Gross Domestic Product (GDP) is the total market value of all final goods and services produced within a country's borders in a given period — the single broadest scorecard of economic activity. "Growth" refers to the rate of change in real (inflation-adjusted) GDP, the number traders actually watch. The core tension for markets is this: GDP is the definitive measure of how big the economy is, yet it is a backward-looking, heavily-revised, low-frequency number, and — counterintuitively — its relationship to stock returns is far weaker than intuition suggests. GDP tells you where the economy was, not where stocks are going.

How it's calculated / measured

GDP can be measured three equivalent ways: the expenditure approach (C + I + G + NX — consumption, investment, government spending, net exports), the income approach (sum of all incomes earned), and the production/value-added approach. The expenditure identity is the one quoted in headlines, and personal consumption (C) is the largest component, roughly two-thirds of U.S. GDP per the Bureau of Economic Analysis (BEA).

Two distinctions matter most:

• Nominal vs. real. Nominal ("current-dollar") GDP is valued at the prices of the period; real GDP strips out price changes using a deflator. Real GDP growth is the figure that drives policy and markets — nominal growth can be high simply because of inflation.
• Headline reporting. U.S. GDP is reported as a quarterly, seasonally-adjusted annualized rate (SAAR) — the quarter's growth compounded as if it ran a full year. Most other countries report quarter-on-quarter or year-on-year, so cross-country comparisons require care.

The BEA releases each quarter in three vintages: an Advance estimate ~one month after the quarter ends (incomplete source data), then Second and Third estimates in the following two months as more data arrives. Annual and comprehensive (benchmark) revisions can move figures years later. The practical consequence: the first print is an estimate, sometimes substantially revised. Because the official number lags so badly, traders lean on nowcasts — the Atlanta Fed's GDPNow and the New York Fed's Nowcast assemble incoming monthly data into a running real-time estimate of the current quarter.

How it's used in practice

GDP and its growth rate function as regime context, not as a trade trigger:

• Business-cycle positioning. The expansion → peak → contraction → trough cycle frames sector rotation and risk appetite. Accelerating growth favors cyclicals (industrials, consumer discretionary, financials); decelerating or contracting growth favors defensives (staples, utilities, healthcare).
• Fed reaction function. Strong real growth raises the odds of tighter policy; weak growth raises the odds of cuts. Markets often trade the implication for rates more than the GDP number itself.
• Earnings backdrop. Nominal GDP is a rough ceiling on aggregate corporate revenue growth — useful for top-down sanity checks on consensus S&P 500 earnings.
• Nowcasting the surprise. What moves markets is the deviation from consensus, not the level. GDPNow and the consensus forecast set the expectation; the release is scored against it. By release day, much of the information is already priced.

Adoption, debate & evidence

GDP is universally adopted as the output measure — by central banks, the IMF, and every macro desk. But for equity investors there is a large, robust, and underappreciated evidence base showing the GDP-growth → stock-return link is weak to negative.

Jay Ritter's "Economic Growth and Equity Returns" (2005) found the cross-sectional correlation between compounded real per-capita GDP growth and compounded real equity returns over 1900–2002 was approximately −0.37 for 16 developed countries — i.e. faster-growing countries delivered lower, not higher, returns. (Over the shorter, lower-quality 1970–2002 window for 19 countries the correlation was a statistically insignificant ~0.08, illustrating how sample-dependent the figure is.) Ritter's updated Is Economic Growth Good for Investors? (2012), covering 1900–2011 across 19 countries, reports a cross-sectional correlation of roughly −0.39 (and ~−0.41 for emerging markets, 1988–2011). MSCI and Dimson-Marsh-Staunton's work reach the same qualitative conclusion: countries that grew faster did not deliver higher equity returns.

Why? Ritter's explanations: (1) growth often comes from new firms and IPOs whose value never accrues to existing shareholders (dilution); (2) technological gains flow to consumers and labor unless firms hold durable monopolies; (3) high-growth expectations get priced in, so realized returns disappoint. The lesson: GDP growth is not a buy signal for the index.

On nowcast accuracy, the Atlanta Fed reports GDPNow's average absolute error of final forecasts is about 0.77 percentage points since 2011, shrinking from roughly 1.1 pp about 90 days before the first GDP release to roughly 0.5 pp just before release. The Atlanta Fed's own assessment is that GDPNow is competitive with survey-based consensus forecasts but not consistently superior — it beats the average individual panelist closer to the release, yet a blend of GDPNow with other forecasts outperforms either alone. Useful, but not precise.

A persistent myth worth flagging: "two consecutive quarters of negative GDP equals a recession." The BEA and NBER both reject this. NBER's Business Cycle Dating Committee defines a recession as a "significant decline in economic activity that is spread across the economy and lasts more than a few months," judged across multiple monthly indicators. The NBER-dated 2020 recession lasted only two months (February–April 2020), too short to register two full quarters; and the 2001 recession was officially a recession despite not having two consecutive quarters of negative real GDP (its negative quarters were non-consecutive). NBER also dates recessions with a long lag — often roughly a year after the fact.

Strengths & limitations

Strengths: the most comprehensive, internationally comparable measure of output; the anchor for cycle analysis, policy, and top-down revenue framing; nowcasts give a usable real-time read.

Limitations: lagging and revised (the Advance print is provisional); low-frequency (quarterly); a summary statistic that hides composition (a strong headline driven by inventory build or a falling trade deficit can mask weak final demand); and — the #1 misuse — treating GDP growth as a stock-return predictor. It is not, at the index level, per the cross-country evidence. GDP also omits distribution, informal activity, and well-being, but those critiques matter less to markets than the return-disconnect.

Sources

• BEA — What to Know About GDP (definition, components, real vs. nominal, release vintages)
• BEA — GDP Release Additional Information (advance/second/third estimates, revisions)
• Atlanta Fed — GDPNow (nowcast construction, accuracy vs. Blue Chip)
• Jay R. Ritter — Economic Growth and Equity Returns (2005) (−0.37 cross-sectional correlation, 16 countries, 1900–2002) — landmark, widely cited
• Jay R. Ritter — Is Economic Growth Good for Investors? (2012, J. Applied Corporate Finance) (−0.39, 19 countries, 1900–2011; −0.41 emerging markets 1988–2011)
• MSCI — Is There a Link Between GDP Growth and Equity Returns? (2010) and Dimson-Marsh-Staunton (qualitatively corroborating evidence)
• NBER — Business Cycle Dating (recession definition; "two quarters" myth)
• Britannica Money — GDP (lagging-indicator framing)

Disputed/contested: the magnitude (and sign) of the GDP-growth/equity-return correlation varies by sample, period, and currency basis — Ritter's long-run developed-market figures are negative (≈−0.37 to −0.39), while shorter sub-periods (e.g. 1970–2002) can be near-zero or slightly positive (≈0.08). The direction of the claim (growth does NOT reliably predict returns) is robust across authors (Ritter, MSCI, Dimson-Marsh-Staunton); the precise coefficient is not. Treat any single correlation figure as sample-dependent.