Norwegian fund could go hunting for 'exotic betas'

Officials at the 2.59 trillion Norwegian kroner ($458 billion) Government Pension Fund-Global will consider a proposal to overhaul the Oslo-based fund's benchmarks into ones that mimic active strategies, a move that eventually could cut the need for external active managers.

Martin Skancke, director general of asset management at Norway's Ministry of Finance, said it was too early to say how revamping the benchmarks would affect external managers, but warned that the proposal won't be implemented any time soon, and that it could include use of some active strategies.

The proposal, which would affect both stock and bond investments, represents an effort to incorporate so-called “exotic betas” in the benchmark, and could be picked up by large funds beyond Norway, experts said. These benchmarks could be replicated through use of long/short strategies.

A trio of professors recommended in a report to the Ministry of Finance that the Norway oil fund create custom, completely liquid benchmarks onto which it would impose factor tilts, thereby capturing exposure to an array of systematic risk factors.

Those factors — sometimes called exotic betas — capture about 70% of the fund's returns from active management; therefore, active management returns “may be more cheaply replicated in-house,” said Andrew Ang, the lead author and professor at New York's Columbia University's business school, in a Jan. 20 seminar in Oslo that was simulcast online.

“I regard this as a very original proposal ... it's quite iconoclastic,” said Roger Urwin, global head of investment content at Towers Watson & Co., Reigate, England commenting on the report. “This is kind of saying a large part of the asset management industry is not needed.”

That's because a traditional active mandate — such as a global equity manager looking to provide 200 basis points better than a world stock index — would have no place in the proposed structure because most of its performance could be replicated by use of these factors.

“This is beta knocking on the door of alpha and becoming a bigger piece of the investment agenda of the next decade,” Mr. Urwin said.

“For a fund that currently uses passive benchmarks, this is a paradigm shift,” said Tristan Froidure, Paris-based managing director and head of research at quantitative “anti-benchmark” index provider TOBAM, commenting on the report in an e-mail to Pensions & Investments. “This approach will without doubt provide (Norway with) greater transparency (and) provide a greater accountability for active management, as factors that have in the past been a significant component of performance but were not included explicitly into the passive benchmark will now be recognized.”

Experts said the report would resonate outside of Norway, and can be seen as a key step forward in redefining how investors define outperformance.

Jason MacQueen, managing director at R-Squared Risk Management Ltd., London, said it's not fair to investors for active managers to use systematic risk factors in their investment processes, but then expect to be measured against a benchmark that only includes market risk — and excludes exotic betas.

“This (report) is quite bad news for active managers,” Mr. MacQueen said. “This lifts the veil on what they've been getting away with for years and years.”

Mr. MacQueen expects other sovereign wealth funds to follow Norway's lead.

In the Jan. 20 seminar, Mr. Ang emphasized that the report calls for the Norway fund to remain in active management. However, he noted that, “overall, the fund is not that active,” and that how the fund manages beta accounts for more than 99% of its returns.

The Norwegian fund now invests most assets in line with its benchmarks, and would continue to do so if the new benchmarks were implemented.

MSCI Barra offers a number of factor benchmarks that could be combined to replicate a similar strategy, said Remy Briand, Geneva-based managing director and global head of index and applied research at MSCI Barra.

Mr. Briand and his colleagues wrote a paper titled, “Portfolio of Risk Premia: a new approach to diversification,” in January 2009. However, he said that the report by Mr. Ang, William Goetzmann and Stephen Schaefer was the first to consider using factors at a portfolio strategy level, rather than for risk management. Messrs. Goetzmann and Schaefer are finance professors at Yale School of Management and London Business School, respectively.

Using factors at the strategic level involves “seek(ing) exposure to those factors, think(ing) about what exposures you should have to those factors and how they should be combined,” Mr. Briand said.

But the proposal's advanced nature might also be its downfall. Mr. Urwin said the challenge of selling a proposal to the Norwegian people “that seems quite abstract could be problematic.”

Mr. Skancke agreed there are limits to how abstract the fund's benchmark could be and still be effectively communicated. The risk that the public might lose confidence in how investments are handled is the biggest risk facing the fund, he said.

The finance ministry commissioned the report as part of a review of active management. Criticism of the Norway fund's performance through the financial crisis focused not on the fund's overall losses but losses from active strategies.

Any reworking of the fund's benchmarks would need parliamentary approval. But the finance ministry will not be ready to recommend such a change when it makes its annual report to the Storting, the Norwegian parliament, this spring.

“A full-fledged transition to a factor benchmark ... is a big step,” Mr. Skancke said. “This is a process that will perhaps take some time.” A date for the report has not been set, but will likely be around Easter.

Mr. Ang and his co-authors suggest the fund create custom benchmarks and use dynamic factor tilts that would tweak exposures. In practice, this could be done with a long/short portfolio.

For example, a value/growth factor goes long value stocks and short growth stocks. If a value stock appreciates to the point of being rebranded “neutral” or “growth,” the factor would drop its long exposure to the stock (and go short if it becomes growth).

The result would be a dynamic benchmark that holds time-varying non-market-weighted positions. The fund would continue to hold a 60%/40% stocks/bonds asset allocation, but these factor tilts would capture systematic returns that otherwise would be labeled as “active,” according to an e-mail from Mr. Ang.

Additional factors could capture liquidity, volatility, credit, small-large and momentum risks.

A key strength to these benchmarks is that the fund's owner, the Ministry of Finance, would set the type and size of exposure to each factor, much like an asset allocation, Mr. Ang said. Plus, the factors would be rebalanced similar to how many pension funds rebalance within asset allocations. Instead of holding an optimal allocation of assets, the fund would hold an optimal allocation of factors.

If implemented, the fund could capture in its benchmark return much of the risk exposure it currently counts as coming from active management. That, in turn, raises the bar for active managers.

The report is available online at www.pionline.com/norwayreport.