HUMAN CAPITAL: DETERIORATION AND NET INVESTMENT

The current practice in national accounting is to exclude from national product investment in schooling and on-the-job training, except for direct costs of schooling which are included in consumption. Foregone earnings, which form the major part of investment in human capital, go unrecorded.
Much is to be gained in consistency and analytical clarity by treating human capital like physical capital in national accounting. Estimating the amount of foregone earnings net of deterioration, that is, net investment, is a step in this direction.
Using the framework of the life-cycle hypothesis of earnings, and assuming declining-balance deterioration of human capital, estimates of deterioration rates in respect of American males by race and education level are computed for 1960. Every such d is, however, the minimum consistent with the respective costs and benefits profile. Hence an upper limit d is assumed. The model generates for each costs and benefits profile and in respect of either d , a year-by-year series of net investment in human capital. These are used to obtain two estimates (which turn out to be close to each other) of aggregate net investment in American white males in 1960. On the basis of these estimates, aggregate net investment in human capital is found to be about equal to net investment in physical assets (including consumers'durables).
It is also found that the Denison method of estimating the contribution of the increase in human capital to economic growth understates this contribution by a ratio approximately equal to net investment in on-the-job training to returns to human capital. This was about 16 percent in 1960.