The NSAHO has implemented some investing ideas that have been around for a while, Jordan said, adding that this theory of risk budgeting and liability driven investing can actually be put into practice—regardless of limited internal resources.

“DB pension plans,” he emphasized, “can be saved by those who are committed to saving them.”

The Good

Beginning with the best of the trio, “good” leverage reduces uncompensated policy risk. This is done through hedging interest and inflation rate risk. Jordan explained that their strategy uses bond forwards and repos to leverage additional bond exposure beyond what would be possible if they only used physical assets. They have 65% fixed income exposure while only using 25% of their physical assets.

Jordan points out “Sixty-five percent may not be the perfect allocation, but I don’t know what the perfect allocation is”. He continued “Don’t miss good looking for perfect. If we had hedged more than 65%, we may have found ourselves overhedged because of the sensitivities that exist in other asset classes.”

Jordan argues that the 40% unfunded part of his bond strategy not only reduces risk, but may also increase expected returns. The expected return enhancement results if the yield curve in the future is upward sloping on average. With an upward sloping yield curve the expected return from the bonds is higher than the expected cost of financing inherent in the derivatives.

But Jordan stressed that any increase in expected returns is a bonus. The point of the exercise, is just to make the assets better match the liabilities from a risk budgeting perspective.

One more bonus of the strategy is that it may indirectly provide solvency relief, said Jordan. If your actuary agrees that expected returns used in your going concern valuation are increased, your current service cost will decrease, leaving more of your contributions left over for solvency special payments.

The Bad

The “bad” part of leverage that Jordan highlighted was the credit spread charged by mortgage lenders on real estate assets. To minimize credit spreads the NSAHO Pension Plan tries to position leverage in the fund where it is cheapest. They keep mortgage debt as low as possible, and in its place access leverage implicitly with synthetic exposures in other asset classes.

Jordan says he’s heard a few arguments against NSAHO’s approach:

• leverage within the real estate portfolio is necessary to get reasonable diversification; and

• a reasonable amount of leverage can help increase returns.

He explained that these arguments don’t hold water, as his strategy doesn’t reduce total fund leverage, but only shifts it to where the credit spreads are cheapest. The gross exposure to real estate beta doesn’t need to be reduced.

The Benign

That leaves the “benign,” increasing expected compensation from the active risk budget (portable alpha).

Jordan compared the traditional approach versus the portable alpha approach using a large cap US equity example. Traditionally, you invest cash in U.S. equities and get the U.S. market index return plus the value added from active management, he said.

With portable alpha, the exposure to U.S. equities comes from an unfunded derivative instrument. The NSAHO Pension Plan uses the cash that is left unused to invest in hedge funds, with the hedge funds selected to have reasonably low equity beta. The alpha is the hedge fund return less the financing cost that is inherent in the derivatives. Jordan argued that even if you assume a fairly low hedge fund return, with today’s low financing costs you may improve your expected compensation from your active risk budget.

Of course the hedge funds could provide negative returns, but Jordan compared this risk to the chance that an active manager would provide negative alpha. It’s a risk but Jordan argued that the potential impact on the total fund is minimal relative to a fund’s policy or beta risks. And if you wish you can limit the risks related to portable alpha by reducing the amount of your portable alpha exposure.

Taking all of their strategies together, Jordan indicated that they estimated that they had increased their expected real returns by about a third, and reduced their expected funded position risk by about a quarter. This is compared to a typical 60% equities, 40% debt asset mix.

Risks of these leverage strategies can include liquidity risk, counterparty risk, headline risk, career risk and key employee risk. Jordan agreed that such risks need to be managed but remained unfazed. “The 60%/40% alternative to these strategies also has its risks” he concluded.

Currency management was the topic of the most recent Canadian Investment Review debate.

Is active hedging the answer? Thanos Papasavvas, head of currency management at Investec Asset Management thinks so, for two reasons. The first is that the orderly world of Bretton Woods is gone. Thus, “in the fixed exchange rates world of the 1950s and 1960s, currency investing was neither necessary nor worthwhile. However, over the last 15 to 20 years, it has developed greatly by way of cross-border portfolio exposure management and as a stand-alone way to manage assets.”

On top of that, currency investors have different reasons for their investment decisions. “Many market participants do not try to maximize their returns when trading currencies, meaning currencies may trade away from their fair value for prolonged periods of time.”

But is active management worth the cost?

Jay Moore, managing director at State Street Associates, thinks otherwise. “[W]hile there is a strong argument in favour of direct investment to currency alpha strategies within a diversified portfolio, I would argue that investors should primarily concern themselves with existing currency risk rather than embarking on a quest for new sources of alpha.”

What’s key here is the likely current exposure of pension funds to international securities. “The primary challenge to active overlay programs in Canada is that these programs impose constraints that limit investment opportunities to exposures that are highly concentrated within USD, EUR, GBP and JPY, and which make up roughly 80% of total foreign exposure.”

As a result, pension plans are constrained from seeking alpha, say in emerging markets, because they are heavily benchmarked to the developed world in their asset allocation. “Until benchmarks are moved away from a market-cap weighted (or similar) allocation methodology to international investing, active currency overlay will continue to be handicapped by the concentration of currency exposures,” Moore says.

So, a passive or an active currency allocation? Debate moderator Michael Lewis, a principal at Mercer, concludes that “we have seen that currencies should be part of the investment decision-making process from the strategic allocations through to implementation…. the discussion of hedging or active currency management may ‘paralyze or polarize’ a board of trustees.”

Paralyze or polarize? Not for participants in the debate.

They voted 57% in favour of active currency hedging.

Why should we use these? Overlay strategies can help to offset the market value volatility in liabilities and short-term financing costs, Geddes explained.

Overlays can increase interest rate hedging in various types of plans, he said, including those with a traditional 60/40 asset mix or those with universe to long duration fixed income. “The reduction of interest rate mismatch risk can be achieved,” Geddes maintains.

Risk Factors

However, there are a number of risk factors in using overlays. Investment risk (portfolio structuring, liquidity, benchmarking) and operations risk (collateral rules/parameters, legal structure and documentation, operational risk infrastructure) are directly manageable risks, he said.

However, counterparty and liquidity risk can be introduced, warns Geddes. And heightened headline risks and unexpected outcomes can also come into play.

Counterparty risk with increasing systemic leverage is also an important consideration. Because if systemic leverage becomes a problem, says Geddes, what is Plan B? It’s important to envision a scenario when hedging is compromised, he added.

There are other effective approaches to overlays, too. Overlay doesn’t necessarily mean interest rate hedging. For example, extra-long duration fixed income (i.e., strips) may be worth exploring, he said.

Governance

In terms of governance, overlay strategies should be shared between the plan, the consultant and the investment manager. Plan sponsors must disclose and discuss the potential risks and should also measure the effectiveness and appropriateness of the strategy.

Geddes said it’s important to remember the following when considering an overlay strategy:

• walk through the overlay solution before implementation;

• remember that a range of structures may also be effective;

• make sure the risk/reward trade-offs are sufficiently enhanced; and

• ensure that your investment committee addresses potential unintended outcomes.

Brooke Smith is managing editor of Benefits Canada. brooke.smith@rci.rogers.com

However, things are changing, said Lior Menzly, director of quantitative research at Nomura Global Alpha LLC, speaking at the 12th annual Risk Management Conference in Muskoka, Ont.

Menzly said there are two recently developed applications that can exploit the strengths of the BL framework while simultaneously avoiding the framework’s weaknesses.

The first is concentration analysis, which allows risk management to take an active role in shaping the portfolio by finding concentrated positions and cheap hedges. Concentration analysis is actually the reverse engineering of BL; it uncovers embedded manager views in the current portfolio, continued Menzly.

The second is strategic analysis, which incorporates a forward-looking macro scenario analysis into the framework. It can be used to embed top-down views about risk factors (instead of bottom-up views on specific securities), and it maximizes expected returns in base scenarios while minimizing losses in tail events, Menzly explained.

Although the BL framework results in a stable intuitive optimal portfolio, according to Menzly, it’s costly to use from an operational standpoint. The framework asks for too much information in a form that isn’t very intuitive, he continued. The manager has to specify all views and express them through securities in the portfolio, and the challenge is to translate the manager’s views into views that are in line with the BL framework, he explained.

But despite the challenges, Menzly maintains that BL is a worthy model. “[BL] allows us as risk managers to be actively engaged in the construction of the portfolio—instead of taking a passing role.”

Brooke Smith is managing editor of Benefits Canada. brooke.smith@rci.rogers.com

But it’s not just one of these, he said. “It’s all of the above and none of the above.” Risk, he said, depends on your situation and what your goal is in addressing these risks going forward.

“We have to look at risk in a pension plan in a similar fashion.” In the recent past, modern portfolio theory looked at risk in terms of market-based theory—beta, tracking errors—he continued, but we’re now starting to look at pension plan liabilities.

Liability driven investing is changing the way plans manage bonds, he said. Market risk and value added are being replaced by interest rate risk to actuarial ratios and in some cases pension cash matching, he continued.

Equity Investing

Even though the bond landscape is changing, equity investing seems to be continuing on the same basis, he said. Equities don’t have the same ratio-matching properties to liabilities; they are largely labelled as mismatching and viewed purely as a return vehicle, he continued.

But since pension plans need equities for returns, they need to diversify their equities as much as possible. Equities should be used from the point of view of what you’re hoping to achieve within the plan, Wilson explained. “Equities will always be volatile.”

Long-term return of equities is very attractive but the pattern is not, Wilson said. In fact, over the 100-year history of the Dow Jones, there were four decades in which there was no return in equity markets—a pattern, Wilson maintains, that is unhelpful to pension plans.

So when considering risk, plans sponsors need to remember the following:

• Risk is situational and should be managed relative to the organization’s liabilities and their impact on the financial risk.
• The organization’s focus should be strong and affordable benefits.
• Bond management is shifting from market-based risk to managing financial risk.
• Although equities continue to be managed to market-based risk, they should move to consider financial risk as market-return patterns are not helpful.
• Low beta approaches may better address pension financial risk (but at some opportunity cost).
• A number of documented anomalies seem to offer better financial risk protections to pension plans.

Brooke Smith is managing editor of Benefits Canada. brooke.smith@rci.rogers.com

Plan sponsors in Canada thought alternatives—private equity, real estate infrastructure, commodities—would get them out of the bad situation. In fact, pension plans have been allocating more to alternatives. In 1999, pension plans allocated 6% to alternatives. In 2009, that number had increased to 22%. “[But] out of the last few years, we didn’t do very well,” said Rabovsky. “We had started to change the way we looked at the capital markets and had very little regulation to support that.”

There were three areas that caused the credit crisis: too much credit, belief that the markets knew best (the idea that the market was going to right itself) and excess of complexity, competition and compensation. In managing risk, then, the tactics used have to be far more complex than what we’ve been using, said Rabovsky.

Plan sponsors need to look at what type of asset classes they want to be in and look for asset classes that are less correlated. Leverage also needs to be reduced. And plan sponsors need to review quant/leverage and refine their manager profiles.

Macro Factors

Some of the bigger issues that may affect global returns include demographics and emerging wealth. However, investors have started to look at geopolitics (terrorism, competition for resources, free trade) and public policy (G8 deficits, managed foreign exchange, inflation/deflation). Out of 15 extreme risks to consider, the first six—depression, hyperinflation, excessive leverage, currency crisis, banking crisis, sovereign default—are related to the concept of leverage.

The challenge for plan sponsors is do they manage for expected outcomes or do they manage for their fears or worst expected outcomes? You need to think about what it is you need to discuss as a committee when things go wrong. How are you going to act? said Rabovsky. “Risk isn’t a four letter word. Unintended risk is.”

Brooke Smith is managing editor of Benefits Canada. brooke.smith@rci.rogers.com

Risk managers, Baker said, aren’t as smart as they sometimes think they are. In fact, they learned some valuable lessons from the crisis:

• assume liquidity is not there when you need it;

• rebalancing only works with liquidity;

• keep leverage away from vital organs; and

• systemic risk is not diversifiable.

Going forward, managers need to focus on liquidity risk. They haven’t done a great job of it, said Baker, but they’re getting better. “They need to understand qualitatively and quantitatively the interactions among all component parts of a portfolio.”

Brooke Smith is managing editor of Benefits Canada. brooke.smith@rci.rogers.com

Among Jorion’s other major conclusions about the role of risk during the crisis — black swans aren’t to blame for all the failures by any stretch of the imagination. While “unknown unknowns” such as regulatory risk (i.e., the ban on short sales), event risk (deleveraging of investment risk) and contagion risk played a role in 2008, there were “known unknowns” like model risk and liquidity risk that could have been dealt with in 2007. Understanding that risk models have their weaknesses is key according to Jorion who noted that the most successful risk managers are able to think beyond the limits of those models.

Stay tuned for the full summary of Jorion’s research and presentation.