It’s a long story, but I have a deeply ingrained aversion to what I call “What could go wrong?” charts.
You know the ones I mean. They’re the charts you see on Twitter (sorry, “X”) that get all the retweets. And also all the “likes” from the pitiable community of would-be macro mavens duped into believing that everybody with “ex-Goldman” in their social media bio actually worked at Goldman and gullible enough to pay thousands of dollars a year to every Tom, Dick and Harry peddling a “research” service (or gullible enough to “pay” millions of clicks per year to every doom profiteer peddling market-themed anarchist propaganda from the comfort of their suburban living room).
“What could go wrong?” charts take many forms, but they have a few things in common. For example, in almost all cases, they plot apples and oranges using two y-axes where the scales are adjusted such that it looks like two unrelated series are correlated. You can build an enormous following by spending your days creating those kinds of charts (or get a job making those kind of charts for someone who already has an enormous following). If you’re the enterprising type, you can build an actual business based on them because — well, because people are gullible.
If you can’t distinguish those sorts of “chart crimes” from visuals that have real informational value, then God help you. There’s no telling how much time you’ve wasted pondering statistical relationships that don’t actually exist.
But not all “What could go wrong?” charts constitute chart crimes, and occasionally you can compare apples to oranges as long as there’s some underlying logic. With that in mind, have a look at the visual below.
The Fed’s senior loan officer opinion survey produces a lot of different series, but that one (i.e., the share tightening C&I standards to middle-market and large firms) is typically considered the headline print. The disparity with junk spreads is pretty remarkable.
Outside of the brief spike around the onset of the pandemic, banks haven’t been this cautious about commercial and industrial loans since the financial crisis. And yet, the premium junk issuers are paying to borrow is negligible considering the inherent risk of high yield debt, and considering other financing options for non-IG firms are drying up.
Relatedly, junk spreads seem to belie the message from the S&P’s outperformance versus small-caps.
That’s another “What could go wrong?” chart, and like the first visual shown above, I think it has some merit — I don’t think it’s necessarily a “chart crime.” If it is, it’s a misdemeanor.
Both of those visuals were highlighted in a Wednesday note from SocGen’s equities strategists. “One should not ignore the inverted yield curve as a recession signal, as it is the inverted yield curve that drives the tightening in lending standards,” the bank said. “This is likely the reason why, despite high-yield spreads coming down, small-caps have continued to underperform large-caps.”
Obviously, all of this ties into the “haves versus have-nots” discussion as it relates to corporates, tighter financing conditions and Fed hikes. “Refinancing risk remains significantly higher for Russell 2000 companies, which are also struggling to improve profit margins,” SocGen went on, warning that even if the outlook for the US economy is relatively good, investors “need to navigate the quality and refinancing cycle.”
Is any of the above actually foreboding? Not really, to be honest. I’ve been doing this for a long (long) time, and I can’t remember a single instance where I’ve looked up six months after an economic or market calamity and thought to myself, “If only I’d heeded that chart.” That doesn’t mean the visuals above aren’t worth a look — I wouldn’t have published them if that were the case. It’s just to say that “What could go wrong?” charts are a dime a dozen, and unlike $1,500/year subscriptions to amateur “research,” you get what you pay for.
Of course, there are charts that do foretell the next crisis, and someone could, in theory, create them. The problem is that we’re all (every, single one of us) inept when it comes to preemptively determining where the next crisis will come from. Even if we can make that determination, we’re not very good at predicting when the next crisis will occur. Usually, we’re inept at both.
So, if your question is whether the definitive “What could go wrong?” chart is floating around out there somewhere, just waiting to be discovered, the answer is, “Probably not.” If that chart does exist, the person who made it either doesn’t realize how profound it is yet, or if they do realize it, and bet the house, that person is probably in the process of learning that being early is the same as being wrong.




As a lay person with a long time interest in mathematics, I was surprised to see Nassim Taleb pop up in Mathematicians of the Day” for September 12 on the St.Andrews University Maths History web site (https://mathshistory.st-andrews.ac.uk/Biographies/Taleb/). H’s readers might find this brief biography interesting, especially the last paragraph quoted here:
“I no longer care about the financial system. I gave them my roadmap. OK? Thanks, bye. I’ve no idea what’s going on. I’m disconnected. I’m totally disengaged. People read 3 million copies of ‘The Black Swan’. The bulk of them before the crisis. And people love it. They agree with it. They invite me to dinner. And they don’t do anything about it. You have to pull back and let the system destroy itself, and then come back. That’s Seneca’s recommendation. He’s the one who says that the sage should let the republic destroy itself. [It used to frustrate me.] Now it doesn’t.”
My feelings exactly.
Generally, I pay little attention to charts of any kind. While they are often very visual but you can’t actually Do anything with them. There are no numbers on them and trying to determine any thing you can put in an equation for analysis is like trying to read an old-fashioned cheap slide rule. I had a very good one and in my youth I could read three decimal points on it. Charts, especially those flipped upside down with different scales on each side, solely designed to appear shocking in order to sell someone’s point of view point of view are typically garbage. Give me the table, the actual input data that made the chart. I’ll do my own analysis. One of the types I really hate is the type that’s invisible, as in: If you drink oat milk you’ll decrease your risk of toenail fungus by 38%, or some such. The real data is, those who don’t drink the product get the fungus at a rate of 13/100,000 pop. Those who do drink it get the fungus at a rate of 8/100,000 pop. Whoopee. Maybe even true. The chart would be huge but show a meaningless change. Drug companies use this canard all the time.