The 4% retirement withdrawal rule has enjoyed a ubiquitous presence in the world of financial planning ever since William Bengen proposed it in 1994. Mr. Bengen's rule states that individuals should withdraw no more than 4% of their retirement savings annually to ensure they will not outlive their assets.
His work came in response to fears that the prevailing rule of thumb at the time, 5%, was too risky. Today, some even argue that the 4% rule is still not safe enough and that 3% is the way to go.
But is this even relevant? What if this guidance is based on assumptions about retiree behavior that are not accurate?
A 2013 study, The Drawdown of Personal Retirement Assets, found that between ages 60 and 69 (i.e., after penalty-free retirement withdrawals begin and before required minimum distributions kick in at age 70 ½) only 17% of retirees made retirement account withdrawals annually. Forget the withdrawal rate: A clear majority of retirees did not withdraw any money at all.
Of course, that changes once RMDs become mandatory. By age 71, 60% of retirees were taking withdrawals, and this percentage increased with age. However, there is ample evidence to suggest that the distributions were taken to comply with the law rather than out of necessity. How do we know? Well, when RMDs were suspended in 2009 as a part of the fiscal stimulus package, withdrawal rates for those between ages 72 and 85 dropped by 15 percentage points — a behavior hardly consistent with any withdrawal rule.
I've noticed this behavior in research I conducted for the Employee Benefit Research Institute last year (Asset decumulation or asset prevention? What guides retirement spending?). The research found that in the first 18 years of retirement, households that started retirement with less than $200,000 in non-housing assets experienced only a 25% drop in their assets. Why? Perhaps households with fewer assets wanted to hold on to that money for medical expenses or other emergencies.
But what about people with more assets who could afford to spend a little more? They spent even less. Retirees who started with more than $500,000 in non-housing assets spent down about 12% on average in the first two decades of their retirement. Since people with higher income and more assets are expected to live longer, perhaps this group put the brakes on spending down due to the fear of outliving their money.
If that's the case, then it may be safe to expect higher spend-down among people who have guaranteed income for life, such as participants in defined-benefit pension plans. But guess what? Pensioners spent even less — only 4% of their non-housing assets in the 18 years after retirement.
Irrespective of where retirees started with their assets, about one-third had more assets after nearly two decades of retirement than they had at the beginning of retirement.
So what does all the data tell us?
There is a lot of evidence to suggest that it's highly unlikely retirees plan to spend down their assets in a manner we assume they will. If anything, the data suggests retirees are intentional about not spending down their assets as long as they can. But if retirees are not following any rule or strategy such as Mr. Bengen's 4% rule, how are they deciding how much to draw down from their assets?
A driving factor for retirees is most likely the preservation of principal. Retirees' behavior shows they prefer to leave their principal intact and will prioritize spending down any income they receive on an ongoing basis first, such as Social Security income, pension income and the income their assets generate.
There could be several reasons why retirees are hesitant to spend down their assets — both financial and behavioral. Financial reasons could include preserving assets for uncertain medical needs late in life, particularly long-term care needs, or hedging against longevity risk.
The desire to leave a bequest for heirs could also come into play. Behavioral reasons could include the inability to suddenly become a spender after building up a lifelong habit of saving, or the challenge of decreasing an account balance after getting comfortable increasing it during one's working life.
How should retirement advisers use this information to build clients' retirement income plans? The short answer is: Assume nothing, and ask clients what they want.
Some of the relevant questions to ask may include: How comfortable is the client with drawing down principal? How flexible are they with adjusting their consumption to avoid spending down principal? Finally, what do they consider to be a financially successful retirement?
Most retirement models assume if people have a dime left in their accounts at the time of their deaths, it's a financially successful retirement. But I highly doubt someone in that situation would consider that a success. By asking these important questions, advisers can gain the insight necessary to guide clients toward rewarding retirement years.
(More: An update on the 4% rule)
Sudipto Banerjee is a senior manager of thought leadership at T. Rowe Price.