Friday, February 19, 2021

Retirement Income for Life (Second Edition)

Those of us not lucky enough to have employer defined-benefit pensions have to save up for a decent retirement.  You’d think the challenge would be over when you’re done working and saving, but deciding how to manage your investments and how much you can spend is a new challenge.  Fortunately, former chief actuary at Morneau Shepell, Frederick Vettese’s book Retirement Income for Life: Getting More Without Saving More (second edition) shows us how to proceed with “decumulation.”  He even provides free online tools you can use anonymously.  I reviewed the first edition, and now Vettese has added new material and made significant updates to this excellent book.

Typical decumulation advice based on the 4% rule can fail.  Vettese goes through five enhancements to this typical advice to greatly improve the odds of having your money last your lifetime, without the need for any more savings.

“Many industry experts and professional associations already endorse the enhancements that are presented in this book.”  So, why is this book needed?  Because “most financial advisors are a conspicuous exception” to this endorsement because the enhancements reduce financial advisors’ compensation.  Even if these enhancements are clearly good for retirees, they cut into financial advisors’ livelihoods.

The Thompsons

Throughout the book we follow a hypothetical couple, the Thompsons, who have $600,000 in RRIFs, retire at ages 65 and 62, and are “the type of people that most readily come to mind when you think of retired people.”  “As Anthony Quinn’s titular character declared in Zorba the Greek, ‘Wife, children, house, everything.  The full catastrophe.’”

Vettese shows that if the Thompsons follow the 4% rule, they could run out of savings just 16 years into retirement if their portfolio earns poor returns.  He bases this on 1.8% MERs and 5th-percentile market returns (for a mix of stocks and bonds) with an unlucky sequence of returns where the market drops early in the Thompsons’ retirement and recovers little.  More technically, the first year return is a 5th-percentile one-year return.  Then the second year return is such that the average return of the first two years is a 5th-percentile two-year return, and so on.

Another assumption included in Vettese’s simulations is that the Thompsons’ inflation-adjusted spending will decline 1% per year in their 70s and 2% per year in their 80s.  I’ve explained before why I think this overstates how much they’ll likely want to spend less as they age.

One of the reasons future returns could be poor “is that the bond portion of portfolios will not do as well in the future because interest rates are so low.”  “It is practically impossible to obtain high returns on bonds when the starting point is low interest rates.”  “Traditional long-term government bonds will not be a great investment.”  I agree.

Throughout much of the book, we follow the Thompsons while they try to protect themselves against poor market returns.  They want to live the good life if markets perform well, but they’re most concerned with not running out of money if markets perform poorly.  The book goes through a few possible solutions before settling on Vettese’s 5 enhancements.

Alternate Ways to Try to Avoid Running Out of Money

The Thompsons could put all their money into a high-interest saving account.  If they are able to almost keep up with inflation, a simulation shows that their money could last 24 years using the 4% rule.  This is an improvement over getting 5th-percentile market returns, but it still leaves one or both of them running out of money, and it eliminates any possibility of higher returns if markets perform well.

Another possibility is to invest in real estate, but “real estate is not for amateurs.  My advice is to steer clear of this investment class (apart from your own home) and stick with stocks for capital gains.”  “The best hope for decent returns in the years to come … is to invest in stocks, risky as they are.”  Vettese settles on an asset mix of 60% stocks and 40% bonds.  In my case, I’ve chosen to put much of the “bond” part of my portfolio into high-interest savings accounts.

What about trying to get better returns by picking your own stocks?  “I used to research the market on my own and trade in individual stocks.  It was hubris to think I was smarter than the crowd.”  “I didn’t beat the market anywhere near often enough to call the experience a success.”  I’ve had a very similar experience.

Enhancement 1: Reducing Fees

The first enhancement is to reduce investment fees.  The Thompsons go from 1.8% MERs to 0.6% MERs using a robo-advisor.  Over 25 years, this reduces the fees bite from 36% to 14%.  “Simply reducing fees adds nearly three more years of RRIF income” for the Thompsons.  It’s possible to reduce fees even further if you can manage your own index ETF investments.  My own portfolio cost including MERs, commissions, spreads, trading fees within the ETFs, and foreign withholding taxes on dividends is less than 0.2% per year.

Enhancement 2: Transfer Risk to the Government by Delaying the Start of CPP Benefits

For some reason, people are jealous of the pensions government workers get, but they’re unwilling to grow their own inflation-indexed CPP payments.  “About the only good reason not to defer CPP to 70 is having insufficient assets to tide you over until CPP starts.”  Vettese goes through many reasons why people want to take CPP early and explains why they don’t make sense.  I’ve done this myself.

Enhancement 3: Transfer Even More Risk with an Annuity

Even though annuities’ built-in returns come from today’s very low interest rates, the ability to transfer longevity risk to an insurance company helps guarantee adequate retirement spending.  Vettese finds that using 20% of savings for an annuity is about right.  This goes up to 30% for those whose savings exceed a million dollars.

In my own simulations, I find that delaying the start of OAS payments to age 70 for a 36% increase in payments works better than buying an annuity.  However, this only works if you have enough assets to to tide you over until both OAS and CPP benefits start.  “High-net-worth couples should also defer OAS to 70.”

As for the type of annuity to buy, “Even if an indexed annuity made some sense, … indexed annuities tend to be overpriced, assuming you could find someone to sell it to you.”  Presumably, this means that even though indexed annuities are more valuable than fixed annuities, their prices are even higher than they should be.  Vettese advocates just buying a fixed annuity that has no annual increases in payments and using other income sources to make up for the fact that the payments will be worth less and less as the years go by.  I’d like to know if annuities with 2% annual payment increases are also overpriced or if it’s just inflation-indexed annuities.  The problem I have is that the annuity market is opaque, meaning it’s hard to get prices.

Although the main purpose of delaying CPP and buying annuities is to protect against poor market returns, Vettese finds that the Thompsons benefit even if they get median market returns.

A concern about the annuity calculations is that “The underlying interest rate for annuity purchases is 2.5 percent.  Historically this has been very conservative, but that is not the case at the present time.”  I try to be open to the possibility that I should own an annuity, but I’d want a simulation that uses actual annuity prices.

Another concern about annuities is inflation.  Vettese’s simulations assume inflation of 2.2% per year.  What if inflation is higher, even by just a little?  This would cause annuity payments to lose their purchasing power faster than the simulations expect.  People have opinions on whether higher inflation is likely, but we can’t know for certain what inflation will be over the next 3 decades any more than we can’t be certain of stock market returns.

Enhancement 4: Knowing How Much Income to Draw

Even if you’re prepared to follow Enhancements 1-3, it’s far from obvious how to decide how much you can safely spend each year.  For this Vettese provides a free Personal Enhanced Retirement Calculator (PERC) that you can use anonymously.

PERC provides three spending plans.  The first is based on 5th-percentile market returns assuming you don’t follow Enhancements 1-3.  The second is based on 5th-percentile market returns assuming you do follow Enhancements 1-3, and the third is based on median market returns and following Enhancements 1-3.  Because markets never perform as expected, you can run PERC every year or so to adjust spending and stay on course.

Vettese recommends choosing a spending level somewhere between the second and third plan.  He thinks it’s fine to base your spending on the third plan hoping for the best.  But you need to be prepared to reduce spending in future years if running PERC again reveals that your portfolio returns have been disappointing.

Enhancement 5: Have a Backstop such as a Reverse Mortgage

“Fully implementing enhancements 1 to 4 make anyone’s decumulation strategy practically bulletproof.”  But if something extraordinary happens and you do run out of money, a possible backstop if you have a home is a reverse mortgage.

With a reverse mortgage “You cannot be forced to move out,” but “You do have to maintain the home.”  This requirement to maintain your home worries me.  The elderly family members I’ve helped got to the point where they couldn’t maintain their homes, but still didn’t want to move.  Reverse mortgage lenders are just growing their business today and wouldn’t want the bad press of forcing old people out of their homes.  But let’s fast-forward a decade or two to the time when lenders have many loans on their books where the debts are approaching or exceeding the value of the home.  It might be profitable at this point to send out inspectors to see if any old people can be forced out because of poor maintenance.

Other Situations

To this point, the book focused on a typical retired couple who are long-lived.  Vettese goes on to examine other situations, such as one spouse dying young.  Even in this case, he finds that delaying CPP and buying an annuity still help.

Another different situation is early retirement.  Vettese finds that if the Thompsons wanted to retire 5 years earlier with the same income, they would need $1.3 million in RRIF assets instead of $600,000.  Part of this increase comes from needing to cover 5 extra years of retirement, and part of it comes from reduced CPP benefits.  When I tried to confirm this with PERC, I found that the Thompsons only needed about a million dollars for the earlier retirement.  I’m not sure why this doesn’t agree with Vettese’s figure.

One may wonder whether it makes sense for younger retirees to delay CPP benefits past age 60 because the lack of CPP contributions from 60 to 65 will reduce benefits.  The answer is that this dilution “is more than offset by the early retirement reduction factor.”  (Under CPP rules, there is no additional dilution from making no CPP contributions from 65 to 70.)

It can be challenging to decide whether to spend from RRSPs/RRIFs, TFSAs, or taxable (non-tax sheltered) accounts first.  In PERC, “I have assumed that income each year in retirement would come from a blend of all the asset classes.”  However, Vettese explains how to spend from some accounts first “to maximize after-tax income.”  I’ve devised a plan for myself to draw from my accounts to increase the amount of after-tax spending I have available.

Other different situations covered include single people and bequests and inheritances.

Other Interesting Comments

On mutual fund Deferred Sales Charges (DSCs), “To me, [charging DSCs] is one of the most odious practices a fund salesperson can perpetrate on an innocent investor.”  “Do not trust anyone who has recommended investment funds with a DSC.”

“Employers have an excellent business case for maintaining a [defined contribution] pension plan for their employees.”  “The same business case strongly suggests that employer support should continue beyond retirement” to help “during the decumulation phase.”

“It is a fair guess that fewer than a dozen people in Canada can explain how the survivor pensions under the CPP are calculated.”  Vettese goes on to explain how to calculate a CPP survivor’s pension.  However, I found that his description is different in one respect from other descriptions I’ve read.


The intended audience for this excellent book is “people who are close to retirement, or who are already retired, and who are going to rely heavily on their own savings to meet their retirement income needs.”  This audience will get solid advice on the best way to handle decumulation of their savings through retirement.  One reservation I have about the five proposed enhancements is buying a fixed annuity.  It may be that buying such an annuity is best, but to confirm this, the simulations need to use current annuity prices and consider a reasonable range of inflation outcomes.  That said, retirees who simply follow all Vettese’s advice and use his free Personal Enhanced Retirement Calculator (PERC) will very likely have better outcomes than if they follow typical retirement spending advice.


  1. Great review and great book by Frederick Vettese. I have both editions. I have used his free calculator and I'm trying to follow his recommendations. As you mentioned, the only recommendation I'm not sure about is the annuity but I have time to decide on this enhancement.

    What are your thoughts about going with a 70/30 split for stocks and bonds assuming spending can be ramped down in down market years? Also do you consider your 40% in a high interest savings account as an emergency fund/cash wedge or do you have additional funds put aside for that purpose?

    We have an existing HELOC which we would use for an emergency fund so I don't expect will have any more cash on hand than the 40% or 30% in bonds, except for a rolling 6 months of cash for current expenses.

    1. Hi Joel,

      Thanks. You'll likely do well following the book and calculator.

      I maintain 5 years or my safe spending level in a combination of HISAs and short-term government bonds. Currently, this amounts to about 20% of my portfolio (which I calculate to include the present value of future CPP and OAS payments). So, I already "vote with my feet" on holding more stocks than Vettese assumes. I do have the ability to reduce spending as necessary because so much of my spending is discretionary. The large amount I have in HISAs makes a separate emergency fund unnecessary. I don't treat any part of my holdings as a cash wedge that is separate from my portfolio. I don't do "buckets" in the way most people mean the word.

  2. Great review of Vettese's book. I also recently read this book and found it somewhat comforting that it really isn't that complicated to plan for a financially successful retirement. The plan can be simple, but of course, isn't necessarily easy to execute.

    I'm all in for reducing MERs (Enh #1 already using all-in-one ETF XGRO), and delaying CPP and OAS until 70 (Enh. #2). However, I still am not sold on buying an annuity, and neither is Vettese as of the writing of this edition, which is quite a change from the first edition. I will continue to watch the annuity market over the next 20 years to see if they make better sense at another point in my life.

    Using the PERC calculator seems like a great idea to validate whether you are still on track or should perhaps make some adjustments to your plan. Will certainly follow this enhancement recommendation.

    I pray I will never to use enhancement #5 - a reverse mortgage. This makes me cringe at the thought, but understand that some may need this option to stay in their home.

    1. Unknown,

      Thanks. Vettese has definitely made it easy to follow his recommendations by using his calculator. I'm not sold on an anuuity because of possible inflation, and because prices are higher than Vettese assumed in the book.

      I've watched too many people cling to staying in their long-time family homes. I hope I'll have the good sense to move into something more suitable for an old person (even if I don't need to raise any money). Even just moving to a smaller house or a condo is better than a big sprawling house.

    2. I agree with you. I'm envisioning selling my house and renting a place in a small community, thus liquifying all of my assets in which to live on for many years.

    3. Hi David,

      If this is your plan, I consider not waiting too long. Few people seem willing to make big changes like this as they get late in life.

  3. From the book "Lifecycle Investing" by Ayres and Nalebuff, "few insurance products operate with less than 30% overhead and profit margin." If you can self insure, then self insure. I'd only think about an annuity if I was getting quite advanced in years, and my financial status wasn't secure. Then the mortality credits might make it worthwhile.

    1. Anonymous,

      The challenge is to figure out when mortality credits exceed overhead, profit margin, and dismal long-term bond returns. Beyond this challenge, the wild card of uncertainty about future inflation keeps me from buying an annuity.

    2. The longer one defers buying an annuity, the more important mortality credits become and the less important inflation becomes.

    3. Anonymous,

      True. The challenge is to figure out when the crossover happens. Buying too early or buying too late both leave money on the table.

  4. I welcome my contribution to CPP, but I'm hesitant about buying an annuity. Why the difference, as essentially they're the same? They're both annuities, which are insurance products, where risk is managed by transferring it to a counterparty. The difference is in the return. IIRC, a Fraser Institute study estimated the average annual real return on CPP contributions of someone born around 1970 at 2.1%. I've not seen an estimate of the return on private annuities, buy my guess is that it's negative. CPP meets my definition of an investment, as well as a way to manage risk. A private annuity is only a way to manage risk. Why the difference? My guess is that CPP is lower cost and has more tax advantages. Also, when you buy an annuity, I believe that the money is invested in bonds. CPP contributions are invested in more diverse products, including those with higher expected return. In fixed income, it's hard to beat a 2.1% real return.

    1. Anonymous,

      The differences between annuities and CPP are even larger than you say. The underlying interest rate on today's annuities is definitely negative in real terms (they appear to be near zero in nominal terms, depending on which mortality tables you use). The Fraser Institute study was biased. The biggest problem was assuming a work history where dropout provisions are useless. Most people get a higher real return than 2.1% on the CPP contributions. Further, the IRR on the investment of delaying CPP to age 70 is a real return in the 4-7% range, depending on the terminal age you use.

      However, I'm sure Vettese understands all this. His focus on blunting the severity of bad retirement outcomes is correct for most people. I'd really like to see his analysis repeated using actual annuity prices and taking into account inflation uncertainty.

    2. Diversification, hedging and insurance are used to manage risk. Usually, diversification is the only risk management method that doesn't decrease return. Hedging, such as asset liability matching, does decrease return. Insurance, such as protective puts and life/disability insurance, is generally a money loser.

      With CPP, you transfer risk (longevity risk, market risk, inflation risk, dementia risk) to the government. And you also get a positive expected return on your contributions. In many ways, it's a unique financial product.

    3. Anonymous,

      I'd be pleased to be able to put my portfolio's bond allocation into CPP to get guaranteed inflation-indexed payments based on CPP's returns.

  5. Could you please explain this a bit more? I'm uncertain what it means, & is a decision we are presently pondering. Thank you.

    "One may wonder whether it makes sense for younger retirees to delay CPP benefits past age 60 because the lack of CPP contributions from 60 to 65 will reduce benefits. The answer is that this dilution “is more than offset by the early retirement reduction factor.”

    1. Anonymous,

      When calculating your CPP benefits, you get to drop out 17% of your working life. The idea is that you remove several of the years where you paid little into CPP, and you get to take the average of your remaining (better) years. If you take CPP at 60, you perform the calculation I described on the 42 years from 18 to 60. If you take CPP at 65, you do this calculation on the 47 years from 18 to 65. If you didn't pay anything into CPP from 60 to 65, then this period would use up 5 of your dropout years, and your ability to eliminate other low-contribution years is limited. This reduces the percentage of the maximum benefit that you'd receive. However, this CPP reduction due to taking CPP at 65 but not working from 60 to 65 is much smaller than the increase you get for taking CPP at 65 instead of 60. As for delaying the start of CPP from 65 to 70, there is a special dropout provision that allows you not drop out the years from 65 to 70 (unless you worked and would benefit from keeping them) so there is no penalty for not working from 65 to 70.

  6. We bought a joint annuity in 2018 at the age of 74, in spite of the warnings: don't give your money to an insurance company; why not just put the money in the bank and draw out so much a year; you lose control of your money; interest rates are too low right now; what if you die next year; you can make 12% in the stock market; Garth Turner thinks it's a terrible idea. These arguments are convincing when you are younger, but when older you are now considering physical and mental issues and investment worries.

    The annuity has given us great peace of mind and we spend the payments freely as there is no fear of running out of money. Our phone call last March with our financial advisor was relaxed as the drop in the value of the investment portfolio did not affect our budget.

    The payment we receive works out to 6% annually on the amount paid in. Since the payout is mostly return of some of our capital, we only pay income tax on the interest portion of the payout. This puts the theoretical return over 6%, compared to all interest being taxable if a similar capital amount were in investments like a GIC.

    We have a 10yr guarantee which will continue to pay our beneficiaries to the 10th year if both of us die before that.

    The fly in the ointment? Once you buy the annuity, the capital is tied up permanently, so you have to consider what proportion of your assets should be invested this way. I have seen the figure of 20%. If I was single with no dependants or family, didn't have a company pension, and didn't care about leaving an estate, then I would maximize the annuity. After all, this is the same as other pensions that stop when you die.

    I understand the inflation argument, but it's outweighed by peace of mind. And as Youngstown Ohio lawyer Benjamin Roth pointed out in his Great Depression Diary, insurance companies paid out through the depression.

    1. Hi Donald,

      I'm glad your annuity has worked out for you. IIRC, annuity prices were lower in 2018 than they are today, which helped you. Investing 20-30% of one's portfolio in an annuity (as Vettese recommends) may not be optimal (depending on annuity prices), but it does buy some peace of mind.

      If by "maximize the annuity," you mean investing almost all savings in fixed annuities, I don't think this is wise, particularly for young retirees. Inflation concerns aren't so bad if they affect 20-30% of your portfolio, but watching almost your entire income erode would undermine that peace of mind. I've seen this happen to elderly members of my extended family who had annuities through the 1970s and 1980s.

  7. I've said this before, and I say it again: Wouldn't it be nice if we could just purchase some additional CPP?