Testosterone May Distort Markets And Hurt The Economy

A new study suggests that steroid use by traders may cause financial bubbles

Image source: this video

I’ve been very interested in stock trading ever since I was little kid, and for a while now I’ve been looking for a good excuse to write about it. So imagine my delight when I saw this gem on Twitter a couple weeks ago:

A study connecting the subject of testosterone levels to the subject of stock trading? This topic was practically tailor-made for me.

It turns out there are a few studies on testosterone and asset-trading behavior. Ready to dive in? Let’s go.

Testosterone Makes People Trade More Aggressively

The study in question was published in 2017 by Nadler et al. It used an experimental method– that is, subjects were given testosterone rather than merely having their natural testosterone levels measured. They also traded in a simulated market in which they were the only participants, meaning the market was entirely moved by their actions– obviously not entirely generalizable to the real world.

In the study, the group that got testosterone bought assets more aggressively, pushing prices higher– well above their intrinsic values– and creating market bubbles.

In other words, they lost track of the distinction between wealth as inherent value, and wealth as market value, i.e. things being worth whatever people are willing to pay for them. That second type of wealth is what the market says things are worth, but it’s also somewhat imaginary and a lot more changeable, and over a long enough time tends to revert back to the actual inherent value of an asset. As Warren Buffet says, the market is a voting machine in the short term and a weighing machine in the long term.

Of course, since this was a simulated market in which all traders were on testosterone (the control group was in a separate simulated market), these asset bubbles persisted far longer than they would have in a real-world market where there are other, more rational traders to burst the bubble. In fact, some bubbles didn’t get around to bursting before the study ended. What about the real world?

A 2008 non-experimental study of London asset traders measured testosterone and cortisol levels at 11 AM and 4 PM– that is, before and after the trading day. Now, a man’s testosterone levels can vary significantly from day to day depending on how well he slept the night before, what he ate over the past 24-48 hours, and stress levels. That’s true for women too actually, albeit less-studied and also subject to menstrual cycle variations– but I digress.

High morning testosterone predicted both more aggressive trading behavior and better returns for the day. There was no correlation between one day’s testosterone level and the next day’s returns, which shows that the effects of testosterone levels are short-lived and can change quickly.

Cortisol– the body’s main stress hormone– was correlated to market volatility, though not to overall returns. The authors speculate that sustained losses would increase cortisol (almost certainly true), though the study didn’t prove that.

So high testosterone produced better results then, so testosterone is good right? Not so fast. I know I said the last study had limitations due to being a simulation, but real-world studies have a different problem– whether trading aggressively is good or bad is heavily dependent on what you’re trading and when.

This study was published in 2008, but it was approved for publication in November 2007. It doesn’t say exactly when it took place– and given the nature of the study it really should– but it had to be sometime earlier in 2007, and the authors do give the following hint:

The U.S. calendar, in particular, is closely watched, and most foreign markets pay more attention to U.S. economic numbers than to their own. For that reason, we conducted the study during a period that led up to and included the most important U.S. economic releases, foremost of which were the Institute of Supply Management Manufacturing Index and the Employment Report.

This would seem to suggest that it the study took place between the end of May and first week of June 2007, although I can’t rule out that it could have been a whole year earlier. In any case, an eight-day duration means the study did not take place under a variety of market conditions. Also, the subjects were mostly trading German assets, particularly German interest rate futures. While it may have taken place after the earliest rumblings of the financial crisis, it definitely was before central banks started rapidly cutting interest rates in 2008.

My point is, if that same study had followed, say, stock traders from late 2007 to early 2009, it would have find high testosterone to be correlated with more aggressive trading and worse outcomes– unless they were short-selling, I suppose. I think it’s best to focus on what behavior testosterone causes rather than what results it produces, unless you can test its effects over a very long period of time.

The other limitation is that subjects were day-trading, and the study was measuring day-to-day results. This makes sense because it’s easier to study, but long-term results are what we really care about.

To their credit, the authors do seem to understand this. In the discussion section they note:

Cortisol is likely, therefore, to rise in a market crash and, by increasing risk aversion, to exaggerate the market's downward movement. Testosterone, on the other hand, is likely to rise in a bubble and, by increasing risk taking, to exaggerate the market's upward movement. These steroid feedback loops may help explain why people caught up in bubbles and crashes often find it difficult to make rational choices.

Finally, a 2015 study by Cueva et al administered either testosterone or cortisol to participants, and found that both hormones increased the appetite for riskier investments. So it seems clear that testosterone increases one’s appetite for risk; the effects of cortisol seem less clear.

At this point we should take a step back from asset-trading studies and take a broader look at the influence of testosterone on personality and behavior.

Testosterone Makes People More Impulsive, Risky, and Better at math, but Also Sometimes Worse at Math

Contrary to popular opinion, the link between testosterone and physical aggression is pretty weak. Studies generally find a weak correlation between testosterone and aggression, but experiments typically fail to find that giving men testosterone makes them more aggressive– in other words, weak correlation, little or no causation.

Men who take testosterone frequently report feeling more confidence, but studies don’t always find that. One study found that men with lower pre-natal testosterone actually were more likely to have unrealistically high expectations about their own performance, while another found that testosterone administration did not improve women’s mood or self-esteem.

On the other hand, studies very consistently find that testosterone increases impulsivity– in rats, and in men making financial decisions, and in men making sexual decisions.

An experimental study in women elucidates some of the results of those stock-trading studies from earlier– testosterone seems to increase sensitivity to reward and simultaneously decrease sensitivity to loss, leading to an increase in risk-taking.

So how about the stereotype that men are better at math than women? Studies on this have been very mixed.

One study suggests that testosterone is related to better math scores but worse verbal/literacy scores on standardized tests. Other studies have found no correlation, or even occasionally that low testosterone was associated with higher math scores.

One study even find that high testosterone was associated with better mathematical and spatial ability in men, but low testosterone was associated with better mathematical and spatial ability in women. A growing number of researchers now believe there is an inverse U-shaped relationship between testosterone and mathematical ability, in which the best math scores are associated with testosterone levels that are moderately high, but not too high.

A 2017 study provides a clue as to why this is: testosterone makes you more prone to fall for trick questions. Subjects in this study were asked trick math questions, like the following:

A bat and a ball cost $1.10 in total. The bat costs $1 more than the ball.
How much does the ball cost?

The correct answer is 5 cents, with the bat being $1.05, but your mind wants to jump to 10 cents at first.

Another example question is the one that goes something like, the number of lily pads in a pond doubles every day, and it takes 48 days for the lily pads to cover the whole pond– how many days does it take to cover half the pond? Your mind might jump to 24 days, but the correct answer is 47 days. More details are provided in this article.

What seems to be going on here is that testosterone has two separate effects on math performance: it increases mathematical intelligence while also making you more impulsive and thus less likely to check your work.

On the whole, I think this makes testosterone a net negative for asset trading. Trading involves math, but the most mathematically-complex strategies promoted by fancy Wall Street types and bullshit– the most effective strategy is the buy-and-hold, invest for intrinsic value strategy exemplified by Warren Buffet and friends, and while it involves some math, the math isn’t too hard, and you can get it right by just taking your time at it. Controlling your impulses is far more important for investing performance than being great at math.

At this point I should note that this whole time, I’ve been talking about testosterone. However, there are other naturally-occurring sex hormones, including estrogen, whose effects on stock trading don’t seem to have been studied.

There are also dozens of other illegal anabolic steroids you can take if you’re so inclined, some of which have much more profound psychological effects than testosterone.

So far we’ve been thinking in terms of asset traders having high-normal vs low-normal testosterone, but that begs another question: how many stock brokers, hedge fund managers and other finance pros are roided up?

Probably a Significant Minority, It Turns Out

We’ll start with the obvious: most steroid users are men. There are a few anabolic steroids that are relatively safe for women, at low doses– not that doing illegal drugs you got off the dark web is ever really safe. But yeah, mostly men, so let’s start by looking at the gender ratio in this industry.

According to one estimate, 73% of finance professionals are men, with an average age of 44. Of course this is highly dependent on what you consider a finance professional– I’ve seen other thats that say the finance industry is around 54% male, but that includes the HR and marketing people and so on, who don’t make trading decisions. So let’s look at specific jobs.

76% of investment bankers are men.

78% of fund managers are men.

85% of stock brokers are men.

So, it’s around 75-85% male for the jobs we’re concerned with, the people who make the big trading decisions. And remember, men are more impulsive and risk-tolerant than women even before we get into steroids.

So, steroids– who uses them? Mostly not athletes and bodybuilders, actually. Most steroid users are just regular dudes who want to look jacked and sexy. From the Results section of that article:

The majority of respondents did not initiate AAS use during adolescence and their NMAAS use was not motivated by athletics. The typical user was a Caucasian, highly-educated, gainfully employed professional approximately 30 years of age, who was earning an above-average income, was not active in organized sports, and whose use was motivated by increases in skeletal muscle mass, strength, and physical attractiveness. These findings question commonly held views of the typical NMAAS user and the associated underlying motivations.

Upper-middle-class professionals in their 30’s, mostly white and male…sounds a lot like the finance guys we’re talking about. Okay, but how many people is this?

According to one estimate, around 2.9-4 million Americans aged 13-50 have used steroids. That’s the general population of the USA– other studies have looked only at people who lift. A Brazilian study of regular gym attendees found that 16.9% of men and 6.5% of women use, have used or plan to use steroids. That sounds high, but steroids are probably more readily available in Brazil from what I’ve been able to find. In another study, 30% of Russian male gym attendees said they use steroids, so this definitely varies by country.

What about the US? Surveys get all kinds of results here depending on methodology. One found that 15-30% of American men who regularly lift weights use steroids, but this survey was posted to a message board that caters mostly to steroid users– obviously a terrible sampling methodology. Other surveys cited in that article find numbers like 2.7-2.9% of all Americans, 9.1% of British men and 2.3% of British women, and so on.

Nobody knows really, but the answer seems to be some small but significant fraction of all men, and a sizable minority of men who are into weightlifting, with female numbers probably being about a fifth to a quarter as much. For traders specifically, you can roughly double those estimates since they tend to be right in the demographic that most uses steroids.

It’s also important to note that aside from illegal steroid use, a lot of men are legally taking testosterone through often-sketchy hormone therapy or “anti-aging” clinics. It’s not hard to get if you want it; you’ll see ads for it all over the place if you spend time reading up on testosterone, and clinics basically promise to prescribe it to you. Much like the recent rise of “ADHD'“ companies that promise to give you speed, they’re often just glorified drug dealers.

Okay, So What Can We Take Away From All This?

A few things. First off, don’t do steroids, or at least don’t trade stocks while you do. “Don’t pin and trade” might be the new “don’t drink and drive.”

Second, and it pains me to say this, be suspicious of financial advisers and stock brokers who look too jacked.

Third, we should probably be testing fund managers, investment bankers, brokers and other finance pros for steroids. And while we’re at it, also stimulants and other drugs that affect the dopamine system– see this article by Scott Alexander on how Sam Bankman-Fried was probably on drugs during his tenure at FTX.

Fourth, develop some safety habits around investing. Familiarize yourself with the warning signs of investing scams, always double-check your work, and perhaps make decisions more slowly– if you decide to make a big trade, sleep on it and see if it feels like a good idea the next day.

Fifth, now that we know that testosterone tends to contribute to risky investing and asset bubbles, we should try to limit that in a more systematic manner. Steroid testing would help of course, but I wouldn’t advocate deliberately reducing fund manager’s natural testosterone levels just for this reason– who’d want that?

The image of a bunch of roided-up fund managers tanking the economy is kind of funny in a dark sort of way, but as tends to happen in these sorts of articles, I’ve meandered through a lot of weird shit only to end up at a very prosaic conclusion: there should be more women in the finance industry, and specifically in the roles that make big trading decisions like fund managers, investment bakers, and so forth. Not for the usual equity-based reasons that people give for hiring more women (although those matter too) but because it seems that in this particular field, a more equal gender balance would probably lead to better– less risky, more rational– decision-making.