Non-tariff Barriers, Regionalism And Poverty: Essays In Applied International Trade Analysis

Essays in Applied International Trade Analysis

L Alan Winters

• Publisher: WSPC
• Year: 2015
• ISBN:
  9789814571289
• Topic: Biological Sciences
• Subtopic: Science General
• Index: Biological Sciences

Non-Tariff Barriers, Regionalism and Poverty is a collection of key articles in three important areas of applied international trade research: measuring non-tariff barriers and their effects, the consequences of regional trading arrangements, especially on the countries excluded from them, and the connection between international trade and poverty. Drawing from 30 years of research and experience, L Alan Winters illustrates the development of techniques of this field and his continued commitment to answering real policy questions at the times at which they are debated. The collection shows the ways in which economic and econometric analysis can be used to answer real-world problems rigorously in the area of international trade and trade policy. Readers will find that some of the research included is of current methodological relevance and some of more historical significance. This volume is invaluable to anyone who is keen on developing their knowledge on trade policy, regionalism or poverty — three pressing issues in today's globalized world.

Contents:

• Introduction
• Non-Tariff Barriers:
  • The Extent of Nontariff Barriers to Industrial Countries' Imports (with JJ Nogués and A Olechowski)
  • Do Exporters Gain from VERs? (with J de Melo)
  • Labour Adjustment Costs and British Footwear Protection (with WE Takacs)
  • Voluntary Export Restraints and Rationing: U.K. Leather Footwear Imports from Eastern Europe (with PA Brenton)
  • VERs and Expectations: Extensions and Evidence
  • Digging for Victory: Agricultural Policy and National Security
• Regionalism:
  • Separability and the Specification of Foreign Trade Functions
  • British Imports of Manufactures and the Common Market
  • Regionalism and the Rest of the World: The Irrelevance of the Kemp-Wan Theorem
  • How Regional Blocs Affect Excluded Countries: The Price Effects of MERCOSUR (with W Chang)
  • Trade and Economic Geography: The Impact of EEC Accession on the UK (with HG Overman)
• Trade and Poverty:
  • Trade Liberalisation and Poverty: What are the Links?
  • Trade Liberalization and Poverty: The Evidence So Far (with N McCulloch and A McKay)
  • Trade Liberalisation and Economic Performance: An Overview
  • Trade Liberalisation and Poverty Dynamics in Vietnam (with Y Niimi and P Vasudeva Dutta)
  • Agricultural Trade Liberalization and Poverty Dynamics in Three Developing Countries (with J Litchfield and N McCulloch)
  • Why Isn't the Doha Development Agenda More Poverty Friendly? (with TW Hertel, R Keeney, and M Ivanic)
  • Trade as an Engine of Creative Destruction: Mexican Experience with Chinese Competition (with L Iacovone and F Rauch)

Readership: Undergraduates, research students and professionals interested in macroeconomics; international trade practitioners.
Key Features:

• Shows the ways in which economic and econometric analysis can be used to answer real-world problems rigorously in the area of international trade and trade policy
• The broad collection of applied work on trade illustrates problems and analyses that others may find a useful base for their own work

Chapter 1

Introduction

This book contains a selection of my articles on international trade and trade policy, written over a period of about thirty years. This introduction is intended to place them in several contexts. It attempts to locate the papers’ contributions at the time they were written and also to ask how much of that contribution remains today. It also briefly desribes their origins in terms of my own voyage, both intellectually and in terms of policy engagement and prescription.

Even after forty years, I find some pieces of economics elegant and exciting in purely intellectual terms — for example, the Stolper-Samuelson Theorem and the Kemp-Wan Theorem still thrill me in almost the same way as Mozart does. But, possibly for a want of ability, theoretical development per se has never motivated my research. If economics were really just a branch of moral philosophy, as a discipline with only theory would basically be, it would be impossible to justify a profession of even one tenth of the size that economics has now attained. Rather, I believe profoundly that economics and economists exist to be useful, and that to be useful we need to relate theory to the real world in as precise a fashion as possible.

Given this slightly grandiose starting point, it is hardly surprising that almost all my research has been driven by either empirical or policy questions — wanting to know whether a particular piece of theory actually reflected the ‘real’ world and/or trying to apply it to answer specific policy challenges. I have found deciding what research to do among the most difficult parts of academic life, and have rarely, if ever, been successful at dreaming up research topics in the abstract or identifying things that will ‘sell well’ in journal terms. Rather almost all my research stems from one of three origins. In some cases I have found statements by fellow economists (academic or policy-based) to be surprising and have set off to test them; in others my own research has posed questions, so that one thing has led to another; and in yet others, I have been asked whether I can solve or offer advice on a particular policy question. While the last motivation is not strongly represented in the papers included here — possibly because many policy questions do not lend themselves to outputs that are attractive to leading journals — I have benefitted enormously from talking to policymakers and policy-advisors and taking their concerns seriously and/or at face value. For example, my research on small countries — e.g., Winters and Martins (2004) — and my work quantifying the potential benefits of liberalising temporary labour mobility (so-called Mode 4 of the GATS) — e.g., Winters et al. (2003) — were both stimulated by insightful but practical questions (and a small amount of funding) from Roman Grynberg, a top-rate practitioner, who then worked for the Commonwealth Secretariat. In the first case, I thought his views were wrong, but the data proved quite the opposite! One of my main pieces of advice to young economists is to listen carefully to policy-makers and try to work out what is behind their questions and views.

The three topics covered by this book have been the major themes in my research life.

Part A: Non-Tariff Barriers

I joined the World Bank’s International Economics Research Division led by J Michael Finger, an extraordinarily wise economist and manager, in 1984 at a time when Anne Krueger was Vice-President and Chief Economist. Together, they had taken several steps towards changing the prevailing view of trade and development, both in the world and in the World Bank, from import substitution to a more liberal and open conception. The original Asian tigers (Hong Kong, Korea, Singapore and Taiwan, China) had demonstrated the virtues of trade as an aid to development, and trade policy reform was beginning to catch on in the Bank’s operational work. A major problem for developing country governments wanting to pursue a more liberal policy — and by so doing to eliminate a number of comfortable rents in their own economies — was that the developed countries seemed to be saying ‘do as I say, not as I do’. Opponents of reform would constantly claim that the developed countries restricted access to their markets especially in the commodities of interest to developing countries. To the extent that this was true, it would not only have set a bad example, but would also have curtailed developing country exports and hence incomes. And in this way, it would tend to weaken the contribution that trade liberalisation could make towards development and undermine the forces arguing for it.

The Tokyo Round of GATT talks, which had ended in 1979, had reduced tariffs in many developed countries quite significantly, and the General System of Preferences (GSP), exempted developing countries completely or partially from many tariffs. But on the other hand, developing country exports of many products were subject to quantitative and other trade restrictions: among the worst, agricultural trade was hugely distorted by a plethora of opaque barriers, clothing exports were subject to restraints by the Multi-Fibre Arrangement, and footwear was also subject to quotas and other measures. One such barrier was the so-called Voluntary Export Restraint (VER), whereby developing country exporters (and also Japan in a number of products) were invited to restrict their exports to the USA and Europe and agreed to do so for fear of facing something worse. Because these restrictions were voluntary, it was argued — speciously — that VERs were consistent with the GATT. They were about as voluntary as the response to the old highwayman’s offer of ‘your money or your life’.

Both to induce some discipline on the part of developed countries, which in the first half of the 1980s restricted trade with almost no provocation, and also to help put the criticisms of developing countries into some perspective (i.e., to explain that Europe might restrict imports of shirts via the MFA, but that this was not the same as having 300% tariffs on many imports as developing countries sometimes had), it became imperative to measure and analyse these non-tariff barriers. For international trade economists, this presented a happy coincidence of interesting analytics and deep policy relevance, and both the World Bank International Economics Division and several leading scholars responded to the challenge.

My first task (I do not think I had heard of non-tariff barriers before joining the World Bank) was to try to measure the extent of developed countries’ non-tariff barriers (NTBs). Scholars in UNCTAD had started to record the restrictions that existed as best they could (they were often declared to be secret), but to make these data work in the policy world they had to be quantified (and implicitly compared across countries). Andrzej Olechowski in UNCTAD and Julio J Nogues in the Bank had started to work on this problem, and I joined them in the autumn of 1984. One job was to persuade UNCTAD to share these (sensitive and possibly valuable) data with the Bank, which Jurek Rozanski and I did in December 1984, and he and I then analysed them over the following spring, ready to present to the World Bank’s Development Committee. By today’s standards the dataset was small, but then it seemed massive and it required huge ingenuity on Jurek’s part and many overnight sessions at a facility in Mclean Virginia, which had a larger computer than the Bank to get them into order.

We also had to devise appropriate measures for NTBs. Guided significantly by the work of Robert Stern and Alan Deardorff (see the references in Chapter 2), we used the frequency ratio and the coverage ratio, the first reporting the share of headings in a standard classification of international trade that were affected by one or more NTB, and the second the proportion of total trade that occurred in the affected headings. Neither is powerful as a measure (for example, not all trade headings are equally important and a prohibition on imports receives a zero weight in the coverage ratio) but it is not clear that we have devised any better measures in the intervening three decades — see, for example, page 82 of UNCTAD (2010).

Clearly just counting up how many headings and how much trade gives a very incomplete picture of trade barriers. What we really want to know is by how much the barriers restrict trade and, from that, what their implications are for the things that matter such as economic welfare and its distribution. So this was the next frontier and one on which we started to work immediately. Many fine scholars such as Rob Feenstra and Jim Anderson were working on this problem, and the decade from 1985 to 1995 produced many exciting estimates and uncovered several different routes through which NTBs — especially VERs — affected economic welfare.

My own efforts hinged substantially around a big programme that I put together with Carl Hamilton and Jim de Melo, both of whom I met via the International Economics Division of the World Bank and Mike Finger. I decided we should focus on the footwear sector: it was technologically fairly simple (unlike, say, autos) and while it was heavily distorted by quantitative restrictions, there were a number of flows and years in which trade was relatively free. These would provide the evidence for a counter-factual analysis — the sine qua non of any policy analysis. All told, we produced nearly a dozen papers on footwear, which caused Carl to quip — more aspirationally than accurately — that footwear would become the fruit-fly of NTB research (a reference to the way in which the fruit-fly under-pinned so much genetic research at the time). Three of these papers appear here as Chapters 3 to 5.

Rob Feenstra had investigated two important aspects of VERs in the USA market for autos where Japanese exports were subject to VERs. (It is salutary to recall today the 1980s fears that Japanese manufacturing was unstoppably competitive and that Japan would dominate the world economy). First, Feenstra, along with several others, showed that, by reducing the supply of imported goods relative to the demand, VERs increased the prices of imports, worsened the USA’s terms of trade, and transferred welfare to the Japanese suppliers. This is part of what made VERs such good politics — the exporter was dissuaded from complaining about selling fewer units by earning higher rents on each of of the units he/she did sell. Second, Feenstra showed that VERs also led to product upgrading. A VER acts like a specific duty on an import (imagine that the licences to export to the USA were auctioned — each would command the same price). The cost of the licence (duty) is proportionately lower, the higher the price of the good, and so producers will tend to use their limited stock of licenses to sell high quality (and, therefore, high price, high margin) variants of their products.

Jim de Melo and I found little evidence of quality upgrading in the case of footwear exports (De Melo and Winters, 1993). We also argued, however, in Chapter 3 of this book that even the rents generated by a VER were not a sufficient compensation for the fact that VERs (a) distorted allocation decisions across markets because the goods that could not be sold in the VER-restrained market were diverted to other markets in which they drove down prices, and (b) reduced the size of the exporting industry. Moreover, even if a VER generated higher profits in the exporting industry, any associated contraction in production reduced wages in it. Thus the generally prevailing view in the 1980s and 1990s that VERs imposed little hardship on developing country exporters was, we argued, misleading. These findings are still relevant today, although quantitative restrictions are rarer and VERs are prohibited by the WTO.

The last result about wages pointed out a further irony about VERs and other quantitative restrictions. They were frequently introduced precisely to protect the interests of workers in the import-competing industry in developed countries — in particular to reverse or reduce the decline in jobs that competition might induce and thus to reduce or avoid the adjustment costs of workers having to find new jobs. Taking these aspirations seriously, Wendy Takacs and I used employment records from a British footwear company to quantify adjustment costs — see Chapter 4. We developed a measure of adjustment costs and estimated it for the British footwear sector and also modelled the labour displacement that was averted by UK quantitative restrictions on footwear imports. Even ignoring the fact that jobs protected in footwear might reduce jobs elsewhere in the British economy, we found that the costs that such protection imposed on consumers far outweighed the adjustment costs avoided by restricting trade.

The main reason that adjustment to reduced labour demand was not very costly in the UK footwear sector was that the sector had high labour turnover: any jobs lost could be fairly well-accommodated by natural wastage as workers moved off to other jobs naturally and, because the firms were in the market to recruit frequently, wages in the sector were close to competitive levels. In addition, the restrictions were not very effective at reducing imports — partly because being country-specific, they could be circumvented by trade diversion. There is not a great deal of work estimating the labour adjustment costs of import competition and the chances of reducing them by import controls, and so this work is still periodically cited.

The third footwear paper in this book — Chapter 5 written with Paul Brenton — was certainly the hardest to produce. Its starting point is that any binding quantitative restriction induces rationing somewhere in some economy. If, for example, producers facing a VER do not raise their prices at all (perhaps because they want to preserve consumer goodwill), consumers are rationed — they want to buy more than they are allowed to at the prevailing price. If, on the other hand and as is more commonly assumed, the prices of the imports are raised to eliminate excess demand (i.e., to clear the market at the exogenously fixed quantity permitted by the VER), producers are rationed in the sense that, at these prices, they wish to sell more in the restricted market. In the former case, consumers’ choices among varieties or among goods are distorted and in the latter case, it is producers’ decisions about which markets to supply that are distorted.

In Chapter 5 we considered the consumer side of the issue—allowing for the possibility that when the UK imposed restrictions on imports of leather footwear from Eastern Europe, prices were not raised to market-clearing levels. Our aim was as much methodological as substantive, showing how one could identify quantity rationing in the data and then allow for this in estimating the parameters of economic behaviour. We identified significant rationing in a number of periods and found that it had strong effects on the quantities imported: for example, we estimated that Poland would have had a 40% share of UK imports of men’s leather footwear in the absence of trade restrictions, compared with an actual share of 10%.

The approaches in Chapter 5 are fairly complex to implement and so we have not seen a great deal of use made of them. However, the results of our exercise do suggest that where rationing is a possibility, it is important to take it into account, and thus I hope that if in future quantitative restrictions were to proliferate, this methodology would be helpful in identifying their costs.

As noted above, VERs were quite common and attracted a good deal of academic attention in the 1980s. Exporters — especially from smaller countries — found VERs very burdensome and pressed hard to ‘outlaw’ them in the Uruguay Round of trade talks that ran from 1986 to 1994. And they prevailed! VERs were prohibited and while the agreement did not seem completely water-tight at the time, it has proved quite effective with many fewer VERs now in evidence. It is not possible to attribute credit for this major improvement in trade policy in any accurate fashion. However, it is not implausible to believe that the academic community’s findings that VERs were costly both to the countries imposing them and to the affected exporters as well — played a significant role in informing wider audiences about the problem and in providing credible ammunition to those who wished to prevent them in the future.

In a final postscript to the work on quantitative restrictions, I observe that export restrictions have become more common in the last few years, especially in response to rising food prices but also with respect to raw materials. Their analysis calls for much the same tools as were developed in the 1980s and 1990s. In 2013 and 2014, I worked on the WTO Dispute between the European Union, Japan, and the USA on the one hand and China on the other, over the latter’s export restrictions on rare earths, tungsten, and molybdenum. In describing how to infer the effects of such restrictions, I found myself drawing on this old literature, including my own papers on footwear, such as Brenton and Winters (1991).

Chapter 6 reflects a different part of the debate about VERs, but actually stemmed less from that than from an attempt to clothe a popular idea in empirical garb. It had been argued that while a VER might curtail exports from a particular source to a particular market, the anticipation of a VER would have the opposite effect. The total level of exports under a VER needed to be allocated across firms in the exporting country, and since obtaining the right to export granted the firm the right to sell in a market that had been artificially restricted and hence would generate rents, firms had strong incentives to get an allocation. In many cases export rights were allocated in proportion to firms’ exports in the last unrestricted year, so higher exports in that year were essentially the key to obtaining rents the next year when the VER had been introduced. In Chapter 6 I showed that this reasoning depended critically on the struc...

Table of contents

• Cover Page
• Title Page
• Copyright
• Table of Contents
• About the Author
• Credits
• 1. Introduction
• Part A: Non-Tariff Barriers
• Part B: Regionalism
• Part C: Trade and Poverty