722THE AMERICAn ECOnOMIC REVIEWJunE 2008
t and the composite residual et 1 vt Measurement error renders OLS inconsistent because Dk w
t , Dkut2 . 0, which biases the coefficient estimate for r 2 rDkut are correlated. First, cov 1Dk w
toward zero. Second, because the numerator (stocks) and denominator (stocks plus riskless assets) t , vt2 Z 0, which could of at are made up by components of wt , we should also expect that cov 1Dk w
also bias the sign of the coefficient. Specifically, measurement error only in stock holdings would t , vt2 . 0; measurement error only in riskless assets would lead to cov 1Dk w t , vt2 lead to cov 1Dk w
, 0. It is not possible to unambiguously sign the combined effect when both stock holdings and riskless assets are mismeasured. Overall, the bias in the estimate of r could go in either direction, depending on whether measurement error in stocks or riskless assets dominates.
To address the measurement error problem, we look for instrumental variables for Dkwt . The identification requirement is that the instruments, zt , are (partially) correlated with Dkwt , but not with et 1 vt 2 rDkut . Given such instruments, we can estimate r consistently with two-stage least squares (TSLS). Our instruments are quantile dummies for income growth from t 2 k to t (similar to Dynan 2000, in a different application), and inheritance receipts (as in Jonathan Meer, Douglas L. Miller, and Harvey S. Rosen 2003) between t 2 k and t. These instruments are based upon survey questions that are different from those for the components of wt . Hence, it is reason-able to assume that the elements of zt are uncorrelated with et 1 vt 2 rDkut .
Unfortunately, it is to be expected that we lose precision compared with the OLS estimator, and so there is a trade-off between potential measurement-error bias and precision. A priori, it is not clear that the TSLS estimator will be closer to the true parameter in a mean-squared error sense. Therefore, we report both the OLS and TSLS results in our tests.
D. Data
We use data from the PSID, obtained from the University of Michigan. It is a longitudi-nal study that tracks family units and their offspring over time. We use data on asset holdings collected in the years 1984, 1989, 1994, 1999, 2001, and 2003. Income data and many house-hold characteristics are available annually until 1997, and every second year from then onward. Appendix A.4 of Brunnermeier and Nagel (2006) describes our data in more detail. Here, we briefly discuss the definition of the variables that we extract from the files. To make magnitudes comparable over time, we deflate all income and wealth data by the consumer price index (CPI) into December 2001 dollars.
Variable Definitions.—We define liquid assets as the sum of holdings of stocks and mutual funds plus riskless assets, where we follow common practice and define riskless assets as the sum of cash-like assets and holdings of bonds. Subtracting other debts, which comprise non-mortgage debt such as credit card debt and consumer loans, from liquid assets yields liquid wealth. We further denote the sum of liquid wealth, equity in a private business, and home equity as financial wealth.
We then calculate two risky asset shares: first, the sum of stocks and mutual funds held, divided by liquid assets (the liquid risky asset share);6 second, the sum of stocks and mutual funds, home equity, and equity in a private business, divided by financial wealth (the financial risky asset share).
Since we don’t have more detailed information on the composition of risky asset holdings, we don’t know, for example, whether households hold stocks with low or high systematic risk or how well they are diversified. However, our focus is on whether households reallocate between stocks and riskless assets in response to wealth changes, not on whether they reallocate between stocks of different systematic risk (which the theory is silent about).6
s Varying Risk Aversion on the Dynamics of Asset Price Fluctuations Baosheng...At each time step t, every agent decides how to allocate his wealth ...
s Varying Risk Aversion on the Dynamics of Asset Price Fluctuations Baosheng...At each time step t, every agent decides how to allocate his wealth ...
热门文档