Risk parity becomes bittersweet flavour of the month

A risk parity approach to asset allocation is flavour of the month, in spite, and because, of the leverage it requires. Amanda White explores the topic with Greg Allen, president and director of research at Callan Associates, and Steve Foresti, managing director at Wilshire Associates.

The public castigation the State of Wisconsin Investment Board (SWIB) received following its decision to use leverage in its new strategic investment direction, is testament to the philosophical leap required by pension boards in considering a risk-parity approach to asset allocation.

On the surface, the theory makes sense: reduce the traditional allocation to equities so that diversification endures via a more equally-weighted allocation, and use leverage to increase the return.

A number of asset managers have been touting this approach to a global diversified multi-asset class portfolio – groups such as Bridgewater, BlackRock, Putnam, AQR and PanAgora, whose chief investment officer, Edward Qian first coined the term in 2005.

But now, as a reaction to the mean variance optimisation framework resulting in under-diversified portfolios that failed to provide risk control when it was needed, public sector funds and their consultants are exploring the approach as a potential asset allocation alternative.

While the theory makes sense, Greg Allen, president and director of research at Callan Associates, says there are many considerations that funds will need to overcome in exploring this approach.

The philosophical hurdles include overcoming the predisposition to favour equities and introducing leverage at the policy level.

“In the US there is nothing worse than being different and wrong at the same time,” he says. “And the return pattern of this approach is completely different which is aggravated by using leverage.”

In addition Allen says leverage is “still a dirty word” – despite its acceptance in strategies, it is rarely so explicit.

The SWIB found this out when they announced their 2010 asset allocation strategy, which it had been considering for some years, which includes shifting its allocation out of equities into fixed income and adding leverage. (Ohio Fire and Police have also recently approved this strategy).

For the SWIB the allocation to equities comprised about half the core fund’s assets but 90 per cent of the total fund volatility, so the rationale was to trade a reduction in the allocation to stocks for the lower risk of fixed income.

In the US, the media put pressure on the SWIB accusing them of gambling with taxpayers’ money.

But according to consultants such as Callan’s Allen, the approach can represent a potentially liquid, transparent, low fee alternative to a hedge fund exposure.

A Callan Investments Institute research paper contrasts the methodology of this approach with the traditional mean variance approach in the context of developing policy portfolios for large institutional investors.

Its analysis determines that in the 20-year period since 1990, a levered risk parity portfolio that delivered an 8.25 per cent return would have done so with about half the volatility of an unlevered efficient frontier portfolio.

But in order to produce an expected return of 8.25 per cent, the paper says that a levered risk parity portfolio employing standard asset classes would require somewhere between 40 and 60 per cent leverage, depending on the expected returns for the unlevered portfolio and the cost of borrowing. This puts a whole raft of operational considerations on the table.

“One of the most important considerations is that you have to be good at borrowing,” Allen says. “This is a skill that is very specialised.”

In addition he says funds can add value by making global asset allocation decisions along the way, and most fund sponsors are not set up for that either.

Wilshire Associates’ paper on the matter, “Risk-focused Diversification”, also highlights that operational considerations become more of a focus under this approach and include incremental management costs and in particular leverage costs, which are variable under periods of market stress.

“Understanding a program’s results involves attributing relative performance to active management, identifying any tactical asset allocation decisions and assessing mechanical factors such as leverage costs.

“For most investors implementation of a leveraged strategy would likely require the retention of a beta overlay manager to execute and maintain the desired leveraged systematic exposures or an allocation of capital to one or more of the off-the-shelf investment products which employ embedded leverage to achieve asset class risk balance.

The managing director and head of research at Wilshire, Steve Foresti, says he views the approach as a “removal of a constraint connected to building a portfolio”.

“The main objective of a risk-focused portfolio is the attempt to maintain diversification at a required rate of return. If you move to the right of the efficient frontier you get more risk for more return but you are sacrificing diversification to get more return.”

This approach attempts to achieve the same level of expected return while maintaining diversification.

“The catch is while you are removing a risk, you are replacing it with other risks,” Foresti says.

When using synthetic exposures through diversification, liquidity issues are very important and need to be well-thought-out: this means cash flow, liquidity and operational-type risks are paramount.

“Risk management becomes a heightened focus and few institutions are equipped to handle it in-house.”

And derivative maintenance issues are a particular consideration in times of extreme market volatility.

Wilshire also notes that the asset class to be increased relative to a traditional portfolio is not necessarily where an investor must have derivative exposure.

Another aspect to consider, highlighted in the Callan paper, is that due to its higher allocations to fixed income, the levered risk parity portfolio will be more sensitive to interest rate movements than an unlevered efficient frontier portfolio with the same expected return.

All these operational risks can potentially be overcome with advances in the monitoring, reporting and risk management tools use by institutional investors.

What may be harder to overcome is the psychological risk of being different, as Callan’s paper highlights.

“One risk that will always remain – by design, the underlying portfolio will have a very different pattern of returns from the portfolio employed by the typical long-term investor. Applying leverage will amplify this difference.

“In periods characterised by rising equity markets, particularly if they are accompanied by flat or inverted yield curves, the levered policies have the potential to underperform peers by thousands of basis points.

“During these periods, fund sponsors who choose to implement this type of approach will need to be able to convince their constituents to maintain a long-term perspective. Ironically that is the same challenge that the proponents of the traditional approach are facing today.”