Over the last couple of days equity markets have intensified their downward trend. Worries about China and falling oil prices are now accompanied by another major issue, the (increasing) possibility of a U.S. recession. But is a recession really on its way? To shed some light on this, here are 13 charts on a U.S. recession.1. Yield Curve
Historically, the yield curve has been one of the most reliable recession indicators. The chart below confirms this. Whenever the yield curve inverts, a recession is near. However, the graph also reveals that the yield curve is far from inverted, currently. Hence, do not expect a recession any time soon.
2. Yield Curve Revisited
If only it was this simple. As known for some time, the great global monetary experiment has massively distorted bond markets, especially at the short end of the curve (although LT yields in Germany, Japan and Switzerland look pretty crazy too.) Because of the major central bank intervention, yield curves are 'artificially' steep. Deutsche Bank takes this artificial steepness into account to construct a more reliable yield curve. Based on this adjusted yield curve the probability of a U.S. recession equals 46% (see chart below), significantly higher than recession probability derived from the traditional yield curve. According to Deutsche, a U.S. recession is too close to call.
3. ISM Manufacturing
Up next is the ISM Manufacturing Index, also a well-respected gauge of economic activity. But also one that is perhaps more useful for signaling the trend in GDP growth than for calling a recession. Just take a look at the graph below. The ISM Manufacturing Index falls below 50 quite regularly without triggering a recession. Historically, much lower levels of the ISM (roughly below 43) are required to successfully forecast a recession.
4. Periods of low ISM
The graph below confirms the limited forecasting power of the ISM Manufacturing Index. Even when the index remains below 50 for a considerable period of time a recession is far from certain. When the ISM index comes in below 50 for four consecutive months, just like we have seen in the last four months, a recession is 'only' 57% likely to occur. Let's call this one even as well.
5. Initial Jobless Claims
While employment-related data are lagging by nature, they too can signal recessions. The graph below shows that initial jobless claims skyrocket in times of negative GDP growth. No surprise there. But it also reveals that troughs in initial jobless claims are often followed by recessions. The important thing to keep in mind here is that the time lag between troughs in jobless claims and recessions is significant. On average the lag is close to 12 months. This implies that if (this is not certain yet) we have seen the bottom in initial jobless claims a recession is likely to be a year out. Hence, a trough in initial jobless claims does not signal imminent recession risk. This raises the question if markets are looking that far ahead? My guess is 'no'.
6. Labor's Share of National Income
This recession indicator hasn't popped up all that much, recently. The reason behind this is pretty straightforward. Labor share of income has been falling until very recently, whereas a peak in labor share often coincides with recessions. Labor scarcity, rising wages, rising inflation expectations, possible overheating, monetary tightening and pressure on company earnings are just some of the characteristics of this labor share recession dynamic. I think it's fair to say that the current environment, as shown in the graph below (apology for the blurry picture), doesn't fit that profile. 'No recession' from a labor share of income perspective.
Since corporate America is often the swing factor of the U.S. economy, earnings are a solid recession indicator. Especially profit margins have an impressive track record for signaling recessions. Whenever profit margins fall by more than 60 basis points, as has currently happened, a recession is very likely. From a top-down profit margin perspective (leaving the whole split between commodity-related sectors and the rest aside) we should expect a recession.
The last two peaks in M&A were followed by a recession within a year, as is shown in the graph below. The idea behind it is pretty compelling. At the height of the economic cycle CEOs tend to get overconfident or, perhaps a more CEO-friendly explanation, organic growth options decline. In each case it's not hard to imagine that a downturn lies around the corner, as the sheer size of this M&A boom makes you wonder. That said, with just two recessions in scope, the significance of M&A as a recession indicator is perhaps less than that of other indicators. On top of that, extreme monetary policy and the continuous lack of growth could distort the data. A peak in M&A could signal a recession, but it's difficult to tell from these two occasions.
Breadth is an interesting dimension of recession. Can one part or segment of the economy cause the whole economy to shrink? This question is, of course very topical, given everything that is going on in the energy sector. But if history is any guidance, breadth is required to push GDP growth into negative territory. As the graph below shows, in previous recessions, production in at least 80% of all manufacturing industries declined. Currently, this number is just 28%. While this is a bit of a lagging indicator, I would say the odds are against a U.S. recession from this angle.
10. Fixed Income Investors
Fixed income markets are better in predicting the economy than equity markets, right? Whatever the answer, spreads surely provide useful information, especially because they tend to start moving before equity markets do. As the graph below shows the current level of high yield spread has historically coincided with either a recession or a (serious) growth scare. According to the graph there have been three of each since 1986. So, while slower growth is pretty much a given, a recession is no certainty based on the high yield spread.
11. Equity Investors
Let's not rule out equity investors all together. The graph below shows that bear markets have often been followed by or coincided with recessions. In six out of the ten recessions since 1950 we also experienced a bear market. In three occasions the S&P 500 Index fell more than 20% without signaling a recession. But there were also three occasions in which the U.S. economy did enter a recession without stocks falling more than 20%. Equity markets can be a bit misleading from time to time. And, since the S&P 500 Index is currently 'only' down 13% from its peak, there is just not much to make of it.
12. The Models
Obviously, models can't be missing when forecasting recessions. And they surely aren't. In fact, there's an abundance of recession forecasting models out there. But hey all do the pretty much the same. That is, they all combine both economic and market indicators with solid forecasting power to estimate the possibility of recession. Their results look very similar as well. The chance of a U.S . recession is very slim.
If severe enough, or if it stays around long enough, the weak sentiment in financial markets itself can provoke a recession. As Anatole Kaletsky neatly points out in his recent article,
posted on the World Economic Forum website,
expectations in financial markets can influence (the probability) of events, hence a recession. To quote his words, 'As a result, reality can sometimes be forced to converge towards market expectations, not vice versa. This process, known as "reflexivity," is a powerful force in financial markets, especially during periods of instability or crisis.' While it is difficult to catch this reflexivity in one single chart, the one below does give a good idea of how this thing could work. Falling markets reduce the number of people who think the economy is getting stronger. Once expectations about the real economy are adjusted downwards, companies scale back their spending, et voila, a self-fulfilling prophecy is born.
So, that leaves the question of what to make of this all? In short, although a new U.S. recession can't be ruled out at this point in time, I would not bet on it. The odds favor 'no recession' in my opinion.