Any plans to rework Bernicke model based on new research?

[intent]

I stumbled across a Kitces post that seemed to imply Bernicke's numbers might be off. A relevant quote from the article:

For instance, a study in the Journal of Financial Planning by Ty Bernicke found that based on the 2002 CES data, for every 5 years older a retiree was, their spending was on average about 15% lower; the cumulative impact meant that those in their late 70s were spending less than half of what those in their late 50s were spending.

Notably, though, the Bernicke study looked at a cross-section of people in the same year to see the differences in spending by age, rather than actually track a cohort of people to see how their spending changed over time. When a subsequent study by the Center for Retirement Research looked at multiple age cohorts in the CES data, though, they still found that retirement spending drops persistently, by about 1% per year as a cohort ages, even after controlling for a number of other factors.

(bolding mine)

Kitces still seems to think the cumulative research shows that spending does tend to decline in retirement, but maybe just not as much as Bernicke believes.

Just curious if FireCalc developers have considered revising the Bernicke model, or maybe adding other reduced spending models based on the more recent research. Or maybe these models already exist in a paid version of FireCalc.

 

[growing_older]

As people age and run low on money, they do spend less. That doesn't mean that people who are not low on money are going to spend less, and certainly not in the same amounts. I have older relatives who fit these models, because as inflation reduced the value of their fixed incomes, they were forced to spend less. I have older relatives who had ample means who increased spending on home maintenance helpers, household help, more comfortable travel and other conveniences until they were very old, and then went into expensive senior care facilities. Their spending never decreased until they were frail and immobile and then medical costs rose considerably.

Taking overly broad averages of spending doesn't consider individual cases enough to make assumptions about an individual's future spending.

 

[intent]

I believe Bernicke's research found that as people aged they spent less (on average) regardless of their level of wealth. The more recent research doesn't seem to dispute that general finding, only the level of spending deceleration is in question.

Of course you're correct growing_older that broad averages are only useful as guidelines when planning as an individual, but we use the data that's available.

 

[haha]

I stumbled across a Kitces post that seemed to imply Bernicke's numbers might be off. A relevant quote from the article:



(bolding mine)

Kitces still seems to think the cumulative research shows that spending does tend to decline in retirement, but maybe just not as much as Bernicke believes.

Just curious if FireCalc developers have considered revising the Bernicke model, or maybe adding other reduced spending models based on the more recent research. Or maybe these models already exist in a paid version of FireCalc.

This "research" is bogus anyway. Best to forget it.

Ha

 

[wmc1000]

To my way of thinking, if most of us on this forum are LBYM's types, we have kept most of the "FAT" out of our spending all these years so not much that we will reduce going forward in life. For DW and I we use the constant spending power to model our retirement in FIRECALC rather than have to plan on ways to spend less as we age.

 

[audreyh1]

We don't plan on spending less as we age either. Maybe it will go to primo accommodations when we are too old to take care of ourselves.

 

[FIREd]

I'd rather not plan on spending less as we age so, like wmc1000, we use the constant spending power model in FIRECalc.

Discretionary spending represents 36% of our budget at the moment, and it is conceivable that our wants might decrease with age (although I don't see this happening in the near future). But on the other hand, healthcare spending is already 22% of our budget and this percentage is very likely to keep increasing over time (probably eating into our discretionary spending, since our other expenses are already pretty lean and incompressible). So I would be really surprised to actually see our overall spending fall - maybe after we both get on Medicare?

 

[razztazz]

You can save money and spend less and less year over year for 15-20 years then blow all that money in your last 2 or 3 years, so it all averages out. Just not in a nice even way, but that's what makes up the "averages."

 

[intent]

While I agree that it is safer to plan for a constant spending model, and delay retirement until you have saved enough to where FireCalc provides a 95% success rate - and that sounds like what most of you who have posted here advocate - FireCalc does offer an alternative model based on Bernicke's research.

The Kitces post references additional research that seems to confirm Bernicke's general findings, if not his specific decelerated spending amounts.

My main reason for starting this thread isn't to suggest that everyone should use the Bernicke model in FireCalc, but rather to find out if the Bernicke model in FireCalc is based on Bernicke's original research findings, if it has been updated to incorporate more recent research, or if there are any plans to update it to incorporate more recent research.

 

[youbet]

I believe Bernicke's research found that as people aged they spent less (on average) regardless of their level of wealth.

Not quite........

Bernicke said that some people will spend more as they age, some the same and some less. The average will be a trend to spend less.

Your job is to determine a priori which group you'll fall into. We fully FIRE'd (no PT work, no PT business, no managing RE, etc.) at DW 55 and me 58. We're now a happy 70 and spending more annually than when we pulled the plug over a decade ago. Apparently we're not "average" in Bernicke's data.

Averages can be useful as long as:

(1) You know you're going to be average.

(2) You know the dispersion of the data around the average.

(3) you understand any circumstances which will impact where you'll be amongst the widely dispersed data.

Personally, I think assuming you're going to spend less (be Bernicke "average") would be highly risky, so I didn't and I'm glad.

Admittedly, we're still about as active now as when we FIRE'd and since the economy and our portfolio have been doing well, we've been going a bit more upscale than we ever planned for originally. It's still possible we'll slow down in the future, although we're inclined to pay more for the necessary services to keep going despite age limitations.

Others will be unique to themselves in how they do it. The important thing is to be able to know well in advance how you'll do (average, more or less) vs the overall population. Or, be conservative and assume your spending will not reduce over time.

 

[MichaelB]

My main reason for starting this thread isn't to suggest that everyone should use the Bernicke model in FireCalc, but rather to find out if the Bernicke model in FireCalc is based on Bernicke's original research findings, if it has been updated to incorporate more recent research, or if there are any plans to update it to incorporate more recent research.

Not to my knowledge.