The behavioural finance of bonds

  • Written by 
  • 19/07/2016

Behavioural biases have an impact on investments. Most studies show the link between behaviour and equity investing. But there are several areas within fixed income where behavioural biases matter, including the reactions to negative rates, the search for yield and the assessment of default risk.

Opportunities in Credit Long/Short 7 of 8

Residents of the United States, please read this important information before proceeding

Over the years we have had a number of articles in Compass describing how investment markets and investor behaviour often create conditions for imperfect decisions. Most of the articles have been about equity markets. The reason for the focus on equities is that this is on one of the best examples of a near perfect market. And finding biases in this market means that you can start to refute traditional theories on investing. But less perfect markets, like the bond market, provide ample opportunity for behavioural inefficiencies.

Imperfect markets

With the exception of some government bonds, there is an absence of a central exchange matching potential buyers and sellers with transparent pricing. Most bonds trade ‘over the counter’ meaning that a broker will quote a price of their choosing and the buyer will either accept or decline. One reason that formal exchanges do not exist is the heterogeneity of the asset class. Bonds differ in the credit rating of the issuer, the coupon they pay to entice people to lend them capital, and the length of time to maturity. Every bond issue is unique making the asset class relatively more complex. Added to this, many bonds are bought to be held to maturity rather than traded. It can be many days or weeks between the same bond issue changing hands and allowing the ‘market’ to discover the latest price.

These are hardly perfect market conditions. A classical economist might argue that bond markets are so imperfect that behavioural biases have to exist. Perversely, because the market can easily feature behavioural inefficiencies, there are only a handful of published articles in this area.

Every bond issue is unique making the asset class relatively more complex

Negative rates and the search for yield

We would recommend that private investors access fixed income asset classes through funds to avoid concentration risks. However, this doesn’t remove the potential for investment themes to lead to biased investment decisions. We’ll look at two themes of fixed income investing that are important in a low interest rate environment – negative interest rates and the search for some yield.

We believe that our individual behaviour places a limit on the effectiveness of low interest rates to stimulate the economy. Negative interest rates provide a negative incentive to hold cash, as the money you hold will be worth less in the future than it is today. This is called the ‘individual time preference about money’. But even though it would be rational to spend that money today before its value is withered further, the reality is that most of us need to continue holding our savings because we prefer a relatively smoothed level of consumption from one year to the next rather than spikes in individual years. With the exception of doomsday preppers storing dry goods in their basements, we don’t typically stockpile years of our consumption needs to maximise the spending power of our wealth. This leads us to the second area of potential behavioural bias – the search for yield. To restore our time preferences for more money in the future being preferred to money today, investors need to search for higher returns. Investment theory suggests that the only way that you can increase your yield from fixed income investing is to accept more risk, either in the form of greater interest rate risk or by looking for bond issuers of a lower credit quality – or ‘greater probability of default.’

Confirmation bias

Because bond prices are inversely related to interest rates, if you think that interest rates will rise, you also expect that bond prices will fall. With the benchmark interest rates in most major economies near zero it is reasonable to expect that interest rates will rise at some point. The questions for individuals to answer are how quickly will any rate rise come, and how much will rates rise? This is the area that our expectations can be biased. In forming our expectations we will be heavily biased by our starting views of the economy and may exhibit confirmation bias by not searching out enough contrary views. The likely result is that we exhibit some over-confidence in our views and place too little weight on the possibility that we could be wrong – leading us to underestimate the risk and potentially get stung if interest rates move further and faster than we expected.

Estimating credit default risk is no less fraught. The difficulty is whether investors can even accurately assess default risk. We are unaware of any evidence that proves or disproves that investors are able to do this with any accuracy. Credit ratings provide a guide but these aren’t a numerical estimate of the probability. They are a classification that allows for the ordering of fixed income securities according to risk. Individuals normally perceive rare events as being more likely than they really are. That would suggest that if people can assess small probabilities of default, they are likely to perceive these bonds to be riskier than they really are, and consequently demand too high of a return as compensation.

We think that investing in fixed income asset classes can be biased by our individual behaviours in the same way that investing in equity markets can be. Individual investors would be well served to consider how professional investment management can help navigate this most heterogeneous of asset classes where the lack of central exchanges make the true value of an investment difficult to discover.