Causal Relations Among Stock Returns, Interest Rates, Real Activity, and Inflation

Abstract

Using a multivariate vector-autoregression (VAR) approach, this paper investigates causal relations and dynamic interactions among asset returns, real activity, and inflation in the postwar United States. Major findings are (1) stock returns appear Granger-causally prior and help explain real activity, (2) with interest rates in the VAR, stock returns explain little variation in inflation, although interest rates explain a substantial fraction of the variation in inflation, and (3) inflation explains little variation in real activity. These findings seem more compatible with Fama (1981) than with Geske and Roll (1983) or with Ram and Spencer (1983). VARIOUS THEORETICAL FRAMEWORKS HAVE been developed to explain the negative correlations observed between inflation and real asset returns inthe postwar period. In light of the Fisher hypothesis, and of the commonly held view that stocks and bonds should be a hedge against inflation, these relations are indeed puzzling. Fama (1981), Geske and Roll (1983), Ram and Spencer (1983), James, Koreisha, and Partch (1985), Stulz (1986), and Kaul (1987) all attempt to explain the negative association between stock returns and inflation; and Fama and Gibbons (1982) attempt to explain the negative association between inflation and real interest rates. Fama (1981) hypothesizes that the negative correlation between stock returns and inflation is not a causal relation but that it is proxying for a positive relation between stock returns and real activity and is induced by a negative relation between real activity and inflation. Geske and Roll (1983) argue that stock returns cause (or signal) changes in inflationary expectations because of a chain of macroeconomic events. When stock prices declinein response to anticipated changes in economic conditions, the government, given largely fixed expenditures, will tend to run a deficit. To the extent that the deficit is monetized, expected inflation will rise. Ram and Spencer (1983), however, find evidence of unidirectional causality from inflation to stock returns. Using a vector autoregressive-moving average (VARMA) model, James, Koreisha, and Partch (1985) examine simultaneously the causal links between stock returns, real activity, money supply, and inflation. They find evidence that stock returns signal both changes in real activity and changes in the monetary base, which suggests a link between money supply and real * Department of Finance, Carlson School of Management, University of Minnesota. I would like to thank an anonymous referee and editor for their many helpful comments and advice. 1592The Journal of Finance activity that is consistent with the money supply explanation offered by Geske and Roll.All of these studies recognize the importance of causal relations and discuss them in one way or another. However, two potential drawbacks remain. First, as Mehra (1978) and Sims (1980b) point out, the causal relations observed based on a bivariate causal test may not be robust when other relevant variables are introduced into the vect...

Stock Returns, Term Structure, Inflation, and Real Activity: An International Perspective

“…Also, no relationship between nominal stock returns and current or future industrial production growth emerges. This is not necessarily in contrast with the evidence of Fama (1981) or Lee (1992) as we consider nominal (as opposed to real) stock r e t urns. Conversely, there is a comparatively stronger link between the slope of the nominal term structure and in ation in all countries (in agreement with Fama (1990b) and with some of the ndings of Mishkin (1990)).…”

“…The explanation of the sign of this correlation proposed by Fama (1981) and Gerske and Roll (1983), based on the anticipatory movements of real stock r eturns for real activity and on a negative link between in ation and real activity, has been empirically examined f or the US by various authors with mixed results (see e.g. James, Koreisha and Park (1985), Fama (1990b), Lee (1992)). …”

“…These problems are reduced, if not solved, by directly taking a measure of the s l ope of the term structure as basic variable of the system. We choose to include nominal returns and an in ation rate instead of converting nominal returns into real ones using proxies for ex-post in ation or expected in ation prior to their use in the VAR (as in Lee (1992)), because results are more transparent in this w ay since the time path of real returns can be easily computed from estimated VAR coe cients. As Lee (1992) and others have stressed, the rate of in ation is very highly correlated with the growth rate of the monetary base (on average for the countries under consideration the contemporaneous correlation i s about 0.7): by including them directly we can interpret in ation surprises as close proxies for monetary s u rprises.…”

“…Geske a nd Roll (1983); James, Koreisha and Partch ( 1 985) or Lee (1992)). Here we maintain the same structure used by these authors except that w e employ a measure of the slope of the term structure in place of a measure of s h o rt term interest rate.…”

15. Some Policy Implications

“…We include forward-looking financial variables -stock market returns, short-term interest rates and the dollar exchange rate -that are thought to embody future economic expectations. In the case of output growth, studies such as Barro (1990), Fama (1990), Lee (1992), Estrella and Mishkin (1998), Hassapis and Kalyvitis (2002), Hassapis (2003) and Panopoulou et al (2005) among others find that stock market returns improve forecasting ability. Stock market returns are not generally found to be useful in predicting future inflation, e.g.…”