Every EB-5 investor must create at least ten full-time jobs for U.S. workers. A direct investor may count only employees actually hired by the new commercial enterprise (NCE). A regional center project may instead rely on an economic model to derive indirect and induced jobs, which is the central advantage of the regional center route. But “jobs the model produces” are not “jobs USCIS automatically accepts.” Understanding how the models work and what limits the law imposes is a prerequisite for assessing a project’s job-creation risk.
Three categories of jobs
| Category | Definition | Example (hotel development) |
|---|---|---|
| Direct | Positions employed by the project itself | Front desk, housekeeping, and restaurant staff after opening; construction workers where construction lasts two years or more |
| Indirect | Positions at other businesses created by the project’s purchases of goods and services | Employees of building-material suppliers, design firms, laundry services |
| Induced | Positions created when direct and indirect workers spend their wages locally | Employees of nearby restaurants, retailers, housing services |
Regional center investors may count all three; direct investors only the first. In USCIS usage, indirect and induced jobs are sometimes grouped together as “indirect,” and the RIA’s cap applies to that combined figure.
How input-output models work
A regional center project’s economic report usually relies on an input-output model. The underlying logic is that each additional unit of final demand in an industry generates a multiple of output and employment across the local economy through purchasing and wages. That multiple is the multiplier, and it varies by industry and region.
Commonly used models include:
- RIMS II (Regional Input-Output Modeling System), developed by the Bureau of Economic Analysis (BEA) of the U.S. Department of Commerce, which publishes multiplier tables by region and industry and is among the most frequently cited in EB-5 reports.
- IMPLAN, a commercial model from a private firm offering finer industry detail and regional data.
- Others such as REMI and REDYN appear in some reports.
USCIS accepts “reasonable methodologies” and does not mandate a particular model, but it expects the economist to explain the choice of model, data sources, geographic scope, and assumptions.
Common inputs
A model needs inputs to produce job counts. In EB-5 projects the typical inputs are:
- Construction spending. Hard costs (materials, labor) and some soft costs (design, engineering), converted to jobs by expenditure or by construction-industry output. Land acquisition, financing costs, and various fees generally cannot be used as inputs.
- Operating revenue. Revenue after opening (hotel room revenue, food and beverage revenue), converted to operating-phase jobs through the relevant industry multiplier.
- Direct operating headcount. Some reports take projected full-time hires as direct jobs and derive indirect and induced jobs from them.
Investors should check whether the inputs match the business plan and capital stack table, whether ineligible construction costs have been excluded, and whether revenue projections are supported by third-party market studies.
The RIA’s two caps
The EB-5 Reform and Integrity Act of 2022 (RIA) imposed two explicit limits on regional center job counting:
- 90% cap. Indirect jobs (including induced) may not exceed 90% of the jobs an investor needs. In other words, at least one of the required ten must be a direct job.
- 75% construction rule. Jobs from construction activity lasting less than two years may account for no more than 75% of the required jobs.
In practice this means a project relying entirely on short-term construction spending, with no operating-phase employment, has difficulty meeting the requirement on its own after the RIA. Either construction must run two years or longer, or operating-phase jobs must fill the gap. When reading an economic report, investors should confirm that the construction-duration assumption is stated and that the per-investor countable jobs are shown after the caps are applied.
Job cushion
The “job cushion” is the ratio of total projected jobs to the total jobs all investors need (number of investors times ten). For example, a project projecting 1,500 jobs and raising from 100 investors needs 1,000 jobs, a cushion of 1.5x.
The cushion absorbs uncertainty: construction may come in under budget, timing may shift, revenue may fall short, and USCIS may challenge some inputs. Industry practice treats a cushion below 1.2x as cause for caution and higher cushions as more robust, though no legal threshold exists.
Two pitfalls deserve attention. First, the denominator should be the maximum number of investors the project may ultimately admit, not the number subscribed so far. Second, the cushion should be computed after the RIA caps are applied; otherwise it is overstated.
Timing of job creation
Jobs cannot be counted whenever they happen to occur. The general principles are:
- Jobs must be created within the relevant period USCIS recognizes. The USCIS Policy Manual (Vol. 6, Part G) sets out when job creation must be complete at the I-829 stage and when an investor may instead show that jobs will be created within a reasonable time; current policy governs.
- For construction-based jobs, the spending generally must actually occur within the relevant period, supported by payment records, contractor certifications, and progress reports.
- For revenue-based jobs, the project generally must be open and achieving the projected revenue; otherwise jobs are prorated to actual revenue.
Investors should understand how their own I-526E timing relates to the project’s construction and opening schedule, and ask the regional center how it will document jobs at the I-829 stage.
Checklist for reading an economic report
- Which model is used, and does the multiplier region match the project location?
- Where do the inputs come from, and do they agree with the business plan, capital stack, and budget?
- What is the construction-duration assumption, and does the 75% rule apply?
- After the 90% cap, how many jobs are countable per investor?
- What is the cushion at the maximum investor count?
- How large is the operating-phase share, and is the revenue projection independently supported?
- Does the report describe the job-creation timeline and how jobs will be evidenced at I-829?
Key takeaways
- Regional center projects may count direct, indirect, and induced jobs derived from input-output models such as RIMS II and IMPLAN, using construction spending or operating revenue as inputs.
- The RIA limits indirect jobs to 90% of required jobs and jobs from construction under two years to 75%.
- The job cushion should be computed at the maximum investor count and after the caps; below 1.2x warrants caution.
- Jobs must actually be created within the period USCIS recognizes and be supported by spending or operating evidence, as detailed in the USCIS Policy Manual.
- Verifying that model inputs match the business plan is the single most important test of a job projection’s credibility.