By Fraser Stark

We apologize, but this video has failed to load.

Try refreshing your browser, or
tap here to see other videos from our team.

Beyond the 4% rule: Improving your safe retirement withdrawal strategies Back to video

We apologize, but this video has failed to load.

Try refreshing your browser, or
tap here to see other videos from our team.

Play Video

The four per cent rule. The 3.3 per cent rule. The 2.26 per cent rule. Whatever your number, over time, these prescribed income level rules of thumb seem to point to lower — and more precise — values.

They all strive to answer the same challenging, timeless question: How much can I safely withdraw from my retirement portfolio each year without the risk of running out of money?

“Running out” is seen as a clear failure, and correctly so. But the premise of these rules is that the opposite — not running out — constitutes success. This is where the logic behind these rules begins to fray.

Evolved thinking around the methodology, updated long-term macroeconomic forecasts and more sophisticated modelling tools are changing how experts evaluate these rules. But honing in on the “correct” value misses the point: the entire premise of holding a basket of assets and drawing from it blindly is a suboptimal approach that often leads to inefficient outcomes for retired investors.

Origins of the 4% rule

Financial adviser Bill Bengen’s seminal 1994 paper arrived at a safe withdrawal rate of four per cent by back-testing various withdrawal levels against historical market return data back to the 1920s. His analysis determined that an investor who started spending four per cent of their original portfolio value and raised the withdrawal rate by three per cent annually for inflation would have not fully depleted their balanced portfolio over any 30-year period.

Bengen’s approach was to trial-and-error using historical data, but it rests on a simple theoretical foundation.

If an investor wants to be assured they can withdraw an income each year, held constant for inflation, no matter how long they live, they must maintain their account balance at a real (inflation-adjusted) level. For example, if they start with $1 million and draw a rising amount from it each year, after a number of years, they will need to have more than $1 million or the rising withdrawal amounts will begin to rapidly deplete their assets.

In order to hold their account balance at a constant “real” level over time, on average, they can only withdraw their real returns: their expected portfolio returns less inflation. While there is some year-to-year variability, portfolio returns over the medium term are likely to average six to seven per cent, and inflation might average two to three per cent. That means the real expected returns — what can safely be withdrawn for several decades — is around four per cent of the initial balance.

Three major flaws with the 4% rule

Withdrawing an amount set well below a portfolio’s expected return may seem prudent, but there are a number of serious flaws with using this approach to determine how much to draw from a portfolio in retirement.

First, the rule fails to account for retirees’ ability to adapt and adjust their spending. People can and do adjust their spending up and down throughout their lives due to changing income or market conditions. Yet this rule rests on the assumption that the investor rigidly holds to a fixed spending level and will not adapt to what’s going on in their portfolio or the costs of the items they buy. It is truly set it and forget it, which is not how people behave.

The rule was also evaluated over a 30-year time horizon. We know that for many retirees today, there’s a reasonable probability that they will live another 35 to 40 years. For example, mortality tables used by many financial planners indicate a greater than 34 per cent chance that a 65-year-old woman lives more than 30 years.

Finally, no one claims the rule always works. The analysis only notes that this approach seems to “not fail” when evaluated over a recent block of market history. It’s based on historical returns data from the United States during the 20th century, which is now known to have been an extraordinary period of productivity gains and wealth creation in the world’s most dominant economy. Is the future certain to mirror the past? Of course not.

Real-life implications

One notable byproduct of following this approach is that an investor should expect to leave to their estate an amount approximately equal to the “real” value of the starting retirement account. For some people, this might nicely align with their personal preferences and wishes, while it might be of little value for others and create suboptimal outcomes, effectively obligating them to leave a sizable estate even if that’s not their intent.

Imagine a couple who invested and sacrificed for their children’s education. Their adult children are now financially successful in their careers and don’t need a large inheritance, while the retiring parents have fewer financial resources than they might have otherwise to fund their retirement.

A knock-on effect is that by choosing to “self-insure” against the small chance of living a very long life, an investor will spend less every year from the very beginning of their retirement. Much like the estate size question above, this constrained spending is of little consequence for some investors — think here of the very wealthy — but it may meaningfully limit others’ comfort in retirement and their ability to hit key life goals.

Since optimizing sustainable income to support their desired lifestyle in retirement is the paramount goal for many people, these “X” per cent rules immediately fail to be the best solution for them.

A better approach?

A more effective approach is to annuitize a portion of your assets at retirement, thereby creating a stream of sustainable income and withdrawing from the rest of your portfolio according to your percentage rule of choice. There are several methods to accomplish that.

One is to delay the start of Canada Pension Plan (CPP) and Old Age Security (OAS) payments, which, in effect, allows you to “buy” more into the pension. This income is highly secure, will last your whole life, is indexed formally for inflation and should partially flow to your spouse.

Another option is to buy a lifetime annuity from an insurance company. This income will last your whole life, may partially flow to your spouse and is guaranteed to neither rise nor fall. Not falling is nice, of course, but not rising means it will lose purchasing power as inflation raises costs year after year. This adds up over 30 years, by the end of which you should expect prices to be two to three times higher than today.

The final option is to invest in a lifetime income fund, creating income that will last your whole life. While the level of lifetime income is variable from year to year (that is, not fixed), such products apply a degree of longevity protection to a portfolio and help match the total asset (how much money an investor will have) with the total liability (how much money an investor will need).

In the case of the Longevity Pension Fund by Purpose, this income can be expected to rise over time, (though it is adjusted annually and can go up or down), and similar attributes apply to many such structures in Canada and globally.

An investor must first decide what overall income level they would like to draw in retirement to cover both non-discretionary (needs) and discretionary (wants and wishes) spending. They can then decide how to allocate their portfolio across various asset classes. If withdrawing three to four per cent annually is sufficient, that can likely be drawn from a traditional balanced portfolio, adhering to the four per cent rule (or something similar).

But for those seeking more than four per cent, an allocation to a lifetime income fund can help. For example, a 33 per cent allocation to the Longevity Pension Fund, yielding a little more than seven per cent in October 2023, allows an investor to draw five per cent from the overall portfolio: seven per cent from the third in the Longevity fund and four per cent from the two-thirds in a traditional balanced portfolio. This 25 per cent increase — from four per cent to five per cent — can have a substantial impact on what life goals can be achieved in retirement while still maintaining a healthy portion of the investor’s net worth for their estate.

Note also that investors will want to consider how much guaranteed income they would like to have. Since most Canadians have some guaranteed and inflation-hedged income via CPP and OAS payments, some degree of variability in the balance of their income portfolio is tolerable.

For anyone seeking additional guaranteed income, life annuities could play this role in raising the spending level above four per cent, adding the benefit of guarantees, but at the cost of no expectation of rising income and lower flexibility.

Perhaps coincidentally, it’s now been nearly 30 years since Bengen’s paper concluded that a balanced portfolio could sustain a four per cent withdrawal rate, rising with inflation, for 30 years without failing. Yet this rule has failed to demonstrate its optimality, and many investors will do better with more bespoke and outcome-oriented strategies.

Much has changed over those three decades. In the face of rising living costs, greater macro uncertainty and continued innovation in financial product design, an optimal outcome for many investors can be achieved by more thoughtfully constructing an initial portfolio to meet their desired outcomes, and by dynamically responding to market and life conditions as the retirement phase unfolds. We deserve no less.

Fraser Stark is president of the Longevity Pension Fund at Purpose Investments Inc.

Bookmark our website and support our journalism: Don’t miss the business news you need to know — add financialpost.com to your bookmarks and sign up for our newsletters financialpost.com.