The Canadian Source Of Employee Pension Fund Investment And Benefits Plan Management

Back Issues

The Magic Of Compound Interest And Risk Budgeting

By: Vaino Keelmann

Budgeting from year-to-year, while keeping an eye on long-term actuarial rates of return, may enable pension plans to take advantage of the ‘Eighth Wonder of the World’ – compound interest. Vaino Keelmann, of API Asset Performance Inc., examines this approach.

Recently I was talking with a senior Defined Benefit pension fund sponsor about downside risk management. They pointed out that by limiting potential investment losses, we would be putting a cap on the ‘upside potential’ of the portfolio. This impression rests on the premise that portfolio returns are symmetrical, or positive, and negative moves are equal, which they aren’t in the real world. Just ask your money manager.

Leaving that aside, let’s turn to the basic mandate of the pension fund, as well as pretty well all DB plans in Canada. That mandate is to pay the bills, not to ‘outperform’ in their investment policy. And the best accepted guess of what the bills or liabilities are comes from the actuary, who has to guess and use mathematics to build up logical longterm future scenarios.

On the investment side, we are seduced into thinking like the actuary and talking about the long-term as well. Indeed Harry Marmer, in his book Perspectives on Institutional Investment Management, points out that, using statistical thinking, the minimum number of years required to be 95 per cent confident of a manager’s performance ranges from 14 years for global equity to 24 years for Canadian fixed income. In the cliché of the business, the assets of the money manager go up and down the elevator every day. In other words, individuals are key to money management. But does any one of us know any person who has not changed significantly over 10 years let alone 14 or 24?

So, I am going to suggest we take issue with our industry’s fixation with the longterm at the expense of short-term prudence. After all, if I can only judge the money manager after 14 years, then I can accept at face value the senior investment counsellor, who in the 1980s, publicly advised clients to “trust me” because the time horizons are so long. That counsellor subsequently melted down before the long-term came to pass.

‘Take Off The Policy Portfolio Corset’

Investment risk guru Peter Bernstein wrote earlier this year: “The time has come to take off the policy portfolio corset and breathe freely – for the indefinite future, not just in today’s world ... The main message is that we must abandon fixed views of what the future is going to be like ... The present structure has a lot of conveniences, it’s a very easy way to organize the way we go about doing our business. It sets up clear-cut marching orders for consultants and other people ... All of that suggests a degree of neatness about the investment process (but) there’s nothing neat about it. It’s very hard.”

To make it easier, we can still accept the legally stipulated actuarial rate of return requirement as a guideline, but we can also start to employ the shortterm budgeting period – one year – that we, as a society, are most comfortable with.

So by budgeting from year-to-year, while keeping our eye on the long-term actuarial rate, we can now work with the ‘Eighth Wonder of the World,’ Compound Interest. As Lord Rothschild most vividly illustrated for us, using interest on interest can enable us to take $1 and double it just 20 times to become a millionaire.

As well, remember, Keith Ambachtsheer and Don Ezra, in Pension Fund Excellence, told us to expect 20 per cent of our pension payouts to come from contributions and 80 per cent to come from investment returns. The magic of compounding is crucial to the 80 per cent for most of us who do not expect to benefit from a windfall to provide that amount when needed.

When we accept the one-year time horizon, and the notion that we are aiming to make the actuarial rate every year, then life becomes more manageable.

We can then step back into Marmer’s statistical world where “One of the major advances in asset liability modelling is the ‘personalization’ of the risk definition to go beyond volatility or standard deviation to more meaningful terms such as downside risk or value at risk.” Instead of the traditional asset/liability modelling exercise, described by Marmer as when the optimal portfolio is “the one you choose just before you cry uncle,” we can take on risk budgeting employing valueat- risk to give us the doorway into the future.

During the fall of 2002, I was on a panel with Valter Viola, of the Canada Pension Plan Investment Board, and he effectively showed us how he and his colleagues like to use this methodology for their asset strategy.

You don’t have to be a megaplan like CPP to adopt proper risk budgeting. Advances in computerization and the simplification of the statistical formulae means that value-at-risk can be economically employed for less than plan sponsors pay for their conventional asset/liability study. No mysterious secrets, the formulae are widely published – see Professor Philippe Jorion’s Financial Risk Manager Handbook for the Global Association of Risk Professionals exams – and used in the financial community.

‘Worst Expected Outcome’

To build up your risk budget to give an independent, objective future scenario with value-at-risk, we take the individual holdings of your current portfolio, not your portfolio historical track record or that of your money manager. Then, take the historic standard deviations or volatility of those securities as well as their correlations. Correlations are unstable over time, so we restrict ourselves to going forward 12 months. From there, we can calculate the downside risk of each portfolio and the total fund with the future picture of ‘Worst Expected Outcome’ to be correct in 95 years out of 100.

At our 95 per cent confidence level, we will be more than meeting the ‘Prudent Person’ requirements and dealing with probabilities at a generally accepted level. Readers keen on watching their pennies should note that historic standard deviation calculations will not give you an equivalent future scenario, so we have to go to the lengths we do in the calculations – so much for the traditional backward looking risk (volatility)/reward quadrants.

Out of all that work, we will come up with the value-at-risk of your total fund (the maximum expected loss for your portfolio for the next 12 months). We can’t give you a rate of return expectation with the same degree of confidence, but we can take the long-term historic rates of return for each asset class as some indication of the relative ranking of asset returns. Input them with value-at-risk (instead of historic standard deviation or volatility) in a traditional optimizer, and the pension fund committee will have the asset mixes most appropriate for their downside risk tolerance going forward. Thus, if you can’t manage losses greater than, say, five per cent in any one year, here are the asset mixes best for you without putting your money unproductively under the mattress.

As you can see, the concept is uncannily simple. Instead of navel gazing and hoping for the best, the pension fund committee can act every year as a ‘Prudent Person’ to minimize losses and capitalize on compound interest.

Vaino Keelmann is a principal and consultant at API Asset Performance Inc.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Subscribe to Daily News Alerts

Subscribe now to receive industry news delivered to your inbox every business day.

Interactive issue now onlineSubscribe to our magazinePrivate Wealth Online