With government bond yields at historic lows in the UK and US, many managers are taking short duration bonds to manage risk and enhance performance in the belief that yields have nowhere to go but up.
Understanding duration
But how does duration affect the risk and performance of a bond portfolio? It is worth at first defining and quantifying what is meant by the term and how it is applied.
Duration is a measurement of how long, in years, it takes for the price of a bond to be repaid by its internal cash flows. It is an important measure for investors to consider, as bonds with higher durations carry more risk and have higher price volatility than bonds with lower durations.
For the two basic types of bonds the duration calculation varies: a zero-coupon bond has duration that is equal to its time to maturity; a bond that pays a coupon will always have a duration that is less than its time to maturity.
Thus on a zero-coupon bond the entire cash flow occurs at maturity, while a bond that pays coupons yearly and matures in, for example, five years will conversely repay the amount paid for the bond sooner.
Duration is also used as a measurement of a bond portfolio’s sensitivity to interest rate movements.
This stems from expectations that stronger economic activity will fan inflation, eroding returns on securities that pay fixed rates of interest. Such worries can spark selling - and, as prices fall, that pushes up yields, which move the opposite way.
Duration and performance
While all of this may appear to be little more than arcane arithmetic, duration can be a useful indicator as to how the fund will perform, as the movement of bond yields is inversely correlated to the performance of the fund.
So, in the portfolio a manager can hold different duration bonds in order to work out a weighted duration.
For example, in Chris Bowie’s Ignis Corporate Bond fund the duration is currently 7.6 per cent, slightly below the index of 7.8 per cent. This means the manager has less duration risk than the index.
In short, this means is that for every 1 per cent that yields rise, the fund will lose 7.6 per cent, while for every 1 per cent that yields fall the fund will rise 7.6 per cent.
Last year Mr Bowie says he had duration greater than the benchmark, but says he had been shortening that position this year to now be slightly underweight the benchmark, adding that his fund is “getting ready to go quite a bit shorter.”.
“Having duration risk will be one of the biggest risks over the next five years because yields are so low there is not much left to go for. I don’t think there is much left on the table.”
At the moment 10-year UK gilt yields are trading around 2 per cent and have dipped as low as 1.9 per cent.
Quantitative easing
Mr Bowie says that quantitative easing was a particular risk that was prompting him to move to short duration. “Printing money leads to inflation and when inflation comes back and rises, yields will rise,” he says.
“I won’t know when yields will rise until I know for sure about inflation. I think inflation will fall more this year so we have not trimmed duration just yet.
“But once inflation starts to rise, that’s when you trim duration, so we are not doing it yet.”
Inflation in the UK has remained stubbornly above the Bank of England’s target of 2 per cent, however it has started to fall since the temporary effects of the VAT hike and high oil prices have fallen.
In March the consumer prices index was up 3.5 per cent year on year, slightly up from 3.4 per cent in February. However, inflation is widely expected to continue falling over the medium term, as it has done since September 2011.
Ian Spreadbury, manager of Fidelity’s £979.9m Strategic Bond fund, said he was also running duration below benchmark to protect against a rise in gilts.
He says recent gross domestic product figures in the UK, which had plunged the economy into a double-dip recession, had increased the potential of further quantitative easing by the Bank of England, which is “storing up problems for further down the road”.
“It is distorting bond markets by keeping yields artificially low. This in turn keeps unproductive areas of the economy afloat when perhaps they shouldn’t be.”
He adds that more QE exacerbates the risk of a potentially “dangerous” inflationary problem, since the policy was largely experimental.
“Putting all this together leads me to believe gilt yields will probably remain low for a long time yet, but there are significant tail risks.
“Currently I’m positioned to protect against a rise in gilt yields by running a duration below benchmark, while focussing on adding value through careful corporate bond selection.
“Investment grade corporate bonds offer value against gilts, but I expect sectors most exposed to the economic cycle to underperform.”
Interest rate risk
Jim Leaviss, head of retail fixed interest at M&G Investments, says duration was one of the “levers” he used to enhance performance in addition to credit and currency and that he is therefore holding a short duration on his fund.
The manager says he had recently increased duration on the £81.5m M&G Global Macro Bond fund on the back of less positive data, but it still remained low.
“Duration remains fairly short as government bond yields are low and I don’t believe there is much room for them to fall further.
“A large part of the duration exposure comes from UK and European bonds due to my belief that interest rate risk in US assets is currently less attractive, with the country’s economy recovering at a decent pace.”
Schroders’ head of global macro, Bob Jolly, says he thought the UK and US were likely to take the “greatest risks” with regards to inflation and that due to this he is taking a short-duration bias.
“We think the US and UK are likely to take the greatest risks with regards to inflation, due to the debt dynamics in both countries. Both have high levels of nominal debt and real assets dominated by property and equity.
“Deflation would be fairly disastrous for both counties, leaving the central banks prone to keeping monetary policy at an extremely accommodative level for too long. We reflect this through a combination of curve steepening and short-duration biases.”
