By LARRY SWEDROE
The bear market caused by the coronavirus led the Federal Reserve to rapidly lower the fed funds rate to effectively zero. The flight to quality and liquidity led to yields on short-term Treasury bills to turn negative. For example, on March 25, 2020, the yield on three-month Treasury bills fell to -0.08 percent. That means that investors buying a three-month Treasury bill and holding it until maturity will, with certainty, experience a loss of 0.08 percent. It would also mean that if yields on very short-term Treasuries remain negative, as portfolios of money market funds owning only Treasuries “roll over” (older bills mature and new ones are added), the net asset value will eventually fall below $1 as losses are incurred.
Given that this would be a first, a “breaking of the buck”, I thought it would be interesting to consider how investors will react. Will the negative yields lead to investors increasing risks in their portfolios in order to avoid a certain loss?
The field of behavioural finance provides us with fascinating insights into individual investor behaviour, including how individuals view risk, as well as the impact of those views on asset prices. Prospect theory plays a major role in explaining investor behaviour. The theory, formulated in 1979 by Amos Tversky and Daniel Kahneman, describes how individuals make choices between probabilistic alternatives where risk is involved and the probability of different outcomes is unknown. Prospect theory, which aims to describe the actual behaviour of people, is a challenge to the conventional expected utility theory, which models the decision that perfectly rational agents would make.
The theory starts with the concept of loss aversion, the observation that people react differently to potential gains than to potential losses—they make decisions based on the potential gain or loss relative to their specific situation rather than in absolute terms. Faced with a risky choice leading to gains, individuals are risk-averse, preferring solutions that lead to a lower expected utility but with a higher certainty (concave value function). On the other hand, faced with a risky choice leading to losses, individuals are loss-averse (risk-seeking), preferring solutions that lead to a lower expected utility as long as it has the potential to avoid losses (convex value function). These examples contradict expected utility theory, which only considers choices with the maximum utility. This is also the impetus for the disposition effect, whereby people would rather hold on to a stock and expose themselves to a greater loss than realise the paper loss, accepting a shortfall from their aspiration (reference point) to break-even (referred to as shortfall aversion).
Maren Baars, Henning Cordes and Hannes Mohrschladt, authors of the study How Negative Interest Rates Affect the Risk-Taking of Individual Investors: Experimental Evidence, which was published in the January 2020 issue of Finance Research Letters, help answer the question of whether negative interest rates would lead to investors taking on more risk. They recruited 316 students from an undergraduate business lecture at a German university for an online experiment (participation was voluntary). German interest rates, both short- and longer-term, have been negative for a while.
The main task given the students was to allocate an experimental endowment of 60,000 monetary units between a risk-free and a risky asset. They chose an experimental design where all participants played six independent rounds in randomised order. Three rounds had a 1-year horizon and three had a 10-year horizon. Two rounds (one for each horizon) had a risk-free rate of 3 percent. In those rounds, stock returns were either 26 percent or -9 percent, each with a 50 percent probability. Two rounds had a risk-free rate of 1 percent. In those rounds, stock returns had a 50 percent chance of being either 24 percent or -11 percent. And two rounds had a risk-free rate of -1 percent. In those rounds, stock returns had a 50 percent chance of being either 22 percent return or -13 percent. In all cases, the expected equity risk premium was 5.5 percent and annual volatility was 17.5 percent. The return of the risky asset was determined by repeated coin tosses. The experiment concluded with a brief survey on financial literacy and the attitude toward financial risks and losses.
The authors found that for the 10- and 3-year horizons, the average share invested into the risky asset was virtually identical in the 3 percent and the 1 percent interest rate scenarios. However, when the risk-free rate fell from +1 percent to -1 percent, risk-taking increased significantly, from 50 percent to 61 percent, and the increase was highly significant (t-stat of 6.9). Results for the 10-year investment horizon were remarkably similar. The order in which participants encountered the different interest rate scenarios did not affect risk-taking.
The fact that the risky asset share increased when the interest rate turned negative indicates increasing loss aversion. And finally, participants with weak loss avoidance increased their risk-taking by only 7 percentage points if the interest rate fell from 1 percent to -1 percent, while participants with strong loss avoidance increased their risk-taking by 14 percentage points. The difference was economically and statistically significant (t-stat of 6.8). In addition, the tendency to increase portfolio risk if interest rates turned negative was found for both risk aversion groups, illustrating the robustness of the findings. The effect was of similar magnitude for both groups. The authors concluded that risk-taking increases if the interest rate turns negative—a disproportionally strong value loss occurs if outcomes switch from the gain to the loss domain.
When interest rates turn negative, investors face a choice between a sure loss relative to their reference (break-even) point, by accepting negative interest rates, or exposing themselves to a greater loss by choosing stocks — attempting to avoid that loss. When interest rates turn negative, there is an increased willingness, caused by shortfall aversion, to make the riskier choice of investing in stocks. That choice is, in a way, analogous to the disposition effect.
These findings, while from a small sample in a laboratory test, are in line with the behavioural finance theory of loss probability (shortfall) aversion. Logically, this should not occur. But we are humans, and are not always logical in our behaviour but “psycho”-logical.
My hope is that this serves as a warning that your risk preference should not shift just because rates turn negative. If you make the mistake of taking on more risk because of loss aversion, the result could be that your portfolio assumes more risk than you have the ability, willingness or need to take. Doing so can lead to panicked selling in bear markets as your stomach screams, “GET ME OUT!”
Also of interest is that this behavioural argument of loss aversion implies a lower equity premium demanded by investors when the returns of risk-free investments turn negative on a widespread basis. The result would be a rise in valuations of stocks (price-to-earnings ratios would rise).
LARRY SWEDROE is Chief Research Officer at Buckingham Strategic Wealth and the author of 17 books on investing, including Think, Act, and Invest Like Warren Buffett.
Want to read more of his work? Here are his most recent articles published on TEBI:
Does indexing really get you average returns?
Nine points to consider on the outlook for equities
Analysts’ earnings forecasts and the value premium
Value investors should follow their heads, not their stomachs