Highest US annual real GDP growth

The starting probabilities of this market were calculated by simulating annual GDP growth paths from 2026 through 2034. Each path starts from current official GDP forecasts, samples annual forecast-error variance, allows recession/rebound shocks, and adds a modest-to-fat AI productivity uplift. The model asks a simple question for each threshold: how likely is it that at least one year in the window reaches this growth rate?

These probabilities assume the market can resolve using official U.S. GDP data.

This market has horizons that resolve independently:

• Before 2030 - highest annual real GDP growth across calendar years 2025 through 2029. The latest observed anchor is 2025 annual-average growth of about 2.1%, so thresholds below 3% are omitted.

• Before 2035 - highest annual real GDP growth across calendar years 2025 through 2034 inclusive.

Reference classes

The main baseline comes from current official forecasts. The March 2026 Federal Reserve Summary of Economic Projections has median real GDP growth of 2.4% in 2026, 2.3% in 2027, 2.1% in 2028, and 2.0% in the longer run; its historical forecast-error bands put roughly 70% of outcomes in a wide range that reaches about 4.5% to 5.2% over the next three years. That anchors the moderate part of the curve but leaves real upside risk. The Fed SEP values are Q4/Q4 growth rates, so they are used as baseline calibration rather than as the exact resolution metric.

Historical FRED GDPC1 data from 1948-2025 give a mean growth rate of 3.14%, with 38 of 78 years at or above 3%, 14 at or above 5%, 9 at or above 6%, and none at or above 10%. Rolling four-year windows reached at least 5% in 49% of post-1948 cases, while rolling nine-year windows reached at least 5% in 60%. The recent economy is lower-trend than the early postwar period, so these historical frequencies are shrunk toward the Fed baseline rather than copied directly.

As a check on the low end, historical years after a start close to 2025's 2.1% have almost always produced at least one later year that beat the starting year. Clearing 3% is less automatic in the lower-growth modern economy, but the combination of ordinary volatility, recession-rebound dynamics, and possible AI/productivity gains argues against leaving too much probability on every future year staying below 3%.

The AI channel can move both GDP and employment. The BLS 2024-2034 projection technical note says faster technological progress would show up as faster productivity growth and therefore higher GDP growth under its full-employment framework. If the productivity gain were broad and uniform, employment need not fall mechanically; if it is uneven across industries, the employment effects are much harder to project. This is why the model gives GDP an upside AI/productivity tail while treating labor-force dislocation separately in the LFPR market.

Why the 2035 distribution sits higher than the 2030 distribution

The 2035 window is twice as long as the 2030 window and has more chances to include one unusually strong year. The same path generator is used for both horizons; the 2035 distribution is higher because it observes more years from each simulated path.

Simulation inputs

Implied reference points

Influential examples

Overlap handling and uncertainty

Recession rebounds, annual volatility, and AI productivity gains are simulated inside the same annual path rather than counted as separate independent reasons to exceed a threshold. The AI tail keeps some probability on very high growth years without treating 10%+ annual real GDP growth as a central outcome.

Source links

• Federal Reserve March 2026 Summary of Economic Projections.

• FRED GDPC1 real GDP series, sourced from BEA.

• BEA fourth-quarter and annual 2025 GDP release.

• NBER U.S. business cycle dates.