Why REMI

Run the economy twice.
Measure the difference.

REMI pairs long-range economic and demographic forecasting with dynamic modeling. A model runs forecasts the region on its current path, the baseline, then runs that same economy with a policy, an investment, or a project in place. The gap between the two is the impact.

Baseline vs. scenarioHow every REMI result is read
Baseline compared with a policy scenarioTwo rising lines share a starting point in 2026 and separate over time. The dashed line is the baseline forecast; the solid line is the same economy with a policy change applied. The widening gap between them is the impact of the policy.ImpactScenarioBaseline2026203520452060Economic activity

This is a schematic, not a forecast. The size and timing of a real impact depend on the region, the policy, and the assumptions behind it.

What you get

The REMI Forecast

Walk the path ahead.
Compare what changes.

An impact isn’t one headline number — it’s a path. Each model reports the difference between baseline and scenario year by year through 2060, across jobs, output, income, and population, so you can see when an effect arrives, how large it gets, and how long it lasts.

  • Year-by-year results
  • Baseline vs. scenario
  • Forecasts through 2060
SEE MORE PUBLISHED STUDIES
View forecast:

Published Study

Regional Jobs Generated by the Gateway ProgramNew York & New Jersey

Step through the years: construction jobs peak in the mid-2030s, then commuting and productivity benefits carry the total once the tunnel opens in 2045.

Checkpoint year: 2060

Jobs in New York and New Jersey generated by the Gateway Program, 2023 to 2060A stacked area chart of jobs in thousands, for the New York and New Jersey region. Construction of Phase 1 and Phase 2, with supporting projects, drives the total to roughly 30 thousand jobs in the mid-2030s. Once the tunnel opens in 2045, commuting, productivity, and operations and maintenance benefits sustain roughly 25 thousand jobs a year through 2060, averaging 24.4 thousand across the period.24.4k avg0k10k20k30k40k50k20232030204020502060Jobs
  • O&M Spending Benefits
  • Productivity Benefits
  • Commuting Benefits
  • Gateway Full Build Scenario / Phase 2
  • Gateway Full Build Scenario / Phase 1
  • Supporting Projects - Construction

Modeled in REMI TranSight by Regional Plan Association and WSP USA: The Economic Promise of the Gateway Program. The chart plots the New York and New Jersey region; the study’s average of 46,100 jobs is the national total.

How it works

Dynamic Economic Modeling

Change one thing.
Watch the rest respond.

A change in one part of the economy creates responses in the others. Employment affects population, population shifts demand, and demand moves output, prices, and trade. Those responses feed back into each other year after year — which is what makes a long-range forecast possible rather than a guess extended forward.

  • Five linked blocks
  • Feedback every year
  • No fixed multipliers
SEE ALL MODEL DOCUMENTATION
The full structure of the REMI model: five blocks — output and demand, labor and capital demand, population and labor supply, compensation, prices and costs, and market shares — connected by arrows showing how each feeds back into the others.
  1. Output & Demand

    Balances supply and demand across the economy — output, consumption, investment, government spending, imports, and exports.

  2. Labor & Capital Demand

    Sets employment and capital stock from output, wage rates, and the cost of capital, industry by industry.

  3. Population & Labor Supply

    Tracks migration, population, and labor force participation as economic opportunity shifts between regions.

  4. Compensation, Prices & Costs

    Follows wages, consumer and housing prices, and production costs as they adjust to the changes above.

  5. Market Shares

    Turns production cost into each industry's domestic and international market share — which feeds back into output and demand.

Static

ChangeResult

One estimate built on fixed multipliers. Prices don’t move, nothing substitutes, and the answer is the same in year one and year twenty.

Dynamic

ChangeResponseFeedbackNew response

A year-by-year path. Firms and households react to price signals, labor and capital adjust, people move — and the implicit multiplier changes as they do.

Why it holds up

Built for Scrutiny

Defend the number.
Show the method.

Results that shape a budget or a statute get challenged. REMI models are built by economists, documented in the published literature, and used by agencies on every side of a question — so when someone asks how you got the number, you can show them.

  • 40+Years of service
  • 2060Forecast horizon
  • 1000+Published studies
  • Economist-Developed

    Built by a team of economists with decades of experience in regional, industry, and policy analysis.

  • Published Methodology

    The economic relationships, equations, data sources, and estimation methods behind the models are documented in peer-reviewed work, not summarized in a brochure.

  • Official Data Sources

    Models are built on official sources — the Census Bureau, the Bureau of Economic Analysis, and the Bureau of Labor Statistics among them.

  • Reproducible Results

    Documented methods and consistent assumptions mean another analyst can rerun your scenario and reach your numbers.

Assume the number will be checked.

  • The model documentation — equations, data sources, and estimation methods, published in full rather than described.
  • The published literature — methods that have been through peer review and are cited in regional economics research.
  • Your own scenario — hand a reviewer your assumptions and the model returns your numbers, not approximately.
REMI allows us to deliver top-notch economic analysis for our members and partners. It enhances our offerings to members and makes us thought leaders when we break new ground with cutting-edge analysis. We couldn’t do all that without REMI.
Curtis Dubay, Chief Economist, U.S. Chamber of Commerce

Bring your question to the model.

Tell us the policy, the geography, and the timeline. We’ll point you at the model built for it — or tell you if there’s a simpler way to get your answer.