Retirement possibilities

By Gerard Hass

CAs are well positioned to provide retirement planning projections, but as with any projections, beware garbage-in, garbage-out

A common theme in retirement projections seems to be that financial planners are often left to guess when estimating the many variables that go into the projections when planning for clients’ retirement and other long-term events. Often they follow statistics and performance numbers that represented periods in the past without applying any forward long-term thinking or understanding the anomalies that may have occurred in the various markets.

For example, many planners and clients may believe that future bond returns will reflect the past 25 years and that Canadian bank stocks are always a sure bet for long-term protection and performance. Yet it was the anomaly of high inflation and interest rates rising to unprecedented levels in the 1970s and early 1980s that gave a tailwind to bonds and financial stocks that lasted 20 years.

With interest rates and inflation back to levels experienced in the decades prior to the 1970s and with aging demographics suggesting the greater need for safety and income, there is less likelihood of investors seeing the double-digit and high single-digit fixed income returns they saw in the past 25 years.

When preparing a retirement projection, there are a number of assumptions and items that need to be considered.

Life expectancy and the planning horizon
Life expectancy is the key determinant when deciding the planning horizon you should be projecting. From an actuarial point of view, a person’s life expectancy is conditional upon their age. The older you currently are, the more likely you’re going to live to a later age. By definition, life expectancy is an estimate of an age beyond which you have a 50% chance of living.

Odds of 50% may not be accurate enough for planning purposes and to complicate matters, the average life expectancy of the Canadian population (both male and female) has been growing each decade. We are expected to live longer than ever, so to use a person’s life expectancy as a planning horizon would likely result in a substantial number of irate clients living beyond the planned period. They may not be irate at being alive, it’s just that they would have little money left to live on.

To avoid the difficulties that may arise with longevity (such as having half your client base living in your basement), the planning horizon needs to be a term longer than the person’s life expectancy. A safer approach would be to use an estimate of the age in which the client has about a 10% chance of living longer. As with life expectancy, the older the client, the longer the planning horizon needs to extend.

Table 1 on page 47 demonstrates the life expectancy and planning horizons of Canadians based on 2005 Statistics Canada data. Other factors that should be considered include the client’s marital status (marriage usually lengthens the time horizon although some would argue that logic dictates the opposite), their gender, longevity of other family members and any medical issues that may be relevant.

TABLE 1

Source: Statistics Canada Life Table 2005

As a rule of thumb, clients aged 65 and younger should have a planning horizon to about age 91 if male, 95 if female, and every five years of age should add one extra year to the planning horizon. Thus, a client aged 70 should have a planning horizon to age 92 for men and 96 for women as a starting reference.

Inflation
The inflation estimate is a cornerstone variable in retirement projections. Not only does inflation erode the purchasing power of clients’ money and hence require them to use increasing amounts of their capital to live but it also affects the rate of return they earn on their investments.

The effects of inflation can be substantial. For example, a $100 basket of goods will cost $192 in 30 years’ time if inflation averages 2% annually. Planners must consider the effect of inflation when determining income flows and any estate capital requirements.

TABLE 2

Source: Statistics Canada

The table above shows the inflation levels, as measured by the Canadian consumer price index for various periods. The level of inflation that occurred in the 1970s through the 1980s was higher than what had been experienced up to that point in time. With the world central banks committing to a program of financial stability in the early 1990s, inflation has averaged approximately 2% in the last decade, as it had from 1924 through to 1973.

While central banks continue to stay committed to stable pricing, planners must keep focused on central bank rhetoric in order to ensure their plans reflect reasonable inflation expectations.

Investment returns, whether from fixed income or equity investments, are influenced by interest rate levels that in turn are affected by existing and estimated inflation levels. A rule of thumb used by planners is that short-term fixed income earns 3% above inflation, long-term bonds about 5%, and equity investments average about 7% above inflation over the long term.

A check of the 20-year, rolling real Canadian bond returns from 1943 to 2005 suggests the 3% to 5% rule of thumb is inaccurate. A better one would be to use the current spread on government of Canada Real Return Bonds, which is currently about 1.6% above inflation. The equity rule of thumb however seems to be very resilient. In his books Stocks for the Long Run and The Future for Investors, Jeremy Siegel shows the equity rule of thumb for US securities to be applicable still. Canadian data seems to suggest a 5% real return is more representative of the Canadian stock market. Remember that the rules of thumb are based on long-term averages and on index returns that may not be appropriate.

A 2% rule of thumb for fixed income and 6% for equities is a good starting point. Regardless of the rate of return you use in a retirement projection, the important thing to note is that inflation will have long-term effects on your plan.

REAL RATES OF INFLATION

Source: Bloomberg

Rates of return
Perhaps the most important variable in a retirement projection is the rate of return projected for the pre-retirement and retirement periods. This is because a small difference in estimate can have a large impact on the portfolio value over a long period of time. In this estimate many planners have been a little exuberant possibly because of the previously mentioned old rules of thumb. In addition to inflation, a client’s rate of return will be affected by the client’s investment objectives and risk tolerance.

If we learned anything from the last bear market, it would be that a person’s risk tolerance should be the dominant factor in deciding an asset mix. The asset mix in turn will be the dominant factor in determining the expected rates of return the client will earn. The more conservative the client, the lower the expected long-term return.

When deciding on a projected rate of return, there are a number of points that warrant attention:

• Return projections should be made in at least two stages — pre-retirement and retirement. This will account for people getting more conservative as they age and their returns will likely change (i.e., being lower in their later years) because of this.
• In the long run, stocks have outperformed bonds, which in turn have outperformed short-term cash investments. The greater performance comes at a price though — higher short-term volatility and loss of guarantees.
• Fees are a fact of life. Most clients deal with professional money managers or with investment advisers who in turn make use of professional money managers. Many managed products don’t outperform their benchmark indices partly because of the fees. But fees are just one of the reasons clients don’t achieve long-term returns suggested by the indices.
• Another reason clients don’t achieve index-like returns is that they, the investment adviser and the portfolio manager are likely haunted by behavioural demons. Chasing performance, having regret for short-term losses and isolating accounts rather than viewing them as a combined entity are just some of the behavioural problems investors may have. These typically result in clients earning lower returns than what the market indices achieve.
• Taxes are a fact of life. Capital gains are currently the lowest-taxed form of income and have a positive effect where tax benefit reductions may exist such as the GST credit and Old Age Security clawback. Canadian dividends are subject to a low tax rate, but their required gross-up may have a negative impact on other tax benefits. In light of the recent tax changes further increasing the dividend gross-up, the old rule of keeping fixed income assets in registered accounts and equity investments outside of registered accounts needs to be rethought. There are different degrees of tax efficiency and volatility amongst managed products. Most advisers recognize this and, recognizing their client’s behavioural issues, take the level of volatility and tax-efficiency into consideration when making recommendations. These features may impact returns.
• There is some evidence that stock markets appear to move in 18-year cycles (a growth cycle and an accumulation cycle). Where we are within the cycles may have an impact on the rate of return you will want to project.
• A study prepared by SEI Investments in the late 1990s found that the greatest contributor to added value in fixed-income investing is duration management, i.e., the lengthening and shortening of bond maturities based on interest-rate forecasting. If an investor was not capable of forecasting it would be less likely that he or she would be able to perform better than the index.
• Using historical returns of popular indices as a proxy for estimated returns can lead to overstating expectations as many funds do not outperform the indices in the long run. By the nature of being diversified, an investor is likely to achieve average returns. Fixed-income funds, such as bond funds, have an even harder time of outperforming their respective indices and recently Canadian equity funds have had little success in beating the TSX composite index.

A revised rule of thumb for fixed income investments, 2% above inflation and a 5% real return on equity investments, is more appropriate when preparing projections. Therefore, on a 50% fixed income, 50% equity asset mix, a real return of 3.5% would be an appropriate return estimate.

(An index the author developed for evaluation of managed portfolios, representing an average diversified portfolio, achieved an average annual return of 6.2% over the past 20 years, while CPI has averaged 2.6% over the same time. The net real return of 3.6% is close to the revised rule of thumb applications and suggests they are a more likely achievement for clients.)

Conclusion
There are many considerations to make when generating retirement projections for clients. Being conservative when estimating planning horizons, inflation and rates of return is a key element to the effectiveness of the projection. By questioning the rules of thumb that may have applied in the past and replacing them with forward-thinking, achievable, revised estimates, we will better serve our clients in their quest for a better life after work.

Gerard Hass, CA, CFP, CFA, is a portfolio manager with Raymond James in Hamilton. His views don’t necessarily reflect those of Raymond James. This article is for information purposes only. Statistics are from sources believed to be reliable, but accuracy can’t be guaranteed. His website is www.WealthTrust.ca

Technical editor: Ian Davidson, MBA, CFP, CA, RFP, vice-president, Assante Capital Management Ltd.