Syndicate content

Worrying Too Much about Brain Drain?

David McKenzie's picture

Brain drain worries policymakers around the world. For example, a search today in Google News gives a host of stories in the past month alone concerning efforts by universities in Vietnam to stop brain drain, demands for wage increases to stop the brain drain of doctors in Pakistan, claims that Malaysia’s brain drain hinders its economic progress, efforts to stem brain drain in Jamaica, a plea to “stop the brain drain” in Cyprus, and even fears of massive brain drain from the state of New York.

But does high-skilled emigration really pose such a threat? The last five years has seen a surge in empirical research on the subject, which John Gibson and I use to answer eight key questions about brain drain in a paper forthcoming in the Journal of Economic Perspectives and now out in the World Bank working paper series.

The 8 key questions addressed are: 1) What is brain drain? 2) Why should economists care about it? 3) Is brain drain increasing? 4) Is there a positive relationship between skilled and unskilled migration? 5) What makes brain drain more likely? 6) Does brain gain exist? 7) Do high-skilled workers remit, invest, and share knowledge back home? and 8) What do we know about the fiscal and production externalities of brain drain?

Some key findings:

  • Brain drain is mostly not about doctors and nurses – scientists, engineers, IT professionals, and academics are the most common occupations, while health professionals only account for 5-15% of high skilled immigrants in some key destination countries. Moreover, the rate of emigration actually is lower for doctors in most developing countries than it is for the average tertiary educated individual. And contrary to popular discussion, hardly any PhDs are taxi drivers!
  • Although brain drain has been increasing in absolute terms, the rate of high-skilled emigration has been flat over the 1975-2000 period and likely fell over the past decade. The reason is that tertiary enrolment rates have grown at a faster rate than emigration in most developing countries.
  • Skilled migration and unskilled migration seem to go hand in hand: So the idea that countries can stem high-skilled emigration while promoting more opportunities for less-skilled individuals to migrate does not meet the empirical reality.
  • Existing empirical evidence on “brain gain” shows the incentive to migrate does increase education at upper secondary and perhaps lower tertiary levels – but evidence is much less clear for higher skilled professions. One key factor is how free supply of education is to respond to demand – so Government limitations on private education and restrictions on the number of University places available for doctors, engineers, and other professionals will reduce the likelihood of brain gain.
  • And finally, for our finance readers – there has been much debate about whether the high-skilled actually remit or not. Recent micro data shows the high-skilled do remit, especially when they come from poor countries (see Figure):

Recent evidence should counteract many of the myths and common concerns about brain drain. Brain drain rates are not skyrocketing. Africa is not the most affected region for brain drain; small island states are. Skilled migrants enjoy massive increases in their living standards as a result of emigrating, and skilled migrants end up remitting back as much as the fiscal cost of their absence.

Ultimately then, given the massive individual gains from migration, any belief that brain drain is detrimental for development must rely on large externalities of high-skilled individuals. This is the area where the existing empirical literature is weakest, but the estimates that do exist suggest that the production externalities of brain drain (at the margin at least) are quite small.

The paper suggests then that worries about brain drain are likely to be overblown, but also that there are still large knowledge gaps on certain impacts, and almost no rigorous evidence as to the impacts of policy actions in this area. Given how prevalent brain drain concerns are, there seems to be plenty of fruitful avenues for future research and policy experimentation going forward – with the paper ending with some ideas on directions for these.


I have a few concerns about this point of view: 1. I think comparing remittances with the fiscal cost of brain drain is the wrong calculation. We should be looking at the long-term change in living standards in the home country that depends on the presence or non-presence of a skilled worker. This doesn't just take into account the worker's own production but also that of the future generations who, as a result of the skilled worker's emigration, will not live in the home country. 2. Rich countries are becoming better at cherry-picking the "best" potential migrants. The highly skilled migrant programs in Australia and the United Kingdom, for example, are greatly increasing the incentive for brain drain to occur without doing anything to address its costs. 3. I think the externalities associated with the presence of highly skilled individuals in their home countries can be quite big. It's not just a question of sharing knowledge and other typical spillovers. These are also the people who can help to form the professional middle class that serves as the anchor for economic and social development. Without them, it will be hard for poor countries to solve their own problems; they'll continue to depend on technical assistance from organizations like the World Bank.

Submitted by David on
Thanks Daniel, On point 1), the fact that future generations will no longer live in the poor country, but grow up in a rich country should be considered a massive plus I would think. Taking into account the increase in welfare of future generations in this way would increase the benefits of high-skilled emigration. Your point 3) is one that is often made, but very hard to quantify. Given the large numbers of skilled migrants that return with knowledge and experience of how institutions can work better abroad, high-skilled emigration can actually support this professional middle class in some cases.

Hi David, On (1), you seem to be maximizing a global social welfare function rather than the social welfare function of the country that's losing people. Surely the latter is the one that the country's government should and will use when making policy. On (3), just because something is hard to quantify doesn't mean that it should be ignored. Also, I'd like to see some statistics on the shares and ages of returning emigres. Are they coming back in numbers? If so, is it for an affordable retirement or truly to help their countries secure a more prosperous future? Nice discussing this with you.

Thanks for this discussion, David and Daniel. Daniel, my personal starting point is that I would feel a grave harm had been done to me if government agents informed me that I was restricted to live only in the place (city, state, or country) where I was born, and that my choice of where to live had been forcibly taken from me. I think most people feel this way, which is why the UN Universal Declaration of Human Rights confers an unqualified right to emigration on all people, skilled or not. We should thus demand a high burden of proof from anyone who feels that this right should be curtailed by any government. Meeting this burden of proof requires answering a long list of questions that we don't now have good answers to. Does skilled emigration necessarily deplete the stock of skilled workers in the country of origin? Rigorous case studies of small countries with very high rates of skilled emigration have cast doubt on this idea (Pedro Vicente and Catia Batista on Cape Verde, forthcoming JDE, and my work with Satish Chand on Fiji, among others). Skilled emigration greatly increases the incentive to invest in skill acquisition, and the net effects are complex. If skilled emigration *does* necessarily deplete the stock of skilled workers, what is the magnitude of human capital externalities? This is extremely poorly understood. Acemoglu and Angrist have trouble detecting any substantial schooling externality in the U.S., and Lant Pritchett has shown that it's very difficult to detect in African data as well. The idea that economic progress results naturally from a group of educated people forced to interact with each other because they can't go anywhere else is one that I would describe as untested. There's better evidence for causation in the other direction: People invest in skill when it has high returns for other reasons including basic institutions of sound political and economic governance. Finally, even if skilled emigration does necessarily deplete the skill stock *and* there are large externalities -- both of which are intuitively plausible but the world seems much more complex than this -- there remains the question of who *owns* those externalities. It is not at all clear to me that people near whom I was born hold a property right in the positive externalities I may or may not convey to others. Why do other Indians hold more of a property right in the external effects of a smart Indian's talents than the Indian herself does? If they own those positive external effects, do other Indians also own the *negative* external effects of a nefarious Indian, so that all Indians should be punished if one Indian commits a murder overseas? What makes the most sense to me is that you own your talents and your shortcomings, and I own mine. This last point regards an assumption about property rights, one that precedes economic analysis. Past AEA president Charles Kindleberger once described this as the debate between those who feel that people should go where they can make the best contribution, and those who feel that people should stay where they belong. I don't think anyone has the right to unilaterally inform you where you belong, and I think the same applies to Indians, Chinese, Filipinos, Malawians --- everyone. I discuss some of these points in more detail in a piece for the Journal of Economic Perspectives that will appear alongside David's, later this year.

Michael, I don't think we should reduce this discussion to moral absolutes. There are plenty of things countries can do to encourage/discourage immigration/emigration without impinging upon human rights. Good examples are the things countries do to try and lure their citizens back to their shores. So let's not throw the baby out with the bathwater - I hope we can still have a meaningful discussion of policies that countries might use to influence the flows of people across their borders, and why (from an economic perspective) they might use such policies. I'm going to look for the papers you mentioned about the effect of skilled emigration on the local stock of skilled workers. I'm curious what the econometric strategies will be, since it's tough to prove a counterfactual when you have a nationwide change in policy.

Thanks Daniel. Here are the two studies I mentioned: The one about Fiji is by me and Satish Chand. There's a video of me presenting it at Stanford, here: ...and the paper is here: The paper on Cape Verde is not mine, it's by Batista, Lacuesta, and Vicente, forthcoming JDE, available here: I didn't espouse any moral absolutes at all in my comment. I said that the right to emigrate is well-established, so if policy is going to curtail that right (i.e. moving away from absolutes) then there should be good reasons for it, which is sensible. And giving good reasons means we need answers to many questions. I didn't say we have those answers (again, no absolutes). I said that we don't have the answers we need in order to feel fine doing definite harm to people by restricting their choices in seeking a better life. I do not think there is a good answer to the question of who owns the positive externalities you convey to other people. I do think that it's not obvious that other people own them. I also think that if society is going to give even partial ownership of your positive externalities to other people, then it also makes sense to give other people partial ownership in your negative externalities as well, so that if you do anything bad to another person, society should be partially punished. Since that state of affairs is something that strikes me as unjust, that makes me question the idea that other people should hold even a partial property right in your externalities -- negative or positive. None of this is absolutist; I'm asking for reasons to change my views. But I don't now see any.

I'm a little confused here. Let's say a government puts in place a policy whereby it offers $10,000 to returning emigres as start-up capital for a new enterprise. This is clearly an incentive to change location, but I can't see how it curtails anyone's rights. The government instituting the policy might think that luring emigres home is a good move to maximize the social welfare of its populace. Anything wrong with that? Or consider a government that's trying to keep its skilled people from emigrating - perhaps it would take steps to make its economy more meritocratic. This doesn't imply impinging on rights at all, as far as I can tell - in fact, it would be a positive move for all the citizens of the country. And yet it would influence someone's decision on where to move. I think there's a difference between telling someone where they should live and influencing their location decision; there's a lot of room for welfare-improving policies that do the latter.

Submitted by Lee on
Great discussion. Anecdotally there is lots of evidence about the positive influence of returnees - such as the number of national leaders educated in the US, or the establishment of the Indian IT industry with support from diaspora - is there any systematic evidence on this?

Submitted by Gabriel Demombynes on
I don't want to speak for Michael, but while he clearly states that he is particularly focused on policies that curtail one's right of movement, most of the points in his first comment apply as well to any policy intended to change incentives to move. Maybe it would be helpful to talk about a concrete policy question. One common case of a policy that curtail's one right of movement is restrictive citizenship policies. Many countries revoke the citizenship if individuals become citizens of another country. I think it is fairly clear that this curtails one's right of movement, by taking something away from people who move and acquire another citizenship. Under its new constitution, Kenya explicitly drops this policy and allows those who obtain another citizenship to remain Kenyan citizens. It is likely that this will increase the gross brain drain, i.e. more educated Kenyans will emigrate, knowing that if they become citizens of other countries they will not have to give up the benefits of Kenyan citizenship. The net effect on the supply of skills of Kenyans in Kenya, however, is uncertain, because the change may also increase the incentives for Kenyans to obtain skills, and also because more Kenyans may return. I wonder if Daniel would argue in favor of restricting dual citizenship to reduce brain drain? Or are there other policies he would propose?

Add new comment