The art and science of portfolio construction

The theory is based on the notion that when some asset classes are doing poorly, others will do well. These divergent returns offset one another, limiting the portfolio’s downside. Avoiding losses leads to larger overall returns over the long term because future gains are compounding on a larger starting amount (see page D5).

Assembling a well-diversified portfolio is a bit like putting together a winning hockey team. A team composed solely of high-scoring talents rarely wins on a consistent basis. While it may provide plenty of offence, it doesn’t offer much in the way of defence. Rather, the best teams have a mix of scoring talent (high-risk, high-reward players); solid two-way players who don’t score much, but who don’t give up many goals against, either; and defensive stalwarts whose sole job is to prevent goals against.

In the investment world, it’s much the same. Betting on a single hot stock or fund may produce some fantastic short-term gains, but it also exposes the investor to the risk of heavy losses should that hot asset cool and drop sharply in value. By assembling a mix of risk/return profiles, an advisor can give a client a larger return for a given level of risk over the long term.

As with any investment process, the first step is “know your client.” The challenge of constructing a successful portfolio starts with a full understanding of the client’s investment goals, risk tolerance, time horizon and other individual circumstances. Nailing down precisely what clients want to achieve — saving to fund their children’s education, looking to finance a far-off retirement or augmenting their current income — is critical. If, for example, they have goals that demand greater risk-taking but they aren’t willing to live with much risk, then expectations must be altered at the outset.

Within these basic parameters, the challenge is to build a portfolio that delivers the best possible risk/reward trade-off. Asset allocation has significant influence over performance, and diversification is critical to maximizing the returns a portfolio delivers for the amount of risk it takes on (see page D4).

That basic mix starts with different asset classes. In the world of investments, equity securities are the offensive superstars; they carry the highest expected return but also hold the greatest risk. Fixed-income instruments are the solid two-way players, with lower returns but also lower risk. And cash is the defensive specialist; your clients almost never lose money in cash, but they aren’t likely to make much, either.

The historical returns of different asset classes in the Canadian markets bear out this risk/reward trade-off theory. Canadian treasury bills have returned, on average, 6.8% a year over the past 45 years, according to data from BMO Nesbitt Burns Inc. In that time, T-bills have never had a negative year, with annual returns ranging from 1.7% to 20.8%.

Bonds have delivered a better average annual return over the same period — 8.3% — but the asset class has also experienced three years with negative returns. Their worst 12-month period was in 1981, when bonds lost 7.9%. Their best period was just a couple of years later, when they gained 40.8%.

Stocks are another step up from bonds in the return department, averaging 10.5% annually over those 45 years. However, the risk is greater; equities have had 13 down years — losing as much as 39.2% in 1982, and gaining 86.9% the following year. This volatility is exactly what gives equities their riskier reputation, although over time, they generate larger average returns.

These historical data put the different risk/return profiles of various asset classes into concrete terms, but it’s important to recognize that the future will almost certainly be somewhat different from the past. Notably, returns across all asset classes may well be lower.

@page_break@For one, the historical data include some extreme periods of high inflation and high interest rates. While it would be foolish to declare inflation dead, the fact that the Bank of Canada has focused on maintaining price stability suggests that these periods of runaway inflation are less likely in the future.

Moreover, the profile of the Canadian population and the economy is changing. As the population ages, labour force growth will slow; economic growth is expected to be lower, so returns from financial assets are likely to moderate.

Economists at TD Bank Financial Group recently predicted long-run return expectations for the three major asset classes, based on their forward-looking economic projections rather than historical experience. TD forecasts the average annual return from cash will be 4.4%; bonds should deliver about 5.6%; and equities will return about 7.5%.

Whether these forecasts ultimately are accurate, the relative risk/return profile of the different asset classes can be expected to persist. So the question is: what proportion of the portfolio should be allocated to which type of asset, in view of the client’s needs and risk tolerances?

In general, a portfolio that’s heavy with equities can be expected to generate higher long-term returns but with higher volatility, so these returns may not be available in the shorter term should the client need cash. Clients who can’t afford to lose money in the short term will have a stronger tilt toward cash and bonds. Investors with a long time horizon and a healthy appetite for risk should be happy with plenty of equity weight in their portfolio.

While setting this asset mix for an individual client is a unique circumstance, basic model portfolios suggest a starting point for different types of investors. For example, Nesbitt Burns’ portfolio strategy report considers three different asset mixes for three different investor profiles. The income portfolio is 5% cash, 70% bonds and just 25% equities. The balanced portfolio is 5% cash, 45% fixed-income and 50% equities. The growth portfolio is 5% cash, 25% fixed-income and 70% equities.

Depending on the circumstances, it may be appropriate to tilt allocation even more dramatically. TD’s most conservative, income-oriented portfolio puts 20% in cash, 50% in fixed-income and the remaining 30% in equities. It projects that, based on its return forecasts, such a portfolio would return 6% annually.

Expected returns increase as the asset mix shifts to take on more risk. TD predicts a slightly riskier mix of 10% cash, 45% fixed-income and 45% equities would return 6.4%. Taking on yet more risk — with 5% cash, 35% fixed-income and 60% equities — would have an expected return of 6.7% a year. An even more aggressive stance — no cash, 20% fixed-income and 80% equities — would generate a 7.2% annual return. Finally, TD projects, a 100% equities portfolio would deliver a 7.6% annual return.

Once the basic asset mix is set, the next step is to diversify within the individual components of the portfolio. The same logic prevails within asset classes as among them; by diversifying among different securities, investors can achieve higher returns at a given level of risk. This sort of diversification is realized by selecting assets whose returns are not correlated.

Whether your weapons of choice are individual stocks, managed products (mutual, segregated or pooled funds), exchange-traded funds or other securities, the overriding goal should be diversifying returns. Any number of basic signifiers can indicate the opportunity for uncorrelated returns — this may mean diversifying by geography, sector, style or market cap. If tech stocks are slumping, perhaps financials will be doing well. When the Canadian markets are sliding, Europe may be buoyant. Growth could be thriving while value slumps, and small-caps may be bouncing when large-caps are struggling.

Beyond that, depending on the size of the portfolio, it may be worth considering so-called “alternative” asset classes. Assets such as hedge funds, private equity and commodities historically have generated returns that aren’t correlated with the major equity markets. While they may be cost-efficient only in bigger accounts, in the right circumstance they can provide still more diversification.

Another fundamental consideration in portfolio construction is cost. Management costs can take a significant bite out of returns. There are managers that justify their cost with their performance, and managed products may be the only cost-effective way to gain exposure to some of the more esoteric asset classes. But advisors should be sensitive to costs, particularly when they are looking simply to add exposure to a market. Index funds and ETFs may be able to deliver that return at a lower cost than an active manager (see pages D6-D7).

Of course, making all these asset mix and security selection decisions is not an end in itself. While a well-diversified portfolio that delivers good returns for a given level of risk is the goal, an investment portfolio is primarily a means to an end, and that end is often a moving target.

Clients may want different things from their portfolios at different stages of their lives. A young family’s needs are often quite different from an empty nester’s or those of a couple on the verge of retirement. As clients move through these stages of their lives, and their needs change, their portfolios must evolve with them (see page D8).

Moreover, asset class correlations aren’t carved in stone. Markets that are typically not correlated may start moving in sync in response to an event, particularly a negative event such as a global recession. Ensuring diversity in a portfolio isn’t a one-time decision. Effective portfolio construction is partly art, partly science, and always an ongoing challenge. IE