Safety vs. Risk in Retirement Post 4 - The Wade Pfau Book
Reviewing the advantages and disadvantages of "Redefining Retirement" by Wade Pfau
This post #4 continues the series already begun of reviewing retirement income books
Post #1 - Introducing the Safety v. Risk in Retirement Series
Post #2 - Pensionize Your Nest Egg by Moshe Milevsky
Post #3 - Retirement Portfolios by Michael Zwecher
Of the three authors reviewed in this Safety v. Risk in Retirement series, Wade Pfau is the most academic as well as the best known. His book Redefining Retirement: A Safe and Secure Way Down The Mountain seems the most targeted to the financial planner and practitioner, rather than the retail prospective retiree.
This is the strongest of the three books, but a large margin. It is also the least “accessible” to the average retail investor. It is meant as a resource for professional practitioners.
Pfau clearly sets up the binary distinction between what he calls the “probability-based” approach to retirement and the “safety-first” approach. Those two approaches may also be categorized as the “investment brokerage” approach and the “insurance company” approach. This binary is quite useful. It properly frames my personal approach (strongly probability-based!) as well as the reason I’m reading these books (to learn )more about the "safety-first” approach. I have approached all three of these books with my strong priors but also a willingness to be persuaded by the elements of a “safety-first” approach.
Depending on which type of professional you put at your disposal, the advice you get may favor one versus the other.
Pfau lays out the case in Chapter 1 against the “probability-based” approach, specifically naming Longevity Risk, Sequence-of-returns Risk, and Competence risk.
Longevity Risk - Because we can’t know the date of our death, perfect longevity planning can prove impossible. We don’t want to outlive our money, which is one side of managing the uncertainty of our life’s longevity. The worst part of longevity risk is the fear of outliving money. But the opposite can and will happen for many: as a result of longevity risk fear, prudence leads us to underspend to guard against an uncertain future. The underspending is then sub-optimal for living our best life in retirement, and we may accidentally leave too much money for after our death.
Sequence-of-Returns Risk - This is a description of the problem that a prolonged (multi-year, severe) market downturn can lead us to either forcibly curtail our expenditures beyond what is comfortable in retirement, or we maintain an overly aggressive withdrawal schedule that permanently impairs our nest egg. Selling stocks when they are down to maintain our lifestyle could put a risky portfolio on an unsustainable path. Then we may outlive our money because we misjudged how much the market could drop and for how long.
Competence Risk - This is the idea that managing our own investments through a probability-based approach depends upon a retiree, or the retiree’s spouse, or the retiree’s caregiver, or the retiree’s trustee, or the retiree’s trusted financial advisor having the wherewithal to make complex financial decisions late into life. A grim shorthand here could be “dementia risk.” It may just be too much to ask a person to track and maintain control over their own self-managed situation.
These are all fine and accurate critiques!
Pfau argues in his book for a “safety-first” approach, which will depend on risk-reducing or risk-eliminating retirement products like annuities, combined with other guaranteed programs like Social Security and a pension. Since these generally are guaranteed for life, the prudent planner simply needs to ensure that known income for life can match known lifestyle costs.
One of the characteristics of a safety-first approach, says Pfau, is that 90% certainty of success is not enough. The goal is 100% certainly that there will be enough money in retirement.
This is an interesting (and maybe key?) point. Because in my own planning and scenario analysis, I have often been left with some uncertainty. Like, I’m optimistic it’s all going to work out. But also, it’s only with about 95% certainly. And, while I think my household funds will last until I’m 95-ish, there’s still the possibility that - in an extreme sequence-of-returns catastrophic downturn - the money runs out on us in our 80s. I have always told myself that that’s “fine” and we’ll deal with it if that worst-case scenario happens. I actually still do think it is fine. But still, not everyone has my (relatively high) risk tolerance. Some people want a 100% guarantee. And those people will tend toward a safety-first approach, rather than the probability-based approach.
Chapter 2 - Fixed Income Assets
I am a bond guy by training, so I know the stuff he is teaching. It’s all accurate as far as it goes. On the other hand, it’s accurate as a theoretical discussion for a financial professional to understand some very basic information about how bonds work, purely in theory.
The practical use of the chapter for a non-financial professional is, in my opinion, zero.
I was having a discussion with a family member recently about a near-retiree of her acquaintance who sought expertise in building a “bond-ladder” for his use in retirement, which happens to be one of the techniques that Pfau discusses as an option for “safety-first” retirees. The bond-ladder, I recently told my family member, is madness. Nobody, and I literally mean nobody, should be building a “bond ladder” for their retirement in 2025. If you want the safety of bonds, you buy a bond mutual fund, and/or a TIPS mutual fund.
Retail investors messing with individual bonds is insane.1
And yet, this is book #3 of this series in which the bond ladder is discussed as the most logical way to create a safe retirement. I continue to strongly disagree.
Chapter 3 - Stocks and Diversified Investment Portfolios
Here we get a comparison of historical returns and volatility of different asset classes (large cap, small cap, long-term corporate bonds, long-term US bonds, intermediate bonds, short-term bonds). We get the 1950s Markowitz modern portfolio theory (MPT) that you can increase your combination of return and volatility by owning a diversified pool of assets (as long as they are not perfectly correlated, you get an improved risk/reward in combination.) We get the accurate critique that households are not the same as institutions, specifically because retired households have to worry about withdrawals and sustainable withdrawal rates in addition to the optimal risk/return, while institutions are more aiming for perpetuity plus have the ability to fundraise.) For that reason alone, we should have a greater concern for safety for households, and of course I agree.
Pfau offers the key insight which I will highlight: Taking into account the worst case scenario for stock and bond returns, there is no historical period beyond 17 years in which equity returns (S&P500 index) didn’t trounce bond returns (intermediate bonds).2 If you’ve got a long time horizon, and you’re holding bonds, you are implicitly saying the most dangerous 5 words in investing: “It is different this time.” You are saying, the odds and history be damned, I want to make a very low probability bet. Pfau doesn’t use this language, that’s me saying it. He comes to a different conclusion, although his data backs up what I say.
Pfau also includes a nice review of the behavioral finance mistakes investors commonly make. If you don’t know what these are, it’s worth your time to look them up and see how many of these sins you should confess: Availability bias/recency effect, Loss Aversion, Overconfidence, Hindsight Bias, Survivorship Bias, Herd Mentality, Ambiguity Aversion, Framing, Home Bias, and the Behavioral Cycle of Investing.3
Pfau says you should always take into account taxes and fees when thinking about stocks and a diversified portfolio, although he does not highlight the absolute key to reducing taxes and fees with stocks and a diversified portfolio.4
Chapter 4 - Income Annuities (Risk Pooling)
Pfau explains the math behind “monetizing mortality,” which I explained in an earlier retirement income book review. Because insurance companies know (on average) when you will die, they will generously quote you an average amount of guaranteed income for your annuity. If you die early, they win. If you die late, they’re also fine because enough people in their annuity pool died early. And that’s a win for hedging your “longevity risk,” the risk that you might otherwise outlive your money.
If you understand discounting cashflows5 and you understand the concept of actuarial probabilities of death, Pfau’s explanations are pretty logical. The amount of money you get is based on your age, prevailing interest rates, and how much money the insurance company wants to make on the annuity.
There are many variations on income annuities - like 2-people covered (called joint-survivorship), or payouts increasing with inflation, or deferred income, or refunds if you die early, or guaranteed numbers of years of payment.
Chapter 5 - Variable Annuities
At the start of this chapter Pfau says the truest thing, in a breakout box called Variable Annuity Caveat. He writes “All variable annuities are not created equal, and many come with excessive fees.” I would shorten that caveat to: “All variable annuities are not created equal, and many come with excessive fees.” I’m glad he’s got a version of that statement up front.
Then he writes the next most true thing about variable annuities, in the first sentence of the chapter “Generally, the most efficient means for balancing protected income and investment upside is to combine life-only income annuities with aggressive stock portfolios.” That’s sort of where my analysis ends and so should yours.
But of course he has more to say. Pfau understands “guaranteed rollup rates”, “vesting into the benefit base”, “step-ups to the benefit base,” “guaranteed withdrawal rates,” “death benefits,” “constraints to the subaccounts,” fees in the form of “contingent deferred sales charges,” “12-b1 fees in the subaccounts,” and “mortality and expense charges” - and you could sort of understand them too if you are a masochist.
I have a hard time getting into the weeds with him because these are bad products and my fondest wish is for them to disappear.
Like, cancer is no doubt interesting to the lab bench scientist who could infinitely study the ways in which human cells growth mutates and goes haywire to the detriment of the whole person. But the human being (as opposed to the scientist) would be better off simply not having the cancer.
Chapter 6 - Fixed Index Annuities
This also starts with a break-out box Fixed Index Annuities Caveat with the same warning as Chapter 5 about the high fees, treading carefully, and not all fixed index annuities are created equal, but that Pfau will “assume these are priced competitively.” I don’t think that’s a great assumption. The whole point is to market these to people who do not know even what questions to ask. These are even more opaque than the Chapter 5 Variable Annuities, and I just can’t.
Chapter 7 - Life Insurance
I have not traditionally understood life insurance as an important retirement product, although of course it can be marketed that way. Pfau explains that the hybrid product of “whole life insurance” allows the investor to build “cash value” inside a structure that will pay out when the person dies.
Here is a neat insurance company trick embedded in this product. When you have (for example) an embedded $500,000 “payout upon death” policy, and some kind of accumulated “cash value” of (for example) $75,000, the insurance company only has to pay out $425,000 upon your death, and then it returns to you your own $75,000 in accumulated savings. Meaning the insurance company is off the hook for that amount. I had not previously realized that was the thing with “whole life insurance,” but it seems like a really neat trick insurance companies play on their customers.
If the purpose of life insurance is a legacy for one’s heirs, then life insurance makes sense. A big benefit is that the payout at death is not taxable, unlike a traditional retirement account, which is taxable.
To the extent a whole life insurance policy includes income-generating assets and the accumulation of a “cash value,” these will be invested in low-return fixed income products. The return on investment, you should expect, will be low.
A related product is variable and universal life insurance and these are just going to be worse still even than whole life insurance. What will actually happen (as a rule, trust me) is that the retiree’s insurance premiums will go up later in life, such that they will eventually stop paying on them. This will cause the life insurance to lapse after years or maybe decades of paying on them, which was the point all along. For the love of God don’t buy these products for your retirement planning. (My words, not Pfau’s. He is just patiently explaining their various features throughout this chapter.)
Chapter 8 - Fitting Income Annuities Into a Financial Plan
This is probably my favorite chapter in the book, in the sense that it can be summarized with a sensible lesson of use to everyone. And it encapsulates the best lessons of the other 2 books reviewed in this series.
Retirees should consider fixed annuities to cover some portion their known essential lifestyle expenses, to reduce longevity risk.
That’s fine, I agree.
Pfau would guide the retiree to calculate their essential lifestyle expenses, their optional lifestyle expenses, and then aspirational or emergency lifestyle expenses. He would further guide the retiree to strongly consider purchasing a guaranteed-lifetime annuity for the essential expenses so that those would never be at “market risk.” For the remainder, a mix of brokerage products could be deployed to generate optional and aspirational income to cover expenses.
Good, I like it.
A fine point about “liquidity” in retirement
Pfau makes a fine counterpoint to my preference for a brokerage-product based retirement, and my strong belief in the superiority this flexibility offers. His point is that a (let’s say) $1 million retirement portfolio in the stock market is not truly liquid and flexible, if it must generate, without fail, a $40,000 annual income. The retiree can’t actually spend any of the $1 million as she chooses since it must be fully invested in the market to produce income sufficient to maintain a sustainable withdrawal rate in retirement.
Something I kind of hate about annuities is that they are not liquid, but Pfau is right to point out that brokerage products may not be as liquid as they seem, if they are needed for covering essential lifestyle costs.
Chapter 9 - Product Alignment for Retirement
Pfau argues that there exists - analogous to the 1950s Markowitz Modern Portfolio Theory about diversifying portfolio assets - an ‘efficient frontier’ in retirement products. According to Pfau’s own analysis, that efficient frontier includes stocks in a brokerage account plus guaranteed income annuities. But, pointedly, not bonds.
If you are familiar with the Modern Portfolio Theory curve, the interpretation of this graph is that you can have an all-stock retirement, or an all-income annuity retirement, or some combination of the two, and in every case you will be better off in terms of both preservation of assets and meeting your spending goals than if you blended bonds into the mix. I don’t consider that Pfau proved that in this book, but I do think individual bonds are so inefficient for retail investors that I tend to agree. I don’t agree that income annuities are better in all cases as compared to a bond mutual fund, which seems to have a tremendous liquidity advantage over annuities. The main argument from Pfau is that by “monetizing mortality” you can achieve more lifetime spending for a given amount invested, which is a fine insight.
Tax treatment of income annuities
This is a key point of chapter 9!
Guaranteed income annuities are pretty tax efficient if purchased with taxable (after-tax) dollars.
If you buy an income annuity in a taxable account (like a regular brokerage, not an IRA or 401(k) type account) then you get a pretty good tax advantage, that works like this:
At retirement (or whenever) you turn over your $100K of premium in order to receive a lifetime series of payments for an annuity. The IRS tells you what your “exclusion ratio” is based on your expected remaining life. The exclusion ratio is how much of your annual annuity payment will be tax free each year.
The “exclusion ratio” calculation compares how much you’re supposed to receive on average compared to how much you paid for the annuity, to separate out (from a tax perspective) how much is income versus how much is repaying the investment.
To be more specific:
Let’s say your $100K annuity would pay your $6,500 per year for life.
If you’re a 65 year-old man with a 20 year life expectancy, the IRS would say you should on average get back $6,500 x 20 = $130 thousand. The exclusion ratio then is the $100K initial investment divided by the $130K expected return, or .7692.
Each year, when you receive $6,500, the first $5K (which = .7692 x $6,500) is tax free, because it represents a return of principal on your annuity investment. The remaining $1,500 is taxed as income.
That pattern continues until you’ve received your full $100,000 returned, which will take, in this example, exactly 20 years. After 20 years, your $6,500 annual annuity payments are fully taxed.
In this way, $6,500 of annuity income is likely more tax-efficient than $6,500 of traditional retirement account IRA income via RMDs.
Yes but remember don’t let the tax tail wag the investment dog
I still feel obligated to point out that the investment return on income annuities is going to be significantly lower than the historical average return on a diversified portfolio of stocks, so you shouldn’t choose an income annuity “for the taxes.” It’s just that the guaranteed aspect of an annuity, plus tax advantages within taxable accounts, make it slightly more attractive, if you insist on covering your lifestyle costs in retirement this way.
It’s difficult and inadvisable, though not impossible, to purchase a guaranteed income annuity inside a retirement account. Difficult, because lot’s of 401(k) plans probably won’t allow it, and it makes for complicated RMD calculations. Inadvisable, because you’ve disregarded the tax-advantage of annuity income.
Concluding thoughts
I liked Pfau’s presentation better than the previously reviewed retirement income books, because he tends to base his views on math rather than fear. He does not make wildly inaccurate commentary about historic stock market returns, which I also appreciate.
I think retirees - on average - are still far better off choosing the “probability-based” approach of building retirement based on brokerage assets rather than the “safety first” approach of insurance products. But that is my remaining bias somewhat based on personality and risk tolerance and life experience. I can understand the appeal of locking in lifestyle costs through “monetizing mortality,” and will keep that in mind more for myself and for folks I advise.
Fine, I’ll explain a little bit. Bonds, for people with less than $10 million to spend, are incredibly inefficient to buy and sell. You will lose tons of value in the transaction as a retail buyer as compared to a wholesale Wall Street buyer. A “Bond Ladder” is like the most inefficient way to invest and earn a return on capital in 2025. It would be like using the fire escape to climb to your penthouse apartment on the top of a 50 story building when there is the option of a nice elevator. Just buy the bond mutual fund for goodness’ sake.
Of course, if you have a short time horizon, it is quite common for stock returns to underperform bond returns. Especially for less than 5 years. Beyond 5 years, it become less common, even in a heavy down markets for stocks.
The most complete way to learn about these may be reading (Nobel Prize winner) Dan Kahneman’s Thinking Fast and Slow, which as a book is a bit of a slog, but worthwhile. For a peppier version of these big behavioral-finance ideas, you could read Michael Lewis’ The Undoing Project, Morgan Housel’s The Psychology of Money, or Belsky and Gilovich’s Why Smart People Make Big Money Mistakes and How To Correct Them, each of which I’ve reviewed on my old website www.Bankers-Anonymous.com.
Index funds are both the lowest fee way to invest in a diversified portfolio of stocks, while simultaneously reducing portfolio turnover, thus minimizing taxes within a taxable account.