Congress created EB-5 in 1990 to buy American jobs with foreign money. The Immigration Act of 1990 built a fifth employment based preference around one trade: an immigrant puts capital at risk in a new commercial enterprise, that enterprise employs at least ten Americans full time, and only then does the immigrant keep permanent residence. The green card is the payment for the jobs. It was never the product on sale, and that single design decision explains almost every EB-5 rule that frustrates investors today.
What the 1990 law actually traded
The category sits at section 203(b)(5) of the Immigration and Nationality Act. Read the employment based preference statute at 8 U.S.C. 1153 and the bargain is visible in the drafting itself. Three things must be true at the same time: the investor places capital at risk, the capital goes into a new commercial enterprise, and that enterprise creates full time positions for at least ten qualifying US workers. Congress then capped the whole category at 7.1 percent of the annual worldwide employment based visa supply, which tells you how modest expectations were in 1990.
The original price was $1,000,000, reduced to $500,000 if the money went into a Targeted Employment Area. Those figures were not chosen to sound serious. They were an estimate of how much capital it takes to support ten jobs in a typical American business. The price follows the jobs, not the other way round. That is still the logic behind today's numbers: $800,000 in a TEA, $1,050,000 outside one, with the first inflation adjustment scheduled for 1 January 2027.
There was a competitive motive too. Other countries already ran investor immigration schemes and were collecting capital that might otherwise have crossed the Atlantic or the Pacific into the United States. The American answer was deliberately harder to satisfy, because the political price of admitting anyone through an investment route was a measurable economic return.
Why the ten job rule is the requirement nobody waives
Most of an EB-5 file has some give in it. Source of funds can be evidenced several ways. A business plan can be amended. Processing can be chased. Job creation has no equivalent flexibility. There is no hardship exception, no partial credit for investing less, no way to substitute taxes paid, rent generated or square feet built. If the jobs are not there when you file Form I-829, the conditions on your residence are not removed and your family goes back to square one.
The definitions are narrow on purpose. Under the EB-5 regulation at 8 CFR 204.6, full time means at least 35 hours a week, and a qualifying employee has to be a US citizen, a lawful permanent resident or another immigrant authorized to work. You do not count. Your spouse and children do not count. Anyone present on a nonimmigrant visa does not count. The rules for removing conditions at 8 CFR 216.6 then ask for proof that the enterprise was created, that the capital was actually invested and sustained, and that the jobs exist.
Apply that as your first filter on any project you are shown. Not "will I get my money back", which matters enormously but comes second, but "who signs the paychecks, and will an adjudicator be able to trace ten of them to me". A deal that works as a business but creates its jobs through assumptions nobody can evidence is a bad EB-5 investment.
Targeted Employment Areas and the geography Congress wanted
The TEA discount exists because private capital does not naturally travel to the places that need it most. Congress offered a lower entry price in two kinds of location. Rural means outside a metropolitan statistical area and outside any city or town with a population of 20,000 or more. High unemployment means an area where the rate is at least 150 percent of the national average. Both tests run on public data: unemployment from the Bureau of Labor Statistics local area unemployment statistics, and the boundaries that decide rural status from the Census Bureau metropolitan and micropolitan area program.
For years that intent was blunted by TEA drawing. Long chains of census tracts were assembled from a prosperous development site out to a distressed neighborhood so that a downtown tower could claim the discounted price. The attempt to close that gap, and the court fight that followed, is set out in EB-5 Modernization: Rule Changes, Lawsuits and the Court Reversal. The 2022 statute finally moved TEA designation to the Department of Homeland Security and limited a high unemployment TEA to a census tract or tracts directly adjacent to it.
The 1992 regional center pilot and the drift from intent
Two years after EB-5 was created, Congress added a pilot program letting designated regional centers count indirect and induced jobs produced by economic modeling. That is the reason most EB-5 money now sits in pooled projects rather than businesses the investor personally runs, and it moved the job test from a payroll record to an input output model. The origin story is in EB-5 Regional Centers: The 1992 Pilot Program Explained, and the consequences that surfaced within a few years, including the precedent decisions that reshaped adjudication at the end of the 1990s, are covered in EB-5 in the 1990s: Early Controversies, Court Cases and Lessons for Today.
Whether modeled jobs honor the 1990 intent is still argued in Washington and never settled. Critics call an indirect job a spreadsheet output. Supporters point out that construction spending and supply chain purchases employ real people whom a narrow payroll test simply cannot see. Both positions have merit, which is why the fight recurs at every reauthorization. The public version of that argument is unpacked in Is EB-5 Just Buying a Visa? Public Perception vs Reality in 2026.
How the 2022 reforms restated the original bargain
The EB-5 Reform and Integrity Act of 2022, enacted as part of Public Law 117-103, reads like Congress repeating itself, louder. Set-asides steer 20 percent of visas to rural projects, 10 percent to high unemployment areas and 2 percent to infrastructure, which is the 1990 geography goal enforced through the visa queue instead of the price tag. Project level review arrived with Form I-956F, filed by the regional center before investors petition. An integrity fund pays for audits and site visits. USCIS sets out how it applies all of this in the USCIS Policy Manual chapter on immigrant investors.
The regional center program is authorized through 30 September 2027, with grandfathering that protects petitions filed by 30 September 2026 if the authority lapses. The full sequence of amendments since 1990 is laid out in Key EB-5 Amendments and Extensions: Timeline from 1990 to RIA 2022.
What 1990 intent means for your own file
Three practical consequences follow, and they are worth more to you than any brochure.
- At risk means at risk. Guaranteed returns, redemption promises and escrow that never really releases are not conservative structuring. They contradict the statute, and USCIS reads them as evidence there was no qualifying investment.
- Evidence beats ambition. A modest project with a defensible job model and clean documentation will clear adjudication ahead of a glamorous one whose numbers depend on optimistic inputs.
- The program lives on its economic record. EB-5 keeps its political license because it can point to jobs. That is why reauthorization debates, quota fights and integrity rules all circle back to the same question, as discussed in Will EB-5 Become Permanent? 2026 Debate on the Program's Future.
Read the 1990 design as a warning label rather than history. Congress was clear that the visa is contingent, and the modern adjudication file simply asks you to prove the contingency was met.