This week we’re focusing on a new book about IMF lending. The IMF – the International Monetary Fund – exists, among other things, to provide policy advice and financial support to governments facing economic difficulties. But are its programmes effective?
The book that we’re discussing in this episode suggests that IMF funding becomes a resource held by local leaders, which those leaders can use to benefit their own supporters to the detriment of the rest of the population.
The book – called IMF Lending: Partisanship, Punishment, and Protest – has two authors, and we are joined by both of them.
- Dr Rod Abouharb is Associate Professor in International Relations here in the UCL Department of Political Science.
- Dr Bernhard Reinsberg is Reader in Politics and International Relations at the University of Glasgow and also a Research Associate in Political Economy at the Centre for Business Research at the University of Cambridge.
Mentioned in this episode:
SUMMARY KEYWORDS
imf, governments, programmes, bernhard, countries, politics, protest, opposition, conditionality, rod, protests, supporters, number, ucl, hypothesis, lending, evidence, argument, approach, conditions
SPEAKERS
Bernhard Reinsberg, Rod Abouharb, Alan Renwick
Alan Renwick 00:06
Hello. This is UCL Uncovering Politics. And this week we're looking at the real-world impacts of IMF lending.
Hello. My name is Alan Renwick. And welcome to UCL Uncovering Politics – the podcast of the School of Public Policy and Department of Political Science at University College London.
This week, we're focusing on a new book about IMF lending. The IMF – the International Monetary Fund – exists, among other things, to provide policy advice and financial support to governments facing economic difficulties. But are its programmes effective?
Well, the book that we're discussing gives reason for doubt. It suggests that IMF funding becomes a resource held by local leaders, which those leaders can use to benefit their own supporters to the detriment of the rest of the population.
The book – called IMF Lending: Partisanship, Punishment, and Protest – has two authors. And I'm delighted that both of them join me here to discuss it.
Dr Rod Abouharb is Associate Professor in International Relations here in the UCL Department of Political Science. And Dr Bernhard Reinsberg is reader in Politics and International Relations at the University of Glasgow, and also a Research Associate in Political Economy at the Centre for Business Research at the University of Cambridge.
Rod and Bernhard: welcome both to UCL Uncovering Politics. It's great to have you both here.
And I think it's perhaps useful if we just start off with a little bit of background here. So many listeners, I suspect, will have heard of the IMF, but maybe don't have much sense of what it actually is. So Bernhard, do you want to start us off just by explaining what is the IMF and what does it do?
Bernhard Reinsberg 01:55
Yeah, of course. Thank you so much for having us.
So the IMF was created, together with the World Bank, in 1944 at the Bretton Woods Conference. And the IMF's role was to oversee the system of pegged exchange rates of member governments and to provide a contingency reserve fund on which countries could draw on in case of balance of payments problems.
Today, the institution has over 190 member states and has three main functions. First is the macroeconomic surveillance of members' economic policies. Second is to provide technical assistance on fiscal affairs, for instance, but also many other aspects of policymaking. And third, and most controversial, to provide loans for countries in economic trouble.
Alan Renwick 02:45
So can you tell us a little bit about the circumstances in which countries end up getting loans from IMF?
Bernhard Reinsberg 02:53
Absolutely. So countries can borrow up to the quota and can go beyond that quota if needed. And this is often to do with excessive fiscal deficits, excessive debt that, you know, prevents countries from meeting their payment obligations, often denominated in foreign currency. So the IMF provides hard currency – fresh capital – when these countries need to repay their debt obligations.
Alan Renwick 03:23
How often does that happen? Do we have a sense of how many countries are currently in receipt of IMF loans?
Bernhard Reinsberg 03:29
There was a heyday of these programmes in the mid '90s when we had the Asian financial crisis, where we had about 68 active programmes. And this number has come off slightly, so to about 40 to 50 in the most recent decade. But currently we are also in that range – between 30 and 40 countries – and the numbers are projected to rise again, as again many countries are suffocating from the massive debt they have accumulated over the course of the COVID 19 pandemic.
Alan Renwick 04:08
And Rod, maybe we should bring you in at this point.
A concept that is relevant for this funding that comes up in the book is the concept of conditionality. Do you want to just explain what that means?
Rod Abouharb 04:21
Sure. Conditionality has been something that has been associated with IMF lending from its very beginning. And essentially, the idea is that, you know, governments who want to kind of borrow above the ceiling that Bernhard was talking about will... There's no such thing as sort of a free lunch, right. So the IMF will demand a range of economic reforms from these member states in order to receive these additional funds.
And these reforms are sort of broadly labelled 'conditionality', but they can range from things like, you know, the privatising of a variety of state assets; they can include, you know, potentially increasing your tax base or having to engage sort of austerity cuts.
And it's usually with a view to kind of balancing the budget. And as Bernhard was saying, to kind of pay back your debtors.
And the underlying principle is that these economies will function much better once they're sort of brought back into balance.
But also I think the other thing that is important to think about with the conditionality is that it's not neutral in its approach. So there is an idea that a smaller state in general is better. One where the state tends to have less involvement in society is better.
And from the, you know... The economists who defend this approach say that it really sort of minimises the probability of corruption, of inefficient state intervention in the economy. So there's a sort of underlying normative set of arguments that economists make about why they should be handling these sort of debt crises in this particular way.
Alan Renwick 06:24
I'm interested that you're saying that policy direction comes from economists, I guess. So I know very little about the IMF. But I guess my hunch would have been that there are power structures in the IMF, and that there are some member states that are more powerful than others, and that those might be kind of directing the policy stance that is taking them.
To what extent is it a kind of technocratic process? And to what extent is it more political and shaped by the interests of the member states?
Rod Abouharb 06:52
I mean it's clearly both. So I don't want to kind of, sort of, overemphasise that.
And we know, you know, there is this kind of long-standing kind of informal agreement that the head of the World Bank is an American, the head of the IMF is European, right. So politics is clearly heavily involved here.
And politics is often also involved in terms of who gets loans and the types of conditions that they have – sort of how stringent they are or not. So are they allies of the US or the UK? And there's a long line of research that looks at the politics behind the lending.
But the economics behind it is also something that has changed over time. And, you know, I think both matter here. But perhaps the clearest example that you've seen in the sort of the change is: there are a number of sort of economic models that they tried to use in the sort of '50s and '60s, primarily at the World Bank, in terms of sort of building large pieces of infrastructure and these other things, and that didn't seem to work very well. So partly driven by that, partly driven by a normative push that you saw in the US and the UK with sort of neoliberal economics taking your forefront, that you actually saw the content of what the bank can fund do change over time.
But that was also a function of who they employ. So there's a long line of research looking at, you know, the fact that if you did economics in Chicago, you got hired, right. If you did heterodox economics elsewhere, you probably weren't going to get hired. So there was a particular ideological direction of certainly the IMF. I think the World Bank is probably sort of broader in its approach to to the sort of development issues.
Alan Renwick 08:53
Interesting. Bernhard, do you want to come in on this at all?
Bernhard Reinsberg 08:56
No-
Alan Renwick 08:57
Okay. Sure. In that case, I will go into another question.
I should have said, if ever you do want to come in and just stick your hand up and you can come in.
Okay, so that's really interesting. So we'll get onto your book in just a moment. But before we do that, it's useful just to understand what we already know from the existing literature. And I guess particularly what we know about the impact of IMF programmes and IMF lending.
And Bernhard, do you want to tell us a little bit about that?
Bernhard Reinsberg 09:24
Yeah. So certainly there is great interest in the social economic consequences of these IMF programmes. So of course economists have often evaluated the IMF by what it's meant to achieve, namely, to stabilise these countries, which I agree is sort of very important dimension. But here there is no consensus really. So whether or not the IMF programmes help countries to get back on track with economic growth is still an open question. Some findings say yes, others say no.
For sure we do have evidence that IMF programmes do not help countries avert future financial crises. In fact, one can show that IMF programmes increased the moral hazard for some countries at least, and with those that are politically well connected to the major shareholders, and that this moral hazard creates reckless behaviour in terms of, yeah, self-insurance with reserves and more reckless economic policies.
Alan Renwick 10:31
Great, thank you. So that's very useful context for your study.
Rod, do you want to explain just what exactly it is that you're focusing on in this book? What's the core question that you're trying to answer?
Rod Abouharb 10:44
So I think the way we began thinking about this book, Bernhard and I had a nice chat during the COVID lockdown. I was sitting in the back garden at that point in time and I think Bernhard was in some sunny location in Glasgow.
And what really struck us about the kind of the existing literature, for me there are a couple of puzzles. So one is that, as you've seen, overwhelmingly kind of negative effects of these programmes on society at large, sort of worsening all sorts of areas of inequality. But also some evidence that politicians stay in power longer when they go into these programmes. And that, to me, seems like a puzzle. So there's austerity, there's protest, there's poverty, but somehow politicians aren't being thrown out. So there's something going on there. So I think that's one puzzle.
And then I think the other two things that really kind of grated on me, and that's because I'm a political scientist and I think about how governments retain power, rather than an economist. So there are some very well-worn arguments in economics about who gets listened to. And these are primarily the groups that are most well organised. And so this is the literature that we know well. But there's also, I think, a tension there by saying, well, look, if government simply listened to those groups who are well organised, then how do they respond to their voters, because they need their voters to stay in office. And I think that's where this interest group approach really underplays the role of governments as political actors in the distribution of pain and benefit from these programmes.
So that's where our argument comes from. And so we think that when governments, you know, face these difficult economic situations, they go to the fund for help. They don't have that much choice about what they can do. So they really have to implement the agreements that they may.
But nevertheless, these increments have discretion. And we think governments use that discretion for their own political advantage. So they will cut spending in opposition areas and they will try to protect their own supporters. So these kind of inequality outcomes that we see, and the fact that inequality worsens, were able to explain who is this being done to.
So in these societies not everyone is in this together. We find pretty strong evidence, as I'm sure we'll talk about, that the opposition is hit hardest here. And then it's the opposition who protests. So I guess the value of what we're doing is it really tries to kind of unpack at the individual level what the consequences of these programmes are.
Alan Renwick 13:59
Great, thank you. So there are kind of two core hypotheses, if I understand correctly, that are coming from that detailed thinking.
So one is the hypothesis that the costs associated with IMF programmes will be concentrated upon the opponents of governments – the opponents of those in power. And second, that that will lead to protest on the part of those people.
And so what you're doing in the book is essentially testing those two hypotheses and seeing whether that's what's happening.
And there's also an argument around the conditionality that we were mentioning there earlier, isn't there? And so I was quite curious about this, because the argument is that the more conditionality there is associated with IMF lending, the worse this problem becomes. And you might think that it would get better – that somehow the conditionality would keep the government on the straight and narrow. But actually the hypothesis is that it makes things worse. Do you want to explain that?
Rod Abouharb 14:55
Sure. And the logic behind that is really that, with all these conditions that governments need to meet, is that oftentimes there is discretion on how you meet the condition.
So, you know, some examples from some of kind of the case illustrations that we have in the book: oftentimes, governments are faced with demands to restrict the size of the civil service.
We saw this in Kenya, we saw this in Ghana. They'll be told, well, you need to reduce the numbers of your civil servants by 20,000, 30,000.
But there aren’t any specifics about who is cut or where the cuts take place. And it's in these areas where, you know, governments have the most control. So particularly in the public sector, that we think you see this kind of distributional politics taking place.
You know, so if the government has the option to close a civil service in a government held constituency in comparison to an opposite opposition held constituency, we think they'll close the one in the opposition constituency.
Now, demonstrating that is really difficult. So, you know, governments don't go around advertising this. And they often don't go around kind of recording this type of information.
But there are a number of observable implications for things like that, you know. So it may well be that when they go under these programmes that actually opposition fares worse in terms of employment, in terms of access to food, in terms of their evaluation of their life chances going forward and these other types of metrics. And if there's a significant difference between how the opposition view that and how the government supporters view that, then we know there's something going on here.
And then I think the really strongest set of tests is: what did those differences look like when states are not under IMF lending? Do you see this – these significant differences – are not?
So it's not that we don't think distributional politics isn't in play in periods before. But we actually think it's much worse when governments go under these programmes for these two reasons.
So one is that they have this sort of curtailed fiscal envelope that they have to deal with there, right. So there's cuts, they have less money. And now they have to choose what to do with this constrained amount of money that they have. And we think that they will try to benefit their supporters first.
So given that, someone has to pay for that. And we think it's the opposition that pays.
And it makes sense, right? Why not? They're not going to vote for you anyway so why not burden them?
Alan Renwick 17:50
So really interesting. I'd like to pursue that further in just a moment. But just before we do so we should perhaps just be clear about the methodology of the study. And you've started to talk there a bit about the evidence that you draw on.
And Bernhard, do you want to take us through the logic of the methodology a bit more. So you've got both qualitative and quantitative evidence that you're drawing in here.
Bernhard Reinsberg 18:13
Yeah, indeed. We do combine qualitative and quantitative evidence. So on the question of how people perceive these programmes under different conditions, we first look at case studies, particularly the case of Ghana.
And this allows us basically to overcome one limitation of common survey datasets, which is that they don't have repeated observations, right. So here we can really draw out how different governments, when facing similar demands from the IMF, implement these conditions, with a view to supporting their own – to rewarding their own supporters and punishing opposition supporters.
This was the case in Ghana, as we said, from when we see the transition from the Rawlings to the coup for administration. I'm sure Rod will be able to tell you a bit more about that. So this was sort of our thinking here to use sort of an intertemporal, or over time comparison, of the case of Ghana.
We used the same method when we look at the case of Kenya, where we draw out the differences in who protests across these two administrations.
Now, we complement that because, you know, with singular case studies it's often difficult to generalise, right. So here the strength of the large-N analysis comes in, and we mobilise evidence from four global barometers with broadly consistent findings actually.
Alan Renwick 19:58
So the barometer you refer to here are to multinational public opinion surveys.
Bernhard Reinsberg 20:04
Exactly. So firstly, the Afrobarometer from 1999 to 2001 – a heyday of structural adjustment – and the Asian Barometer from 2005 to 2008, and the Latin American Barometer from 2005, as well as the global longitudinal World Value Survey dataset.
What is our approach here? We have a staged approach really where we first try to show the general patterns in the data between opposition supporters and government supporters with respect to their valuations of these IMF programmes and protest behaviour.
Now, this can of course be an ongoing feature of these societies. So that alone doesn't tell us much, which is why we also bring in this additional aspect of whether or not countries are under an IMF programme. And it's precisely in those cases that we see an intensification of the distributive politics.
We also exploit variation in the design of IMF programmes, finding that these programmes with more conditions, which leave governments more discretion to lump adjustment burdens on the opposition, also see a greater divergence between opposition and government supporters.
Alan Renwick 21:30
So you've kind of answered the questions that I was about to ask you there, which was that: so the hypothesis is that the costs of IMF programmes will be concentrated upon government opponents. And in order to test that you're looking at public opinion. And you're seeing that opposition supporters are less likely to be supportive of the programme. But of course, you could say, think, well maybe that's just because they're opposition supporters and they're not going to like what the government does. So it might not be an indicator of objectively what's actually going on.
But what you're saying there, if I understand, is that crucially we find this effect only where the IMF are involved. We don't find the same effect where the IMF are not involved – where there is some kind of crisis and there's a need for structural adjustment on the part of the government, but the IMF are not involved. And you're saying, in those cases, we don't get the problem. So it seems to be specifically where the IMF is involved that we get this effect. Is that correct?
Bernhard Reinsberg 22:32
Exactly. Indeed. So of course this is a super difficult challenge to untangle these perceptive biases with the real effects that these programmes may have. But we undertake several empirical strategies to circumvent this challenge.
So just looking at the outcome, for instance, that the structural adjustment programme made my life worse, right, we do find, of course, baseline differences between different partisans. But this difference increases when country is under an IMF programme. So that gives us some indication that it must be related to the problems that come along with these IMF programmes and with the increased pressure for distributive politics that governments mount in these cases.
So just to give you an idea of the numbers here, the baseline difference is 15 percentage points, but it is 20 or 25 percentage points under an IMF programme between the two partisans. And once you look within those countries that already have a programme, you will find a 10 percentage point difference between the programmes with few conditions and the programmes with many conditions. So again, further intensification of this effect here in the case of the Afrobarometer.
Now a second very convincing test that we do, at least from our perspective, is that we look at material outcomes. So of course, these are perception-based measures that we typically have in the surveys. But we have no indication that people would lie to the interviewers when asked about whether they've faced material hardship, such as going without food, going without medicine, or going without income.
And here we find, again, very strong effects. So about 2.2 index points on a 1-to-12 scale.
If a country has third tests, we also show that these results hold along ethnic lines. So we would expect this, given what we know about African politics. This is also a finding that is consistent with the distributive politics hypothesis that we advance.
Alan Renwick 24:53
That's all on the first hypothesis. Rod, do you want to tell us what the evidence is with regard to protests, and the impact of these patterns upon protest?
Rod Abouharb 25:03
Sure. So as Bernhard was saying, distributional politics is sort of not taking place outside of these financial crises, but we think that the Fund makes these issues worse.
And you see that with respect to protest as well. So, you know, as the number of conditions increase, so does the protest.
And this fits well with our argument that government is burdening opposition with the costs of these programmes. And then people naturally protest that, right, because they don't think it's fair, because they see it.
And, you know, there's evidence, you know, with respect to the case studies as well. So, Kenya is a kind of a really good example.
As Bernhard was mentioning, we have this sort of intertemporal approach where we examine countries that were under IMF lending, but the government changes from one type to another. And then with the change in government, even though the loan conditions remain almost exactly the same, the behaviours of the government changes.
And so you see this in Kenya: under the Moi administration, he is backed by kind of a relatively narrow section of the population, the Kalenjin ethnic group. And he seems to be okay with people protesting him, as long as it's not his voters.
So you see, these types of protests erupt because of the cuts in public sector spending, in the civil service, that he makes. And then, for example, for the civil servants that are left, they get pay increases. So even in times of austerity, there is the concentration of wealth in particular areas.
And if you look at where the protests were taking place in Kenya, these protests were taking place in opposition held areas of the country. And oftentimes, these were civil servants that hadn't been paid or had faced sort of a wide variety of cuts, whereas other civil servants seemingly weren't protesting. So there was a real sort of geographic distribution in the protests that took place.
And then if you compare, let's say, the Moi administration with the Kibaki administration that followed next, his support base was much broader across Kenya. And actually what you saw, because of that, there was a deep reluctance to actually engage in any cuts precisely because he was worried about the consequences of a protest.
And in some cases, this is understandable. In other cases, you saw reforms that probably would have benefited the Kenyan economy then not happen. So you saw arguments about kind of liberalising the maize sector not taking place because it was a locus of support for him, even though Kenya was dealing with all sorts of kind of food insecurity, because the maize sector of the economy was relatively close to that point.
Alan Renwick 28:40
Okay, so we're getting close to the end of our time, alas. But there are two really important questions that I still want to get in.
One of them is still about these results. So I understand the broad picture and I understand the argument and the evidence. And that seems very clear. I guess what I don't yet get is: why is it that the IMF is actually making things worse here? So you're comparing countries that go into IMF lending with countries that also have crises and also need to go through some kind of adjustment programme but don't have the IMF lending.
And why is it not the case that those... Those countries also have governments that try to concentrate the costs among opponents. Why is it that we're particularly seeing this effect among those countries that need to go to the IMF? Bernhard, do you want to start on that?
Bernhard Reinsberg 29:29
Yeah. So one might think that the IMF reduces discretion for governments, which is a popular argument in the literature. But in fact we show that the opposite is actually true. And this is because there are certainly different layers of discretion that governments have, right, and once they face more demands from the IMF, more conditions, they can also lump more burdens upon opposition supporters. So in some sense this is an idea window of opportunity to speak from a perspective of a politically motivated leader to, yeah, to play these distributive politics out in a much greater extent than they would otherwise be able to do.
Alan Renwick 30:16
I guess that leads on to the final question. Rod, what are the policy implications here? What should the IMF be doing differently as a result of the analysis that you present in this book?
Rod Abouharb 30:28
Well, it's a really good question and we've thought a lot about this.
And I think one of the problems here is there's plenty of blame to go around. So if you like, the local political economy story is about politicians acting like politicians, right. They're sort of fairly badly behaved; they want to make this for their own advantage. And some people within their societies benefit from that, but a lot of people don't.
And, of course, then the problem is that if you then reduce the discretion in these loans, which is the obvious thing to do, and being kind of much more prescriptive about what these governments are allowed to do in order to receive these this funding, I think the politics of this then becomes one of claims around sort of neocolonialism and these other types of claims.
You've seen this made by Kenyan politicians and others. There are a number of episodes where the IMF wanted them to put through anti-corruption bills into parliament, because they knew large amounts of money was being syphoned off by these corrupt politicians. And then this gets framed as some kind of post-colonial enterprise by the IMF against developing economies.
And so this is very, I think, politically very difficult. I think we can identify what is going wrong here, in terms of the discretion that then politicians are using for their own advantage. But whether or not that's politically feasible to remove that discretion, because then governments won't go to the IMF for loans – they'll go to the Chinese for loans, right, they'll go to other actors for loans.
And that also has political implications. You know, we know that this funding is used for kind of soft power purposes, as well. And so if the US and its allies lose that opportunity, that political opportunity, with this institution, and it goes elsewhere, then is that something that they're willing to support? Even if it means less corruption, less inequality, less protest.
Alan Renwick 33:00
That's fascinating. So it's a real policy quandary facing the IMF there.
Well thank you so much, Rod and Bernhard. Alas we have to finish it there. But it's been a really interesting conversation. I confess that when I saw that this topic was coming up, I thought, gosh, that's quite dry. But actually, it's incredibly important, and it's really fascinating to hear you talking about the book.
So we have been discussing the book IMF Lending: Partisanship, Punishment, and Protest by Rod Abouharb and Bernhard Reinsberg. It will be out imminently, published by Cambridge University Press as part of the Cambridge Elements series. And it's available now for preorder online.
As ever, you will find all of those details in the show notes for this episode.
Next week is reading week here at UCL, so we'll be taking a short break. But we'll be back in two weeks’ time to discuss some intriguing new research on the philosophy of language. Can you, for example, make a promise if the person you're speaking with doesn't recognise that you have done so?
Remember, to make sure you don't miss out on that or other future episodes of UCL Uncovering Politics, all you need to do is subscribe. You can do so on Apple, Google Podcasts or whatever podcast provider you use.
I'm Alan Renwick. This episode was produced by Alice Hart and Eleanor Kingwell-Banham. Our theme music is written and performed by John Mann.
This has been UCL Uncovering Politics. Thank you for listening.