GLOBAL - St Bride's CEO Robert Houston has called on pension funds to end their "fixation" with benchmarking, arguing that schemes are blinded by established wisdom to follow benchmarks even to the detriment of their own portfolios. But is he right?
"Criticising benchmarking for breeding mediocrity is a fair enough point. But what is the alternative?" said Andrew Baum, professor of land management at the University of Reading and chairman of the investment committee for CBRE Global Investors' global multi-manager team. "If managers expect to be rewarded for good performance, how do they prove it?"
Whether pension funds should benchmark is the easy question. More difficult ones are what to benchmark, and how. "In the old days, pension funds basically had the FTSE for equities and a similar index for bonds and that was it," said pension fund adviser Georg Inderst, who two years ago wrote an OECD report on (the lack of) infrastructure benchmarks. "With alternatives it's a bit more difficult. How do you benchmark real estate or private equity or commodities? In infrastructure, there's no established index of any form."
In a paper she wrote earlier this year on how Australian investors might go about benchmarking their property portfolios, Russell Investments senior analyst Samantha Steele argued that, despite the availability of more sophisticated indices, it was up to investors to reflect on how they used them. "Benchmarks should be used with a healthy degree of scepticism," she wrote, pointing to the need for additional metrics linking benchmarks not only to return objectives but also to risk elements including asset quality and leverage.
Arguably, we are already seeing a healthy degree of scepticism among pension funds whose managers have beaten the market benchmark but still left them struggling to pay pensions. "If the benchmark has fallen low enough, it won't pay pensions," said Inderst, pointing to an increased use of absolute return, inflation-plus and LIBOR-plus indices among pension fund infrastructure investors. Among infrastructure investors the most common benchmarks are 8-10% absolute return or CPI +5% or LIBOR +4-5%, he added.
On the risk issue, Steele has some pretty vocal support. "If your benchmarks are all down because markets got slammed, who cares if you beat your benchmarks? How are you mitigating downside risk?" said Leo Kolivakis, a blogger and former senior investment analyst at both Caisse de dépôt et placement du Québec and the Public Sector Pension Investment Board in Canada.
Kolivakis posits liability-driven investment as an alternative benchmarking approach, with its focus on limiting downside risk. "But even there you need benchmarks. The critical thing is to be transparent, disclose benchmarks in all activities and only reward risk-adjusted returns, not leveraged beta."
In past blog entries Kolivakis has argued that apparent outperformance often marks an inadequate approach to accounting for risk in the underlying portfolio. Even given that each pension fund has its own risk profile so benchmarks are not always the same, Kolivakis said in an email, "pension fund managers should be compensated on risk-adjusted basis".
The NOK3trn (€388bn) Norway Pension Fund Global currently benchmarks its real estate portfolio against the IPD's 15-country European property index and increasingly against the IPD global index. Its two-year strategy to 2013 will see it move away from a benchmark focus in favour of long-term risk-weighted real return.
In his presentation, Houston hinted at herd behaviour among pension fund investment committees unwilling to back a benchmarking trend. He also bemoaned the absence at his presentations of consultants, who he holds responsible for not enlightening pension funds. Those contacted by IP Real Estate did not respond to requests for comment on his view. But in fact benchmarks - given enough of them - may mitigate the herd tendency.
Proliferation of benchmarks makes comparing pension fund portfolio performance more difficult. But that is not particularly important. "The most important thing for pension funds is that they match their assets and liabilities, not that they can see what their neighbour is doing," said Inderst.
But in fact, he added, it is less an argument for no benchmark than for a proliferation of benchmarks, as multiple metrics potentially create a diversification effect in the market.
"If they're all using the same index, there is a risk of common behaviour," he said.
He contrasted this with an equities index: when pension funds use different benchmarks, a company may enter or exit one index without doing the same across all indices.
In any case, even if benchmarks measure only against the market, they are indispensable to investors, according to Baum.
"If there are no market benchmarks, or if they are ignored, then recent history shows that managers can get rewarded for delivering mediocre returns and for taking big risks, including leverage, to get there," he said. "It's truly shocking that after the value destruction we've seen managers feel thay should not be benchmarked."