Transparency

Data sources & methodology

Good teaching tools show their work. Here is exactly where the data comes from and how the simulator calculates its results.

Historical data (History Replay & quizzes)

SeriesSourceCoverage
S&P 500 / Composite stock index (monthly average)Robert J. Shiller, Irrational Exuberance online data (Yale University), via the open datasets/s-and-p-500 package1871–present
Consumer prices & inflationU.S. Bureau of Labor Statistics CPI-U (1913–present); Shiller CPI series before 19131871–present
Long-term interest rateShiller long-term government bond yield (to 1953); Federal Reserve 10-year Treasury constant maturity (GS10) after1871–present
Real GDP growthU.S. Bureau of Economic Analysis, NIPA real GDP (chained dollars) — annual 1929–1947, quarterly year-over-year after1929–2024
Unemployment rateBLS Current Population Survey annual averages (1941–present; ages 14+ before 1947); Stanley Lebergott’s estimates as published in Historical Statistics of the United States (1929–1940); monthly BLS data for 2020–211929–present
Federal funds rateFederal Reserve (FRED: FEDFUNDS), annual averages, with monthly values for 2008, 2020 and 20221955–present
Home pricesS&P CoreLogic Case-Shiller U.S. National Home Price Index (not seasonally adjusted)1975–present
World GDPWorld Bank, World Development Indicators (current US$)1960–present
Recession datesNational Bureau of Economic Research (NBER) Business Cycle Dating Committee1873–present

When the site administrator connects the Federal Reserve Bank of St. Louis FRED API, official monthly series (UNRATE, CPIAUCSL, FEDFUNDS, GS10, MORTGAGE30US, GDPC1, CSUSHPINSA, M2SL) replace the built-in values. Annual figures are drawn as step lines so students can see their frequency. Data can be revised by the agencies; check the original sources for research use.

How the simulator works

The simulator is a small, transparent teaching model of the U.S. economy that runs in your browser, one quarter at a time for five years. It is built from relationships taught in introductory macroeconomics, with round-number coefficients chosen so effects have realistic signs and sizes. It is not a forecast.

  1. Aggregate demand (IS curve). The output gap moves gradually toward a target set by the real interest rate relative to its neutral level, mortgage rates, fiscal policy (spending raises demand roughly one-for-one; tax changes a bit less), confidence, credit conditions, money growth, world growth, tariffs, energy prices, disasters and wealth effects from stocks and housing.
  2. Okun’s law. Unemployment moves toward the natural rate minus half the output gap.
  3. Phillips curve with expectations. Inflation = expected inflation + 0.28 × output gap + supply shocks (energy, tariffs, disasters, productivity) + excess money growth. Expectations adapt to actual inflation and are gently anchored to 2%.
  4. Monetary policy. The Fed holds the rate you choose, or (with Auto on) follows a Taylor rule: neutral rate + inflation + 0.5 × (inflation − 2) + 0.5 × output gap.
  5. Interest rates. The 10-year yield blends the current policy rate with the neutral rate plus expected inflation and a term premium that rises with deficits and debt. Mortgage rates add a spread that widens when lending tightens.
  6. Housing and stocks. Home-price growth responds to mortgage rates, credit, zoning, population, inflation and the output gap. Stock prices follow nominal GDP and profit margins, discounted by bond yields plus a risk premium that rises with fear.
  7. Government budget. Deficit = spending + automatic stabilizers + interest − revenue (which rises in booms). Debt grows with deficits and shrinks relative to a growing nominal GDP.
  8. Status labels (recession, depression, stagflation…) are rules of thumb applied to the simulated numbers each quarter.

The model’s full source code is in assets/js/engine.js and can be read by anyone. The History Replay “what-if” lines add the simulator’s estimated difference between a changed and unchanged economy to the real historical path — a rough illustration, not a precise counterfactual.