Keeping the show on the road

It’s one of the longest bull runs ever, but that doesn’t mean it will be cancelled any time soon. 

1 October 2018

With the long post-financial crisis market rally reaching into its first decade, more and more clients are asking us the same question: are we nearly there yet? Or, rather, are there signs of recession on the horizon? 

Some of those questions may have been prompted by a flat US yield curve, which became a little flatter still over the quarter. More and more newspaper articles are referencing it. A negative or “inverted” yield curve (when short-term debt yields more than longer-term) is a reliably ominous portent: before each of the last nine US recessions, 1-year Treasury yields rose higher than 10-year yields, typically nine months ahead of the downturn. There has been only one false signal, way back in 1967! (Interestingly, other economies’ yield curves are not good predictors of recession.) Having a nine-month lead is  a boon to asset allocators. Equity markets tend to lead the economy by rather less time – three to six months before recession hits – giving us time to adjust before share prices react.

But herein lies the problem: following the earlier signal from the yield curve could cause you to miss out on rallying equity markets for rather a long time. Or more: the average lead time from curve inversion to recession may be nine months, but the range of lead times is rather large. 

For risk-aware investors, missing out on the final few months of returns is often prudent. But missing out on the final few years may be less forgivable. The curve remained “flat” around today’s levels for over two years leading up to the dot-com recession in the early 2000s. In other words, taking a flat yield curve and pre-empting inversion is unnecessary and the opportunity cost potentially large. 

And we can go a step further, applying a statistical tool to transform the signal from today’s yield curve into the probability that the US economy falls into recession in the next 12 months – currently it’s just a 17% chance. 

In economics, the yield curve is often thought of as a proxy for the difference between the cost of capital and the return on invested capital. This gap can be estimated more specifically using national accounts data, which also suggests that the business cycle is not yet nearing its end. 

Adding other macroeconomic indicators to the yield curve can produce a signal that’s much easier to interpret. It has remarkable accuracy, but the trade-off is a much shorter forecasting window – just four months ahead. Today the probability of recession in the next four months is less than 2%. Of the many indicators included (e.g. housing starts, new orders), the only one sending a signal even approaching contraction is the money supply impulse (changes in the flow of money). 

Some gloomy commentators have cited the falling price of industrial metals, which were down 15% year-on-year. Industrial metals are a useful ‘real-time’ indicator of economic activity, but they aren’t really a leading indicator. They are also exceptionally volatile and prone to false signals.

Our leading indicator for global GDP and our business cycle indicators suggest that the economy is still in expansion mode. This favours equities over bonds and cash. However, our global leading indicator is lower today than it was a year ago (only the US component is powering ahead). That suggests some caution toward cyclical companies and sectors is warranted.

Brexit means confusion

To help us filter out the day-to-day nonsense and make sense of how Brexit could develop over the coming year, we’ve created a simple decision tree.

It starts with the overarching question: Can the government strike a plausible deal with the EU? We believe there is a strong probability that it can – there’s just too much at stake on both sides of the Channel. As for whether they can do it in a timely manner, we think the chances are a coin toss (mathematics for don’t know). And if the government does arrange a deal before 29 March 2019, would it pass Parliament? That’s another don’t know. But if it were to, we think it would be with the cross-party agreement of everybody but the Conservative hardliners led by Jacob Rees-Mogg. We call that a “sandstone Brexit” – the softest rock (or hard Brexit) you can find. If the government and the EU cannot come to terms at all, there are two choices as we see it: a hard Brexit or a second referendum. The probability here is a coin toss.

When we calculate the total probabilities we find that by far the most likely result is a “never ending story”. The politicians agree on principles but not the details. In short, they keep kicking the can down the road. Sterling would likely rise, potentially benefiting investors who own domestically focused UK assets and hurting those with foreign investments and the FTSE 100, which both tend to do poorly when the pound strengthens. A hard Brexit has the second-highest probability. A no-deal break with the EU would send sterling tumbling as low as $1.20 or more, we believe. Again, holders of overseas assets and the FTSE 100 would probably do best. Then comes our sandstone option, followed by no Brexit. A sandstone agreement would probably make sterling amble lower, as the greater barriers to cross-border business will hamper the UK’s future prospects. However, a continued trading relationship would mitigate the pound’s fall. 

It’s always worth remembering that the FTSE 100 is not the UK economy and it has historically been a relatively defensive basket of global companies. Of all the major indices, the UK has the lowest sensitivity to global economic activity. It is has been one of the least sensitive to changes in peripheral European bond yields since 2010, and it weathered the Italian political crisis well. At a time when many leading economic indicators outside of the US are starting to roll over, a defensive bias within equities makes sense to us.

Midterm mutiny?

On Tuesday 6 November Americans will vote in the midterm elections. Sometimes known as off-year elections, the midterms occur every four years, between presidential elections. Politicians in the House of Representatives only serve two-year terms so all 435 of their seats are up for grabs. Senators serve six-year terms and work a staggered system whereby a third are elected every two years: this year 35 seats are on the table.

History does not favour the chances of the sitting President’s party at midterm elections – losses seem almost inevitable. It is rare for Republicans to be in the White House and control both chambers of Congress, and even rarer for them to hold on to this position after midterms. George W Bush achieved this feat in 2002; the last Republican to achieve it before that was Calvin Coolidge in 1926.

At first glance it may seem that the Republicans have a better chance of retaining the House than the Senate, where they have a much slimmer majority. But the staggered senatorial system favours the Republicans this year. Although 35 Senate seats are up for election, 26 of them belong to Democrat incumbents. Of those 26, 12 have a significant chance of swinging to the GOP, but only six seats held by Republican incumbents have a chance of swinging to the Democrats.

Republicans could lose the House for five reasons: (i) the spread of swing seats favours Democrats – according to the Cook Political Report/NY Times 88 Republican seats have a chance of turning blue (the Democrats only need a net 24); only 14 Democrat seats have a chance of turning red (3%); (ii) the extraordinary number of retiring Republicans, thereby giving up the incumbency advantage; (iii) historic losses for the Presidential party; (iv) Donald Trump’s approval rate – over 40% of the variation in House midterm results can be explained by Presidential approval ratings. Currently, Mr Trump’s approval rating is 41% (using the Gallup poll), putting the Republicans on track for a 9% loss; (v) the economy may be strong but it’s likely not strong enough to override the previous four reasons.

Of course forecasting election results is far more art than science and we certainly don’t recommend basing an investment strategy on election predictions. So will a Democrat victory even matter to financial markets?

Analysing the last 50 years, we find evidence that US equities tend to be rather directionless in the six months before a midterm election, with a little extra volatility. Over the subsequent six, equities tend to rise steadily. But here’s the rub: it doesn’t seem to matter one jot whether or not either party loses, gains, kept or never had control of Congress in the first place.

We found no evidence that midterms alter the path of US equities relative to global equities. The dollar doesn’t appear to deviate from the path it was already set on. Similarly, there’s no relationship between dollar performance and which party does well, poorly or indifferently. We’ve dug into sector performance too and again no pattern emerges.

Now to be clear, there just aren’t enough data points to draw firm conclusions from the historic performance around midterm results. In other words, this time may be different (heaven forbid). The point we wish to make is that if the Democrats do retake the House, be careful not to overestimate the magnitude of the impact on financial markets and, in particular, equity sectors. Investors may just be glad to get back to studying earnings reports and economic releases. After all, the third year of a President’s term has delivered the most consistently positive returns.

If, however, the Democrats retake both the House and the Senate, equity markets may not be so sanguine.

US stocks have received a substantial boost from Mr Trump’s tax plan and if Democrats controlled both chambers they could undo it. Most Democrat lawmakers are keeping quiet on their intentions here. US governments rarely reverse the previous one’s tax changes outright, but a redistribution away from the very wealthiest beneficiaries of Mr Trump’s tax plan seems inevitable. The key question is: will a Democrat Congress raise taxes and improve the budget balance, or raise taxes and spend the proceeds elsewhere? The latter is arguably more likely; Democrats have talked about an infrastructure programme. But despite the large economic gains possible from improving infrastructure, markets will not want to see it funded by a much higher corporate tax rate. 

Trading places

For us, the big risk is what a Democrat victory – in either the House or the Senate – does for Mr Trump’s trade agenda. Hamstrung in Washington, unable to enact the remaining policies on his to do list, Mr Trump may focus all of his attention on the one thing for which he doesn’t require Congressional approval: stoking the trade war.

The trade war escalated over the summer as the White House pressed ahead with tariffs on another $200 billion imports from China. The latest round brings us to a cumulative $250bn worth of tariffed imports from China. The tariff on the latest $200bn will be 10% but that will increase to 25% in January, as China has retaliated with its own tariffs on another $60bn of imports from the US. 

Mr Trump has threatened tariffs on an additional US$267 billion worth of Chinese goods. That would bring the total amount of tariffs threatened or imposed by the US on China to US$517 billion, accounting for essentially all Chinese exports to the US. In 2017, the US imported US$505 billion worth of products from China. 

A range of economic models predict that the impact of $250bn of tariffed trade would cumulate to c.-0.3% of US GDP after three years, and a little more than -0.3% for China. Of course, as a proportion of GDP growth, that means a bigger hit to the US than to China. But it’s still not all that much of an impact (because the $250bn is still just 8.5% of total US imports in 2017), and certainly won’t put either economy in danger of recession, all other things being equal. 

Of course, all other things may not be equal. There’s the risk that supply chain disruptions and the threat of even more tariffs will lead to higher business uncertainty, derail capex, crimp productivity growth and lead to much larger economic consequences. 

In applying his tariffs, Mr Trump has targeted Chinese exports of technological and intermediate, or component, goods. This means US companies which stand to suffer most are the hi-tech manufacturers. 

There are two ways they will be affected: (i) higher input costs; (ii) many Chinese exports are re-exports of goods with a lot of American value-added content or goods made by Sino-American joint ventures; so the tariffs are effectively a tax on US manufacturers! This sector is currently booming and responsible for a considerable amount of expected investment spending.   

Much of Mr Trump’s agenda is bent toward re-shoring manufacturing jobs which flowed out to developing nations over the past few decades. Under the renegotiated NAFTA deal, agreed with Mexico last quarter, 75% of a car’s components will have to be made in North America to cross the border duty-free, up from 62.5% under the old agreement. The kicker is that roughly 45% of components must be made in a high-wage environment ($16/hour). That means US workers will be much busier or Mexican workers will be breaking out the tequila because their average wage is about to shoot skywards from $3 an hour. The latter is perhaps less likely. Meanwhile, Apple complained that US tariffs would hurt its business and force it to raise prices for Americans. This was met by indifference from Mr Trump, who told them to build in the US instead. “There is an easy solution where there would be ZERO tax, and indeed a tax incentive. Make your products in the United States instead of China,” he tweeted. Apple’s primary manufacturing partner, Taiwanese-owned Foxxcon, has actually broken ground on a $10bn factory in Wisconsin just this year, so Mr Trump may get his wish. 

Other businesses may not be so eager. The latest US Federal Reserve minutes highlighted reports from regional agents that the trade war is weighing on CFOs’ minds. Anecdotal evidence accompanying the ISM business survey data shows companies are thinking about postponing capex because of the brewing conflict. The anecdotes don’t yet seem to be making it into the numbers, however – quantitative surveys of capex intentions and the actual fixed-asset investment data are still very decent. The impact will be more severe if the trade war escalates further. The Bank of England (BoE) estimates that if the US placed a 10% tariff on all imports, from all of its trading partners, US GDP could be around 5% below where it would otherwise be after three years. That’s -2.5% from the trade channel (disrupted supply chains, higher input costs, lost demand) and potentially another -2.5% from second-round effects on business investment. As uncertainty weighs heavy and financial conditions tighten, businesses need higher returns to compensate for their risks. This means potential projects get shelved because they are no longer worth the time, money and effort.

The BoE’s work makes a similar conclusion to that of an IMF study from 2015. The UK is one of the world’s most immune economies, while the possible cumulative impact on global GDP is -2.5%.

The effect of tariffs on the global economy is a present and growing risk, but not one that would compel us to shun stock markets today. There have been plenty of problems and set-backs for nations, economies and markets over the years since the last recession. There was the eurozone debt crisis of 2011, oil prices bottomed out at $25 in 2016, the vote for Brexit and another for Mr Trump. There was even a bubble in volatility derivatives that popped earlier this year and sent equities sharply lower. But if you had sold at any of those points, you would have missed out on substantial returns. 

It may be a very long-running show – and it may have become boring to some – but we don’t think this bull market will be cancelled in the coming months.