Study: Men Given Testosterone Generate Price Bubbles, Overvalue Assets More Than Those With Placebo

By:
Chris Gaetano
Published Date:
Oct 12, 2017
Pumping Iron At Desk

A recent study that gave some men testosterone and others a placebo found that, when buying and selling financial assets to earn real money, those who got the hormone were more likely to generate price bubbles and overvalue assets than those who got the placebo.

"Much like sentiment, which causes entire asset classes to move in the same direction, endogenous changes in testosterone can cause synergistic co-movements in financial markets: traders winning in bull markets likely experience an increase in the endogenous production of testosterone. The resulting increase of testosterone can fuel overpricing and bubble formation as shown in this experiment," said the paper's conclusion. 

The study, which was published in the journal Management Science, involved 17 sessions of continuous double-auction markets, each consisting of three rounds of 12 trading periods, which each lasted 1.5 minutes. During this time, traders bought, sold, bid and asked for shares of a financial asset. Of those taking part in the study, 84 were given a testosterone gel and 56 were given placebos. Each session took place over two days to allow testosterone levels to increase and stabilize following exogenous application on the first day. The researchers controlled for factors that would affect testosterone levels by doing things like using only male research assistants and imposing a uniform time schedule. 

Aggregating results across all three rounds, the researchers found that there was a 114 percent larger Mann-Whitney test amplitude (the primary measure of bubble size) between those who got the testosterone and those who got the placebo. The researchers also found that average testosterone levels positively correlated with amplitude (the maximum normalized difference between average prices and fundamental value during a trading period) and market value amplitude (the volume-weighted average price above the asset’s fundamental value). Turnover, the number of trades divided by the number of shares in the market, was not affected. 

Testosterone levels were also seen to have affected bids and asks: there were significantly higher average bidding prices among traders who'd received testosterone versus those who did not. Similarly, ask prices were also higher among testosterone-treated traders versus the placebo traders. Buying and selling turnover (the number of bids and asks divided by the total number of shares) were also significantly higher in the testosterone group than the placebo group, as were spreads. 

The researchers also said that testosterone also affected overall trading preferences. Traders were assigned a positive point for every bid below the asset's fundamental value and a negative point for every bid above the asset's fundamental value. Those with positive points were categorized as "fundamental bidders" and those with negative points were categorized as "bubble bidders." It was found that testosterone-treated traders were primarily bubble bidders while placebo bidders were primarily fundamental bidders. Being in a market with primarily testosterone traders was found to increase the odds of bidding above the asset's fundamental value by a factor of 1.65. 

In terms of trading performance, the researchers found a positive correlation (with the caveat that it is uncorrected for multiple hypotheses) between testosterone-treated traders and percentile earnings, with no significant correlation among placebo traders. This was based on later analysis of trading activity. During the sessions, testosterone-treated traders thought that other traders were buying too low while the placebo-treated traders thought that they were buying too high. On the other side of things, traders in the placebo group thought others were selling too high and buying too high relative to those in the testosterone group. 

Click here to see more of the latest news from the NYSSCPA.