significant issues raised by the public comments relating to the initial regulatory flexibility analysis and the small entities impact analysis are discussed in section III.D of this preamble.

3. Description of and Estimate of the Numbers of Small Entities to Which the Rule Would Apply

To estimate the small entities affected, DHS uses the generally accepted Office of Management and Budget, Economic Classification Policy Committee, North American Industrial Classification (NAIC), pursuant to 44 U.S.C. 3504(e), and the size determinations by the Small Business Administration (SBA) for SBA and other programs. 13 CFR 121.101(a); 121.201; 121.902 (size standards promulgated for SBA programs and applicable to other agency programs). The definition of what constitutes a small business varies from industry to industry and generally depends on either the number of employees working for a business or the amount of annual revenue a business earns.

DHS requested information from SSA to assist in better identifying the number of small entities that could be expected to establish safe harbor procedures. Specifically, DHS requested that SSA provide the names and addresses of the companies already identified by SSA in its preparation to release no-match letters in September 2007. This raw data would have permitted DHS to conduct research to determine the North American Industry Classification System industry to which the specific companies belonged, to research the annual revenue and/or the number of employees of these companies through standard sources, and thus to apply the appropriate small business size standards. With these analyses, DHS anticipated that it would be able to provide a rough estimate of the number of employers expected to receive a no-match letter that met the SBA's definitions of small businesses.

However, SSA informed DHS that it was unable to provide DHS with the names and addresses of the employers expected to receive a no-match letter, citing the general legal restrictions on disclosure of taxpayer return information under section 6103 of the Internal Revenue Code of 1986, 26 U.S.C. 6103. DHS also approached the Government Accountability Office (GAO) and the Small Business Administration, Office of Advocacy, to seek any data that these agencies might be able to provide, and to consult about the analysis to be included in this IRFA. GAO supplied some additional data, but SBA informed DHS that it had no data-other than general small business census data-that was relevant to this rulemaking and that could assist in the analysis for purposes of this IRFA. Consequently, DHS does not have the data necessary to determine the precise number of small entities expected to receive a no-match letter.

Nevertheless, SSA was able to provide some general information. SSA provided a table showing a distribution of the number of employers that were slated to receive a no-match letter for Tax Year 2006, according to the number of Form W-2s filed by the employer. As this data did not exclude small entities, DHS believes that the universe of small entities that would have received a no-match letter for Tax Year 2006 is contained within the table that SSA provided. Even though this data did not provide the number of small entities, this data was useful to DHS while conducting the small entity impact analysis contained in the docket. See ICEB-2006-0004-0232, Exhibit A.5. DHS was not able to determine what share of the affected small entities would be small businesses, small non-profit organizations, or small governmental jurisdictions. Absent some reason to believe small non-profits or small governmental jurisdictions might implement the rule's safe harbor procedures differently from private employers, the cost structure for such entities would be no different from small firms. DHS is unaware of any data to suggest there would be a difference, and the public comments did not suggest there would be any difference.

4. Proposed Reporting, Recordkeeping, and Other Compliance Requirements

The proposed rule suggests, but does not require, that employers retain records of their efforts to resolve SSA no-match letters. This suggestion is based on the possible need of an employer to demonstrate the actions taken to respond to a no-match letter if and when ICE agents audit or investigate that employer's compliance with INA section 274A, 8 U.S.C. 1324a. While the rule encourages employers to document their eligibility for the safe harbor by keeping a record of their actions, the rule does not impose any requirement for an employer to make or retain any new documentation or records.

Companies that choose to adopt the safe harbor procedures in the rule would reasonably be expected to incur costs related to administering and implementing those procedures. Company-level costs could include the labor cost for human resources personnel, certain training costs, legal services, and lost productivity. A detailed analysis of safe harbor-related costs that companies may incur is contained in the Small Entity Impact Analysis available in the docket of this rulemaking. While several commenters have expressed concerns about the costs to businesses relating to the termination and replacement of unauthorized workers, DHS finds that those costs cannot properly be considered costs of this rule. The INA expressly prohibits employers from knowingly hiring or knowingly continuing to employ an alien who is not authorized to work in the United States. If an employer performs the due diligence described in the rule, and loses the services of unauthorized employees as a result, those costs of terminating and/or replacing illegal workers are attributable to the INA, not to this rule.

Table 1, below, summarizes the average cost per firm that DHS estimates will be incurred by businesses that receive a no-match letter and choose to adopt the safe harbor procedures set forth in this rule. Because DHS does not have adequate data to estimate the percentage of unauthorized employees whose SSNs are listed on no-match letters, for the purpose of this analysis, DHS estimated costs based on various ratios of authorized to unauthorized workers (i.e., 20% unauthorized—80% authorized). As Table 1 shows, the expected costs of adopting the safe harbor procedures in this rule are relatively small on an average cost per firm basis. In interpreting these costs, these estimates were based on a series of assumptions which are explained in detail in the small entity impact analysis included in the docket. Consequently, the costs a specific firm incurs may be higher or lower than the average firm costs estimated in Table 1.

Table 1—Total Costs per Firm by Employment Size Class

Employment size class	Percentage of current no-match employees assumed to be unauthorized
	10	20	40	60	80
5-9	$4,560	$4,454	$4,244	$4,033	$3,822
10-19	4,847	4,716	4,455	4,194	3,933
20-49	6,818	6,597	6,155	5,712	5,270
50-99	8,890	8,582	7,966	7,350	6,734
100-499	24,785	23,426	20,709	17,992	15,274
500+	36,624	34,496	30,239	25,983	21,726

Table 1 does not reflect the termination or replacement costs of unauthorized workers. The termination and replacement of unauthorized employees will impose a burden on employers, but INA section 274A(a)(1), (2), 8 U.S.C. 1324a(a)(1), (2), expressly prohibits employers from knowingly hiring or knowingly continuing to employ an alien who is not authorized to work in the United States. Accordingly, costs that result from employers' knowledge of their workers' illegal status are attributable to the Immigration and Nationality Act, not to the August 2007 Final Rule or this supplemental proposed rule, and its provision of a safe harbor. Similarly, any costs incurred by seasonal employers that face difficulties in hiring new employees in the place of unauthorized workers whose SSNs were previously listed on SSA no-match letters are attributable to the Immigration and Nationality Act bar to knowingly hiring workers who are not authorized to work in the United States.

In summary, DHS does not believe that this safe harbor rule imposes any mandate that forces employers to incur “compliance” costs for purposes of the Regulatory Flexibility Act. Even assuming that the safe harbor rule requires certain action on the part of employers that receive no-match letters, DHS does not believe that the direct costs incurred by employers that choose to adopt the safe harbor procedures set forth in this rule would create a significant economic impact when considered on an average cost per firm basis. To the extent that some small entities incur direct costs that are substantially higher than the average estimated costs, however, those employers could reasonably be expected to face a significant economic impact. As discussed above, DHS does not consider the cost of complying with preexisting immigration statutes to be a direct cost of this rulemaking. Thus, while some employers may find the costs incurred in replacing employees that are not authorized to work in the United States to be economically significant, those costs of complying with the Immigration and Nationality Act are not direct costs attributable to this rule. DHS has not formally certified the rule as not having a “significant economic impact on a substantial number of small entities” as allowed under section 605(b) of the Regulatory Flexibility Act. Instead, DHS has prepared this Final Regulatory Flexibility Analysis as described in the Regulatory Flexibility Act, 5 U.S.C. 604.

5. Significant Alternatives Considered

DHS has considered several alternatives to the proposed rule. For the most part, however, the alternatives would not provide employers with necessary guidance and assurances against liability under the INA, nor would the alternatives improve employers' compliance with INA section 274A, 8 U.S.C. 1324a.

a. No action. Taking no action to clarify employers' responsibilities under INA section 274A, 8 U.S.C. 1324a, was considered. Taking no action, however, would not resolve any of the problems identified and addressed by this proposed rule. Employers will remain confused and unsure how to act to resolve no-match letters in a manner consistent with their responsibilities under current immigration law, and will continue to face possible liability based in part on their failure to respond to no-match letters. Employers would continue to employ aliens unauthorized to work under federal immigration law.

b. Specific industry or sector limitations. DHS considered limiting the proposed rule to specific industries previously noted to be at high-risk of abuse of Social Security numbers in employment, including agriculture, services and construction. See, e.g., Government Accountability Office, Social Security: Better Coordination among Federal Agencies Could Reduce Unidentified Earnings Reports, Administrative Record at 400 (GAO analysis of SSA data noting 17% of ESF filings by eating and drinking places; 10% by construction, and 7% by agriculture). DHS also considered promulgating a rule that applied only to critical infrastructure employers because of the increased need to prevent identity fraud by employees in high-risk facilities. None of these alternatives was acceptable because none addresses the larger population of aliens working without authorization or the need for clear guidance for employers in other sectors of the economy. These alternatives would also offer unfairly selective assurances to employers in certain sectors against liability under INA section 274A, while depriving other employers of the same protection.

Focusing on the three economic sectors with the most egregious violations of the immigration laws might have had an impact on a significant portion of the alien population that illegally enters the United States to work. As discussed more fully in the small entity impact analysis in the docket, the degree to which specific industry sectors violate the bar to employment of unauthorized aliens is, however, speculative. DHS does not have access to the data files indicating the number of employers by industry sector who would receive no-match letters under current SSA policies. DHS requested industry-sector-specific data from SSA but was informed that SSA does not possess this data. Non-empirical, anecdotal evidence, such as the admissions of the President of the Western Growers' Association, supra, that between 50 to 80% of their employees are unauthorized aliens, is a less reliable guide for agency action than empirical evidence. Even if such anecdotal evidence is sufficient to guide decisions about investigation and enforcement priorities, it is not an adequate basis for limiting the effect of formal agency guidance to a specific sector of the economy. Partial enforcement tends, moreover, as a matter of experience, to have the effect of redirecting unauthorized workers into areas where the law is unenforced or underenforced.

A critical-infrastructure approach provided other benefits, focusing on high-risk facilities and organizations. Critical infrastructure encompasses, however, segments of industries that are not entirely discrete. Focusing on critical infrastructure would have had salutary effects in certain areas, but the inefficiencies and inequities that result from other types of partial enforcement would remain unchanged. Moreover, DHS has already taken, and continues to take, other steps in working with critical infrastructure partners to improve employer compliance with the INA and reduce the employment of aliens not authorized to work in the United States.

Another variation suggested that DHS adopt special provisions for short-term, seasonal, or intermittent employees and employers that have high turnover rates. This variation applies, as the commenter pointed out and DHS has previously noted, to the agriculture, construction, and service sectors (such as restaurants or hotels). The commenter particularly noted that agricultural employers hire many employees for 60-day periods and, because SSA sends no-match letters on an annual W-2 wage reporting basis, most of these letters will arrive long after the term of employment has ended. The commenter further suggested that, because the employee no longer works for the employer, the employer's responsibilities should end there. The commenter requested that DHS clarify that employers are not required to track and contact past employees for whom they receive no-match letters.

DHS agrees with certain points made by the commenter, but disagrees with the commenter's suggested alternative. The commenter is correct that when an employee is terminated, the employer does not have any further responsibility for tracking down the employee and resolving the mismatch. DHS does not agree, however, that this scenario requires any special rule. The focus of this rulemaking is on reinforcing the INA's prohibition on continued employment of aliens not authorized to work in the United States. The issue of whether an employer acquires constructive knowledge from receipt of a no-match letter or possesses constructive knowledge at a later time when the employer hires the same employee for another cycle of work is not addressed by this rule. Employers' hiring practices must comply with the INA, and no safe harbor or specific guidance is offered by this rule.

Most significantly, none of the alternatives for limiting or tailoring the applicability of the rule to specific industries or sectors would mitigate the rule's impact on small business. Accordingly, DHS rejected the industry-specific approach as insufficient to accomplish the goal of improving overall employer compliance with immigration law and reducing the population of aliens illegally working in the United States, and as ineffective in limiting the impact on small employers.

c. Phased implementation for small employers. DHS considered phasing in the implementation of the rule by delaying its applicability to small entities. Comments suggested that by imposing the rule on large entities first, many of the errors thought to exist in the SSA database could be corrected over time and best practices for resolving no-matches could be developed. A commenter suggested that this experience could then be used to ease small entities into the process. The commenter suggested that large entities (including both private sector and governmental employers) that receive no-match letters have sophisticated human resources departments that are capable of handling no-match letters, but that small entities with limited human resources capacity do not have this capacity.

DHS has concluded, after further review, that such an approach would still harm, not help, small employers. All employers, including small entities, are already subject to the legal obligation not to knowingly employ unauthorized workers and the constructive knowledge standard for employer liability, both of which flow from the INA. DHS cannot exempt small entities from the INA, and so delaying the applicability of this rule for small entities would not excuse small employers from their existing legal obligations. Instead, limiting the guidance and the safe harbor protection offered in this rule to large employers would effectively leave small employers exposed to greater liability risk and would not address the illegal employment of unauthorized aliens by small employers.

d. Extended time allowance for small employers. DHS also considered further extending the time periods in the rule for small employers that wish to obtain the protection of the safe harbor to check their internal records to confirm the no-matches were not the result of some administrative error by the employer. Several commenters supported this alternative, with some suggesting that small employers in rural areas may find their employees have difficulty resolving their mismatches with SSA. Proposed alternatives included providing small entities with 180 days to complete the steps outlined in the rule, or establishing a tiered approach with different timeframes based on the size of the employer (with smaller employers receiving more time to comply), or based on the distance to the local SSA office. One commenter also suggested that DHS consider suspending the running of the timeframes when an employee is actively working with SSA to correct the discrepancy. DHS considered each of these variations, but does not believe that they would provide meaningful benefit to small employers or maintain the rule's effectiveness.

The timeframes set forth in the August 2007 Final Rule were extended significantly from those contained in the proposed rule published in 2006, in response to comments from large and small employers expressing concern that the timeframes initially proposed were too short. In particular, the time allotted for an employer to review its own records for errors was doubled from 14 days to 30 days. The commenters provided no evidence that small employers, with small payrolls, would need more time to review their records than would large organizations with thousands of employees. Several comments submitted during this supplemental rulemaking suggested DHS extend the timeframe for an employee to resolve a mismatch with SSA, citing distance to the nearest SSA office as a concern for workers in rural areas. But the comments provided no evidence or concrete support for the claim that the 90 days allotted under the rule would be insufficient. SSA has approximately 1,300 local offices nation-wide, and provides public assistance in locating the closest office both on-line and by telephone, along with advice on the documents required to resolve a mismatch.

Moreover, undue extension of the time period for an employee to resolve his or her mismatch would substantially weaken the effectiveness of the rule by frustrating employers' ability to be confident in the legal status of their workers. If the timeline in the rule were extended to 180 days, for example, unauthorized workers (possibly with encouragement from unscrupulous employers) would be more likely to simply go through the motions of contacting SSA in order to extend their time on the job for a full six months, while law-abiding employers that suspect, but lack conclusive proof, that some of their employees are illegally working without authorization would be forced to stand by and worry that the listed employees may leave without warning or that the employer might be subject to a worksite enforcement or investigation effort by ICE. The suggestion to suspend the running of the timeframes while an employee is “actively” working to resolve his mismatch suffers from these same flaws and adds another: There would be no clear way for either the employer or DHS to determine whether an employee had in fact been actively working in good faith to resolve the mismatch, and an employer could not be confident that its conduct met the requirements for the safe harbor, effectively eviscerating the value of the rule for law-abiding employers.

e. Mandatory steps without assurances of safe harbor. DHS also considered requiring all employers to take specific actions whenever they received a no-match letter and their records indicated that a social security number was used in Form I-9 processing. Requiring employers to take affirmative steps to resolve social security no-match letters (as outlined as discretionary steps in the proposed rule) could result in fuller compliance with the prohibition against employment of aliens who are not authorized to work in the United States. But such a mandatory scheme implies that the steps set forth in the rule are the only reasonable response to a SSA no-match letter, a conclusion that cannot be supported by the evidence currently before DHS. Furthermore, the relative gains from a mandatory scheme, in the absence of additional statutory authority to impose sanctions for violations of that mandate, are likely to be very small. Employers that consciously or recklessly violate the INA will not alter their behavior under either a mandatory or voluntary safe harbor regime, while responsible employers that want to comply with the INA will benefit from the guidance provided in the proposed safe harbor rule and will improve their hiring and employment practices to ensure compliance with the INA.

f. Elimination of the time limit for resolving no-matches. One commenter suggested that DHS adopt what was described as a simpler, more straightforward rule for small entities that receive a no-match letter, in which the employer would: (1) Complete an internal investigation to determine whether the source of the discrepancy is the employer's own clerical error; (2) if not, inform the affected employee of the discrepancy; and, (3) if the employee challenges the discrepancy, require proof that the employee has been in contact with SSA to resolve the discrepancy. Under this scenario, the commenter suggested that a reasonable employer could assume that the employee was resolving the discrepancy with SSA and need not inquire further unless another no-match letter was received the following year (or some other adverse information arose). The commenter suggested that this approach would reduce the burden on small entities. The commenter also believed that this would eliminate what it perceived to be a presumption that receipt of a no-match letter puts the employer on notice that the employee may be unauthorized to work in the United States.

This alternative essentially eliminates the timeline for an employee to resolve the mismatch, and deprives the employer of any assurance that the questions raised by the no-match letter have been answered. The comment also mistakenly assumes that such a rule would negate the well-established fact-conceded in the record of this rulemaking even by this rule's opponents and endorsed by the district court in the ongoing litigation over this rule-that a no-match letter is a legitimate indicator of possible illegal work by unauthorized aliens. Such a rule would offer a carte blanche safe harbor to employers without requiring the employer to take any meaningful steps to answer the questions raised by the employees' appearance on a no-match letter. DHS cannot give the benefit of a safe harbor when there is no assurance that the mismatch has been resolved.

g. DHS resolution of no-matches. A commenter suggested that DHS, rather than employers and employees, resolve mismatches involving the employees of small entities. The commenter suggested that small entities could be sent to DHS for investigation of any mismatches that remained unresolved after the rule's timeframe expired. The commenter argued that such a system would give DHS notice of the existence of the no-match discrepancy, but not require that the employee be terminated until DHS has had an opportunity to investigate the matter. A variation on this alternative suggested that DHS create a special office or appoint an “ombudsman” to assist employees in resolving “no-matches” where the employee has been unable to resolve within the requisite timeframe. The commenter suggested that such an approach could lead to an intra-governmental correction process with direct lines of communication to investigate no-matches and correct the SSA database, relieving employers and protect authorized employees from automatic termination.

This alternative is not practically feasible. DHS does not have access to the information contained in no-match letters, nor does DHS have the personal information about individual employees that SSA needs to resolve mismatches. Taken to its logical end, this is a proposal to eliminate the SSA no-match letter program entirely-an undertaking that is far beyond DHS's regulatory competence.

6. Minimization of Impact

The RFA requires that an agency provide “a description of the steps the agency has taken to minimize the significant economic impact on small entities consistent with the stated objectives of applicable statutes * * *” 5 U.S.C. 604(a)(5). This requirement presumes that the agency finds that the rule will have a significant economic impact on small entities and is normally treated in conjunction with the discussion of alternatives (see above) required by paragraph (a)(5). Although DHS, after reviewing the record, does not make a finding that the rule will have a significant economic impact on small entities, DHS believes that explaining the existing means by which a small entity may minimize any impact of the rule, and certain additional steps that DHS is taking to assist them, will be useful to small entities.

(1) DHS and its subsidiary components ICE and United States Citizenship and Immigration Services (USCIS), already provide substantial support for employers that wish to ensure the work eligibility of their workforce. The primary tool DHS makes available to employers is the E-Verify program, which is an Internet-based system for electronically verifying employment eligibility that is operated by U.S. Citizenship and Immigration Services (USCIS), in partnership with the SSA. The requirements for obtaining access to E-Verify and procedures for the use of E-Verify are established by DHS and USCIS. Before an employer can participate in the E-Verify program, the employer must enter into a Memorandum of Understanding (MOU) with DHS that sets out certain features of the program and enumerates specific responsibilities of DHS, SSA, and the employer. This MOU requires employers to agree to abide by current legal hiring procedures and to ensure that no employee will be unfairly discriminated against as a result of the E-Verify program. Employers participating in E-Verify must still complete an Employment Eligibility Verification Form (Form I-9) for each newly hired employee, as required under current law. Following completion of the Form I-9, however, the employer enters the employee's information into the E-Verify Web site, and that information is then checked against information contained in SSA and USCIS databases to confirm the employee's work eligibility with much greater rigor than is possible with the Form I-9 process alone.

E-Verify first sends the information to SSA for verification of the name, SSN, and date of birth, and SSA confirms these elements as well as U.S. citizenship based on the information in SSA records. USCIS also verifies through database checks that any non-United States citizen employee is in an employment-authorized immigration status. E-Verify will then confirm the employee is employment-eligible.

If the information provided by the employee matches the information in the SSA and USCIS records, no further action will generally be required, and the employee may continue employment. E-Verify procedures require only that the employer record on the Employment Eligibility Verification Form I-9 the verification ID number and result obtained from the E-Verify query, or print a copy of the transaction record and retain it with the Form I-9. Verification of the employee's name and SSN through E-Verify sharply reduces the likelihood that individuals checked through E-Verify will appear on an SSA no-match letter.^[15]

(2) In addition, the ICE Mutual Agreement between Government and Employers (IMAGE) program permits companies to reduce unauthorized employment and the use of fraudulent identity documents, thereby reducing the likelihood of receiving a no-match letter. As part of the IMAGE program, ICE and USCIS provide education and training on proper hiring procedures, fraudulent document detection, use of the E-Verify employment verification program, and anti-discrimination procedures.

ICE provides employers in IMAGE with an “I-9 audit.” This free audit is similar to the services commercially provided by law firms and others for a fee.

IMAGE also provides employers with a catalogue of “best practices” including:

• Use of E-Verify for all hiring.
• Establish an internal training program, with annual updates, on how to manage completion of Form I-9 (Employee Eligibility Verification Form), how to detect fraudulent use of documents in the I-9 process, and how to use E-Verify.
• Permit the I-9 Employment Eligibility Verification and E-Verify process to be conducted only by individuals who have received this training—and include a secondary review as part of each employee's verification to minimize the potential for a single individual to subvert the process.
• Arrange for annual I-9 audits by an external auditing firm or a trained employee not otherwise involved in the I-9 and electronic verification process.
• Establish a self-reporting procedure for reporting to ICE any violations or discovered deficiencies.
• Establish a protocol for responding to no-match letters received from the Social Security Administration.
• Establish a Tip Line for employees to report activity relating to the employment of unauthorized aliens, and a protocol for responding to employee tips.
• Establish and maintain safeguards against use of the verification process for unlawful discrimination.
• Establish a protocol for assessing the adherence to the “best practices” guidelines by the company's contractors/subcontractors.
• Submit an annual report to ICE to track results and assess the effect of participation in the IMAGE program.

To help ensure the accuracy of their wage reporting, ICE assists employers participating in the IMAGE program to verify the Social Security numbers of their existing labor force through SSA's Social Security Number Verification Service (SSNVS). IMAGE participants also verify work eligibility of their new hires through E-Verify. All of these steps reduce the potential for employer created errors in wage submittals to the IRS and SSA, reducing the potential for the employer to receive a no-match letter. See http://www.ice.gov/​partners/​opaimage/​index.htm.

C. Unfunded Mandates Reform Act of 1995

This rule will not result in the expenditure by State, local, and tribal governments, in the aggregate, or by the private sector, of $100 million or more in one year, and it would not significantly or uniquely affect small governments. Therefore, no actions were deemed necessary under the provisions of the Unfunded Mandates Reform Act of 1995, Public Law 104-4, 109 Stat. 48 (1995), 2 U.S.C. 1501 et seq.

D. Small Business Regulatory Enforcement Fairness Act of 1996

This rule is not a major rule as defined by section 804 of the Small Business Regulatory Enforcement Act of 1996, Public Law 104-121, 804, 110 Stat. 847, 872 (1996), 5 U.S.C. 804(2). This rule has not been found to be likely to result in an annual effect on the economy of $100 million or more; a major increase in costs or prices; or significant adverse effects on competition, employment, investment, productivity, innovation, or on the ability of United States-based companies to compete with foreign-based companies in domestic or foreign markets.

E. Executive Order 12,866 (Regulatory Planning and Review)

Because this rule considers interests of a number of different agencies and provides guidance to the public as a statement of policy or interpretive rule, the final rule was referred to the Office of Management and Budget pursuant to Executive Order 12866, as amended. Multiple agencies reviewed and considered the draft. This rule reflects that consultation. OMB has determined that this rule will not have an effect on the economy of more than $100 million.

F. Executive Order 13,132 (Federalism)

This rule does not have 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, in accordance with section 6 of Executive Order No. 13,132, 64 FR 43,255 (Aug. 4, 1999), this rule does not have sufficient federalism implications to warrant the preparation of a federalism summary impact statement.

G. Executive Order 12,988 (Civil Justice Reform)

This rule meets the applicable standards set forth in sections 3(a) and 3(b)(2) of Executive Order No.12,988, 61 FR 4729 (Feb. 5, 1996).

H. Paperwork Reduction Act

Under the Paperwork Reduction Act of 1995, 44 U.S.C. 3501, et seq., all agencies are required to submit to OMB, for review and approval, any reporting requirements inherent in a rule. While employers seeking to establish eligibility for the safe harbor are encouraged to keep a record of their actions, this rule does not impose any additional information collection burden or affect information currently collected by ICE.

List of Subjects in 8 CFR Part 274a

• Administrative practice and procedure
• Aliens
• Employment
• Penalties
• Reporting and recordkeeping requirements

Accordingly, for the reasons stated in the preamble to this supplemental final rule, the Department of Homeland Security reaffirms the text of the final rule issued on August 15, 2007, 72 FR 45611, and makes one typographical correction as set forth below:

PART 274a—CONTROL OF EMPLOYMENT OF ALIENS

1. The authority citation for part 274a continues to read as follows:

Authority: 8 U.S.C. 1101, 1103, 1324a; 8 CFR part 2.

§ 274a.1

[Amended]

2. In § 274a.1(l)(2)(iii) remove the phrase “(l)(2)(i)(B)” and add in its place the phrase “(l)(2)(i)(C)”.

Michael Chertoff,

Secretary.

Footnotes

1.  See INA Section 274C, 8 U.S.C. 1324c.

2.  See, e.g., United States v. Gonzales, 2008 WL 160636 (N.D. Miss. No. 4:07-CR-140, Jan. 18, 2008) (finding no-match letters admissible at trial, and upholding a search warrant obtained on the basis of information, including copies of social security no-match letters, received from a confidential informant, treating no-match letters as “documentary evidence supporting the allegation” of the confidential informant); United States v. Fenceworks, Inc., No. 3:06-CR-2604 (S.D. Cal.), D.E. 16 (judgment of probation and forfeiture of $4,700,000 in case involving multiple Social Security no-match letters) (related cases Nos. 3:06-CR-2605 (probation and fine of $100,000); 3:06-CR-2606 (probation and fine of $200,000)); United States v. Insolia, No. 1:07-CR-10251 (D. Mass), D.E. 1 (complaint; attachment, ¶¶ 25-32, February 2007 probable cause affidavit detailing history of employer's no-match letters from 2002 through 2005 and other investigative methods and facts); 34 (indictment); United States v. Rice, No. 1:07-CR-109 (N.D.N.Y), D.E. 1 (complaint; attached probable cause affidavit) (¶¶ 64-66, detailing results of matching analysis and SSA letters received by defendant's employer), D.E. 17 (plea agreement).

3.  Social Security Administration, Performance and Accountability Report, Fiscal Year 2007 at 67-8.

4.  For example, DHS rescinds conclusive statements from the preamble of the August 2007 Final Rule such as “employers who follow the safe harbor procedures * * * will not be found to have engaged in unlawful discrimination.” 73 FR at 15950, citing 72 FR 45613-14.

5.  The Small Business Administration had provided additional guidance. See Office of Advocacy, Small Business Administration, A Guide for Government Agencies: How to Comply with the Regulatory Flexibility Act (2003). It states, in pertinent part:

The RFA requires agencies to conduct sufficient analyses to measure and consider the regulatory impacts of the rule to determine whether there will be a significant economic impact on a substantial number of small entities. No single definition can apply to all rules, given the dynamics of the economy and changes that are constantly occurring in the structure of small-entity sectors.

Every rule is different. The level, scope, and complexity of analysis may vary significantly depending on the characteristics and composition of the industry or small entity sectors to be regulated.

Id. at 14.

6.  Social Security Administration, Performance and Accountability Report, supra n.2, at 190.

7.  While the Earnings Suspense File is an electronic repository for wage items that cannot be matched to an individual worker's earnings record, the database that SSA uses to match a wage item to a worker is the Numident database.

8.  Social Security Administration, Office of the Inspector General, Congressional Response Report: Accuracy of the Social Security Administration's Numident File (No. A-08-06-26100, Dec. 2006).

9.  One commenter suggested that a DHS-funded study of the Basic Pilot or E-Verify program shows that a larger share of individuals listed in no-match letters will need to visit SSA, claiming that “only in 30% of the time were tentative non-confirmations caused by either solely an error with the date of birth or the name.” ICEB-2006-0004-07321.1 at 27 (citing to Westat, Findings of the Web Basic Pilot Evaluation, supra at 51). After re-reviewing the Westat report, DHS disagrees. The passage of the Westat report cited by the commenter examines the approximately 5% of individuals who receive a final non-confirmation from the E-Verify system and breaks that population down by the type of mismatch that caused the system to flag each person with an initial tentative non-confirmation. That analysis is graphically represented in Exhibit III-6 of the Westat report, which shows that 17% of those found unauthorized to work who claimed U.S. citizenship were flagged as “DOB not matched” and 13% of those found unauthorized to work who claimed U.S. citizenship were flagged as “Name not matched.” It appears that the commenter added 17% to 13% to arrive at the claim that “30%” of tentative non-confirmations are caused solely by errors in date of birth or name. The comment misses the mark for a number of reasons. First, the passage of the Westat report cited by the commenter looks at individuals who received a final non-confirmation stating that they were not authorized to work, and sorts individuals not by actual citizenship status but by citizenship status claimed by the individual. The population of unauthorized workers includes large numbers of individuals who falsely claim U.S. citizenship. By definition, the population relevant to the SEIA's calculation of no-match resolutions is entirely different, since it is limited to work-authorized persons. The comment also assumes, without explanation, that the workers with either a mismatched date of birth or a mismatched name correlate to the population that will be able to resolve the mismatch without visiting SSA. The passage of the Westat report cited by the commenter does not shed any light on the question of how many employees listed on a no-match letter will need to visit a Social Security office to resolve their mismatches. E-Verify and SSA's no-match letter program are distinct programs that rely on different input data sources and that examine different things. And the data summarized in Exhibit III-6 of the Westat report is simply not related to the subset of authorized employees that will choose to visit SSA.

10.  See SEIA, Appendix C: Estimation of Weighted Average Turnover Rates.

11.  See Technical Notes for May 2006 OES Estimates, “Estimation methodology” at http://www.bls.gov/​oes/​2006/​may/​oes_​tec.htm.

12.  SEIA, at 30-31, citing Institute for Survey Research, Temple University, and Westat, Findings of the Basic Pilot Program Evaluation (June 2002) at 170; Westat, Interim Findings of the Web-Based Basic Pilot Evaluation (Dec. 2006) at IV-17.

13.  A “tentative non-confirmation” can occur when an employee's name, date of birth, or social security number does not match SSA's records or if a death indicator is present in SSA's database.

14.  This average was calculated from the information DHS obtained from SSA by dividing the total number of mismatched SSNs listed in EDCOR letters by the total number of EDCOR letters.

15.  E-Verify also provides a thorough procedure for contesting and correcting records. If SSA is unable to verify information presented by the employee, the employer will receive an “SSA Tentative Nonconfirmation” notice. Similarly, if USCIS is unable to verify information presented by the employee, the employer will receive a “DHS Tentative Nonconfirmation” notice. Tentative nonconfirmation notices issues are issued for a variety of reasons, including mismatches of name, date of birth, invalid SSNs, mismatches in citizenship status or alien work authorization status or if a death indicator is present in SSA's database. If the individual's information does not match the SSA or USCIS records, the employee may contest the tentative nonconfirmation. To contest the tentative nonconfirmation, the employee must contact SSA or USCIS within eight federal government work days to try to resolve the discrepancy. Under the E-Verify program requirements, the employer is prohibited from terminating or otherwise taking adverse action against an employee who has contested a tentative nonconfirmation while he or she awaits a final resolution from the federal government. If the employee fails to contest the tentative nonconfirmation, or if SSA or USCIS concludes that the individual is not work authorized, the employer will receive a notice of final nonconfirmation and the employee may be terminated.