Bad Retirement Spending Plans
A recent research paper by Chen and Munnell from Boston College asks the important question “Do Retirees Want Constant, Increasing, or Decreasing Consumption?”. The accepted wisdom until recently was that retirees naturally want to spend less as they age. This new research challenges this conclusion.
What we all agree on is that the average retiree spends less each year (adjusted for inflation) over the course of retirement. However, averages can hide a lot of information. The debate is whether this decreasing spending is voluntary or not. However, it’s important to recognize that the answer is different for each retiree. Some don’t spend less over time, some spend less voluntarily, and some are forced to spend less as their savings dwindle.
I’ve been saying for some time that not all spending reductions by retirees are voluntary and that this affects the average spending levels across all retirees. I’ve discussed this subject with many people, including a good discussion with Benjamin Felix, who was good enough to point me to the new Chen and Munnel research. (Larry Swedroe also discussed this research.)
Research Findings
“On average, household consumption declines about 0.7-0.8 percent a year over retirement. However, consumption for wealthy and healthy households is virtually flat, declining only 0.3 percent a year over their retirement. Thus, at least in part, wealth and health constraints help explain the observed pattern of declining consumption.”
“Retirees likely prefer to enjoy constant consumption in retirement. The results suggest that a retirement saving shortfall exists since consumption declines are larger for households without assets.”
Resistance
Some commentators want to believe that it is safe to assume declining spending in a retiree’s financial plan. They dismiss involuntary reductions in observed retirement spending as insignificant. However, this new research makes it clear that retirees’ preferred spending levels are much flatter than the observed spending data. (For the record, Ben Felix says he assumes flat inflation-adjusted spending in his clients’ retirement plans.)
The idea that we’ll want to spend less as we age is seductive; it means we don’t have to save as much for retirement, can retire earlier, and can safely overspend in early retirement. What’s not to like? The problem is that average retirement spending data shows spending declines right from the first years of retirement. Does it make sense that people still in their 60s suddenly want to just sit around inside their homes? It’s plausible that retirees tend to become homebodies deep into their retirements, but not in the early years.
The clever-sounding justification for planning for spending declines is that our retirements consist of “go-go years, slow-go years, and no-go years.” However, this contradicts the observed early spending declines. In reality, some unfortunate retirees have not-enough-money-left years.
Other Factors to Consider
Another important consideration is that once retirees do start slowing down deep into their retirements, the possibility of needing expensive care increases. It’s not clear that we can ever count on wanting to spend less during retirement.
Yet another factor is the tendency for wage inflation to exceed price inflation. Over time, society becomes slowly wealthier. If you plan for flat inflation-adjusted spending through retirement, your spending level will slowly fall behind your younger neighbours. If you plan for declining inflation-adjusted spending, you’ll fall behind faster.
Conclusion
Average retiree spending declines over time in part because some retirees spend too much early on and are forced to cut back. It doesn’t make sense to me to plan my own retirement using statistics that include data from retirees who have overspent. The end result for me is that I’ll assume my spending desires will grow with inflation over my retirement. Anything less is risky.
Now what do you do if you oversaved while working and just can’t get excited about increasing spending? Does it get better the more you do it, like your golf swing?
ReplyDeleteYour question assumes facts not in evidence; my golf swing does not get better the more I do it. I’d certainly rather have saved too much than to live fearing the day when my savings run out.
ReplyDeleteMy thoughts have always directed me into increased spending as we age. I expect my portfolio to increase over the course of 30-35 years to meet that demand. The biggest unknown factor is taxes, which I really do not know how to accommodate.
ReplyDeleteIf you mean in nominal terms, then I agree, because I expect my spending to increase with inflation. If you mean increased consumption in real terms, then I'm not planning for this. However, this means I may be missing out on the latest Apple Vision Pro, or whatever comes along in the future that doesn't fit in my budget.
DeleteNot an apple fan so I wouldn't buy it either :-) But my target for the portfolio is complete salary replacement with yearly salary increases built in. Wish me luck, I need it.
DeleteThat plan wouldn't have worked for me, because my family spent much less than my salary, and it would have had me delaying retirement indefinitely. But everyone's situation is different. Good luck.
DeleteI understand completely.
DeleteMy own personal thoughts on this (totally personal, not backed by any research) are that I might guess and prepare for my personal spending in retirement to SLIGHTLY decrease in real terms over retirement (because I will likely "slow down" somewhat) but then potentially very suddenly rise towards the end / last few years (as potential health-related expenses increase - hopefully not badly so - but reality is what it is).
ReplyDeleteI think it's a potential mistake to assume the spending curve will be monotonic. I think it's likely to have the two phases I mentioned above: slowly decreasing in real terms, for "most of" retirement, followed by an unpredictable and potentially high phase near the end.
(As I think about it, maybe that's an insight into how to plan - treat as two separate "needs" - the "steady retirement" phase 1 vs. the "near the end" phase 2 where needs & costs may change rapidly)
You're right that these thoughts aren't backed by research, but you get to choose your own plan. A challenge I see is trying to assign numbers to this plan. Say you're retiring at 60. Will your slow spending decline start right away, or will the decline only kick in at some later age? If so, at what age? While your spending is declining, will you be happy with a plan that prevents you from spending any money to compensate for whatever is stopping you from continuing to enjoy life as you had in earlier years?
DeleteI find future spending plans difficult to come to any conclusions. Why year over year expenses have never really deceased, so why would I expect that to happen if I retired from work? My thinking is we don't see spending declining at all. Rather I expect to see consistent increases in rent, food and health. Perhaps in my most senior years I'll give up driving, but I will still need to by jeans and eat.
DeleteThe article is written as if the individual has no control? You can plan for reduced spending, and choose to do it. Or not. If you really understand your own spending over time, and your needs and desires, then there shouldn't be a big surprise. There are some costs that may go down depending on your life situation: work expenses, savings, child expenses; and many more that are directly under your control: housing, car payments, travel. It's just important to not assume your spending will go down if you haven't thought about it in detail.
ReplyDeleteI don't think a 60-year old has a good handle on how his or her spending desires will change through the coming 30+ years. There are just too many possible surprises. The point of looking at historical data from other retirees is to get a sense of how their desires changed, so that it can inform how your desires might change.
Delete