Nearing a recession

The US economy continues to march forward; after posting robust growth numbers in the second and third quarter, there is little doubt that the immediate future poses little threat to one of the longest economic expansions in modern history. But are we starting to see some signs of a slowdown on the horizon?

Although unemployment is hovering around post World War II lows, GDP growth is on the verge of logging an annualized growth rate of 3%, and inflation shows no sign of leaving its goldilocks range, I can see the country falling into a recession in approximately a year or so.

If we’re going by definition, a recession would require two consecutive growths of negative GDP growth. So theoretically we could experience a recession without recognizing it until several months past its worst stretches. The definition is similar to that of a bear market (categorized by a 20% drop from the latest peak): pretty arbitrary and rather a good rule of thumb.

So why may a recession be near? I’ll present my argument with two graphs. Let’s look at fixed residential investment as a share of GDP. Residential investment, as its name suggests, represents the demand for residential structures and equipment.

The important takeaway from this graph is not that residential investment falls during recessions, but rather that declines in investment tend to be a precursor to slowdowns. The gray shaded areas represent recessions, and it is evident that almost every recession was foreshadowed by a fall in residential investment. Our current period is not necessarily in a decline, but it has failed to show meaningful growth for several quarters.

The relationship should make some intuitive sense: buying a house usually requires other additional purchases to supplement the house, and that demand will lead to greater income for all. Considering the magnitude of housing purchases, if you expect prices to decline you may be likely to hold out and wait for a more affordable opportunity. Unfortunately this can become a self fulfilling prophecy if everyone sours on the market.

Additionally housing purchases are financed by leverage and can help us see the current state of credit markets; economic expansions are founded on available credit while slowdowns are connected to tightening lending standards. When conditions make it more difficult to service debt payments, individuals may be less likely to assume the financial responsibilities of making regular mortgage payments. With the Federal Reserve on pace to raise interest rates a couple(?) more times over the next year, the days of easy money policy are in the rearview mirror.

But how do we know that residential investment simply isn’t in a lull that will eventually result in another rally? Let’s examine new one family houses sold:

We understandably see a similar pattern, but now it is evident that house sales are on the decline. Given the clear relationship between housing purchases and residential investment, we can conjecture that there will be a fall in the latter in upcoming quarters.

As to why these declines are occurring, maybe Millenials are simply opting to take alternative paths of living, in which case we would have to evaluate how those demographic trends may impact data. But the tax bill is starting to wear off, and several provisions in the new law discourage home-ownership through limited deductions. There are few signs that productivity will climb out of its decade plus long rut, and slowdowns in many developed economies have experts wondering how long the US can succeed in near isolation. Pair these concerns with geopolitical issues around the globe and the economic outlook over the next couple years increasingly appears to include a recession in the agenda.

A quick but important note: looking at the graphs may cause one to wonder why there was an instance during the 1960s where residential investment fell but a recession did not follow; in addition we have the early 2000s recession, which was not preceded by a slowdown in investment. These two instances are known as “False Positives” and “False Negatives” respectively. The 1960s recession was averted in part because spending for the Vietnam War helped carry the country’s economic load. Separately the early 2000s recession was more an isolated business recession than anything else; the bursting of the tech bubble exposed weaknesses in corporate America that lead to a faltering over the next few years. So with those two facts in mind we can say that there is no ramp up in war spending coming to save the US this time around.

With all of that considered, I could see a recession beginning in the fourth quarter of next year or in the beginning of 2020. Of course nobody can predict the future and even the brightest minds in economics often fail miserably at forecasting, but given the current data I would not be confident betting against the possibility.

-Brian Proferes