How will the future of globalisation play out? - Capital Economics
Emerging Markets Economics

How will the future of globalisation play out?

Emerging Markets Economics Focus
Written by Andrew Kenningham
Cancel X

While the downside risks of a global trade war are often overstated, we think the bigger concern is that the world splits into competing regional blocs that do not cooperate on trade or investment. This could cause global supply chains to splinter and technology transfer to seize up, ultimately leading to much weaker global growth. It would also be fertile ground for other unorthodox economic policies.

  • While the downside risks of a global trade war are often overstated, we think the bigger concern is that the world splits into competing regional blocs that do not cooperate on trade or investment. This could cause global supply chains to splinter and technology transfer to seize up, ultimately leading to much weaker global growth. It would also be fertile ground for other unorthodox economic policies.
  • The future of cross-border trade and investment could take many forms, each of which would have different implications for economic growth and inflation. In this Focus we consider four broad scenarios.
  • The best outcome for global growth would be a fresh wave of globalisation driven by new technological breakthroughs and a resumption of international cooperation. That would lift productivity growth and be particularly helpful for those emerging economies that are best placed to adopt new technologies and have low labour costs. In practice, though, we think that such an outcome is unlikely.
  • A more plausible scenario is that globalisation has already reached its natural limits: supply chains have expanded as far as possible and new technologies, such as 3D printing and robotics, along with changing consumer preferences, are causing them to be shortened. This scenario would be neutral for global growth but would have big distributional implications: it would make life harder for many low-income countries.
  • The third scenario is a trade war in which many governments erect tariffs barriers. The damage from an all-out trade war would probably be less many assume. But we still estimate that it would reduce global GDP by 2-3% in the near term and that trend economic growth would slow by around a tenth.
  • Perhaps the most likely scenario is that we are moving to a world of regionalisation, though this itself could take various forms. A mild version would be less damaging than an all-out trade war. A lot of trade already takes place within regions and would be unaffected. Regions as big as North America, Europe or emerging Asia, would be able to support large multinational firms and enough competition to sustain economic growth. And technological progress would continue more or less unhindered.
  • A more severe version of regionalisation could be highly damaging. Anything that caused foreign direct investment to dry up would be a considerable drag on economic growth. A technological iron curtain, an end to transfer of intellectual property, or a divided currency and financial system could lead to much lower global productivity growth and a significant divergence between regions.
  • In practice, these scenarios are not mutually exclusive and elements of most of them are already visible. The net impact of all these changes is likely to be a longer period of subdued global growth and a tougher environment for many emerging economies.
  • There could also be significant indirect effects from de-globalisation. Unorthodox policies are likely to spread as peer pressure exerted by multilateral organisations and the discipline of financial markets are reduced. And if – as seems likely – China decouples from the rest of the world, it may devote more resources to strategic objectives and double down on its state-led development model. These policy shifts could have bigger implications for economic growth and inflation than the direct effects of protectionism.

How will the future of globalisation play out?

While there are numerous ways in which cross-border trade and investment might develop, we have grouped them here into four broad scenarios: a new wave of globalisation; globalisation reaching its natural limits; a global trade war; and the world splitting into competing economic regions. For each, we set out the likely implications for economic growth, at a global and regional level. And we discuss how a process of de-globalisation may influence economic policy.

This Focus is part of a series of reports about the end of globalisation and the impact of this on the global economy and financial markets. (See here for more.)

Scenario 1. A new wave of globalisation

The first scenario is one in which there is a new wave of globalisation. This could, for example, be due to a technological breakthrough which further reduces travel or communication costs and gives a new lease of life to global value chains. And it would only be possible if there were a reversion to greater international economic cooperation.

In this scenario, global potential economic growth would be higher than it would otherwise be. And experience suggests that emerging economies would benefit the most. (See Chart 1 and our Global Economics Focus, “Globalisation and its effects on the world economy”, 10th October.)

Chart 1: GDP per Capita (5yr Average, % y/y)

Sources: World Bank, Refinitiv

Many policymakers still seem to believe that the world is likely to revert to this course in the near future, as suggested by a recent World Bank report, for example. But we think this is quite unlikely, for several reasons.

First, as we demonstrated in an earlier Focus from our end of globalisation series, such technological waves are rare and, in any case, take many years to transform the economy. (See our Global Economics Focus, Lessons from the history of globalisation”, 8th October.) Second, many of the latest innovations have been trade-reducing. Third, transport and communication costs are already so low that there is little chance of further reductions triggering a big change in the production process. And fourth, several other factors are pushing the world towards shorter supply chains: these include households purchasing more non-tradable services and mounting policy uncertainty. (See our Global Economics Focus, Does globalisation have a future?”, 15th October.)

Scenario 2. Globalisation reaches its natural limits

The second scenario is that globalisation reaches its natural limits. In this case, cross-border trade and investment are either stable or falling as a share of GDP, because of new technologies and changing consumer preferences.

For the world as a whole, we think this scenario would be broadly neutral. Companies in the developed world would re-shore some activity and consumers would benefit from more customised products and shorter lead times. The world economy would simply be responding to changing technology and consumer preferences rather than being saddled with policy-driven barriers to trade and investment.

While the aggregate impact may be neutral, there could be big effects on the distribution of income between countries.

Just as offshoring particularly benefitted some emerging economies, so reshoring may have the opposite effect. Most advanced economies and some of the more developed emerging economies, including China, would be well placed to exploit new technologies that are causing supply chains to shorten, such as robotics, Artificial Intelligence (AI) and 3D printing.

As an example, Chart 2 shows the number of robots per capita in a handful of major economies. Among other things, it shows that developed economies in Asia are particularly heavy users of robots. Indeed, there are more than seven times as many robots per person in Korea as there are in the US.

Chart 2: Industrial Robots Per Million Population (2017)

Source: International Federation of Robotics

Things will change over time, but we suspect that the early adopters of these technologies will continue to do well from the spread of AI and robotics at least for the next few years.

In contrast, such technologies will probably make life more difficult for many of the world’s poorest countries. After all, the most common development path for low-income economies begins with labour-intensive manufacturing in sectors such as textiles and footwear, followed by electronics and food processing. These new technologies will cause some of this work to migrate to wealthier emerging economies or to the developed world.

For example, Adidas now produces some training shoes in Germany and the US using 3D printers. That is a threat to countries such as Vietnam, Cambodia and Albania, which are competitive in that sector as shown, for example, by the importance of exports of footwear in their economies. (See Chart 3.)

Chart 3: Footwear Exports (% of GDP)

Source: Intracen

Of course, concerns about the end of globalisation are not primarily focused on new technologies, or for that matter on the continued growth of the services sector. Instead, the big worry is the spread of protectionist policies.

Scenario 3: A global trade war

Last year, we assessed the economic damage from a trade war in which every country levies a tariff of 25% on the rest of the world. (See our Global Economics Focus The damage from a global trade war”, 9th July 2018.) This assessment had two parts.

The first is that a trade war would deliver a one-off hit to output, which would affect small open economies the most. We think the reconfiguration of supply chains would knock two or three percentage points from world GDP over a year or two, relative to what it would otherwise have been.

Chart 4: Goods & Services Trade (% of GDP)

Source: Refinitiv

The costs for individual countries would vary. As Chart 4 highlights, countries such as the Netherlands, Switzerland and Sweden are particularly vulnerable because they are so trade dependent. Brazil, the United States, and even Japan (which is often thought of as an export-oriented economy) are less vulnerable.

The second part is the impact which an every-country-for-itself trade war would have on potential economic growth. To get a handle on how big a drag on global growth across-the-board 25% tariffs might be in the long run, we have decomposed average world GDP growth from 1995 to 2017 into contributions from various factors. (See Chart 5.) This gives us a sense of the portion of potential growth that is at stake from a lurch to protectionism.

The single largest contribution to global GDP growth since the mid-1990s has been employment, with growth in hours worked accounting for about a third of GDP growth. A rollback of globalisation should not affect the contribution of labour to GDP growth in the future, since this will continue to be determined instead by demographics, trends in labour market participation and preferences for work rather than leisure.

Chart 5: Estimated Decomposition of Average
World GDP Growth (1995 – 2017)

Sources: Penn, World Development Indicators, PRS, Capital Economics

Growth of the capital stock, particularly of machinery & equipment and ICT, has added another big chunk to global GDP growth. This leaves roughly 1.5%-points, by our estimation, unexplained by growth in the quantity of labour and fixed capital inputs. This residual is typically known as total factor productivity (TFP).

Part of TFP growth can be accounted for by improvements in human capital – i.e. in a better educated and healthier global workforce. Institutional changes have had an influence too. For instance, data from Political Risk Services indicate that many economies have experienced increases in the independence and the quality of public sector bureaucracies, in addition to more effective regulations governing investment.

If we strip the estimated contributions of human capital and institutions out of TFP, we are left with 0.7%-point of average global GDP growth that is still unaccounted for. This will include the effects of trade and financial openness on global productivity growth. But it will also include the effects of myriad other factors such as changes in product market regulation and competition, increased migration and labour market flexibility, development of financial systems and improvements in corporate governance. Moreover, there will be a whole host of innovative ideas and processes (forms of so-called “intangible capital”) which are difficult to pick up in the national accounts data but are nevertheless important for putting production inputs to effective use.

So, the actual contribution of past globalisation, and the potential hit from future protectionism, to trend GDP growth is likely to be much smaller than the 0.7%-pt residual shown in pink in Chart 5.

Admittedly, in such an extreme scenario as the all-out trade war that we are currently considering, there probably would be long-run spill-overs to other drivers of growth such as fixed capital formation. But we must also bear in mind that most of the hit to productivity growth from higher trade barriers would come in the tradable goods sectors. There would be only limited knock-on effects to non-tradable sectors which make up 70% of global economic output. For example, public services – the largest industry at the two-digit level of classification – would scarcely be affected at all. (See Chart 6.)

Chart 6: Trade as a % of Industry Value Added vs. Industry Share of Total World Value Added (%)

Sources: OECD ICIO, Capital Economics

All told, across-the-board tariffs would dampen investment growth and reduce competition, but many of the drivers of growth would not be much affected. Our best guess is that across-the-board 25% tariffs would subtract only around 0.3%-points from trend global GDP growth. This is roughly a tenth of the annual rate which we would otherwise expect to world economy to grow.

Chart 7: World GDP (2020 = 100)


Chart 7 brings together the two elements of our estimate of the effects of a trade war. The higher, solid line shows our projection for world GDP assuming there is no trade war and the lower, dashed line shows our projection if there is a trade war.

In the trade war scenario, world GDP is roughly unchanged for just over a year, reflecting the impact of the reconfiguration of supply chains. It then resumes its upward trend but at a slightly slower pace. While this may not look too bad at first sight, the costs add up over time; by 2030 the world economy would be 4 or 5% smaller than it would otherwise have been.

In practice, while some elements of this trade war scenario have already taken place, an extreme trade policy free-for-all is not really on the cards. After all, Europe is likely to remain a largely integrated economic area. Moreover, governments of small or medium-sized countries ranging from Hong Kong and Singapore to Thailand and Mexico, are unlikely to impose high tariffs on their key trade partners because of the damage that would do to their own economies. A more plausible scenario is one in which the world is increasingly organised into competing regional blocs.

Scenario 4: Regional splits (mild version)

There are numerous ways in which the world economy could split into regions. Perhaps one of the most likely is that it divides into three broad areas: a US-led bloc, a European bloc and a China-led bloc.

Clearly, the consequences of such a split would depend on the extent of any barriers between regions. At one end of the spectrum, this could entail disengagement from multilateral institutions and a gradual build-up of regional trade defence policies. At the other is a clamp-down on technology transfer or full-blown Soviet-style autarky.

Provided the split were not drastic, there are several reasons why regionalisation should be less damaging than a country-level trade conflict. First, a lot of trade takes place primarily within regions and would not be much affected. This is illustrated in Chart 8, in which we have split the world into five regions: Asia-Pacific; Europe; NAFTA; South America; and the Rest of the World. For example, Mexico and Canada trade overwhelmingly with the US. Similarly, Australia trades mainly with emerging Asia. And the Netherlands, like most European countries, trades mainly with its neighbours.

Chart 8: Within-Region Share of Total Exports (%)

Source: Refinitiv

Second, the regions would probably be big enough to sustain companies that achieve maximum economies of scale. And they would be large enough to ensure that there was plenty of competition within each market.

Third, technological progress would not grind to a halt, even in the emerging world (which has typically relied on technology diffusion from the G7). Research and development is less concentrated than it was in the past – the share that takes place in advanced economies has fallen steadily since around 2000, at the expense of an increase in the share in emerging Asia, most of which is in China. (See Chart 9.)

Chart 9: R&D Expenditure (As a % of Global R&D)

Source: Refinitiv

In addition, we think governments would struggle to prevent technology from spreading between regions even if they wished to do so. That would require a reversion to the sort of extreme authoritarianism which prevailed during the Cold War – when East German auto manufacturers famously produced Trabants while those in West Germany made BMWs.

Earlier we said a trade war might reduce trend global growth by around 0.3%-points. A mild form of regionalisation would reduce it by less.

Scenario 4: Regional splits (severe version)

The damage from a split into regions would be much bigger if other economic barriers were erected. In Table 1 we have listed some of the possibilities.

Table 1: Extreme Regionalisation

Policy Risk


Crackdown on foreign investment.

Less competition and technology transfer.

Technological iron curtain.

Splinternet, banning foreign parts in high tech goods.

Financial sanctions.

Payments system frozen.

Travel restrictions.

Hit to tourism, higher education, multinationals.

Source: Capital Economics

Chart 10: Output of Subsidiaries Operating in Other Countries (% of Total Output, 2014)

Source: OECD

For a start, the consequences of a tougher crackdown on foreign direct investment, or FDI, could be very serious. As shown in Chart 10, production by subsidiaries operating abroad is particularly significant for countries such as Switzerland, the Netherlands and Sweden. But it is also highly significant for US-based multinationals, including the major tech firms, as well as the larger German, French, British and Japanese firms.

A reduction in inward investment would also dampen domestic competition in many markets. The European auto industry is a good example. European car firms have been sheltered behind high tariff barriers for decades, but the presence of firms like Nissan and Toyota means that they have to keep up with Japanese innovations. It follows that the damage from regional protection would be much larger if foreign-owned companies were heavily taxed or even banned outright.

Another possibility is that a sort of technological iron curtain comes down between China and the rest of the world. Already, technology firms such as Google and Facebook don’t operate in China. And because of the US-China trade tensions Huawei’s latest smartphone is being launched without features such as Google Maps or Youtube. Things could go much further, impeding companies from sharing data or communicating securely, which in turn could make some global supply chains difficult to sustain.

On top of this, the banking system could split into two zones, possibly using different currencies, echoing what happened when the rouble was used almost exclusively in the Soviet bloc. After all, the US has used its control of the SWIFT payments system to prevent non-US companies doing business with Iran: it could presumably do the same to China.

For its part, China is attempting to reduce its dependence on the dollar by promoting trade in renminbi with Iran, Russia and elsewhere. And the current protests in Hong Kong may lead China to further reduce its reliance on the dollar-based financial system. (See our China Economics Focus, “Decoupling, and its impact on growth”, 17th October.)

Inflation and policy

Having touched on several scenarios, we now turn to the question of how de-globalisation might affect the outlook for inflation and the impact that it may have on economic policy.

The fact that the latest wave of globalisation coincided with a sustained fall in inflation, as shown in Chart 11 overleaf, may suggest at first sight that a reversal of globalisation will have the opposite effect.

However, we think that the impact of globalisation on inflation over the past few decades has often been overstated. The big reduction in inflation took place in the 1980s as a result of tighter monetary policy, before the third wave of globalisation really got under way. As we have argued before, changes in technology, the labour market and central banking have been at least as important in achieving price stability. (See our Global Economics Focus, “Globalisation and its effects on the world economy”, 10th October.)

Chart 11: CPI Inflation in Advanced Economies (%)

Source: Refinitiv

With the possible exception of independent monetary policy, these forces seem likely to remain in place for the foreseeable future, so there is little reason to fear that the end of globalisation will lead to a period of high or rising inflation.

We are more concerned about the way in which protectionist policies may influence broader economic policymaking in the coming years. For a start, the roll-back of globalisation may create conditions in which it is easier for governments to turn away from the set of policies and norms often referred to as the Washington Consensus.

Admittedly, countries such as Argentina, Russia and Turkey have resisted much of this agenda in any case. But the spread of protectionist policies and withdrawal of support for multilateral institutions may erode norms of economic orthodoxy, meaning that others are more likely to break rank and follow suit. This could encourage politicians to adopt looser fiscal policy, to take back control of monetary policy and to rely on “financial repression” to inflate away public debts. They may also step up their involvement in commercial decisions and use of trade barriers for foreign policy purposes.

The Trump administration is leading the way on much of this. It has put pressure on the Fed to cut interest rates; ordered some companies not to do business with Mexico or China; and threatened trade barriers on countries such as Turkey in pursuit of foreign policy goals. In many emerging economies, including Brazil, Mexico and the Philippines, the rhetoric has shifted away from unconventional policies in recent years. So far, policy hasn’t been as radical, but it may be only a matter of time.

A world which splits between a China-led and US-led economic bloc may also encourage more inefficient policymaking in China, where the strategic imperative to catch up with the US in technology could lead to even more resources being ploughed into politically-driven projects.

Globalisation scenario outcomes

Table 2 shows the scenarios with which we began this Focus. The columns show the likely impact of each scenario on growth, inflation and the equilibrium real interest rate, or R*, relative to the current situation.

Table 2: Summary of Outcomes





New wave

Natural limits

Global trade war

Regionalisation (mild)

Regionalisation (severe)



Source: Capital Economics

In brief, we think that a new wave of globalisation would boost world GDP growth and help to reduce inflation. To the extent that higher potential GDP growth ought to lead – all else equal – to higher equilibrium interest rates, a new wave of globalisation should cause R* to rise, as we’ve indicated in the Table.

However, this relationship isn’t so clear cut in practice. If a new wave of globalisation were to exacerbate income inequality, for example, then higher desired savings would exert downward pressure on R*. So, the net effect is somewhat ambiguous and would depend to some degree on what was driving the new wave.

Globalisation reaching its natural limits would be broadly neutral for the world economy, although there would be winners and losers with respect to reshoring. Some of the poorest emerging economies would be the worst affected.

A trade war and a mild form of regionalisation would both have a negative impact on global GDP growth but little impact on inflation.

And an extreme regional split between China and the rest of the world could do a lot of damage to economic activity. This in turn could cause real interest rates to fall.


In practice, the world does not fit neatly into scenarios and, in any case, those which we have considered are not mutually exclusive. We are already witnessing some elements of a trade war, the natural limits to globalisation and regionalisation. Most likely, there will continue to be some mix of these different scenarios.

Nonetheless, this Focus has suggested three main conclusions.

  • First, fears about the economic fallout from protectionism are often overdone. In most of the scenarios we have considered, the effects for the world as a whole would be manageable and would probably be outweighed by other factors.
  • Second, just as emerging economies generally benefited from globalisation, so they are likely to lose out from de-globalisation. That is particularly true of countries which have not yet managed to get their foot firmly on the development ladder and of those which do not have a large domestic market to drive growth.
  • And third, the de-globalisation scenario that is most concerning is a deep split between a China- and US-led economic bloc. This could potentially lead to a permanent reduction in trend GDP growth relative to the status quo.

Andrew Kenningham, Chief Europe Economist, +44 20 7708 4698,
Simon MacAdam, Global Economist, +44 20 7808 4983,