Volume 5, Number 1
July 1999
Measuring Women's Assets
In the discussion on women's land rights (see section above), Alexander had also asked if anyone had suggestions about ways to measure asset ownership and changes in asset ownership in terms of gender. Quisumbing replied that this is something that researchers at IFPRI have been spending a lot of time on. For modeling intrahousehold allocation, her colleagues like to collect information on "assets brought to marriage," which are independent of decisions made within marriage, and could be a measure of bargaining power of the husband and wife. Depending on what the purpose of the study is, current assets may be of greater interest. Quisumbing offered a few thoughts about the type of asset information to collect.
- Any module collecting asset information can be disaggre-gated by gender. This just means adding one column in the questionnaire regarding who in the household owns the asset. This is a fairly straightforward but often overlooked item.
- Quisumbing likes collecting data on assets brought to marriage, and she prefers to collect these for men and women. People usually remember important life-cycle events like marriage and death of parents, probably the two major life-cycle points when there is an intergenerational transfer of resources. Preliminary qualitative work (like the work of Duncan Thomas and Elizabeth Frankenberg of RAND in Indonesia) can identify what the cultural practices are surrounding the transfer of assets at marriage as well as whether or not individuals can "walk away" with these assets at the time of divorce. You can usually obtain a reasonable estimate of the value of the asset at the time it was acquired. If not, collect data on the date of acquisition and the current market price, and use a CPI deflator to adjust assets received at different points in time. You will need to do this anyway if you want to compare changes in assets over time.
- The usual form researchers at IFPRI use is an inventory of assets. In IFPRI surveys in Bangladesh and Ethiopia, researchers collected these data for pre-marriage assets as well as transfers at marriage. The categories included livestock, land, utensils, jewelry, consumer durables, farm equipment, and so on. These were fairly detailed and were designed with reference to previous agricultural surveys as well as qualitative studies on transfers at marriage.
- To track changes in asset holdings, you can, of course, do a panel survey and go back to the same households. But this is likely to be expensive. Another option is to do retrospective questionnaires about the acquisition and disposal of assets. It is important to record the type of transaction, the date when it occurred, and the value of the transaction. Retrospectives have their drawbacks (recall errors, etcetera) but a well designed retrospective can provide very useful information.
- Lastly, one of the most important assets that women may
have is social capital: access to networks, access to kin,
membership in groups, and so on. Lawrence Haddad and John
Maluccio, both at IFPRI, have designed a survey in South Africa
that aims to look more closely at gender differences in social
capital and how these may affect intrahousehold allocation. One
question though: how do you impute a value to a common
property resource that may indeed be one of women's most
important assets?
Cecile Jackson (Cecile.Jackson@uea.ac.uk), School of Development Studies, University of East Anglia, England, commented further that in her experience collecting meaningful data on assets is quite challenging. For a start the Western concept of ownership does not always translate—in one research situation she worked in there was no word to correspond to “own” in the local language, but only the term to “have,” which is clearly very different. Another word to take great care with is “you,” which has both singular and plural meanings in English. A simple question such as “do you own any trees?” can be answered as you personally, and you collectively in a household.
Second, rights and claims to assets, and the boundaries between personal and joint property, are fuzzy and may be represented very differently in a context where a woman is conjugally committed compared to one in which she is ceasing to be (you may not readily think of which of the goats, records, etcetera are yours unless you are thinking of leaving). So latent asset claims are often underestimated.
Finally, asset claims (and valuation of property brought into a marriage) are usually delicate issues, and the expectation that women will be ready to declare and value assets in what are usually public or semipublic discussions (whatever the methods manuals advise) might be misplaced.
