Skip to main content

Advertisement

SpringerLink
Log in

HIV and fertility in Africa: first evidence from population-based surveys

  • Original Paper
  • Published:
Journal of Population Economics Aims and scope Submit manuscript

Abstract

The historical pattern of the demographic transition suggests that fertility declines follow mortality declines, followed by a rise in human capital accumulation and economic growth. The HIV/AIDS epidemic threatens to reverse this path. We utilize recent rounds of the demographic and health surveys that link an individual woman’s fertility outcomes to her HIV status based on testing. The data allow us to distinguish the effect of own positive HIV status on fertility (which may be due to lower fecundity and other physiological reasons) from the behavioral response to higher mortality risk, as measured by the local community HIV prevalence. We show that although HIV-infected women have significantly lower fertility, local community HIV prevalence has no significant effect on noninfected women’s fertility.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. Bloom and Mahal (1997) run cross-country regressions of growth of GDP per capita on HIV/AIDS prevalence and find no effect. Papageorgiou C, Stoytcheva P (2008, What Is the Impact of AIDS on Cross-Country Income So Far? Evidence from Newly Reported AIDS Cases, unpublished) find negative effect on the level of income per capita in a similar framework. Werker, E D, Ahuja A, Wendell B (2006, Male Circumcision and AIDS: The Macroeconomic Impact of a Health Crisis, unpublished) instrument HIV/AIDS prevalence by national circumcision rates and show that there is no effect of the epidemic on growth of the African countries.

  2. Kalemli-Ozcan and Turan (2011) shows that Young’s identification from time-series data may not be appropriate given the existing trends in South African data due to abolition of apartheid and the ongoing demographic transition.

  3. Among HIV-positive women, 29 % are widowed, separated, or divorced as opposed to 7 % among HIV-negative women.

  4. The community-level results were produced independently and at the same time in an earlier version of our paper, Juhn C, Kalemli-Ozcan S, Turan B (2008) HIV and Fertility in Africa: First Evidence from Population Based Surveys and a working paper version of Fortson (2009).

  5. In the follow-up paper that uses country by cohort variation, Young (2007) reports a range of coefficients from − 1.54 to − 0.60 (high–low). These coefficients translates into a reduction in fertility of approximately 154–45 % as a country goes from 0 to 100 % prevalence. As discussed in Young (2007), the size of the coefficient appears to be sensitive to the inclusion of the country-specific time trends.

  6. While not directly related to HIV/AIDS, a recent paper by Acemoglu and Johnson (2007) find no effect of life expectancy on level and growth of per capita income. They instrument changes in life expectancy with dates of global interventions in disease prevention. Their results suggest that an increase in life expectancy leads to an increase in population, and fertility responses are insufficient to compensate. It may be the case, however, that many of the countries in their sample have not yet completed the demographic transition. Ashraf et al. (2008) show that the effects of health improvements on income only emerge for half a century after the initial improvement in health.

  7. While the focus of our study is the fertility channel, an equally important question is the effect of HIV/AIDS on human capital investment. A large number of papers cover this topic and generally find substantial negative effects. Meltzer (1992) argues that AIDS raises mortality of young adults, which is going to have the biggest effect on the rate of return on educational investment. He claims that for a 30 % HIV-positive population like Botswana, there would be a 6 % reduction in the rate of return to education relative to no HIV. Bell et al. (2006), using household survey data from South Africa, argue that the long-term economic costs of AIDS could be devastating because of the cumulative weakening from generation to generation of human capital. Fortson (2011), using data similar to ours, shows that children currently growing up in Africa, including non-orphans, will complete 0.3 fewer years of schooling compared to the case of zero HIV prevalence. Akbulut and Turan (2013) show that HIV prevalence in the community impairs the intergenerational human capital transfers even if the mother is HIV-negative.

  8. While their estimates are somewhat higher than other estimates, Gray et al. (1998) is often cited as the study that comes closest to identifying the effect on fecundity. The study interviewed a representative sample of women in their homes and obtained blood samples from 91 % of the women. Most importantly, women did not know their HIV status at baseline because access to testing prior to the survey was not available in the communities surveyed. Contraception and abstinence were also very rare in these communities.

  9. In the 2000s, antiretroviral regimens to prevent mother to child transmission (MTCT) became more widely available even in resource poor countries in sub-Saharan Africa. While a full-scale analysis incorporating differences across regions and time in the availability of these drugs is beyond the scope of our paper, it is important to consider how the omission of this information may bias our results. The availability of drugs designed to reduces MTCT may encourage unprotected sex and higher fertility among HIV-infected women but it may also reduce the precautionary move towards having protected sex among noninfected women making it difficult to forecast a priori the bias in our individual-level regressions. In our community-level regressions, one possibility is that communities with higher infection rates also have more access to drugs (under the plausible scenario that health organizations concentrate their efforts in the most infected areas) and to the extent that the availability of these drugs reduces precautionary motive for protected sex; this would likely lead to a positive bias, confounding the true underlying negative effect of community-level HIV risk on individual behavior.

  10. One important limit of our data is that we observe HIV status at time t while our fertility variables refer to births last year or earlier. One of the implicit assumptions is that infection at time t is a reasonable proxy for infection in previous years.

  11. Each fertility measure has advantages and disadvantages. On the one hand, since HIV status refers to the survey year, number of births last year provides the closest match between treatment and outcome variables. On the other hand, number of births last year is more subject to idiosyncratic noise, and cumulative birth measures may be better indicators of an individual woman’s total fertility. We have also investigated the effects for older women aged 35–49 who may be close to their desired fertility levels except for the marginal child. These results are reported in Appendix Table B-1. The table shows that the results are similar for this group of women.

  12. Urban/rural residence is arguably more endogenous due to migration. In practice, however, we find that including or excluding urban/rural residence has little impact on the size of the HIV coefficient.

  13. The preponderance of zeros as well as the nonnegative and discrete nature of the dependent variable suggests a Poisson specification may be more appropriate. Our Poisson estimates yielded very similar results and are available upon request.

  14. One concern is that there is insufficient overlap in the distribution of covariates. To investigate this issue, we estimated impact of own HIV status on births using propensity score matching. We use the propensity score from the probit estimation (results reported in Appendix Table A-3) and use one-to-one nearest neighbor matching without replacement. We implemented the STATA 9 procedure developed and described in Leuven and Sianesi (2003). We also conducted simple t test on differences in characteristics between HIV-positive and HIV-negative women in our matched sample and did not find statistically significant differences. The results are reported in Appendix Table B-2. The negative impact of HIV is slightly smaller ranging between 15 and 17 %.

  15. Another concern is survivor bias. It may be the case that the HIV-positive women in Table 3 are not a representative sample of all women who ever contracted the virus and that women who contracted the disease earlier had already died. For there to be negative effect on fertility that is due to survivor bias, however, the HIV-positive women who died must have had higher fertility relative to even women who are HIV-negative. There is little to indicate that women who were early contractors of the disease would have had higher than average fertility. For example, those who are more likely to contract the disease are better educated and urban, characteristics that are associated with lower than average fertility.

  16. In results we do not report, we have run the same regression as in Table 4 but on a sample of women who reported positively to “ever had intercourse,” who reported never being tested for HIV and who lived in rural areas where regional HIV prevalence exceeded 15 %.

  17. In a previous version, we defined a community as a country by region by urban/rural residence cell. However, DHS samples are not representative at the disaggregated level. We therefore use country by region cell to define communities in this version while still controlling for urban/rural residence. We thank Jane Fortson for pointing this out to us.

  18. The numbers in Table 5 may also differ from those in Table 1 due to the fact that we report unweighted averages across communities in Table 5.

  19. The one coefficient that is negative in sign (column 3) has a 95 % confidence interval of − 0.489 and 0.069. Since the average number of births in the last 5 years is 0.722, the largest negative effect we estimate is a reduction of approximately 67 % (− 0.489/0.722) which is still smaller than the coefficient in Young (2005).

  20. To calculate the TFR for our sample of women with HIV status instead of all the women in DHS survey sample, we follow the method used by the DHS, which uses information on births over the last 36 months for each woman based on the fertility histories. The numerator of each age-specific birth rate is the total number of births over the previous 36 months for women in each 5-year age category based on age at birth. The denominator is the total number of women-years in each 5-year age category. Then we summed up all the age-specific fertility rates and multiply it by 5 (since each woman is present in each age group for 5 years) to end up with the TFR as done by DHS. To adjust TFR for differences in observable characteristics between all and negative HIV women, we run the fertility regression pooling HIV-positive and HIV-negative women as specified in Eq. 1, predict fertility by age group, and add back residuals for HIV-negative women.

References

  • Acemoglu D, Johnson S (2007) Disease and development: the effect of life expectancy on economic growth. J Polit Econ 115(6):925–985

    Article  Google Scholar 

  • Akbulut D, Turan B (2013) Left behind: intergenerational transmission of human capital in the midst of HIV/AIDS. Journal of Population Economics. doi:10.1007/s00148-012-0439-3

    Google Scholar 

  • Ashraf Q, Lester A, Weil D (2008) When does improving health raise GDP? NBER Macroecon Annu 23:157–204

    Article  Google Scholar 

  • Allen S, Serufilira A, Gruber V, Kegeles S, Van de Perre P, Carael M, Coates TJ (1993) Pregnancy and contraception use among urban Rwandan women after HIV testing and counseling. Am J Publ Health 83(5):705–710

    Article  Google Scholar 

  • Becker GS, Barro RJ (1988) Reformulation of the economic theory of fertility. Q J Econ 103(1):1–25

    Article  Google Scholar 

  • Bell C, Shantayanan D, Gersbach H (2006) The Long-run economic costs of AIDS: theory and application to South Africa. World Bank Econ Rev 20(1):55–89

    Article  Google Scholar 

  • Bloom S, Banda C, Songolo G, Mulendema S, Cunningham A, Boerma JT (2000) Looking for change in response to the AIDS epidemic: trends in AIDS knowledge and sexual behavior in Zambia, 1990 through 1998. J Acquir Immune Defic Syndr 25(1):77–85

    Article  Google Scholar 

  • Bloom D, Mahal A (1997) Does the AIDS epidemic threaten economic growth? J Econom 77(1):105–124

    Article  Google Scholar 

  • Caldwell J, Orubuloyeb IO, Caldwell P (1999) Resistances to behavioural change to reduce HIV/AIDS infection in predominantly heterosexual epidemics in third world countries. Health Transition Centre

  • Carpenter LM, Nakiyingi JS, Ruberantwari A, Malamba SS, Kamali A, Whitworth JAG (1997) Estimates of the impact of HIV infection on fertility in a rural Ugandan population cohort. Health Transit Rev 7(2):113–126

    Google Scholar 

  • Cheluget B, Baltazar G, Orege P, Ibrahim M, Marum LH, Stover J (2006) Evidence for population level declines in adult HIV prevalence in Kenya. Sex Transm Infect 82(1):121–126

    Article  Google Scholar 

  • Corrigan P, Gloom G, Mendez F (2005) AIDS crisis and growth. J Dev Econ 77(1):107–124

    Article  Google Scholar 

  • Ehrlich I, Lui FT (1991) Intergenerational trade, longevity, intrafamily transfers and economic growth. J Polit Econ 99(5):1029–1059

    Article  Google Scholar 

  • Fortson J (2011) Mortality risk and human capital investment: the impact of HIV/AIDS in sub-Saharan Africa. Rev Econ Stat 93(1):1-15

    Article  Google Scholar 

  • Fortson J (2009) HIV/AIDS and fertility. Ame Econ J: Appl Econ 1(3):170–194

    Article  Google Scholar 

  • Fylkesnes K, Musonda RM, Sichone M, Ndhlovu Z, Tembo F, Monze M (2001) Declining HIV prevalence and risk behaviors in Zambia: Evidence from surveillance and population-based surveys. AIDS 15(7):907–916

    Article  Google Scholar 

  • Galor O, Weil D (2000) Population, technology and growth: from Malthusian stagnation to the demographic transition and beyond. Am Econ Rev 90(4):806–828

    Article  Google Scholar 

  • Gray RH, Wawer MJ, Serwadda D (1998) Population-based study of fertility of women with HIV-1 infection in Uganda. Lancet 351(9096):98–103

    Article  Google Scholar 

  • Hunter SC, Isingo R, Boerma JT (2003) The association between HIV and fertility in a cohort study in rural Tanzania. J Biosoc Sci 35(1):189–199

    Article  Google Scholar 

  • Kalemli-Ozcan S (2003) A Stochastic model of mortality, fertility and human capital investment. J Dev Econ 70(1):103–118

    Article  Google Scholar 

  • Kalemli-Ozcan S, Turan B (2011) HIV and fertility revisited. J Dev Econ 96(1):61–65

    Article  Google Scholar 

  • Lagarde G, Pison E, Enel C (1996) Knowledge, attitudes and perception of AIDS in rural Senegal: relationship to sexual behavior and behavior change. AIDS 10(3):327–334

    Article  Google Scholar 

  • Leuven E, Sianesi B (2003) PSMATCH2: stata module to perform full Mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. http://ideas.repec.org/c/boc/bocode/s432001.html. Accessed 7 Mar 2009

  • Lindan C, Allen S, Carael M, Nsengumuremyi F, Van de Perre P, Serufilira A, Tice J, Black D (1991) Knowledge, attitudes, and perceived risk of AIDS among urban Rwandan women: relationship to HIV infection and behavior change. AIDS 5(8):993–1002

    Article  Google Scholar 

  • Lucas RE Jr (2000) Some Macroeconomics for the 21st century. J Econ Perspect 14(1):159–168

    Article  Google Scholar 

  • Luke N, Munshi K (2006) New roles for marriage in urban Africa: kinship networks and the labor market in Kenya. Rev Econ Stat 88(2):264–282

    Article  Google Scholar 

  • Meltzer D (1992) Mortality decline, the demographic transition and economic growth. Ph. D Dissertation, University of Chicago

  • Mwaluko G, Urassa M, Isingo R, Zaba B, Boerma JT (2003) Trends in HIV and sexual behavior in a longitudinal study in a rural population in Tanzania, 1994–2000. AIDS 17(18):2645–2651

    Article  Google Scholar 

  • Ng’weshemi JZ, Boerma JT, Pool R, Barongo L, Senkoro K, Maswe M, Isingo R, Schapink D (1996) Changes in male sexual behavior in response to the AIDS epidemic: Evidence from a cohort study in urban Tanzania. AIDS 10(12):1415–1420

    Article  Google Scholar 

  • Noel-Miller CM (2003) Concern regarding the HIV/AIDS epidemic and individual childbearing, evidence from rural Malawi. Demogr Res 1(10):318–349

    Google Scholar 

  • Oster E (2005) Sexually transmitted infections, sexual behavior, and the HIV/AIDS epidemic. Q J Econ 120(2):467–515

    Google Scholar 

  • Soares R (2005) Mortality reductions, educational attainment, and fertility choice. Am Econ Rev 95(3):580–601

    Article  Google Scholar 

  • Stoneburner R, Low-Beer D (2004) Population-level HIV declines and behavioral risk avoidance in Uganda. Science 304(5671):714–718

    Article  Google Scholar 

  • Tamura R (2006) Human capital and economic development. J Dev Econ 79(1):26–72

    Article  Google Scholar 

  • Temmerman M, Moses S, Kiragu D, Fusallah S, Wamola IA, Piot P (1990) Impact of single session post-partum counseling of HIV infected women on their subsequent reproductive behavior. AIDS Care 2(3):247–252

    Article  Google Scholar 

  • Williams B, Taljaard D, Campbell C, Gouws E, Ndhlovu L, Van Dam J, Carael M, Auvert B (2003) Changing patterns of knowledge, reported behavior and sexually transmitted infections in a South African gold mining community. AIDS 17(14):2099–2107

    Article  Google Scholar 

  • Young A (2005) The gift of the dying: the tragedy of AIDS and the welfare of future African generations. Q J Econ 120(2):423–466

    Google Scholar 

  • Young A (2007) In sorrow to bring forth children: fertility amidst the plague of HIV. J Econ Growth 12(4):283–327

    Article  Google Scholar 

Download references

Acknowledgements

We thank Janet Currie, Angus Deaton, Jane Fortson, Emily Oster, Adriana Lleras-Muney, seminar participants at Princeton, University of Texas Austin, the 2007 AIDS Workshop at Amsterdam Institute for International Development, the 2008 Society of Labor Economists Meetings, the 2009 Health and Macroeconomy Conference in University of California at Santa Barbara, the 2009 Health and Macroeconomics Conference in Madrid, and the 2009 American Economic Association Annual Meetings session on “HIV/AIDS and Economic Development” for valuable comments and suggestions.

Author information

Authors and Affiliations

  1. Department of Economics, University of Houston, Houston, TX, 77204-5019, USA

    Chinhui Juhn & Sebnem Kalemli-Ozcan

  2. NBER, Cambridge, USA

    Chinhui Juhn & Sebnem Kalemli-Ozcan

  3. Department of Economics, University of Maryland, 4118D Tydings Hall, College Park, MD, 20742, USA

    Belgi Turan

Authors
  1. Chinhui Juhn
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Sebnem Kalemli-Ozcan
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Belgi Turan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Belgi Turan.

Additional information

Responsible editor: Junsen Zhang

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

(PDF 948 KB)

(TEX 17 KB)

Rights and permissions

Reprints and permissions

About this article

Cite this article

Juhn, C., Kalemli-Ozcan, S. & Turan, B. HIV and fertility in Africa: first evidence from population-based surveys. J Popul Econ 26, 835–853 (2013). https://doi.org/10.1007/s00148-012-0456-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00148-012-0456-2

Keywords

Mathematics Subject Classifications (2010)

Search

Navigation