The Impact of Individual Investment Behavior for Retirement Welfare: Evidence from the United States and Germany

Thus, for lower levels of net worth, the gap between optimal and empirical asset
allocation becomes larger, and the welfare losses are also often larger.

Varying the gender variable from male to female increases the individual’s life
expectancy, and increases the empirical risky asset share for U.S. investors (see
Table 3).²² The variation in life expectancy yielded ambiguous results in the
normative model and produced only minor differences in asset allocation. Thus, the
optimal asset allocation for men and women according to the benchmark model is
almost identical. But for women, the gap between optimal and empirical asset
allocation (risky asset share) is smaller due to the positive coefficient for gender.
Consequently, the welfare losses are smaller for women.

The impact of different assumptions for the coefficient of relative risk aversion γ is
shown in Figure 4.

Figure 4: Welfare Losses, ∆W0 / W0, for U.S. (SCF)   Data;

Gender = 0 (male), γ = 1, 2, or 3, δ = 0.97, Age = 50, Education = Middle

--- put Figure 4 here ---

Higher risk aversion is associated with lower losses in welfare.²³ The reason for this
is that while increasing risk aversion, the empirically measured asset allocation stays
constant, but the benchmark investment in the risky asset decreases. As a result, the
gap between the optimal risky asset share and the empirical asset share decreases.

The influence of a variation in the subjective discount factor δ is shown in Figure 5.

²² The growth rates for income are assumed to be identical in our U.S. income profiles.

²³ As can be seen from Figure 4, the impact of changes in income and net worth are
confirmed. The gender effect is also confirmed. Because this is also valid for the
following variations (except age), we will no longer refer to these effects (except for age).

25

<<   23   24   25   26   27   28   29   >>

More intriguing information