View Single Post
Old 09-16-2019, 09:58 AM
Corry El is offline
Join Date: Jan 2013
Posts: 3,948
Originally Posted by Sam Stone View Post
Predicting a recession is a lot like trying to predict whether the next grain of sand dropped on a pile will cause it to collapse. It's impossible, because of complexity. But even if it's impossible to literally predict, you can certainly see conditions that suggest that things are starting to get unstable.
That's where we are at now. Record debt, trade tariffs, a slowdown of the economy in Europe and Asia, interest rates near the zero bound, moribund long-term investment. We can see that the pile is getting pretty big, and pebbles are starting to roll down the sides, but we have no idea when the collapse will come, other than it sure looks like we're creeping up on it.
You can't predict timing I agree. And it's not entirely clear we're that near the actual casual factors of previous recessions. Which have been first and foremost Fed tightening cycles* and increase private sector (not govt or total) debt loads. The Fed *was* tightening, though now loosening again, so if there is a recession in the next year or two it might be argued it follows past cases. But not very aggressive tightening and now going the other way. On debt, the relative cost of servicing debt for households and corps in the US is still significantly lower than 2007. Debt in general in the world is higher, but the kind of increase in debt servicing cost relative to income in the US that's preceded most US recessions, not so much right now.

OTOH the gradually greater internationalization of the economy, or even the recent backsliding of that (trade tensions) makes past correlations of US-centric measures to US recessions less reliable. Which could cut either way now. On the pro-recession side there's definitely slower growth in a lot of the world, apparently much higher chance for example Eurozone goes back into recession soon, and China not recession as in negative growth but negative compared to recent growth. On anti-recessions side the world financial system has become more USD-centric even in recent years, even since the 2008-9 crisis, even as US GDP as % of world GDP goes down, and even as import/export is a relatively low % of the US *real economy*. Meaning on one hand the shape of the US treasury yield curve is more influenced by overall global financial conditions than it used to be, but the US real economy while certainly affected by world economic conditions, isn't affected to the same degree. This is a notable caveat IMO for the 'yield curve shape predicts recession' argument which also as others have pointed out is based on actually pretty few examples statistically speaking.

And that's even assuming the statistical process is stationary, which it's not. Derivations of the supposed underlying statistical process in finance and economics often assume it's like inferring the odds of a dice or roulette table by sampling outcomes. But in case of finance/economics the 'dice' or 'table' is probably also changing shape between trials. This tends to defeat prediction by means of 'what's always happened' to a degree people tend to underestimate IMO, even people who should know better or are even those paid to make such predictions.

*inverted yield curve being in part a symptom of relatively tight Fed monetary policy, they directly control short term not long term rates.