Earlier this month, at an event hosted by Thomson Reuters, we presented some of the highlights from our Global Economic and Markets Outlook for 2018 Q3. We were joined by former Bank of England policymaker, Dame DeAnne Julius, and Shamik Dhar, Chief Economist at the Foreign Office.
The event began with a presentation of Fathom’s central, or most likely scenario, which sees the US economy begin to overheat after several years of above trend growth, triggering a recession in 2020. This is not a rerun of the global financial crisis, rather it is a story of good old-fashioned boom and bust. With most of the advanced economies unable to loosen materially either monetary or fiscal policy, the US takes much of the global economy down with it.
For now, there is plenty of good news about the US economy. Our US Economic Sentiment Indicator (ESI) has been reporting near-record levels of confidence among US firms and households for the best part of two years. This has finally begun to translate into firmer GDP data, as we suspected that it would. Growth in the second quarter, at 4.2% on an annualised basis, was the fastest in almost four years.
The difficulty, as we see it, is that the US economy is growing and is set to continue to grow at a rate that is unsustainable. In our judgement, the trend rate of growth of the US economy is somewhere in the range 1.0%-1.5%, shown shaded in our second chart. Since early this year we have consistently forecast growth of around 3% in the US both in 2018 and in 2019. And the consensus, as measured by the Reuters Poll, has moved towards us — either in full (for this year), or in part (for next year).
Books have been written about the difficulties involved in producing real-time estimates of the output gap. It is not easy! Our view is that the US economy was operating a little above potential by the end of last year. This certainly chimes with official estimates of the Non-Accelerating Rate of Unemployment, or the NAIRU. According to the Congressional Budget Office the US unemployment rate dropped below the NAIRU in early 2017, shown in our third chart, and has fallen further below the NAIRU since then.
If we are right, both about the starting point for the output gap at the end of last year, and about the pace of economic growth in the US relative to trend both this year and next, then output will move substantially above potential over the next year or two. Other things equal, that will cause US inflation to rise over the coming months. There are already signs in the New York Fed’s underlying inflation gauge, which tends to lead both core and headline CPI measures by a year or two, that pressures are building.
Recessions are notoriously difficult to predict. They are infrequent, non-linear events, often characterised by a rapid reassessment of future economic prospects on the part of firms and households. Spotting recessions may not be easy, but that does not prevent economists from trying. Back in 1996, two economists at the Federal Reserve Bank of New York, Arturo Estrella and Frederic Mishkin, published what is still considered by many to be the benchmark when it comes to assigning numerical probabilities to US recessions. Their model is simple. It has only one input, and that is the slope of the US yield curve — the difference between the yield on ten-year US Treasury notes and on three-month Treasury bills. They find that the less steeply upward sloping is the US yield curve, or the more steeply downward sloping, the more likely is the US to suffer a recession. The intuition is quite straightforward. When investors sense that a recession is around the corner, they are likely to revise down their expectations about future policy rates, and future rates of inflation, putting downward pressure on long-term interest rates relative to short-term interest rates.
Currently the Estrella and Mishkin model is barely flashing amber, with the probability of a US recession within the next twelve months standing at just under 15%. However, we question whether their model is quite what it used to be. With short-term interest rates still close to the lower bound, it must be the case that the slope of the US yield curve cannot become as negative as it once was. Does that mean that the chances of a US recession cannot become as high as they have been before? Surely not.
In light of these concerns, we have developed our own model of recessions, not just in the US but across 17 advanced economies. It is similar in spirit to the model of Estrella and Mishkin, but includes, in addition to the slope of the yield curve, a range of other macroeconomic and financial market indicators, such as the level of debt in the economy relative to GDP, and our own estimate of the output gap. Feeding in our forecasts for these indicator variables we find that, by the end of 2019, the chances of a US recession will be as high as they have been in the past 60 or more years. This empirical analysis supports the decision we took three months ago to make a US-led global recession in 2020 our central macroeconomic scenario.
We invited our panellists to give their own assessment of the macroeconomic outlook. DeAnne Julius and Shamik Dhar both felt that, although a US-led global recession in 2020 was a material risk, it was not, for either of them, the most likely outcome.
DeAnne Julius started by outlining the factors that, in her view, were supporting US growth. She cited the tax cuts, promised by Donald Trump during his election campaign and enacted earlier this year. But she also felt that the President’s broad deregulatory agenda had a part to play. There was in train a process of lessening the degree of environmental protection and easing the quantity of red tape in the provision of financial services. She felt that both these factors were likely to remain in place for some time to come. Moreover, there was the possibility of a virtuous circle, with animal spirits feeding through to faster economic growth, and higher corporate earnings encouraging more investment in fixed capital. DeAnne Julius felt that the biggest threat to the US economy came from what she described as a ‘Trump Tantrum’. This is the idea that the US President, currently engaged in a high stakes economic game with other countries, might overplay his hand. One example she gave was of an ill-judged decision to impose tariffs on European car imports. An all-out trade war risked triggering a severe downturn for the US, and indeed the global economy.
In his remarks, Shamik Dhar highlighted the fact that the two most recent US recessions had each been preceded by a financial crisis. The bursting of the dot-com bubble around the turn of the century undoubtedly contributed to the relatively mild recession through 2001 that has been identified by the National Bureau of Economic Research (NBER). The much more severe recession that, according to the NBER, ran from the end of 2007 until the middle of 2009 followed a realisation that many financial institutions around the world were holding CDOs — bundles of US mortgages — that were identified by the ratings agencies as low risk yet, in truth, were unlikely ever to be repaid. A growing lack of confidence in the financial system produced a sharp fall in the supply of credit, and with it a severe contraction in global economic activity.
Mr Dhar felt that, were the US to suffer a more traditional recession, where a period of overheating leads to rising inflation and higher interest rates, then that would in fact be a good thing. It would mean that the US economy had escaped a world of secular stagnation, where deficient demand makes it very hard to generate rising inflation. He felt, on balance, that we were still in that world, which is why he did not expect to see large increases in US inflation or US interest rates. He did, however, see some alternative mechanisms that might produce a global recession within the next year or two. The first of these was a financial crisis in the emerging market economies. As US interest rates drift slowly higher, and the dollar strengthens, less well-off economies that have relied on cheap dollar funding might be found wanting. The second was a more generalised trade war. A combination of rapid economic growth and a strong dollar meant, in his view, that the US current account position was likely to deteriorate further, perhaps prompting the US President to extend his protectionist measures to other countries, and to other products.
As usual, we closed the event by inviting our audience to express their own views on the macroeconomic and financial market outlook. Interestingly, when asked to put a figure on the chances of a US recession by 2020, the audience was split down the middle. Precisely one half of those who voted saw the probability as less than 50%, with the remaining half judging that the probability was greater than 50%, perhaps confirming that recessions are indeed hard to spot!
Financial time series database which allows you to identify and examine trends, generate and test ideas and develop view points on the market.
Thomson Reuters offers the world’s most comprehensive historical database for numerical macroeconomic and cross-asset financial data which started in the 1950s and has grown into an indispensable resource for financial professionals. Find out more
We have revised higher our forecast for US Q1 GDP growth, from an annualised rate of 2.0% ...
The latest reading from Fathom's China Momentum Indicator (CMI 2.0) combines the first ...
China has enjoyed a formidable economic ascent. At the turn of the millennium, China’s ...
After experiencing their worst quarterly net outflows ever in Q4 2018 (-$10.1 billion), ...