Skip to content
Menu
Finabase Learn

How markets and economies work Lesson 6 of 7

Economics · Foundations

Potential GDP and the output gap: how far the economy is from capacity

What CBO’s potential GDP and the output gap measure, how the gap moved through US recessions since 1949, how the Fed uses it, and why real-time estimates get revised.

On this page
  1. What potential GDP is
  2. How the gap is calculated
  3. What the data show
  4. How the Fed uses it
  5. Revisions: the biggest caveat
  6. Pitfalls
  7. The gap and related measures
  8. How Finabase does it
  9. Frequently asked questions
  10. Summary
  11. Check your understanding
  12. Exercise

Potential GDP is an estimate of how much the economy could produce if its workers and equipment were used at a high but sustainable rate: not flat out, but without the strain that pushes inflation up. The output gap is the difference between what the economy actually produced and that estimate. A negative gap means idle capacity: people who want jobs and cannot find them, factories running below their normal pace. A positive gap means the economy is running hot.

The gap is one of the most important numbers in macroeconomics and one of the least certain. It cannot be observed, only estimated, and the estimates change, sometimes by enough to flip the sign years later. In the third quarter of 2026, real GDP was $24,611 billion at an annual rate, and the Congressional Budget Office's estimate of potential was $24,200 billion: a gap of +1.70%. This lesson explains where those numbers come from, what the gap did in every US recession since 1949, how the Federal Reserve uses it, and how much the estimates have been revised.

What potential GDP is

Real GDP is what the economy produced, adjusted for inflation (see GDP explained: how gross domestic product is measured and read and Real vs nominal GDP: deflators, chained dollars and why both matter). Potential GDP is a model estimate of what it could produce at a sustainable rate of use of its labor and capital. Because it is a rate that can be exceeded, actual GDP sometimes runs above potential, at the cost of rising inflation pressure.

Long-run potential growth is, roughly, growth in hours that people can sustainably work plus growth in output per hour (Productivity: output per hour, unit labor costs and living standards). That is why potential grew much faster in the post-war decades than it does now: from 1949 to 1973 CBO's series grows about 4.0% a year, while from Q4 2007 to Q4 2017 it grows about 1.8%. In the projection loaded into FRED on February 27, 2026 (from CBO's Budget and Economic Outlook), CBO projects potential growth of about 2.0% a year from Q4 2025 to Q4 2035.

The labor-market twin: the noncyclical rate

CBO pairs potential output with an estimate of the unemployment rate that would prevail without swings in demand: the noncyclical rate of unemployment (earlier CBO publications called it the "natural rate of unemployment"; close to the idea of the NAIRU). It was about 5.3% in 1949, peaked at 6.24% in 1978 and stood at 4.39% in Q3 2026. CBO projects 4.16% for the end of 2036. The difference between the actual unemployment rate and this estimate, the unemployment gap, moves as a mirror image of the output gap, the relationship known as Okun's law.

How the gap is calculated

What the data show

US output gap and unemployment gap, quarterly, 1949 – Q3 2026 -8% -4% 0% 4% Q4 1982: −7.3% Q2 2020: −8.8% Q1 1966: +6.1% Real GDP minus CBO potential, % of potential -2 0 2 4 6 8 Unemployment rate minus CBO noncyclical rate, points 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025
Output gap (real GDP, GDPC1, versus CBO potential, GDPPOT) and unemployment gap (UNRATE quarterly average minus CBO NROU), Q1 1949 to Q3 2026. Shaded: NBER recessions. Data via FRED, downloaded 2026-10-30. Calculation: Finabase Research.

The gap falls in every recession on the chart. Some readings from the current data:

Quarter Output gap What was happening
Q1 1966 +6.1% Late-1960s boom; inflation began to climb
Q4 1982 −7.3% Trough of the 1981–82 recession
Q4 2007 +1.3% Peak before the financial crisis
Q2 2009 −5.0% End of the Great Recession
Q4 2017 0.0% Gap finally closed, eight years later
Q2 2020 −8.8% Pandemic lockdowns
Q3 2026 +1.7% Latest

Two patterns stand out. First, deep negative gaps can last for years: after the 2008–09 recession the gap stayed negative until the end of 2017. Second, the late 1960s show what a large positive gap does: the economy ran above its estimated potential for most of 1965 to 1969, and inflation rose through the decade (see Inflation: what it is, how it’s measured and what it does to money). The gap after 2020 closed within about a year and a half; it was +2.1% by Q4 2021.

How the Fed uses it

The Federal Reserve's mandate is maximum employment and stable prices (see The Federal Reserve: structure, mandate and policy tools). Both halves rely on some notion of capacity. The FOMC's own strategy statement, reaffirmed in January 2026, is careful about it: "The maximum level of employment is not directly measurable and changes over time." The Committee says it considers a wide range of indicators rather than one number.

Still, the output gap appears throughout policy analysis:

Revisions: the biggest caveat

Potential GDP is revised every time CBO updates its outlook, and actual GDP is revised by BEA for years. The gap moves with both. The St. Louis Fed's ALFRED archive keeps each vintage, which lets us compare.

Potential GDP as CBO saw it in early 2008 vs today, and actual GDP (Q4 2007 = 100) 95 100 105 110 115 120 125 130 128.8 119.1 117.5 Potential, Feb 2008 vintage Potential, current estimate Real GDP, current data 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Potential GDP as published in the 2008-02-01 vintage (CBO projection, chained 2000 dollars) and as currently estimated (chained 2017 dollars), with current real GDP, all indexed to Q4 2007 = 100. Data: CBO and BEA via FRED/ALFRED, downloaded 2026-10-30 and 2026-11-18. Calculation: Finabase Research.

Part of each revision comes from GDP itself and part from CBO learning that the trend had changed. After 2008, slower growth in the labor force and in productivity led CBO to lower its path for potential a great deal. A shortfall that looked like temporary slack partly turned out to be a permanently lower path. That is the core difficulty: in real time it is hard to tell a cyclical shortfall from a change in trend.

Pitfalls

  • Treating the gap as data. It is a model estimate built on other estimates. Quote it with its source and date.
  • Using today's estimate to judge past policy. Policymakers in 2008 saw a different gap from the one in today's data. Judge decisions with the information available then.
  • Ignoring the unemployment gap. When the two gaps disagree, the labor market often gives a clearer read on slack.
  • Different estimators. CBO, the IMF, the OECD and private forecasters all publish output-gap estimates, and they differ. This page uses CBO's because it is public and quarterly.
  • Supply shocks move potential. A pandemic or an energy shock can lower potential itself, so a fall in GDP does not always open a negative gap of the same size.
Measure What it tells you Source
Output gap GDP relative to sustainable capacity CBO estimate
Unemployment gap Unemployment vs the noncyclical rate BLS and CBO
Capacity utilization Share of industrial capacity in use Federal Reserve
Real GDP growth Change in output, not its level vs capacity BEA
Sahm rule Rise in unemployment from its recent low BLS data

Capacity utilization (see Industrial production and capacity utilization: the Fed’s G.17) is a measured cousin for industry; the The Sahm rule: a recession signal from the unemployment rate uses unemployment to flag recessions.

How Finabase does it

Finabase does not publish an output gap; the gap figures and charts on this page are our own calculation from CBO, BEA and BLS series. In the app:

  • The Economy page charts the unemployment rate, CPI inflation and the Fed funds rate, the numbers the gap is usually read beside, and its calendar lists scheduled US releases, GDP among them, with the forecast and the actual figure.
  • Expectations about how hot the economy is running show up first in short-term yields; the 2-year Treasury yield has its own page and chart.
  • The Fed news topic collects coverage of how officials describe slack.

Frequently asked questions

What is the output gap?

The output gap is the difference between actual real GDP and potential GDP, as a percent of potential. A positive gap means the economy is producing more than its sustainable level; a negative gap means it has idle capacity. On CBO's estimate it was +1.70% in Q3 2026.

Who estimates potential GDP?

In the US the most widely cited estimate is the Congressional Budget Office's, published with its Budget and Economic Outlook and available on FRED as GDPPOT. The Fed, the IMF, the OECD and private forecasters make their own estimates.

Is a positive output gap good or bad?

Neither in itself. It means more output and jobs today, but if it persists it tends to push wages and prices up, which is why central banks lean against it.

Why do output gap estimates get revised?

Because both inputs change. BEA revises GDP for years, and CBO revises potential when new data change its view of the labor force and productivity trends. In February 2008 data, Q4 2007 was 1.1% below potential; in today's data it was 1.3% above.

Summary

Potential GDP is CBO's estimate of what the economy can produce at a high, sustainable rate of use of labor and capital, and the output gap is actual real GDP's distance from it. The gap fell in every US recession since 1949, reaching −7.3% in Q4 1982 and −8.8% in Q2 2020, and stood at +1.70% in Q3 2026, with unemployment slightly below CBO's 4.39% noncyclical rate. The gap feeds policy rules, inflation models and budget analysis. But it is an estimate, and the real-time version can be badly wrong: at the end of 2007 it looked like slack and is now recorded as a boom.

Check your understanding

4 questions on this lesson. Each answer comes with the reason, and nothing is scored.

  1. Real GDP is $20,400 billion and CBO estimates potential at $20,000 billion. What is the output gap, and what does it suggest?

    1. +2%: the economy is running above its sustainable level, a sign of upward pressure on wages and prices
    2. −2%: the economy has idle capacity
    3. +$400 billion, which says nothing about inflation
    Show the answer

    A. +2%: the economy is running above its sustainable level, a sign of upward pressure on wages and prices (20,400 − 20,000) ÷ 20,000 = +2%. A positive gap means output above what CBO judges sustainable, the usual sign of inflation pressure.

  2. In data published in February 2008, Q4 2007 GDP was 1.1% below potential. In today’s data it was 1.3% above. What is the main lesson?

    1. BEA made a mistake in 2008
    2. Real-time gap estimates can be revised enough to change their sign, because both GDP and potential are revised
    3. The economy was in recession in Q4 2007
    Show the answer

    B. Real-time gap estimates can be revised enough to change their sign, because both GDP and potential are revised CBO later lowered its path for potential sharply, and GDP was revised. Judging policy in hindsight with today’s gap uses information policymakers did not have.

  3. The output gap is positive, but unemployment is well above CBO’s noncyclical rate. What should you conclude?

    1. The economy is clearly overheating
    2. The two measures of slack disagree, so treat the gap estimate with extra caution and look at other labor-market data
    3. The unemployment rate must be wrong
    Show the answer

    B. The two measures of slack disagree, so treat the gap estimate with extra caution and look at other labor-market data Normally the two gaps move as mirror images (Okun’s law). When they diverge, at least one estimate is off, and the labor market is often the clearer read.

  4. Why did potential GDP grow about 4% a year in 1949–73 but under 2% a year in 2007–17?

    1. Because recessions were more frequent after 2007
    2. Because potential growth is roughly growth in sustainable hours plus growth in output per hour, and both slowed
    3. Because CBO changed the base year of its dollars
    Show the answer

    B. Because potential growth is roughly growth in sustainable hours plus growth in output per hour, and both slowed Potential depends on labor-force growth and productivity growth. A slower-growing workforce and slower productivity growth lower the sustainable pace of the whole economy.

Exercise

One task to do in Finabase. Nothing is graded or saved.

Open the Economy page and read the unemployment rate chart. CBO’s noncyclical rate, its estimate of normal unemployment, was 4.39% for Q3 2026. Is the latest reading above or below it? Then find the lowest point on the five-year chart, how far it sat below 4.39%, and check the CPI chart for what inflation was doing in the same months.

Open the Economy page
What you should find

The lowest unemployment on the chart, in 2022–23, sits roughly a point below CBO’s rate, a tight labor market that matches the positive output gap (+2.1% in Q4 2021) and came with the highest inflation of the period. Unemployment a little below 4.39% now is the same message as the +1.70% gap. Both are estimates and get revised.

Sources and further reading

  1. Congressional Budget Office — Real potential GDP, chained 2017 dollars (FRED series GDPPOT; Budget and Economic Outlook data, FRED update of February 27, 2026), downloaded 2026-10-30
  2. Congressional Budget Office — Noncyclical rate of unemployment (FRED series NROU), downloaded 2026-10-30
  3. BEA — Real gross domestic product, chained 2017 dollars (FRED series GDPC1), downloaded 2026-10-30
  4. BLS — Unemployment rate (FRED series UNRATE), downloaded 2026-10-30
  5. ALFRED (Federal Reserve Bank of St. Louis) — GDPPOT and GDPC1 as published on 2008-02-01 and 2018-05-01, downloaded 2026-11-18
  6. FOMC — Statement on Longer-Run Goals and Monetary Policy Strategy (reaffirmed January 27, 2026), retrieved 2026-11-18