SWR for healthy 40-year-old

[REWahoo]

Don't get me wrong, I think FIRECalc is an outstanding tool, but users need to be aware of all the variables and their substantial influences.

Yep, that variable is history. If the future is unlike the past, FIRECalc outputs will be inaccurate.

 

[smjsl]

My portfolio yields about 3.25% (100% equity) If I only spend the dividends how can I run out of money? Sure all the companies can go bankrupt but other than that and considering they are all the bluest of blue chips and virtually no dividend cuts in the last 75 years? I was one of the guys who said 1%

Danmar, with this approach you could *possibly* run into 2 issues:

(1) With a 60-year-horizon (if you were 40-year-old), you are betting these companies will still be around and continue to pay that dividend for a very long time... Was not BAC, C, GM, and many other companies also considered "bluest of blue chips" not that long ago?

(2) Even if you do continue to get 3.25%, high inflation may make these 3.25% less than what you need to live on requiring you to sell some equities. (Of course, at least in US history, principal appreciation and historical dividend growth took care of this, and so this had not been a problem, at least in the past in US... unless someone corrects me here.)

Since you responded 1% to the poll, then you clearly do not believe 3.25% might be a "safe" WR (for a 40-year-old)... So, I am not sure how to reconcile you poll vote with your reasoning then... On a second thought, I think I do - you mentioned that you consider dividends separately somehow. Most folks think of them as part of the growth of the portfolio and independent of how much you decide to withdraw; so I wonder if your real answer would then be 4.25%.

 

[youbet]

Yep, that variable is history. If the future is unlike the past, FIRECalc outputs will be inaccurate.

Not to mince words, but FireCalc outputs will be 100% accurate even if the future is unlike the past. FireCalc does not predict, it clearly states what it is doing and that is testing a set of FIRE assumptions against historical market and inflation performance.

However, I know what ya mean....... If FireCalc tests your assumptions and says that your portfolio would have historically survived 95% of the time, you'd like to think that means there is a 95% chance your portfolio will survive in the future. But alas, FireCalc makes no such claim and clearly states so in the instructions.

A little leap of faith is in order. Not a lot different from the one we take when we leave the womb into who knows what historical and circumstantial conditions......

 

[Danmar]

Danmar, with this approach you could *possibly* run into 2 issues:

(1) With a 60-year-horizon (if you were 40-year-old), you are betting these companies will still be around and continue to pay that dividend for a very long time... Was not BAC, C, GM, and many other companies also considered "bluest of blue chips" not that long ago?

(2) Even if you do continue to get 3.25%, high inflation may make these 3.25% less than what you need to live on requiring you to sell some equities. (Of course, at least in US history, principal appreciation and historical dividend growth took care of this, and so this had not been a problem, at least in the past in US... unless someone corrects me here.)

Since you responded 1% to the poll, then you clearly do not believe 3.25% might be a "safe" WR (for a 40-year-old)... So, I am not sure how to reconcile you poll vote with your reasoning then... On a second thought, I think I do - you mentioned that you consider dividends separately somehow. Most folks think of them as part of the growth of the portfolio and independent of how much you decide to withdraw; so I wonder if your real answer would then be 4.25%.

Thanks for the thoughtful response. I have to reassess my poll choice. More like my current div yield-say 3%. I agree that inflation is my biggest risk. Dividends only represent about half my projected spending and the spending levels were based on pension and portfolio yield, so I could easily reduce spending if necessary. Also have about 3 years of spending in cash. 60 on my next birthday so I don't see running into problems. My real issue is how much more to spend or give away.

 

[DivinDave]

I'd aim for 2% or less. From age 40 you've got to make it last a long time.

 

[Midpack]

Yep, that variable is history. If the future is unlike the past, FIRECalc outputs will be inaccurate.

That my problem, I don't believe the future will mimic the past. I think there is plenty of cause for optimism in our future, just not as much as history would suggest. Hence a small additional margin of safety beyond history for my decision, say 1% in WR. I'm at 2.6% now (still working), not comfortable yet, but very close. YMMV

 

[Gone4Good]

This is very disturbing to an old retired engineer. I want solutions to be stable and relatively insensitive to input parameters. If the solution is on a part of the curve with a very steep slope, like it seems to be here, I get nervous.

Here it seems like the difference between 3% and 4% withdrawal is like the difference between easy street and eating cat food.

Have I screwed up (again)? Any Firecalc gurus out there?

I don't think you screwed up. I think the problem you're seeing is with FIRECalc's definition of "success". Success means having $1 left at the end of the period. So there are potentially a lot of very marginally successful portfolios in any FIRECalc run. Change any assumption and those "successes" tip easily to "failures". Most of us wouldn't consider running down our portfolio to the last dollar to be very "successful". One way to adjust for this is to set a minimum net worth requirement in FIRECalc. That also helps to highlight the benefit of lowering the SWR below levels where the standard run produces 100% success rates.

 

[walkinwood]

I use a constant percentage of portfolio like Audrey and a few others stated. I currently use 4%, but if I include the management expense of my mutual funds, it is closer to 4.5%. With possible longevity of 50+ years, I do not feel comfortable that there are enough unique periods to trust any predictions based on historical data.

I initially thought of using the Clyatt 4%/95% solution, but have shied away from using the 95% backstop in 2009. We managed with 4% of our much reduced portfolio - which has given us a great deal of confidence in facing the future.

I haven't found a tool that shows the variance in annual withdrawals using historical data for a constant % withdrawal. For example, for a 60/40 portfolio and a 4% withdrawal rate, what would the withdrawals for each year look like historically?

Anyone know of such a tool?

 

[walkinwood]

To test the variation of SWR using a constant percentage of portfolio method, I put a simple spreadsheet together using two funds. Vanguard's total market and intermediate bond index.

You can change the allocation & SWR and see the effect of an SWR as a constant % of portfolio. Values are adjusted for inflation.

Please check the calculations - I did this very quickly and there may be errors. You can unprotect the sheet - it isn't password protected. There are no macros.

I got the data from Simba's spreadsheet
Bogleheads :: View topic - Spreadsheet for backtesting (includes TrevH's data)

 

[MichaelB]

40 years old means 60 years in retirement with no improvement in standard of living. Imagine living the “Leave it to Beaver” lifestyle today. Not satisfactory.

Shiller PE10 says future returns will be low. I’d want some room to at least let my standard of living rise along with the rest of the world. A 40 year old has a very good chance of a life-impacting event – like marriage, divorce, kids.

My 3% is not a recommendation, it is the absolute minimum.

Has this ever been discussed before on this forum?

 

[jIMOh]

This has been my strategy also. But when we talk of SWR aren't we talking about principle not income. eg If I earn 3% dividend yield isn't the SWR is on top of that? Also if we never sell anything aren't we going to leave a pretty big estate?

It depends...
I see Two answers...

If you have a $1 M portfolio the day you retire, and need $30k in income, that is 3% SWR. Its not "on top of" anything... my thought was to position as much of the base portfolio into dividend paying stocks, and have the dividends create the 30k... then use the balance to invest in cash as a fall back plan.

For example if I have a group of stocks which can yield 3.75% and need the 30k income and have the 1 M portfolio...

I would invest 800k into the stocks (yielding 3.75%) and then invest 200k into cash and bonds. (for safety). Considering in this example that 200k is almost 7 years expenses, the probability market goes down, takes the dividends with it, and does not recover for 8 years is minimal to non existant (and if this happens, there will be bigger problems than finances IMO).

The dividends are just one aspect of the equity portfolio- they should confidently payout 30k in dividends whether the 800k grows to 840k (5% gain) or shrinks to 700k (12% loss). The focus is on the 30k payout, and that payout should increase on average year over year.


My second answer would then mention using buckets

For example put 90k (3 years expenses) in cash- this is bucket 1
put 800k (the dividends) in bucket 2
put the balance 110k into a more diversified portfolio of stocks and bonds

So the 800k kicks off 30k in annual dividends, which is spent each year. If the 800 stops supplying enough dividends (market drop) a portion of bucket 3 is liquidated to add to bucket 2 so there are enough dividends coming out of bucket 2.

Bucket 1 has enough cash that there is a 3 year window for a 2008 type market event to spend that cash and wait for dividends to recover without selling anything at a 25-50% loss.

 

[walkinwood]

It depends...
I see Two answers...

Sounds reasonable.

How well did this method work historically?

 

[jIMOh]

Sounds reasonable.

How well did this method work historically?

I am still w*rking and accumulating. I have a dividend paying mutual fund as core of my portfolio, but have not ventured into dividend paying stocks (yet) because that takes more time than I have to manage (right now).

The two strategies I listed came from reading various strategies on this board and adjusting them for my own risk tolerance. I believe a couple people in this thread (clifp and others) have put portions of what I suggest into practice (because they are already retired).

 

[clifp]

Since the question was raised in a new thread about how much should a very early retiree should take as a SWR. I thought it would be interesting to see if the current bull market has changed anyone opinion.

I debated between 3 and 3.5% at the time of the thread. With the market hitting all time highs I'd definitely cut back to 3% for someone in their 30 or 40%.

But my real answer is that very early retirees withdrawal rates should be less than or equal to their income, with some allowance for dipping into cash/cd reserves during bad years likes 2008/2009 when both dividend and interest income are cut back. While using great years like 2013 to refill their cash/cd reserve or in my case going ahead with deferred house projects. 40 it is just to young to start dipping into your principal.

As it turns outs a few months before this thread was started (in late 2009) my dividend and interest income hit a multi year low. Since that time my income is up 29% but of course my portfolio is up more than 60%. My current yield on my portfolio is 2.75%.

My observation was that during the crash the income volatility was significantly lower (roughly 1/2) than portfolio volatility. So It was interesting to see that is true on the way up also.

This does require one to construct a portfolio with a emphasis on income. But this doesn't mean having to buy individual dividend stocks like I have. Good old Psst Wellesley has current yield of 2.75% and trailing yield of 3% while Wellington is .5% less.

 

[Ready]

Firecalc tells me that with a 50/50 portfolio and a 50 year time frame, I can withdraw 3.16% with 100% success. So for those who are voting in the 3% range, there seems to be some historical evidence to suggest this is a good strategy. I understand there are fewer runs possible when you expand the time frame to 50 years, but even dropping it down to 45 or 40 years, the numbers don't change that much.

One exercise I ran to make me a bit more comfortable about the long time frame was to assume a 30% drop in equities as of tomorrow, the beginning of the 50 year run. While the amount available to me to withdraw clearly drops, it still stays within a range that I can live on. So even if we have a terrible bear market tomorrow, Firecalc says I will be OK. Not quite as comfortable as before the drop, but still OK. That gave me some comfort to see.

 

[Zathras]

My experience also supports clifp's observation.
Income stream from interest and dividends is much less affected by both UPS and downs in the market than portfolio balance.
With the experience of going through the last recession intact, we are more confident than ever that out 3% withdrawal (all dividends and interest) is sustainable.
Of course, nothing is ever 100%...

 

[audreyh1]

I do a fixed % withdrawal each year, and in "good" years like this one (knock on wood) I'm glad that means taking out a bigger absolute chunk now before is possibly goes "poof" and my withdrawal goes down.

My experience with withdrawals is my tax rate seems to go up in good years and down in bad, so my after-tax "spendable" income is much less variable than my pre-tax withdrawal.

FWIW - mid 50s and 3.33% withdrawal of whatever my portfolio value is each Dec 31.

 

[Major Tom]

Just saw this thread and I don't think I have contributed to it yet.

I began withdrawing from my portfolio about 2 1/2 years ago at the age of 47 at a rate of 2.5% of the initial value. Hopefully my retirement will be over 40 years (fingers crossed). I am going to hold off on inflation adjustments to the withdrawals for as long as possible, but eventually hope to be able to give myself a worthy raise. The promise of SS should help to give me confidence in raising my WR a little (maybe to 3%, or I may even do a reset based on a new portfolio value).

I voted for 2.5%.

 

[scrabbler1]

Just saw this thread and I don't think I have contributed to it yet.

I began withdrawing from my portfolio about 2 1/2 years ago at the age of 47 at a rate of 2.5% of the initial value. Hopefully my retirement will be over 40 years (fingers crossed). I am going to hold off on inflation adjustments to the withdrawals for as long as possible, but eventually hope to be able to give myself a worthy raise. The promise of SS should help to give me confidence in raising my WR a little (maybe to 3%, or I may even do a reset based on a new portfolio value).

I voted for 2.5%.

This s pretty much my story, too. I ERed 5 years ago at age 45 and have had a SWR of about 2.5% from 2009-2011, dropping below 2% last year mainly because I had switched to a bare-bones HI plan which will be discontinued at the end of this year thanks to the ACA (I am actually happy about his because I will no longer be underinsured next year).

Inflation adjustments have been tough to predict because I have had some one-time increases and decreases in various expense items unrelated to inflation (such as the switch to the HI plan I mentioned above). Furthermore, once two of my three reinforcements (SS and my frozen company pension) kick in starting in about 9 years, I expect my SWR to decrease.

 

[Danmar]

Just saw this thread again. I now understand how divs play into SWR. I am only spending income from my portfolio. ER at 56. Our divs have increased significantly since 2010. Current yield is about 3.6%. My portfolio had no significant div decreases during the crises although it isn't as diversified as I would like. I keep a fairly large cash balance that could cushion any decreases if they occurred. I think the chances of our running out are very low and certainly don't justify reducing current spending.

 

[RetireAge50]

Depends on the timing of other sources of income. Early in retirement I expect to take 5% from portfolio then later between 2% and 3% when pension and social security kick in.

 

[robnplunder]

Retiring at 40 means that your money may have to last 50 years (or more). That is a long time and a lot can happen in that time - just look back on the last 50 years and consider some of the things which have happened and how they would affect you if they were to reoccur - much higher tax rates (in many countries in the 1960s and 190s), prolongued double digit inflation (1970s), a very long period of deflation (like Japan). It's also a long period for your personal circumstances to change as well - disability, old age, medical issues etc.

I will be retiring in my mid-40s and intend to maintain the real value of my portfolio indefinitely. Any other approach is, IMHO, too risky. Most of my assets will be in equities and real estate (some bonds) and I intend to live off less than 100% of the income (which i hope will grow in line with inflation over the longer term).

Given that you have social security and health care in the US, you probably have room to be less conservative than me.

+1. A lot can happen in 40 - 50 years to the US/world economy and to a person. I don't think there is a such thing as SWR. Stay on top of investments, and be flexible with WR year to year is the way to go for me if I were so lucky to retire at age 40.

 

[panhead]

Ah yes, the eternal question.....
When i first started actually planning ER and started thinking about what withdrawal rate I would use, it started at 5%. This was 15 years ago, and my risk tolerance was much higher. That reduced to 4% over the next several years, then 3% not that long ago, and now finally, 2.5% which basically matches the income from my portfolio. This is also what I voted. I'm 43 and still accumulating. I'll tell ya, the size of the portfolio looked much better at a 5% draw, and that probably helped to keep me motivated at that time. I'd like to add that even at 2.5%, I would want (and plan on) having options to either raise more cash or reduce expenses. This may involve something as simple as drinking lower cost rum and cancelling cable, to selling my house and buying a cheaper one, or even moving into an RV or overseas, and yes, even some part time w*rk if it presents itself. At a 2.5% draw (basically just dividends) combined with these options, I find it highly unlikely that I (or anyone) would run out of money. If I can keep close to that draw rate until SS kicks in, I would be golden, but inflation could obviously rear it's ugly head which "might" require me to dip into principal or exercise some of those options mentioned. Tools like firecalc are awesome, but I think what it takes to make ER successful is to have other options, and the ability to put them into action if it becomes necessary.