Finding the Right Retirement Withdrawal Strategy

Christine Benz: Hi, I'm Christine Benz for Morningstar.com. Many retirees have gotten the message that they should make their withdrawal rate sensitive to their portfolio's fluctuations. But how specifically should they do it? Joining me to discuss that topic is financial-planning expert Michael Kitces.

Michael, thank you so much for being here.

Michael Kitces: Great to be here. Thanks.

Benz: You have written about the 4% rule, as have a lot of folks recently. The guideline has gotten a lot of scrutiny. You have actually argued that, in many market scenarios that might unfold during retiree's lifetimes, it's actually too conservative. Let's talk about that.

Kitces: The whole framework and origin of the safe withdrawal rate is interesting. I find a lot of people just aren't even familiar with where that number really came from. The starting point for the number was a study done all the way back in 1994 that said, "Let's just look historically at what kinds of withdrawal rates would've worked throughout history." Sometimes you get a really good retirement and you can withdraw 7% or 8%; sometimes you get a bad retirement and you would have to withdraw less. So, in Bengen's original study, all he did--and I don't mean that in a belittling way, it was very creative--but all he did was say, "Let's look at all the different withdrawal rates that worked through history. Some were good; some were bad. The average was like 6% or 6.5%. And let's find the worst one that's ever happened."

Benz: The worst environment.

Kitces: The worst environment that has ever happened when we go back through history. So, what happens if you retire right before the '73, '74 bear market? What happens if you retire in the '60s? What happens if you retire right before the crash of '29? So, he looked at all the different possible historical markets that were out there and said, "I am going to take the absolute worse one--the single worst one we've ever had--and I am going to make that the baseline spending." That was where the 4% rule came from. And that 4% withdrawal rate would have actually still worked--just barely--but that was the one that would have actually still worked in the worst-case environment that we'd ever seen in history. Now, the caveat to that is, most of the time, you don't actually retire at the worst moment in history.

Benz: Right. If you are unlucky--

Kitces: If you are unlucky, you do. Everybody else does not. So, when you actually look at how that flows historically, withdrawal rates have actually varied anywhere from 4% to 10%. The median historically was between 6% and 6.5%. So, when you actually cut from 6% or 6.5% down to 4%, you are taking a really big haircut. You are taking lifetime spending that's a third below what would have worked on average just to defend against the possibility of this bad scenario.

What ends up occurring if you actually run a 4% withdrawal rate throughout history, regardless of when you retire, what you find is in one of those scenarios you just barely make it to the end of 30 years--that was the one that originated the money. Ninety-six percent of the time, you actually finish with 100% of your principal left over. And on average, you almost triple your wealth. So, if you started with $500,000 at retirement, you finish with $1.5 million left over on top of doing all of that lifetime spending. And that's just a recognition that while there was one sequence historically that was horrible, the other 99% of them are not that bad. And in fact, what we really see from that data is virtually none of [the sequences] are that bad. Ninety-six percent of the time, you still end up with all of your principal left over, even after the whole time period.

<TRANSCRIPT>

Benz: So, let's just clarify before we move further: When we talk about 4%, that's 4% of my initial portfolio balance. And then I'm just keeping that dollar amount static except for taking that little inflation adjustment every year.

Kitces: To make the math easy, if I've got a $1 million portfolio, in the first year, I spend $40,000. For every year for the rest of my life I spend $40,000 adjusted for inflation. So, we don't keep recalculating every year because that gets really volatile. If we recalculate every year, then you get into weird scenarios like, for example, it's 2008 you buy your retirement home; it's 2009, you have to sell your retirement home; it's 2011, you can go buy the retirement home back if the people will sell it back to you. So, we don't keep recalculating. The whole point of it is how conservative do we have to set the number in the first place so that, even if horrible things happen, we're going to make it to the end with stable spending? And then if a better situation occurs, either we'll pass away with a larger account balance left over or in the real world at some point we'll sit down and say, "This is going a little bit better than I had feared it might; I think it's OK to spend a little bit more."

Benz: A lot of the important retirement-planning research over the past, say, five years has really been in this area of sustainable withdrawal rates. I think a recurring theme among a lot of the research is that people are better off trying to tether their withdrawal rates to what's going on with their portfolios and what's going on with the market. You need to be willing to put up with these periodic adjustments. But one thing I know is that retirees really struggle with how to do this--how to get in line with what's going on with their portfolios, while not having to take catastrophically reduced distributions from their portfolios. So, let's talk about some practical strategies for people setting these withdrawal rates on an ongoing basis.

Kitces: Great question. As we just said, if you take this 4% recalculation of a volatile account balance every year, your spending is as volatile as your portfolio. That kind of becomes untenable in the real world. Sell your house, buy it back--ridiculous scenarios.

Benz: Stop going to movies.

Kitces: Those are things you would have to do if you were going to execute on that strictly. So, the kinds of methodologies I like for this really originated with a planner and researcher out of Minneapolis named Jonathan Guyton, who did a series of studies in the mid-2000s that looks at what he calls the "guardrails approach." I analogize it to going bowling with bumper lanes.

So, we're going to set some guardrails on either side of your retirement path. We know, on average, your spending as a percentage of your portfolio should rise gently over time simply because you are getting older and your time horizon is getting shorter. But if I do steady spending from a volatile portfolio--if I actually calculate each year from this stable spending with a volatile portfolio what my actual spending rate is this year--it bounces around. It's kind of like calculating the yield of a bond; as the price bounces around, the current cash flow yield moves around.

So, Jon's idea was to start with a target--for example, we're going to start at 4% or 5%, and we're going to assume it creeps up over time. We are going to recalculate and check on it every year. And as long as it's relatively close to where we were targeting, we'll hang tight. But if it gets really far off, we are going to start making changes. And so the way he structured it--which I think is actually still a very good approach--is to say, "We are going to start with a withdrawal rate of 5%," so $50,000 in a $1 million portfolio.

Each year, we'll recalculate how we are doing. If the portfolio growth is outpacing the inflation adjustments to spending rate, then my withdrawal rate will drop. So, my spending goes from $50,000 to $52,000, but my portfolio went from $1 million to a $1.5 million because we got this great two-year bull market. So, my withdrawal rate that started at 5% falls down below 4%, and he labeled that as the "prosperity rule." So, if you start with a withdrawal rate of 5% and it falls all the way down below 4%, you are so far ahead it's time to take a spending increase, and his target spending increase was going to bump up 10%.

And so, as long as portfolios are great and you keep setting this withdrawal rate that falls lower as the portfolio grows, you lift it up again with a prosperity-rule adjustment, and now you are kind of staying in the channel over time. If the market goes the other direction, so, for instance, you start at a 5% withdrawal rate, but the portfolio is flat or even declining while your inflation-adjusted spending is rising and suddenly your withdrawal rate is 5% and then it's 5.2% and then it's 5.5% and then it goes over 6%. If it goes over 6%, now you need to do a spending cut. You take a 10% dollar-for-dollar spending cut, so maybe you were up to $55,000 with inflation adjustments. You chop yourself back to $50,000, so you take a 10% real cut in spending. But now you've moved back into the channel again; your withdrawal rate had crept over 6%, and now you are back at around 5% because you trimmed your spending back to be in line with your portfolio.

So, as you execute that over time, what you find is you've got this withdrawal rate that's a little bit bumpy--because the markets are bumpy--and if it moves too far one way or the other, you adjust it back into the lane so that it can keep going. If it goes too far down, you lift it up to keep it in the lane. But then you can kind of move forward on that progression, and it just gives you a framework to do it.

So, if you start at 5%, you make adjustments at 4% and 6%, and your spending adjustments are 10% in either direction when you need to do it, you get a reasonably steady path in retirement. And this is research that Jon had done and, ironically, I guess I'll take a little responsibility for it as someone who does research in this space as well. I think we actually have lagged a little bit in doing follow-up research on this to try to build more robust rules or a wider range of rules--what are other ways that you can do this? But I still think Jon's framework is actually a really good starting point.

Benz: When we look at retiree portfolios today--say they had maybe a 50% equity, 50% bond portfolio at the outset, say, in early 2009--they've seen really nice appreciation in the portfolio. What should retirees be doing today? Should they be giving themselves a raise or would you say that they should just get on this year-by-year program?

Kitces: I think, from a practical perspective, you would give yourselves a raise. Granted, it would have been an unpleasant time to do it, but if you had actually pulled the retirement trigger in 2009--maybe you still thought you had enough after the bear market. Maybe you didn't have a choice because your employer went out of business, and you said, "Well, we're retiring with what we've got."

When you get the kind of market run that we've had over the past five years, in general, what we find is that anybody who started out with a relatively moderate withdrawal rate of even something like 4% or 5% at the market bottom in 2009, that withdrawal rate has fallen well below 3% right now if they've only been adjusting for inflation.

Even with a balanced portfolio, equities run over 200%, your portfolio is well ahead of where it was originally. I've actually done a recent follow-up study on this. Historically, when we look at the scenarios where you had to spend something like a 4% rule, what path of returns did you get?

We know, in general, you tend to get bad returns at the beginning--we call that sequence-of-return risk. Those tend to be the scenarios that create low withdrawal rates. The returns are so bad for the first 10 or 15 years, and you spent so much in the first half of your retirement that you've just got to be conservative or you're going to run out in the second half. But as it turns out, when you look at those historical scenarios, anytime that your portfolio ever gets more than 50% above where it started, you never come close to a 4% rule. You are virtually always at 5% or more at that point.

So, if you map that onto an environment like today, I think it really does take you to a point where, realistically, you could be ratcheting your spending higher today. Even if the market falls back, it only falls back so far. You've already lived several years of your retirement, and you are already so far ahead that it's still safe to do so. And I've taken to calling these "ratchet strategies"--kind of like the ratchet on a wrench. It only goes in one direction; it doesn't go the other.

So, there is a part of the Guyton rules that says, "If times are bad, you have to cut." But not everyone wants to do that. Granted, some people just don't like to do that, in general; but for many of us, they would say--and we hear this all the time from clients--"I'd rather just spend a little bit more conservatively today, from the start, than spend more and get used to that lifestyle and then have to cut later, which would be more painful. If I'm not using the money, it doesn't feel traumatic. If I'm using it and then I have to give it up, it's a lot harder."

So, what does a prosperity-and-capital-preservation rule look like from John Guyton's research? If you get rid of the preservation rule, you can just say, "I'm going to spend so low that even if times are horrible, I won't have to cut." But now, if times are good--and it's even more likely that times will be good because you're spending is so conservative--how do we start ratcheting upward?

Again, what I found from some of the research is that you can make relatively simple rules. For instance, if I start out at $1 million, anytime my portfolio is above $1.5 million--if it has grown that far, however long it takes--you can bump your spending up 10%. And every three years that it continues to stay above $1.5 million, you can continue bumping it up 10%. When we play those scenarios out, even throughout everything we've seen in the history, if you're that far ahead, even when the next bear market comes, you don't actually fall far enough behind to threaten that spending.

It really just become another version of Jon Guyton's prosperity rule, which is when you get far enough ahead and your withdrawal rate has fallen back far enough, it's OK to start lifting your spending because you can only fall back so far. In the short term, markets are a voting machine; in the long term, they are a weighing machine. They only fall so far before eventually every stock will have a 10% dividend, and people will be buying them because they just give amazing cash-on-cash dividends; that's what happens if equities fall far enough, and it starts lift them up again.

So, I think that's what we're going to see more of going forward--different ways to put these rules together. But as a starting point, I think Guyton is one approach. So, we're going to start at a withdrawal rate like 5%; if we fall outside the bands of 4% or 6%, we're going to start taking spending adjustments to put us back within the guardrails. Or a more ratcheting-style rule: I'm not going to start at 5% and adjust up or down; I'm going to start down at 4%, and I'll just adjust up if I get really far ahead--however long that takes. If I don't get really far ahead, I'll sit tight at 4%. If things get better, then I can ratchet up as I go.

Benz: Michael, this is such an important topic. Thank you for being here to share your very practical wisdom.

Kitces: Hope it helps. Thanks.

Benz: Thank for watching. I'm Christine Benz for Morningstar.com.