Divide the project's total projected jobs by the maximum number of investors it will admit. If that ratio comes out at 10.0, the deal has no margin for error and you should not invest. Insist on a cushion above the ten jobs per investor the statute demands, so that a delayed opening or a soft first year does not drag the count under the line. A fifth above the minimum is a sensible floor. Everything else in the analysis, from the economist's model to the operating forecast, exists to test whether that cushion is real.
Start with jobs per investor, not total jobs
A headline of 2,400 jobs means nothing by itself. Do the division.
A $60 million EB-5 raise at $800,000 per investor admits 75 investors and needs 750 jobs, so 2,400 projected jobs give a cushion above three times the minimum. Push the same 2,400 jobs against a $180 million raise and you have 225 investors needing 2,250 jobs, a cushion under 7 percent. Identical project. Wildly different risk.
Use the maximum authorized raise rather than the amount subscribed today, because the sponsor is entitled to fill it. Investors who ran the math on a half raised offering have watched their cushion evaporate as later subscribers came in.
Where the jobs come from, and which kinds are riskiest
Three categories exist. Direct jobs are actual W-2 positions at the job creating entity. Indirect jobs are modeled positions at suppliers. Induced jobs are modeled positions created when wages get spent in the local economy.
Only Regional Center investors receive indirect and induced credit. A direct investor counts payroll employees and nothing else, which is why job counts across the two paths are not comparable, and Regional Center vs Direct EB-5: Two Projects and Their Business Plans sets the structures side by side.
Modeled jobs are not fabrications. They are outputs of a model whose inputs you can inspect, and the inputs are where the risk lives.
Reading the economic impact report without an economics degree
The report will run a recognized input output model. RIMS II multipliers come from the Bureau of Economic Analysis. IMPLAN and REMI are the other two names you will meet. Which model got used matters far less than what was fed into it.
Four inputs deserve your attention.
- Hard construction cost. Land acquisition creates no jobs and must be stripped out of the expenditure figure, along with soft costs the model does not accept.
- Revenue assumptions behind operating jobs. A hotel modeled at 78 percent occupancy in its first year is telling you something about the economist's optimism.
- Geography. A county multiplier differs from a metropolitan statistical area multiplier, and the wider the defined region, the larger the number that falls out.
- Timing. A report assuming stabilized operations 18 months after groundbreaking is assuming nothing goes wrong.
Then perform the one reconciliation an ordinary investor can do unaided. Compare the expenditure figure in the economic report against the construction budget in the offering documents. If the report used a larger number, ask why and keep asking until somebody explains it in writing. Our explainer on EB-5 Economic Impact Reports: How Job Numbers Are Calculated and Checked walks through the arithmetic.
Construction jobs versus operating jobs
Construction jobs arrive early and then vanish. Operating jobs arrive late and persist. For your petition, early wins.
The EB-5 Reform and Integrity Act of 2022 capped how much of a count may come from each source. Indirect jobs may account for no more than 90 percent of the total. Positions arising from construction activity lasting less than two years may account for no more than 75 percent. A project leaning on a 14 month build for the bulk of its jobs is running near that ceiling, and the statutory language sits in 8 U.S.C. 1153, the employment based preference statute.
Where construction activity runs two years or longer, those positions have long been treated as direct jobs rather than being squeezed under the 75 percent limit. That is one quiet reason large ground up developments keep dominating the Regional Center pipeline.
Timing: can the jobs land inside your own two years?
The jobs have to be in place when Form I-829, the petition to remove conditions is adjudicated, and conditional residence runs two years from the date you obtain it. USCIS can accept jobs created within a reasonable time after that period, but the allowance is discretionary and no schedule should depend on it. Line the construction schedule up against your personal timeline instead of against the sponsor's marketing calendar.
Ask when escrow releases. Ask when the first shovel goes in the ground. A project that has not started when you file Form I-526E for Regional Center investors has not started its job clock, which for an applicant from a backlogged country can work in your favor rather than against you.
The opposite risk is nastier. Where a project finished its construction spend three years before your money arrives, those construction jobs may already belong to earlier investors. Request a job allocation table showing which jobs have been assigned to which subscription tranche. A sponsor who cannot produce one has told you something important.
Troubled business deals and job preservation
Buying a failing company changes the test entirely. Under 8 CFR 204.6, which defines the troubled business, the enterprise must have existed for at least two years and have lost at least 20 percent of its net worth during the 12 or 24 month period before the petition. Preserving existing jobs at no less than the pre-investment level for at least two years can then substitute for creating new ones.
Verification is document heavy. Pull payroll for the 24 months before the investment and count the full time positions yourself. Check whether the turnaround plan addresses whatever actually caused the losses. A rescue that leaves the underlying problem untouched preserves nothing beyond the first year.
Ask these before you wire $800,000
- What is the maximum EB-5 raise, and what does the job cushion look like at that maximum rather than at today's subscription level?
- How many jobs have already been allocated to investors admitted ahead of me?
- What share of the count comes from construction, and does the construction period run at least two years?
- Which economist prepared the report, and will the sponsor share the model inputs rather than the summary page?
- What happens to the count if the project opens 12 months behind schedule?
None of those questions is impolite. A sponsor who treats them as impolite has answered the most important one already.
For the broader framework, EB-5 Due Diligence 2026: Beginner Checklist to Avoid Bad Projects and Fraud covers the elements that have nothing to do with jobs, while I-829 Evidence: Proving EB-5 Job Creation Requirements and Capital at Risk shows what you will eventually have to demonstrate. The USCIS Policy Manual guidance for immigrant investors is the authority your lawyer and the economist are both working from, and reading a few pages of it will sharpen every question above.