Consensus estimates of the probability of a U.S. recession have receded and our analysis of monthly economic data imply that risks of a recession are negligible for the time being.
An alternative model run by the New York Federal Reserve suggests there is a 25.2% probability of a recession over the next twelve months; we do not get concerned until that model rises above 40%. The U.S. economy, barring an issue with liquidity, an error on the policy front or a large exogenous shock, should continue slow and steady growth throughout the year.
The U.S. economy should continue slow and steady growth throughout the year.
The threat of an outright economic recession faded along with the rancor of the North American trade war episode and the lull in the trade disputes between the United States and China and the United States and Europe. Domestic politics has been the most likely catalyst for China rapprochement and for the sudden end of confrontation with the European Union.
We remain concerned about the pace of growth in the first quarter due to the shutdown of 737 Max production at Boeing and the spread of the COVID-19 virus, which will impact supply chains across autos, aerospace and retail in the period. RSM’s monthly GDP model implies a 1.5% pace of growth to start the year, and we expect growth in the first quarter to slow to near 1% due to the aforementioned risks.
2019 recession worries
During a six-month span beginning in the middle of 2019, it did appear as if the U.S. economy was ready to slide into recession. The manufacturing sector was already in recession, global growth was slowing and the constant threat of disruptions to the global supply chain had brought business investment to a standstill. The U.S. Treasury yield curve was inverted — a closely watched signal of the financial market’s assessment of future growth and considered a necessary — but not sufficient — condition of recessions.
Despite the lull in policy uncertainty, by the third quarter of 2019, there was no denying that the manufacturing sector was in decline and the total real GDP growth had begun to slip. The consumer sector was a clear holdout, having responded to a late-cycle fiscal boost in the form of tax cuts that juiced the stock market and underpinned high-income household spending.
Even so, real GDP growth remained above 2% throughout 2019. Our model suggests real GDP growth of below 2% throughout 2020 and into 2021, hampered by second-order effects from the global manufacturing slowdown and the lack of high-paying U.S. jobs that continue to be a drag on the export sector and disposable income. There is low likelihood of recession, as the figure below suggests, but it comes with an asterisk: The outlook could change for the worse if the COVID-19 virus is not contained as expected, an asset-price bubble happens to burst or our trading partners in South America, Asia and Europe were to fall into recession. On the positive side, growth could remain above 2% if U.S. military spending were to suddenly increase before the election.
RSM Monthly Index of Economic Activity
That low level of economic activity has been masked by surprisingly high consumer and government spending (and most recently by a drop in imports, which works to boost net exports). The RSM Monthly Index of Economic Activity suggests underlying monthly GDP growth of 1.6% at the end of 2019. (That low-growth scenario coincides with the range of consensus estimates for the first quarter, according to the Atlanta Federal Reserve’s NowCast.)
The estimate of underlying growth comes despite the January increase in payrolls and a similar bump-up in hours worked. While the 225,000 payrolls number drew considerable attention, jobs increases were confined to the service sector – where wages are considerably lower than goods-producing jobs — and the long-term growth trends in both data series continue to drift lower.
These downward trends in the labor market form the basis for our forecast of eventual consumer belt-tightening, particularly for the lower-income population, whose high multiplier effect of spending tends to ripple through local economies and has ramifications for the economy as a whole.
Components of the probability model
Recessions are officially determined by the National Bureau of Economic Research, which has the formidable task of determining the month in which the economy stopped or started growing. Our recession probability analysis uses a statistical model, which uses monthly economic and financial data to forecast the probability of whether the economy is or is not in a recession — a binary condition – rather than a standard model of the level of real GDP growth.
Age of the economic expansion — At some point, all recoveries run out of steam and become prime candidates for a major economic or financial event to send the economy into recession. We are currently in month 127 of the longest and perhaps most moderate of economic recoveries in the post-war era. All good things come to an end, but when will depend on how much the government decides to spend on a fiscal stimulus.
Residential investment — The U.S. economy is increasingly dependent on the consumer sector to provide demand for goods and services, with home sales forming the basis for much of that spending, with the total demand for housing (rental or purchased) adding to the price of all residential housing. As the figures below illustrates, residential investment tends to drop sharply as the risk of an economic slowdown increases, and then rises again once the economic crisis has ended and employment opportunities are restored.
Cumulative residential investment shows an even more dramatic pattern of rising throughout the expansion and then reaching a peak in the months prior to the next recession.
In the current cycle, residential investment recovered after the 2007-09 housing bubble, but has drifted lower since 2010, either because of buyers being priced out of the market or because of the paucity of high-wage employment.
Labor market expansion – We represent the degree of expansion in the labor market by the change in the unemployment rate and the number of hours worked by manufacturing employees. Unemployment tends to decline as an economic recovery matures, and then climbs rapidly again as employers shed jobs during periods of low demand. Conversely, the number of hours increases during a recovery before plummeting during a recession.
In the current cycle, unemployment continues to fall, with most job gains in the service sector. And to prove the point, manufacturing hours began to drop when the first trade tariffs were announced in April 2018, and have continued to drop as the current manufacturing recession continues.
Corporate risk – The risk of corporate default can be measured by the difference between the yield of corporate bonds and 10-year Treasury bonds, with U.S. government debt considered to be a safe and guaranteed investment. Corporate risk would be assumed to drift lower as an economic recovery took hold, and then jump sharply higher as corporate defaults increased during an economic downturn.
In the current business cycle, corporate bond spreads began to push higher again, once attention turned to the administration’s trade policies. Spreads dropped when confidence began to rise again along with the upsurge in the stock market in 2019.
