Duration

This article takes us into the clinical, mathematical world of bonds. We leave behind the cosy warm furriness of equities and the statistical nonsenses of fund analysis and dive full length into the icy waters of bond valuations. Away with you, foul sentiment. Break out the abaci, the programmable calculators and the spreadsheets. This is the utilitarian world of Bounderby and Gradgrind.

I will make what I think is a fair assumption that many of the features of bonds are familiar territory. Yield, coupon, maturity and maybe even price. All these are intuitive and mostly jargon-free. But what about ‘duration’? The word is readily bandied about and almost always adjectivally appended. A browse through the IMA bond sectors tells us that duration may be short, reduced, low, low average, long, ultra-long and even hedged.

This does not help very much. Halfway through this list of descriptions duration could have been a brand of margarine. An ultra-long or hedged butter-substitute however is not so appetising. We need to backtrack briefly and to grasp what duration is not. And it is not maturity. Maturity (the length of time until a bond is repaid) also appears in fund titles, but with the appendage ‘dated’. This needs a much simpler set of adjectives, the list usually comprising only short, medium and long. Hence the M&G Short Dated Corporate Bond Fund, for example will invest into bonds which typically have less than five years to maturity.

So, do not confuse your short-dated with your short-duration. Oh-ho no. The concept of duration is quite simple, its calculation moderately complicated and multifarious, but above all it is misnamed. Duration refers to the sensitivity of a bond to a change in interest rates. It is the calculation that tells us how much the price of a bond will move in response to a change in interest rates. There may be reasons why it is not so, but it would be more accurately named ‘sensitivity’.

This is genuinely useful information, verging on the compulsory. But it is easy to get lost in the detail. Should you be looking at Macaulay’s or the modified duration of a bond? But what about the effective duration? Surely that must be better than the ineffective duration? This is where it starts to get tricky. The good news is that unless you are taking an exam, it doesn’t matter.

Instead we need to have an understanding of which types of bonds are most and least sensitive to changes in interest rates. If you have a view that European quantitative easing will mean that yields are going fall even more, then you would want to own bonds with the highest, or longest, duration. That is, bonds with the greatest sensitivity to interest rates. And vice versa: if you believe inflation is parked round the corner and is revving up its battle tanks your bond exposure should be in low, or short, duration bonds.

Sparing the mathematical niceties, the rules of thumb are the lower the coupon and/or the longer the maturity of the bond, the greater the duration. Thus the most volatile bond prices tend to be long-dated index-linked gilts. At the other end of the spectrum, the nearer a bond is to maturity and the higher its coupon, the less sensitive its price will be to a small change in interest rates.