Do GDP, Inflation, and Unemployment Predict Presidential Elections?

Not reliably. In a study of all U.S. presidential re-election bids, researchers found “no significant relationships between the incumbent’s vote margin and inflation or unemployment.” GDP does predict the incumbent’s vote in simple regression, but it is “rendered insignificant when combined with the stock market in multiple regression.”

Across every test, the stock market’s prior net change (β = 0.573 over three years) outperformed all of the “big three” economic variables. As the authors put it, “none of the alternative measures we test are as powerful as the stock market in predicting U.S. presidential re-election outcomes.”

Source: Prechter, R. R., Jr., Goel, D., Parker, W. D., & Lampert, M. (2012). “Social Mood, Stock Market Performance and U.S. Presidential Elections: A Socionomic Perspective on Voting Results.” SAGE Open. Read the complete study on SSRN →

The Economic Variables Political Science Relies On

For decades, election forecasters have leaned on economic data to predict how incumbents will fare. The paper calls these the standard tools of “economic voting.”

The “big three” popularity functions traditionally mentioned in the literature are economic growth, inflation and unemployment.

— Prechter et al. (2012), p. 5

The researchers set out to test whether any of these variables predicts re-election outcomes better than the stock market. Their framing question:

Is the stock market a unique predictor of presidential re-election outcomes, or do any of the “big three” economic variables yield stronger results?

— Prechter et al. (2012), p. 27

Economic Voting
The hypothesis, dominant in political science, that changes in economic variables such as growth, inflation, and unemployment cause changes in public mood and voting results, so voters reward or punish incumbents for economic conditions. Socionomic theory reverses this causal arrow, treating economic data as a lagging result of social mood rather than the driver of the vote.

— Derived from Prechter et al. (2012), pp. 4–5, as cited in Prechter et al. (2012)

Gross Domestic Product: Weaker Than the Market — and Lagging It

GDP is the strongest of the three economic contenders, yet it still trails the stock market on every measure the authors tested.

Substituting the three-year percentage change in nominal GDP for the DJIA to predict the incumbent’s popular vote margin produces weaker relationships across the board, including a reduced linear regression beta weight (0.33 as opposed to 0.57 for the DJIA) and a smaller Spearman’s rank correlation (0.27 vs. 0.59). Similar results occur when substituting real GDP as the predictor (β = 0.47 vs 0.57, ρ = 0.46 vs. 0.59).

— Prechter et al. (2012), p. 28

When the outcome is simply whether the incumbent won or lost, GDP fails entirely: “When predicting the dichotomous criterion of election win/loss, neither real GDP nor nominal GDP emerges as a significant predictor in logistic regression.”

The authors also tested whether GDP might be the hidden force driving both the market and the vote. The data point the other way — the market leads GDP, not the reverse:

When lognormal yearly changes in the DJIA are used to predict the subsequent year’s lognormal movement in GDP, we observe a significant beta weight of 0.25 (p < 0.0002), versus no relationship at all (β = -0.02, p = 0.81) when we use lognormal yearly changes in GDP to predict the subsequent year’s lognormal movement in the DJIA. Thus, GDP does not predict stock market movement, but—in accordance with socionomic theory—the stock market predicts GDP.

— Prechter et al. (2012), p. 27

Inflation and Unemployment: No Significant Relationship

Inflation and unemployment fare worse still. Neither reaches statistical significance as a predictor of the incumbent’s vote margin.

Using the three-year percentage change in the PPI rather than the DJIA to predict an incumbent’s popular vote margin results in a sizably worse beta weight (0.16 vs 0.57 for the DJIA) and Spearman’s rho (0.15 vs. 0.59), neither of which is statistically significant. The unemployment rate, with data available from 1940 to the present, fails to register as a significant predictor in the same simple regression analysis. It yields a beta weight of -0.11 and Spearman’s rho of -0.18. Neither the PPI nor the unemployment rate is able to predict better than at chance levels with regard to overall election win/loss.

— Prechter et al. (2012), p. 28

The authors’ summary judgment is blunt:

Changes in unemployment and inflation rates (as measured by the PPI) in the one-, two-, three-, and four-year periods prior to elections have no discernible relationship to presidential re-election outcomes.

— Prechter et al. (2012), p. 29

Combine Them All — the Stock Market Still Wins

Perhaps the economic variables work together even if none works alone? The authors ran multiple regressions on every combination. The result was the same.

In a large number of analyses, the DJIA remains the only significant predictor of election outcomes when combined with nominal GDP, real GDP and/or the inflation rate. With popular vote margin as the criterion, none of eleven combinations of various independent variables registers as a significant predictor; the DJIA remains a significant predictor in all such combinations.

— Prechter et al. (2012), p. 29

To rule out the possibility that these results were an artifact of running many tests, the authors applied an omnibus Simes’ test for multiple comparisons. The verdict held: “none of the alternative measures we test are as powerful as the stock market in predicting U.S. presidential re-election outcomes” (p. 30).

Predictor Variable	β (3-year, pop. vote margin)	Significant?
Stock Market (DJIA)	0.573	Yes (p = 0.001)
Real GDP	0.47	Weak; insignificant alongside the DJIA in multiple regression
Nominal GDP	0.33	Weak; insignificant alongside the DJIA in multiple regression
Inflation (PPI)	0.16	No
Unemployment	−0.11	No

Why the Economy Lags: The Socionomic Explanation

Why would a stock index beat hard economic data at forecasting votes? Socionomic theory’s answer is that both the market and the economy are downstream of the same hidden variable — social mood — but the market registers mood changes almost instantly, while the economy takes months to catch up.

Investors… can buy or sell in the stock market almost immediately in response to social mood, so its effects appear there prior to appearing in macroeconomic indicators.

— Prechter et al. (2012), p. 11

Under this view, GDP, inflation, and unemployment are not causes of the vote at all. They are delayed, partial reflections of the same mood that the stock market captures first and more cleanly. That is why substituting economic data for the market consistently weakens the prediction.

---