The Interesting Times: The Midwit Trap

It’s a funny and effective meme because it’s often true.

I often cite the Oliver Wendell Holmes quote which sort of gets at the same thing:

"For the simplicity on this side of complexity, I wouldn't give you a fig. But for the simplicity on the other side of complexity, for that I would give you anything I have."

I’m always interested that if you meet people who are really good at their job, they are often using simple heuristics.

You would be amazed at how much traders managing huge amounts of money don't use some complex optimization algorithm but rather rely on rules of thumb like "Sell half of the position when it doubles and let the rest ride."

An observation I had not considered that this post made is that this meme really is designed for what Nassim Taleb called Extermistan, environments where outcomes are highly skewed (like wealth) as opposed to Mediocristan where outcomes tend to be more normally distributed (like height)

The Midwit Trap is to some extent just an application of Extremistan. The effectiveness of a simple solution depends on having a problem where one cause is responsible for most or all of the negative consequences. This is likely to be the case where there is a power law distribution or strict dependencies, but not so much elsewhere. We are used to solving problems in Mediocristan, where we usually cannot achieve a significant outcome with "one simple trick!" and so we are dismissive of simple solutions or simple explanations.

As Gerd Gigerenzer has argued persuasively, simple heuristics are often more robust "out of sample" than complex algorithms.

Wolves in Yellowstone National Park [Twitter Thread]

Dr. Clayton Forrester

As long-time readers are aware, I love a good story of a complex system with interdependent feedback loops. This is a great thread on how wolf populations dramatically impacted Yellowstone National Park in the Northern U.S.

All hail the butterfly effect.

Hari Krishnan - Market Tremors & Tail Hedging (S5E11) [Podcast]

Flirting with Models

On the topic of complex systems with interdependent feedback loops, let us turn to Exhibit A, financial markets.

This interview was with Hari Krishnan, a trader with a PhD in complex, dynamic systems about some of the models he has developed for thinking about market behavior and how they differ from more standard models.

Consider this argument that it may be theoretically impossible for the Fed to even control inflation at all.

A lot of people like to attribute malice or some crazy political theories to the Fed behavior, but the argument that they’ve just modeled the feedback loops wrong is a much more interesting critique in my opinion.

The Federal Reserve has a dual mandate: stable inflation of about 2% and low unemployment.

Right now inflation is running at about 8.5% (high) and unemployment is running pretty low so the Federal Reserve is hiking interest rates with the hope that this will slow inflation without causing too much unemployment.

However, this can likely only go so far.

If they decide to raise rates … they're going to hit growth. And given that we're not growing that quickly, if they hike too much they'll hit their loss function and they'll have to stop hiking.

This sort of balancing between unemployment and inflation kind of makes sense, it's like trying to get a shower to right temperature. If it's too hot, you turn the knob a little bit towards cold and vice versa.

But, what if the feedback loops for unemployment and inflation are very different?

Now the problem, as I see it, is the difference in delays. If the Fed raises interest rates by a point, it takes longer for that, on average, to affect inflation, especially sticky inflation than it does to hit growth.

So the inflation dampening that they're hoping for may never occur. It won't be observable before we hit the skids.

If raising interest rates causes unemployment to go up fairly quickly, but inflation doesn’t decline as quickly as unemployment goes up then the Fed is in a real pickle - they have high unemployment AND high inflation.

This would be like if you were trying to get the shower to the right temperature but it took 5 seconds for the cold water to ramp up but 15 seconds for the hot water. You'd keep making it too hot because the feedback loop was too long.

As Hari also points out, this is made even more difficult by the fact that these feedback loops are actually dynamic. Maybe in one environment, it takes 12 months for rate hikes to hit inflation and 6 months to hit growth.

In another environment, maybe it takes 24 months to growth and another 36 months to hit inflation.

You are now trying to get the shower to the right temperature even as the lag on turning the hot and cold knob is constantly adjusting. This would be very hard!

Here are a few more things you might find interesting:

Newsletter Past Editions: Read past editions of The interesting Times Newsletter.

Interesting Essays: Read my best, free essays on topics like bitcoin, investing, decision making and marketing.

Consulting & Advising: Are you looking for help with making decisions around scaling your company from $500k to $5 million? I’ve been working with authors, entrepreneurs, and startups for half a decade to help them get more out of their businesses.

Internet Business Toolkit: An exhaustive list of all the online tools I use to be more productive.

Futures and options trading involves a substantial risk of loss. You should therefore carefully consider whether such trading is appropriate for you in light of your financial condition. Unless distinctly noted otherwise, the data and graphs included herein are intended to be mere examples and exhibits of the topic discussed, are for educational and illustrative purposes only, and do not represent trading in actual accounts. Opinions expressed are that of the author. The mention of specific asset class performance (i.e. S&P +3.2%, -4.6%) is based on the noted source index (i.e. S&P 500 Index, etc.), and investors should take care to understand that any index performance is for the constituents of that index only, and does not represent the entire universe of possible investments within that asset class. And further, that there can be limitations and biases to indices such as survivorship, self-reporting, and instant history.

DISCLAIMER

This email provides information regarding the following commodity pools: The Long Volatility Fund LLC and The Cockroach Fund, LLC (collectively the "US Funds") and Mutiny Funds Cayman Ltd. (together with the US Funds, collectively the “Fund(s)“), which are managed and operated by Attain Portfolio Advisors LLC and Mutiny Funds LLC (the “Managers”). Investments in the US Funds are only available to Accredited Investors as defined in Rule 501 of Regulation D of The Securities Act of 1933. This content is being provided for information and discussion purposes only and should not be seen as a solicitation for said Fund(s). Any information relating to the Fund(s) is qualified in its entirety by the information included in the Fund’(s)’ offering documents and supplements (collectively, the “Memorandum(s)”) described herein. Any offer or solicitation of the Fund(s) may be made only by delivery of the Memorandum(s). Before making any investment in the Fund(s), you should thoroughly review the Memorandum(s) with your professional advisor(s) to determine whether an investment in the Fund(s) are suitable for you in light of your investment objectives and financial situation. The Memorandum(s) contain important information concerning risk factors, including a more comprehensive description of the risks and other material aspects of an investment in the Fund(s), and should be read carefully before any decision to invest is made. This site is not intended for European investors, and nothing herein should be taken as a solicitation of such investors.