The recession that keeps stubbornly not happening

This morning, I'm going to anthropomorphize macroeconomic events arbitrarily defined by the coincidences of where we draw a line in reference to Earth's orbit around the sun and the aggregation of the purchasing decisions that we can measure at the national level.

Hello?!  Recession?  Where are you?!

Let's all remember what a recession is, in technical terms.  It is a period consisting of two consecutive quarters (six months, lining up with specific calendar periods) of negative GDP growth.  Or, as George Costanza would put it, shrinkage.  Lots 'o folks have been waiting for that Trump recession since he took office, but more particularly, I'm interested in the last year or so.  Why?  Two things.  The trade war, obviously, but even more particularly, the "yield curve."

Remember the "yield curve?"  Probably not.  Why not?  It stopped being a story many months ago.  And that's why I'm writing about it this morning.  Last Friday, we had a positive jobs report, and stocks went up, and blah, blah, and that got me to remembrin' and a folksy, old-timey way...  You see, I like to do that.  Not misspell things intentionally, although I do occasionally enjoy that.  Rather, I like to think about whatever everyone else isn't thinkin' 'bout.  That's the "yield curve."  Let's take a step into the wayback machine.  Come along, Sherman!

The yield curve refers to the gap between the interest rates on short-term and long-term treasury bonds.  If you are going to have your money locked up for two years, how much interest do you want on that?  OK, now if you are going to have your money locked up for ten years, what kind of interest are you going to demand for that?  Obviously, you are going to demand a higher interest rate on the longer-term investment, right?  That's basic.

Every once in a while, though, something weird happens.  You see, every once in a while, investors start trading bonds in a way that is messed up.  You start seeing lower interest rates on, say, the ten-year bonds than on the two-year bonds.  Huh?!

Yeah, that's right.  It's screwy.  We call that an "inverted yield curve."  When the yield curve inverts, it is a pretty good statistical predictor of an on-coming recession.  The basic thinking behind that is that if you think the economy is going to tank, you want your money protected, and you'll pay a premium for that protection, in the form of the security you get from US Treasury bonds.  Or, like, something, 'cuz we gotta reverse-engineer an explanation for the fact that inverted yield curves happen, and that they tend to presage recessions.  Yay for microfoundations!  [Obscure, jargon-laden zing!  Trust me.  That was, like, a "sick burn," or something.]

Anyway, the yield curve inverted a bunch of times, for brief periods recently.

Oh no!  Batten down the economic hatches, put all of your money into gold, and... say goodbye to Trump!  ... R... Right?

Or, not.

So, what did I go around telling people when the yield curve inverted?  Two things.  First, an inverted yield curve can't cause a recession, and second, the idea that it is a mystical recession-predictor depends on the notion that bond investors, rather than, say, stock-investors, have mystical, recession-predicting powers.  After all, stock investors, businesses, and other economic actors aren't acting like a recession is coming.  And... bond investors aren't actually a whole separate set of actors!  Investors diversify, depending on their risk tolerance, proximity to retirement, and so forth.  There are fund managers, but fund managers pretty much all suck anyway.  So... what's the deal?

If you go around looking for "stuff that has happened right before past recessions," you'll find stuff in the data.  We have a lot 'o data!  It reminds me of an old scam.  It goes as follows.  Put together a list of, say, 10,000 names and addresses.  Send half a "newsletter" predicting that a stock will beat the market, and the other half a "newsletter" predicting that the stock will underperform.  Given that short-term fluctuations in stock prices are basically random, half will get a newsletter with a great prediction!  Among that 5,000, send half another newsletter the next week predicting a stock will overperform, and the other half another newsletter that it will underperform.  Half-- 2,500-- will have gotten great predictions two weeks in a row, on two different stocks!  Do that a third time, and 1,250 get three amazing predictions, three weeks in a row!  They'll think you're a genius!  Ask them to pay up to keep getting your "newsletter."  Tell 'em how much money they can make by beating the market using your amazing powers of prognostication!

Do NOT do this.  This is called "mail fraud," and it's a federal crime.  I'm just making a point here about prognostication and how easy it is to find a prognosticator if you go looking for one.  If you go looking in the economic data for a thing that happens before recessions, you'll find it.  Like, maybe, an inverted yield curve!  But, once you pay for that newsletter, you might find you've been had.

And if you put your electoral expectations on a recession based on the observation that the yield curve started intermittently inverting a few months back, you might find yourself with a newsletter good for little more than bird cage liner.  It should be defecated upon by creatures who mimic sentience, but lack understanding of the phonemes that escape their vocal cords.

You know, those critters on cable news.

Is a recession coming?  I do not know.  It keeps not happening, though, and we should not be surprised.  The only potential cause has been the trade war, but most of it so far has been rhetoric, with escalation being a slow build.  We have new tariffs coming into effect, but the macroeconomic effects of those will be slow to show in the data.  How do we read claims of a China deal?  Cheap talk.  We have a slow build on tariffs, and maybe there will be a deal eventually, but the point is that the tariffs are limited right now.  So, that has kept recession as a possibility, but it hasn't tanked the economy.

From a political perspective, what matters is Q2 of 2020.  That's comin' 'round the bend.  Could a recession happen that quickly?  Yes.  Is that likely?  Remember that recessions have causes, and an inverted yield curve can't cause a recession.  It is a thing that usually presages a recession.  Which... hasn't come 'round the bend.  So, what's the cause?

I don't know if a recession will happen, but the core lesson for today is to not forget yesterday's story.  This is hard.  The fact that an inverted yield curve was big economic news for a while, getting Democrats' hopes up, and Republicans nervous, can be quickly forgotten amid political news, but that fact itself is what you must remember.