Twenty years ago, Norway established a sovereign wealth fund to manage the public wealth generated by its oil revenues. Today the Government Pension Fund–Global is valued at well over $400 billion, larger than the size of Norway’s annual GDP. With only five million residents, the fund represents close to $100,000 per citizen, and, as such, is understandably prominent in Norwegian life. Consequently, there was a great deal of public debate about how the fund should be managed after Norway’s fund, like many funds, experienced significant losses as a result of sharp declines in global asset prices in 2008 and early 2009.
As part of the debate, some members of Norway’s parliament proposed pulling the fund’s active management mandate and shifting the entire fund into a passive index. But it was not entirely clear that passive investing was the right answer for Norway, so the Ministry of Finance commissioned an extensive external review designed to assess how well the fund had been managed and to make recommendations for going forward.
The centerpiece of the review was produced by Professor Andrew Ang, who worked with William Goetzmann of Yale and Stephen Schaefer of London Business School. The three-part report first sought to assess the performance of the fund’s managers and answer the question of whether the fund should pursue active or passive strategies.
“Academic theory says that if markets are efficient then investors should simply hold index funds,” Ang says, “and that those pursuing active management will lose money because of the transaction costs associated with paying fees to fund managers.”
This view — the efficient markets hypothesis — has evolved since its inception in the 1960s, and its more recent incarnations do take into account transaction costs, financing and other frictions, and that fund managers are either more or less advantaged based on their skills or their access to superior information about the market. Ang and his coresearchers determined that, consistent with these realities, there is no reason for the fund — or any large fund — to pursue a purely passive strategy.
“The fund’s large size confers advantages that Norway should make the most of,” Ang says. “Large funds have the advantage of being able to do things on a truly global scale and to hold assets with very long payoffs that short-term investors or smaller funds cannot.” A large fund can also provide liquidity — the Norwegian fund was one of the largest buyers of equity in the last part of 2008, worldwide — and can buy cheap when most others are selling.
Ang and his coresearchers next conducted a historical evaluation of the fund’s long-term performance and assessed whether the fund was being correctly and efficiently managed.
“We found that the fund’s managers were doing precisely what they should have been doing: collecting as many sources of risk premiums as possible,” Ang says. “There are many strategies for exposing a fund to risk premiums — value investing, exploiting various sources of credit risk or illiquidity risk, selling volatility protection or providing liquidity when others want to sell. The fund should have exposures to these risk factors.”
Importantly, the researchers found that the majority of the fund’s losses in 2008 and 2009 could be attributed to its pursuit of these factor risk premiums. But the common intuition that the resulting losses were bad for the fund is a superficial one, according to Ang. “It is no mystery that exposure to these is going to result in losses at some point. It is precisely because factors — like credit, liquidity or volatility risk — do badly during certain times that they pay off handsomely in the long run.”
Where the fund stumbled, Ang says, was in its failure to clearly communicate the types of active strategies that it was pursuing. “The Norwegian parliament and public are sophisticated about these matters. For example, when equity markets fell dramatically in 2008, there was no outcry because the Norwegians understood that equity markets could crash.”
In the case of the more recent discord over losses in the assets under active management, Ang suggests that the Norwegian public should have been made aware of the active investment strategies the fund was pursuing prior to 2008. “They would have understood that such strategies might provide low returns or even losses in the short term, and been steeled to expect the downside,” he says.
The final part of the report — and the one of greatest interest to the wider fund management community and to large pension funds in particular — made recommendations on how the fund should be set up to exploit its advantages. The central recommendation is to pursue risk-factor exposures through a passive rather than active strategy, freeing up active management resources to pursue other premiums and mispricing opportunities in the marketplace. There are many sources of dynamic risk premiums other than passive holdings of assets and these should be harvested by investors.
But there is no index fund that provides exposure to, for example, volatility risk. “Part of Norway’s problem, and the problem that many investors have right now,” Ang says, “is that their benchmark is an asset-based benchmark composed of things like an equity or bond mix, without additional components that would give long-run returns on, for example, volatility risk, value-growth risk, credit risk or other factor exposures.”
Ang and his coresearchers developed a framework that provides a way to engineer risk factors into the passive component of a fund — a practical tool capitalizing on factor benchmarks rather than asset benchmarks. The framework also provides a way to holistically look at an investor’s portfolio in terms of its underlying risk properties, rather than viewing it solely through the lens of asset classes.
Shifting factor exposure into passive benchmarks shifts decisions about how much factor risk to bear from the fund manager back to the investor. It also allows investors to monitor risk more closely. Active managers should be pursuing strategies that cannot be replicated by factor exposures and thus bringing factors into the benchmark reduces costs to the investor.
“Factor benchmarking is better for investors from a back-pocket point of view,” Ang says, “because investors should only be paying for active fund management that cannot be engineered passively.”
Andrew Ang is the Ann F. Kaplan Professor of Business in the Finance and Economics Division at Columbia Business School.
Read the Research
"Norwegian Government Pension Fund–Global"