Has globalisation deepened inequality?

Paul Johnson

Hello and welcome to this edition of the IFS Zooms In with me, Paul Johnson. And I’m joined today by two experts in trade, and in particular the impact of trade between countries on inequality. I’m joined by Peter Levell, who’s an associate director of the Institute for Fiscal Studies and Penny Goldberg, Professor of Economics of Yale University and formally chief economist at The World Bank.

And we’re looking at the impact of trade in inequality in the context of the big and very exciting IFS Deaton review of inequality, this big review that we’re doing, chaired by Nobel Laureate Angus Deaton. Penny is a colleague of mine on the panel overseeing that, and Peter, co-author of a chapter within that on trade and inequality.

Before we kick off, first of all, apologies for the fact that if you were looking forward to listening to us last week, I’m afraid that illness intervened, so, for those of you who can’t do without your fortnightly dose of the IFS Zooms In, you’ll have to make do with this slightly delayed edition.

So, let’s kick off, I mean we all know I think that the last thirty or forty years have been characterised by what’s been known as a period of hyper-globalisation, with huge increases in trade, particularly between low-income and high-income countries, and particularly of course, trade with China. And that’s led to certain tensions, particularly the US/China trade war and the, indeed big changes in the UK’s relationship with the European Union which of course is by far our biggest trading partner. And all of that has had an impact on all of us, and we’re here particularly interested in the question of trade and inequality.

And let me come to you first, Penny, so to really ask the question about why trade and inequality? We think about trade in many ways, but it’s not the first thing that springs to mind, perhaps for many people when thinking about explanations for inequality. So why, why, why trade and inequality?

Penny Goldberg

The are three main reasons. The first one, Paul you mentioned that this is not the first thing that comes to people’s mind when they think about inequality, but his is actually something that immediately comes to mind, to the mind of a trade economist. Because trade theory teaches us that while trade is good, is beneficial for the average person, trade also creates winners and losers, this is one of the basic insights of economic theory. So, whenever were interested in inequality, it’s very natural to turn to trade as a potential source of this inequality.

Second, as you mentioned, the last few decades have been a tremendous increase in global trade, so especially imports in advanced countries from low-income countries, and especially China. And at the same time, we saw a very large increase in inequality, many, many dimensions of inequality. So, we have this temporal corelation and of course we know that correlation doesn’t imply causation, but a very strong correlation is a flag. So again, it’s very natural to ask the question: are these two phenomenon linked?

And the third reason is that, whether it’s true or not, there is a very wide perception in many countries, in Europe and in the United States, that open boarders whether this is through trade, or immigration, have been responsible for dissemination in real wages, for the demise of unions, for the hardship that workers have faced in these countries. And this in turn has had important political consequences, we saw parties, both at the extreme right and extreme left, exploiting this view, we saw Brexit in the United Kingdom, we saw the increase in tariffs that you mentioned in the United States. In other European countries we see, we saw a backlash against immigration, so again, whether this is true or not, this is something that we need to take seriously. If we think there are benefits to trade and to open borders and we want to keep these benefits, it’s important to also address the potential concerns that people have about the unequal effects of trade.

Paul Johnson:

So, lots of reasons to think that trade benefits us on average, but also lots of reason to think that there’ll be winners and losers from within that. I mean Peter, what do we know about who the winners and losers have been? Particularly in the UK, but in rich countries generally. Let’s start with the winners – who has got better off as a result of this big increase in international trade?

Peter Levell

Well the standard trade theory that Penny talked about suggests that the winners from trade should be, or in particular growth in trade, between poor countries and rich countries, should be, in rich countries there’s skilled workers who aren’t directly competing with imports coming in from low countries which predominantly will be produced using low skilled labour, it will be the skilled workers in those rich countries who benefit from the lower prices but don’t face that increase in competition. That’s what led economists to try to assess the trades impacts on inequality by comparing real wages for skilled workers and low skilled workers.

However, it didn’t look, at least when economists were first looking at these things in the ‘90s and 2000s like trade played an important role in explaining the growth and the gap, the wage premium between collage educated and not collage educated workers. What’s since emerged, a lot to do with research by my co-author on the chapter, David Dorn, is that while it wasn’t, there weren’t broad impacts of trade on different groups, there was some quite disruptive impacts on particular industries and particular regions. So low-income countries, in particular China, specialised exports of particular goods and parts of the UK, United States, other western European countries, both geographically and in terms of particular types of workers, specialised in producing those goods. So, it looks like in terms of the losers, there were highly concentrated groups of workers who lost out. And when you look at the evidence to how these workers fared after that big increase in imports, we see them not doing very well in a lot of dimensions, like sort of employment, on earnings, but also other urr impacts on health and other social outcomes.

Paul Johnson

Just to sort of break that down a bit, don’t the, in one sense everyone gains from trade, because it takes prices down, that’s the first point. But then among the workers there’s a group who, because they’re being, as it were, out competed by urm, people in other counties, they’re the losers?

Peter Levell

Yes, that’s broadly correct. It’s not a foregone conclusion that everyone will benefit from the price reductions you get from cheaper goods to the same extent. So, it might be the case, you know rich people happen to consume more of the goods that are imported than poor people, or vice-versa. When we looked at the evidence on who we felt were benefitting from increases in cheaper imports from the 2000s from China in particular, what we’d found was that poorer consumers tend to benefit more from imports of, of for example clothing and textiles where they tend to spend more of the budget; but richer consumers tend to benefit more from the fact that they spent more on electronics and luxury products and that, which also came form China. And when you compare the impacts across hose different groups, it, it worked out quite similarly. So, it wasn’t the case that those poorer workers who are also tending to lose out from growing import competition, received greater benefits from lower prices than rich consumers

Paul Johnson

And what do we know about the scale of the losses for those lower skilled, less well-off workers? I mean do we know how much, how many of them lost their jobs or how big the impact might have been on their wages? I mean either in the UK or elsewhere?

Peter Levell

In the UK, we did an exercise that suggests that between a hundred and fifty and three hundred and fifty thousand manufacturers were displaced by growing import competition.

Paul Johnson

So, a couple of hundred thousand, something like a couple of hundred thousand people lost their jobs? They have got other jobs, that was the cost in one sense and the benefit was we’re all a little bit better off because we’re paying less for the stuff that we’re importing. That’s an important message I think that these cheaper goods we’re getting came at a cost of maybe up to three hundred and fifty thousand people’s jobs, is that, again is that a reasonable way to look at that?

Peter Levell

Yes, I mean you have to remember this is the degree to which the manufacturing sector shrank that we’ve attributed to import competition, at the upper end of the range it was around three hundred and fifty thousand workers. It was the people whose jobs have been displaced, so as you say they may have found work elsewhere, and that’s, that’s part of the cost of globalisation. So, they may not have found equally well-paid work, equally well-paid work, there would have been a cost of them finding new work.  

But on the other side, there were these benefits to consumers that we also document, and those represent the benefits of increased trade between rich and poor countries. We estimate the scale of those benefits as well, and between the period we’re looking at, between 1999 and 2007 - which happens to be the period of most rapid growth of imports from China to the UK - those benefits amounted to about £400 per family over that period.

Paul Johnson

That’s £400 per year?

Peter Levell

Yeah, so over the whole period, the cost of their basket was £400 less as a result of these cheaper imports coming in.

Paul Johnson

Penny, I mean obviously people have looked at some of this from in the US perspective as well, and indeed our chairman on The Deaton Review, Angus Deaton, has suggested that the concentrated costs of some of the additional trade with China on particular communities has been very big indeed.

Penny Goldberg

I agree with this, and this has been a message that was quite uncomfortable to many economists. To add to what was said before, we all benefit, we all believe, very strongly that trade and openness is beneficial to an economy, and you mentioned the price effects. But in addition to the price effects, which is something that we can hope to quantify more or less, there are many challenges, but still with careful analysis and with a lot of data, we can try to put some numbers to the price effect.

In addition to this price effect, there are also other effects, dynamic effects, longer effects that may be harder to quantify but are nevertheless important. So, for example trade exposes counties to more competition. It forces countries to, to be on their toes, to try to innovate to stay ahead of the game, to stay at the frontier of technology. Without any external pressure, without competition, it’s not clear that this is ging to happen. It’s very hard to quantify this channel but we all believe, based on first principles that at some level it is important that import competition with low-wage countries has contributed to innovation, to technology adoption in the United States and in Europe.

So, no one denies this benefit. But at the same time, we’ve seen displacement, we’ve seen disruption, and this has been concentrated, and as Peter said, the big insight that research has had in the last ten years is that this concentration applies to space. So, it’s the negative effects as spatially concentrated in particular communities. I emphasise that because this is one dimension of inequality that previous work have not focused on.

So, traditionally it was focused on the difference between educated and less educated workers, skilled and non-skilled workers. We did not focus on the difference across different communities; the reason for that is that the premise, among economists, was that it was easy for people to move in space. So, if an area of your district of your community is adversely impacted by trade or some other, some other shock, then you pack up and you go somewhere else and you seek a job in this other community that is doing better. And what this research has demonstrated is that this is definitely not the case, people are not mobile. Even within countries there is very little mobility. We still need to understand why this is so, but the fact is, it is so.

And that and because of that, we’ve seen, starting with a global financial crisis in the United States, but also, as a result of trade, we’ve seen that, whenever there is a negative demand shock, the effect of this shock tends to persist, and they tend to be as you mentioned, spatially concentrated with many knock-on effects.

Paul Johnson

And can you, so you talk about spatial concentration, and I don’t know if you or perhaps Peter could just put a bit of colour on that because certainly some of the work that Angus Deaton and others have done suggest that, that, you know, that can have a really, I mean really bad effect on those working communities.

Penny Goldberg

First of all, what do we mean by communities? One aspect of David Dorn’s work that was very surprising to me is that they found this spatial concentration to apply for commuting zones in the United States. This is a very narrow definition of the spatial unit, and it’s really surprising that people could not move across commuting zones. So, and then what the workers documented is that first we saw job displacement, so people lost their jobs, this is something that you might expect, the wages also stagnated; but then there were follow up social effects. We saw crime in these areas increasing, we saw mental illness in these areas increasing, we saw many of the negative health effects that Angus Deaton and [unclear 14:35] talk about in their books, being concentrated in precisely those communities that were adversely, among other things they were adversely affected by trade. And I don’t want to, to imply here that trade was the only culprit, that trade was responsible for the depths of despair – there were certainly other factors including supply-side factors, there were doctors who were prescribing opioids in a very irresponsible way so there were supply-side factors as well as demand-side factors. But nevertheless, there has been also a very strong correlation between the adverse effects of trade in these communities and all these negative effects that Angus Deaton talks about.

Paul Johnson

So, so, Penny, we’ve talked about the negatives here, we’ve talked about, you know, in the UK potentially hundreds and thousands of people losing their jobs and probably more in the US and other developed countries. We’ve talked about the scale of the impact this can have on individuals and communities. I mean is it worth it, I mean was it a mistake to allow globalisation to happen in the way that it has happened over the last twenty or thirty years?

Penny Goldberg

My answer is definitely not, it was not a mistake to encourage trade, to encourage open borders. Perhaps what was a mistake was to completely dismisses any concerns about potential adverse effects of trade on inequality. What perhaps was also a mistake is that many of these changes happened too fast, without giving countries and communities time to adjust. So, I think if we were to do this over again, we might have taken more time to think carefully about the consequences of opening up that fast. I think that we may question that part, but not the fact that we opened up.

I think however many concerns we have about inequality, it’s also undeniable that trade and openness have had significant, significant benefits, beneficial effects on the world economy. So, both on rich counties and also on poorer countries.

Paul Johnson

So, it was worth doing despite the cost that people have faced. But as we’ve seen there’s clearly been some political backlash on those costs. What are the lessons for the future? I mean we don’t want the world to close down, we don’t want to move away from international free trade and globalisation, so do we know very much about how to do it better in the future, in a way that won’t have these negative effects?

Penny Goldberg

So, I believe there are some lessons we’ve learnt. One I already mentioned we need to be, we need to acknowledge the fact that trade does have distributional impacts. This is something that was always, that was often denied even though economic theory, trade theory, explicitly teaches us that trade does have these distributional impacts. So, the first step I would say is to acknowledge that there is a potential problem.

Second, I think we need to have policies in mind that may help with the transition. At this point we all agree that the solution to the problem is not the return to the past. Reshoring or bringing heavy industry back to the United States or Britain is not the solution, it’s not going to make workers better off and at the same time it may conflict with other objectives that governments have. So, for example environmental objectives, or the desire to improve the quality of jobs, nobody would think that mining jobs or jobs in heavy industry are good jobs from today’s point of view.

So, I don’t think that’s the way urr to solve the current concerns. What’s a better way? A better way is to help the transition, to make transitions easier for people and how we exactly do that, this is a question for policy. Right now, we need a lot of experimentation, we need to experiment with different ways of assisting people with different programmes and see which programmes work well and which programmes don’t.

But another general lesson we learnt, again, is in many cases, even when people from governments mean well, it might be hard to identify the winners and the losers. It takes some time. So, I mentioned one example and until very recently we’re always thought it would be very easy for people to move across space, to move to a different city, within a country. So, economists never worried about this very much. In the meantime, we realised that this actually an issue. If you believe this is an issue, place-based policies may be called for, place-based policy that’s a loaded term in Europe, because they’re often associated with huge inefficiencies. But design place-based policies that make sense it something that might help. Developing better social protection systems, you know systems that ensure people don’t fall through the cracks if they’re faced with an adverse shock, again might help. And it would help because they won’t require, want to identify immediately what the source of the income loss was. That’s the problem with current schemes for example the trade adjustment assistance in the United States, it’s tied very closely to you having lost your job because of trade, it takes some work to figure this out, it takes some time, but in the meantime, people fall through the cracks. So having a strong social protection system might help one to solve this issue.

But overall, I would say the one big lesson we learnt is that that, we need to think about the distribution of consequences in advance, you know. Not go headfirst with the attitude, “let’s liberalise, let’s open up the economy and we’ll worry about the negative impacts later.” This won’t work in the future.

Paul Johnson

Yeah, there’s a really strong message there, I think you could use that message for quite lot of policies that economists and others have talked about over the, over the years and indeed I suspect some people from the last labour government think that’s about the scale of immigration, at least in Europe in the early 2000s, not the issue of it but the speed of change that it engendered.

Place-based policies, Peter, of course are very much top of the agenda here at the moment as government thinks about, in inverted commas, “levelling up” and policies to achieve that. I mean you looked a little in the work that you did at some of these policies that might mitigate the impact of globalisation, I mean where did you end up on, in your view on what might be effective going forward? Do we yet have a good grasp of how you can actually mitigate these threats?

Peter Levell

So, I’m afraid I’m going to have to disappoint you and say I’m not sure we do have a good grasp on the basis of current evidence on which sort of policies, be they training policies or place-based policies, are particularly effective at helping people get back in to work, complete these transitions as Penny says. And that’s because the evidence we have on a range of place-based or training policies is quite mixed. There have definitely been some policies that show some evidence of having an effect in terms of what they set out to achieve and a large number of policies that have also not turned out to be cost effective or have come with a heavy administrative burden and so on. And it’s not really clear on the basis of the current evidence that we can point to a particular set of things that work very well and a set of policies that don’t work as well.  

So, this is an area where we do need to think about impacts in advance, as Penny suggests, but we also need a degree of evidence gathering and experimentation and humility when it comes to understanding how we should respond to these sorts of rapid changes. I should add that we’ve been talking about rapid changes associated with trade and trade shocks, a lot of these are associated with, for example the rise of China in the 2000s. Looking ahead it’s not clear that there’s another big developing country in the next ten years that’s going to explode on the scene in quite the way that China did, with quite the same disruptive effects, particularly as a lot of manufacturing employment in developed countries has already been kind of hollowed out, if not by trade shocks, by, by automation.

But there are another series of shocks that might have similarly spatially concentrated impacts that we ought to be keeping an eye on, and these are, as Penny said for example things to do with the transition to net zero when we’re going to increase the cost of energy greatly, is this going to have particular effects on employment in particular areas for particular types of workers. These are things that we need to be thinking about now, we need to be thinking about how we’re going to help people adjust out of those industries and out of those areas if necessary, ahead of time.

Paul Johnson

So that’s exactly the issue I want to pick up from here, Penny, what, what next in terms of globalisation? I mean have we been though a one-off shock as China has entered the trading world and as far as globalisation is concerned at least, we’re now in a period of relative calm? Or are we going to be moving backwards as trade barriers go up, or is there more globalisation coming over the horizon?

Penny Goldberg

Well, no one has a crystal ball. But I won’t say that globalisation has run its course, I mean if we agree that right now, we are in a calm period, partly because there was this backlash against hyper globalisation, but the new frontier in globalisation is the service sector. The service sector remains highly protected, there is relatively little trade in services, even though it’s very hard to measure. But suppose we globalise service trade, and there was true free trade in services across the world, the consequences would be mind boggling, and that one reason no one wants to go in that direction very fast. But imagine that you can hire a doctor or a lawyer in another country, you can hire their services in a different country, or an architect? There will always be some service that require the person to be onsite, but there are many other services that do not require the on-site presence of a service provider. So, if this happens then we are talking about a very different world.  

There are many good reasons, that this hasn’t happened, one is that the sector is protected; the other reason is that for service trade to become global, there would have to be a certain degree of regulatory convergence across countries. So right now, the regulations are very different, and this reflects partly different value systems in different countries, and different preferences, but to a certain extent it also reflects protection, you know, precisely because people are concerned about what would happen if all of a sudden you opened up the borders and you let service trade be completely free.

Another issue that Peter mentioned is carbon pricing and the way that we deal with the environment in the future, so if, if there is, the devil I always think is in the detail, so again no one knows precisely how all these carbon taxes will be implemented at this point. But the one thing we can anticipate is that carbon prices are going to change relative prices across countries. And this is going to have very big distributional impacts across countries, but even within countries across different groups. And we can think of carbon pricing the same way we think about trade, it has a mostly beneficial effect to the world as a whole, but the distributional efforts are going to be enormous, we have to make sure, as we think about the future, that whatever beneficial effects we want to realise, they don’t happen at the expense of this huge increase in inequality that may bring the world as we know it down.

Paul Johnson

Yeah, a really important cautionary note the, it’s so easy whether you’re talking about trade or immigration or climate change, or for any other policy, to think about the obvious big overall economic benefits and forget that, well as we know, economic loses to individuals are much more hardly felt than benefits are enjoyed. And particularly when those are concentrated loses, when they’re loss of jobs, when they’re loss of livelihoods and given the, I think, slightly disappointing state of our knowledge on actually how effectively to protect or to support individuals in the communities that are badly affected by some of these changes. Given that we don’t know those, strikes me that the things we need to be very careful of are speed of change in particular. 

But also, and it’s been very interesting listening to both of you, also we as economists really do need to, sort of questions some of our basic assumptions. It was very striking reading some drafts of the chapter that Peter and David Dorn wrote, how, how bizarre from today’s perspective, some of the literature, the economic literature of twenty, twenty-five years ago looks as you were saying Penny earlier on, just this assumption that people would move if local areas are effected. Just the assumption, almost, that trade would help if not everyone, at least it would not be too difficult to offset some of these negative effects.

But I think we are at the end, sadly, of our allotted half hour or so, so thank you ever so much, to Peter Levell and to Penny Goldberg for an incredibly interesting discussion about trade and inequality. It’s one of nearly twenty studies that we are doing as part of The IFS Deaton Review of inequalities, covering everything from things like trade and the role of firms, things you might not expect to be central to studies of inequality; to things you are more likely to think as central which is around race and gender, income inequality factors, welfare, education and so on.

And we’ll be covering more of those topics in later editions of The IFS Zooms In, but for now, thank you very much for listening, and we’ll be back with you in the new year.