Rising trade tensions weigh on Fed outlook
- Trade tensions increased in May, when the US raised tariffs to 25% on $200bn of Chinese goods.
- We expect negotiations to resume between the two countries after a meeting between their Presidents at the end of this month, despite recent hawkish rhetoric.
- Yet the US is likely to continue to provoke further trade tensions over the coming 18 months with China, Europe and beyond, as part of the President’s political strategy.
- Restrictions on trade with Chinese tech look likely, regardless of developments over headline tariffs.
- Recent trade developments lead us to lower our US GDP growth outlook to 2.4% for 2019 (from 2.5%) and 1.6% (from 1.8%) for 2020.
- We now expect the US Federal Reserve to cut rates in September and December, to 1.75-2.00%.
- This outlook is trade development dependent and we discuss scenarios around this central view.
- In China, we expect developments to result in further easing. China GDP growth looks set to remain above 6% for 2019 and 2020. However, we expect the renminbi (RMB) to rise above 7 to the US dollar.
Trade policy back to the fore
For the first five months of 2019 markets allowed concerns about US protectionism and trade wars to fade. Progress appeared to be made towards an accommodation of US and Chinese interests in trade negotiations and the US had not made any announcement on the Section 232 report that could pave the way to auto tariffs.
Yet trade concerns re-emerged at the start of May. President Donald Trump announced that Chinese leaders were trying to renegotiate previously agreed aspects. The US increased tariffs on $200bn of Chinese imports to 25% from 10% – a rise we had been pleasantly surprised had not emerged as first scheduled on 1 March (Exhibit 1). The US added Chinese tech company Huawei to its “entity list” – which restricts trade it deems could compromise national security – and threatened similar for other large Chinese tech companies.
Meanwhile China, rejecting accusations of reneging on previous agreements, argued that US demands continue to grow. It retaliated by imposing tariffs on $60bn of US imports from 1 June, created its own “unreliables” corporate blacklist, began investigations into US company FedEx and implied it could disrupt rare earths exports.
US and Chinese teams have not met for further negotiations in recent weeks, and doubts have risen around a meeting between Chinese President Xi Jinping and President Trump on the fringes of the upcoming G20 meeting on 28-29 June.
Meanwhile, the US increased trade pressure further by threatening a rising scale of tariffs on Mexico1 until its “illegal immigration problem is remedied” before a subsequent agreement suspended the plans. Separately it also ended India’s “preferential trade status” – a scheme allowing around 20% of Indian goods to enter the US duty free – because of disputes over US access to Indian markets.
In the following note we consider these developments and their implications for global trade tensions and the likely economic impact and possible financial market reactions.
The China Syndrome
Renewed tensions between the US and China have revived geo-political concerns about a rising global power and the existing incumbent. There are worries of ‘Thucydides Trap’, a theory that power displacement often leads to war – in this case, potentially a proxy trade war. While these tensions doubtless underpin current disputes, more prosaic factors, including protection of intellectual property protection, the role of state subsidy, protection of tech supremacy and narrow political advantage are also all likely to play a role.
Upcoming trade negotiations are an illustration of game theory. We consider the relative costs to both sides to try to determine whether a co-operative or non-cooperative outcome is most likely, with Exhibit 2 illustrating where we see relative negotiating advantage.
President Trump has often claimed a trade war with China would be easy to win. He points to receiving 4.5 times more exports from China than the US ships to China ($540bn in 2018) as evidence of this. In pure trade terms, the US appears to have an advantage. Exhibit 3 illustrates the breadth and depth of relative tariff coverage (tariffs x coverage) by the US, and retaliation by China. It shows that in early 2018 China matched US tariff hikes, but that it has subsequently implemented them at a slower pace. This in turn likely reflects the greater proportion of US imports affected by retaliation, which should have larger negative impacts on China’s own economy.
Moreover, while China was able to let its currency depreciate in response to previous tariff increases (as theory suggests2), dropping by around 10% as tariffs emerged in 2018, it is unlikely to want to see a similar scale of depreciation without a substantial escalation of trade tensions. Chinese authorities had been concerned that depreciation above seven yuan to the dollar could lead to increased capital outflows from the Chinese economy, threatening financial stability. However, more recently the Peoples Bank of China (PBoC) Governor Yi Gang suggested that Beijing may be more flexible about this level, as defending it could incur more costs than gains. We therefore believe that the renminbi (RMB) will rise above 7 (to the US dollar) at points this year.
The economic impact of escalating trade tensions is also likely to be larger on China than the US. In 2018, we published two research notes34 estimating the impact of escalating trade tensions. We estimated that tariffs rising to 25% on $200bn of China imports would reduce US GDP growth by around 0.25 percentage points (ppt) over two years, but would reduce Chinese GDP growth by 0.7ppt. In case of an escalation to full coverage of China imports this would rise to a 0.5ppt impact on the US, but 1.5ppt on China. These estimates were conducted on a comparative static basis, attempting to assess the economic impact of tariff increases alone, and not allowing for any subsequent offsetting policy actions. Nevertheless, the estimates suggest a more material impact on China.
Both channels suggest that the impact of tariffs would be felt more by the Chinese economy and people. However, we think a key aspect of these negotiations is political. Regardless of the economic implications of the trade war, China’s President Xi will remain in power over the coming years. In contrast, President Trump faces re-election next year. We believe that politics is a key driver of US trade policy, with the White House’s current policies reflecting the core of President Trump’s support and keen to avoid Democrat criticism of a ‘soft’ deal with China. This is why we have long argued that trade war rhetoric was always likely to flare up again, although considered it more likely in the actual election year, 2020. However, while President Trump’s base may applaud this trade-hawkish outlook, it will be difficult to win re-election against a backdrop of a weakening economy – and close to impossible if the US suffers recession.
This political consideration should make the US wary of escalating the trade war further. The White House might ‘ideally’ keep trade rhetoric “on the boil” for political effect, but not so much as to cause lasting damage to the US economy. Yet this is a difficult balance to strike. The US might try and achieve this by leaving tariffs at current levels – deferring further escalation and resuming negotiations. It might even consider reducing tariffs back to 10% as negotiations resume, but subject to implementation/progress reviews at agreed future dates. Both outcomes would be more benign than trade escalation, and would provide a ready vehicle to renew trade tensions. Yet both would fall short of returning to the status quo ex ante. This is because uncertainty about future protectionism has risen, particularly considering Chinese tech trade and Mexican tariffs. We argue that this increased uncertainty will have a lingering impact on business investment decisions, globally and in the US, which is likely to act as an additional headwind to growth.
The political aspects of US trade policy are also in part grounded in a poor general understanding of the mechanics of tariffs. President Trump tweeted on 13 May “We will be taking in tens of billions of dollars in tariffs from China”. A detailed analysis of the impact of the tariffs across 20185, however, concluded that the 2018 “tariff changes have had little to no impact on the prices received by foreign exporters”, having been “almost entirely passed through into domestic prices”. The tariffs have thus reflected a combination of transfers from the US consumer to government, with some additional deadweight welfare loss – diverted purchases to items that would have been disregarded before the tariffs, or foregone purchases. Unless the US government spends this additional revenue in an equally social welfare benefit-enhancing manner, the “losses to taxpayers could rise by as much as the full value of their tariff payments”. Admittedly this more thorough examination of tariffs is not immediately intuitive, but as a broader distribution of this understanding takes place, even those currently supporting tariffs may become less keen.
Is China tech a special case?
China trade tensions have evolved on several levels. While we argue that a full-blown tariff war is one that might have economic and political consequences that quickly become unpalatable to the US administration, we are also clear that the US will maintain some level of tension – and this looks set to be focused on China’s tech industry.
Beyond tariffs, current trade tensions have quickly focused around tech companies. The US added Huawei to its “entities” list threatening to severely restrict Huawei’s access to international tech imports, as well as limiting its export options. Several other Chinese tech firms have been threatened with a similar listing, including iFlytek, Hangzhou Hikvision Digital Tech and Zheijang Dahua Tech. China has responded by creating its own “unreliables” list. However, President Trump also suggested on 23 May that it was “possible Huawei even would be included in some kind of a trade deal”. Huawei may yet prove to be this year’s version of ZTE, which last year saw a seven-year export ban introduced after it breached sanctions with Iran and North Korea. ZTE announced it was ceasing “major operating activities” in May 2018, before the US lifted the ban the next month, albeit following a $1.2bn fine, management overhaul and outside compliance team oversight.
While the current pressure on Huawei and others may simply be additional levers to broker a wider deal, we would still expect some tech trade tensions to persist even in a more benign trade scenario. We see four key motives:
- Addressing concerns about forced tech transfer and intellectual property protection
- Adversely impacting China’s ‘catch-up’ productivity growth. In turn, slower productivity growth should adversely affect China’s trade surplus (although not necessarily the US-China bilateral surplus)
- Sustaining US tech dominance
- Echoing Soviet-era policy, which included tech transfer restrictions
This attempt to isolate China from international tech flows could simply enhance China’s drive to develop its own technological prowess. Indeed, the Soviet Union broadly kept pace with US military technological developments during the Cold War; separately South Africa developed synthetic fuels during its apartheid sanctions. However, both economies suffered a clear cost from this approach. Moreover, Exhibit 4 illustrates that despite China’s significant R&D spending it still lags the US at the technological frontier.
As such, full technology restrictions should significantly slow Chinese technological development over the coming years. Slower productivity growth should reduce export competitiveness and add to the deterioration in its current account position – a trend underway due to internal structural development. It would also reduce China’s trend growth rate. Exhibit 5 illustrates that with Chinese labour supply contracting over the coming decade, headline growth will be increasingly dependent on productivity growth. Slower trend growth would also impinge on other areas of China’s development and impact its longer-term debt levels.
Beyond the strategic connotations of technology restrictions, the US position may also have a mercantilist angle. Although these policies could result in the establishment of an independent Chinese technology eco-structure, in the short term it is likely to prolong US tech dominance. Even in traditional markets the persistence of dominance allows for continued extraction of economic rents. However, tech industries often include network externalities – ‘winner-takes-most’ returns that provide significant rewards to the remaining dominant player and increases the chances of prolonging that dominance.
That said, restrictions on China tech trade would also have material implications for the US. China is a major customer of US tech exports and restricting US sales to one of its largest customers would have a significant impact on US company profits. The scale of this impact is illustrated by Russell Vought’s (Acting Director of the Office of Management and Budget) letter seeking a delay to the impact of the ban on Federal government agencies contracting with Huawei, or companies that do business with Huawei. He suggested an additional year to provide US companies that currently deal with the Federal government time to make alternative arrangements. China is also a major part of the supply chain of US companies, the most obvious being Apple. Significant restrictions would pose difficulties necessitating a costly shift in supply chains, most likely to alternate third parties.
There is also likely a knock-on effect across broader emerging Asia, beyond China. Over the medium-term tech trade restrictions would create some winners and losers as some countries gain from tech manufacturing sourced from outside China, while other countries would suffer reflecting the intricate supply chain network of China’s own tech industry. While the medium-term effect is likely to be mixed, in the short term the realignment of current supply chains represents pure cost to the global tech industry. Moreover, the uncertainty of trade policy is already likely to be having a dampening effect on investment in these industries.
Amid the broader China question, on 2 June President Trump announced intentions to raise tariffs on Mexico from 10 June. However, on 7 June, the US and Mexico announced an agreement to permanently defer such tariff increases reflecting a number of Mexican initiatives to address migration flows. While the outcome of this week’s worth of uncertainty was to leave trade policy with Mexico ostensibly unchanged, the implications of this series of announcements may have a more lasting impact.
The economic impact of these tariffs would likely have been extremely damaging. For the US, Mexican imports constituted 12% of total imports in 2018 (China 18%). A comparative static analysis on the impact of such tariffs suggests an economic impact of around 0.4ppt if tariffs rose to the full 25%. However, this may underestimate the negative impact on the US economy given the larger share of US exports sent to Mexico (16% of total US exports, more than double the amount sent to China) and the deeply integrated supply chains between Mexico and US state economies. Certainly, the economic impact on the Mexican economy would also have been large, with the US constituting 80% of Mexican exports.
This was also the first time the US threatened tariffs for ostensibly non-trade objectives, in an attempt to influence Mexican migration policy. Migration is a big political issue in the US and for the Trump administration particularly. Indeed, these tariffs were the latest in a series of policies focused in this area, including building “The Wall” and threats to “close the border” earlier in the year. Echoing the President’s claim that the US is receiving “billions” from China it is even possible that President Trump might have claimed that these tariffs would fund “The Wall”. Moreover, we were unnerved by the coincidence of this sudden populist action on the border at a time when Robert Mueller, US Special Counsel, appeared to be clarifying his expectation that Congress progress his evidence in impeachment hearings. It also appeared unusual as it occurred on the same day that the USMCA6 trade deal (to replace NAFTA7) was put to national congresses. This unusual timing raises questions about whether the Mexican tariffs were announced with a motive of distraction from developments in Washington, rather than with US economic concerns in mind.
This combination of erratic policy-making, with difficult to discern motives and potentially large economic ramifications, likely comes at a broader economic cost. Even as Mexican tariffs have been deferred, investors are likely to remain concerned about what the next move in trade barriers will be, with an increase in US auto tariffs already on the cards. This uncertainty over future trading arrangements is not helpful for long-term business planning or spending – something that has been illustrated in the UK’s trade barrier uncertainties reflecting Brexit, which has resulted in business investment stagnation at an estimated cost of around 2¼% of GDP in the 2½ years to end-20188.
A US recession tipping point
On our assessment, a full escalation of trade war with China could reduce US GDP growth by 0.5ppt and threatened Mexican tariffs by 0.4ppt. There are additional economic costs associated with an uncertain level of restriction on trade with China (and potential trade partners) and an adverse impact on business spending because of a permanent elevation of policy uncertainty. Additively, one can see the combined impact that this could have had on US GDP growth of around 1ppt – and we consider even excluding Mexican tariffs an impact of around 0.6ppt.
However, economic shocks, and shocks to other complex systems, are often not captured by simple additive – linear – processes. Rather, non-linearities can lead to modest shocks having marginal effects, but bigger shocks having disproportionately larger impacts. We fear this would be a case in point with current trade policies.
Exhibit 6 attempts to illustrate this quantitatively. Our recession probability model9 has suggested that the likelihood of a recession over the coming 12 months has increased in recent quarters, primarily reflecting the flattening and inversion of the US yield curve. We believe the materialisation of tariffs as currently threatened would further invert the curve (that is, lower 10-year UST yields) and widen credit spreads. This would likely raise estimated recession probabilities above those preceding previous recessions over the past fifty years.
Intuitively, the global economy is presently vulnerable, having suffered a material deceleration in global trade growth in 2018, which we expect to have reached a nadir in H1 2019. The US economy is also in a vulnerable deceleration phase, reflecting the fading fiscal stimulus compared to last year. Negative shocks – which would include a material increase in trade barriers and disruption to global supply chains – at points of economic vulnerability tend to result in recession.
Yet we do not expect the current tariff threats to be fully implemented. Mexico is a case in point, but we also doubt the likelihood of a full escalation of US-China trade tensions. Rather our central expectation is for a meeting at the G20 to announce a resumption of negotiations, a deferral of the broadening of tariffs on the additional $300bn of China exports and perhaps even some hopes of a reduction in the current tariffs. Such progress towards a “deal” would in our minds be more of an “extended truce” than “lasting peace”. We fully expect the US to maintain tariffs at some level, possibly requiring verification of implementation of Chinese agreements – as discussed a potential future source of trade tensions. These ongoing tariffs and any ‘verification’ are likely to be an ongoing source of trade tensions.
Yet we consider some lasting damage from the fluctuations in recent trade policy – which we expect to persist in the run-up to the US Presidential elections. Such erratic policy-making has increased the uncertainty surrounding future policy developments, with very real risks that China tensions could rise again, auto tariffs could easily be announced or further unforeseen trade tensions arise. Such uncertainty has an insidious effect on business investment and is likely to remain a permanent drag on US activity. This is something we believe that the US Federal Reserve (Fed) will be forced to react to with the likelihood of some policy easing.
Fed to the rescue
Exhibit 7 summaries the scenarios we consider over the coming months and our estimated impact on US growth. We expect each scenario10 to reduce the outlook for US GDP growth, either insidiously, as uncertainty weighs on business sentiment, or more directly through a supply shock. In each case we envisage a case for the Fed to ease policy to offset the estimated growth impact. Admittedly, supply shocks would be expected to affect both output and prices, presenting something of a trade-off for the Fed. However, against a backdrop of subdued inflation pressures, the Fed is likely to be more focused on short-term growth implications.
As illustrated, we see the scale of any Fed easing to be dependent on developments in trade policy. Using ‘rules of thumb’ from the Fed’s macro model, we assess that one rate cut would add around 20 basis points (bps) to US growth over two years. Assuming the Fed was not concerned about the inflationary impact of tariffs, we would argue that one or two 0.25% cuts from the Fed would address the expected growth shortfall associated with current tariffs or escalation, respectively. However, we assume that additional concerns of elevated uncertainty and persistent downside risks would see the Fed loosen by one cut more than this baseline.
As discussed, there is additional risk that the escalation scenario is consistent with a further deterioration in market outlook (i.e. the market does not fully price a bad trade outcome). In this case, escalation of trade tensions could result in pushing the US into recession. In the last three US recessions, the Fed has eased policy rates by 500bps. If trade tensions push the US into recession the Fed would likely ease policy sharply this year, but continue to ease policy across the course of 2020, likely back to the zero lower bound.
US monetary policy outlook appears materially dependent on the medium-term outlook for US trade policy. As such, it was little surprise that the Fed provided no suggestion of an immediate cut ahead of its June meeting. Rather, Fed speakers referred to rising uncertainty associated with trade policy and Fed Chair Jerome Powell stated that the Fed was monitoring trade developments closely and would “act as appropriate to sustain the expansion”. We expect the Fed to react to trade developments, which could include an easing as soon as its next meeting on 31 July, but we think more likely in September.
The Fed also needs to manage fragile market sentiment. While we expect sentiment to improve if trade negotiations are resumed between the US and China, the current level of financial conditions assumes three cuts by year-end. The Fed will thus have to judge the scale of easing it provides based on the assumption that failure to fulfil market expectations will result in a further tightening in financial conditions, which will additionally weigh on the economy.
Taking all of this together we revise our outlook for the Federal Reserve to now expect two rate cuts this year (September and December), which would take the federal funds rate (FFR) to 1.75-2.00% by year-end. We do not expect further easing at this stage in 2020. We also acknowledge that trade uncertainty is currently elevated and we expect to see significant developments across the course of this year – this may result in additional fine-tuning of our view over the coming months.
Considering financial assets more broadly, the current roster of trade concerns readily explains the apparent fragility of financial markets, with US 10-year Treasury yields currently trading just over 2½-year lows. Other safe-haven instruments including gold, the Japanese yen and Swiss franc have also been boosted recently. We tend to the view that markets continue to have a gloomy outlook for trade and on balance, our trade outlook remains somewhat more upbeat – at least insofar as a scenario that avoids the worst-case further escalation. As such, financial market reaction should be poised for a positive reaction in our main case scenario. However, we also note that with policy-by-tweet being the key channel of information distribution, there is an ongoing risk of continued volatility, particularly as we get nearer to the likely all-important G20 meeting on 28-29 June.
- Mexican tariffs are suggested to begin at 5% on 10 June and then to rise by 5% at the start of each month to 25% by October. 12% of US imports came from Mexico in 2018.
- Mundell-Fleming framework suggests international economies will see exchanges rates rise (fall) as tariffs are increased by them (on them), so long as there is no accompanying monetary accommodation.
- Yao, A. and Shen, S., “China: Quantifying the impact of trade war”, AXA IM Research, 16 August 2018.
- Page, D., “Trade wars: an old hope”, AXA IM Research, 9 August 2018.
- Amiti, M., Redding, S.J., Weinstein, D., “The impact of the 2018 trade war on U.S. Prices and Welfare“, CEPR, March 2019.
- United States-Mexico-Canada Agreement
- North American Free Trade Agreement
- Source: AXA IM Research.
- Page, D., “Is the yield curve pointing to recession?”, AXA IM Research, October 2018.
- Technically we do consider an upside “deal” scenario that removes all tariffs and resolves outstanding issues, but we consider this a tail risk and hence do not include it in the table.