Sustainable Withdrawal Rates

[steves]

And here, with the math (including taxation) is what Georgia would need to earn on her $500K RRSP alone, in order to make it out (just) to age 95..... Georgia's plan earning 3.28%

[crystals]

Suppose she takes $100K of RRSP money and buys an annuity with a 10-year guarantee. http://www.canadianbusiness.com/my_mone ... /index.jsp says she can get $618.27/month from Manulife.

Are these annuity payout rates indexed to inflation?
If not, is there any tool I can use to find out how much premium I need to get an annuity which guarantees a certain indexed monthly income for life?

[Flights of Fancy]

Generally speaking, discount the expected payout (from the link above) by 25% for a rough rule-of-thumb "indexation" premium. Or get a quote from an annuity broker. And if you want average monthly payouts for males and females at varying ages, go here. Those tables are populated with data from CANNEX.

[crystals]

Ok, so we have Mr. Novice and Mrs. Novice. They want to know the premium they need at age 65 to purchase an annuity for joint life.

Let's say they want to have $70K to spend anunally. (everything in today's dollar).
They need $80k income before tax. ($15K each no tax, $25K each taxed at 20%).
Let's say they have together $30K per year from OAS, CPP, and a fuly index pension.
So they would like to get an annuity that gives them $50K per year indexed payout.
That's $4167 per month payout from inflation-indexed annuity.

Now use this table, a joint life for 65 year olds is about $540 per month for $100K premium.
That means the couple need $77.2K premium to get a monthly payout of $4167.

Now use the "discount the expected payout by 25% for a rough rule-of-thumb "indexation" premium" suggested by Flights of Fancy. The premium the couple need would be $77.2/0.75=$1.03M.

Is the above more or less right?

[Flights of Fancy]

More or less right, but at this point you are dealing in so many generalities I am starting to get uncomfortable.

One big issue is that the payout rates for annuities differ quite substantially depending on whether you purchase the funds with registered (RRSP) dollars or not.

I suggest you get some quotes from an annuity broker.

Edited to say I realize I am repeating myself here with the "get thee to a broker" advice.

Coming back one more time. I'm curious to know why you want to annuitize ALL your income needs, instead of just some. You will increase the sustainability of your income, but at the expense of any financial legacy.

[BRIAN5000]

One of the calculators I'm playing with allows you to adjust the Std. Dev of your proposed Inflation rate. What would be a reasonable percent Std Dev....be with a 4% inflation rate?

[Flights of Fancy]

Well, inflation over the past 60 years has ranged from 12% (1980) to zero (1950). The SD will depend on how many years you want to include in your sample.

Not that you asked, but longevity is actually much more variable (subject to a bigger SD) than either inflation or stock market returns.

[Bylo Selhi]

Time to replace the 4% withdrawal rule?

Conventional wisdom suggests that you withdraw on average 4% adjusted for inflation. Now comes a paper co-authored by William Sharpe, the winner of the 1990 Nobel Prize in Economics, challenging the conventional wisdom.

"It is time to replace the 4% rule with approaches better grounded in fundamental economic analysis," wrote Sharpe in his paper "The 4% Rule -- At What Price?"

"Supporting a constant spending plan using a volatile investment policy is fundamentally flawed. A retiree using a 4% rule faces spending shortfalls when risky investments underperform, may accumulate wasted surpluses when they outperform and, in any case, could likely purchase exactly the same spending distributions more cheaply."...

Sharpe's study, in essence, shows that retirees waste money by adopting the 4% rule. "The 4% rule's approach to spending and investing wastes a significant portion of a retiree's savings and is thus prima facie inefficient," Sharpe wrote...

E. Tylor Claggett, a finance professor at the Perdue School of Business at Salisbury University, said the question of whether it's time to toss the 4% rule in the circular file is an old one and it has always been perplexing.

"The truth is, no one has a crystal ball," he said. "Therefore, no one knows how long the retiree will live, what his or her actual future financial needs will be (due to health issues and the like) and the future year-to-year performances of the various capital market components. If all of these were known, we would not have to have this discussion. Instead, we are left with looking for 'rules-of-thumb' to increase the probability that a retiree's needs will be met given his or her asset base at the time of retirement."...

So instead of slavishly following any approach, Spiegelman says it's always a good idea to remain flexible. "The 4% rule is easy to understand and follow, and provides a good starting point for the average investor looking for a ballpark idea of how much they need to save or, conversely, a ballpark estimate of how much they can safely withdraw," he said.

Meanwhile, Stephen P. Utkus, a principal with the Vanguard Center for Retirement Research, agrees that the 4% rule is flawed. But he also notes, as did Sharpe, that there's no practical mechanism to replace it with and that further research is required...

[steves]

I seem to recall not too many years/decades ago, the feds announced that they would "try" to keep inflation at 3%. So far, they seem to have been successful. I wouldn't factor in much SD into that determination, certainly not basing it on those wild and woolly inflation rates from back in the day.

[BRIAN5000]

OMG, ok I'm on page 16 of the 24 page explanation, I think I'm lost.

[couponstrip]

Simple question I hope. Is the 4% (of the original portfolio value at retirement) safe withdrawal rate "rule of thumb" meant to be used before or after tax on the withdrawal?

For example, if I decide that $60,000 will cover my yearly retirement expenses, and if my retirement portfolio of 1.5 million is entirely registered could I decide I have enough and go sailing for 30 years? Assuming a 25% tax on the RRSP (I don't know what the tax at this income is, but I'm just putting up numbers for example), this would require a withdrawal of $80,000. Or with my registered portfolio of $1.5 million would I have to settle for a $45,000 lifestyle ($60,000 less $15,000 tax) assuming I couldn't work another day for some reason?

Obviously real life is even more complicated than this with multiple types of taxes, fixed income/dividend distributions, and things like OAS clawback coming into play.

[Shakespeare]

Before all taxes and expenses.

(But I think it completely misrepresents spending patterns; most people should withdraw at a higher rate while they are still able to travel.)

[kcowan]

...
Obviously real life is even more complicated than this with multiple types of taxes, fixed income/dividend distributions, and things like OAS clawback coming into play.

And the 4% is based on many years of history. Many people believe that the current environment is historically different and that aiming for a lower SWR like 3% is more prudent.

Also historically, people retired at 65. So the 4% rule might be good for 15 years and not apply to someone taking early retirement and requiring an income stream for 30 years, for example.

[adrian2]

Also historically, people retired at 65. So the 4% rule might be good for 15 years and not apply to someone taking early retirement and requiring an income stream for 30 years, for example.

IIRC,using historical US-based data, 4% SWR would last "forever". For example, having a portfolio 100% in RRB's @ 1.5% yield and consuming the capital, the money should last approximately 40 years. That's good enough even if retiring at 50, taking the chance that you'll be on social assistance if you live past 90.

[kcowan]

...For example, having a portfolio 100% in RRB's @ 1.5% yield and consuming the capital, the money should last approximately 40 years...

I am pretty sure that RRBs were not a part of the original study.

I take your point about the original study. The common belief among the FIRECalc users is that an SWR of significantly less than 4% would be recommended for early retirees. BWTFDIK?

[Shakespeare]

The common belief among the FIRECalc users is that an SWR of significantly less than 4% would be recommended for early retirees.

Early retirees have to be cautious before other sources of income (CPP, OAS) kick in because a major bear market could do irreparable harm to their portfolio. Once they hit 60, however, they can probably tolerate higher withdrawal rates.

[adrian2]

...For example, having a portfolio 100% in RRB's @ 1.5% yield and consuming the capital, the money should last approximately 40 years...

I am pretty sure that RRBs were not a part of the original study.

That's correct, I've just pointed out that exhausting the capital in combination to RRB's is a pretty much guaranteed way to surpass the 30 years portfolio survivability you've mentioned for an early retiree. With 4+% RRB yields, which were available a decade ago, a 4% SWR would have lasted "forever" (ignoring, for simplicity, that after 30 years you'd have to reinvest at the new RRB yields). Even then, using a hypothetical 0% RRB yield in 2030, an early retiree in year 2000, with 100% in RRB's at 4%, withdrawing 4%, would have the portfolio last 55 years (30 years with no depletion, 25 years drawing it down to 0).

[couponstrip]

Before all taxes and expenses.

(But I think it completely misrepresents spending patterns; most people should withdraw at a higher rate while they are still able to travel.)

Thanks.

Also historically, people retired at 65. So the 4% rule might be good for 15 years and not apply to someone taking early retirement and requiring an income stream for 30 years, for example.

IIRC,using historical US-based data, 4% SWR would last "forever". For example, having a portfolio 100% in RRB's @ 1.5% yield and consuming the capital, the money should last approximately 40 years. That's good enough even if retiring at 50, taking the chance that you'll be on social assistance if you live past 90.

This brings me to my next question. Does the math for the 4% SWR (with all assumptions and caveats noted) include distributions as part of what is safe or is it intended to suggest you can withdraw 4% of total capital, and the distributions (fixed income, stock dividends etc) are gravy?

For example, if somebody has a completely non-registered portfolio of $1 million and they have $500,000 of bonds that yield 3%, is the $15,000 pre-tax income from the bonds in addition to the 4% that can be withdrawn from the portfolio's capital ($15,000 pre-tax from bonds, $40,000 from sale of portfolio capital), or is it meant to be part of the 4% that is used from the portfolio for that year (ie only 2.5% of capital=$25,000 can be removed from the portfolio after accounting for the $15,000 bond income)?

[Flights of Fancy]

On the topic of spending more earlier in retirement: I know this has been posted before, but Milevsky and Huang released a paper in March which addresses and critiques the SWR. From the precis:

Our main practical conclusion is that counseling retirees to set initial spending at constant 4% of their nest egg is consistent with lifecycle theory only under a limited set of...parameters. We show exactly how the optimal behavior in the face of personal longevity risk is a plan that adjusts consumption downward in proportion to survival probabilities – adjusted for pension income and risk aversion -- as opposed to blindly withdrawing the same inflation-adjusted income for life.

(emphasis added)

[adrian2]

Does the math for the 4% SWR (with all assumptions and caveats noted) include distributions as part of what is safe or is it intended to suggest you can withdraw 4% of total capital, and the distributions (fixed income, stock dividends etc) are gravy?

The original studies included the distributions as part as what is safe.

[MaxFax]

All the discussions (I have ever seen) about safe withdrawal rate already have the income being generated factored into the calculation.

Regarding the ability of a 1.5% RRbond to sustain a 4% withdrawal (increasing with inflation - which is another assumption most all discussions include) --- really it will only sustain a 3.4% withdrawal. Use a spreadsheet with assumptions:
3.5% return
2.0% inflation
0% tax
$1,000,000 principal
40 yrs to death

[BRIAN5000]

withdraw rates, oldy but may be of interest

The Dynamic Implications of Sequence Risk on a Distribution Portfolio

•While a distribution portfolio's exposure to sequence risk changes over time, sequence risk never really goes away unless the withdrawal rate is constrained considerably

http://www.fpajournal.org/CurrentIssue/ ... Portfolio/


LOOKS like I can take 3% with a 35/65 FI/EQ and not have to worry.

[bones1]

LOOKS like I can take 3% with a 35/65 FI/EQ and not have to worry.

Why so low? You can get a higher withdrawal rate with a 100% FI portfolio of RRBs, and it's completely safe (assuming the federal government doesn't default).

If all you're aiming for is a 3% withdrawal rate, you shouldn't bother with equities at all. Why take equity risk if you don't need it?

[BRIAN5000]

I just haven't figured out what I'm going to do yet.

[Mouly]

If all you're aiming for is a 3% withdrawal rate, you shouldn't bother with equities at all. Why take equity risk if you don't need it?

Well there are tax considerations, unless the entire retirement portfolio is in an RRSP or TFSA then the 3% interest will be taxed the hardest compared to capital gains and dividends. You'd have to withdraw more than 3% of RRBs to have the same after tax amount of other investment types. So you'd need a bigger starting nest egg.

And bond yields can fall below 3%.