Theses Doctoral

Macroeconomic Volatility and Asset Prices

Ermolov, Andrey

This dissertation investigates, both theoretically and empirically, how does the macroeconomic volatility, in particular, consumption growth, GDP growth and inflation volatility, affect asset prices in equity, bond and currency markets. In all three chapters of the dissertation I use the Bad Environment-Good Environment structure of Bekaert and Engstrom (2014) to model macroeconomic volatility. The key advantage of the approach is that it allows to model non-Gaussian features important in macroeconomic dynamics while yielding closed-form asset pricing solutions and being relatively efficient to estimate.
In the first chapter of the dissertation I show that an external habit model augmented with a heteroskedastic consumption growth process reproduces well known domestic and international bond market puzzles, considered difficult to replicate simultaneously. Domestically, the model generates an upward sloping real yield curve and realistic violations of the expectation hypothesis. Depending on the parameters, the model can also generate a downward sloping real yield curve and predicts that the expectation hypothesis violations are stronger in countries with upward sloping real yield curves. Internationally, the model explains violations of the uncovered interest rate parity. Unlike a standard habit model, the model simultaneously features intertemporal smoothing to match domestic real yield curve slope and bond return predictability and precautionary savings to reproduce international predictability. The model also replicates the imperfect correlation between consumption and bond prices/exchange rates through positive and negative consumption shocks affecting habit differently. Empirical support for the model mechanisms is provided.
In the second chapter, coauthored with my advisor Geert Bekaert and Eric Engstrom of Board of Governors of the Federal Reserve System, we extract aggregate supply and demand shocks for the US economy from data on inflation and real GDP growth. Imposing minimal theoretical restrictions, we obtain identification through exploiting non-Gaussian features in the data. The risks associated with these shocks together with expected inflation and expected economic activity are the key factors in a tractable no-arbitrage term structure model. Despite non-Gaussian dynamics in the fundamentals, we obtain closed-form solutions for yields as functions of the state variables. The time variation in the covariance between inflation and economic activity, coupled with their non-Gaussian dynamics leads to rich patterns in inflation risk premiums and the term structure. The macro variables account for over 70\% of the variation in the levels of yields, with the bulk attributed to expected GDP growth and inflation. In contrast, macro risks predominantly account for the predictive power of the macro variables for excess holding period returns.
In the final chapter, I embed the macroeconomic dynamics from the second chapter into an external habit model to analyze the time-varying stock and bond return correlations. Despite featuring flexible non-Gaussian fundamental processes, the model can be solved in closed-form. The estimation identifies time-varying "demand-like" and "supply-like" macroeconomic shocks directly linked to the risk of nominal assets and matches standard properties of US stock and bond returns. I find that macroeconomic shocks generate sizeable positive and negative correlations, although negative correlations occur less frequently and are smaller than in data. Historically, macroeconomic shocks are most important in explaining high correlations from the late 70's until the early 90's and low correlations pre- and during the Great Recession.



  • thumnail for Ermolov_columbia_0054D_12756.pdf Ermolov_columbia_0054D_12756.pdf application/pdf 1.42 MB Download File

More About This Work

Academic Units
Thesis Advisors
Bekaert, Geert
Ph.D., Columbia University
Published Here
May 12, 2015