Banks Starting to Cut Back on the Cheap Money
More evidence of erosion in credit this month as the Fed’s lending standards surveys come out. The Fed releases these in 3 series: for large and middle industrial and commercial loans; for small industrial and commercial loans; and for consumers. Each gives a different set of information.
First off, why lending standards matter. In the Classico-Austrian understanding of the cycle, easy money subsidizes poor-quality investments – malinvestments – that are sustainable only during the “tissue-fire” demand conditions during the boom.
If this easy money is restricted, two things happen: first, those demand conditions get worse – consumers and investors aren’t so flush with cash as they were. Keynesians see this part, so little disagreement, and the market digests this part.
But the demand story is relatively unimportant for Austrian. Instead, the recession timing is dominated by a second effect, coming from the tightness of capital itself. Because tight money pushes those borderline businesses not only lose revenue momentum, but they can’t find new investors to sustain any losses, while what capital is available is going at a higher rate.
Put these factors together, and the lowest-quality malinvestments begin to liquidate. If those liquidations cluster, we call it a recession.
So, in this narrative, the key factors is how easy money is – how available, and how cheap. Cheapness of capital, to a first approximation, can be read off the Fed Funds Rate.
As the base rate, the Fed’s rates sets the bar for capital across the economy. Firms will typically pay a fairly constant factor above this rate, depending on credit-worthiness, with small variations coming from overall market sentiment. So these premiums will vary – in 2008, for example, spreads actually rose, as marginal firms paid rising premiums above the Fed Rate — but the Fed’s target rates gives us a pretty good approximation in most cycles.
Now, the problem today is that we’re getting abnormally little information out of Fed rates. Because they’re stable and low, instead the fluctuations in easy money is shifting towards non-rates policy such as quantitative easing. Remember the clustering of malinvestments is based on changes in ease of money, not just the base rate.
So, if we were to do a naïve extrapolation, basing on past recessions, we’d simply conclude that rates are so low it’s all blue skies ahead. But if we look at the changes, embedded in non-rates manipulation, we’ve got changes in tightness of money.
We want, then, to dig deeper, to see how credit’s actually playing out in the real economy. We want to move from saying, “winters are cold, so if it’s winter it must be cold” to, instead, buying a thermometer.
So, with that, let’s plunge in to the Fed’s lending survey.
First, large and medium industrial and commercial loans. This chart’s telling us what percent of bankers are telling the Fed, in a survey they mail out, that they’re tightening or loosening lending standards on commercial and industrial loans to large and middle-market firms. After 5 years of perfect calm, standards have finally started tightening. These numbers were just out this month, but they’ve got tightening starting in Q4 2015.
The first questions we should ask about any indicator is how often it false-alarms, and how early is it. After all, we don’t want to cycle-in and cycle-out every time we get noise, so we want to ignore false alarms. Meanwhile, an early indicator can be worse than no indicator at all, if it makes you miss the boom entirely.
First, the false alarm. Looking back to the 1980’s, lending standards to large-and-mediums does spike without follow-through. But those spikes typically last a single quarter. In 1998, for example, they leapt in Q4, but then dropped the following month. This may be due to idiosyncratic factors – things like tax rates or bankruptcy bills – that we don’t want to extrapolate as cycle indicator.
However, what makes this tightening worth talking about now is the confirmation: rarely does a tightening jump sustain for two quarters, usually reversing course the following quarter. In this case, that’s exactly what we got in Q1: the jump sustained. At a slower pace, so probably not time for panic, but sustained nonetheless.
So we don’t want to completely ignore this as noise.
The next question, though, is how early of an indicator this is. Turns out, lending survey tightening is pretty good. In the dotcom bust, for example, stripping out the one-quarter jumps would put you selling after the Q4 1999 numbers. So actually selling around November 1999 at roughly 1400. Considering the market topped out less than 10% above that in early 2000 before plunging to 800, this is not bad at all.
In the next 2008 recession you actually got even better timing. You got the sell signal after Q4 2007 numbers, again putting you at a November sellout. This time it was nearly the precise market top, selling in the 1500’s before it marched down to 800 by early 2009.
Past results are not guaranteed and all that, but it’s a tidy indicator that’s born out of theory and kills it in the real world.
Making the key questions at this point confirmation and magnitude: how believable is the narrative that banks are finally tightening standards, and will keep doing it. After all, if the spike reverses, we’ll simply be looking at another false-alarm on the way to higher equity highs.
So to get a bit of confirmation, first let’s take a look at small firms. This is the same survey, asking bankers about their commercial and industrial loans to small firms.
The story we’re getting here is also the same; loosening standards for five years, now with two consecutive quarters of tightening. Small lending is actually higher variance than lending to large and mediums, so we get higher highs and lower lows. And small is confirming the tightening story among large and medium lenders.
Consumer Credit
Finally, let’s take a look at consumer credit.
Even here, we are getting confirmation. Not that we care very much, but at least it’s an insanity check in case business loans are tightening while consumer is not, which would indicate loose money simply changing its route of entry.
Pulling all 3 together on a single chart, we’ve got agreement that banks are tightening loans, the first time this has turned sustainably since the last recession.
The major caveat, so far, is that it’s still a small move. Banks have barely cleared the net tightening line. So your response to this will partly depend on your transaction costs, including tax impact, of selling out in a false alarm.
This means we want to be pretty certain, and not jump at shadows. So here’s two charts zooming in on a comparison of today with the lead-in to the 2008 crisis.
First, 2008:
And, second, today:
Comparing the two, in both periods we had a long period of progressively loosening standards, with a sustained increase starting in late 2007, and again starting in late 2015.
If we continue confirmation on tightening — if the pattern holds — we can probably expect a rougher 2016 ahead. The next lending standards are out in early May, so until then we’ll see.
If your assets are exposed to the next recession, read more about recession indicators, and if you’re hoping to shelter from the economic storm, check out this article on overseas diversification.
If you haven’t already, sign up for free updates in the box above, and if you’ve got more than $100k invested in cycle-sensitive assets, look into a subscription to my Austrian Investment Monthly.
Happy investing!