Canadian Money Forum banner

Another take on SWR

19K views 93 replies 22 participants last post by  steve41 
#1 ·
#2 ·
Somehow we all got bitten by the SWR bug? I've been reading material for the last few days. Here's one view:
http://financialmentor.com/retireme...o-i-need-to-retire/safe-withdrawal-rate/13192

The first point that has me really thinking is that the oft-cited work is based on US markets during the years of incredible US expansion. The analysis has also been attempted on international markets:

Using 109 years of data for each of 17 different developed countries, Pfau determined that a 4% withdrawal rate with a fixed 50/50 asset allocation would have failed in all 17 countries. Yes, a 100% failure rate.
That's obviously a very important result. How come I don't hear this mentioned more often? The US market is unique in the world, as the US has expanded to become a global empire. But here are all of us, basing all of our stock market expectations on the US's history of stellar performance.

To me, it seems unlikely to be repeated. I'm getting the impression that SWR models (and stock market projections in general) are wildly optimistic, going forward.
 
#3 ·
Building more on the international theme, which is relevant both to us as Canadians but also for everyone (since the unique US expansion cannot repeat), see this other article
http://www.mcleanam.com/probability-4-rule-ensure-clients-retire-safely/

This mentions that for the UK, the SWR was 3.43%. Then with more global diversification, he found SWR was 3.26%

I recently exchanged emails with my friend who is an algorithm developer at a major Bay Street investment bank. I described to him my reasoning for why an expected SWR in our situation is 3.3% to 3.5% and he agreed with that range. I actually made that estimate before I saw the above article about the UK & global results, so now my analysis has confirmation from both my investment banker friend, and the international study.

Therefore I am recommending that my parents plan for SWR of 3.3% to 3.5%
 
#4 ·
Yes James, 3% SWR is the new 4%. This is not a big revelation. It's been the consensus in the retirement planning community for a few years now. Future investment returns are very likely to be lower than the past returns.

Note that academic SWR strategy is very rigid. You withdraw a certain percentage of your portfolio in the first year of retirement. From there on, you index your withdrawal amount to inflation, without paying any attention to investment performance of your portfolio. This model may be correct from the academic point of view, but it looks rather silly in real life. Most of us have the flexibility to withdraw more or less depending on the investment performance. Consume more in the bull markets. Consume less in the bear markets. Staying flexible reduces the risk of grinding your portfolio to zero.
 
#5 ·
Variable Percentage Withdrawal

Has anyone investigated Variable Percentage Withdrawal (VPW) strategies?
http://www.financialwisdomforum.org/forum/viewtopic.php?f=30&t=117200
http://www.bogleheads.org/wiki/Variable_percentage_withdrawal

It adjusts withdrawal amounts annually based on portfolio value, and puts a rigorous method on the "Consume more in the bull markets. Consume less in the bear markets." approach mentioned by GoldStone. It really lessens the probability of running out of money in retirement. Also fixed percentage withdrawal (like the 4% rule) tends to leave behind a large estate if the retiree experiences a long period of normal or high investment returns.

Using such a method requires that you have the ability to cut back spending when returns are low, which is fine if you have low non-discretionary spending with areas of discretionary spending like travel or hobbies. It would not work well for LBYM people whose non-discretionary spending is high relative to total spending.
 
#7 ·
In addition to being able to vary spending from your portfolio annually, other decumulation methods will affect a portfolio SWR calculation. For example:
i) Building up a cash wedge to fund your first 5 years or so.
ii) Buying an annuity so that it plus cpp/pension income are enough to cover your basic living costs. Annuities are better bought around age 70. Then withdrawls from your portfolio need only cover your additional discretionary costs.
 
#10 ·
This page is part of a larger article where it compares international markets and their ability to successfully sustain a 4% SWR: http://www.fa-mag.com/news/why-4--could-fail-22881.html?section=47&page=2

I've also seen mentioned a few times that as a corollary to sequence of returns risk, one is more likely to be successful if one retires when the markets are in correction mode rather than at all time peaks.
 
#13 ·
My response to SWR is very simple. Try to find someone who has followed that for 30 years. You won't find anyone. In all of the comments and all the links, I find only one real piece of worthwhile information, ' and then there is the real world.'

Why do people want to find a SWR? Answer, because it would appear to provide a way of knowing if you have enough money to retire on. Why is a 4% SWR so appealing? Answer, because current rates of return are so abysmally low, it looks like most people will never be able to afford to retire unless they can count on making more than they are currently making on investments. Why are people so fixated on stocks and bonds? Are they the only way to invest money?

When I retired 26 years ago, there is no way I could have managed on a 4% return on investment. Nor is there any way anyone would have planned to do so. Even bank interest after adjusting for inflation paid more than that. So what was the thinking back then? Well SWR as a term did not exist for starters. So what anyone knew was that they wanted X amount of income and had Y amount of capital from which to earn it. So the question was how to invest the capital to provide the income required. NO ONE was thinking I can spend more in a year than I earn because there is a study that says I can do so. Think about it, spending more than you earn. Right now in your working life (as most of you seem to be), what does spending more than you earn result in?

I knew how much capital I had to invest. I knew what income I would like to earn from that as a minimum but I also knew there was no guarantee that I would earn that much. I knew that inflation had to be taken into account. I knew I did not want to run out of capital and have to try and go back to work. Those factors all had to be dealt with before I decided to pull the plug.

So here's what I figured. Rule #1. You have to live on what you earn just like you should be doing (but many don't) during your working years. Rule #2. You NEVER spend the capital. Those are the only 2 rules you need to follow. That doesn't make it simple to do of course.

Back then (1989), I could easily count on making 10% nominal (before accounting for inflaltion) in many ways. Even bank interest was around 8%. Bonds, GICs would return more than 10%. Stocks were never on the radar. Stocks are always a gamble and could result in breaking rule #2. Not an option. But interest rates, bonds and GICs would not necessarily provide proof against inflation which could result in a lower income and making it had to stick to rule #1.

So I needed an inflation proof investment that paid 10% or more. As luck would have it, I had an 'in' that allowed me to invest nominal amounts ($25-50k) in commercial and industrial real estate. Rental income is pretty inflation proof in that you can increase rents to cover inflation. Industrial and commercial properties means that you do not have to be a landlord, that is done by property management companies. Smaller investments in multiple properties rather than a large investment in one property means you can average the income without worrying about one property being vacant for a period of time. This is nothing like relying on a couple of rental houses for your income.

That strategy allowed me to pull the plug and it worked fine for several years. But then life took a turn and it was no longer feasible for me to continue in that way. Life has a habit of making changes that you never anticipated. That doesn't stop when you retire. That's also a reason why talking about SWRs makes me laugh. For example, suppose you are a married couple and you retire on your 4% SWR. After a few years, you divorce. What happens to your SWR? Answer, it is down the tubes. There are many things that can happen in life that you cannot predict and that can affect your finances. Where are they in SWR studies?

In the last 26 years, I have never had to return to work but my income has varied and my means of deriving that income has varied as well. I've learned I cannot plan beyond a year financially. Just as I re-visit my budget each year, I re-visit my investment strategy. Some people have commented you can do a 4% SWR but maintain flexibility. That's a laugh. Either you are following a 4% rule OR you are flexible. You can't do both. The answer as far as I am concerned is forget the SWR and just realize you have to be flexible.

If you want to know if you have enough to retire, the answer is that no one can answer that. What you can answer is if you will have enough income for the next year. Beyond that is crysal ball time. This is no different than when you are working for a living. Is there anyone who could not lose their job tomorrow? What then? If there is no guarantee in your working life, why would anyone think they can get a guarantee in retirement? Yet that is in fact what people are looking for when they want to believe in a SWR.
 
#27 ·
Nothing wrong with spending capital. Also, you can find property managers for residential real estate as well.

People are talking about low rates of return. What are these people investing in? TSX returned 20% last year and the US market has tripled since the collapse in '08. Back test the 4% for your portfolio and see if it worked out. I bet it would have.

For those interested, here is Mr money mustaches take on it: http://www.mrmoneymustache.com/2012/05/29/how-much-do-i-need-for-retirement/
 
#14 · (Edited)
Yet another take on SWR as it applies to Canada by Morningstar:

http://cawidgets.morningstar.ca/ArticleTemplate/ArticleGL.aspx?culture=en-CA&id=796865

Only scanned it, but if you want $50k pa of investment income (plus inflation), you need about $1.5 $Million when you retire at 65 (based on a couple and not running out of money in 30 years.

The full report is here: http://video.morningstar.com/ca/Safe_WithdrawalRates_ForRetirees_CA_010517.pdf

One thing pops out - you need a good equity allocation if you use the suggested withdrawal rates.
 
#15 ·
Some people would say you can get $50k/annum from a 100% dividend stock portfolio of $1.5 millon yielding 3.33%, potentially with dividend growth exceeding inflation. And not touch the capital. And not including CPP/OAS. I'd suggest something closer to $1 million is all one needs IF one uses a very cost efficient dividend stock portfoio. Even something like XDV might give that to you.
 
#25 · (Edited)
XDV currently has a yield of about 3.7% or $37,000 on a $1million portfolio. So we draw dividends and some capital for 30 years and don't run out of money??
This type of analysis should really be done after tax, which is of course different for each person/couple. Dividends payed within registered accounts will be taxed same as interest when withdrawn. And once retirees get to 71, they are required to start withdrawals. The minimum withdrawal was reduced, but if taxpayer has a substantial RRIF, those taxes can be significant.

So just trying to say that a 3.7% equity yield doesn't give you $37,000. It gives you substantially less, so withdraw at 4% and you will likely run out of money sooner than later. CPP/OAS will of course help and can be in region of $30k pa (before tax) if both have worked. A million might get you by, IF you are able to live off family income of under $50k.
 
#17 · (Edited)
I personally believe, rightly or wrongly, a tax efficient dividend portfolio worth $1 M, will churn out a healthy $35,000 or so as cash for life. You would not need to touch the capital since stocks such as RY, TRP, CSH.UN, CIBC, etc. that raised their dividends in February, will raise them again to help fight inflation. TD raised theirs in early March and a host of others have raised their dividend since the new year. Will this happen every year for every blue chip stock going forward? Maybe not, probably not but many will raise their dividends over time and you'll get some capital appreciation as well.

Don't want to go this income route? Most of these stocks are in the ETF XIU, a very tax efficient ETF itself that could reasonably churn out $30k per year every year for life.

Top considerations for Canadian dividend ETFs include the one AR pointed out, and XEI, CDZ, VDY, ZDV, and a few others - most of them yielding between 3-4%.

Add in CPP and OAS and for most 60-somethings, that's $50k per year after tax or about $4,000 per month without touching the capital or selling their house (yet). That's pretty good.
 
#31 · (Edited)
I personally believe, rightly or wrongly, a tax efficient dividend portfolio worth $1 M, will churn out a healthy $35,000 or so as cash for life.
Since SWR refers to an initial amount that's inflation adjusted upwards, if you're talking about 35K (constant and not inflation adjusted) out of 1M that's less challenging than 3.5% SWR.

Let's say you add inflation adjustment to the initial 35K, then you are exactly talking about a 3.5% SWR. Yeah, I'd say that's feasible but remember the SWR studies only say that the money lasts for 30 years -- not forever. If your aim is to have the money last 30 years, then I agree with My Own Advisor that this is doable.

The part of MOA's statement that I will challenge is the part about not needing to touch the capital of the stocks, and also the "cash for life" part. Again -- this is not what the SWR studies say. If you're taking 3.5%, the studies don't assure that you will be left with capital. You may or may not totally deplete the portfolio down to zero by the time 30 years is up.

If the aim is to actually preserve capital and not deplete the portfolio, you probably need to go closer to 3.0% SWR

However I retract that statement if MOA literally meant $35,000 without inflation adjustment. Since that is less than 3.5% SWR, it's more doable and I can see how it can continue for life.
 
#33 · (Edited)
And here I am! I almost didn't see this because I've been on flights all day. How lucky that I saw it, eh?

I can see the reactions forming in people's heads, to my post. "If you're only taking out 3.5% using the dividend, then how on earth would you deplete capital given that many stocks easily pay dividends of that size?"

(1) there's still a pervasive misunderstanding of dividends, treating them like newly generated free money. That's not what dividends are. SWR studies don't treat dividends in any special way... they are analyses of total returns, because that's all that matters. Dividends are not free money; they are part of total return.

(2) sequence of return risk and LUCK. This is the reason that capital doesn't last forever even though you're extracting less than the average long term growth rate. If you get unlucky and get a bad series of losses, then you can deplete the capital even at 3.5%. Heck, you can even deplete capital at 3.0% withdrawal. It's a matter of luck... SWR studies sweep a very wide range of probable outcomes.

(3) the sad reality is that the SWR studies have told us that "portfolio failure" (capital depletion) is a possibility even if you're only extracting 3.0% to 4.0% of the initial amount, plus inflation adjustment. Yes, I really do mean capital depletion... you are NOT assured to preserve your capital, despite dividend growth, despite long term high stock returns. But it's about probability of portfolio failure. Then the question becomes, how confident do you want to be that your money will last? Maybe 90% probability that your money will last 30 years is good enough for you.

(4) Remember that the studies look at 30 years of capital preservation. Not indefinite. If their simulation shows that you end up with $2 left in year 30, that's considered success. This is probably not what many people interpret when they hear the term "sustainable withdrawal rate"

(5) because sequence of return risk is an important factor, and it's not all about high returns, fixed income is also recommended. Taking a pure stock allocation does not make your capital last longer. Source: SWR studies. People around here seem to love the idea of outlandishly high stock allocations, even though SWR studies say that your money will last longer if you have some fixed income. Maybe as much as 50% fixed income.

(6) the good news is that you will be OK as long as you can be flexible about how much you extract from the portfolio. If you reduce your withdrawals in bad years of the stock market, the money will last longer. Taking a dividend is a form of extracting cash from the portfolio. Reinvesting the dividend instead would be leaving the money in the portfolio.

(7) let me add this just to stir up more trouble. Taking a cash dividend is just as destructive to a portfolio as selling shares during a steep drop in prices. They are equivalent. So dividends do not add inherent safety or sustainability to a withdrawal regime, nor do they insulate you from a market crash. If you have a stock that pays big dividends, and you take those dividends as cash instead of reinvesting during depressed prices, you are causing harm to your portfolio.

Proof of (7) is, what would happen if you reinvested those dividends instead of taking them out as cash? At the very depressed stock prices, your reinvestment of the dividend will produce a very high return going forward. This is the return you deprive yourself of by taking the dividend out as cash.
 
#20 ·
No, it isn't mainstream thinking AND not something the vast majority of people could, or would, take a risk on. Most investors need a balanced portfolio in retirement to be able to sleep at night AND avoid the inevitable roller coaster rides of financial crises, etc.

That said, there are at least a few here that believe strongly in a 'very high' dividend stock portfolio and further believe the dividend income stream (and some growth in it) will carry the day. It is not for the faint of heart AND it will take a cast iron stomach when (or if) bond yields significantly rise. We all know that dividend growth rate on many dividend stock sectors will stall in event of material interest rate increases. [Think capital intensive sectors like telecoms, pipelines and utilities for some examples]. Others like banks and insurers will continue to grow their dividends.
 
#22 ·
For sure. I carry about a 15-20% component in FI (bonds, debentures, GICs, HISA) to supplement withdrawals in 'bad' years in the equity markets. Also circa 8% of my equity is pretty reliable REIT income, and another 7% in preferreds. That is about as secure as I feel I need to be.
 
#26 ·
I agree it depends on one's definition when they talk about annual cash flow needs. When I consider numbers like $37k or $45k as mentioned in this thread, I assume BT dollars and I have to pay income tax out of that. IOW, income taxes are a 'spend' just like property taxes, food, etc.
 
#34 · (Edited)
James, you're taking the discussion too literally. $1million today generating $35/yr of income a year will climb each year too.... i.e. the $1 mil will grow with capital appreciation and the $35k/yr will grow with dividend growth, likely outpacing inflation. If a person does not have to tap into capital (and that is the caveat), the portfolio will only get bigger over time!

Added: There will be some down years in capital of course, i.e. the vagarities of the market. It is highly unlikely the dividend stream will decrease in a bear market and even if it does, it does so marginally.... easily within the retiree's ability to adjust living expenses to compensate.

Added: What any smart retiree will do is to use the VPW methodology, i.e. Variable Percentage Withdrawal. Easy enough to play with $1 million that way.
 
#36 · (Edited)
James, you're taking the discussion too literally. $1million today generating $35/yr of income a year will climb each year too.... i.e. the $1 mil will grow with capital appreciation and the $35k/yr will grow with dividend growth, likely outpacing inflation. If a person does not have to tap into capital (and that is the caveat), the portfolio will only get bigger over time!
Counter-example to show that this is not assured:

Let's say that you start this withdrawal regime at the beginning point of a 10 year bear market. Let's say this plays out as a 50% stock decline, followed by depressed prices for 10 years before any recovery happens.

Portfolio is 1000K
Portfolio is down to 965K after you take out your 35K in the first year
Portfolio crashes to 482K because you refuse to own bonds
Now you have nine more years of a sideways market and no stock gains.
Your 482K shrinks to about 160K after the withdrawals.
Portfolio is now 160K
And now there's explosively high stock growth, huge bull market
Ten years later, average stock growth is still the usual 7% annual

This is a damned ugly scenario, and like it or not, this is a possibility. Just 10 years into this retirement, your 1000K portfolio is down to 160K.

That's why it's not true that "the portfolio will only get bigger over time". There are a wide range of possible outcomes.
 
#35 ·
AltaRed, why are you eyeballing this stuff when several people have done detailed simulations on exactly this, in studies that have been repeated multiple times? I know it feels like dividend growth will make the capital last forever, but that's not what the simulations show.

In other words, don't take my word for it. Read the studies. And those studies are the basis for all the wealth management and retirement planning... they are important results, especially because the results are somewhat counterintuitive.
 
#37 · (Edited)
Not true if one is also getting a $35k dividend stream. There is no decrease of capital to 965k for that reason at all. It might be due to a bear market, but then one is not touching capital, or at least not that much.

And it is not nearly as bad if you use VPW methodology because the withdrawal would drop substantially if the capital value of that portfoilo went down. IOW, one learns to cut back their spending and intuitively, anyone with more functioning brain cells than Trump would cut back spending anyway. VPW is a logical common sense approach to withdrawals. I'd be willing to go with an all equity portfolio with that concept starting at 3.5% withdrawal rate.

Added: Eyeballing is crystal clear when I have looked at VPW work extensively. Success does depend on what withdrawal rate you start with of course, and what one has their investments in. If my starting point is a dividend stream of 3.5% yield and my starting withdrawal is 3.5%, the only way that can go south at all is if the dividend stream falls and I don't march lockstep with that. Basic arithmetic.

Added yet again: James, I am among the group that cheerleads appropriate asset allocations, re-balancing and (some) traditional (or modified) rules of thumb for the bulk of investors. Which is why VPW is much better than the classic SWR. But those who can set themselves up with a realistic dividend stream that is not reaching stupidly for yield, i.e. blue chips only, and don't have to touch capital really don't need anything else in their portfolio. I will have to check final numbers but I think I am almost dead on 3% yield on my entire portfolio allowing me to pretty much NOT have to touch capital except for discretionary extras.
 
#41 · (Edited)
I didn't forget. The dividend is just part of total return. My numbers are with total return, i.e. 0% total return followed by -50% total return. And then the good side of the recovery, with a whopping 17% total return in the recovery years. All those numbers are plausible and within the range of what the market has delivered in the past.

The 50% crash that occurred in 2008 was still -50% even including dividends

As mentioned, this sequence of returns I demonstrated ends up giving about +5% average annual return. Very feasible.

So in my numbers, the dividends still exist, and they still pay out as you expect, but the total return goes from 0% to -50% to +17%.

And you're right of course, variable strategies give a ton of flexibility and are the best solution. But I post this stuff at midnight because I worry about what happens if people think that dividends guarantee preservation of capital. As I demonstrated in post #38, there are sequences of returns that can happen, even with dividends doing their expected thing, that deplete your capital awfully quickly.
 
#42 ·
I can see where you and I have passed each other in the night. The data stream I was using was capital only, not the income stream thrown off by it and that is how a number of folk who work on the 'dividend stream' concept do it. They pay attention only to the cash flow thrown off by the portfolio and don't spend more than that. Cannew will go on and on about that exact methodology.

For myself, I look at total return of course, but I sure do look at my total income for the year and measure my spend against it....so that I know when/if I am dipping into true capital, and can decide if that fits my 'plan'.
 
#43 · (Edited)
The 4% SWR determined by the Bengen and Trinity studies is based on the maximum portfolio withdrawal rates from a low-cost balanced portfolio that had a high probability of surviving 30 year retirements during the worst investment returns of the study periods, for people that retired in years from 1926 through 1976. Those worst periods were retirees in 1929, 1937, and several years in 1963 through 1969. Portfolios of people that retired in most other years could have survived a much higher SWR.

Interesting dichotomy is some people saying 4% is no longer a sustainable safe withdrawal rate because current valuation levels and interest rates mean future portfolio returns will be lower than the Bengen & Trinity study periods. Other people say no-problemo... have a 4 or 5% dividend yield so assuming dividends grow at least at inflation I can withdraw that much and never spend any capital. But remember that SWR is determined on what is needed to not exhaust a portfolio through really bad economic times, not the current sunshine and roses 8th year of a bull market always increasing yields return environment.

During the study periods, most people that set an initial 4% WR and adjusted it for inflation each year would have ended their retirement with way more money than they started. They would over-save before retirement and under-spend during retirement, just in case they hit a really bad patch early in retirement. Which demonstrates the strength of variable withdrawal strategies, especially VPW.
 
#44 ·
Interesting dichotomy is some people saying 4% is no longer a sustainable safe withdrawal rate because current valuation levels and interest rates mean future portfolio returns will be lower than the Bengen & Trinity study periods. Other people say no-problemo...
The posters here may have said that, but I would go back and read the Morningstar report that we were discussing. I don't think everyone here did. It is a Canada oriented update on the B&T studies and worth reading. For those in retirement with say 10-30 yr outlook, using current yields on equity and FI and adjusting the 4% down a bit, seems to me to make more sense that data from the past century.

It's not surprising that we have different views on a forum like this. Some here likely have enough saved that they can live off their dividend and interest income (plus CPP/OAS). Others won't have saved enough to do that, and will have to spend some of their capital. The latter group are the ones that need to adjust their lifestyle to suit expected income. And use a more conservative withdrawal rate.
 
#45 ·
+1^ GreatLaker, good post to put things back into perspective.

A person's view of the world and where they sit with respect to SWR, VPW, etc. depends partly on where they are at on the spectrum of accumulation (or withdrawal), their investing acumen, the size of their portfolio and their safety net of CPP and OAS in particular. Those with DBs are in yet another domain. That said, I am not a fan of classic SWR since few qualify its use, e.g. low cost balanced portfolio, 30 years, etc. and few understand the rigidity of its classic interpretation. The vast majority of investors who likely would adopt it 'hook, line and sinker' are also likely the ones with high cost portfolios of MFs and/or portfolio churn, and that could be its undoing.

We also need to understand the wealth management and retirement planning industry have a conflict of interest in perpetuating investor fear and dependence.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top