Popular withdrawal strategies, such as the 4% rule, treat spending as rising throughout retirement. These strategies increase the absolute size of withdrawals in accordance with the rate of inflation. The logic is to tie replacement income to inflation. In doing so, the strategies seek to prevent a retiree’s purchasing power from diminishing.

What these strategies fail to consider is that actual spending patterns evolve throughout retirement. A growing amount of data shows retirees changing their spending patterns over time. Some studies show spending dropping gradually as a retiree ages. Research by David Blanchett of Morningstar suggests more of a “smile” pattern: real spending declines a little early in retirement, accelerates its decline during the middle point of retirement before slowing its rate of decline late in retirement. Behind the changes are lesser spending on discretionary items as a retiree ages and higher spending on health care expenses late in life.

One strategy for accounting for such spending patterns would be to project retirement spending to decrease by 10% each decade (e.g., at age 70, at age 80 and at age 90). A key problem with such an approach is that individuals age differently. The timing of health care expenses can be outside of forecast ranges. As such, forecasts based on this type of strategy could be off by several years, as financial planner Michael Kitces points out.

An alternate approach would be to project retirement spending to increase by 1% less than the rate of inflation used to increase withdrawals. The strategy assumes retirement spending will decline, but without the big changes of the first strategy.

Kitces points out that decreases in retirement spending actually tend to reflect reductions in leisure and discretionary spending. Such decreases do not fully offset increases in health care expenses. He argues this tendency makes a category-based spending approach a better strategy. Spending on leisure and travel could be assumed to decline at faster rates per each decade of retirement, while health care spending could be projected to increase during each decade. The upside of this approach is that both category spending and the rate of inflation expected for each category are adjusted at different rates.

Source: “Kitces: A Better Baseline for Retirement Planning,” Michael Kitces, Financial Planning, December 30, 2016.