family planning and the negative likely impact of the HIV/AIDS epidemic on national savings rates. Although labor-force growth would be slowed in both scenarios, the third row of the table shows that the family planning slowdown would enhance labor productivity, while the HIV/AIDS epidemic will have six distinct negative effects on labor productivity, both now and in the future. By increasing employers' costs of labor relative to capital, the epidemic will bias employers toward labor-saving and away from labor-using technologies.

The fourth row of the table points out that family planning and HIV infection could both change the mix of workers in the same direction—toward a less-skilled work force. In the case of the family planning program, unless a national policy of targeting the program to the poorest households is extraordinarily effective, higher-income households will demand more family planning services and make more use of the program than will the poor. By slowing the growth of the educated labor force more than that of the uneducated, the program will have the unintended consequence of decreasing the proportion of the work force with more education.5

For HIV, the argument rests first on the ubiquitous observation that infection rates are higher in urban areas, where average levels of education are the highest, and second on the hypothetical link between male socioeconomic status and casual sexual activity. Evidence from recent surveys of sexual behavior supports the hypothesis that men of higher status (specifically, men with higher educational attainment) have more casual sex partners per year (Caraël et al., 1994). The finding holds for men throughout the world, and for eight of the nine surveyed African countries (Caraël et al., 1994). Thus it is not surprising that where data on male HIV prevalence by socioeconomic status exist, they often show higher prevalence at higher income or education levels (see Ainsworth and Over, 1994b, for a recent review of the evidence on the current rate of HIV infection in sub-Saharan Africa by economic group). Since African populations are known to have improved greatly their understanding of the causes of AIDS since the risk behavior that resulted in the higher infection rates at higher social levels occurred, it is possible that this relationship has since been reversed. However, no data have yet been collected to demonstrate that more highly educated African men have responded to their improved knowledge of the causes of AIDS by (1) having fewer casual sex partners than less-educated men, (2) having safer sex more often than less-educated men, or (3) becoming infected at a lower rate than less-educated men. Moreover, even if such changes have occurred, the pattern of the epidemic for the next decade or two has been set by past risk behaviors and prevalence patterns. Those patterns make it likely that HIV will kill relatively


 Of course, if the higher-income families reduce their children by a larger percentage than the poor, they will also save more educational expenses and will therefore have more resources available to be rechanneled into the education of the children they do have. This effect may partly offset the change in mix described in the text.

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement