The article “Why You Should Think ‘Smile’ When Planning Your Retirement Spending Path” was originally published on Financial Post on January 4, 2017.

Rudimentary retirement rules of thumb are prevalent in a financial industry that prefers to spend more time focused on stock picking than retirement planning. On that basis, it is incumbent upon aspiring Canadian retirees or those who are already retired to make sure they understand what will happen to their spending in retirement — and how to plan for it.

Financial planners spend much of their time trying to help people answer the question: how much do I need to retire? Often, the answer is that it depends. It depends on the length of retirement, investment returns, tax rates, inflation and countless other factors. With the exception of retirement length — since it is difficult to know how long we will live with any certainty — spending is by far the most influential piece of the retirement puzzle.

Early retirement spending theories like Modigliani and Brumberg’s 1954 life-cycle hypothesis (LCH) still have a strong influence on retirement rules and financial planning tools. The LCH presumption is that people will have stable spending during their lives, planning their saving and retirement in such a way to accumulate and decumulate assets from the cradle to the grave smoothly.

The theory behind life-cycle spending appeals to the economist in me, but in practice, people spend more variably. So it is important to look at studies of real life experience.

Countless academic analyses have shown that declines in spending are much more modest than many might otherwise expect. Ameriks, Caplin and Leahy (2007) found that while U.S. pensioners anticipated spending would drop by 11 per cent on average upon retiring, the actual average decline was only 5 per cent when they were later polled during retirement. Hurd and Rohwedder (2008) also used U.S. data and determined that spending falls by 1 to 6 per cent, varying depending based on the measure.

Internationally, Italian research by Miniaci, Monfardini and Weber (2003) and Battistin, Brugiavini, Rettore and Weber (2007) found similarly modest declines in spending.

A study from Christensen (2004) found no decline in spending in Spain.

In a much less scientific career-long study, Heath (2002-2016), I have worked with thousands of clients and digested volumes of retirement planning observations. I find that most people expect much larger declines in spending in retirement than the studies suggest. I generally encourage people to assume their spending on living expenses (excluding taxes, mortgage payments, pension contributions, children’s education, etc.) does not decline in retirement. The academic evidence is supportive of this level of conservativism.

If we focus on Canadian data, a 2015 BMO Financial Group study found that average Canadians spend $28,800 per year. Retirees in Atlantic Canada were found to spend the least ($23,700) and those in Ontario spend the most ($30,408). A 2016 Sun Life survey found the average retiree budget in Canada to be $31,332. But little to no Canadian data exists to show trends in spending in retirement.

One of the more interesting retirement spending observations I have seen comes from a recent article in the Journal of Financial Planning entitled “Exploring the Retirement Consumption Puzzle”. According to author David Blanchett of Morningstar Investment Management, although the U.S. “retiree consumption basket is likely to increase at a rate that is faster than general inflation — a fact that can largely be attributed to the higher weight to medical expenses for retirees — actual retiree spending tends to decline in retirement in real terms.”

For clarity, Blanchett’s research suggests that spending increases by about 1 per cent less than the rate of inflation, so by about 1 per cent per year in a 2 per cent inflation environment. That is, it effectively decreases over time when adjusted for inflation.

He has also studied an intuitive phenomenon called the “retirement spending smile,” whereby expenses often decline during the early years of retirement, before reaching an inflection point and rising in the second half of retirement — in the shape of a smile. If someone has the fortune, or the financial misfortune, of living a long life, spending tends to rise by more than the rate of inflation in later years on average. This makes sense, given how few 95-year-olds cut their own grass, live independently in their home or avoid prescription drugs.

The retirement spending smile trend and the real life retirement spending data are important for Canadians who are approaching retirement or already retired. The consolidation of information suggests that spending may not decline significantly upon retiring, but that over the course of retirement, spending may decline modestly in the early years. Planning for rising costs in your 80s and 90s may then be more important than planning for long walks on the beach in your early retirement. This risk is particularly pertinent for the one-quarter of Canadian retirees who do not own a home that could otherwise be considered a partial insurance policy on living too long and sold to fund long-term care costs.

Retirees and retirement planners alike should take note of academic research on spending trends. It may have an impact on investment portfolio construction, product selection and ultimately on retirement budgeting assumptions.

While you may not be the average Canadian and the dollar amount of your annual spending may be considerably more or less than other typical retirees, it is not unreasonable to assume that the trajectory of your spending will decline in inflation-adjusted terms particularly during your 70s. It is important to plan for expenses to increase in your 80s in case your own retirement spending trend looks like a smile, so as to ensure a happy retirement.

Jason Heath is a fee-only Certified Financial Planner (CFP) and income tax professional for Objective Financial Partners Inc. in Toronto, Ontario.