What Investment Rate of Return do you Use?

[Danmar]

That depends on how much margin for error is in the portfolio and on a person's risk tolerance. Those without pensions will naturally be more cautious.

Agree, but could you try to explain what you would do if you decided to reduce your forecast of future real gains by one percent. Would you reduce your WR by a corresponding amount?

 

[Meadbh]

Agree, but could you try to explain what you would do if you decided to reduce your forecast of future real gains by one percent. Would you reduce your WR by a corresponding amount?

Yes.

 

[Danmar]

Yes.

If I may ask, what is your long term real return forecast and your corresponding WR? Is there a mathematical link between the two? Maybe, this would be a good question for others? If they can be linked, I might change my mind on this issue. Thanks.

 

[BBQ-Nut]

This may need a separate thread, but who uses SWR as an actual withdrawal methodology vs a check?


My approach is my expenses are what I base my withdrawals on and then I calculate what the SWR is as a check only. Then that is tempered against what the immediate past real returns were against next year's expected real return.


Sometimes resultants SWR may be higher or lower than the 4% 'rule of thumb'. But I have projected out different scenarios across different planning tools, including my own and using a range of real returns, my SWR is overall less than 3%. But I do not use SWR as the amount for withdrawal planning.

 

[Meadbh]

If I may ask, what is your long term real return forecast and your corresponding WR? Is there a mathematical link between the two? Maybe, this would be a good question for others? If they can be linked, I might change my mind on this issue. Thanks.

Well, in recent years we have been hearing that equities are likely to yield mid single digits over the next decade. Say 5%. Bonds are predicted to yield something in the region of 1%. Let's simplistically assume that a balanced portfolio yields an average of 3% over the next decade. Long term inflation rates have been ~3%, but let's assume 2% for now. The average predicted real return of a balanced portfolio would then be 1%. In the first 3 years of ER, my WR has been ~3.5%.

Let's say I program that into my spreadsheet and populate it with my expected lifestyle expenses, debt repayment and taxation, all adjusted for inflation. I can now predict what my net worth may be in a decade.

Next, I take that average net portfolio return and introduce some variability. An average of 2% could be a sequence beginning with -4%, -10%, or +6%. Obviously, the degree of volatility and the sequence of returns will influence the bottom line. I vary the inflation rate. One or more of these scenarios will yield numbers that make me worry. Then I will look at my expenses to see what I can do to minimize the chances of failure. Strategies include cutting back on WR, building income streams from real estate, taking CPP early, and tax smoothing by withdrawing judiciously from tax sheltered accounts prior to encountering RMDs. Going back to work is not an option.

There is no mathematical formula involved here, just scenario building. I am not suggesting that you use my methods. You and I are in different financial leagues and you have far more financial security than I do. I need reassurance that I can handle whatever comes my way.

 

[Danmar]

Thanks Meadbh. That's useful. I agree with you that it is quite conservative.

BBQ, also useful. So your WR is determined bottom up as desired spending? Also interesting. In my case if is more iterative. Spending increased to match funds available but funds available is a judgement call. Since my portfolio is 100% equities, I think a WR of 3.25-4% (being my div yield) is reasonable for a portfolio real return of maybe 3%. For the last 10 years it has compounded at about 10% real so this is a real decrease. Anyway, many of you apparently spend a lot more time on this kind of thing than I do. Good for you.

 

[youbet]

This may need a separate thread, but who uses SWR as an actual withdrawal methodology vs a check?


My approach is my expenses are what I base my withdrawals on and then I calculate what the SWR is as a check only. Then that is tempered against what the immediate past real returns were against next year's expected real return.


Sometimes resultants SWR may be higher or lower than the 4% 'rule of thumb'. But I have projected out different scenarios across different planning tools, including my own and using a range of real returns, my SWR is overall less than 3%. But I do not use SWR as the amount for withdrawal planning.

It looks like you might be confusing SWR with WR in some of the above.

And I'd say you are very much planning your WR based on SWR calculations. You've got the "Chicken or egg" thing going on. But, whether you withdraw and spend and then compare that spending to a calculated SWR and make ongoing adjustments or you calculate a SWR, withdraw and spend that, you're taking calculated SWR into consideration either way.

Everyone has their own way of doing it. But as long as you're calculating a SWR and taking that into consideration for future WR's, it's part of your planning. IMHO anyway.

Interestingly, FireCalc never directly recommends an SWR as a percentage of portfolio. It asks for a spending amount and then tests that amount historically vs your ongoing income sources such as SS, pension, deferred comp, etc., plus withdrawing any additional funds needed from your portfolio and projects portfolio balances.

 

[BBQ-Nut]

It looks like you might be confusing SWR with WR in some of the above.

And I'd say you are very much planning your WR based on SWR calculations. You've got the "Chicken or egg" thing going on. But, whether you withdraw and spend and then compare that spending to a calculated SWR and make ongoing adjustments or you calculate a SWR, withdraw and spend that, you're taking calculated SWR into consideration either way.

Everyone has their own way of doing it. But as long as you're calculating a SWR and taking that into consideration for future WR's, it's part of your planning. IMHO anyway.

No, I understand SWR vs WR.

My question is in regards to those using SWR as their actual withdrawal methodology.

I only use whatever the flavor of the year is by pundits and 'experts' as a check.

I get the impression from this thread and others, that some here actually base WRs on an SWR (whatever the number is given its point in time).

 

[CaliforniaMan]

I think the issue here is what does it means to be "like the past"? There are multiple ways of defining "like" that could lead to widely different forecasts.

For example, consider the following two approaches

(1) You believe that the future stock returns will be like the equally weighted average of returns in the past. So to create your expectation you take a simple average and use that as your forecast.

(2) You believe that future stock returns will be like past years when the market had similar Schiller P/E values (or whatever market condition you think is important). So to create your expectation you take a weighted average (giving higher weight to past years with similar Schiller PE) and use that.

In both cases, we are trying set expectations based on the idea that the future will be like the past. It's just that the definition of "like the past" is a little different in each case.



In some cases like with Schiller PE there's a mountain of evidence that this is a real effect and not some spurious result. Other methods (like using the butter production in bangladesh to predict S&P 500) may have more limited support.



The people who make models to forecast stock returns are all basing it, directly or indirectly, on past data. The model might produce different predictions than the historical average but that's because current conditions are not historically average.



I will not be happy either with such low rates of return. But forecasting is extremely difficult and even the best models (which tend to be pessimistic now) do not exclude good outcomes when you look at the range of expected results instead of a point estimate.

+1 Excellent post.
We of course have no way of knowing the future, and anything we do know we would have to assume is pretty much already priced in. So we base all of our models on what we know of the past. Other than guessing, that is all there is. We just pick and choose what part of that past history we want to emphasize and what parts we want to minimize. Our assumptions control the model, so in some cases, why do the model, we already know from our assumptions how it will turn out.

One reason I like historical estimates like FIRECalc is that there are so few assumptions made, just cranks out how well we would have done in the past, to answer a simple question - are we anywhere near reasonable.

As an individual, I don't see why we want to spend any time trying to predict an unknowable future rate of return. Certainly we cannot forecast sequence of returns, and that could prove a far greater problem for us.

It is not that I am against modelling, I love it and have done a lot, stochastic, ARIMA, multivariate, etc. I just have a hard time figuring out the point of doing it to figure a SWR. We put in our assumptions (based on some info from the past) and churn out results. How it changes our estimated SWR, other than what we can already get from simplistic FIRECalc-like estimates, I haven't a clue. We are just pretending we know more than is possible.

Now if you are a pension fund, well things are different, and you have a whole set of additional problems, modelling is a must. But for individuals, not so much.

 

[youbet]

No, I understand SWR vs WR.

My question is in regards to those using SWR as their actual withdrawal methodology.

I only use whatever the flavor of the year is by pundits and 'experts' as a check.

I get the impression from this thread and others, that some here actually base WRs on an SWR (whatever the number is given its point in time).

Sounds like you might be a bit addicted to the "talking heads" and other so-called financial guru's the media splashes around these days. I doubt that many on this board are grabbing the many and various numbers flying around the Internet and airwaves these days!

I do, however, use a WR which I test with FireCalc and a few other popular tools, as well as my own back-of-the-envelope calculations for safety. That is, test my WR for survivability over time. So, yes, I guess SWR is an important component of my withdrawal planning.

You make choosing a SWR sound quite cavalier with terms like "whatever number" and "pundit."

I don't think your methodology is as unique as you are portraying it BBQ. Many of us follow the FireCalc methodology which is to specify a spending level and then test it vs your projected income and portfolio.

 

[BBQ-Nut]

Sounds like you might be a bit addicted to the "talking heads" and other so-called financial guru's the media splashes around these days. I doubt that many on this board are grabbing the many and various numbers flying around the Internet and airwaves these days!

I do, however, use a WR which I test with FireCalc and a few other popular tools, as well as my own back-of-the-envelope calculations for safety. That is, test my WR for survivability over time. So, yes, I guess SWR is an important component of my withdrawal planning.

You make choosing a SWR sound quite cavalier with terms like "whatever number" and "pundit."

I don't think your methodology is as unique as you are portraying it BBQ. Many of us follow the FireCalc methodology which is to specify a spending level and then test it vs your projected income and portfolio.

I think we are actually close in agreement for withdrawal methodology.

I only listen to my Talking Heads CDs and LPs, thank you very much.

You do agree though, that there is a body of study and articles that describe and detail an actual retirement methodology based on "Safe Withdrawal Rate"?

And, would you agree, that it is possible that some members here, or others retired, that they actually use SWR as their own methodology?

 

[youbet]

One reason I like historical estimates like FIRECalc is that there are so few assumptions made, just cranks out how well we would have done in the past, to answer a simple question - are we anywhere near reasonable.

As an individual, I don't see why we want to spend any time trying to predict an unknowable future rate of return. Certainly we cannot forecast sequence of returns, and that could prove a far greater problem for us.

Nicely said.

I'm a fan of basic back testing vs historical data thus a big fan of FireCalc. And I'm a big proponent of giving strong consideration to "sequence of returns" and thus again a big fan of FireCalc.

One of my favorite attributes of FireCalc is that the huge amount of variability we are subject to is put right in our faces with the output graph. Will I just make it? Will my results be closer to average? Will I wind up with multiples of my original portfolio value? All possibilities. Makes the game fascinating!

An earlier (and I thought better) version of FireCalc that included a graph of ending values (Y axis) vs beginning year (X axis). You could glance and see which year(s) were the beginning years of test periods that resulted in high or low ending values. If you had two or three failure years, you could glance and see which years they were and think about conditions in those time periods. For example, I was surprised to see that test periods that began with high inflation were the toughest on retirees, even tougher than the Great Depression. Etc. It was quite educational. You can still figure out the same thing in current FireCalc, but you have to drop a spreadsheet and do a little work.

 

[youbet]

You do agree though, that there is a body of study and articles that describe and detail an actual retirement methodology based on "Safe Withdrawal Rate"?

No, not as you describe it. I've seen references to the original studies that historically tested WR's vs portfolio depletion. And, more commonly, I've seen the 4% figure thrown out as a tool to calculate the goal portfolio size a person should shoot for during accumulation stage. But I'm really not aware of a "body of study" (as you put it) that suggests a person withdraw some set percentage of their portfolio regardless of other sources of income or their spending needs. I'm sure some talking head or popular media writer has stated things that way, but they've stated anything you can think of in ways not originally intended. But no, "no body of study" calling for some set percentage WR regardless of spending needs or other sources of income.

And, would you agree, that it is possible that some members here, or others retired, that they actually use SWR as their own methodology?

Hey, out of the thousands of members, we probably have one who does it that way while dressed as Henry VIII and smoking dope. But that's not the common trend I read here.

I think you'll agree that there's nothing unique or special about your methodology. You start with a withdrawal level and then "check" it. That's just what FireCalc does. FireCalc backtests vs historical data. You say you have your own methodology. Fine. In ten or twenty or thirty years we can all meet here and compare notes on how we did. There will be no way to know the outcome until the time has passed.

 

[BBQ-Nut]

No, not as you describe it. I've seen references to the original studies that historically tested WR's vs portfolio depletion. And, more commonly, I've seen the 4% figure thrown out as a tool to calculate the goal portfolio size a person should shoot for during accumulation stage. But I'm really not aware of a "body of study" (as you put it) that suggests a person withdraw some set percentage of their portfolio regardless of other sources of income or their spending needs. I'm sure some talking head or popular media writer has stated things that way, but they've stated anything you can think of in ways not originally intended. But no, "no body of study" calling for some set percentage WR regardless of spending needs or other sources of income.

Hey, out of the thousands of members, we probably have one who does it that way while dressed as Henry VIII and smoking dope. But that's not the common trend I read here.

I think you'll agree that there's nothing unique or special about your methodology. You start with a withdrawal level and then "check" it. That's just what FireCalc does. FireCalc backtests vs historical data. You say you have your own methodology. Fine. In ten or twenty or thirty years we can all meet here and compare notes on how we did. There will be no way to know the outcome until the time has passed.

Ok. A simple google search turns up quite a body of study on the SWR methodology.

But, so as not to go down the pedantic hole - I retire from the field.

 

[youbet]

Ok. A simple google search turns up quite a body of study on the SWR methodology.

Yes, just not as you're describing "SWR methodology."

 

[daylatedollarshort]

This may need a separate thread, but who uses SWR as an actual withdrawal methodology vs a check?

We have mainly fixed income investments and just use a real return percent in our planning. We started out with 2% real return and have adjusted that downward as global interest rates have decreased and it doesn't look like they are coming back up again any time soon.

At a a zero real return over fifty years we can still have a 2% SWR (100 / 50 = 2). We plan to spend that or less. If we spend more we might run out of money.

 

[Major Tom]

The great thing about retirement planning, to my mind at least, is that it conforms to the 80/20 rule (as does pretty much everything in life!) Anyone who puts some thought into basic planning will be way ahead of anyone who lives their financial life by the seat of their pants with no forethought.

I don't use spreadsheets and predicted rates of return, and don't use any tool other than FireCalc. It might seem foolhardy to some to rely on just one financial tool but my WR is ~2.05% of my current portfolio value. That figure doesn't take SS into account, which will only help in the years to come. With what I believe is a relatively conservative WR, I don't feel the need to run my situation through multiple speadsheets and other predictive tools.

Even during the accumulation phase, I didn't perform much in the way of analysis. My general modus operandi was to save as much as I reasonably could, as well as always keeping an eye out for any opportunities to save and invest. A few luckily-timed property purchases and subsequent sales in a fast-rising market added to the pot as well. That, plus the fact that I am single and willing to turn on a dime in order to adapt, was my entire retirement plan.

I definitely understand that some folk are more numerically analytical by nature, but even if you run many detailed predictive scenarios, you never know when or if your situation will change in the future - thus the need to be adaptable.

 

[youbet]

I don't use spreadsheets and predicted rates of return, and don't use any tool other than FireCalc. It might seem foolhardy to some to rely on just one financial tool but my WR is ~2.05% of my current portfolio value. That figure doesn't take SS into account, which will only help in the years to come. With what I believe is a relatively conservative WR, I don't feel the need to run my situation through multiple speadsheets and other predictive tools.

12 to 15 yrs ago, as I was beginning to get interested in FIRE, I did lots and lots of spreadsheets which used varying rates of inflation to increase spending over time and varying rates of return to add to the portfolio over time. These spreadsheets were my favorite thing to think about and play with almost everyday. Recently, while getting ready to junk an old desktop, I found the files on the HD. I hadn't looked at them in years. Today they're only worth a smile and a memory. It all turned out to be totally worthless except for the practice they gave me at thinking about the impacts of inflation and returns over time. And, importantly, that the failure to account for sequence of returns in my formulas made it all pretty worthless.

Today, like you, I primarily rely on FireCalc and try to manage my portfolio (55-40-5) and control expenses the best I can. Ten yrs into FIRE, it's working out fine so far, despite the Great Recession hitting me square in the face two years in. We make no attempts to fine tune spending or investing (other than rebalancing), try to always remember time is more important than money and that Dr Pareto rules!

That, plus the fact that I am single and willing to turn on a dime in order to adapt, was my entire retirement plan.

You and I are at opposite ends here since I'm inundated with family and family issues and was born that way. Still, we've mostly all pulled the plow in a surprisingly coordinated way and I can't really say having the family has been too much of a hindrance. Probably cost me a year or two or three in terms of reaching FI. There are different ways to skin the cat as they say......

By my take of the hundreds of your posts I've read, you've done very well and, importantly, seem happy and satisfied with life. You're being more than quantitative enough, IMHO.

 

[Major Tom]

12 to 15 yrs ago, as I was beginning to get interested in FIRE, I did lots and lots of spreadsheets which used varying rates of inflation to increase spending over time and varying rates of return to add to the portfolio over time. These spreadsheets were my favorite thing to think about and play with almost everyday. Recently, while getting ready to junk an old desktop, I found the files on the HD. I hadn't looked at them in years. Today they're only worth a smile and a memory. It all turned out to be totally worthless except for the practice they gave me at thinking about the impacts of inflation and returns over time. And, importantly, that the failure to account for sequence of returns in my formulas made it all pretty worthless.

I did a more primitive version of this, at a similar stage in my financial development. I have always liked, and felt very comfortable with numbers. For some reason, I have never taken the familiarity and comfort with numbers beyond the basic stage and to the level of slightly more advanced analysis. The most "planning" I ever did was to fill both sides of many, many pieces of scrap paper with handwritten sums, calculating the gradual compounding of my monthly contributions at various fixed rates of return. The bits of paper are long gone now but if I were to come across any of them they would, like your early calculations, elicit a chuckle and flash of recognition. As with you, they were very useful mainly for fixing the idea of the importance of saving and accumulating in my head, as well as the effects of inflation and compounding. Unlike you, I had not even heard the phrase "sequence of returns" and at that point, didn't know what it meant.

 

[I was misinformed]

I'm curious -- do you know why he is going for multiple PhDs? The marginal utility of an extra phd degree is basically zero. Once you have one, you meet the minimum for hiring requirements. In fact, if I was hiring and saw that a candidate had 2 or more PhDs I'd consider this a red flag unless there was a very good reason.

Generally most folks I know that want to move from one research area to another just do it by doing the research, finding new collaborators, etc. No need to get another PhD and subject yourself to being a grad student again.

I agree that pursuing multiple Ph.D degrees is a strange strategy in anything like a conventional career. However, this reminds me of something I heard occasionally when I was young, and Albert Schweitzer (1875-1965) was much more famous than he is today (I suppose), namely that Dr. Schweitzer had set the "world's record for most earned doctoral degrees" - counting his medical degree, and Ph.D.s in each of Music, Theology and Philosophy. I don't know if it is true that he had all these earned doctoral degrees but you can find this claim repeated on some websites today (http://www.mincava.umn.edu/documents/kidshurt2much/bkp051702/Mincava/Book2/AlbertS.asp).

Anyways it seems Dr. Schweitzer spent till age 38 getting his formal education and then the next several decades in a very active life of scholarship and humanitarian service. Such a career reflects a very different world from today I suppose.

Speaking of a different world, I believe this claim about Dr. Schweitzer was made on the Merv Griffin show when Merv had a now largely forgotten figure, Charles Templeton (https://en.wikipedia.or/wiki/Charles_Templeton) as his guest. Try to imagine someone like Templeton, talking about his admiration for Albert Schweitzer's philosophy, on a major late-night TV show today!

Anyways I am drifting far, far from the subject of the thread....

 

[kcowan]

That depends on how much margin for error is in the portfolio and on a person's risk tolerance. Those without pensions will naturally be more cautious.

Along with those of us with non-COLAd pensions!

This may need a separate thread, but who uses SWR as an actual withdrawal methodology vs a check?

My approach is my expenses are what I base my withdrawals on and then I calculate what the SWR is as a check only. Then that is tempered against what the immediate past real returns were against next year's expected real return..

Yes me too. I think of SWR as a planning approach to be tempered with reality.

 

[papadad111]

Along with those of us with non-COLAd pensions!

Yes me too. I think of SWR as a planning approach to be tempered with reality.

Just keep in mind there are millions who did little if any retirement planning and are suddenly or not so suddenly retired or soon to be retired .

I would not doubt that for many of them it's set and forget based on SWR - and some smart author will change SWR to "recharge amount" or some such term. For this population it's spend as much as we can until it's really unsafe. So, it serves as a stop sign, a cautionary warning, a red flag.

Of COURSE for us it's just a planning tool..but by nature we are ...planners.

 

[photoguy]

One reason I like historical estimates like FIRECalc is that there are so few assumptions made, just cranks out how well we would have done in the past, to answer a simple question - are we anywhere near reasonable.

This is also one very strong reason that I like Firecalc - there's little in the way of parameters that a modeler can tweak so the outputs can't easily be manipulated. I remember reading one MC paper by Pfau and seeing ~20 parameters used to generate stock/bond returns. Maybe that's okay for a paired study where two withdrawal methods are being compared but I wouldn't trust any absolute numbers about what was a safe withdrawal percentage.

It is not that I am against modelling, I love it and have done a lot, stochastic, ARIMA, multivariate, etc. I just have a hard time figuring out the point of doing it to figure a SWR. We put in our assumptions (based on some info from the past) and churn out results. How it changes our estimated SWR, other than what we can already get from simplistic FIRECalc-like estimates, I haven't a clue. We are just pretending we know more than is possible.

I don't think we get a lot of precision from financial models (either expected return or MC models for determining SWR). But I do think they may be helpful directionally, (i.e.if a historical study suggests 4% was safe, then we should reduce/increase it based on portfolio expectations). I'd also be concerned if my W.R was lower than my expected portfolio return.

I don't consider myself a market timer but if the equity risk premium got too low (outlier levels), I would considered reducing my stock allocation in favor of bonds/cash/something else.

 

[Meadbh]

For those already retired why do you forecast total returns at all? I've been retired 9 years and just spend what divs I get. These are very easy to forecast. Some years total return is great others not so much. Why forecast total returns other than to project what my heirs might get? Actual divs seem way more useful and certain? Even if you are not a div investor, why forecast returns once retired?

Here's an article from today's Globe and Mail on spending dividends.

Why high-net-worth investors should avoid living on dividends alone - The Globe and Mail

 

[Danmar]

Here's an article from today's Globe and Mail on spending dividends.

Why high-net-worth investors should avoid living on dividends alone - The Globe and Mail

Thanks. That was a good article. Agree that my income approach is not optimal. Problem is I kind of got stuck with it, ie I needed to own a lot of my employer's stock up until 6 months after I retired. By then the shares had appreciated so much, the cap gains tax would have been prohibitive to switch out. Couple that with the obvious knowledge and comfort ( not to mention continuing excellent performance) I have with this stock, I have been slow at switching out. So do I stick with the company I know, or look for something else, probably in the US market.

The recently announced tax increases also make my income approach less advantageous. Divs are taxed at about 30% while cap gains would be 24%. Also cap gains are only on the actual gain, not the whole proceeds.

Just hard to find something that might be better than what I have. CAGR total return since 1997 is about 12%. Got any ideas? Doubt index ETF's would match this, especially based on real returns people are forecasting in this thread.