A few weeks ago, I wrote about the procession of data to watch in the event the inversion in the treasury yield curve were to lead to a producer-led recession. In this post, I consider the same procession as it might apply to a consumer-led recession. As I will argue in this post, there is nothing in the present data suggesting a progression to such a consumer-led recession.
As an initial matter, recall that a yield curve inversion is not infallible (see 1966 and 1998). And per my missive of several weeks ago, right now the yield curve is the only long leading indicator of the seven that make up my major forecasting system that is negative. The rest are either positive or neutral. But to return to the main point of this post, with the inversion of the 10- to 2-year spread last week, the only positive part of the curve is that from 5 years out to 30. Virtually all other comparisons are inverted.
The mechanics of a consumer-led recession
Turning to the consumer, there is another nearly infallible metric that typically appears about one quarter before the onset of a recession: a negative YoY change in real retail sales. The below graphs show the relationship of the - yield curve (blue) with real retail sales (red):
The only two false positives of negative real retail sales are the same as for the inverted yield curve: 1966 and 1998. There are no false positives since 1960.
But since there can be a year or more between the yield curve signal and the real retail sales signal, and since the latter occurs very close to the onset of a recession, is there anything in between? There are two reasonable candidates: real consumer durable goods purchases, particularly of vehicles; and bank tightening of consumer credit. The first comes from the recession model of UCLA Prof. Edward Leamer. The second comes from a noted BIS study.
First, let’s compare overall durable goods spending (blue) with motor vehicle spending (red):
Note that I have used CPI as the deflator, partly because it is used to deflate retails sales and partly because no GDP deflator goes back long enough on a monthly basis.
Spending on motor vehicles typically has turned negative one quarter before total consumer durables spending.
Now let’s compare light vehicle sales (blue) with real durables spending on vehicles, adjusted for inflation (red). Note the first is the numbers bought, whereas the second is the dollars spent:
These have tended to move in tandem. Since inflation rather than deflation has been the rule for many decades, this means that sales have turned down before nominal spending on vehicles (not shown).
So now, let’s bring in bank lending practices.
First, banks have typically raised interest rates for auto loans (shown YoY and inverted, so that negative = increased interest rates) even before the treasury yield curve has inverted:
The next graph compares motor vehicle sales (blue) with banks’ willingness to lend to consumers (gold), and specifically with bank lending standards for auto loans (violet diamonds - sorry, that’s the only way FRED displays that metric!):
But overall, banks have been increasing, if slightly, their willingness to make consumer loans. Over the past 40 years, before recessions, this metric has turned neutral, with no more than a 3+ reading, several quarters before a downturn.
More generally, bank unwillingness to lend, whether to consumers (blue) or to businesses (red) has been a consistent precursor of recessions after a yield curve inversion:
So, the fact that no such tightening of bank lending conditions has occurred means to me that the mechanism for transferring the effect of an inverted yield curve onto bank lending practices has not taken place.
To summarize, the typical order leading up to a consumer-led recession has been: (1) interest rates rise; (2) the yield curve invests; (3) lending standards tighten; (4) motor vehicle sales decrease, usually by 10% or more; (5) durable goods sales generally decrease; and (6) real retail sales overall decrease - all before a recession starts. In the past few years, we can see that parts (1) and (2) have been repeated, while (3) is only neutral rather than truly negative for now, and (4), (5), and (6) have not tipped over yet.
In other words, if a yield curve inversion ultimately telegraphs a consumer-led recession, typically the process appears to be transmitted through a tightening of bank credit to consumers, which shows up in a decline in consumer purchases of vehicles and in consumer durable goods purchases generally, and then, finally, retail sales turn negative YoY.
None of those precursors to a consumer-led recession is in evidence. Unless and until they are, I find it very difficult to believe that a consumer-led recession is on the way.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.