Ebb tide

Ebb tide


In the summer of 1999, the tide of trade liberalisation turned. Less than five years after the creation of the World Trade Organisation (WTO), the crowning moment in the long struggle to lower tariffs and police trade practices, the Seattle meeting of its members collapsed amidst violent protests against globalisation. Though the following 20 years saw more trade deals (and the admission of China to the world trading system), it has become increasingly clear that trade liberalisation is under pressure. And just at the moment that British Brexiteers place their faith in “WTO rules” as a substitute for membership of the European single market, the WTO itself is under attack by the leadership of the world’s largest economy. This article reviews the economic rationale for free trade, the rise in protectionism, and the search for some “better answers” for the future.  

The global trading system ended the 1930s encrusted by protectionism, even before the Second World War brought naval blockades on the participants. As the war ended, economists led by John Maynard Keynes urged governments to stimulate world economic recovery by dismantling barriers to trade and creating a global institution capable of policing this – a third pillar of global economic architecture, to buttress the International Monetary Fund and the World Bank. It was a long time coming.


The eighth postwar round of global trade negotiations began with no more fanfare than the previous ones, in September 1986. When trade negotiators representing the countries party to the General Agreement on Tariffs and Trade (GATT) gathered in the city of Punta del Este in Uruguay, there was very little media coverage of the event. It would be another three years before Tim Berners-Lee brought the world wide web into being – and conference participants recall that the few fax machines available in Punta del Este worked only intermittently.

The MIT economist Lester Thurow had dismissed the GATT, the supposedly temporary agreement signed in 1948, as the “General Agreement to Talk and Talk”. It took a lot of talking – the Uruguay Round lasted over seven years – to replace it with something better. But in the end this round led to the biggest ever single extension of global rule-making on trade, as well as cuts in tariffs, an extension of trade rules to new sectors, and above all, in 1995, the launch of the WTO.

Barely was the ink dry on the WTO’s founding agreement, when the backlash against globalisation began (see Box 1). To answer the “why?” and “what next?” of these political and economic changes, we begin by revisiting the theory of trade, review the evolution of international trade in practice, and then focus on present tensions, with the aim of identifying where new policies need to be developed.

Barely was the ink dry on the WTO's Founding Agreement, when the backlash against globalisation began






The presumption that free trade is a good thing stems from faith in competition to drive down prices, stimulate innovation and increase economic efficiency. It is a faith that gained strong adherents in the postwar world, inoculated against a belief in protectionism by memories of the 1930s.

It was the philosophy that led even those suspicious of the “European Project”, such as Margaret Thatcher, to support the development of the Single European Market. It was the stimulus for President Ronald Reagan, in his 1980 campaign, to propose a North American common market; and for President Bill Clinton to bring the North American Free Trade Agreement into being in 1993. And it was the reason for all the major developed economies to work for 50 years towards the creation of a global trade authority (see Box 1).

Ironically, free trade with the rest of the world is now the cause espoused by many of “Thatcher’s children” who are determined to leave the European single market. But it is very questionable how many Brexit voters yearn for free trade. Many of those in high Brexit-voting, low economic-performing regions of the UK saw international competition as against their interests.

They were not alone. In other countries too, during the 20 years of Frontier’s existence, globalisation has become a dirty word. This is most obvious in the United States, one of the chief architects of postwar economic governance, where President Trump has couched his policies in terms of a struggle against “globalists”. France, one of the main drivers of European integration, was rocked in late 2018 and early 2019 by a series of spontaneous, violent protests. While the genesis of the “yellow vest” movement lay in a specific grievance against a fuel tax, it quickly morphed into a broader set of grievances on the part of those seeing themselves to have been “left behind” by globalisation. In this article, we attempt to bring economic thinking back to the aid of those involved in the tensions and tangles of current arguments about trade policy.  




The first GATT agreement, in 1948, was signed by 23 member states; the WTO was launched in 1995 with 123. The Uruguay Round Agreements brought in deep tariff cuts, and tighter rules on subsidies and regulation. They brought agriculture under trade rules for the first time. They started to address services and intellectual property, which account for a larger proportion of developed economies’ output than merchandise trade. And they replaced the supposedly-only-temporary GATT with a new, more powerful body.

The name “World Trade Organisation” was settled on after the Americans had pressed for the deletion from its working title of “multilateral”, an unpopular word with Congress. The WTO was to oversee a binding system of dispute settlement, allowing retaliatory measures against those of its members found to be in violation of rules.

It established its credentials with some important early judgements demonstrating it could take on even the United States; notably the case brought by Brazil and Venezuela on US gasoline import restrictions. But perhaps inevitably, the main litigants proved to be the US and the EU; developing countries, while queuing up to join the WTO, continued to complain that the odds were stacked in favour of those with deep pockets for lawyers.

By 1999, however, this was only one element in a disparate coalition of anger against globalisation. The protesters who brought the members’ meeting to an end in the “Battle of Seattle”, amid a haze of tear gas, seemed to include every strand, from rich-world labour unions through environmentalists to poor-world commodity producers, as well as an international mixture of the anti-capitalist left.

Seattle was an early warning shot across the bows of governments and international institutions. They needed to respond to those the economist Joseph Stiglitz called the “discontents” of globalisation. But it was a warning that largely went unheeded.

 Instead, frightened governments began to lose their appetite for multilateral trade liberalisation. The WTO continued to attract important new members, including China and Russia; China’s accession being the major development in multilateral trade in the last two decades. Bilateral and regional trade agreements continued (most recently, in February this year, when a major deal between the EU and Japan came into force). But the Doha Round of talks, launched in 2001 and (over-ambitiously) dubbed as a “development round”, fizzled out with only an agreement on trade facilitation in 2013 to show as a new addition to the multilateral rule-book...

The attention of national leaders became focused on other economic issues, notably those spawned by the global financial crisis of 2008, and the impact of international terrorism on global security. And meanwhile domestic pressures for protection of specific industries and regions, particularly in the United States, began to gather strength. And while President Barack Obama supported the Trans-Pacific Partnership (TPP), despite congressional opposition, his successor refused to ratify it and quickly began the scaling-back of the North Atlantic Free Trade Agreement (NAFTA). President Trump’s administration also blocked appointments to the WTO’s appellate body, threatening to derail the dispute settlement mechanism.

More generally, developed country governments began to suffer a form of “buyer’s remorse” at having allowed China into the scope of multilateral trade rules. Their expectations that China would over time become a “normal” market economy appeared not to have been borne out. The growth of China and its participation in world trade has been to the benefit of the world economy as a whole. But the perception – even if exaggerated – that it plays by its own rules has contributed to liberalisation fatigue. Meanwhile, other emerging countries (such as India, Brazil or South Africa) remain unwilling, largely for reasons of institutional culture, to show leadership in trade governance at a multilateral level, even though it is in their long-term interests to do so.  




Despite the current scepticism about “experts”, the sensible place to start is with trade theory. This offers a perspective on the presumption in favour of competition that takes account of borders; and seeks to analyse how countries set policies to compete with each other.

Most first-year students get their first taste of this with David Ricardo’s principle of comparative advantage, coined at a time – the early nineteenth century – when tariffs were the norm. In his own country tariffs became painfully plain to see in the Corn Laws, which protected landowners from foreign competition and kept up the food prices faced by the poor, making an exposition of the beneficial dynamics of trade that were sorely needed.

Today’s students are taught to apply the principle of comparative advantage within a general equilibrium framework developed over a century later, through highly stylised models of trade whereby two countries trade two commodities based on two factors of production (labour and capital).

This was developed by Eli Heckscher and Bertil Ohlin, and extended by Paul Samuelson, and it is a neat way of presenting a powerful principle. But it has its obvious limits in a world in which, among other things, similar countries trade multiple varieties of similar goods, and where as countries developed, they tended to diversify rather than specialise.

Moreover, neo-classical economics was primarily concerned with efficiency rather than distribution. Predictions using the Heckscher-Ohlin trade model, that exports of labour-intensive goods would rise from labour-rich developing countries, and fall from advanced (more capital-rich) developed countries, roughly fit with our experience of rising Chinese wages and exports. But though even very basic models of trade allowed for distributional consequences, economic thinking on such effects lagged behind. At most, they were thought of as temporary (the unemployed in industries exposed to greater competition would be reallocated to other sectors).

Standard modelling techniques did not help, since they assumed that all markets – and that included labour markets – cleared instantaneously. Even when the unrealism of these assumptions was acknowledged, distributional changes were seen as transient (euphemistically known as “adjustment effects”) and it was assumed that governments would do something to address them – though it was never made clear what that “something” might be.

In the 1980s, Paul Krugman and others introduced new ways of thinking about trade that drew on the theory of industrial organisation, and particularly the concepts of increasing returns to scale and product differentiation. They also used these concepts to marry the study of economic geography and trade theory. All these innovations helped to explain observed patterns of trade, and why growth in trade (and growth itself) was unevenly distributed. Paul Krugman won a Nobel Prize in 2008 for his efforts, although another Paul (Romer) had to wait a further ten years for a Nobel Prize to recognise his work on growth that paralleled the new trade theory.

But even Krugman, who had begun by arguing that jobs lost to imports would be quickly replaced (“the limit keeps being pushed out, as trade creates “new” labour-intensive industries through the fragmentation of production”), had to rethink some of his arguments as the extent of the distribution effects became plain. By 2007, he was warning in published articles of the need for free traders to have “better answers”.

To begin with, trade itself was changing. Several decades of liberalisation at a global level had brought tariffs to historically low levels and made big dents in restrictions on services trade. Transport costs had fallen. And finally, information and communications technology had, in line with Moore’s law, developed at an exponential rate – powering further potential growth in trade. And yet the rate of growth in trade was falling. The principal effect of these developments has been what some economists have called the “great unbundling”: the emergence of global value chains.  



Value chains are hardly new. Firms have long had different ways of arriving at a final output, ranging from a soup-to-nuts production line in a single factory in one country to a chain of contracts between different companies across the globe. But the cross-border scale of today’s value chains, managed by multinational behemoths, is a relatively new phenomenon: by 2015, multinational companies accounted for around a tenth of global GDP.

Four centuries ago, Adam Smith argued that the gains from specialisation are limited by the size of the market. Breaking up value chains brings countries global opportunities, irrespective of whether the final product is actually sold in their home market. And it permits suppliers to specialise in specific tasks, improving their efficiency. This can have self-reinforcing effects: places that become good at integrating their contributions into value chains see this advantage reinforced over time.

From a trade theory perspective, what this means is that countries have come to “trade in tasks”, rather than barter products, singly or in bundles. Countries have specialised in particular links in the value chain. Inputs or intermediate goods are brought in, combined or worked on, and sent onwards for further development until the final product is completed and sold.

The iconic example of this mode of production is Apple’s iPhone. A schematic representation of its value chain and the provenance of value is depicted in Figure 13.1.

The emergence of global value chains has also blurred several traditional boundaries: between exports and imports, and between services and manufacturing. This complicates the data, and how policy-makers should respond.

Consider the iPhone once again. Suppose that inputs into its production priced at $24 are exported from the US to China, and that the factory-gate price of an assembled smartphone, ready for export from China to the US, is $194. In the balance of payments this would be treated as a $170 addition to the US’s bilateral trade deficit with China.

But if we take into account inputs from the rest of the world, the “value added” in China in this example is only about $6.50. Moreover, if we then consider the other elements of value added, such as services (marketing, distribution) and intangibles (intellectual property, branding), we discover that they account for the bulk of the final $600 retail price.

This is not only true of “techy” products. Researchers at the Fung Institute in Hong Kong analysed the value chain behind the sale of a “Chinese” suit in the US, and concluded that 91% of its value was attributable to intangibles – and all but two percentage points of that were captured by the United States.

Data prepared by the OECD through its Trade in Value Added Project bear out some of these trends at an aggregate level.  

Figure 13.2 reports the share of domestic value added by different countries in the exports of their foreign partners, as a proportion of those countries’ exports. For example, the share of US value added in the exports of all of the US’s trade partners is equivalent to around 22% of the US’s total gross exports. That is slightly higher than the average for all OECD countries and for the European Union. The shares are an indication of “forward linkages” in value chains, and of how tightly the export prospects of one country are intertwined in the prospects of partners. Figure 13.3 illustrates the interconnections.

The first observation to be drawn from these interconnections is that headline trade statistics (gross exports versus gross imports) do not give us the full picture. The idea that exports are good and imports are bad was always lousy economics, but these interdependencies make it particularly foolish.

The second point is that protectionism will be counterproductive. The basic case against protectionism is that it diverts resources to less productive activities. But this damage is amplified by the interdependencies that exist within value chains. If significant amounts of domestic value added are embedded in imports from foreign countries, targeting these will bite the hand that blocked them.

The third point is that the boundary between services and manufacturing is not watertight. Services are increasingly embedded in manufacturing. This means that more competitive services sectors can boost manufacturing, and thus services liberalisation is an important part of any strategy to overhaul the competitiveness of manufacturing.

Fourth, as the iPhone example demonstrates, value is not evenly distributed along value chains. In particular, knowledgeintensive activities tend to appropriate a higher fraction of the value created.

The fifth, related point is that unbundling has distributional effects within nations as well as between them. Businesses and regions that are active at the higher-value points along the chain gain relative to those who see their parts of the value chain relocated elsewhere.

These points have important implications for policy-makers. The distributional effects explain the pressures on policy-makers to capture more of the value chain. But as we have shown, such knee-jerk responses as attempts to repatriate manufacturing are likely to be counterproductive. So where do we go for Krugman’s “better answers”?


Source: Frontier Economics calculations based on OECD Trade in Value Added database.




Source: OECD, Trade in Value Added (TiVA) Database.




Source: WID.world (2017). See wir2018.wid.world/methodology.html for data series and notes.
In 2016, 37% of national income was received by the Top 10% in Europe against 61% in the Middle-East.



Source: WID.world (2017). See wir2018.wid.world/methodology.html for data series and notes.
In 2016, 47% of national income was received by the top 10% in US-Canada, compared to 34% in 1980.




Historically, income inequality has been low in Europe, higher in North America, and higher still in Sub-Saharan Africa, Brazil, India and the Middle East. But in the past 40 years it has increased almost everywhere, and markedly in the United States.

...The World Inequality Report, compiled in 2018 by researchers at the Paris School of Economics, indicated that by 2016 the top 10% of incomes in North America (the US and Canada) accounted for 47% of national income, up from 34% in 1980.

In the US taken by itself, the share of national income accounted for by the top 1% rose from 11% to 20% over that period, and the share of the bottom 50% fell from 21% to 13%. Inequality was still lower in the US than in some oil-producing countries and much of the developing world, but it showed a marked shift in the period, helping to explain some of the backlash against globalisation in rich countries as well as poor. 


That the US labour force may have had difficulties in competing with lower-cost labour in China seems obvious. But at the same time, we have witnessed an increase in inequality in China, to a level not far short of that in the US. So how much of these shifts in distribution is attributable to trade?

In pursuit of Krugman’s “better answers”, it helps to look at a body of trade theory (led by Marc J. Melitz’s 2003 paper, The impact of trade on intra-industry reallocations and aggregate industry productivity) that began to focus on the differences in the performance of firms and workers. This literature helps to explain why the benefits from openness to trade are not evenly spread, so that while the economy as a whole benefits, there are losers – at least in the first round of effects – as well as winners. The gains are concentrated in the more productive firms, who self-select into trade.

These firms tend to have a larger number of highly skilled, talented or capable workers. Looking for such higher-quality workers is costly, particularly if their skills are in short supply. This increases the wages of productive workers, and moreover increases firms’ willingness to pay to attract and retain skilled workers, thus widening pay distributions.

Exposure to competition further reinforces the efficiencies of these firms, enabling them to pay more and/or expand and demand more skilled workers. Meanwhile, firms that are less productive either contract or exit, and the pay they offer falls relative to that offered in firms that can compete globally.

These findings from research into the microeconomics of trade coincided with the insights from research into the geographical distribution of economic activity. Businesses have a tendency to cluster. This in turn is because such clustering reduces search costs for skills, and creates economies of scale. The clustering effect means that regions that already have location advantages find those advantages reinforced. Economists refer to these as agglomeration effects. These effects tend to be self-reinforcing and can thus further increase the distributional effects described above.

Collectively, the results from these two lines of research help to explain why inequality is on the rise within both developed and developing countries. It is consistent with the unbundling of production we have seen in global value chains and the closer integration between services and manufacturing. It fits in also with the analysis in the previous section of how the returns from intangible assets, particularly intellectual property, have increased relative to other links in the supply chain.    

These two lines of research also yield us some insights into the political economy of protectionism and protest. Historically, those lobbying for tariff protection carried clout because they were concentrated in politically visible constituencies – and this political leverage (by, say, steel and aluminium producers) is still evident in US congressional districts today. The classical response was to mobilise a broader coalition of those who benefited from tariff reduction elsewhere (exporters) and at home (consumers).

But in the new world of technology and complex value chains, what dissenters – such as gilets jaunes, and many Brexit supporters and Trump voters – are seeking is access and participation in economic growth. If the gains from liberalisation are themselves concentrated on the few, creating a broad-based coalition in favour of liberalisation will be harder. And it will be made harder still by the fact that modern trade negotiations are less about tariff reductions than they are about negotiating regulatory changes – some of which press hard on national sensitivities. This helps explain why even governments, let alone protesters, resiled from the latest round of global trade negotiations.  

In the new world of technology and complex value chains






The ease with which protectionist rhetoric has flowed into political debate makes regrettably clear the failure of governments to either persuade voters of the benefits of trade or to come up with solutions to ensure that its benefits are broadly distributed. And in President Trump, protectionist constituencies have found an advocate, whose actions have caused a degree of uncertainty in the world trading system not seen since the Second World War. Support for these actions reflects a deeply rooted disenchantment with trade in the US, which was reflected across the political spectrum in the 2016 elections: Trump’s opponent, Hillary Clinton, was no more than lukewarm on trade, and her own opponent in the primaries, Bernie Sanders, had actively picketed against NAFTA.

To withdraw from the 12-country TPP, one of the most ambitious attempts at making rules on trade outside the European Union, was among President Donald Trump’s first decisions. But it was the renegotiation of NAFTA, and its reformatting as the United States-Mexico- Canada Agreement (USMCA) without the dirty words “free trade”, that made plain the Trump administration’s ambition to “repatriate” value chains to the US.

The negotiations mainly turned on the question of rules of origin. These rules are part of any free trade agreement, and specify the circumstances in which goods are eligible for duty-free treatment. The US wanted to rewrite the rules to require greater “US content” in motor vehicles that could be imported duty-free from Mexico. To these were added minimum wage requirements that would have the effect of disallowing duty-free exports to the US from many manufacturing operations in Mexico.

But Mexico was not the only, and some would say not the main, target of the proposed changes. The other target was China. NAFTA rules of origin already had the effect of reducing Mexican imports from non-NAFTA countries. Further tightening was negotiated in order to reduce the scope for Mexico to use Chinese inputs in manufacturing operations. For good measure, the USMCA included a clause permitting one party (i.e. the US) to withdraw from the agreement completely if another party signed a trade deal with a “non-market economy”, (i.e. China).

Starting in early 2018, the administration began imposing tariffs on imports from all major trading partners, including China, the EU, Japan and Canada. The US tariffs were progressively expanded to cover nearly half the value of the Chinese goods that had been imported by the US in 2017, justified as tit-for-tat for the fact that the Chinese impose tariffs on over 70% of their – much smaller – imports from the US. A proposal to impose tariffs on automobiles imported by the US from various trade partners, including the EU, came into consideration at the end of 2018. But at the time of writing, the US was still talking about carrots as well as sticks: lifting tariffs in return for further tariff cuts elsewhere.

An optimistic view of this would be that by talking tough, the Trump administration might actually achieve further trade liberalisation. And the view that China should liberalise more, subsidise less and crack down on piracy and the infringement of intellectual property rights is widely shared. But attempts to bludgeon trading partners into liberalisation, even if they worked, would come at the expense of the rules-based trading system, on which the US will need to rely more as Chinese economic power grows.  


On the other side of the Atlantic the UK’s referendum vote to leave the EU is less straightforwardly linked to concerns about the effects of trade. There are free traders as well as protectionists in the Brexiteer camp – one of the reasons for its difficulty in agreeing on an exit route. Moreover, there is little evidence of a link between the extent of exposure to EU trade and the propensity of different regions in the UK to vote for Brexit.

There is, however, evidence that the intensity of exposure to import competition from China, and the loss of manufacturing jobs, had a strong correlation with voters’ inclination to tick the “leave” box. It is not difficult to see their votes as at least in part a resentment of the distributional effects of international trade.

Voters may have been aiming at the wrong target, and indeed unconsciously voting to make their situation worse, since the Brexiteers’ necessary ambition is to increase trade with those non-European economies which are in fact more likely than the EU to compete British factories out of a job. But the Leave slogan – “taking back control” – played right into anti-globalisation sentiment.

The problem this leaves for post-Brexit policy-makers was already evident in the disagreement between two leading Brexiteers – the UK’s Trade Secretary and its Environment Secretary – about the need for agricultural tariffs after leaving the EU. In the event, when the UK published the temporary tariff schedule it would apply if it were to leave the EU without an agreement, it became clear how difficult it would be to walk the line between cushioning consumers from price increases and protecting producers from sudden exposure to global competition.

Any future trade agreement between the UK and the EU is unlikely to be as deep or as comprehensive as the EU’s single market. So regardless of the final form Brexit takes, it will be the first major episode in modern times of a country choosing to reverse a process of deep economic integration with its major trading partners.  


Although developments in the US and the UK are the most high-profile disruptions of the global trading system, they are not isolated occurrences. We have seen a pronounced increase in global protectionism since the global financial crisis. Figure 13.6 indicates the balance of liberalising measures versus restrictive measures, implemented between 2009 and 2017, and imposed or agreed by the top five trading blocs (the United States, China, the EU, Japan and Canada) plus Brazil, Russia and India.

This is a very crude measure of protectionism – it is not weighted by strength or scale. So one single seriously restrictive import measure may have much greater effects than several supposedly liberalising ones, and vice versa. But on this measure at least, all bar one of the major trading nations are pulling back – i.e. restrictive measures outnumbered liberalising ones. The result is true globally by a ratio close to 2:1.

Nonetheless, the US stands out. As the Trump administration frequently argues, it started with freer trade than its competitors: the US has a lower average level of tariffs than any of the other countries in Figure 13.6. But it makes extensive use of non-tariff barriers such as trade “remedies”, and procurement measures, and did so well before the election of President Trump. The share of global imports affected by these measures rose from 2.4% in 2009 to 12.3% in 2016; to this must now be added the increase in tariffs.

For several decades, annual growth in trade outstripped growth in global GDP by a fair margin. This changed with the global financial crisis. Between 1999 and 2007, the annual average rate of growth in world trade was about 6.5%; from 2008 to 2016, it was about 2.8%, only slightly above the average growth in GDP (see Figure 13.7).

This is not, of course, all about protectionism. It may also be cyclical and structural. By 2014, China accounted for 10% of world trade; when its imports slowed, the rest of the world suffered. We are all more dependent on the Chinese economic cycle than we were, although stronger demand from other major developing countries – Brazil, India – may help to rebalance things in the future. It is also argued that the slicing-up of value chains described above, which artificially increased recorded trade, may have slowed as the limits of unbundling were reached. Nevertheless, the impact of trade disputes on policy, the retreat from internationalisation and the appetite for protectionism cannot be ignored. 


Source: Frontier Economics calculation based on data provided by the Global Trade Alert.
Note: Measures for EU are reported country by country. Some measures are country-specific; others are EU-wide. The EU total is based on the most liberal EU member state and the most restrictive EU member state.


Source: Based on World Bank World Development Indicators.


While Brexit and Trump’s protectionism may spring from similar roots, there is one aspect of the turbulence in the global trading system where their interests are very far from aligned. Much is made by Brexiteers – particularly those who favour “no deal” with the EU – of the safety net provided by the WTO, an institution barely mentioned in British political debate before (or even during) the referendum campaign.

“Falling back on WTO rules” is a position frequently touted since the referendum. By this is really meant that the WTO’s “Most Favoured Nation” (MFN) clause provides the UK a global trading platform, at least until bilateral deals can be negotiated. Few who make this argument, however, understand that MFN would mean the UK would have to impose the same level of tariffs on every country with which it traded, until such deals were made. Moreover, the working assumption behind this argument is that the UK could take advantage of a benign, open trading system buttressed by the liberalising influence of its largest members, and policed by an institution they supported.

That was already an unsafe assumption, and has become more so since. Consider, for example, the WTO’s dispute settlement system, often dubbed the “jewel in the crown”, because of its powers for enforcing compliance in a manner not available to other international tribunals. This, and the WTO’s surveillance functions, have helped – so far – to mitigate the risk that the global financial crisis would trigger a 1930s-style protectionist response.

But the system is close to breaking point. The US’s refusal to approve members of the WTO’s appellate body has raised the possibility that the dispute settlement mechanism will grind to a halt after 2019. In the meantime, US tariff measures on steel and aluminium have brought the WTO into troubled waters. This is because the US has couched its actions in terms of national security.

If the WTO were to rule against the US, the US would be likely to claim the organisation had overreached its jurisdiction, pick up its bat and ball, and leave. If the WTO were to rule in the US’s favour, other parties might follow suit, and negotiate bilaterally rather than rely on the WTO. Bilateralism is naturally favoured by the Trump administration, since it plays to the US’s strength.

Moreover, national security is not the only get-out clause trading nations like to be able to deploy. One country’s environmental, health and safety regulations are another one’s non-tariff barriers; restrictions on the import of (say) hormone-fed beef or chlorine-washed chickens may come under pressure in bilateral trade agreements, even more than they would before an impartial WTO panel.

Like all solutions to collective action problems, the WTO needs to meet a participation constraint (it must be worthwhile for parties to sign up) and a compliance constraint (it must be worthwhile for parties to accept punishments for breaking the rules, and to remain within the system). Clearly both have frayed as far as the US is concerned, and no global economic institution can maintain authority without the support of its largest member.

There are steps that can be taken to re-engage the US in support for the WTO, and indeed to revitalise the enthusiasm of other major members. But since the US always wins from bilateralism, it takes an administration with a strong appetite for global fairness to let go of the big stick. So while institutional reforms of the WTO are certainly worth discussing by its members, re-engaging the US will not be easy.  

On 23 June 2016, UK voters chose by a margin of 52% to 48% to leave the European Union







On 23 June 2016, UK voters chose by a margin of 52% to 48% to leave the European Union. Once the shock of the result had subsided, minds turned to the question as to what “leave” actually entailed. The referendum had proposed a binary choice (“in” or “out”) to the public on a matter that did not have a binary solution. What kind of relationship with our biggest trading partner should replace EU membership?

The prime minister’s “red lines” were drawn according to her understanding of the politics: abandoning EU-wide commitments on free movement, leaving the jurisdiction of the European Court of Justice (ECJ) and acquiring autonomy in domestic legislation and regulation, and being able to strike independent trade agreements.

From an economic perspective, the key challenge was how far these objectives could be pursued while maintaining as deep a level of integration with the EU as possible. Analysis after analysis showed that the adverse effect on UK trade of leaving the EU’s single market and reverting to trade with the EU on MFN terms simply could not be compensated for by any possible number of new free trade agreements with other countries.

Also lost in the rhetoric on immigration and autonomy was that losing access to EU labour markets and skills would penalise UK businesses, particularly knowledge-intensive services, while regulatory divergences would reduce the efficiency of cross-border supply chains. Meanwhile the UK faced another unique set of complications associated with its land border with the Republic of Ireland.

Unsurprisingly, therefore, the withdrawal agreement and political declaration, setting the framework for negotiations on future arrangements, that the prime minister hammered out found little favour with MPs before 29 March 2019 – the deadline the prime minister herself set by invoking the now-famous Article 50 of the European Treaty. As a result, she had to beg the EU for an extension. And when, or whether, she eventually succeeds in scraping together parliamentary support, this would only be the beginning.  



Whatever measures are taken to “reform” the WTO are unlikely to solve the distributional problems that have created the backlash against globalisation. For those, reform must begin at home.

A fundamental principle of economic policy-making, associated with the Nobel Prize-winning economist Jan Tinbergen, is that the number of policy instruments needs to match the number of policy variables targeted. And it is clear that trade policy, by itself, does not have the appropriate number or range of instruments to target the underlying distribution issues.

The policies that are needed fall under three headings:

  • Understanding the data and the economics well enough to ensure that the cure is not worse than the disease. In other words, making sure you do no harm.
  • Undertaking measures designed to mitigate distribution effects, or concentration of the gains from trade ex ante.
  • Supplementing these with measures that mitigate distribution effects, ex post, i.e. compensating losers for excessive losses.


Actions under the first heading include an effective re-engagement with the arguments by policy-makers used to taking the benefits of trade as a given. It is important to remake the basic case that trade benefits the economy as a whole, even if the benefits are not equally distributed.

Protectionism will not magically change low-productivity firms and workers into higher-productivity ones, while reducing international competition will restrict growth opportunities for more efficient firms. But at the same time, the argument has to recognise that there are winners and losers from trade – and that there is a role for government in mitigating these effects.

This does not mean succumbing to the old “lump of labour” fallacy, and assuming that all that matters is the distribution of work (and by implication reward); only a recognition that adjustment in economies that are being changed by globalisation can be painful for some.

This heading also covers a more sophisticated analysis and understanding of data that were created to inform policy in a simpler world. As the iPhone example showed, it is dangerously easy to reach the wrong conclusions about value added from balance of payments data. Tariffs based on such misunderstandings can be counterproductive.

Given that the value chains are cross-border, taxing one part of this affects the competitiveness of the value chain as a whole. That includes, for example, US suppliers of components to Mexico that are worked on and then sent back to the US.

It is also important to create a better understanding of the global trading system; the “rules” under which it operates and the constraints they impose, as well as their limitations in governing the trading practices of major players. Trade policy is complicated by all kinds of other policy imperatives – national security, environmental protection, health and safety – and reconciliation with these is an essential response to the backlash against globalisation.

In particular, concerns about the impact of increasing international trade on energy use and climate change have to be openly and thoroughly addressed. (This may be not be easy for a post-Brexit British government needing to shift UK trade from its European neighbours to more distant markets.)

From the UK’s perspective, one step the authorities ought to take, if they are serious about running an independent trade policy after Brexit, is to establish an independent statutory body that will have the mandate to provide objective analysis of policy questions. That would go some way in providing a basis for informed discussion of trade policy options, something in which the post-referendum debates have been shown to be particularly lacking. A good model has been developed in Australia, with its Productivity (formerly Industry) Commission, which has (including under its previous incarnation) played a central role in supporting the trade and microeconomic reform agenda in that country.

At an international level, the UK and other G20 nations should invest in beefing up the WTO’s surveillance system, under the auspices of the Trade Policy Review Mechanism (TPRM). The creation of that mechanism was one of the key achievements of the Uruguay Round, and it was supposed to sit alongside the WTO’s dispute settlement mechanism and the General Council as one of the key planks of trade governance. The objective was that peer review would facilitate not only a better understanding of trade practices, but encourage reforms through the dissemination of best practices, separate from the politically charged arenas of dispute settlement and trade negotiations.  


Policies to be developed under the second heading are, in fact, only a more trade-focused version of those needed to see that any form of economic growth is reasonably well spread. The self-reinforcing effect of clusters – of the concentration of skills, education and high earnings – are general problems that require government intervention if groups of society and even whole regions are not to be heavily disadvantaged.

Under this heading therefore come:

  • Education policy: the need to ensure schools, skills training and universities in less advantaged areas are supported to meet the standards of those benefiting most from trade opportunities.
  • Regional policy: the need to stimulate and support “clusters” of economic activity in less prosperous areas, with ideas like George Osborne’s “Northern Powerhouse” in the UK.
  • Infrastructure: the need to ensure all regions have sufficient physical and digital communication networks to enable them to participate in the opportunities offered by international trade.
  • Industrial policy: empirical research shows the important role played by “superstar” exporting companies in the structure of trade. One study showed that in 32 countries, the largest exporter accounted for an average of 14% of the total, while in one country that rose to about a third of total exports. Such companies have long value chains, creating business among suppliers in their home countries as well as abroad.

This last factor raises more complex policy questions. Governments cannot usually engineer the emergence of such superstars. But they can create conditions that make their emergence more likely. The key here is to pay attention to the economics of geography, and especially how agglomeration and clustering effects develop. The deeper such clusters, the greater the possibility that superstar firms may emerge, and the greater the chances they create an ecosystem of firms around them.

Immigration policy may also play an important part. Trading companies stress to government their need of easy access to specialist skills, for which there may also be global competition. The free movement of people under the EU single market provisions has played a very important role in developing the UK’s position in financial and professional services. Although the UK Government has stressed its willingness to develop a skills-based immigration policy, concerns remain about the effect of withdrawing from the single market in labour.  


The third category of measures – ex post correction of distributional effects – is clearly less satisfactory than the second. Even if high taxes on the companies and people on the upward spiral of trading success did not create a risk of blunting incentives, the ease with which, assisted by digital technology, economic activity can shift from one tax jurisdiction to another sets a constraint on tax levels. And this, in turn, sets constraints on the scale of benefits. Nonetheless, a tax and benefit system that is seen as fair is critical to the maintenance of social cohesion.

There remains a sense of helplessness and alienation in areas denuded of jobs by international competition which no government hand-outs can cure. Governments, therefore, would do well to focus on measures of well-being, and policies targeted on improving these as much as incomes, to give effective ex post support to the losers from trade. The growing sophistication and policy uses of such measures are another trend of the past 20 years that forms the subject of another article in our 20th Anniversary collection.  


This article began with the story of the WTO’s creation. Its subsequent travails should not obscure its successes over nearly a quarter of a century of existence. The number of members has continued to rise, from 123 at its birth to over 164 today. Over 500 trade disputes between them have been brought to the WTO for settlement, and about 350 judgements have been issued, making it the world’s most active mechanism for the settlement of global inter-governmental disputes.

Bringing China into the rules-based international trading system, perhaps the most important event in the WTO’s history, has helped to boost the global economy, even as it has created some of the tensions described in this article. But, as this article also makes clear, the system is now under unprecedented stress.

The backlash against globalisation is strong. It will take all the “better answers” Krugman ruefully called for to prevent a slide back into protectionism, and a consequent loss of the growth benefits from expanding trade.