Abstract:

We study the lifecycle allocation of financial wealth for an individual investor facing a general income process, access to multiple asset classes, and short selling constraints. We develop analytical techniques that approximate the optimal portfolio choice at any individual history using sufficient statistics that can be directly estimated from micro-level labor income data and macro-level asset return data. A key advantage of our approach is that it avoids the curse of dimensionality that typically arises in incomplete market models with many assets. Leveraging standard tools from labor economics and finance, we construct these sufficient statistics for widely available financial assets. Our quantitative findings indicate that a representative U.S. investor should allocate heavily toward value stocks (i.e., high book-to-market equities) early in life. As retirement approaches, the portfolio gradually shifts toward a combination of long-duration domestic and international bonds, a modest position in a broad stock market index (S&P 500) and, potentially, gold. In contrast, the optimal portfolio includes no exposure to housing ETFs (REITs) or indices tracking international equities.