Dissertation

Figure 3.12: Efficiency profiles from the 2001 CPS and Hansen (1993).

where s is model age and s ≤ 40. Similar to the baseline calibration, beyond actual age 65

(i.e. for s > 40), I use the following quadratic function

ln e(s) = −f₀ − f₁s − 0.01s² (3.67)

Student Version of MATLAB

Efficiency
and parameterize f₀ and f₁ such that e(s) is continuous and once differentiable at s = 40. This yields f₀ = 14.7984 and f₁ = −0.7685. The age-dependent efficiency profile from

Hansen (1993) is compared to the initial profile computed from the 2001 CPS in Figure

3.12.

Following the same approach used in sensitivity analyses with respect to the capital’s

share in total income, I first compute a new baseline equilibrium with the new age-dependent

efficiency profile, and then I re-simulate demographic experiments 1, 2 and 3. The unknown

preference parameters and the associated model generated target values for the endogenous

variables for the efficiency profile estimated from Hansen (1993) are compared to the initial

72