A Robust Illusion

Precise estimates of financial variables, making money by betting on sports, a Fed that can control inflation to a narrow target, and other articles of faith

Photo by Jacob Rice on Unsplash

Hello,

Welcome to Known Unknowns, a newsletter about the statistics we shouldn’t believe and the games we shouldn’t play.

Inflation is sticky after all

So, it turns out the last mile is the hardest. Inflation is looking pretty stubborn, still around 3%—and considering that goods deflation has run its course and that wage and service inflation is still high, paired with an otherwise booming economy, I don’t see how it goes down. The tragic bridge collapse in Baltimore may not cause too much economic damage, but it certainly won’t help matters either, as it will probably increase shipping costs and car prices.

I discussed on Squawk Box what the Fed should do. Is it worth damaging an otherwise strong economy to get inflation from 3% to 2%? There is nothing wrong with 3%, other than the fact the Fed’s target is 2%. I know, credibility is important. But let’s be honest, when have they ever hit their target? Their credibility is about faith more than history. They were below target pre-pandemic and now just above. Maybe a wide range is a better idea, or we just admit the Fed does not have precise control over these things. Though I also worry that giving up now could destabilize expectations.

Sometimes we all need to believe something, even if it is not true. It makes for a stable(ish) equilibrium.

Faith in data

Speaking of misplaced—but useful—faith in things that don’t exist, let’s discuss the robustness of financial variable estimates. A few weeks ago, Bloomberg wrote about a paper that noticed that estimates of investment factors, like small cap and value, are very sensitive to data updates. Well, yeah. Such is financial data. They are also sensitive to the time frame or how you define your factors.

I wrote for Bloomberg on this topic, sharing what I learned from Fama and French when I was at DFA. The estimates suggest that factor investing has some value, no matter what data you use, to me suggests that it is robust. We just can’t put a precise number on that value, but that is true of pretty much every financial variable we estimate.

And worse, this is all based on historical data, and that tells us little about the future. Maybe in a world where the most valuable kind of capital is intangible and scale is super important to success, small-cap and value factors won’t hold up in the future. All we can do is strive for transparency and simplicity, but after that, we are just nailing Jello to a wall when we try to say how anything will perform in the future.

And this will be true as we enter the big data world and try to put odds on every decision we face. And these problems will still arise even if AI is better at estimating data and making assumptions than we are. Data is, well, noisy and arbitrary, but defendable. It’s the best wet got, but never a pure truth.

We also need to believe financial estimates are more accurate than they are, it makes for liquid markets (most of the time). A data-powered AI world will end up being a world that runs of faith too.  

More Madness

I am not a big sports fan, but as a third-generation Storrs, Connecticut native, I have a soft spot for the NCAA tournament. I think what makes it special is the drama, intensity, and shocking upsets. Also, while pro teams always come from big cities, college sports often represent small towns, like Storrs. Sadly, I’ve never participated in an office bracket, originally because I had no office and later because I was not in the habit. But I support the concept, as low-stakes gambling, even with a negative expected value, can be a healthy way to add some excitement to the tournament, and it brings people together.

But times have changed. No one (except me, who is still living in 1990) would call UCONN a Cinderella team, and the gambling has gotten out of control. Sports betting apps have taken off in a big way since 2018 and are now in 30 states. And problem gambling is up, especially among young people. I can’t believe universities even promote gambling to their students. I am too libertarianish to say gambling should be illegal—but should it be so easy? In-game betting apps seem like a bad idea. And it’s remarkable we restrict or discourage risk in every other walk of life—like moving states or starting a business. But we are actively encouraging sports gambling—which is also zero-sum, and the house usually wins.

I am not naive enough to believe more productive forms of risk-taking will scratch the sports betting itch entirely. But maybe we should make buying an S&P 500 index fund easier than betting on the game.

Rent is too damn high

I made this chart for City Journal and can’t stop looking at it.

 Even accounting for inflation, it is extraordinary how much the rent of a studio apartment has increased since the pandemic. Especially when you compare it to the average income level divided by 40 (or the income you need to comfortably rent that space). Even a Bronx studio is unaffordable.

New York is a bit of an outlier, but in the last few years many cities have become unaffordable for a single person living alone. The question is why. I blame constrained supply and inelastic demand. There are many restrictions in cities like New York on building new housing, and rent controls limit the size of market-rate apartments.

And then everyone wants to live here, especially young ambitious people looking to acquire career and social capital. Even if they leave the city one day, they are making an investment (or in some cases a gamble) in their future by living here today. I suppose that makes paying a fortune for 300 square feet worth it.

In other news

I spoke about the introvert economy and how to manage career risk on one of my favorite podcasts, Cardiff Garcia’s The New Bazaar.

Until next time, Pension Geeks!

Allison