Predicting risk in investments
The global fixed income markets process huge amounts that far exceed those processed on public equity markets. The interbank market consists of interest rate derivatives, which are financial instruments whose primary purpose is to manage interest rate risk. The credit market primarily consists of the bond market, which links investors to companies, institutions, and governments with borrowing needs.
In his thesis Optimization-Based Models for Measuring and Hedging Risk in Fixed Income Markets Johan Hagenbjörk uses a method known as stochastic modelling to manage the risks in these two markets. “Stochastic” describes mathematical methods used to model and analyse a phenomenon that involves randomness, an example of which is fluctuation in interest rates.
Johan Hagenbjörk in the corridor outside his office.
“Many people believe that we are trying to predict the future, but that’s not what it’s about”, says Johan Hagenbjörk.
“But we do try to predict risk. If we can generate, for example, a thousand different scenarios that are all equally probable, we can also determine the risk of a financial portfolio more accurately. In that case, we have predicted everything and not missed anything, and we are satisfied with that.”
Twenty years of comparisons
The thesis defines six systematic risk factors in the interbank market that can explain nearly all variation in interest rates. By modelling the dynamism in the risk factors, possible future trajectories of interest rates can be simulated.
Johan Hagenbjörk has used data from the past 20 years and examined their relationship to the values that the optimisation models calculate. It has become clear that the models can accurately capture the historical development of interest rates without overestimating or underestimating the risk.
The accuracy is better than that achieved by the models normally used by banks and other financial institutions today. This means that risks are lower and that the prices can be set more exactly.
“It’s considerably better than traditional models”, says Johan Hagenbjörk, who is, however, reluctant to assess the probability that the method will be used in practice.
“It may be a disadvantage that the model is rather complex, and it must be adapted for each region. Most people today use simpler models, which, even though they are less accurate, can be used in several regions.”
The Riksbank’s limited influence
The thesis identifies a further four risk factors in the credit market that explain essentially all differences in the number of defaults (bankruptcies). In order to calculate the possibility of recovery (that the creditors will get back their invested capital), it is sufficient to use a single risk factor. The difficulty of separating the risk of default from the possibility of recovery is a well-known problem.
Economic mathematics is a complex field, which is often difficult for non-specialists to understand. To illustrate this, Johan Hagenbjörk usually takes the example of mortgage rates, which many people believe to be fully under the control of national banks.
“The Riksbank can significantly affect short-term interest rates, but not the long-term rates. In this case, the credit risk is an additional factor that the Riksbank has very little influence over.
"Consider the Euro crisis in 2011, for example. At the time, Germany and Greece had the same risk-free interest rate curves, but completely different levels of credit risk. This led to much higher interest rates in Greece, 30%, than in Germany, 2%.”
Translated by George Farrants
