Does the level of, or change in, the annual U.S. federal deficit systematically influence the U.S. stock market, perhaps by stimulating consumption and thereby lifting corporate earnings (bullish) or by igniting inflation and thereby elevating discount rates (bearish)? To check, we relate annual stock market returns to the annual surplus/deficit (receipts minus outlays) as a percentage of Gross Domestic Product (GDP). We align stock market returns with deficit calculations (federal fiscal years, FY) as follows: (1) prior to 1977, we calculate annual returns from July through June; (2) we ignore the July 1976 through September 1976 transition quarter; and, (3) since 1977, we calculate annual returns from October through September. Using deficit data, augmented by actual GDP data for FY20, and returns for the S&P 500 Index (SP500) as a proxy for the U.S. stock market during FY 1930 through FY 2020 (90 years), *we find that:*

The following chart tracks federal FY deficits as a percentage of GDP and FY SP500 returns over the sample period. The chart truncates the extremely large +146% SP500 return for FY 1933. There are few surpluses. The pronounced deficits around World War II stand out as very unusual. Overall, results suggest there may be a negative relationship between the two series.

For insight, we examine quantitative relationships between the two series.

The following chart relates SP500 next-FY return to FY federal surplus/deficit as a percentage of GDP over the available sample period. The Pearson correlation between the two series is -0.16, and the R-squared statistic is 0.03, indicating that variation in the deficit as a percentage of GDP explains 3% of variation in next-FY stock market return. In other words, increasing the deficit one year may slightly stimulate the stock market next year.

The outliers to the left side of the distribution come from World War II (FY 1942-1945). Excluding 1942-1945 does not change the sign of the correlation or the value of R-squared. The outlier at the top of the chart is FY 1933 noted above. Separately excluding 1933 does not change the sign of the correlation and changes R-squared to 0.04.

For a different perspective, we look at ranked fifths (quintiles) of FY deficits.

The next chart summarizes average SP500 next-FY returns by quintile of FY deficits as a percentage of GDP over the available sample period. In case the trend in deficit rather than its level is critical, the chart also summarizes average SP500 returns next FY by quintile of change in annual deficits as a percentage of GDP. Results suggest that large government deficits and deepening deficits stimulate the stock market the following year.

However, based on the variabilities in the two series, quintile subsamples are very small (18-19 observations each). For example, excluding the SP500 FY 1933 return of +146% lowers average SP500 return from 20.7% to 13.3% for the most negative deficit quintile.

Might deficits reliably lead stock returns over horizons longer than a year?

The final chart summarizes correlations between SP500 FY return and FY deficit (level of and change in) as a percentage of GDP for various lead-lag relationships over the available sample period, ranging from stock market returns lead deficits by four years (-4) to deficits lead stock market returns by four years (4). Results suggest that:

- A relatively strong (weak) stock market over the past few years tends to drive the government toward surpluses (deficits), perhaps due to associated variations in tax receipts.
- As noted above, a relatively high/increasing deficit tends to be good for the stock market the following year.
- There is probably no relationship between the federal deficit and stock market returns more than one year in the future.

In summary, *evidence from simple tests offers a little support for belief that very large/increasing federal deficits are good for stock returns the next fiscal year.*

Cautions regarding findings include:

- Tests are in-sample, with sample size too small for meaningful out-of-sample testing.
- As noted, based on the large variabilities in stock market returns and deficits, sample and subsample sizes are small.
- Distributions may not be tame (normal-like). To the extent they are wild, interpretations of simple statistics break down (as suggested by the large effect of 1933 in the quintile breakdown above).
- The connection between U.S. stock market return and federal deficit may be stronger for large firms (as in the SP500) than small firms.