This paper incorporates endogenous evolution of illiquid human wealth into the standard framework of optimal portfolio selection. The presence of human capital investment (rather than an exogenous stream of labor income) helps explain some patterns of the life-cycle asset allocation observed in the data, as well as generate new predictions.
Despite the conventional wisdom of financial planners as well as the normative prescriptions of portfolio theory, which suggest that the share of stocks in households’ portfolio should decrease with age, young people hold very little stocks. This observation is the driving force behind the approach of Constantinides, Donaldson, and Mehra (2002) to addressing the equity premium puzzle. Moreover, conditional on participation, researchers have found that the share of equities in investors’ portfolios display increasing or hump-shaped profiles over the life-cycle (e.g. Poterba and Samwick (1995) and Ameriks and Zeldes (2001)). Again, this is in stark contrast with the standard life-cycle portfolio advice that young households should allocate most of their portfolio to equities, and reduce this allocation as the present value of their future labor income decreases with age (eg. Jagannathan and Kocherlakota (1996)).
Introducing endogenous human capital accumulation allows me to refine the predictions of portfolio theory about the effect of labor income on asset allocation in the different stages of the life cycle. In particular, the presence of an option to invest in human capital alters the optimal allocation to risky assets for a given life-time present value of human wealth. Moreover, it allows for endogenous heterogeneity in life-cycle labor income profiles, as well as in portfolio holdings.
There are several key features of human capital investment that yield interesting implications for asset allocation. First, educational investment is indivisible (one cannot get one quarter of an MBA). Second, besides the direct financial cost (e.g. tuition), education has large opportunity costs, since it is hard to both work and study at the same time (evening and part-time programs form an important exception). Most importantly for the life cycle considerations, given the finite life span, the value of the option to invest in human capital declines over time, since there are fewer years to collect the rewards. Finally, the crucial difference between human capital and most other types of investment is liquidity. Given that future labor income is non-tradeable for reasons of moral hazard and adverse selection, it may be hard to borrow in order to invest in education (although various financial aid programs exist in order to help individuals overcome this constraint). This restriction is even more severe for other forms of human capital investment (such as health).
The predictions of my model with respect to the stock holdings of the young investors are two-fold. I find that for the wealthier young investors the presence of the option to invest in human capital depresses risky asset holdings, which thus peak in the middle of the life-cycle, after the option has been exercised. On the contrary, for the poorer young investors the presence of such an option can increase their risky asset holdings (generating the usual pattern of declining equity shares over the life-cycle), thus potentially deepening the puzzle of low stock holding among the young.
Download
PDF Ebook Human capital investment and portfolio choice over the life-cycle
