The games Donald plays

 

By now, anybody who is anybody – and plenty who aren’t – has had their say on the Trump tariffs and their possible implications. Much ink has been spilled documenting the twists and turns of the Trump Administration’s actions and the subsequent market movements. If there is one lesson to be learned from recent weeks, it is not to overreact to latest events because likely as not, there will be a rollback which renders the previous position invalid. That is not to say markets are wrong to sell off/rally as Trump’s whims dictate – after all, investors have to take a position based on available information because each signal from the White House alters the landscape of risks, costs and growth expectations. But it is no way to run an economy.

More than a game

One way to think of Trump’s strategy is through the prism of game theory, and there are some elements which echo the main findings of Thomas Schelling’s classic 1960 book, The Strategy of Conflict – a seminal work in game theory and strategic decision-making, particularly in the context of conflict, negotiation, and deterrence. One of the key takeaways of Schelling’s work is that conflicts are not merely power plays that pit one side against another, but are instead contests that highlight strategic interdependence – in other words, the actions of one player have an effect on the strategy of the opponent.

Schelling also highlighted the role of brinkmanship in which pushing one party further outside their comfort zone forces them to change their own strategy. This is very clearly the case in the context of the US-China trade row which threatens to inflict significant economic damage unless cooler heads soon prevail. China responded to the 145% levy on exports to the US with a tariff of 125% on US imports, highlighting the impact of brinkmanship. But the strategic interdependence element also came into play when the US exempted Chinese-made smartphones from higher import levies – the number one Chinese export to America by value – when the inflationary impact of smartphone tariffs became clear. It was a small de-escalation in a much bigger dispute but it highlights that there are some people in the US Administration who appear to understand the risks of escalation.

Another game theory concept highlights what can happen if actors in the dispute focus purely on the pursuit of their goals – Martin Shubik’s dollar auction problem which I originally used as a demonstration of the irrationality of Brexit negotiations. In brief, the dollar auction problem is a two-person thought experiment in which a dollar is initially auctioned off at a substantial discount to its face value. In subsequent rounds the price is bid up, with the highest bidder winning. But the twist is that the loser must also pay their bid. Consequently, both bidders have an incentive to continue bidding beyond the value of the prize in order to minimise losses, which results in a lose-lose outcome.

In the trade context, neither China nor the US wants to look weak in the eyes of the world leading to a cycle of escalating tariffs, even when both countries incur economic losses. The rational solution to this problem is to back down early or find a co-operative solution which minimises losses for both sides. Unless this approach is adopted, the key lesson highlighted by the dollar auction paradox is that once the bidding process becomes entrenched, the competing parties lose sight of their original goals and produces an outcome where the costs of winning outweigh the cost of defeat.

Credibility and how to lose it

One aspect of Schelling’s work that has not been adhered to is the principle of credibility. Schelling made the point that in strategic interactions, the ability to credibly commit to a course of action conveys a decisive advantage and that threats and promises must be credible to be effective. In the case of trade, however, Trump has not acted credibly. He has backtracked on his tariff policies on a number of occasions, leading many to wonder whether he is bluffing. While Trump sets great store by the art of the deal, the art of the bluff can take you a long way in life, love and poker. But bluffing only works if opponents believe the bluffer will follow through. While Trump has the power to impose tariffs, the question is whether doing so is politically or economically wise. According to a recent poll, 56% of respondents disapprove of his handling of the economy while 75% believe tariffs will push prices higher in the short-term. If Trump were attempting a further run for the White House, it would be politically dangerous to go further with the tariff policy. But because he is not bound by this constraint, further trade escalation cannot be ruled out, especially since 34% of poll respondents believe Trump's economic policies will pay off in the long run (admittedly down from 41% at end-March).

In game theoretic terms, a key weakness of Trump’s approach is that he takes a zero-sum approach to strategy – in other words, winner takes all and the loser gets nothing. This might work in the casino business – although Trump’s record as a casino owner is terrible, having twice filed for Chapter 11 bankruptcy – but it does not work in politics. One reason for this is that politics is not a one-shot game in which strategy is formulated without any concern for future consequences (as the dollar auction problem illustrates). Geopolitical decision making is a multi-shot game in which today’s actions have an impact on future decisions. For example, the US may ultimately need Europe onside if it is to successfully pursue its economic policy with regard to China. But by damaging relations with Europe today through its approach to both trade and security, it becomes less likely that Europe will side with the US in future.

We cannot continue in our game theory mode without talking about a Nash equilibrium – a situation where no player can improve their outcome by unilaterally changing their strategy, assuming the other players keep their strategies unchanged. Since high US tariffs hurt domestic consumers, the US might improve its welfare by lowering tariffs — even without immediate Chinese reciprocity. This suggests that the current position does not represent a stable Nash equilibrium. But since US tariffs also make China worse off, China has an incentive to reciprocate by reducing its own tariffs. Rolling this forwards, both sides engage in tariff reduction until the point at which we reach a stable welfare position. Such a strategy works because it builds trust gradually and is credible, even in the absence of explicit cooperation.

Trust busting

Looked at from a logical standpoint, Trump’s trade policy appears inconsistent because (a) it will make US consumers worse off and (b) he has not followed through on his calls, undermining their credibility. Against that, his policy could be seen as a form of calculated ambiguity or strategic brinkmanship, with Trump attempting to pressure trading partners into concessions and rallying domestic support by appearing strong on trade. By merely postponing tariff implantation, Trump keeps everyone off balance, thus enhancing his leverage and keeping his threats alive without fully committing.

Perhaps the best definition of the current situation is one of strategic ambiguity, based on the goal of retaining flexibility in order to exert pressure and manipulate expectations. Such a strategy might work if the short-term volatility serves a longer-term purpose, but other than putting maximum pressure on China, it is hard to see what the longer-term plan is. At some point Trump has to commit on tariffs one way or the other but when he does, he will have to accept the consequences. America’s putative allies may be amenable to Trump’s threats but China most certainly is not.

Last word: Common sense is not so common

All of this may sound like obvious common sense – don’t escalate conflicts that hurt both sides, don’t bluff unless you’re willing to follow through, and don’t ignore the long-term consequences of short-term moves. Yet time and again, these lessons are ignored in the heat of political battles. This is where game theory proves its value: not as an abstract academic exercise, but as a practical framework for understanding strategic behaviour and anticipating outcomes. By highlighting the incentives, interdependencies, and likely responses of all players, the theory of games offers a rational lens through which to view even the most irrational-seeming policies.

Trump, tariffs and beyond

As the world now knows, Donald Trump followed through on last weekend’s tweet promising to raise tariff rates on USD200 bn of Chinese imports from 10% to 25%. The fact that some of the heat appeared to go out of the tariff wars towards the end of last year suggested that both sides realised there was nothing to be gained from continually ramping up the rhetoric. After all, nobody ever won a tariff war. But once again Trump has upended conventional wisdom and those who continue to underestimate him should now be fully aware that he means to push forward with his ‘America First’ agenda, irrespective of how damaging it might appear at first sight.

We need to view Trump’s actions in a domestic context. Simply put, they can be seen as the opening shot in his 2020 re-election campaign and his strong-arm tactics are broadly popular at home. The US also has a point about technological expropriation and the requirement for the Chinese to open up their domestic markets in a reciprocal manner, in line with their WTO commitments. But we also have to view this from the Chinese perspective which sees itself as reasserting its rightful place on the world stage after two centuries of political and economic humiliation by the west. Finding a resolution that accommodates both sides will not be easy.

Market reaction to Trump’s actions has been somewhat muted. Admittedly, US equities fell 2.2% during the course of last week but that is not a huge decline and only puts the S&P500 back where it was a month ago. One interpretation is that markets are clinging to the belief that a resolution to the tariff war will somehow materialise within the next couple of weeks. After all, only goods leaving China after the Thursday midnight deadline will be subject to the new tariffs – those currently in transit will not – so if a deal can be brokered within the two weeks it takes for goods to make the journey by sea, the impact of the latest tariff spat will be limited.

That may be a very complacent view. Indeed, recent events might just prove to be another shot in the long war against globalisation from which there are no economic winners. Consider first the tariffs themselves. They are in effect a tax on imports and the one thing we do know about product taxes is that they are borne by the end-consumer. Companies that import certain items from China will now have to pay 25% more for them (although the impact so far has been partially offset by dollar appreciation). Households consuming those Chinese goods on the list will face a similar problem. But as this paper by Pablo Fajgelbaum and his co-authors point out, the direct impacts of last year’s tariff hikes cost the US economy just USD69 billion (0.3% of GDP). And once we account for the substitution away from Chinese imports towards domestic alternatives and the gains from higher prices received by US producers, the total impact is a mere USD7.8bn, or 0.04% of GDP.

But this is to underestimate the longer-term damage that an escalation of tariff wars could inflict on the global economy. Unfortunately, the recent actions might encourage the Trump administration to believe that tariff wars are indeed “good and easy to win” as the President said in March 2018 which (a) reduces the chances that the US will offer any concessions to the Chinese in the current dispute and (b) encourage the US to target European exporters, where German auto manufacturers are widely concerned that they will be the next in the line of fire.

With regard to point (a), we do not know how the Chinese will respond. Given that the US imports more from China than it exports, China’s direct ability to engage in a tit-for-tat tariff escalation is limited. It could, of course, levy additional duties on US agricultural products. But more damagingly it could target the US tech sector. China is less dependent than the rest of the world on Amazon, Apple and Google given the local dominance of Tencent, Huawei, Baidu and Alibaba. The Chinese companies start from the advantage of a bigger domestic market and are already formidable competitors in third markets. In a phrase reminiscent of the thinking during the Brexit referendum campaign, the likes of Apple need China more than the Chinese need them.

Also we should not overlook the fact China is the biggest buyer of US Treasury debt. Whilst it is unlikely to sell its current holdings, a buyers strike may push up US interest rates and thus have the opposite effect to what Trump wants (he has, after all, called for the Fed to lower interest rates). However, it is widely believed that the Chinese have no incentive to exacerbate the trade dispute in the short-term – particularly since the Communist Party wants to sell a rosy view ahead of the 70^th anniversary of the People’s Republic in October and the party’s centenary in 2021. But if there is no quick fix or if they are forced to make too many concessions, the Chinese will not easily forgive or forget.

With regard to point (b) there is a relatively easy fix. The EU could adopt a policy of pre-emptive tariff equalisation by reducing the tariffs on auto imports from the US from the current 10% to the rate of 2.5% which the US levies on EU imports. But the wider concern is that the US will become a less reliable ally than has been the case over the past 80 years. Europe and the US have common international interests and cooperation has led to better outcomes for both sides. This would be put at risk in the event of policy divergence. For example, one of the biggest issues rising up the policy agenda are environmental concerns which require international cooperation – no single economy can fix things on its own.

The current environment is increasingly one of mutual suspicion which does not bode well for finding global solutions to global problems. There is also a risk that local issues could become flashpoints for bigger problems, as we experienced during the Cold War. For example, China regards the region bordering the South China Sea as its own sphere of influence and is increasingly less tolerant of US interference in the region. The US does not see it the same way. But having dominated the geopolitical arena since 1945, the US may be forced to cede some control and the manner in which it does so will have a great bearing on the history of the 21^st century. The tariff wars could turn out to be a Gavrilo Princip moment. Or they may simply be a Cuban missile experience. Either way, there is a lot more at stake than import taxes.

China crisis

The announcement that the US government plans to impose tariffs of 25% on $50 bn of goods imported from China has set the cat amongst the pigeons, with equity markets turning sharply lower and safe havens such as gold gaining ground. But whilst this may look like a simple application of economic nationalism, led by a president who clearly has no appreciation of the damage that he may be about to unleash, it is worth considering the underlying US grievances. First and foremost, the Administration believes that China has gained from the unfair expropriation of US intellectual property and that some form of recompense is required. But this is not only a US problem: Many western governments are beginning to worry about the asymmetries which their companies face in doing business in China – the US is simply the first to take action.

Three weeks ago The Economist, which has long been a cheerleader for global free trade, expressed reservations about the direction in which China is heading. In its view, “the China of Xi Jinping is a great mercantilist dragon under strict Communist Party control, using the power of its vast markets to cow and co-opt capitalist rivals, to bend and break the rules-based order.” The article went on to point out that “Chinese markets are opened only after they have ceased to matter” whilst regulators “take away computers filled with priceless intellectual property and global client lists.” A week later, the Financial Times ran a story headed “Backlash grows over Chinese deals for Germany’s corporate jewels” following the news that Geely has acquired a 10% stake in Daimler which has raised fears that Daimler’s know-how in the field of electronic vehicles will filter back to China without appropriate compensation.

Industrial espionage is not new, of course, and we should remember that some of the earliest examples of such activity involved the transfer of Chinese technological advantages into western hands. For example, in the 1800s China had a monopoly on tea growing until a British botanist, acting on behalf of the East India Company, smuggled tea plants and seeds to India and established an industry whose output eventually eclipsed that of China. But this does not assuage current western concerns that the heavy-handed techniques employed by the Chinese are backed by the government. Whilst western companies have long been required to hand over technological secrets before being allowed to conduct business in the Chinese market, the fear is now that many of China’s major foreign acquisitions have been funded by state-backed institutions. Indeed, the FT reports that the message conveyed by the Geely chairman to the German media, that the company wanted to cooperate with Daimler, was not the one he gave to his home audience which was that the action was designed to “support the growth of the Chinese auto industry” and “serve our national interests.”

Seen in these terms, it is hardly a surprise that the US feels that it needs to take some form of action. But despite Trump’s rhetoric to the contrary, it is hard to believe that the US really wants to embark on a major trade war. The fact that China responded with tariffs on only $3bn of imports from the US suggests that it is not willing to escalate the problem either. In the grand scheme of things, the US actions will have next to zero impact on Chinese GDP. A 25% tariff on $50bn of exports amounts to a total hit of $12.5bn which is insignificant in a Chinese economy whose output is valued at $12 trillion. But it is what comes next that matters.

In assessing the outcomes, I am indebted to some modelling analysis conducted by Bloomberg analysts using the NiGEM global macro model. In its first scenario, Bloomberg assumed that the $50 bn figure becomes a US revenue target to reflect the estimated damage done to the economy by intellectual property theft. But the impact of such an outcome, which is four times more significant than what we believe likely to happen, only costs 0.2% of Chinese GDP by 2020. Bloomberg thus concluded that China would be better off not retaliating because the economic losses resulting from inflation generated by higher import tariffs would exceed this amount. In a second simulation exercise, Bloomberg tried to assess the impact of a 45% tariff on all Chinese imports – a figure that Trump happened to mention on the campaign trail. This resulted in a 0.7% hit to GDP by 2020 which they concluded would “not be disastrous.”

But the real problem comes when the tariff war goes global and pulls in countries other than China. An across-the-board rise of 10% in US import tariffs which is met with a similar response by all the US’s trading partners results in a 0.5% drop in Chinese output but a 0.9% drop for the US. This highlights the self-defeating nature of tariff wars and results from the fact, as Bloomberg pointed out, “the tariffs affect 100% of US trade, but for China and other countries, [they] only impact bilateral trade with the US.”

On the surface, the optimal Chinese response to higher tariffs which in aggregate terms amount to little more than a gnat’s bite, would be to ignore them. Since it continues to grow faster than the US it will – in the not-too-distant future – overtake the US as the world’s largest economy and will be in a position to retaliate more effectively. In any case, there are other ways to respond. China is also the largest holder of US Treasury securities (chart). It could thus tweak the tail of the US by selling Treasuries and put upward pressure on US longer-term interest rates.

But this issue is about more than just economics. This is a tale of two alpha economies demonstrating their political muscle, which runs the risk of miscalculation. Unlike Japan in the 1980s, when the US tried to exert pressure using similar tactics, China is a much more potent economic and political rival. It is also not politically allied with the US in the way that Japan was. I have never subscribed to the idea that China and the US are doomed to fall into the Thucydides Trap. But this is a time for cool heads and as I noted in late-2016, it is at times like these that we will miss the rationality of an Obama.

Steeled for trouble

The announcement last week by Donald Trump that he intends to levy tariffs on US imports of steel and aluminium, of 25% and 10% respectively, was the first indication that the President intends to follow up on his campaign promises. The announcement came a matter of hours after I sat in a client meeting and said something to the effect that Trump had so far not implemented the worst of his campaign promises, thereby demonstrating my great prescience.

It came at a bad time for markets, which were beginning to recover from the wobble at the start of February and the S&P500 is currently around 5.7% below the high achieved in late January. What particularly spooked markets was Trump’s claim that “trade wars are good, and easy to win.” Nothing could be further from the truth, as the experience of the 1930s demonstrated. They are nasty and do not result in any winners – everyone loses. Obviously, the steel tariffs will matter because the US is the world’s largest steel importer (26.9 million tonnes in the first nine months of 2017). But who will pay the price? Initially, it will be US industry which uses the steel as an input but ultimately it will be consumers – primarily in the US but also those elsewhere which buy US products using imported steel as an input.

The initial kneejerk reaction was that this was a way of hitting back at China, which has been the focus of the President’s displeasure for some time. Admittedly, China was accused by the EU of dumping steel on the world market at artificially low prices. Only last April the EU introduced levies ranging from 18.1% to 35.9% on certain types of Chinese rolled-flat steel products for a five year period. But China is not even in the top 10 sources of US steel imports. Canada, Brazil, South Korea, Mexico and Russia (in order of importance) account for 57% of the total – and the irony is that two of these countries are NAFTA partners (see chart). With regard to aluminium imports, Canada alone accounts for 56% of the US total, followed by Russia (8%) and the UAE (7%), with China lagging behind in fourth with a mere 6% share.

Trump also turned his focus on the EU at a press conference yesterday, saying “The European Union has been particularly tough on the United States … They make it almost impossible for the United States to do business with them. And yet they send their cars and everything else …” Spot the EU exporter in the list! In fact, Germany is the only EU country which manages to get on the steel importers list, coming in ninth, accounting for 3% of the US total.

We are still waiting to hear which countries will be affected by the tariffs and it really does look like the President has lashed out without regard for the consequences of his actions (why should we be surprised?). There has been speculation in the media that Trump’s actions were nothing more than an angry reaction following the resignation of communications director Hope Hicks, and a series of other incidents.  If true, it certainly raises a concern about the state of mind of the man with his hand on the nuclear button.

The damage from the trade action is likely to be twofold. On the one hand, there will be some limited form of retaliation from US trade partners, and even though an all-out trade war is unlikely, it is still a very bad sign. Second, it may raise questions about the quality of people prepared to serve in the President’s Administration. Gary Cohn, Trump’s highly rated chief economic adviser, has already resigned in opposition to the plan and a number of other cabinet members are believed to be opposed, including Treasury Secretary Mnuchin and Secretary of State Tillerson.

Perhaps more importantly, it calls into question the rules-based system that underpins the global economic order which has served the western world so well for 70 years. If the US, which has acted as guarantor for so long, no longer appears inclined to play by the rules, why should the likes of China or India, which are set to become major economic powers in the course of the 21st century? It will certainly give China greater moral authority to write a set of trade rules to suit itself. And on this side of the Atlantic, at a time when the UK has decided that it no longer wants to be part of the EU, the customs union or single market, it should give those pushing for trade deals with the rest of the world pause for thought about who our friends are and where our interests lie.