Case Study: Redesigning A $5.2 Billion Fund

June 18, 2007

Rdesigning a $5.2 Billion Fund

As the gap continues to widen between assets and liabilities in the pension world, pension managers are faced with the daunting task of trying to match asset flows to obligations. Public pensions, which tend to be more conservative than their private counterparts, are nonetheless experimenting with new ways to design their investment portfolios. Sophisticated products that graft hedge fund strategies onto index-sensitive products are weeding their way into these plans.

North Dakota, known for its prairies and badlands, has one of the more innovative public plans in the country. Ten years ago, it began building an index-sensitive portfolio across its two asset classes: equities and bonds. Today, the pension plan employs complicated strategies that remain rooted in the index world. But whatever benchmark is used, whether it is the Standard & Poor's 500 or Russell 2000, subfunds in each corner of the portfolio seek returns above and beyond the index.

A Radical Overhaul

When Steve Cochrane arrived at the North Dakota Retirement and Investment Office (RIO) in 1997, he was faced with a very stodgy investment strategy that provided very mediocre benchmark-like returns with higher-than-expected volatility. The State Investment Board (SIB) of the Bismarck, North Dakota-based pension fund asked him to revamp its traditional 60 percent equity and 40 percent fixed-income portfolio into one that could provide more alpha while lowering its overall standard deviation.

At that time, the pension was returning double digits, driven by the robust stock market of the 1990s. But, in comparing it with other public funds country wide, North Dakota ranked within the 61st percentile, according to consultant Callan Associates.

Cochrane, who is now RIO's executive director, is responsible for all of North Dakota's state defined public pension assets. When combined, the assets under management total upwards of $5.2 billion, and fall either under the $3.9 billion pension pool, which includes, for example, the Teachers' Fund for Retirement, or the $1.3 billion insurance trust, used to run the state's workers' compensation fund. The two plans are run similarly but discretely.

As soon as Cochrane arrived, he crafted a radical overhaul of the investment strategy with long- and short-term allocation goals. His first proposal was to jettison active management in some asset classes and replace it with index-sensitive strategies. He then introduced non-traditional asset classes to the mix, primarily in complementary, low correlation strategies, including hedge fund overlay and 130/30 strategies, as well as real asset, private equity and venture capital investment strategies.

Today, the success of Cochrane's design has made the North Dakota RIO plans one of the more successful public plans in the country. At the end of June 2006, the plans fell within the top 2 percent of returns for all public funds, according to Callan.

The Large-Cap Conundrum

Cochrane tackled the domestic large-cap portfolio first, and with good reason. For the past three, five, and 10 years, domestic large caps underperformed their small-cap brethren as well as the broad U.S. market, according to Russell. And, because active managers generally do not unearth information the market doesn't already know in heavily traded segments like large caps, market inefficiencies tend to increase further down the capitalization spectrum. Which means, says Cochrane, "it's extremely difficult to outperform the market in large-cap domestic equity, risk adjusted."

He convinced the SIB to go the index-sensitive route for this slice of the portfolio, and threw out anything with a tracking error of 6 percent or above—which was nearly everything. At the time, it was a startling orchestration of assets, and nearly a decade later, it is still pretty unusual.

What remained in the domestic large-cap portfolio was relegated to one of three tranches, the first of which held a pure index product with a zero percent tracking error. Into this tranche went the plan's S&P 500 index product, which was about 10 percent of the $1.1 billion domestic large-cap portfolio. It has since been eliminated and its assets redistributed between the two remaining categories.

Find your next ETF

Reset All