THE BLOG
08/01/2008 05:12 am ET Updated May 25, 2011

The Scourge of the IMF

Tuberculosis, a treatable disease, kills 1.7 million people a year worldwide.

TB incidence, according to the World Health Organization, seems to be correlated to broad social factors, like access to clean water and sanitation, HIV incidence and national health expenditures.

A just published study in the journal PLoS (Public Library of Science) Medicine investigates the role of different possible explanatory factor: the International Monetary Fund (IMF). The researchers' study focuses on the period 1991 to 2003 for the former Soviet Union and Eastern Europe, a region for which there is robust data.

The results: The researchers concluded "that IMF economic reform programs are strongly associated with rises in tuberculosis mortality rates in post-communist Eastern European and FSU [former Soviet Union] countries, even after correcting for potential selection bias, tuberculosis surveillance infrastructure, levels of economic development, urbanization, and HIV/AIDS."

"We estimated an increase in tuberculosis mortality rates when countries participate in an IMF program, which was much greater than the reduction that would have been expected had the countries not participated in an IMF program. On the other hand, we estimated a decrease in tuberculosis mortality rates associated with exiting an IMF program."

In other words: When countries entered IMF programs, TB rates went up. When the programs ended and countries escaped from IMF influence, TB rates went down.

OK, but the region was in chaos after the fall of the Soviet Union. Economies crashed and per capita income plummeted. Crime rose, incarceration rates jumped, HIV spread. Aren't these the real factors behind rising TB rates?

Explains Sanjay Basu of Yale University, one of the study authors: "First of all, not all of these countries in this region were dependent on the former Soviet Union. Many of them actually had an increase in GDP after the fall of the former Soviet Union. Several were not part of the trading bloc. And in some of the key countries where TB rates rose, we actually saw an increase in economic growth. So economic downturns could not explain, as the WHO itself has stated, the trends of tuberculosis in that regions. Something else was going on."

"The reason we use such heavy statistics is precisely to factor in these other issues -- incarceration, HIV, changes to the economy, changes to the healthcare infrastructure. We found a statistically independent effect of the IMF. That's not to say that the IMF was the only cause of TB in this region. The economy, incarceration, HIV -- these are all very important, but those factors could not fully explain TB in the region."

(An interview with Basu follows this column.)

The PLoS study found that participating in an IMF program correlated with increases in tuberculosis incidence of 13.9 percent and an increase in TB mortality rates of 16.6 percent. Basu says that, if the study results are valid, they suggest "we would have averted tens of thousands of deaths and hundreds of thousands of new cases" if countries in the region had never entered IMF programs.

The theory of the study authors is that IMF programs drive down healthcare spending, and this reduced investment in healthcare explains the rise in TB incidence and death. Basu emphasizes, correctly, that the issue is not so much the IMF directing countries to spend less on health. Rather, it imposes a set of policy constraints -- including overall limits on government spending, and needlessly low inflation targets -- that inevitably result in countries spending less on health.

There are always variations between regions, but there is nothing about the PLoS researchers' story that suggests things are any different in Africa, the region where the IMF now exerts the most influence.

Not surprisingly, the IMF has rejected the PLoS findings. "Severe methodological shortcomings limit the scope of these results and prevent any causal interpretation," asserts an IMF response that is much more subdued than comments from spokespeople. "The fundamental problem is that this study does not take properly into account that countries implement IMF-supported reforms in times of economic distress."

Says the IMF response: "The authors do not take into account that the economic and social instability following the collapse of Soviet Union may have had a direct impact on TB incidence in the 21 transition economies considered in the study."

The problem with this line of argument is that it is not true. The authors did take the economic and social instability into account.

Can anything be done about IMF policies with such harmful impacts?

Yes. The IMF is a human creation, not a force of nature.

The United States Congress will next year have a unique opportunity to influence IMF policy. The IMF needs approval from the Congress to go ahead with plans to sell some of the gold it controls. This gold would be used to fund the IMF's administrative costs -- a new income stream the IMF desperately needs. Interest payments from middle-income countries previously paid for administrative costs, but these countries have paid back their loans in order to escape from IMF influence.

As the U.S. Congress looks to approve gold sales to finance the IMF, it must insist that the IMF first end the mandates that effectively restrict countries' health spending, and force borrowing countries to implement a discredited market fundamentalist policy agenda.

----

An Interview with Sanjay Basu,
Department of Epidemiology and Public Health, Yale University
(Co-Author, "International Monetary Fund Programs and Tuberculosis Outcomes in Post-Communist Countries," PLoS Medicine, July 2008, Vol. 5, Issue 7)

Question: Can you give a thumbnail sketch of what you found?

Sanjay Basu: The hypothesis we wanted discussed was the effect of IMF programs on health outcomes. Tuberculosis is a good test because it has long been viewed as a social indicator. When your society breaks down, tuberculosis rates rise pretty quickly, particularly death rates from TB.

We tested two decades of data from the entire region of Eastern Europe and the former Soviet Union. That region was of interest because they had relatively similar public health infrastructure, training systems and general design of their TB program. Their TB rates were improving just prior to the collapse of the former Soviet Union, when many of them received IMF loans of different size and duration. Therefore, we could test the impact of these IMF loans. And they also received other loans from non-IMF sources, of similar size and duration, as a kind of control group.

What we found was that even after correcting for the impact of the economic downturn in some of these countries -- although for some of them, GDP actually increased -- and for HIV rates, incarceration rates and dozens of other variables, IMF loans profoundly affected tuberculosis rates.

We wanted to make sure this wasn't a false association, that we weren't blaming the IMF for responding to a worsening situation. Using something called a control function approach, which gives us a sense of whether someone was responding to a problem, or whether they were generating a problem, we found that the IMF seemingly precipitated the worsening in tuberculosis rates.

One thing that was really strong and surprising and rarely happens in social science data, was a dose-response relationship. The higher that the IMF loan went, the higher the TB rates went in proportion. Similarly, when the IMF loan ended, the TB rates fell to the same degree, and the duration of the loan corresponded exactly to the TB rates. And they corresponded well also to public health funding that was cut during the period of the loan, the number of doctors available and the coverage of WHO-recommended treatment. Meanwhile, the non-IMF loans did not have these negative impacts either on health spending or on tuberculosis outcomes.

Question: Is your hypothesis that the mechanism by which the IMF loans were harmful was that they reduced spending on healthcare?

Basu: Yes. That's what all arrows pointed to.

Question: The IMF says it does not require countries to reduce health spending.

Basu: Well, the IMF doesn't actually directly tell countries what to do, but it provides inflationary targets for them. And implicitly the countries will have to cut health and education funding in order to reach those targets.

It's like saying, you can have our loan but you have to follow our investment advice. We're not going to tell you exactly what to invest in, but here's how it should end up looking.

It's kind of like tying a person's hands behind their back and telling them to eat pie, and when they stick their face in the pie, saying, well, you could have used a fork.

So no, the IMF doesn't directly tell countries what to do, but implicitly, there's a political reality.

Question: There were 21 countries that you studied, and all but one had the IMF loans?

Basu: Right. But we weren't just comparing whether a country did or did not take an IMF loan. There could be many other factors -- including whether a country needed to take a loan, and variations between countries -- that could explain tuberculosis rates. So we did what is called a within-country analysis. It holds all the other factors within a country the same so that you can see the effect of one factor in particular -- in this case, the IMF loan. After correcting for the economic factors and background of those countries, we found that countries that took small loans did better, and those that took the larger loans did worse.

It's kind of like what we do for pharmaceutical studies. If a person is taking a drug for high blood pressure, maybe their blood pressure drops a lot and somebody else's doesn't. But to know if the cause is the drug or not, you have to correct for who that person is. Maybe they have a kidney problem that prevents their blood pressure from decreasing. We corrected for those factors.

Question: In responding to your study, the IMF has argued that you found effects on TB rates too quickly. If IMF loans were responsible for worsening TB, they say, there should have been more lag time between when countries took loans and when the impact on TB was evident.

Basu: That's a fallacy. Tuberculosis mortality in particular is a very quick responder to social change. If you close a hospital, then people who already have tuberculosis will die, and they will die pretty quickly. That doesn't take years to evolve.

Changes in incidence and prevalence are slower, but certainly over two decades we see enormous changes. In Eastern Europe and Central Asia we saw a reverse impact over 20 years. This is a pretty long duration of time over which to see large macroeconomic effects.

Question: Are you able to say what the increased mortality rates mean and what the increased incidence rates mean in terms of overall numbers?

Basu: In a statistical sense, you can say how many statistical deaths have resulted. In terms of deaths, it was on the order of tens of thousands. In terms of incidence of new cases, it was on the order of hundreds of thousands. Of course, these are all jumbled in reality, but in a statistical sense, that is part of the effect due to the IMF programs.

Question: Does that mean that, had there not been IMF programs in these countries, that you would have seen tens of thousands of people living who did in fact die?

Basu: If the associations do indeed point to causation, and everything we did said that it indeed was a causal effect -- and if we're not confounding the results by some unforeseen variable -- then yes, we would have averted tens of thousands of deaths and hundreds of thousands of new cases. And the TB rates before the IMF came were indeed stable and declining.

Question: One intuitive skepticism about your results is that there was a lot of chaos after the Soviet Union collapsed -- declining economies, rising crime and incarceration, social breakdown. How can you really blame the IMF or associate rising TB with the IMF when all of these overarching macro developments were taking place?

Basu: That's why it's important to do statistics in these studies. First of all, not all of these countries in this region were dependent on the former Soviet Union. Many of them actually had an increase in GDP after the fall of the former Soviet Union. Several were not part of the trading bloc. And in some of the key countries where TB rates rose, we actually saw an increase in economic growth. So economic downturns could not explain, as the WHO itself has stated, the trends of tuberculosis in that regions. Something else was going on.

And the reason we use such heavy statistics is precisely to factor in these other issues --incarceration, HIV, changes to the economy, changes to the healthcare infrastructure. We found a statistically independent effect of the IMF. That's not to say that the IMF was the only cause of TB in this region. The economy, incarceration, HIV -- these are all very important, but those factors could not fully explain TB in the region. T he IMF was found to have a profound impact and explain a lot more of the variation than some of these other factors.