Suppose we want to compare two very different bonds, where the spread durations are one year and 10 years, and the credit spreads are 500 bps and 50 bps, for bond A and bond B, respectively. Both have a DTS of 500 and therefore will have the same expected credit volatility.

DTS only predicts the credit risk of the bonds, not their default risk. In our example, investors clearly perceive bond A to be riskier, as its credit spread is the highest of the two bonds. To predict default risk, one could use credit ratings or distress risk measures, such as distance-to-default. Moreover, DTS does not say anything about interest rate risk, because it only predicts risk driven by credit spread fluctuations. To measure a bond’s exposure to changes in risk-free interest rates, an investor can simply use the interest rate duration.

The DTS concept has various advantages. First and foremost, DTS is a more accurate predictor of future volatility than methods previously used by investors. Second, DTS is very simple to calculate, which makes it easy for portfolio managers to use in their daily work. Third, changing market circumstances are reflected in the DTS: if credit spreads double from one day to the next, so does the risk estimate.

Why should one multiply duration and spread? Why not use another formula? One can easily prove that by using DTS as the risk measure, we assume that credit spreads move in a relative fashion rather than a parallel fashion. In our example, if the credit spread of bond A moves from 500 to 550 bps (i.e. a 10% increase), then the credit spread of bond B will move from 50 to 55 bps (also 10% increase), and not from 50 to 100 bps. Our empirical research shows that relative spread changes indeed reflect credit markets more accurately than parallel spread changes.

DTS was originally developed by Robeco researchers in 2003. Shortly thereafter, we started using it to monitor the credit risk of all of Robeco’s credit portfolios. A joint project with Lehman Brothers led to the publication of the results in The Journal of Portfolio Management in 2007.¹ DTS is now widely accepted among investors, and it has been implemented in leading risk management software, including MSCI RiskMetrics and Bloomberg Barclays POINT.

Footnote

¹Ben Dor, A., Dynkin, L., Hyman, J., Houweling, P., Van Leeuwen, E., and Penninga, O., ‘DTSSM (Duration Times Spread)’, The Journal of Portfolio Management, 2007, vol. 33. no. 2, pp. 77-100