Proponents of E‑Verify, the Internet‐​based system to verify that a person is eligible to work in the United States, often tout its supposed speed and reliability. A recent Freedom of Information Act (FOIA) request from Cato has shed some light on how long it takes for the government to resolve contested tentative non‐​confirmations (TNC). The data should temper some enthusiasm for the system.

Our FOIA revealed that in 2012, the most recent year for which data are available, there were 68,775 contested TNCs through E‑Verify. A TNC is an initial E‑Verify determination that a worker is unlawful. Of those, 21,007 were handled by the Social Security Administration, with an average turnaround of 3.42 business days after the TNC was contested.

The Department of Homeland Security handled the other 47,768 contested TNCs, with an average turnaround of 6.01 business days. SSA deals with a lower volume of cases and deals with them in almost half the time that it takes DHS.

The information received as part of the FOIA included further breakdowns of resolution time:

As can be seen, not only was DHS slower in resolving its cases, but nearly half of its cases took longer than eight business days to resolve. Eight days is almost two full business weeks, but that is the minimum possible delay for that category. It could be weeks or months until those TNCs are resolved. Some 55.3 percent took over three business days for DHS resolve.

About 52 million American workers were hired in 2012 and they were subjected to 21.1 million queries. This gives an indication that today’s relatively short delays would increase if tens of millions of additional American workers are forced to use E‑Verify annually.

Worse, Sen. Jeff Sessions (R‑Ala.) has offered an amendment for national E‑Verify that would require verification of all existing employees who were not previously verified—with a three‐​month deadline. There is a total civilian labor force of roughly 155 million people. Running more than 100 million workers through E‑Verify within three months, as Senator Sessions’ amendment requires, would dramatically increase the wait time for resolving contested TNCs.

The 68,775 contested TNCs in 2012 came from a pool of 21.1 million total E‑Verify queries nationwide, about 0.33 percent. Importantly, that is the percentage of contested TNCs to actual queries. It’s likely that individual workers were run through E‑Verify by the same employer more than once, increasing the number of queries relative to the number of workers checked through E‑Verify. That means that the rate of contested TNCs per worker run through E‑Verify is actually higher than indicated here.

Applying that same rate to the 155 million workers Sessions envisions would produce 511,500 contested TNCs. About 36 percent of those, or 184,140, would take eight days or more to resolve—assuming (laughably) that a seven‐​fold increase in the number of people run through E‑Verify would have no effect on speed.

The economic costs of nationally mandating E‑Verify caused by TNC resolution delays would be higher than many of its proponents care to admit. That and the other economic costs of E‑Verify, the worrying privacy issues surrounding it, its lackluster performance, and how it makes everyone ask for permission from the government to work should temper enthusiasm for this scheme.

This post was written with the help of Scott Platton.