"4% is too low." How much will you leave on the table?

[Nords]

I read a lot of financial blogs, and I can't remember any other economists who blog as frequently as Wade Pfau. Maybe William Bernstein, although of course his doctorate is in a completely different field.

So it's nice to read small doses of economic research from someone who writes like a "regular guy", rough drafts and napkin sketches and all, rather than yet another ivory-tower academic who has his stone tablets hauled down the mountain every 3-4 years for everyone to scratch their heads over and try to decrypt into plain English. The only other popular-reading economist I can think of is Milevsky, and I don't even know if he blogs-- but at least he writes readable books. I'm trying to think of more economist's writing that I read regularly, and I'm drawing a blank.

Pfau seems to be spending his time in an area that I think holds a lot of promise-- figuring out how human behavior and variable spending affect portfolio survivability. I think it's very interesting to see him try to put realistic numbers on how much of a portfolio should be annuitized for bare-bones survival, and to try to translate "risk tolerance" into concepts that real-life retirees can sleep with at night.

He's co-author on an article in the March "Journal of Financial Planning" that speculates the 4% SWR may be too low. That's right, we may not be spending the money fast enough. Spending? Spending?!? We're so loss-averse and so risk-shy that we can't HANDLE the spending!

The Cliff Notes version of the article is summarized by Scott Burns here:
Life: How Much Will You Leave On the Table? - Registered Investment Advisor

Quote:

How would you like to double your retirement spending? Well, Michael Finke is working on just that. He thinks we may be able to live better in retirement than most professionals have thought for nearly two decades. The Texas Tech University associate professor, along with two other researchers, Wade D. Pfau and Duncan Williams, has examined William Bengen’s well-known 4 percent safe spending rule and found that some retirees, perhaps many, can spend a lot higher on the hog. Simply raising the spending rate to 6 percent means you can spend 50 percent more. “By emphasizing a portfolio’s ability to withstand a 30 or 40 year retirement,” the researchers write in the March issue of the Journal of Financial Planning, “we ignore the fact that at age 65 the probability of either spouse being alive at age 95 is only 18 percent.” As I pointed out in a recent column, it’s silly to have 95 percent confidence in your income when your chance of being alive is much smaller. Excessive caution, he told me in a recent interview, means we buy long-term security at the expense of giving up many things we’d like to do today. We leave estates that are larger than planned and feel remorse for experiences we’ve missed.

The full article is here, with graphs:
Spending Flexibility and Safe Withdrawal Rates

Quote:

What is missing from the shortfall literature is the consideration of what is lost when withdrawal rates are overly conservative. By emphasizing a portfolio’s ability to withstand a 30- or 40-year retirement, we ignore the fact that at age 65 the probability of either spouse being alive by age 95 is only 18 percent. If we strive for a 90 percent confidence level that the portfolio will provide a constant real income stream for at least 30 years, this means that we are planning for an eventuality that is only likely to occur 1.8 percent of the time. And even that figure assumes that clients are unable to make adjustments to their spending later in retirement. So by relying on standard historical or Monte Carlo simulations to determine a safe withdrawal rate, clients may be unduly sacrificing much of their desired lifestyle early in retirement. The failure to include a client’s willingness to adjust is an important shortfall of the shortfall literature. A common thread in the analysis is that all failures are counted the same, without regard to when the failure occurred or what percentage of the client’s stated aggregate spending goal was funded. Such an all-or-nothing approach to retirement simulation is inconsistent with the way trade-offs are framed in retirement. In practice, advisers often help their clients prioritize spending goals with basic living expenses, insurance premiums, and debt payments receiving top priority. Other goals, such as travel and vehicle purchases, are scalable and may even be reasonably expected to disappear entirely late in life. Different spending goals have different priorities and importance (Curtis 2006). Some clients may reasonably prefer a higher travel budget in their 60s and 70s, even if it means a higher probability of having to cut back on their dining and vehicle budgets in their 80s. This would be considered failure in most shortfall risk analyses. If a couple does not have a very strong desire to leave a liquid bequest (or has planned for the bequest through life insurance), the result of relying on standard simulations is that the vast majority will die with a lot of unspent money that they had intended to use to support their lifestyle. Ideally, we would like to include these unspent funds, and the happiness they could have provided if spent, in a calculation that also considers the serious implications of experiencing a shortfall. Fortunately, both can be modeled by using utility theory—the same concept that underlies modern portfolio theory. [...] For planners, the most significant insight is that a client’s willingness to take portfolio risk before retirement is equivalent to a willingness to accept shortfall risk after retirement. A risk-averse investor should choose a lower withdrawal rate in order to reduce the probability of having to reduce consumption later in retirement. This result is similar to the findings of the typical shortfall minimization strategy; however, the traditional approach fails to capture the preferences of a client who is willing to accept the risk of a diminished income in order to live better in retirement. The authors recognize that it may be difficult to consider a strategy that results in an increased shortfall risk; however, it may be helpful to encourage a client to choose among possible conservative strategies (say, between 4 percent and 6 percent) by articulating the shortfall risk/lifestyle return trade-off. By increasing the size of the income floor in retirement, for example by investing a portion of assets in a guaranteed income product, an adviser is able to recommend both a higher asset decumulation rate and greater portfolio risk. In other words, a client will be better off with a riskier portfolio when the downside drop in spending is not as severe. The magnitude of guaranteed income may then be viewed as a client’s decumulation risk capacity. A larger pension or other source of annuitized income provides a cushion against the loss in quality of life if investment returns are unfavorable or the client outlives his or her assets. Advisers seeking a way to increase expected return in retiree portfolios may be best served by looking into the advantage of mixing a risky investment portfolio with products that protect a minimum level of income. The economic framework used in this paper provides a scientific approach to a philosophical issue that has long been discussed in the planning community. Practitioners are often torn between a strict interpretation of the safe withdrawal guidance and a looser interpretation that allows retired clients to spend more on things that bring meaning to their lives while accepting greater risk. The implications of the analysis in this paper can give practitioners a framework from which they can engage clients in conversations about sensible trade-offs in retirement.

I'm not trying to claim that Pfau has all the answers-- not yet anyway. But I feel like it's 1994 or 1998 and I'm reading about the SWR all over again for the first time. Keep an eye on this guy when you tweak your retirement spreadsheet.

[Danmar]

Thanks for posting this. Very interesting. This thinking supports mine in the sense that I can assume more risk given the backstop of a pension. Also, in the future we will probably ramp up our spending if things look promising. This paper may go a little against the grain here though as many people(most?)here don't want to spend any more regardless of the chances of their portfolio surviving. Now, as you may recall , I on the other hand could spend or give away a lot more if I thought I would not run out.

[haha]

Quote:

Originally Posted by Danmar Thanks for posting this. Very interesting. This thinking supports mine in the sense that I can assume more risk given the backstop of a pension. Also, in the future we will probably ramp up our spending if things look promising. This paper may go a little against the grain here though as many people(most?)here don't want to spend any more regardless of the chances of their portfolio surviving. Now, as you may recall , I on the other hand could spend or give away a lot more if I thought I would not run out.

I would be happy to spend 4x more. What would not make me happy is being transformed into a poor person. This author seems completely to have failed to understand the principle of marginal utility. "Only 4 years of being broke". That would give plenty time to get hungry.

Ha

[Gone4Good]

Quote:

Originally Posted by haha I would be happy to spend 4x more. What would not make me happy is being transformed into a poor person. This author seems completely to have failed to understand the principle of marginal utility. "Only 4 years of being broke". That would give plenty time to get hungry. Ha

No need to worry. It is typically the man who makes the financial decisions, it is the female who mostly suffers the consequences.

Quote:

Figure 2 shows, however, that this couple should expect to spend only about the final 3 percent of their collective lives with no remaining wealth. More often than not, a widowed female will be the one to endure this outcome.

I should say I do think these inquiries are both interesting and useful. However, I always come away feeling like the authors are striving for more precision than is possible. No Monte Carlo simulation in the world can optimize the future.

[pb4uski]

I wonder if I can use this thread with DW to convince her that i can buy that little two-seat convertible I have been looking at?

[Running_Man]

Having witnessed what can happen first-hand this is nonsensical thinking in my experience. My parents took their retirement savings and spent freeely - my father since age 82 is now living alone after the death of my mother and has no money other than Social Security (he is now 86). Yes if studied he will be one of the people that "prove" you spend less later in retirement. But the increased spending in the early part was totally foolish in retrospect. ( my parents spent at about a 7 percent rate)

My inlaws on the other hand actually doubled their savings in the same time span. (my wife's father and my father retired within a year of each other) They now live in a active living retirement community for which they rent an apartment for $40,000 per year, which includes one very high quality meal per day and maid service in their apartment. So as they are slowing down they are getting a lot of very high-quality care to help them in retirement which would not have been possible without careful planning. I am pretty sure in the early years of retirement their spending was in the 3% range. With twice the portfolio they are now over 4 percent per year. They have Kindles are active with a very nice computer,cell phones and have a great HD TV. All of the furniture is updated and well cared for after originally moving to their retirement home the same furniture they had for over 20 years. After seeing they had enough money when the moved to the new retirement community at about the same time my father ran out of money they purchased all brand new furniture.
A few years of possible poverty in exchange for more today? I have witnessed the fable and the reality of the grasshopper and his ways, no thank you.

[W2R]

Quote:

Originally Posted by haha I would be happy to spend 4x more. What would not make me happy is being transformed into a poor person. This author seems completely to have failed to understand the principle of marginal utility. "Only 4 years of being broke". That would give plenty time to get hungry. Ha

Quote:

Originally Posted by Running_Man A few years of possible poverty in exchange for more today? I have witnessed the fable and the reality of the grasshopper and his ways, no thank you.

+1 to both

I wonder if he is making these two assumptions:

(1) Retirees have an infinite, insatiable craving to spend, spend, spend. No matter how much they are already spending, they are eager/desperate to spend more.

(2) Despite this, when a retiree is in his/her last years, presumably in his/her 80's or 90's and essentially helpless, extreme poverty won't be tough to bear at all. Not nearly as tough as flying economy instead of first class would be right now.

I know for a fact that (1) is not true for all retirees. And I don't believe (2) is true, either.

On the other hand, some people welcome a higher level of risk in retirement planning than I would ever recommend. Each to his/her own.

When markets are up ("Wheee!"), people do seem start talking gleefully about higher withdrawal rates. I remember some of those recommendations during the 1990's as I'm sure the rest of us here do. However, having been through two crashes in the past ten years, I am not ready to think like that at all right now. When the markets are up, it seems to me that perhaps we should be talking about lower SWRs, not bigger, because one's nestegg is likely to shrink at some point later on in the cycle.

Perhaps we should use (or at least consider) our lowest nestegg values at the bottom of the 2008-2009 crash when computing an SWR.

[arebelspy]

Quote:

Originally Posted by W2R Perhaps we should use (or at least consider) our lowest nestegg values at the bottom of the 2008-2009 crash when computing an SWR.

I think this sort of mentality is exactly why he thinks people may die with way more than they need. Planning a 3-4% SWR on a portfolio size of only 2/3 (or whatever) of what it actually is?

The key though, as has been said over and over on these forums: do what helps you sleep at night.

I personally appreciate Mr. Pfau's analysis, but I think I sleep harder than most.

[W2R]

Quote:

Originally Posted by arebelspy I think this sort of mentality is exactly why he thinks people may die with way more than they need. Planning a 3-4% SWR on a portfolio size of only 2/3 (or whatever) of what it actually is?

Consider what happens to a 6% SWR (as I have heard proposed again more than once recently), when one's portfolio shrinkls like that and it becomes 9%. If you can sleep with a 9% SWR, great! At least you have considered it and won't be blindsided.

I think this sort of mentality that won't even consider something like that is... is.... well, is just different from mine!

[CyclingInvestor]

I do not consider any surplus "more than I need". The extra 2 years I worked (past when I reached FI) buys me peace of mind, and is worth far more to me than an extra $10-20K a year to spend.

[arebelspy]

Quote:

Originally Posted by W2R Consider what happens to a 6% SWR (as I have heard proposed again more than once recently), when one's portfolio shrinkls like that and it becomes 9%. If you can sleep with a 9% SWR, great! At least you have considered it and won't be blindsided. I think this sort of mentality that won't even consider something like that is... is.... well, is just different from mine!

It's a tough balance, I admit. I think what makes me the most comfortable right now (not being ER yet) is running the numbers, and studies like this.

If FIRECalc says I'm good, I don't feel a big need to have a huge cushion to help me sleep at night. Merely seeing those results do it for me.

I concede that everyone's tolerance is different, but a study like this can help broaden some mindsets on both sides.

[Nords]

Quote:

Originally Posted by W2R (1) Retirees have an infinite, insatiable craving to spend, spend, spend. No matter how much they are already spending, they are eager/desperate to spend more. (2) Despite this, when a retiree is in his/her last years, presumably in his/her 80's or 90's and essentially helpless, extreme poverty won't be tough to bear at all. Not nearly as tough as flying economy instead of first class would be right now.

Here's an idea worth studying.

Let's say you have enough to retire at a 4% SWR, but you're worried about the .00001% longevity risk issue (or whatever significant figures you want to use).

The challenge would be figuring out a baseline bare-bones budget that would support your elder years. Somewhere better than cat food and Goodwill furniture (although our house is full of Craigslist finds) but not up to the level of luxury assisted living or first-class flights. And medical expenses are still an issue to be worked out.

Let's say you annuitize the part of the portfolio required for the bare-bones budget.

Now how do you handle the rest of your portfolio? With so much of it annuitized, perhaps it makes more sense to invest the rest of it in 70-80% equities. If the market goes into recession then you drop your spending to the annuity and don't touch the equity portfolio. If the market goes "Whee" then you spend a reasonable percentage of your equity portfolio.

You're probably spending 5-7% of your equity portfolio, but you're not doing that every year. And you annuitized a large chunk of the rest of your ER portfolio, so I don't know whether that SWR is amortized over your life expectancy or just considered to be the amount of the annuity income.

But with a portion of your portfolio annuitized, I bet your overall SWR is higher.

I would suspect that Social Security is not enough annuitized income. But a military pension probably puts this plan well into Otar's green zone.

[MichaelB]

Quote:

Originally Posted by W2R +1 to both I wonder if he is making these two assumptions: (1) Retirees have an infinite, insatiable craving to spend, spend, spend. No matter how much they are already spending, they are eager/desperate to spend more. (2) Despite this, when a retiree is in his/her last years, presumably in his/her 80's or 90's and essentially helpless, extreme poverty won't be tough to bear at all. Not nearly as tough as flying economy instead of first class would be right now. I know for a fact that (1) is not true for all retirees. And I don't believe (2) is true, either. On the other hand, some people welcome a higher level of risk in retirement planning than I would ever recommend. Each to his/her own.

1) They acknowledge the decreasing utility of spending. 2) They assume everyone in their model knows the trade off between higher spending today and the effect of impaired finances during the final years.

Nords, thanks for the link. This is a good effort to explore an area that needs to be studied, and I think they do a good job. Hopefully they continue to build on this effort. Their base assumption is telling: the most risk averse retiree has 50% of total income derived from SS or pension and the other half from portfolio. As the assumption of risk increases, so does the ratio of portfolio to total income. Those of us that depend totally on portfolio may find this useful if only because it forces us to look differently at the way we assess risk.

[donheff]

The earlier you retire, the more risky this approach since you could be forced to scale back for a long time. But the concept has merit for the huge number of people who will be hitting their mid 60s or early 70s with insufficient savings. And it could have value to ERs who want to stretch their withdrawals a bit to enjoy some travel (or whatever) while health allows. I think the key they discuss is one we talk about all the time here -- understanding your basic needs (and what is basic for you may be entirely different than for me) and trying to cover that base with relatively bullet proof funding (e.g. pensions, SPIAs, bond ladders). For example, as long as you have the basics well covered, is it so dangerous to splurge $20 grand for that trip to the far east you always dreamed of? If the market turns down, scale back. If it stays down forever, fall back to basics. Just make sure the basics you cover are not restricted to dog food.

[Gone4Good]

Quote:

Originally Posted by W2R I wonder if he is making these two assumptions:

I don't belive so. More likely his target audience isn't the typical member of ER.Org, but financial planners who need to deal with a potentially infinite number of client situations and preferences. I can see from the FP's perspective that having the flexibility to customize a retirement plan along more demensions than just asset allocation and probability of success is tremondously valuable.

While these strategies aren't right for me, and maybe not for most of us here, I can see an FP wanting to have more to offer a 65 year old manual labor worker with a little cash saved, a little SS, and a job that is kicking his ass than a recommendation to simply "work longer."

[H2ODude]

Great topic. As said, to each his own. Trying to absorb all the "facts" and so called predictability of SWR is something each of us has to do and reconcile to our own life style and risk tolerance, not to mention what our sources are. Some of us, myself included, are fortunate enough to have a defined benefit pension. If I include SS (I'm 61) in a calculator, DW and I can go till 95 with only enough savings to get to when we claim SS; living on same after tax income as we spend now. Which means all the investments can be put totally at risk. Well, no, that's not going to happen.

The other issue, discussed here, is this one of changing spending. We have family issues that restrict our travel right now, but if we could, I'd be spending all or more of that 4% SWR while we can enjoy it. Might even move or get a vacation place. At least that's what I say, I'm very risk averse and would find it hard to eat into the egg if markets were dropping. But, to me it's insane to think that we would sustain that kind of spending in our eighties. I guess there are calculators out there that allow plugging in a reduction of expense, but I'm not that enthused about the "precision predictability" that this implies.

I know many are forced to cut this decision very closely, as in they want "out" and need to seek a SWR that will allow them out as soon as possible. What looked pretty doable to a 401k dependent pre retiree in early 2008 didn't by the end of the year; I ran into a colleague forced to return to work in that scenario (ugh). I think building in flexibility to your plan is essential. That, and if you don't have anything like a pension, consider an immediate annuity but perhaps not right now given the low rates, for the absolute essential income you'd need to avoid an unacceptable level of lifestyle.

Anyway, just my thoughts on a Sunday morning, Happy Easter all...

[Midpack]

I need to read it again, but it sounds like some replies are missing the annuitization aspect. Aren't they essentially saying if you want to actually die broke (or some fixed target), the more you have annuitized (pension, Soc Sec, SPIA), the more you can withdraw from your (remaining) portfolio during retirement? "Broke" isn't anywhere near broke in Burns' articles context.

Even though Burns mentions going broke before you go poof, he later mentions this is all based on having an assured floor income. Exaggerating to make the point, if you were able to buy an annuity that provided 2X your annual spending needs with COLA or have a similarly generous pension, you could be pretty reckless with whatever money you had left. Your portfolio could go to zero, but your needs would still be covered no matter how long you lived.

As someone who would like to die broke and not run out of money, the article has merits, though I plan to more seriously consider this in about 15 years depending on real returns and proximity to my own annuitization hurdle.

By definition, the 4% SWR approach is conservative, in most cases a large residual will result, which is not desirable to many retirees. So what do to about it? The more you have annuitized, the more you can safely withdraw from your portfolio. The less you have annuitized, the more conservative portfolio withdrawals need to be and the greater the chances of a large portfolio residual (fine if you want a large residual to leave to charity, family, etc.). I think most here already know that.

Thanks Nords, though I am probably reading Wade Pfau's blogs and articles (among other sources) as often as you are now...

[Danmar]

Interesting replies. @Nords&Midpack. Yes I think the important point is the annuitization. Without this most of us would by necessity by very conservative and almost forced into a larger than desired bequest on death. I have a large pension but at some point in my later years may buy a SPIA to reduce longevity risk further. In the meantime it's only divs(3.6% SWR) for me.

[Midpack]

OK, I read the main article more carefully and it's a great insight into maximing retirement spending optimizing annuitization and SWR for those of us who want to attempt to actually die broke (or with a set residual) without ever running out of money. Even if one doesn't annuitize portfolio funds at all, it provides some insight into an appropriate WR factoring in how much assured floor income Soc Sec will provide - something overlooked in many studies.

Not an easy read and I need to better understand how the 0-10 risk aversion factors apply to me, but a keeper IMO. Thanks again Nords!

[tryan]

So I was doing Mom's taxes last week .... she's been living within the required mandatory IRA withdrawls (~4% for her age : 74). But she complains that she feels poor ... "can't even afford a sweater". Little bit of background, a pension and SS easily cover her living expenses .... just not enough for travel or "luxury" items.

So I looked at the balances on the IRA and told her pull 7% for the next 10 years. As she says "80 is the WALL" ... so let's enjoy what she has while she on her feet.