Opinion & Analysis

Is globalisation really doomed?

Globalisation is under attack; to preserve its benefits, healthy domestic social contracts are essential.

By André Sapir

There is a view that globalisation is in retreat. It has produced immense wealth, but has also generated huge inequality, leading to an economic and political backlash. Not everyone agrees with this view, though globalisation is seen as under geopolitical threat, in particular because of the rivalry between China and the United States, and the war between Russia and Ukraine, which both pit autocracies against democracies. So, is the current wave of globalisation that started three decades ago already over or at least doomed, and if so, should we rejoice or lament?


Two waves of globalisation

The earlier wave of globalisation of the modern era started in the mid-1800s. It resulted from technological advances in transportation and communication that dramatically reduced transactions costs, and from political decisions, mainly by Europeans, to liberalise their own markets and force open the markets of others through colonialism or imperialism. This period lasted four to five decades and ended in 1914. It was followed by a period of deglobalisation marked by strong constraints on trade and investment flows that lasted three decades.

After the Second World War, a period of re-globalisation started. By the early 1980s, global trade integration had returned to the level that it had reached in 1914. The main driving force for this integration was trade liberalisation in the advanced economies, while vast parts of the world (mainly China and the Soviet Union, but also India) continued to be marked by significant state intervention and even autarkic policies.

Then, in the early 1990s, the world entered a period of hyperglobalisation, for the same reasons that underpinned the first wave of globalisation: technological advances in transportation and communication, and political decisions – though this time by countries including China and India acting autonomously rather than under the constraint of foreign powers. Together, these two factors allowed manufacturing firms based in advanced economies to source labour-intensive products or components from low-cost labour locations. Global supply-chain trade between advanced and developing or emerging economies was the main driver of the process of hyperglobalisation, accounting in 2008 for more than half of world merchandise trade.

Historically, the ratio of world trade (in goods and services) to world GDP had never reached more than 20%. After 1990, it grew steadily, peaking at 31% in 2008, at the start of the Great Recession (Figure 1).

After the Great Recession, hyperglobalisation stalled and was followed by a phase that has been described alternatively as deglobalisation or ‘slowbalisation’, depending on the indicator used. The ratio of world trade to world GDP has remained around the level it reached at its peak in 2008, although it dipped again in 2020 because of COVID-19(Figure 1).

However, looking separately at trade in goods and trade in services shows a different picture than for goods and services trade together. Deglobalisation seems to have already started for merchandise trade, with the ratio of world trade to world GDP down significantly from its peak in 2008. But for trade in services there seems to be neither deglobalisation nor slowbalisation, but rather an increase in globalisation, with the ratio of world trade to world GDP continuing to increase after 2008, though with a COVID-19-related dip in 2020.

Like the first wave of globalisation, the period of hyperglobalisation produced huge economic benefits. Deeper trade specialisation associated with complex global supply chains have enhanced productivity and income growth, which helped convergence between emerging and advanced economies.

But hyperglobalisation has also produced major economic and political tensions between and within nations that have cast doubt on its sustainability.

The old international division of labour between advanced and developing economies, which lasted about a century and saw the former specialising in manufactured goods and the latter in agricultural products or raw materials has transformed dramatically. Today, developing countries have become major competitors for advanced economies in many manufactured products, while also being important buyers from these countries in many other products.

The system has also struggled increasingly to deal with the tensions generated by the greater international competition between advanced and developing economies. The ‘system’ in the present context refers to both domestic labour market and social institutions that can attenuate domestic social tensions, and global governance institutions, especially the World Trade Organisation, that can reduce international trade tensions. Unfortunately, some of these domestic and international institutions have been weakened in recent years, as the next two sections discuss.


The political economy of globalisation in the advanced economies

It did not take long for hyperglobalisation to produce a backlash first in the United States and then in other advanced economies.

Dani Rodrik in 1997 in his book Has globalisation gone too far? set out convincingly that, with increased trade and capital mobility, “the services of large segments of the [American] population can be more easily substituted by the services of other people across national boundaries.” Consequently, globalisation reduces the bargaining power of workers, who experience increasing instability in earnings in countries like the United States, where labour markets are flexible. For Americans “who lack the skills to make themselves hard to replace,” Rodrik argued, globalisation means “greater insecurity and a more precarious existence.” The median American voter lost wages, resulting in labour’s fierce opposition to globalisation.

In Europe, the situation was very different. Because in Europe the welfare state is more generous and markets are less efficient than in America, globalisation generated less wealth but also less income inequality and adjustment problems. Hence, the European median voter suffered relatively little compared to America. Unemployment increased, but mainly among ‘outsiders’: the young and immigrants. Organised labour in Europe thus voiced less opposition to globalisation than in America. Sustaining globalisation thus posed different challenges to America and Europe. In the United States, the main challenge was to ensure better distribution of income and to reduce poverty. In Europe, it was to reform welfare states so they could confront rising demands and dwindling resources.

There are of course substantial differences between national welfare states in Europe, and some systems are better equipped than others to handle the social, economic and political consequences of globalisation. A 2005 study assessed social systems on the basis of two criteria: efficiency and equity. A system can be considered efficient if it provides sufficient incentives to work and generates relatively high employment rates; it can be deemed equitable if it delivers a relatively low risk of falling into (relative) poverty. The pre-2004 enlargement European Union countries fell into four categories: those with a social model that delivered both efficiency and equity (Austria, Denmark, Finland, the Netherlands and Sweden, labelled as ‘Nordic’); those with a social system that delivers neither efficiency nor equity (Greece, Italy and Spain, labelled as ‘Mediterranean’); and those with a trade-off between efficiency and equity, with two variants: those with equitable but inefficient models (Belgium, France, Germany and Luxembourg, labelled as ‘continental’; and those with efficient but inequitable social models (Ireland, Portugal and the United Kingdom, labelled as ‘Anglo-Saxon’) (Figure 2).

This typology of European social models suggested that there was a strong case for reforming European labour market and social policies, especially in continental and Mediterranean countries, where social systems were judged inefficient. By relying on strict employment protection laws at a time of rapid change when old jobs and practices were no longer warranted because of technological change and globalisation, these models discouraged adaptation to change and preserved the status quo. The result was low employment rates, which endangered the viability of the welfare state.

The desired change towards greater efficiency in Europe has largely occurred since 2004 (Figure 3). In nearly all pre-2004 enlargement EU countries, the employment rate increased – sometimes substantially – between 2004 and 2019. In most countries, an increase in the efficiency of the social model has not been accompanied, as may have been feared, by an increase in inequality. In particular, Germany, the country with the biggest increase in the employment rate (a gain of 11 percentage points between 2004 and 2019) did not see any change in the risk of its population falling into poverty. In one country (Ireland) the increase in the employment rate was accompanied by a substantial reduction in inequality, but in two others (Luxembourg and Sweden) it was accompanied by a substantial increase in inequality.

Figure 3: Changes in employment rates and the probability of escaping poverty in European social systems, 2004-2019 (percentage points)

The outlier in Figure 3 is Greece, where the employment rate has fallen, but thankfully inequality has decreased too. The situation in the two other countries previously classified as Mediterraneans (Italy and Spain) is also not very brilliant, with a small improvement in the employment rate but at the expense of a small increase in inequality. Altogether, therefore, most of the countries in Figure 3 were in a much better position to deal with globalisation in 2019 than in 2004, with the exception of Greece, Italy and Spain, which continue to have social models that are neither efficient nor equitable and hence remain fragile, not only in the face of globalisation but also other major transformations like the green and digital transitions.

Therefore it can be concluded that the view that globalisation was doomed because the social contract in the advanced economies is broken is generally more relevant for the United States than for Europe. Dani Rodrik, in a 2016 article, was right that politicians “should focus on restoring the domestic social contract,” but also that [h]ealthier polities produce — and can withstand — more globalisation.”


The geopolitics of globalisation

The reversal of hyperglobalisation may be in part driven by domestic politics, but geopolitics plays a major role too.

With the end of the Cold War, there was a widely shared sentiment in the West that the world had reached the End of History: “That is, the end-point of mankind’s ideological evolution and the universalization of Western liberal democracy as the final form of human government,” according to Francis Fukuyama. This was one of the reasons why WTO members decided to admit China in 2001. The prevailing view in Brussels and Washington, at the time, was that China’s WTO membership was bound to transform the country into a Western liberal market economy and democracy. When it became clear (after the Great Recession and the rise to power of President Xi) that neither would happen soon, there was huge disappointment in many capitals. In Washington there has also been fear that, having already become the world’s largest exporter of goods and perhaps soon the economy with the largest GDP, China is fast becoming a major geopolitical rival.

Against this background, the dependency of the United States on imports from China, and of China on exports to the United States, led the two rivals to adopt measures to curtail their bilateral trade after the Great Recession, especially after President Trump assumed office. The COVID-19 pandemic was a further reason for many economies, including the European Union, to adopt measures to gain ‘strategic autonomy’ and to reshore production, mainly from China. The final nail in the coffin of hyperglobalisation was Russia’s invasion of Ukraine, which has reinforced the voices of those claiming that trade dependency on economies with illiberal political regimes had gone too far and that near- or friend-shoring is essential.

Similar calls for self-reliance abound in China, where President Xi believes that his country has become too reliant on liberal democracies as markets for its exports and suppliers for crucial inputs. All this may or may not qualify as deglobalisation, but it certainly sounds like partial decoupling. Gone is the time when trade was simply a matter of comparative advantage and mutual gains as was the case during the era of hyperglobalisation. Geopolitics has not entirely replaced economics in shaping trade (and international investment) flows, but it definitely plays a much bigger role than during the heydays of the second wave globalisation.



A healthy domestic social contract is essential in general, and to sustain globalisation in particular. But fears are increasing that in countries where the domestic social contract is lacking, policy will deliver protectionism rather than improved social policies. If this happens, deglobalisation will not reverse the trend towards greater income and wealth inequality seen in many advanced countries, though more in the United States than in Europe, during the relatively short period of hyperglobalisation. Instead, by decreasing international competition, deglobalisation risks worsening the plight of workers through lower labour productivity and higher prices for some of the goods they consume. Those who genuinely seek better social protection for workers – to better withstand not only globalisation but also the digital and climate transitions – would do well to remember that this will require better social policies rather than restrictive trade policies.


This is a shorter version of an article originally published in Spanish by the Barcelona Centre for International Affairs.

About the author

André Sapir, a Belgian citizen, is Senior Fellow at Bruegel. He is also University Professor at the Université libre de Bruxelles (ULB) and Research Fellow of the London-based Centre for Economic Policy Research. Between 1990 and 2004, he worked for the European Commission, first as Economic Advisor to the Director-General for Economic and Financial Affairs, and then as Principal Economic Advisor to President Prodi, also heading his Economic Advisory Group. In 2004, he published ‘An Agenda for a Growing Europe’, a report to the president of the Commission by a group of independent experts that is known as the Sapir report. After leaving the Commission, he first served as External Member of President Barroso’s Economic Advisory Group and then as Member of the General Board (and Chair of the Advisory Scientific Committee) of the European Systemic Risk Board based at the European Central Bank in Frankfurt.André has written extensively on European integration, international trade, and globalisation. He holds a PhD in economics from the Johns Hopkins University in Baltimore, where he worked under the supervision of Béla Balassa. He was elected Member of the Academia Europaea and of the Royal Academy of Belgium for Science and the Arts.

Access the original publication here