The perfect storm: Deglobalisation’s headwinds

Globalisation - the rising interconnection of the world's economies and populations, driven by international trade in technology, services, goods as well as the flow of investment and information - is facing at least five different head winds. Let me discuss them in turn and indicate how I believe they could be tackled.

Distribution effects

This is an old theme in the discussion of trade.  Trade makes countries better off but some workers and some firms lose from it. This not an incidental implication; trade is about producing where it is cheaper to do so, with the implication that some, more expensive, domestic producers will suffer.   

We know that trade is not the only, or even the main source of job losses. But it is easier to identify the source of the losses and attribute the blame when it comes from higher imports or from the relocation of firms to other countries.

Why is the argument more salient now? One reason is the overall failure of the programmes designed to help those workers who lose jobs because of trade. Another is that the culprit behind the job losses is easier to identify when one can point to a specific country, as has happened with the ‘China shock’ narrative. Reputable estimates suggest  gross job losses as a  result of rising Chinese import competition during the 1999–2011 period were in the range of 2.0–2.4 million¹  - and given the distribution of these losses across states, plausibly resulted in Donald Trump’s election win in 2016.  Another example is that of French farmers mobilising against cheaper agricultural imports from Poland and Hungary and blaming European Union (EU) rules.

As is the case with the other forces discussed below, these concerns cannot be dismissed. Trade adjustment assistance programmes have not worked, and probably cannot work. And it may make sense to protect some sectors or workers on distribution grounds. For example, it may make sense to protect small French farmers, even if their products are more expensive. This translates into higher prices for French consumers, but maybe this is still a socially acceptable price to pay to keep the French countryside alive and avoid the rise of ‘geographic deserts’. The obvious issue is how to do it without losing most of the gains from trade.

National security

In a world where geopolitical tensions have increased, it obviously makes sense to avoid depending on non-friendly countries, be it the dependence of the US on trade with China, the dependence of the world on rare earth minerals coming in part from politically unstable countries, or the dependence of the world on Taiwan for semi-conductors.

The experience of Germany in decreasing its dependence on Russian gas over the last three years suggests that firms and countries can adapt, change the method of production, and find alternative suppliers. Still, adapting in the very short run to an embargo or worse can be difficult, and it makes sense to take measures to avoid finding oneself in that situation.   

The risk is that the argument is used to protect a sector, for bad reasons. The US administration’s objections to the purchase of US steel by Nippon Steel is a case in point. Japan is not an enemy of the US.  And the Pentagon buys less than 3% of the overall US steel production.

Economic security

Although the proportion of trade in intermediate goods has remained roughly constant over time, around 60% of total trade, global supply chains have become more complex, and thus potentially more exposed to failures along the chain. Floods in Thailand, the Fukushima disaster, pandemic disruptions, the Houthi attacks in the Red Sea, the risk of closing of the strait of Hormuz, all made the point. This suggests an increased need for on-shoring, or at least friend-shoring some activities.

The role of policy is less clear in this case, as one would assume that firms are best placed to assess the risks and reorganize their supply chains. But network issues, not unlike those that played during the global financial crisis, may be relevant.  Firms may know about their suppliers but not about their suppliers’ suppliers, and so on. Or they may have a plan B which happens to be the same as that of some other firms and find out that the alternative supplier they counted on cannot supply all. Thus, policy can play a useful role in examining the resilience of the system as a whole. 

Industrial policy  

For a long time, industrial policy had a bad name among economists. It was seen as more likely to help lobbies than to make the right technological choices. Most economists have now a more open mind - for at least two reasons. The earlier position was too extreme, and there was in fact often an economic case: China’s strategy of strong subsidies for the solar panel and the electric vehicle (EV) industries has shown that moving faster than other countries can pay off. The Advanced Research Projects Agency Network (ARPANET), the forerunner of the internet, is widely considered a great success, with a large economic impact. Also, the new technologies, in particular artificial intelligence (AI), seem more likely to come with substantial externalities, and strong increasing returns, two factors that justify state intervention.

The issue is that the earlier worries are relevant, and industrial policy may well be used for other purposes, namely, primarily to improve the competitiveness of domestic producers. This is clearly a relevant motivation behind many of the subsidies in the Inflation Reduction Act, from final assembly requirements for EVs or for batteries.

Measures to fight global warming

The need for governments to take measures to fight global warming is widely accepted. It is also agreed that the main tool should be carbon taxes or carbon fees on the use of brown energy. But, because such taxes are unpopular, an attractive alternative, if one ignores the budgetary implications, is subsidies for the use of green energy. Even leaving aside the budgetary implications, subsidies are not quite equivalent to taxes, but they can clearly give the right incentives to increase the share of green energy in energy production.

The issue here is again implications for competitiveness. Europe has largely gone the way of carbon taxes and carbon fees.  The US instead has gone the way of subsidies. The result, at the prevailing exchange rate, is a clear loss of competitiveness for European firms, a problem which is not solved by the carbon border tax, were it to be put in place: if a US firm and a EU firm both achieved zero emissions, the first as a result of taxes, the second as a result of subsidies, no border tax would be collected, but the EU firm would lose competitiveness vis a vis the US one. Trade tensions are already appearing, and tariff wars may well follow. 

Challenges ahead

Each of these forces has its own implications, but all are likely to decrease trade. (One could add to the list two forces I have intentionally left out: The use of tariffs as a substantial source of revenues, which is simply a bad idea. And the old mercantilist view that countries should have positive current account surplus, a bad idea that never dies). Some may come into conflict with each other. Should governments in their fight against global warming buy cheap Chinese solar panels, or should they put tariffs to protect their domestic producers and pay more (I think, in this case, the fight has been lost, and the best is indeed to use the Chinese panels). If the US puts higher tariffs on Chinese EVs, leading China to try to increase sales elsewhere, especially to the EU, should the EU also put higher tariffs as well, at the risk of Chinese retaliation against EU companies in China? Should higher tariffs on Chinese cars come with restrictions on foreign direct investment (FDI) by Chinese car companies in the EU, or instead encourage such tariff triggered FDI?

The correct economic answer cannot be an absolutist defence of free trade. Behind each of the five forces discussed above are good and bad reasons for it to happen. It may sometimes make sense to protect some sectors at the expense of consumers. It surely makes sense to worry about national and economic security. Industrial policy cannot be rejected out of hand and, want it or not, has implications for competitiveness. If carbon taxes are a political nonstarter, but the fight against global warming is existential, it may make sense to use subsidies. But, at the same time, in each case, there is a scope for misbehaving, for using the argument as a pretext for old fashioned protection.  

Defining the right policies is thus a case-by-case job, working out both the cooperative and non-cooperative outcomes. In other times, the World Trade Organisation would have been the natural instance to assess and adjudicate what was fair trade or not.  It is too weak and not in a position to do so today. The risks of retaliation, escalation, and trade wars are high, and with them an inefficient and costly deglobalisation. The effects have been limited so far, but I am afraid that the process has just started.