Skip to main content
SpringerLink
Log in

On the heterogeneity of dowry motives

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

Abstract

Dowries have been modeled as pre-mortem bequests to daughters or as groom-prices paid to in-laws. These two classes of models yield mutually exclusive predictions, but empirical tests of these predictions have been mixed. We draw from historical evidence that suggests a bifurcated marriage market, in which some households use dowries as a bequest and others use dowries as a price. The competing theories of dowry allow us to structure an exogenous switching regression that places households in the price or bequest regime. The empirical strategy allows for multiple checks on the validity of regime assignment. Using retrospective marriage data from rural Bangladesh, we find evidence of heterogeneity in dowry motives; that bequest dowries have declined in prevalence and amount over time; and that bequest households are better off compared to price households on a variety of welfare measures.

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

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Notes

  1. We are aware of only one exception: Ambrus et al. (2010) survey households in a different part of Bangladesh about ownership rights over dowry, allowing them to disaggregate bequest vs. groom-price.

  2. Recent examples of this approach include Kopczuk and Lupton (2007), which investigates heterogeneity in bequest motives of the elderly; Vakis et al. (2004), which looks at heterogeneity in the separability of household consumption from production; and Guo (2014), which studies heterogeneity in bequest receipt as a signal of parents’ credit constraints when estimating intergenerational consumption persistence.

  3. If a bride has no brothers (or if her brothers are below the age of majority at the time of her marriage), the incentive problem that forces an inefficient pre-mortem bequest is vitiated. Any bequest will be given upon her parents’ death or at the time when any brother reaches the age of majority and assumes responsibility for managing the parents’ estate.

  4. The reasoning, although not always explicitly stated, is that dowry improves the bride’s welfare by increasing the overall amount of resources available to her (new) family, as well as by improving her threat point or bargaining position within the marriage. Zhang and Chan (1999) uses whether a husband helped with the chores as a measure of wife’s welfare; Brown (2008) expands the indices of welfare to include household consumption of women’s goods; Esteve-Volart (2004) tests Becker’s (1981, pg. 28) observation that one function of dowry is to protect women from divorce; and Suran et al. (2004) looks at levels of reported domestic violence.

  5. Considerable evidence from other dowry-giving societies around the world similarly points to the coexistence of bequest and price dowries, and the transformation from bequest dowries to price dowries over time (Kaplan 1985; Nazzari 1991).

  6. Such discussions abound in South Asian literature. For instance, early in Kamala Markandaya’s 1954 Nectar in a Sieve, a South Indian mother assesses potential grooms for her daughter: “At last, we found one who seemed to fulfill our requirements: he was young and well favored, the only son of his father from whom he would one day inherit a good portion of land.

    “They will expect a large dowry,” I said regretfully. “One hundred rupees will not win such a husband, we have no more.”

    “She is endowed with beauty,” Old Granny said. “It will make up for a small dowry—in this case.” (New York, John Day, pp. 39–40).

  7. The Survey on the Status of Women and Fertility in Uttar Pradesh and Tamil Nadu (Smith et al. 2000) asked randomly-sampled brides about the control they exercise over different forms of dowry. In marriages where cash dowry was given as dowry, fewer than 10 % of respondents reported that they had the “major say” in how the cash was spent, and more than 70 % reported that they had no say at all. In marriages where jewelry, gold, or silver was given, more than 47 % of respondents reported that they had the major say in how it was used or whether it could be sold, while fewer than 25 % reported that they had no say at all.

  8. There is some evidence of a rise in post-marital extortion of dowry beginning around the late 1990s—after our 1996 survey (Suran et al. 2004), a practice which may have spread from India (Bloch and Rao 2002).

  9. Anderson (2004) separates the sample of Pakistani households into rural and urban; Esteve-Volart (2004), using the same dataset as the one used in our paper, separates the sample into Hindu and Muslim. Both papers estimate separate dowry functions by subsample, and conclude on the basis of parameter estimates that dowry indeed functions differently in the groups.

  10. More precisely, throughout, we assume that such marriages in which positive dowry is given more likely to be in the other regime.

  11. Further discussion of the data construction is in the Data Appendix.

  12. For this reason, we use the eligible sex ratio to test Rao’s (1993) “marriage squeeze” hypothesis. See the Appendix for details of the construction of the series.

  13. This is the empirical strategy implicitly adopted by Rao (1993), Edlund (2001), and Brown (2008). In Rao’s case, α is set to 0, while Edlund and Brown both set α to 1.

  14. The log-likelihood from the pooled regression is −1690, while the log-likelihood from the switching regression is −1560. Twice the difference between log-likelihood for the pooled and mixture models is 260. As discussed in the previous section, a comparison of log-likelihoods is feasible because the pooled specification is nested in the likelihood function for the mixture. However, when the switching model is constrained to yield the pooled model—that is, when the regime parameters are restricted to be equal—several parameters are unidentified. Monte Carlo results indicate that a conservative likelihood ratio test can be constructed by using the χ 2 distribution and setting the degrees of freedom equal to the number of unidentified parameters plus the number of constraints (Goldfeld and Quandt 1976). For this specification, we have 67 degrees of freedom; at 1 % significance level, this yields a critical value of 96.83, well below our likelihood ratio statistic.

  15. Here, we are following Hartley (1978) in assigning regime membership using the cutoff probability of 0.5. That is, bequest regime households are those households that have a probability of greater than 0.5 of being in the bequest regime.

  16. Placing this result together with the findings from the regime separation that bequest dowries tend to be much larger may explain the overall absence of dowry inflation observed by Logan and Arunachalam (2014).

  17. In Table 5, we label the regimes as “price” and “bequest” for convenience.

  18. Summary statistics are computed using the full sample of marriages, which includes 21 for which year of marriage cannot be recovered.

  19. Reported dowry amounts between husband and wife are reasonably correlated; the Pearson correlation coefficient is 0.61. All results are qualitatively similar when we exclude husbands’ dowry reports.

  20. Volumes compiled before partition but published afterward are simply called Statistical Abstract, India.

  21. Details of the sources as well as unit conversions are given in Table 6.

References

  • Ahmed R, Naher MS (eds) (1987) Brides and the demand system in Bangladesh. Centre for Social Studies. Dhaka University, Dhaka

  • Alfano M (2014) Daughters, dowries: deliveries: the effect of marital payments on fertility choices in India, Working Paper

  • Ambrus A, Field E, Torero M (2010) Muslim family law, prenuptial agreements and the emergence of dowry in Bangladesh. Q J Econ 125(3):1349–1397

    Article  MATH  Google Scholar 

  • Amin S, Cain M (1998) The Rise of Dowry in Bangladesh. In: Jones GW , Douglas RM , Caldwell JC , D’Souza RM (eds) The Continuing Demographic Transition. Oxford University Press, pp 290–306

  • Anderson S (2003) Why dowry payments declined with modernization in Europe but are rising in India. J Polit Econ 111(1):269–310

    Article  Google Scholar 

  • Anderson S (2004) Dowry and Property Rights, BREAD Working Paper No. 080

  • Anderson S (2007a) The economics of dowry and brideprice. J Econ Perspect 21(4):151–174

    Article  Google Scholar 

  • Anderson S (2007b) Why the marriage squeeze cannot cause dowry inflation. J Econ Theory 137(1):140–152

    Article  MATH  Google Scholar 

  • Arunachalam R, Naidu S (2014) The price of fertility: marriage markets and family planning in Bangladesh. Working Paper

  • Banerjee K (1999) Gender stratification and the contemporary marriage market in India. J Fam Issues 20(5):648–676

    Article  Google Scholar 

  • Basu S (ed) (2005) Dowry and inheritance. Zed Books, London

  • Becker GS (1981) A Treatise on the Family. Harvard University Press, Cambridge

    Google Scholar 

  • Bloch F, Rao V (2002) Terror as a bargaining instrument: a case study of dowry violence in rural India. Am Econ Rev 92(4):1029–1043

    Article  Google Scholar 

  • Bloch F, Rao V, Desai S (2004) Wedding celebrations as conspicuous consumption: signaling social status in rural India. J Hum Resour 39(3):675–695

    Article  Google Scholar 

  • Botticini M, Siow A (2003) Why Dowries?. Am Econ Rev 93(4):1385–1398

    Article  Google Scholar 

  • Brown PH (2008) Dowry and intrahousehold bargaining: evidence from China. J Hum Resour 44(1):25–46

    Google Scholar 

  • Caplan L (1993) Bridegroom Price in Urban India: Class, Caste and ‘Dowry Evil’ Among Christians in Madras. In: Uberoi P (ed) Family, Kinship and Marriage in India. Oxford University Press, Delhi, pp 357–382

    Google Scholar 

  • Dalmia S (2004) A Hedonic analysis of marriage transactions in India: estimating determinants of dowries and demand for groom characteristics in marriage. Res Econ 58:235–255

    Article  Google Scholar 

  • Dasgupta I, Mukherjee D (2003) Arranged marriage, dowry and female literacy in a transitional society, CREDIT Research Paper 03/12. University of Nottingham

  • Dempster A, Laird NM , Rubin DB (1977) Maximum likelihood from incomplete data via the em algorithm. J R Stat Soc Ser B 39:1–38

    MathSciNet  MATH  Google Scholar 

  • Dickens WT, Lang K (1985) A test of dual labor market theory. Am Econ Rev 75(4):792–805

    Google Scholar 

  • Edlund L (2000) The marriage squeeze interpretation of dowry inflation: comment. J Polit Econ 108(6):1327–1333

    Article  Google Scholar 

  • Edlund L (2001) Dear Son—Expensive Daughter: Do Scarce Women Pay to Marry?, Working Paper. Columbia University

  • Edlund L, Lagerlöf N-P (2004) Implications of Marriage Institutions for Redistribution and Growth, Working Paper. Columbia University

  • Esteve-Volart B (2004) Dowry in Rural Bangladesh: Participation as Insurance against Divorce, Working Paper. York University

  • Garcia R, Lusardi A, Ng S (1997) Excess sensitivity and asymmetries in consumption: an empirical investigation. J Money Credit Bank 29(2):154–176

    Article  Google Scholar 

  • Goldfeld SM, Quandt RE (1976) Techniques for Estimating Switching Regressions. In: Goldfeld SM , Quandt RE (eds) Studies in Non-Linear Estimation. Ballinger, Cambridge, pp 3–35

    Google Scholar 

  • Goode WJ (1963) World Revolution and Family Patterns. Free Press, New York

    Google Scholar 

  • Goody J (1998) Dowry and the Rights of Women to Property. In: Hann CM (ed) Property Relations: Renewing the Anthropological Tradition. Cambridge University Press, pp 201–213

  • Grossbard-Shechtman S (1993) On the Economics of Marriage: A Theory of Marriage, Labor, and Divorce. Westview, Boulder, CO

  • Guo S (2014) Switching regression estimates of the intergenerational persistence of consumption. Econ Inq 52(4):1503–1524

    Article  Google Scholar 

  • Hartley MJ (1978) Estimating mixture of normal distributions and switching regressions: comment. J Am Stat Assoc 73(364):738–741

    Google Scholar 

  • Jappelli T, Pischke J-S, Souleles NS (1998) Testing for liquidity constraints in euler equations with complementary data sources. Rev Econ Stat 80(2):251–262

    Article  Google Scholar 

  • Kabeer N (2001) Ideas, economics and the sociology of supply: explanations for fertility decline in Bangladesh. J Dev Stud 38(1):29–70

    Article  Google Scholar 

  • Kaplan MA (1985) The marriage bargain: women and dowries in European history. Institute for Research in History. Haworth Press, New York

    Google Scholar 

  • Khan AR, Hossain M (1989), The Strategy of Development in Bangladesh. Macmillan and OECD Development Centre

  • Kishwar M (1989) Dowry and inheritance rights. Econ Polit Wkly:587–588

  • Kopczuk W, Lupton JP (2007) To leave or not to leave: the distribution of bequest motives. Rev Econ Stud 74(1)

  • Lee L-F, Porter RH (1984) Switching regression models with imperfect sample separation information—with an application on cartel stability. Econometrica 52(2):391–418

    Article  MathSciNet  MATH  Google Scholar 

  • Lindenbaum S (1981) Implications for women of changing marriage transactions in Bangladesh. Stud Fam Plan 12(11):394–401

    Article  CAS  Google Scholar 

  • Logan TD, Arunachalam R (2014) Is there dowry inflation in south asia? an assessment of the evidence. Hist Methods 47(2):81–94

    Article  Google Scholar 

  • Maddala GS (1983) Limited-Dependent and Qualitative Variables in Econometrics. Cambridge University Press

  • Maertens A, Chari AV (2012) Learning your Child’s Price: Evidence from Anticipated Dowry Payments in Rural India, Working Paper

  • Maitra S (2009) Can Population Growth Impact Marriage Payments? Theory and the Indian Evidence, Working Paper

  • Maitra S (2011) Population Dynamics and Marriage Payments: An Analysis of the Long-Run Equilibrium in India, Working Paper

  • Majumdar R (2004) Snehalata’s death: dowry and women’s agency in colonial Bengal. Indian Econ Social Hist Rev 41(4):433–464

    Article  Google Scholar 

  • Menski W (ed) (1998) South Asians and the Dowry Problem. Trentham, London

  • Morduch JJ, Stern HS (1997) Using mixture models to detect sex bias in health outcomes in Bangladesh. J Econ 77:259–276

    Article  MATH  Google Scholar 

  • Mukherjee D, Mondal SS (2006) Dowry Revisited: A Human Capital Approach. In: Bharati P , Pal M (eds) Gender Disparity: Manifestations, Causes and Implications. Anmol Publications, New Delhi, pp 278–297

    Google Scholar 

  • Nazzari M (1991) Disappearance of the Dowry: Women, Families, and Social Change in São Paulo, Brazil (1600–1900). Stanford University Press, Stanford

    Google Scholar 

  • Oldenburg VT (2002) Dowry murder: the imperial origins of a cultural crime. Oxford University Press

  • Quandt RE (1958) The estimation of the parameters of a linear regression system obeying two separate regimes. J Am Stat Assoc 53(284):873–880

    Article  MathSciNet  MATH  Google Scholar 

  • Rahman M, DaVanzo J, Razzaque A (2002) When Will Bangladesh Reach Replacement-level Fertility? The Role of Education and Family Planning Services. In: Completing the Fertility Transition. United Nations Population Division, New York, pp 343–355

  • Rahman O, Menken J, Foster A, Gertler P (1999) The Matlab Health and Socioeconomic Survey: Overview and User’s Guide, DRU-2018/1. RAND, Santa Monica, CA

    Google Scholar 

  • Rao V (1993) The rising price of husbands: a hedonic analysis of dowry increases in rural India. J Polit Econ 101(4):666–677

    Article  Google Scholar 

  • Rao V (2000) The marriage squeeze interpretation of dowry inflation: response. J Polit Econ 108(6):1334–1335

    Article  Google Scholar 

  • Rao V (2007) The Economics of Dowries in India. In: Basu K (ed) Oxford Companion to Economics in India. Oxford University Press

  • Sekhri S, Storeygard A (2014) Dowry deaths: Response to weather variability in India. J Dev Econ 111:212–223

    Article  PubMed  Google Scholar 

  • Setalvad AM (2001) Paper Laws. In: Thorner A (ed) 10 Daniel Thorner Memorial Lectures: Land, Labour & Rights. Tulika Books, New Delhi, pp 58–69

    Google Scholar 

  • Sharma U (1984) Dowry in North India: Its Consequences for Women. In: Uberoi P (ed) Family, Kinship and Marriage in India. Oxford University Press 1993, Delhi, pp 341–356

    Google Scholar 

  • Smith HL, Ghuman SJ , Lee HJ, Mason KO (2000) Status of Women and Fertility, Machine-readable data file

  • Srinivas MN (1984) Some Reflections on Dowry. Centre for Women’s Development Studies, Delhi

    Google Scholar 

  • Suran L, Amin S, Huq L , Chowdury K (2004) Does Dowry Improve Life for Brides? A Test of the Bequest Theory of Dowry in Rural Bangladesh, Policy Research Division Working Paper 195, Population Council

  • Tambiah SJ (1973) Dowry and Bridewealth and the Property Rights of Women in South Asia. In: Goody J, Tambiah SJ (eds) Bridewealth and Dowry. Cambridge at the University Press, pp 59–169

  • Tertilt M (2005) Polygyny, fertility, and savings. J Polit Econ 113(6):1341–1371

    Article  Google Scholar 

  • Vakis R, Sadoulet E, de Janvry A , Cafiero C (2004) Testing for separability in household models with heterogeneous behavior: a mixture model approach, University of California, Berkeley, CUDARE Working Paper 990

  • Zhang J, Chan W (1999) Dowry and wife’s welfare: a theoretical and empirical analysis. J Polit Econ 107 (4):786–808

    Article  Google Scholar 

Download references

Acknowledgments

We are grateful for helpful comments from Pranab K. Bardhan, Chang-Tai Hsieh, Jennifer Johnson-Hanks, Lung-Fei Lee, Ronald D. Lee, David I. Levine, Ian McLean, Manisha Shah, and numerous seminar and conference participants at UC Berkeley, The Ohio State University. We would particularly like to thank J. Bradford DeLong, Audrey L. Light, and Edward Miguel for several useful discussions and encouragement throughout. Hao Shi and Ranjan Shreshtra provided excellent research assistance. We also thank the anonymous referees and the editor for thoughtful comments which improved the manuscript. The usual disclaimer applies.

Author information

Authors and Affiliations

  1. Department of Economics and Population Studies Center, University of Michigan, 238 Lorch Hall, 611 Tappan Avenue, Ann Arbor, MI, 48109-1220, USA

    Raj Arunachalam

  2. Department of Economics, The Ohio State University, 410 Arps Hall, 1945 North High Street, Columbus, OH, 43210, USA

    Trevon D. Logan

  3. National Bureau of Economic Research, Cambridge, Massachusetts, USA

    Trevon D. Logan

Authors
  1. Raj Arunachalam
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Trevon D. Logan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Trevon D. Logan.

Additional information

Responsible editor: Junsen Zhang

Appendices

Appendix A.1: Matlab Data

The dataset is described in Rahman et al. (1999). There are 4942 women for whom year of marriage is reported or can be reconstructed from age and age at marriage. In addition, we have 808 previous marriages reported, giving a total of 5750 marriages.Footnote 18 Of these, 1722 women report a positive dowry amount (including in kind transfers). Husbands’ reports are rarer due to the sampling design of the MHSS, and also because for previous marriages only women’s reports of the dowry are available. Of the 1722 positive dowry marriages, in 566 the husband also reported a dowry amount. In an additional 111 marriages, the husband reported a dowry even when the wife did not. This gives a total of 1833 marriages in which a dowry was reported. When both spouses report a dowry amount, we take the average of the reports.Footnote 19 Due to other missing covariates for some marriages, the estimation sample uses 1627 marriages. Some additional notes on variable construction:

  • We use height and weight to calculate Body Mass Index (BMI) and categorize individuals as underweight, normal weight, or overweight/obese using World Health Organization guidelines. The vast majority of the sample (97 % of women and 98 % of men) have a BMI that is classified as normal or underweight.

  • Level of school attainment (primary, low secondary, or high secondary) are assigned from the years of schooling, in accordance with Rahman et al. (2002). The omitted category in the regression results is “wife/husband did not attend school.”

Appendix A.2: Rice Prices

No single source contains rice prices for the entire period, so we constructed a series using the Statistical Abstracts for British India Footnote 20 for the period 1910–1946; the Pakistan Central Statistical Office’s Statistical Yearbook 1955 for 1949; the Pakistan Central Statistical Office’s 20 Years of Pakistan in Statistics for 1950-1967; and the Statistical Yearbooks of Bangladesh for 1965-1996.Footnote 21 Duplicate observations for the years 1964 to 1967 were used to convert between takas and rupees, and overlapping medium and common rice prices for the years 1950 to 1967 are used to convert between rice qualities. We used a linear interpolation to cover the missing years (1947–1948). Real dowries are expressed in (medium quality) rice kilograms.

Appendix A.3: Sex Ratio

Constructing a sex ratio time series has proven treacherous in previous dowry work (Edlund 2000; Rao 2000). We adopt a simple approach that takes account of the mortality trajectory in the years between Census records. We calculate the sex ratio from the Census Records of India (1931, 1941), Pakistan (1951, 1961), and Bangladesh (1974, 1984). Where possible, we use the lowest level of geographic detail to correspond with the district in the Matlab study, although this was not possible for early years of the Census records, particularly the Indian Census records. Taking the count of males and females in the Census year, we construct the adjusted sex ratio as described by Rao (1993), where the adjusted sex ratio gives the number of males aged 20–29 divided by the number of females aged 10–19. We calculate the relevant ratio for each age group, from ages 10 to age 90, using the same ten year differential.

For the years between the Census years, the cohorts’ age and mortality, particularly child mortality in the earlier years of our records, is pronounced, so the relevant sex ratio for each age must be adjusted for mortality each year. We use the actual number of males and females in the Census year, and then use the period lifetable constructed with the Census records to derive an estimate of survivorship to the next age ( n q x ). Since these values will differ for both men and women, the sex ratio will change for the years between Censuses. For example, males aged 20–29 in 1961 will be aged 25–34 in 1966, and the best available estimate for the fraction of men surviving to that age would be the period survival rate for men aged 25–34 in 1961. For the final Census we use, we create a quasi-synthetic cohort and assume that the course of mortality over the subsequent years will be the same as for the later cohorts whose mortality trajectory is recorded in the Census.

Table 6 Sources for rice price
Full size table

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Arunachalam, R., Logan, T.D. On the heterogeneity of dowry motives. J Popul Econ 29, 135–166 (2016). https://doi.org/10.1007/s00148-015-0544-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00148-015-0544-1

Keywords

JEL Classification

Search

Navigation