 Good morning at the birth of wider in 1985 was a rather unusual time for economists in this area of trade We were not dismal. We're optimistic and we actually had a fair amount of consensus and The consensus was that opening to trade was good tend to reduce inequality in developing countries Why did we all believe this? Well, we had two reasons one was theory Hextroll in theory Told us that it's the abundant factor that gains Predominantly from opening to trade and of course unskilled labor is the abundant factor in the south and it's also the poorest factor So opening to trade was in theory was going to reduce inequality. That was very clear And we had some very powerful evidence the evidence from the four little tigers in the 1960s and 70s These were the leaders of export net growth These were the countries that really had opened up to trade and in these countries unskilled workers had gained Both in real and in relative terms. So that was how it felt 30 years ago What happened? What happened over the next 30 years taking us up to now. Well It wasn't quite like that the effects of trade Which it's quite difficult to disentangle from other effects on inequality But except for the moment that we can disentangle them. We're very mixed in some countries The theory things went worked out according to the theory inequality went down But in a lot of countries inequality went up and we first spotted that in a big way in Latin America in between the mid 80s and the mid 90s when Latin America opened up and the more countries Opened up and the more we studied them the more did we realize that the direction of the effect was very mixed and Moreover that in most of these countries unskilled workers were losing relative to skilled workers a skilled differentials and wages were tending to widen in most cases and It's nicely summed up the experience of the 30 years by this Diagram which I've plucked from a dollar and cray paper in the economic journal in 2004 What you've got plotted on the horizontal axis is changes in openness On the vertical axis changes in inequality and each of these dots is a particular country in a particular period You see the picture most of the observations are to the right of the vertical axis So most countries were opening up to trade, but the distribution of dots Above and below the horizontal axis is pretty even right Half of them inequality went up and half of them inequality went down and in fact this picture understates, I think the degree to which the effects of trade were mixed because it doesn't pick up the wide That the impact on horizontal inequalities the impacts are very different between different genders people living in different places people with different levels of education and Above all it didn't pick up that the effects of trade on the poor were very mixed In the long run if growth triggered growth if trade opening to trade triggered growth along the lines that Alan was suggesting The poor tended to gain but the short to medium term impacts were very mixed some poor people gained some poor people lost So that was the reality in relation to what we expected 30 years ago. Why were we wrong? Right well, we were wrong for two reasons One was well no unskilled labor isn't always abundant in the South as we'd assumed that wasn't part of the theory That was an empirical assumption and it turns out that the four little tigers had a rather special set of factor endowments they were land scarce these were densely populated countries and They had a high level of literacy their unskilled workers all had a basic education most of them had a basic education That wasn't true of the rest of the developing world Many developing countries are land abundant and what does the theory tell us there? No, it's not labor. That's going to gain. It's a land that's going to gain from opening and the effect on inequality Is going to be on critically on the distribution of the ownership of land is it widely spread among peasants? Is it concentrated in in plantations or is this natural wealth in the form of mines and who owns the mines and and upper middle income countries were skill abundant in the sense that say the countries of Latin America had Average levels of education that were above the world average So they were in the same boat as developed countries and the final point of course is that unskilled labor if it's to participate in the kind of process that happened in the four little tigers needs a basic education Illiterate workers are really not employable on a large scale in export at oriented manufacturing or services So that's one reason we were wrong just empirical Second reason is that Haksha Olin is only part of the theory and we learned that Very importantly opening up to trade is linked to processes that often bring Skill-biased Technology into a country new processes or skill-biased activities You know direct foreign investment integration into value chains causes the establishment of activities in the country Which may not be particularly skill intensive by global standards, but they're more skill intensive than what's already going on in the country So that boosts the relative demand for skilled labor We also learned that exporting is a relatively skill intensive activity In particular sectors if you want to export you've got to produce unusually high-quality products And that requires an unusually highly skilled labor force. So again boosting the demand for skill and the third bit of the theory which was old which we'd forgotten is imperfect mobility of Factors among sectors Sectors and localities the Intersectoral immobility bit. This is standard. This is specific factors. This is Ricardo Weiner theory. We should not have forgotten this Some factors can't move at all between sectors some sorts of natural resources in particular others move only slowly and During the process of transition you then get temporary gains and losses and temporary can extend out to decades In addition, you've got imperfect mobility of people among geographical locations And one of the things I think we we woke up to although there were plenty of reminders in developed countries already Is that opening up to trade can blight an entire locality? I mean think about Livingston the town of Livingston in Zambia after liberalization of the trade regime in the 1980s the Employment in the textile industry in Livingston dropped from 10,000 workers to 300 That was basically the end of Livingston as an economy apart from the Victoria Falls So that's why we were wrong What should we do what should we do about the distributional effects of trade on income distribution distributional effects of trade on inequality Well, the first thing we should I think we've learned we should do is to anticipate the distributional effects better I mean the fact that the effects of opening to trade were very varied among countries Doesn't mean that we can't predict them just means we have to look more closely and think a bit harder about the circumstances of the particular country and We also need to watch out for changes in world prices Because once a country is relatively open to trade It's not just further changes in trade barriers that affect income distribution. It's changes in world prices and In thinking about just the effects of opening to trade and changes in world prices on distribution Watch out particularly for food price food prices and particularly food price rises Because the poor spend the bulk of their incomes on food So a rise in food prices associated with opening to trade is likely to reduce poverty not for all poor people Some of them are food producers, but on the whole the poorest are going to lose So anticipate better, but do something about it. So what can we do? What policies can we pursue? To reduce the adverse distributional effects of trade Well, let's start off by being realistic It is simply a delusion to think that we can somehow pursue policies that will avoid all the adverse inequitable unfair unacceptable effects of trade These are just consequences of change all development is change Development is a nasty process. It always has been a nasty process think back to the 19th century Trade is just one of the forces that is disrupting the Disrupting the livelihoods the lives of many many people. So we can't eliminate it But we can reduce these effects and here I'm focusing specifically on reductions that are relevant in the area of opening to trade One is to provide advance warning of changes in trade policies A lot of these windfall gains and losses arise because the changes that caused them were unanticipated if if governments can advise people in their populations their businessmen in advance What they're going to do with trade policies and when they've done it then stick to it That is going to reduce a lot of this there's going to reduce a lot of these undesirable Distribute short-term medium-term distributional effects, and I said here where possible, but you know No way are governments going to be able to predict what's going to happen to world prices over the next two or three years Another thing governments can do is to subsidize factor mobility particularly labor mobility But also capital where markets fail and markets fail in some key Respects that are relevant to factor mobility information provide more information provide assistance with retraining particularly for skilled workers and perhaps one very important case to note is coordinate investments in The resuscitation of localities that have been completely blighted by trade. I mean these are three three examples What else can we do? Well? There he's very clear you tax the gainers and you redistribute the income to the losers and so you get a real Pareto improvement and not just a potential Pareto improvement Well, unfortunately, that's not feasible It's not fees. It's not it's ethically a bit difficult as well because Not all the losers were poor and not all the gainers were rich So it's a difficult problem there But the infeasibility is really that the effects of trade both the gains and the losses are very widely diffused many of them are Indirect so it's extremely difficult to identify who the gainers and losers are Let alone tax or subsidize them. I mean countries including the United States have experimented for a long time with trade-related Adjustment assistance. I think that's been politically helpful But it hasn't been terribly effective. I think in reducing the adverse distribution of effects so we have to fall back on General sort of second best policy measures and of course progressive taxation however you design it is going to be a good thing It's going to pick up not just the gainers are going to pick up the rich gainers and they're the people who you actually want to tax Conversely you want social protection you want safety nets to protect poor people and again This is you're picking up the losers you want to help who are the poor losers, of course Both these instruments are also going to deal with the wider problem I've I've I've mentioned which is that all development all change generates gains and losses so these two instruments are Going to be useful not just in the context of trade, but they're going to be of course more generally useful and Finally those of you who want to as it were follow the literature on this in a bit more detail You can pursue this your leisure afterwards. Thank you