Working Paper: NBER ID: w2392
Authors: Mervyn A. King; Jonathan I. Leape
Abstract: Empirical tests of the life cycle model have focused on its implications for the level of a household's total net worth and paid little attention to changes in portfolio composition over the life cycle. In this paper, we examine a new survey of the asset holdings of 6,010 U.S households and show that there is a pronounced life-cycle pattern to both the number and value of assets held by U.S. households. Direct survey evidence suggests that incomplete information is a significant determinant of household portfolio composition. We test the hypothesis that information about investment opportunities arrives stochastically over time, estimating a Poisson model for the arrival of new information.
Keywords: Asset Accumulation; Life Cycle; Household Portfolio; Incomplete Information
JEL Codes: D91; G11
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Incomplete information about investment opportunities (D82) | Lack of diversification in household portfolios (D14) |
| Age (J14) | Accumulation of information about investment opportunities (G11) |
| Accumulation of information about investment opportunities (G11) | Increased likelihood of owning a greater number of assets (D14) |
| Age (J14) | Increased likelihood of owning a greater number of assets (D14) |
| Wealth + Age (E21) | Increased likelihood of owning a greater number of assets (D14) |
| Transaction costs + Age-related wealth profiles (D15) | Patterns in asset ownership (G51) |