IMF critics allege that its programs unduly constrain health spending in poor countries, even when external financing is potentially available. The IMF argues that countries set their own spending priorities while the Fund monitors overall spending and fiscal sustainability. The issue has become more pressing, as countries seek to utilize scaled-up aid, including billions of dollars for prevention and treatment of HIV/AIDS. CGD recently launched a new working group to establish what actually happens under IMF programs and to make practical recommendations for improvements. Working group chair David Goldsbrough, a CGD visiting fellow who was previously deputy director of the IMF’s Independent Evaluation Office, chaired the group’s first meeting last week. The day-long study session brought together a diverse group of experts with experience in analyzing and implementing macroeconomic and health sector policies and setting budgetary priorities. Goldsbrough discussed the group’s task and how it will proceed. Q: What is the single most important concern about the impact of IMF programs on health spending that the group will study? A: The key issue is whether the IMF’s approach in low-income countries has unduly constrained the policy options available to governments as they formulate and implement their macroeconomic plans. This could be because the IMF takes too conservative view on what is needed for macroeconomic stability and for longer-term fiscal sustainability or because some of the policy instruments it uses in its programs have unintended consequences for the health sector. Of course, even with debt relief and the prospects of higher aid, countries still face resource constraints that require governments to make difficult choices that will not please all stakeholders. So if health spending in a country is not growing as fast as some critics would like (e.g. to keep on track to meet the MDGs), this does not necessarily mean that the IMF has acted inappropriately. This is why it is useful to couch the key issue in terms of whether IMF actions have unduly narrowed the policy space. Q: Critics say that the increased use of public sector wage ceilings required by IMF programs are an important part of the problem. How do you plan to evaluate whether this is indeed the case? A: Wage bill ceilings in IMF programs have become more frequent in recent years, especially in Africa: slightly over half of all IMF arrangements negotiated in 2003-2005 had some form of ceiling on the wage bill. We plan to use a small number of case studies to investigate in depth how and why such ceilings were derived, what provision they made for increased health sector recruitment, and what the consequences were in practice. Q: The group’s terms of reference say that you will also consider ways to improve the predictability of aid flows. Isn’t this mostly up to the World Bank and bilateral donors rather than the IMF? A: Yes, it is mainly up to the donors to improve aid predictability, and some the recommendations of the working group may be directed at bilateral donors or the World Bank. But the group will also explore ways in which the treatment of aid flow projections in IMF programs might be improved. Q: You plan to consider ways to improve the negotiation of IMF programs. What exactly will you be looking at in this regard? A: We will try to look at two aspects of the IMF “way of doing business”. First, in many low-income countries, macroeconomic stability has been largely restored and the key macro policy challenge is to help manage effectively a scaling-up of aid –financed spending. This requires integrating more fully analysis on sector-level expenditure choices with macroeconomic assessments, in a longer-term timeframe. Second, we will look at ways of introducing greater transparency about the analysis and assumptions underlying IMF programs and policy advice. Q: These issues are both complex and quite controversial, with strongly held views on both sides. How will you ensure that the group’s work is evidence-based? A: First, I should emphasize that the purpose of the group is not to revisit such issues as the appropriateness of particular development models or aspects of the so-called “Washington Consensus” set of policies. Nor will the group try to address questions such as the appropriateness or impact of IMF conditionality in general. Second, we will base our conclusions on two types of evidence: (i) cross-country evidence on what actually happened to health spending and fiscal targets in countries with IMF programs; and (ii) in-depth case studies of how recent IMF programs in a small number of countries have interacted with health sector budgeting, drawing upon a detailed review of IMF and country documents and interviews with key stakeholders. Learn more about CGD’s Working Group on IMF-Supported Programs and Health Spending including working group terms of reference and bios of working group members.

Q: Much of the recent migration debate in the U.S. has focused on the impact of illegal immigrants--or shall we say undocumented workers--on the U.S. economy. What about the impact of migration on the sending countries? Is there a consensus among development economists about whether emigration is good or bad for development? A: Ever since mass emigration from Ireland in the 19th century caused wages there to rise, it has been clear to most economists that the departure of low-skill workers largely benefits the sending country (see, for example, Migration as Disaster Relief: Lessons from the Great Irish Famine). A UN balance sheet of the economic effects of migration found that movement of low-skill workers offers sending countries many benefits and at essentially no cost (see the World Economic and Social Survey 2004: International Migration (PDF) ). But there's a raging controversy about the effects of emigrating high-skill workers. People worried about so-called brain-drain argue that skilled migrants take with them not only their abilities but also public investments in their education. But some skilled émigrés return home, many send money, and many assist in technology transfer and the creation of trade and investment networks. Their departure may even change the education decisions of those remaining behind--for example, by causing people to study harder and invest more in education. Some economists claim to show that the net effect of skilled worker emigration is actually positive (see Brain Drain and LDC’s Growth: Winners and Losers). Q: Compared to the current arrangements, would a U.S. guest worker program have any effect on the development impact of migration? A: There is likely to be little development impact of such a law on the migrant-sending countries, because the primary effect will be to regularize a small fraction of those already moving rather than to change who moves. There are 11 million undocumented immigrants in the U.S. now, with another 700,000 arrivals every year, but various proposals foresee caps of just 100,000 to 400,000 on the total number of visas--most of which will be used by people who would have migrated anyway (see the recent brief (pdf) from the Migration Policy Institute). If there is to be a development effect from any new law, it might come by making immigration somewhat more politically palatable to U.S. citizens (e.g. by shifting workers' medical expenses from tax payers to employers) thereby preventing a backlash against new immigration down the road. Q: Within the development community, there is a lot of interest in remittances. By one estimate, remittances from the U.S. in 2004 were $47 billion, more than twice the nearly $20 billion that the U.S. provided in official assistance. What is known about the impact of remittances on development? A: In El Salvador and Honduras, remittances exceed 15% of national income, and the positive effect of remittances on health and education in recipient households has been documented in several countries (see, for example, Remittances, Household Expenditure and Investment in Guatemala). But while remittances exceed foreign aid globally, this is not the case in most low-income, migrant-sending countries. Sub-Saharan Africa gets about $8 billion in remittances per year but gets more than $25 billion in aid. Many of the countries sending the fewest migrants--and thus getting the lowest levels of remittances--are also the poorest. Q: Your own research has been focusing on the causes and effects of skilled migration. What is your hypothesis, and how are you going about testing this? A: I study the consequences of skilled-worker migration on developing countries of origin. To make the question manageable I focus on one profession and one region: African doctors and nurses. I have measured the movement of health professionals out of 53 African countries, allowing me to observe the relationship between emigration and health outcomes across countries and over time. My hypothesis is that many African countries have both very poor health and increasing emigration of health professionals because of underlying problems with their health systems--not because of migration itself. I recently returned from Kenya and Rwanda, and all the top health policy officials I spoke with in these countries agreed with this basic premise. Q: Give Us Your Best and Brightest: The Global Hunt for Talent and Its Impact on the Developing World by Devesh Kapur and John McHale suggested four policy options for rich and developing countries so that the benefits of migration are more equally shared with the developing world. One proposal is compensation: that rich countries compensate developing countries for the loss of skilled workers, for example, by headhunters paying a fee to the sending country, or that developing countries impose an exit tax, to recoup the costs of educating skilled workers who leave. In your view, does such an approach make sense? A: The answer will surely vary depending on the case, but it is not at all obvious that this is appropriate. Consider migrant health professionals. The only academic study to focus on the remittances of these workers measured how much money is sent home by Tongan and Samoan nurses in Australia. The study found that the typical nurse sends home well over U.S. $2,000 every year, regardless of whether he or she has been in Australia for 5 years or 25 years. So Tonga and Samoa get back far, far more than they spent to educate those nurses. These developing island nations experienced a large financial gain--not loss--due to their public investments in departed nurses. Why should those sending countries be compensated for having made an investment with a huge financial return? Certainly there are complexities of the domestic distribution of those resources, but it is not clear that those are best handled through the blunt instrument of international government-to-government compensation checks.

John Nellis, co-editor of Reality Check: The Distributional Impact of Privatization, answers questions about the book and his work's implications for policymakers: Q: Are there any generalizations that can be made about recent privatizations in developing countries? Do any interesting groupings emerge? A: The most contentious privatizations are those in the utility sectors such as telecommunications, transport, energy and, especially, water. These are the areas in developing countries where it is more difficult to involve private providers, where the chances of something going wrong are higher, and where errors produce large, negative social consequences. The much more numerous privatizations of commercial, industrial and service firms in developing countries have not generated as much political debate; and they have been generally judged as beneficial, even in low-income, institutionally-weak countries. In "Reality Check" we tried hard to reach conclusions based on data and reasoning, rather than on pre-conceived ideas. The big, well-documented conclusion that comes out of this approach is that the privatization of infrastructure - particularly evident in the best-studied Latin American cases - has NOT contributed to the continuing poverty and maldistribution of income. This is an important if modest defense of privatization. A stronger defense would be that privatization was closely associated with improved growth and reduced poverty and inequality. We did not find that, however. Regarding regional differences and groupings, our findings are much more solid regarding Latin America than in the Asian or post-Communist transition cases examined. Partly, this is due to more and better quality information on the Latin cases. Still, we see more than enough in the transition country cases to make us think that privatization did contribute to the massive increases in inequality and poverty seen in that region since the demise of Communism. Q: Is there enough evidence to judge the respective merits of moving quickly to privatize vs. a slower approach? A: Yes, economically speaking, slower is better. The arguments for rapid privatization have been mainly political (e.g., if we Russian reformers do not sell off everything quickly the communists will return and defeat the transition); financial (if we do not find a private investor we will have to fund the railroad ourselves and there is no money in the budget for this); or some combination of the two (e.g., if we do not complete the privatization of X, then we will fail to comply with Bank or Fund conditions and a desperately needed injection of funds will be delayed at best, eliminated at worst). These arguments have often turned out to be spurious (though we must admit they rarely seem so at the time they are put forward!) Russia, for example, with a more cautious pace of sales and transfers, and earlier and greater attention to the creation of legal institutions, might have avoided some of the more egregious injustices in its early privatization program. Our conclusion is that more time is generally available than is thought, and that the costs of speed outweigh the benefits. Q: Can you highlight a few examples of developing countries that have managed privatization well? What did they seem to do right? A: Chile has done a good job of privatization, particularly in infrastructure. In line with the conclusion above, the Chileans took their time. Years were spent on building regulatory bodies and subjecting the natural monopoly firms to regulatory supervision, well in advance of ownership change. Ultimately, they decided that private ownership was still required to tap private capital markets for network expansion. But by the time the new private operators came on board, the Chilean regulators were experienced and ready to deal with them. Q: What do you think is new or important about this book? A: We've already mentioned the important finding that privatization is, in Latin America at least, and probably in other non-transition regions, not the big social villain usually portrayed. The reason is that, in many countries, privatization has resulted in more poor people obtaining electricity and phone services, and to a lesser extent, water. The downside to getting connected to the network is that customers have to pay, and these costs can be high. A major finding of the book is that, in terms of distribution, the effects of increased access are more important than the effects of increased costs. That's a huge plus. Q: What, if any, policy recommendations would you have for rich countries trying to help developing countries through the privatization process? A: "Reality Check" focused much more on what is happening in the case countries as a result of privatization, rather than on how the countries' privatization policies were or were not influenced by donor countries. Nonetheless, the findings offer some guidance for rich countries wishing to assist the process and minimize the potential negative effects on distribution. First, and most obviously, don't rush to sell the infrastructure firms. A corollary is take the time to enact and perfect regulation prior to sale. The stronger and more competent the regulatory system in place, the higher the likelihood that the social impact of an infrastructure privatization will be positive. Second, some sectors are easier to privatize than others. For example, private operation of telecommunications' companies is much less problematic than in water, due mainly to the enormous differences in the technology of delivery systems in the two sectors. Private telecom companies are now the world norm; in water, in contrast, there have been a number of cancelled or disputed contracts with private providers. Electricity falls somewhere in the middle.