Time to Adjust Those Risk Models?

Let's begin with Karen Mazurkewich's article in the Financial Post, Risk models under scrutiny:

The spectre of huge losses at large Canadian pension funds has people in the financial sector calling for a re-evaluation of risk models adopted by the Canadian institutions.

Fuelling calls for a re-examination was last week's newspaper report suggesting that the Caisse de depot et placement du Quebec may announce a whopping $38-billion in investment losses when it releases its 2008 results at the end of the month.

If it's true, "that's a staggering loss," says Tawfik Hammoud, partner and managing director at The Boston Consulting Group in Toronto. "It's unclear to me how you get out of that hole," he adds.

While few seem to have had it as bad as the Caisse under former chief executive Henri-Paul Rousseau, pension funds assets around the world have taken a beating. In December, it was reported that private pension funds among the OECD nations registered losses of nearly 20% on their assets -- an equivalent of US$5-trillion.

In Canada, the full extent of these losses is slowly emerging. A report by Watson Wyatt last month paints an equally bleak picture at home, stating that the Canadian pension solvency funded ratio at the end of 2008 was 69%, compared with 96% at the beginning of that year.

With such gaps in their balance sheets, pension experts like Keith Ambachtsheer, director of the Rotman International Centre for Pension Management at the Rotman School of Management, are calling for an adjustment to the risk models employed by funds.

Most of the simulations run by our pension plans "were based on an underlying assumption of normal probability distributions that don't have these outside, black swans (unseen catastrophic events) on them," says Mr. Ambachtsheer. "I think people should consider events of things that don't get modelled so easily."

Adjusting mathematical models to include doom-and-gloom scenarios means setting a lower bar for returns. Mr. Ambachtsheer admits that getting funds to change their models is tricky: "It's a delicate balance because it's a very competitive environment out there and everyone likes to be number one on the league tables." Then there is the question of whether you reset these things now? The answer to that, he says, is "frankly not obvious."

With all due respect to Keith Ambachtsheer, the problems in pension risk management run a lot deeper than tweaking the models to incorporate "disaster scenarios".

Too little too late. These multi-billion dollar pension funds losing 20%, 30% or 40% in a year, are screwed because it will take them five to ten years or longer before recouping those losses.

Let's not sugarcoat what led to this pension crisis. Clearly there was a breakdown of risk management at one or several levels.

Using 'sophisticated risk models', these pension funds thought they can diversify away all risks, fell asleep as they shoved billions into alternative asset classes like hedge funds, private equity and real estate, bought into the 'cult of equity' , and voila, they all got clobbered because they all forgot about protecting against the biggest risk of all, namely, systemic risk!

From my insider's viewpoint, it was all risk management theater that gave the appearance that they were on top of things, but the catastrophic losses expose the truth.

Importantly, risk measurement is not risk management, and most senior risk officers at pension funds are either woefully incompetent or they are just patsies who have no power whatsoever to stop catastrophic losses before they occur.

The solution to risk problems at pension funds is not simply about 'tweaking the risk models'. If pension funds are to address risk in any meaningful way, the governance model has to change once and for all.

The most important change to the governance model is that you need to have your senior investment and operational risk officers report directly to the board of directors and not to the president & CEO.

Think about it, if the senior risk officers are competent and report directly to the board, they would feel free to unequivocally state their apprehension about certain risks taken in various investment activities without worrying about pissing off the president or some senior investment officer.

They should have the authority to close the books of any trader or portfolio manager who has exceeded their VaR limits or who has taken outsize positions that expose them to huge losses. There should be clear risk guidelines for every single internal and external investment manager who are running liquid strategies. For illiquid private markets, economic exposures should be clearly reported with conservative valuations that do not pad the actual values of these investments.

In fact, I would force the risk officers to write a weekly report to the board and a monthly report to the stakeholders that is publicly available where they clearly outline all the risks in public markets and provide economic risk of private asset classes that are illiquid. Stakeholders should know all exposures and maximum losses if a three standard deviation move occurs. They should also develop systemic risk indicators that are based on timely credit risk indicators and solid financial economic analysis of the fundamental drivers of leverage.

Finally, it is high time that pension funds realize that if they all do the same thing, blindly chasing "alpha" with little or no regard to how they are contributing to systemic risk, then they are doomed to keep repeating the same mistakes.

This insatiable appetite for yield led them down this self- destructive path and even the most sophisticated risk models couldn't protect them against the stupidity of blindly following the herd.

Also, you can blame mark-to-market for exacerbating the crisis, but Lloyd Blankfein is right, if you had adhered to it, you would have reduced risk early on. The key is that you had to have the discipline to adhere to it.

[Of course, Mr. Blankfein neglected to mention that Goldman Sachs had no problems recommending all sorts of alternative assets and esoteric products to their institutional investors as long as they were raking in those huge fees and taking the opposite side of the trade!]

Let me end this comment by stating that risk is not being addressed seriously at public pension plans. Stakeholders are to blame as well because they should provide pension fund managers with clear risk guidelines that state "you cannot lose more than X% in any given year".

And fixing risk at pension funds requires a lot more than just 'tweaking risk models'. It requires wholesale change in the governance structure, giving the senior investment and operational risk officers more power and a direct link to the board of directors who are ultimately responsible for managing pension assets without taking undue risks.

Unless risk governance is taken seriously, pension funds are doomed to repeat the same mistakes that led to the current pension crisis.