Interesting article on SWR

[Ready]

I found an interesting article on personalcapital.com on withdrawal rates. I am not sure how to post a link to it because it appears you have to be logged in to an account on the site to see the article. I won't post the entire article here to avoid copyright issues, but hopefully someone else can figure out how to paste the table in the article. They suggest that a 50/50 portfolio has a 50% failure rate over 40 years. What I can't figure out is that if they based their assumptions on 8% returns on stocks, 4.5% for bonds, and 3% inflation, what type of simulation are they using such that their sequence of returns exceeds the losses of the great depression, which Firecalc includes in it's analysis and still says I can have a 100% success rate over 40 years at 3.23% WR.

If you haven't used personalcapital.com's web site, it's similar to Mint.com. It aggregates your financial info and gives you a nice snapshot. I find it nicer to use than Mint and actually pretty useful in for doing analysis.


From the article...

https://home.personalcapital.com/pag...retired-people

Odds of Portfolio Survival

First, let’s look at some generic odds of investment portfolio survival with some basic assumptions.

Assumptions used are an 8% annualized return for stocks (20% annual standard deviation), 4.5% for bonds, and 3% inflation. Withdrawals stay constant, but are adjusted for inflation. No fees or transaction costs were included and the assumption was that all money is taxable, and that the tax rate on gains is 15%. In real life, expect higher fees and taxes. Remember taxes may still be due on deferred accounts such as IRAs, so you may have less money than you think.

These assumptions may prove wrong, but they are a good starting point for analysis.

This table shows the chances of a portfolio running out of money at some point before the given time horizon expires. It is segmented by each annual withdrawal rate, starting at 2%, and shows results for several different stock/bond allocation mixes.

Odds of Portfolio Exhaustion Table:


Can someone else show me how to paste the table here? I haven't figured out how to paste graphics into these threads.

[target2019]

Find the article in Google, and paste the link. Here it is :
http://blog.personalcapital.com/whit...etired-people/

[W2R]

Thanks, Target2019.

I think that both the assumptions here, and the assumptions of FIRECalc, are too optimistic for my way of thinking. Nothing guarantees that past performance (either in average return or in sequences) will persist into the future. Still, these estimates of success are helpful IMO, as only a small part of the useful information available before deciding on a WR.

[Midpack]

I don't think many people here would be surprised at a significant failure rate of 4% inflation adjusted withdrawals over 40 years. But it's conspicuous that Personal Capital doesn't disclose how they did their "analysis" to arrive at 50% (FIRECALC does and doesn't pretend to predict the future), and then the PC pitch goes on to say (excerpts):

Quote:

What About Actual History? Looking at historical returns, survival rates are much higher. This makes sense because equities have generally done very well in the last century, better than our forward looking assumptions. It is quite common to see financial planning articles suggest 4% as the maximum safe withdrawal rate. From the tables above, it is easy to see why. It is a good rule. History suggests 5% is also a relatively safe for people with a 35 year time horizon, as long as they own a majority in equities. With our assumptions, this withdrawal rate is pretty risky, but not unreasonable.

It strikes me as a slick sales pitch above all else, showing higher than historical failure rates without providing their methodology, so the reader thinks 'I need these guys help!' YMMV

And note their "simple" fees...
First $250,000:0.95%
Next $250,000:0.90%
Next $500,000:0.85%
Next $4,000,000:0.80%
Remaining:0.75%

A review, upshot is good free tools, too expensive to sign up http://cashcowcouple.com/service-rev...apital-review/

With no methodology the tables are suspect, so I'd hesitate to use them. But here's the table the OP wanted shared for 3% and 4% withdrawal rates, the other WRs have more limited appeal (time horizon across the top and AA on the left).

[Ready]

Target2019, Midpack - thanks for posting the link and table. I have to learn how to do those things!

What surprised me was the 3% SWR predicting a 15% failure rate over 40 years. Firecalc says 3.23% on a 50/50 AA would be 100% success over 40 years. If that includes the great depression, what are these guys predicting will happen over the next 40 years?

[REWahoo]

Quote:

Originally Posted by Ready Firecalc says 3.23% on a 50/50 AA would be 100% success over 40 years. If that includes the great depression, what are these guys predicting will happen over the next 40 years?

One thing they probably factored in is their portfolio killing fee schedule.

[Midpack]

Quote:

Originally Posted by REWahoo One thing they probably factored in is their portfolio killing fee schedule.

Funny, and probably true...

[haha]

Quote:

Originally Posted by Ready Target2019, Midpack - thanks for posting the link and table. I have to learn how to do those things! What surprised me was the 3% SWR predicting a 15% failure rate over 40 years. Firecalc says 3.23% on a 50/50 AA would be 100% success over 40 years. If that includes the great depression, what are these guys predicting will happen over the next 40 years?

There assumptions are completely spelled out, and seem reasonable to optimistic. From the article:

"Looking at historical returns, survival rates are much higher. This makes sense because equities have generally done very well in the last century, better than our forward looking assumptions."

Ha -

[Live And Learn]

according to the OP: "No fees or transaction costs were included and the assumption was that all money is taxable, and that the tax rate on gains is 15%."

I recently had the financial analysis offered at my financial institution revised and it gave me a 22% failure rate for my 50/50 portfolio for 36 years with a WR of 3.25% which sounds similar to what the analysis the OP refers to.

It is obvious that you can skew a Monte Carlo Simulator to show less favorable results. Unfortunately for those inclined to OMY syndrome it's another thing that keeps us on the j*b.

[youbet]

Quote:

Originally Posted by haha There assumptions are completely spelled out, and seem reasonable to optimistic. From the article: "Looking at historical returns, survival rates are much higher. This makes sense because equities have generally done very well in the last century, better than our forward looking assumptions." Ha -

Thanks Ha.

These folks seem to have a POV similar to many here: Equity returns going forward are likely to be a bit lower than those of the past. I have trouble arguing with that view. Opportunities to exploit natural resources, technology, industrialization, population growth, etc., may not be the low hanging fruit of the past century.

I'm also assuming slower growth and corresponding investment returns going forward. That translates to assumed lower SWR's.

One manifestation of this was when I found myself using cheap box wine to make mulled wine for the holidays. Somehow deciding to reduce our WR to less than 3% seemed incongruent with pouring good wine into a sauce pan with sugar and spices totally changing its taste. And, indeed, the box wine + mulling spices turned into just fine mugs of spiced wine in front of the wood stove last night.

Everything will be OK.

[audreyh1]

Quote:

Originally Posted by Live And Learn according to the OP: "No fees or transaction costs were included and the assumption was that all money is taxable, and that the tax rate on gains is 15%."

Well, that is funny business. You have to assume you pay taxes out of what you withdraw. Otherwise, you are in fact testing a higher withdrawal rate. Trying to add taxes into the mix for modeling is a huge can of worms.

Those of us using these models treat taxes as expenses covered by the money we withdraw, not as an additional drag on the portfolio.

[sengsational]

Quote:

Originally Posted by youbet Thanks Ha. These folks seem to have a POV similar to many here: Equity returns going forward are likely to be a bit lower than those of the past. I have trouble arguing with that view. Opportunities to exploit natural resources, technology, industrialization, population growth, etc., may not be the low hanging fruit of the past century. I'm also assuming slower growth and corresponding investment returns going forward. That translates to assumed lower SWR's. One manifestation of this was when I found myself using cheap box wine to make mulled wine for the holidays. Somehow deciding to reduce our WR to less than 3% seemed incongruent with pouring good wine into a sauce pan with sugar and spices totally changing its taste. And, indeed, the box wine + mulling spices turned into just fine mugs of spiced wine in front of the wood stove last night. Everything will be OK.

I agree that everything will be OK, or better. But I'm on the optimistic side for equities. Not that they'll outdo history, but be on par. I think it's human nature to say everything is used-up, but only because we can't see the cool stuff yet to come. Recall the ideas proposed in the past like, the world only needs a few big computers. And I'm not sure the prosperity of the 50's was agreed upon before the fact. We have some great discoveries now that will turn into huge opportunities, and things that will be discovered that can make life better and make cool things possible. There's no reason to think that this random time in history is the time that humanity will quit innovating.

[target2019]

Quote:

Originally Posted by Ready Target2019, Midpack - thanks for posting the link and table. I have to learn how to do those things!

The original link you posted had "how-much-can-you-spend-for-retired-people" in it. I retyped that in Google, and left out the hyphens.

That did not get me there!

So I added the domain name personalcapital I think.

That got me the direct link. You can look at your link and google's and see what was different.

[Ready]

Quote:

Originally Posted by haha There assumptions are completely spelled out, and seem reasonable to optimistic. From the article: "Looking at historical returns, survival rates are much higher. This makes sense because equities have generally done very well in the last century, better than our forward looking assumptions." Ha -

Maybe I'm missing something, but since they specifically spell out their assumptions - 8% equities, 5.5% bonds, 20% standard deviation - what numbers are they using to generate the tables if not these numbers? And if they are using these numbers, how are they getting to these failure rates? They are coming up with higher failure rates than Firecalc, which includes the runs during the great depression.

[Ready]

Quote:

Originally Posted by REWahoo One thing they probably factored in is their portfolio killing fee schedule.

This brings up an interesting point. Part of their software analyzes the expenses in your mutual funds. My funds overall average .26% but some are as high as .9%. They state that anything over .30% is too high, and they recommend I change to index funds to lower overall fees. This all sounds great, until you look at their pitch to sell their advisory services, which range from .75-.95% of assets. What's the point of changing out to index fees and then paying those kind of advisory fees? Seems a bit silly to me.

[Midpack]

Quote:

Originally Posted by haha There assumptions are completely spelled out, and seem reasonable to optimistic. From the article: "Looking at historical returns, survival rates are much higher. This makes sense because equities have generally done very well in the last century, better than our forward looking assumptions." Quote: Originally Posted by Ready Maybe I'm missing something, but since they specifically spell out their assumptions - 8% equities, 5.5% bonds, 20% standard deviation - what numbers are they using to generate the tables if not these numbers? And if they are using these numbers, how are they getting to these failure rates? They are coming up with higher failure rates than Firecalc, which includes the runs during the great depression.

Average returns and standard deviation aren't enough to qualify as "completely spelled out" as you know (sequence of returns for one), so I must be missing your drift.

[W2R]

Like so many of these articles discussing withdrawal rate, they seem to assume that the retiree is going to withdraw the specified amount each year, like a robot or automaton.

Perhaps this is the most unrealistic of their assumptions, or so I gather. From what our members here have said, during a crash or down year they plan to tighten the belt considerably.

[Ready]

Quote:

Originally Posted by Midpack Average returns and standard deviation aren't enough to qualify as "completely spelled out" as you know (sequence of returns for one), so I must be missing your drift.

Well, the figures they do give us of 8% and 5.5% are very rosy. They don't specifically state whether they are using Monte Carlo Simulation, or some other proprietary method, to simulate sequence of returns risks. And they just generally state that they think the future won't be as good as the past, but they don't spell out specifically how they are factoring that in to their calculations to get to the failure rates in the tables. I guess it just seems odd to me to point out such rosy figures to start with, end up with such high failure rates, and leave out the rest of the details of how they started with those figures and got to the failure rates in the tables. Maybe they just don't expect the average reader to analyze their figures in great depth.

[aim-high]

FYI . . . a word about Personal Capital.

When you put your portfolio in the system, you are giving their FA's access to your balances at a minimum. I have my suspicions about them actually being able to see your specific holdings.

[galeno]

Use the following expected gross real returns: FI = 0%. Lg Caps = 4%. EM stocks = 6%. Sm Caps = 6%. REIT = 6%.

Subtract portfolio expense and tax ratios = expected net portfolio returns.

E.g. our 2014 60/40 port's expected net real return = 3.0% - 0.2% - 0.4% = 2.4%.