The Transatlantic at INET Conference in Cambridge
Part 1: «The World Bank is a very stubborn animal and it’s difficult to change» – Branko Milanovic
Branko Milanovic is Lead Economist at the World Bank’s Research Unit on Poverty and Inequality and Senior Associate at the Carnegie Endowment for International Peace in Washington D.C. His research focuses on income distributions, democracy and globalization. Since 1996 he also teaches as adjunct professor at the School for Advanced International Studies at John Hopkins University. Ronan Sato from The Transatlantic met with Branko Milanovic at the INET Conference in Cambridge.
Professor Milanovic, you have been quite critical of the lack of attention paid to inequality in mainstream economics today. In particular, it seldom enters the analysis of economists who are not specialists in the area. What’s wrong?
It is very true that mainstream economics has not paid much attention to the issue of inequality. Two reasons come to mind. Firstly, and most obviously, economists find it difficult to integrate inequality into their models. Not because it’s difficult to measure, but because the role it plays in people’s economic lives is not as well studied. However, inequality is recently undergoing a bit of a revival, due in most part to increasing inequality in some of the rich countries. One new movement is to incorporate in models behavior that is a function of one’s income-position. This may be one way to get rid of representative agent problems, which would be a huge leap toward more accurate models. But I’m not the modeler so I can’t comment on how feasible this would be. Besides, the difficulty of integration is not the only problem I see. In particular, I suspect many economists harbor a degree of ‘ideological reticence’, a feeling that somehow the issue of inequality is not as central as other questions. The new approaches I mentioned are still very much the exception and whether their work will be accepted in the mainstream is a different issue altogether. Even 20 years ago there were attempts to incorporate the same kind of factors into computable general equilibrium models as part of the social accounting movement, but it kind of fizzled out. And there’s no guarantee that it won’t happen again.
When you referred to rising inequality in rich countries are you referring to the financial crisis? What has been the effect of the financial crisis in your line of work?
Actually I was referring to broader changes, which have been taking place over the last 25-30 years. To be honest, the financial crisis itself has not had a huge impact yet. The key reason is that the data we use, mostly household surveys, have about a 2-3 year lag before they become available from a sufficient number of countries. Of course we do know certain things. We know that China and India have continued to experience high growth rates, and that both Latin America and Africa have maintained overall positive growth. Insofar as data is available, the effects of the crisis seem to be restricted to the rich Western countries and the Eastern Europeans. In terms of global inequality, therefore, the crisis probably hasn’t had much of an impact. In fact, though I don’t know for sure yet, it is possible that global inequality may have declined in the past two years. However, now with high unemployment in the US, UK, France, Spain and so on, it is very difficult to make claims about global changes. In the rich countries, despite the first effects of the crisis being mostly restricted to people with large capital incomes and high salaries, the effects have trickled down and continue to do so. So we really need to wait until we get a clearer picture that will allow us to place what is going on in context of broader changes.
Speaking of broader changes, the one you probably encounter the most in your line of work must be globalization. Perhaps you could talk a little bit about your conception of globalization.
Well my definition of globalization is very standard. Simply put, it’s the increased ability of individuals and companies to engage in trade, move capital around and access new technologies, as well as a general increase in the mobility of the labor force. Now, the effects of globalization on global inequality are very difficult to figure out because globalization affects countries through multiple channels. Firstly, and most obviously, income distributions within nations can be affected. Traditional (Heckscher–Ohlin) trade-theory says that poor countries should in principle benefit more, while the rich should benefit less, but the evidence has not given full support to the predictions based on this analysis. Secondly, globalization should have different effects on the growth rates of different countries. Again, rich countries and poor countries should be affected differently, but here too the empirical evidence doesn’t fit the predictions very comfortably. Thirdly, globalization has different effects on countries depending on their population sizes. This channel, compared to the previous two, has been relatively unexplored. Nonetheless, it seems clear to me that what is going to happen to inequality in a country as it globalizes is largely dependent on these three channels. Which channels are stronger and which are weaker, however, is a contingent issue and we can’t generalize.
In your presentation earlier today, you made it quite clear that you think high growth rates are a good thing. Can you expand?
Often people ask me about that, and I make my position on the issue very clear. It is my belief that if the benefits of globalisation are really going to be as they have been touted since the 1980s, there really needs to be a pickup of growth in the middle-income and poor countries. It’s easy to get carried away by looking at China and India alone because they have been the success stories. And of course from a welfare perspective we have to focus on these two countries, but we should also keep in mind certain other nations which have not done as well as we expect them to. And for these countries growth clearly is the key factor. Both in terms of globalization and reduced inequality, growth is essential. I don’t see any contradiction between these three. They are the most crucial elements of development.
Do you have anything to say about sustainable growth?
Not much actually. I’m in a distinctly unpopular side of the debate. I certainly have a lot of sympathy for the advocates, but they seem to me to be willing to dismiss growth too easily. There certainly is a trade-off at some point, but that trade-off is at different levels for different countries. And it’s my opinion that for poor and middle-income countries the sacrifice we are asking for is too great. So I support the poor countries that are not willing to go along with targets, which seem to be excessive and inappropriate.
You said earlier that issues of justice are very important for you. This seems to me perhaps quite distinct in a discipline that is often considered amoral.
It’s certainly true that when I’m looking at data, ethical questions don’t really play a role. In my line of work, philosophy comes after the work on data. My work is such that once you finish analyzing data, you are immediately faced with important political philosophy issues. Rawls’ 1999 book The Law of Peoples has been particularly influential in this context, because while he doesn’t use the technical terms that I use, he posits certain rules of international justice, which should exist between people across borders. It is an important book to read because the rules he advocates are quite different from what followers of Rawls put forward, whether they be Cosmopolitans like Peter Singer or Statists like Thomas Nagel. I may not agree on all the points, but Rawls’ work in general corresponds very clearly with what I am doing, and I am increasingly attracted to it because I think it’s part of the natural progression. First you work with data, but then once you’ve done the empirical studies, you have to face the issue of ‘what does it all mean?’ and ‘what can we do about it?’
So you believe that economic analysis leads to a horizon where moral judgements must be made? Does this mean morality is actually integral to economics?
Yes, that’s a good point. It’s a good way to put it. I do believe that eventually we have to address these issues. When I teach at university probably about one third of my class is spent analyzing the link between the empirical work and the translation of that empirical work into the world of political philosophy. The relationship between political philosophy and economics is definitely not a pro forma thing. And once you look at the link closely, it’s almost a two way process. Philosophical positions shed light on your empirical work, and your empirical works sheds light on philosophical positions. So I do think that at the end of the day, the horizon has to be certain ethical judgements.
You mentioned teaching – do you have a stance on economics education? Many attendees of the conference are advocating a change in the way economics is taught. What do you have to say?
To be honest, not much. Although obviously there are certain things I would re-do if I were going to study economics all over again. My preference would be for more history of economic thought and economic history. When crises like this one happen, there is suddenly recourse to the past. We suddenly lose sight of what is happening and so we look back for antecedents of what has happened in the past in order to explain the present. The fact that we are here in King’s College, Cambridge, is an excellent example of how this happens. But the problem is that we do so only after the prevalent models break down. When things are going fine, the study of economic history and history of economic thought are too often neglected. It’s also for this reason that I think it’s vital to study classical economists from Smith to Ricardo, as well as Marx. Smith and Marx are particularly important because their approach is very much historical.
What about new movements within economics? Obviously most are in the developmental stages so they may not be fit for incorporation into economics education immediately, but do you have anything to say about, for example, the study of happiness?
Happiness has huge potential implications for economics, in the sense that economics as a whole, until quite recently with Sen and Rawls and so on, was really based on utilitarianism and welfareism. And it’s still there implicitly. But ‘utility’ shouldn’t really be the end of your analysis. In other words, why do we think that income is good? Because higher income makes us happier. My personal opinion is that the link to happiness is as follows: First, there is income. And income buys ‘utility’ in a narrow sense, and that that narrow utility then translates into happiness. Each of these are more abstract stages, and more difficult to work with. This is part of the reason why happiness has such measurement issues. Also while income has no upper bound, happiness is a scale. Your rating on the happiness scale depends very much on your relative position. And one’s relative position is very often calibrated by the individual within a given income distribution at any given time. Most of it is not, except at the very bottom, dependent on the absolute level of income.
You say that economics has been and still is based very much on utilitarianism. But it seems to me that economics has based itself on a very narrow, perhaps even naïve utilitarianism at that.
I would agree, but I would go even further. Rawls’ very clear rejection of utilitarianism was an eye-opener to me because his positions are very different from the ideas you pick up by just doing economics. What I like about political philosophy and how it interacts with the study of inequality is that political philosophy does have something to say about acceptable or even desirable types of inequality. It doesn’t tell you that the Gini-coefficient should be 40 or 20, but it tells you what are the responsibilities of people and what are the different types of societal arrangements and how might they be superior to other arrangements. Once you start to engage in political philosophy, it becomes clear that economic questions have to be put into context. Utilitarianism is a useful way to look at things in economics, but I don’t think this idea that only utility is important stands up to scrutiny.
Your conception of economics seems to me quite distinct from that of most university academics. Is working at the World Bank very different from working in academia?
Well, my experience with academia is limited, but I enjoyed engaging with students. I didn’t have much to do with the academic world in terms of having lots of departmental duties and so on.
On the other hand the research department of the World Bank is different from the rest of the bank because you have much more freedom to do what you want. And in most cases you really don’t have a very strict policy line that you have to follow. So the World Bank is a reasonably liberal place in terms of research. I would not say that it is as liberal as a university setting because universities don’t have a political line at all. In the bank the pressure does exist but it is quite weak and bearable.
Does work at the World Bank have more of a direct link with policies in the real world?
Unfortunately, I would say no. To an extent it depends on the type of research that one does. It could, in areas like education or AIDS, and perhaps even climate change now. But in areas that are very diffused like inequality or global poverty, that link is quasi non-existent. Also, you can’t deny that to a certain extent the World Bank is driven by an ideological stance, which is actually very much dependent upon the work being done at universities. In other words, when your generation will actually go to the World Bank to work, you are going to influence what the World Bank does to a large extent because your ideas will come from your beliefs formed at university. This also implies somewhat of a lag, because many changes in attitudes or ideas only come to the bank with the arrival of new people. So as of now, it’s still a very Washington Consensus type of place.
Are you ok with that?
No I’m not, but I think the World Bank is a very stubborn animal and it’s difficult to change. Joseph Stiglitz is right when he points out that simplicity is both the advantage and disadvantage of the Washington Consensus. It’s a very simple framework, which has been remarkably successful in what it sets out to achieve; the problem is that it has led to too many other problems because it has been a very narrow view of the world. Of course it’s important to focus on key variables, but the context is just as important. In all honesty, I feel it was an uncertain turn that economists took when they started to value abstract theories à la Ricardo (only now put in a highly mathematical language) and in effect discarding work by people like Stone or Kuznets who were very careful empirical economists.
Kuznets is probably the most famous economist to have dealt with inequality. How influential is his work to you?
Kuznets’s work is some of the most careful and data-driven work there is. His work followed an extremely careful and deductive process so that despite the relatively small amount of data he was dealing with, he was able to develop robust conclusions, which have survived to this day. Of course we are not exactly sure if reality is as he described, but the theories have survived extremely well for work that is over half a century old. So he was extremely influential to me in the careful approach to data and the way it is documented and discussed.
So you would like to take the best from Kuznets and Rawls?
Yes. I think that a very detailed and careful attitude towards data work, as well as a consistent view of the world, a Weltanschauung, are what together make for good economics. You can’t do economics without any view of the world. But at the same time, you have to be humble. Unlike many economists historically or now, Kuznets actually thought, “ok let the data tell us the truth”. The reason I am not very enchanted by abstract modeling is that too often, it seems like the models are putting words in the data’s mouth. Perhaps I feel this way because the study of income distribution, by being so data-intensive, manages to be refractive to simple theorizing. When you work with so much data, you realize that understanding what the data is saying is no easy job. You have to really get your hands dirty to understand the data.