Via Capital Economics today:
Our view that the world economy will undergo a reasonably sharp slowdown over the next couple of years stands in contrast to the relatively rosy consensus. (See Chart 1.) And it has led several clients to ask the obvious question: what causes the downturn in our forecast that others may be missing?
This was the theme of a series of presentations we gave to clients in Europe last week. One point we emphasised was that it’s rare for economic downturns to be caused by a single factor – most tend to have their roots in several different areas.
Chart 2 shows our assessment of the factors that have contributed to recessions in the G7 economies since 1960. Monetary policy tightening contributed in a majority of cases – 29 out of 45. But numerous other factors played a role including the bursting of credit bubbles, oil shocks, property collapses, and so on. The downturn in our forecast is no different.
A handful of countries are facing relatively sharp adjustments. In the emerging world, previous financial excesses mean that Turkey and Argentina are now set for deep recessions. And among the advanced economies, we expect Australia, Canada and Sweden to experience sharp slowdowns as frothy housing markets cool. But while our GDP forecasts for all of these countries sit well below the consensus, they are too small to be at the epicentre of the next global downturn.
Likewise, while a trade war has the potential to inflict substantial damage on the world economy, it would require a significant escalation from the measures seen so far for it to be the trigger of the next global downturn.
Instead, the slowdown in our forecast is led by the world’s two largest economies – the US and China.
In the US, we expect the downturn to be caused by old-fashioned monetary tightening. Real interest rates have already risen by 2%-pts since 2015, with further increases likely as the Fed continues to hike policy rates (we expect three more 25bp increases in the Fed Funds rate in this cycle). Of course, even after the additional tightening in our forecast, real interest rates in the US would still be relatively low by past standards. But it’s likely that the neutral level of real interest rates has fallen in recent years (more on this later). And, in any case, history suggests that it is the change rather than the level of real interest rates that may be more important.
As Chart 3 shows, there have been four previous occasions since the mid-1980s when real interest rates in the US more than 2%-pts – and three of these ended in recession. With the effects of this year’s tax cuts also likely to fade, we expect growth to slow sharply in 2019.
Policy tightening is also behind the slowdown in China, which we believe is already well underway. Admittedly, you can’t see much of a slowdown in the official data and probably won’t over the quarters ahead either. But our China Activity Proxy, which we believe provides a better indication of the health of the economy, shows that growth has already slowed from around 6% y/y a year ago to 5% y/y now.
Policy is now being eased in China in response to signs of fading momentum, but it will take time to have an effect. And when it does, we think that stabilisation at around 4% y/y in the middle of next year is more likely than a rebound. Stepping back, a lack of commitment to market-centred structural reforms by China’s leadership means that long-run growth prospects are continuing to worsen. We expect trend growth to drop to 2% over the next decade.
So far, so gloomy. But while the slowdown in our forecast puts us at odds with the relatively rosy consensus, it’s also relatively mild compared to previous cycles. Severe downturns tend to be triggered by the bursting of asset price bubbles and a subsequent credit crunch in the real economy.
Yet the normal warning signs are not flashing red. Banks have reduced their riskiest forms of lending and are now better capitalised. Meanwhile, although asset prices look stretched in a number of areas, it’s difficult to make the case that they are dangerously overvalued if we accept the premise that neutral real interest rates are now much lower. Accordingly, in our view the next global downturn is more likely to look like the one in 2001 – which was relatively small and short-lived – than the one in 2008-09.
Where might we be wrong? Pinning down the timing of the next recession is more of an art than a science and it’s possible that world growth may continue to chug along at decent rates through 2019. But 10 years into the global recovery, the bigger risks lie on the downside. Three stand out. First, debt ratios remain high in most major economies, which increases their vulnerability to shocks. China gets most headlines, and a financial crisis there would have global ramifications. But Italy may now be the weakest link in a fragile chain, not least because it operates within the constraints of monetary union and the sheer size of its bond market would hamper the ability of the ECB or IMF to provide assistance should it get into trouble.
The second risk relates to the neutral level of real interest rates, which I’ve already referred to a couple of times. This is a theoretical concept but one that has critically important implications.
As I’ve noted, there is a widespread consensus among economists that R* has fallen in the world’s major economies in recent years. But the size of the fall is unclear. This creates risks in both directions for policymakers. If R* is lower than central banks believe, then the effective monetary policy stance will become tighter at lower levels of real interest rates – meaning that modest hikes in interest rates could trigger unexpectedly sharp downturns in activity.
In contrast, if R* turns out to be higher than is generally believed then – depending on the drivers of that rise – the current level of many asset prices could become difficult to justify. This raises the possibility of a sharper downturn in the global economy, triggered by steep falls in asset prices against a backdrop of high levels of debt.
The final risk is that a series of events that, on their own would have a relatively small effect on the global economy, interact to produce effects that are much larger than the sum of their parts. I argued earlier that, it would require a significant escalation in trade protection from the measures seen so far for it to do serious damage to the global economy. But it’s possible, for example, that a modest escalation triggers a sharp fall in some equity prices, which in turn creates financial dislocation that feeds back to economic activity. This is the nature of a global economy that is now 10 years into the cycle and where high levels of debt increase vulnerability to shocks. Viewed this way, our forecast for a modest downturn over the coming years starts to look less bearish…
I’ll take the latter point. Although there are no obvious asset bubbles beyond the usual suspects in small economy housing markets, there are some pretty nasty imbalances in Chinese and US corporate debt. I’d expect them to cause significant equity market dislocation when the moment comes which will exacerbate the underlying slowdown.
No repeat of the GFC, certainly, unless somewhere like Italy causes a massive monetary shock. But still enough to cause a global recession (considered anything below 2% growth) and one that is difficult to recover from as monetary and fiscal policy will need extensive innovation first.
For Australia it looms as a very difficult test as the above overruns household debt.