MethodLogical is now at

MethodLogical is now at

Due to some persistent technical issues we've been having with Blogger, we're now posting at Please update your RSS feeds, etc. For the time being, our new posts will automatically be mirrored here, but you'll have to visit the new site to comment.

Thursday, November 18, 2010

Health --> Life Expectancy --> Education?

Last week, Methodlogical contributor Ben Elberger published a fascinating analysis of why we see lower than expected take-up rates of microinsurance in Africa. Microinsurance seems to be a concept that makes so much sense in paper, yet as Ben shows us through his savings and credit cooperative (SACCO) example, doesn’t always transpose into reality:

The story often goes that a charismatic individual starts up a SACCO offering loans at great rates if you save X% upfront. You save over a number of weeks as it's difficult to plunk down all of the money in one go. Some people don't even want loans, they just save with the SACCO because the interest rate offered on savings is also good. The SACCO continues to grow rapidly and then the General Manager takes the savings and runs away. It's a ponzi scheme and the recoverability of savings is next to nil. Those with loans outstanding greater than their upfront savings win but the broad majority have net savings with the SACCO and lose it all.”

The lack of certainty that your microinsurance provider will actually still exist in three months is a powerful worry that underscores the reality of what Kenyans and Ugandans have witnessed over and over again. Similarly, selling someone on any product that has a time horizon of years is difficult when your potential customers are living day-to-day. Uncertainly plagues long-term investment decisions.

But we haven’t yet discussed perhaps the most crucial long-term investment and longest time-horizon investment of all: education. Education involves a huge up-front investment of years of life that could otherwise be devoted to earning income or other productive uses. Furthermore, with such large societal returns to education, especially for primary education, encouraging greater investment in education should be in the best interest of the individual, the community, and society as a whole. But still, uncertainty plagues education investment decisions, as well. A longer time horizon increases the value of investments that pay out over time – so the uncertainty of how long one’s time horizon will be (life expectancy) could play a crucial role in determining how much of an investment in education one chooses to make.

What sort of evidence is out there to show that maybe there does exist a causal chain of health -> life expectancy -> education? In Lorentzen et al’s “Death and Development,” the authors find that a one standard deviation increase in adult mortality is associated with a 9% decrease in educational enrollment rates in Africa, which is about 1/3 of the average enrollment rate. Furthermore, higher adult mortality is associated with increased levels of risky behavior, higher fertility, and lower investment in other dimensions of physical and human capital.

There has also been compelling evidence found in Sri Lanka, as well. Jayachandran & Lleras-Muney examined a sudden drop in mortality risk in Sri Lanka between 1946 and 1953. During those years, Sri Lanka experienced a 70% reduction in the maternal mortality risk and an increased female life expectancy at age 15 by 1.5 years (4.1%). This data is particularly compelling because maternal mortality rates (MMRs) could realistically affect education investment decisions, unlike infant mortality risk, which is realized before educational decisions are made. The authors found that female children in Sri Lanka received on average 0.11 extra years of education per year of life expectancy increase.

The policy implications of these findings are compelling: not only do we see quantifiable and large increases in educational investment decisions as life expectancy increases, but this also implies that the benefits of health interventions that improve life expectancy may be larger than we think. The spillover benefits of life expectancy-improving health interventions need to be fully factored into the cost-benefit analysis of investing in health programs, because in terms of educational investments, evidence shows us that the benefits may be substantial.

Especially in costly treatments, such as ARVs, that can substantially increase the life expectancy of HIV/AIDS-infected individuals, the spillover benefits on educational attainment can be substantial. In Africa, 40% of deaths ages 15-49 are AIDS-related. We know that increased life expectancy would lead to higher educational investments – but what we still need to figure out is how to fully account for and quantify the benefits of this higher educational attainment into our ARV investment decisions.


  1. There's a fascinating disconnect between people's subjective estimates of the risk of contracting HIV (which tend to be extremely high) and their actual responses in terms of safer sex (which tend to be close to nil). One proposed explanation is a sense of "invincibility" or a misunderstanding of how their personal risk relates to the average risk. Another idea is that people expect to live short lives regardless.

    If we're hoping to see changes in educational or other investments as a result of ARV treatment, we need a firm grasp of how people are actually thinking about AIDS and their own life expectancy and future. This is an area where I'm hoping to do more research in the future.

  2. There seems to be some, albeit imperfect, evidence that education can drive life-expectancy, at least in the U.S. ( I guess the hope is if we can enter a positive cycle where education and longevity reinforce each other.

    In terms of what Jason was saying, in bioethics we learned the difference between "understanding" and "appreciating" the consequences of actions. Most people understand that smoking causes disease and unprotected sex raises the risk of HIV infection. However, most of us fail to appreciate that when WE smoke WE are more likely to get sick and when WE have unprotected sex that WE are more likely to get HIV. Maybe it's because the consequences of these actions are not immediately apparent (in fact, they may occur decades later). But, that doesn't explain why people don't wear seat belts...


Note: Only a member of this blog may post a comment.