Interest Rates & Equity Returns
With interest rates starting to rise from all-time lows, investors have expressed increasing concern about how rising rates might affect equity investments. Equity markets had a wonderful run during the thirty-year period of declining rates that started in the early 1980s, and many now worry that if falling rates were good for markets, rising rates will be bad.
The impact of rates on equities is a question that can be evaluated empirically with data. And we have looked at the long-term history of US markets to answer two simple questions: do changes in interest rates affect equity returns? And do rising interest rates reduce the returns of a strategy that targets leveraged equities? Based on 52 years of empirical evidence, the answer is “not historically.” Across multiple interest rate cycles in the United States between April 1965 and December 2017, there has been no meaningful relationship between changes in interest rates and returns on the broad equity market, small value stocks, or leveraged small value stocks.
As a starting point, here is a table that shows annualized equity returns for the S&P 500, Fama-French Small Value, and Leveraged Small Value during ten major periods of rising 10-Year Treasury Yields since 1965.
Figure 1: US Equity Returns in Rising Rate Environments
Sources: CapitalIQ, FRED (Federal Reserve Bank of St. Louis)
In 50% of these rising rate environments, the S&P 500 outperformed the consumer price index (CPI). In 70% of these rising rate environments, small value outperformed the S&P 500. And in 70% of these environments, leveraged small value outperformed small value. Though there has been significant sturm und drang about the impact of rising rates on equities, the data does not support a particularly bearish outlook. But let’s dig in further.
Theoretical Foundation
Stock prices reflect aggregate expectations of companies’ future cashflows and a discount rate that investors apply to those expected cashflows.
When interest rates rise, investors may apply a higher discount rate to leveraged companies to reflect a higher cost of debt capital. This denominator effect would tend to reduce the prices of leveraged equities. On the other hand, interest rates generally rise in good economic times that are characterized by higher wages and increased consumer spending. Higher consumer spending increases demand for the goods and services provided by companies, suggesting that investors would also increase their expectations of future cashflows. This numerator effect would tend to increase equity prices, potentially canceling out the downward price effect of a higher discount rate. The effects of higher expected cashflows and a higher discount rate counteract each other, and it is unclear which effect would dominate. Therefore, theory does not provide a prediction of how changes in interest rates would affect equity returns.
Empirical Evidence
We now look at the empirical evidence to determine whether there has been any relationship between changes in interest rates and equity returns. Over the past 52 years, there have been multiple interest rate cycles in the United States. During this time, the aggregate stock market has compounded at 10% per year, small value stocks have compounded at 15% per year, and leveraged small value stocks have compounded at 19% per year, as summarized in Figure 2.
Figure 2: Summary Statistics: April 1, 1965 to December 31, 2017
Sources: Ken French Data Library, The Center for Research in Security Prices (CRSP), and Verdad Research
Figure 3 presents scatter plots of the monthly returns of the US Market Index, US Small Value Index, and the US Leveraged Small Value Strategy against contemporaneous monthly changes in the 10-Year Treasury Yield between April 1965 and December 2017. In each case, there does not appear to be a relationship between changes in interest rates and equity returns. Regressions of these monthly equity returns against contemporaneous changes in the 10-Year Treasury Yield have an R-squared ranging from 0.23% to 1.68%. In other words, more than 98% of the variation in equity returns is unexplained by changes in interest rates.
Figure 3: Monthly Equity Returns vs. Changes in 10-Year Treasury Yields: Apr 1965 – Dec 2017
Sources: Federal Reserve Bank of St. Louis, Ken French Data Library, CRSP, and Verdad Research
What if we look over longer horizons? Would changes in interest rates over longer time periods have a closer relationship with equity returns? Figure 4 addresses this question by sorting equity returns in every rolling decade between April 1965 and December 2017 into deciles according the change in 10-Year Treasury Yield during each rolling decade. For example, the tenth decile represents rolling decades where 10-Year Treasury Yields increased the most from the beginning of month 1 to the end of month 120. Specifically, 10-Year Treasury Yields increased by 4.0% to 9.4% in the decades that are included in the tenth decile. Conversely, the first decile represents rolling decades where 10-Year Treasury Yields declined the most (-4.2% to -7.5%) from the beginning of month 1 to the end of month 120.
Based on the evidence in Figure 4, average 10-year equity returns do not appear to be related to changes in the 10-Year Treasury Yield within the same decade. We do not see any pattern of equity returns systematically increasing or decreasing by decile of interest rate change. Therefore, even if someone had perfect foresight into the level of interest rates ten years into the future, that information would not be helpful for determining whether to expect higher or lower equity returns over the next ten years.
Figure 4: Average Rolling 10-Year Returns by Decile of 10-Year Treasury Yield Change (1965–2017)
Sources: Federal Reserve Bank of St. Louis, Ken French Data Library, CRSP, and Verdad Research
It appears there are only two clear patterns in Figure 4. First, the Leveraged Small Value Strategy and the Small Value Index seem to have higher average returns than the Market Index in every decile of ten-year interest rate change. Second, the Leveraged Small Value Strategy appears to outperform the Small Value Index, with higher average returns in nine out of the ten deciles.
As we saw earlier in Figure 2, the Leveraged Small Value Strategy outperformed the Small Value Index by 3.8 percentage points on an annualized basis between April 1965 and December 2017. But the annualized volatility around that “leverage premium” was 10.9% per year. Therefore, we would expect to see some decades where the Leveraged Small Value Strategy underperforms the Small Value Index simply by chance.
Interest Rates and the Leverage Premium
What if we consider the relationship between changes in interest rates and the leverage premium? If changes in interest rates have any effect on equity returns, one would imagine that this effect would show up in the premium for holding leveraged equities versus an unlevered benchmark.
We can define the leverage premium as the difference between monthly Leveraged Small Value Strategy returns and monthly Small Value Index returns. This leverage premium was 3.8% per year between April 1965 and December 2017. Figure 5 presents scatterplots of the monthly leverage premium versus contemporaneous monthly changes in interest rates. A variety of interest rates are used (Federal Funds Rate, 1-Year, 5-Year, and 10-Year Treasury Yields) in order to check for robustness. In all cases, there appears to be no relationship between changes in interest rates and the leverage premium.
Figure 5: Monthly Leverage Premium vs. Changes in Interest Rates (1965–2017)
Sources: Federal Reserve Bank of St. Louis, Ken French Data Library, CRSP, and Verdad Research
Figure 6 quantifies these scatterplots by regressing the monthly leverage premium against contemporaneous monthly changes in interest rates between April 1965 and December 2017. Regressions 1, 2, 3, and 4 each have an R-squared below 0.2%. This means more than 99.8% of the leverage premium’s variation is unexplained by changes in interest rates. Moreover, the effect of changes in interest rates on the leverage premium is indistinguishable from zero in all cases.
Figure 6: Regressions of Monthly Leverage Premium vs. Changes in Interest Rates (1965–2017)
Sources: Federal Reserve Bank of St. Louis, Ken French Data Library, CRSP, and Verdad Research
Based on Figure 6, we find no evidence that changes in interest rates affect the leverage premium. The intercept in each regression is reliably different from zero, and it ranges from 0.29% to 0.30% per month, suggesting an annualized alpha of ~3.6% per year, which is similar to the 3.8% leverage premium that was observed between April 1965 and December 2017.
Conclusion
Financial theory does not provide a prediction of how changes in interest rates would affect equity returns. Based on 52 years of data, we do not find evidence of a meaningful relationship between changes in interest rates and equity returns. The direction of interest rates is very difficult—if not impossible—to predict. However, our results indicate that even perfect foresight into future changes in interest rates would not be helpful for predicting equity returns or the leverage premium. We believe that the best way for investors to capture the premiums offered by the equity market is to remain invested over the long haul and tune out the noise from interest rate prognosticators.