A March 9 USA Today article reported that studies by conservative think tanks stated that "[t]ens of thousands of jobs created by the economic stimulus law could end up filled by illegal immigrants" and quoted a Center for Immigration Studies spokesperson saying a requirement that employers use E-Verify, a program that allows employers to check the eligibility status of newly hired employees through government databases, “could have deterred this.” The article later stated that “business groups and immigrant advocacy groups argue that the E-Verify database is riddled with errors that could result in millions of workers being wrongly identified as not authorized for work.” However, the article did not note that this “argu[ment]” by “business groups and immigrant advocacy groups” is backed up by federal government reports.

A September 2007 report on a pilot program for E-Verify commissioned by the Department of Homeland Security (DHS) found that “the database used for verification is still not sufficiently up to date to meet the IIRIRA [Illegal Immigration Reform and Immigrant Responsibility Act of 1996] requirement for accurate verification, especially for naturalized citizens.” In addition, evaluations of the E-Verify program by the Government Accountability Office (GAO) and Social Security Administration (SSA) have also found that the databases used by the E-Verify system contain errors that can misidentify millions of qualified workers, especially foreign-born citizens, as being potentially ineligible for employment.

The 2007 report prepared for DHS found that:

The erroneous tentative nonconfirmation rate for employees found to be work-authorized at any time during the Web Basic Pilot process in the first half of fiscal year 2007 was less than 1 percent (0.53 percent); the estimated erroneous tentative nonconfirmation rate for all cases sent to the Web Basic Pilot in this timeframe was 0.81 percent. Although the erroneous tentative nonconfirmation rates for all employees are fairly low, they are much higher for foreign-born citizens than for U.S.-born employees and noncitizens. (The erroneous tentative nonconfirmation rate for ever-authorized foreign-born citizens between October 2004 and 2007 is approximately 10 percent.)

The report also stated that as a result of errors in the databases used by E-Verify, the pilot program “increased discrimination against work-authorized foreign-born employees after hiring.” The report later explained how these errors resulted in discrimination:

The Web Basic Pilot increased discrimination against work-authorized foreign-born employees after hiring because foreign-born employees, especially foreign-born citizens, are more likely than U.S.-born employees to receive tentative nonconfirmation findings, with the attendant burdens of contesting erroneous findings. The burden of erroneous tentative non-confirmations on employees is exacerbated by the failure of some employers to follow Basic Pilot procedures designed to protect employee rights. This failure not only results in additional discrimination but also contributes to non-protected groups being denied their rights.

The report also found that “the rate of employer noncompliance [with pilot program rules] is still substantial.” Types of noncompliance included screening job applicants before hire, which is prohibited because employers may turn away applicants receiving potentially erroneous rejections; not notifying employees of the rejection finding; not adequately explaining the process for contesting the finding; discouraging employees from contesting the finding; and taking adverse actions against employees in the process of contesting the rejection, such as reducing pay or requiring them to work more hours.

A June 10, 2008, GAO report about the challenges facing implementation of a mandatory E-Verify system stated: “About 7 percent of the queries cannot be immediately confirmed as work authorized by SSA, and about 1 percent cannot be immediately confirmed as work authorized by USCIS because employees' information queried through the system does not match information in SSA or DHS databases.” The report stated that the majority of these tentative nonconfirmations occur “because employees' citizenship status or other information, such as name changes, is not up to date in the SSA database, generally because individuals have not notified SSA of information changes that occurred.” The report continued:

SSA updates its records to reflect changes in individuals' information, such as citizenship status or name, when individuals request that SSA make such updates. USCIS [U.S. Customs and Immigration Services] officials stated that, for example, when aliens become naturalized citizens, their citizenship status, updated in DHS databases, is not automatically updated in the SSA database. When these individuals' information is queried through E-Verify, a tentative nonconfirmation would be issued because under the current E-Verify process, those queries would only check against SSA's database; they would not automatically check against DHS's databases. Therefore, these individuals would have to go to an SSA field office to correct their records in SSA's database.

In December 2006, SSA assessed the accuracy of its databases and concluded that, based on the number of errors found in the sample of 2,430 records reviewed, approximately 17.8 million records (4.1 percent) of the SSA's 435 million records need to be updated and therefore “could result in incorrect feedback” when submitted through the E-Verify system. Moreover, the report stated:

We are particularly concerned with the extent of incorrect citizenship information in SSA's Numident file for the foreign-born U.S. citizens and non-U.S. citizens we tested.

[...]

[W]e noted that 57 (7.0 percent) of the 810 non-U.S. citizen Numident records we reviewed were currently misclassified -- because the individuals had become U.S. citizens after obtaining their SSN but had not updated their records with SSA.

From the USA Today article:

Tens of thousands of jobs created by the economic stimulus law could end up filled by illegal immigrants, particularly in big states such as California where undocumented workers are heavily represented in construction, experts on both sides of the issue say.

Studies by two conservative think tanks estimate immigrants in the United States illegally could take 300,000 construction jobs, or 15% of the 2 million jobs that new taxpayer-financed projects are predicted to create.

They fault Congress for failing to require that employers certify legal immigration status of workers before hiring by using a Department of Homeland Security program called E-Verify. The program allows employers to check the validity of Social Security numbers provided by new hires. It is available to employers on a voluntary basis.

“They could have deterred this, but they chose not to,” said Steven Camarota, director of research for the Center for Immigration Studies.

He said a federal requirement that employers use E-Verify would have reduced, if not eliminated, the hiring of immigrants in this country illegally.

[...]

The version of the stimulus bill passed by the House of Representatives included a provision requiring employers to check immigration status with the E-Verify system before hiring. The Senate did not include such a provision, and it was not in the version sent to President Obama. The Obama administration has delayed until at least May 21 a Bush administration executive order requiring federal contractors to use the E-Verify system in hiring. It had been scheduled to take effect in January. The U.S. Chamber of Commerce filed suit seeking to block the requirement, joined by the Associated Builders and Contractors and other business organizations.

The business groups and immigrant advocacy groups argue that the E-Verify database is riddled with errors that could result in millions of workers being wrongly identified as not authorized for work. They say requiring its use before hiring would impose a cost burden on employers and open them to lawsuits.