David Rosenberg went from saying we were in "the last inning" six months ago to declaring that we're already in a recession. Considering he was the one who called the Great Recession while everyone else was buying a third house with a NINJA loan, I tend to listen to him.

I get a couple newsletters that are basically "random assortment of graphs that mean something." One of them is aimed at hedge funds from an extremely sketchy hedge fund that moved from Zurich to Liechtenstein when Swiss banking transparency laws changed. The other is the WSJ's Daily Shot, which is basically a hundred-or-so context-free graphs. Seeing the raw stream of data drives home the point that really, it's all fuckin' tea leaves. To whit:

Indicator 1: The Unemployment Rate

The last time we had a recession, you were "unemployed" if you had sent out three resumes in the past week and hadn't gotten a job. If you were doing odd jobs for cash, you weren't considered unemployed. If you filled out a survey for money, you weren't considered unemployed. If you were sick, you weren't considered unemployed. This time around if you deliver a meal for TaskRabbit you aren't unemployed. If you drive a shift for Uber you aren't unemployed. If you spend 20 minutes on Mechanical Turk you aren't unemployed. The definition of "unemployed" has been gerrymandered to "have earned no money, was in perfect health, was looking for work by sending out resumes, was not in any kind of vocational training and was not affected by seasonal considerations for the past two weeks." Using it as an indicator of the strength of the economy is pretty ridiculous but the economists are loath to let go.

Indicator 2: The Yield Curve

This is the one everybody jumps up and down about because it's so mathy. Remember: "technical analysis" in equities doesn't mean "analyze the data based on technical facts associated with its creation" it means "perform phrenology on the graph itself and scry for trendlines." The New York Times doesn't want to simplify it to the point where it makes sense: An "inverted yield curve" means you make more profit loaning money in the short term than you do in the long term. Even then we have to be twitchy about it; not just any yield curves will do, it has to be the 90 day and the 10 year because reasons. So that you trust those reasons they put lipstick on the pig by doing fancy shit like this:

...but it doesn't change the fact that all we're really talking about is investors keeping their powder dry. And most of those investors are computers, and they can't make any money keeping their powder dry, and a disturbing percentage of the world is under negative interest rates (and a disturbing percentage of the world's institutional funds are legally bound to buy bonds that cost them money because capitalism) and what you're left with is a problem every bit as big as the whole "driving an Uber once this week means you're employed" problem.

Indicator 3: The ISM Manufacturing Index

The ISM is survey-based, its weighting is arbitrary, and every regional Fed has their own index that seldom agrees with the ISM. More than that, the ISM isn't a number, it's an essay that gets picked apart and dissected every bit as controversially as the Psalms. It's not uncommon for an ISM report to have a great number and a bunch of dire prognostications underneath it which brings us to

Indicator 4: Consumer Sentiment

This is one reason the hedge fund guys like to talk about "smart money" and "dumb money" and their divergence; the entire profession of investing is basically attempting to emulate the "smart money" even if the "dumb money" is just buying an index fund because otherwise you look foolish at the club. The "consumer sentiment" on the Great Recession was "zomg nobody could have predicted" when if you look at the housing commentary in the 18 months leading up to the crash, everyone was talking about it. The "consumer sentiment" on the 2008 financial crisis was

While everyone paying any attention was getting all the money out of the system they could before the rules got changed back to something that wouldn't predict catastrophe. "Consumer sentiment" is a trailing indicator. Ask the man on the street "do you think our economy is fucked?" and he'll say "absolutely." Ask the man on the street "do you think you'll make more money or less a month from now?" and he'll wonder what the fuck you're talking about because he hasn't gotten more than a cost-of-living increase in six years but the headlines will proclaim "CONSUMER CONFIDENCE STEADY" and Beyond Meat will have a higher market cap than Con Agra.

Indicator 5: Choose Your Favorite

Remember - Goldman Sachs based their credit default swap business on the maxim that a national decline in housing prices was a "six sigma" event. And that's what's driving the machine right now - algorithms. Algorithms that aren't paying attention to the fundamentals, algorithms that have no idea what pork is, what bellies are, or what futures are but can calculate the odds of a rise in pork belly futures faster than you can blink.

One of two things are going to happen: it's all going to come crashing down in a severely ugly way, or we're going to stop at nothing to prop it up.

The Swiss National Bank is basically a hedge fund that can print its own money. You think the Trump Fed won't buy Boeing stock if it needs to hold up the economy? You think Boris Johnson won't invest in Barclay's if Brexit tears too hard?

Tell me what your models are doing now.