Financial-planning tools largely assume retirement spending is relatively predictable, and that it increases annually with inflation regardless of an investment portfolio’s performance. In reality, retirees typically have some ability to adapt spending and adjust portfolio withdrawals to prolong the life of their portfolios, especially if those portfolios are on a declining trajectory.

Our latest research on the perceptions of retirement-spending flexibility provides evidence that households can adjust their spending, and that adjustments are likely to be less cataclysmic than success rates and other common financial-planning-outcomes metrics imply.

This suggests that spending flexibility needs to be better incorporated into the tools and outcomes metrics that financial advisers use to advise clients.

Flexible and essential expenses
Investors are often flexible on their financial goals. For example, a household’s retirement liability differs from a defined-benefit (DB) plan’s liability. DB plans provide a fixed sum for employees at retirement.

While DB plans have legally mandated or “hard” liabilities, retirees typically have significant control over their expenses, which could be perceived as “soft” to some extent. This is important when applying different institutional constructs, such as liability-driven investing (LDI), to households.

Most financial-planning tools today still rely on the static modelling assumptions outlined in William P Bengen’s original research. This results in the commonly cited “4 per cent rule”, in which spending is assumed to change only due to inflation throughout retirement, and does not vary based on portfolio performance or other factors.

The continued use of these static models may primarily be a function of their computational convenience. But it could also be due to a lack of understanding on the nature of retirement liability, or the extent to which a retiree is actually comfortable adjusting spending as conditions dictate.

In a recent survey of 1,500 defined-contribution (DC) retirement plan participants between the ages of 50 and 70 in the US, we explored investor perceptions of spending flexibility. We found that respondents were much more capable of cutting back on different expenditures in retirement than the conventional models suggest. The sample was balanced by age and ethnicity to be representative of the target audience in the general population.

According to traditional static-spending models, 100 per cent of retirees would be unwilling to cut back on any of the listed expenditures. In reality, though, respondents demonstrate a relatively significant ability to adjust spending, with notable variations across both expenditure type and households.

For example, while 43 per cent of respondents would not be willing to cut back on healthcare at all, only 6 per cent would say the same about clothing. In contrast, certain households are more willing to cut back on healthcare expenditure than vacations.

A spending cut’s potential cost may not be as severe as traditional models imply. For example, models generally treat the entire retirement-spending goal as essential. Even small shortfalls are considered “failures” when the probability of success is the outcomes metric. But when we asked respondents how a 20 per cent drop in spending would affect their lifestyle, most said they could tolerate it without having to make severe adjustments.

For example, only 15 per cent said that a 20 per cent spending drop would create “substantial changes” or be “devastating” to their retirement lifestyle; 40 per cent said it would have “little or no effect” or necessitate “few changes”. Retirees appear to be far more sanguine on a potential reduction in spending than traditional models would suggest.

The clear ability to cut spending in the first chart, and the relatively small implied potential impact on retiree satisfaction or utility in the second – at least for a relatively small change in spending – have important implications when projecting retirement-income goals.

While understanding each retiree’s spending goal at a more granular level of expenditure is important, so too is having a sense of the amount of spending is “essential” (needs) versus “flexible” (wants) when mapping out assets to fund retirement liabilities.

Spending flexibility is critical when considering the investment portfolio’s role in funding retirement spending. The portfolio could be used to fund more flexible expenses, which are a very different liability than what is implied by static-spending models. Those suggest that the entire liability is essential.

Overall, our research demonstrates that retirement spending is far more flexible than implied by most financial-planning tools. Retirees have both the ability and the willingness to adjust their spending over time.

That is why incorporating spending flexibility can have significant implications on a variety of retirement-related decisions, such as the required savings level – broadly lower – and asset allocations. Generally, more aggressive portfolios may be acceptable, and certain asset classes become more attractive.

The writer, PhD, CFA, CFP®, is managing director and head of retirement research for PGIM DC Solutions.