Chris Blattman

The other resource curse: Oil and women’s rights

Why has the Middle East lagged behind the rest of the world in women’s rights and political participation? The strictures of Islam receive a fair share of the blame. Michael Ross of UCLA tells us otherwise. The real reason, he argues, is oil.

Michael spoke today at Yale about a new paper of his in the American Political Science Review. He argues that women’s participation in the formal labor force is a driving force in the development of women’s rights and participation. Oil production tends to crowd out local manufacturing, and so oil crowds out job opportunities for women. That is, the discovery of oil in a less developed country, he argues, sideswipes the development of women’s rights. The discovery of oil might even set back previous gains.

It gets more interesting. If you ignore oil, Islam tends to be associated (statistically) with poor women’s rights. After accounting for oil, that Islam-women’s rights correlation goes away. Variation in oil production seems to explain much of the variation in women’s rights within the Middle East, as well as between the Middle East and the rest of the world.

A number of cases inside and outside the Middle East support Michael’s hypothesis—he documents, for instance, how manufacturing and women’s participation in Nigeria waned as oil wealth grew. He has a nice contrast of oil-rich Algeria to Morocco and Tunisia.

I don’t think Michael’s analysis is ironclad (nor, I think, does Michael). One must always be cautious with cross-country statistical correlations, and case evidence does not always yield systematic patterns that are true outside the case. But Michael’s logic resonates, and the combination of statistical and case evidence is persuasive.

I think a next step would be to investigate his claim further using within-country sector-level data on output and employment, to see if growth in the formal sector is really associated with the growth in women’s employment, and in turn to women’s rights and participation. It would be ideal if we could identify exogenous changes in employment opportunities for women, and the consequent changes in women’s status in those areas. Perhaps variation in the reach of the Africa Growth and Opportunity Act (AGOA) can provide such variation?

It could also be important to get experimental evidence on the link between women’s employment and their rights and collective action (as I am attempting, with some colleagues, in northern Uganda). The Nike Foundation is also embarking on a number of women’s formal sector employment initiatives. That would be an interesting exercise to assess.

If Michael is right, programs to promote women’s employment, especially in the formal manufacturing sector, might be the key to improving their rights. Put crudely: sweatshops may be empowering.

This is a controversial but plausible finding. I often feel that those who decry low-wage manufacturing have a rosy view of the alternatives available to the average developing country woman. Of course, economists often have a rosy view of life in a factory (and possibly overestimate the negative consequences of improving workers rights).

These are questions awaiting some hard evidence.

Short story: programs seeking to increase worker wages and rights in the developing world (an important cause) may have need to strike a careful balance between rights and wages for a few now, versus rights and wages for many later. This is yet another example of the conundrums that makes the process of development so complex, difficult, and ethically vexing.

9 Responses

  1. I did read it and I’d love to find the time to read it and critique it more thoroughly. Maybe I will.

    At this point, I’ll just say I’m very concerned that the simplified media take on this (ie. “oil oppresses women – low wage labour liberates them“) has serious political implications. It justifies the continued demonization of countries that seek to develop and control their oil sectors. It makes people who profit tremendously from exploitative labour practices paint themselves as a somwhow noble part of building a progressive polity.

    If Ross were simply stating that oil was a factor in women’s political involvement I might not be so concerned. Instead, he’s pimping his study with comments like the ones in the above-linked article: “Petroleum perpetuates patriarchy” (and goes on to note that this explains the lack of female influence in countries like Chile – where Micelle Bachelet is President and there are no oil deposits of note).

    I realize this argument is being used as a counterpoint to simplistic blaming of Islamic culture but I think the central message here is very dangerous to nations that are seeking to develop and control of their resource sectors.

  2. @Rudy: I’m not sure if you looked at the paper, but it addresses some of these points. Ross’ work shows an average correlation across all countries, which does not preclude exceptions. After all, the determinants of most social phenomena are complex and multi-dimensional. That is why we care as much about the standard error on any estimate.

    Large standard errors imply that the variability is so great that an average relationship is meaningless. This is not the case in Ross’ paper, where the standard errors are modest, suggesting that while there are many exceptions (some of which you point out) there is a reasonably robust average relationship. Ross’ paper is an excellent example of the scientific method.

  3. Also.

    It’s worth noting that in the UN statistics, Algeria (supposedly hamstrng by its oil wealth) has a little less than 5 per cent of government roles filled by women.

    While Morocco (supposedly making great leaps towards gender equality because it exploits women in the export textile business) has a little less than 1 per cent.

    That’s right Algeria (with oil) is doing FIVE TIMES better than Morocco (with low-wage women getting exploited).

    Sorry, but this is, at best, garbage science – at worst a really offensive attempt to justify some bad economic policies.

    – Rudy

  4. The UN compiled statistics of female representation which are here:

    As I noted over here several countries that are heavily reliant on oil have some of the higher levels of participation for women. Venezuela (a founding member of OPEC) is well above the international average – and pretty comprable with Canada.

    Meanwhile, some of the worst countries for female participation have no oil reserves to speak of – Afghanistan, Djibouti, North and South Korea, Monaco, Nepal, Somalia.

    Furthermore, the premise that low wage manufacturing jobs lead to women’s emancipation (in addition to being deeply offensive) isn’t reflected in the statistics either. Countries like China and India which have become synonomous with low-wage manufacturing have participation rates well below the international average.

    This theory is a cute way to justify econimic exploitation but it’s got a limited to the facts.

  5. Resource curse, pure and simple.

    A ruling class, be they priests or princes, has no incentive to develop the human capital of the country if they can sell the mineral concessions to foreigners and bank the proceeds in Switzerland.

    Wmoens’ rights are part of that, but not all of that. But a close look at womens’ rights in nations who lag behind can throw up some truly horrifying examples of oppression and injustice.

  6. The in-country variation seems like a good place to check. Also, moving outside of the Middle East: how does Kazakhstan compare to its neighbors on women’s rights? Malaysia? Venezuela?

    My gut is that regional/cultural differences play a much greater role than oil production alone.

Why We Fight - Book Cover
Subscribe to Blog