The Perfect Storm

The month of August saw a dramatic shift downward of the South-African inflation-linked bond yield curve.

by Shaun Levitan
Published July 14, 2015
Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on LinkedIn0Email this to someone

The month of August saw a dramatic shift downward of the South-African inflation-linked bond yield curve. It is the kind of market event that doesn’t feature in the South African news or financial press headlines. Maybe it’s because inflation-linked bond movements are not nearly as exciting as movements in the All Share Index (unless you work at Colourfield!). The reality is that the impact of this yield curve move is significant for all stakeholders of defined-benefit (“DB”) pension funds. That’s because it’s the kind of market event that has a dramatic effect on the financial solvency of a typical DB fund and on the corporate sponsor’s ability to underwrite it.

Since many funds have on-going liabilities linked to inflation, a market value is assigned to these liabilities based on the yield at which inflation-linked bonds trade. A fall in the yield of inflation-linked bonds results in a higher market value of these liabilities and vice versa. Put simply, every time an inflation-linked bond has a price change, so too does the value of the DB pensioner liability change. Unfortunately, this impacts on the ability of the fund to grant pension increases, affects the solvency of the fund, and the cost at which the employer recognises the DB scheme in its accounting statements.

August Market Movement

The long-dated R202 inflation-linked bond traded at a yield 0.40% lower on 25 August 2010 than it did at the start of August 2010. For the average DB pensioner liability, this seemingly ‘small’ rally in the bond price caused the liabilities to increase by more than 3.5%! The reason for the ‘multiplier effect’ i.e. a small drop in yields resulting in a significant increase in liabilities, is the sensitivity of long-dated liabilities to a change in real interest rates. This is what is referred to as a fund’s “interest rate risk”.

Trustees would hope that fund assets would provide the necessary protection from such increases in liability. However, with equities falling by 4% over the same period, funds with a large exposure to equities would have experienced a perfect storm: a rise in liabilities accompanied by a fall in assets. An investment strategy with any equity allocation would thus have experienced a large fall in their on-going solvency levels (the ratio of assets to liabilities). The perfect storm manifests as a threat to the ability to pay increases in line with inflation, employers experience painful volatility in their accounting statements and all that trustees can do is hope for a quick recovery.

Funds that have a strategy focusing on short to medium term obligations with inflation-linked assets would not have been protected. The liabilities for obligations after the first 15 years actually increased by 10% during the recent market movements. Longer dated obligations accounted for two thirds of the liability increase. This highlights the interest rate risk present in the tail (i.e. that long dated obligations are much more sensitive to changes in rates than those due in the short to medium term).

Similarity of Pension Funds to Hedge Funds

Many pension funds do not invest in so called long/short hedge funds on the grounds that they are a “risky” investment. The events unfolding in August show, that in fact, most defined benefit funds themselves, mimic such hedge funds: they are “long” equities (their assets) and are “short” inflation-linked bonds (a measure of the cost of their liabilities). This means that DB funds which do not invest in inflation-linked bonds, seek to benefit on a relative basis when equities rise and inflation-linked bond prices fall (since the bonds are now cheaper to purchase at this later stage). How many trustees realise they are assuming this risk?

The Least-Risk Portfolio

It is not uncommon for funds to invest in a diversified portfolio consisting of an allocation to multiple asset classes to minimise the risk of a market fall in any one particular asset class. This is normally supported by a detailed Asset Liability Modelling study. Contrary to conventional thinking, a diversified investment strategy is NOT the least risk position when liabilities are taken into account. An investment in the appropriate portfolio of inflation-linked bonds would remove the investment, interest-rate and inflation risk from the fund. The fund would be indifferent to the yield curve movements experienced recently.

The efficient frontiers below set out the risk return dynamics of an asset strategy (on the left) which ignores the fund’s liabilities and one which considers the liabilities (on the right).

The frontier on the left shows that a cash investment produces the most stable returns and hence is the ‘least risky’ option (as measured by the volatility of expected returns) but is expected to produce only pedestrian returns. A pure equity investment is the ‘cowboy portfolio’ which is expected to produce the highest return but not without exposing investors to the highest risk. A diversified multi-asset class strategy sits comfortably between the two and provides a reasonable expected return at a reasonable level of risk. All three strategies appear on the efficient frontier and are all viable options depending on the fund’s tolerance to risk and return expectations (and would depend on the regulations governing the fund). Importantly, there is no consideration for any obligations required of the investor.

The bond portfolio would not feature on this efficient frontier and hence would be excluded from consideration since the multi-asset class portfolio would expose investors to the same level of risk but with a higher return expectation.

The frontier on the right incorporates the on-going obligations of a DB pension fund and paints a very different picture. The multi-asset class portfolio is no longer efficient since the risk, as measured by volatility of the fund’s surplus (assets less liabilities), is unjustifiably high for a given level of surplus. The experience in August 2010 is a case in point where the diversified portfolio added to risk through its mismatched strategy with the liabilities. As equities and inflation-linked bond yields decreased in tandem, so the surplus was hit by a double whammy since liabilities rose when assets fell.

Importantly, a carefully selected portfolio of bonds now appears on the efficient frontier and is by far the least risky option (even less risky than cash!). In August, this portfolio would have risen as the liability value rose and the surplus position would have been unscathed by these adverse market conditions.

The efficient frontier shown in many Asset Liability Modelling exercises do NOT liabilities into account and hence do not illustrate the minimum risk option available to the Fund. Trustees may decide to forego the risk-free government bond portfolio in search of higher returns, but the starting point when designing an investment strategy should be their risk free position.

Conclusion

The month of August 2010 was a perfect storm for DB pension funds which were exposed to equities as part of a diversified portfolio and which had no inflation-linked bond exposure. It is possible to protect against market movements by investing in a tailored portfolio of government bonds. This would have provided an ability to emerge from this perfect storm unscathed. The minimum risk portfolio in respect of a fund’s liabilities should always be the starting point of any investment strategy.

Back to top