Thursday, January 10, 2019

The Ages of the Investor

Different stages of life call for different investing approaches and the need for transitioning from one stage to the next. William J. Bernstein addresses this challenge in his short book The Ages of the Investor: A Critical Look at Life-cycle Investing, the first book in his four part Investing for Adults series. His suggested method of transitioning toward retirement is very sudden.

Bernstein splits life-cycle investing into opening moves, middle game, and endgame. When you’re young, he suggests taking on as much risk as you can handle. Because this book is aimed at “investing adults,” he presumes the reader knows that stock picking and market timing are losing strategies. So, in this context, “risk” means compensated risk that comes with higher expected returns and comes mainly from stocks. Young investors are best off taking as much risk as they can handle without losing their nerve and selling out when stocks hit a difficult patch. “The young investor’s first encounter with a significant market decline serves mainly to ascertain her true risk tolerance.”

The book covers the life-cycle investing approach promoted by Ayres and Nalebuff that has young investors taking on 2:1 leverage using margin investing or call options on stock indexes. Bernstein lists some problems with this approach but leaves out the most serious problem: your ability to save in the future (and even to avoid withdrawing from savings) is highly risky. Losing your job during a stock market crash could completely wipe out leveraged savings.

An alternative to leverage is stock market investing using factor tilts. I tend to be skeptical that factor tilts will give future results that match past apparent success. Perhaps a slight tilt to small value stocks is sensible, but there is nothing wrong with an unleveraged position in low-cost stock indexes.

Because a mortgage is like a “negative bond,” it’s best to “Pay the damned thing off” rather than own bonds at the same time as holding a mortgage.

Bernstein portrays value averaging (VA) as an alternative to dollar-cost averaging (DCA) and describes VA as “a clever technique pioneered by Michael Edelson.” Value averaging doesn’t work. The main problems are 1) the return calculations supporting VA ignore opportunity costs of cash on the sidelines and interest on borrowed money, and 2) VA can lead to unsafe levels of leverage. I did some experiments to examine actual VA returns if we modify the strategy to eliminate these problems.

As you near retirement (the “endgame”), Bernstein suggests splitting your portfolio into two parts, a “liability matching portfolio” (LMP), and a “risk portfolio.” The idea is that the LMP contains completely safe assets that cover your basic needs for the rest of your life. The challenge is to create a LMP free from inflation risk, longevity risk, and counterparty risk (insurance company bankruptcy).

After examining several LMP approaches that all have risks, Bernstein chooses a ladder of inflation-protected bonds as the best option. This means Real-Return Bonds (RRBs) in Canada and Treasury Inflation-Protected Securities (TIPS) in the U.S. The challenge I see here is that we still have longevity risk. How many years of RRBs are safe?

During the middle game of your investing life “If, at any point, a bull market pushes your portfolio over the LMP “magic number” of 20 to 25 times your annual cash-flow needs beyond Social Security and pensions, you’ve won the investing game. Why keep playing? Start bailing.” The author calls for a sudden sell off of risky assets to buy the LMP. Then any new savings can be invested as riskily as you like.

This LMP idea would be more appealing if it handled longevity risk. How do I know that 20 to 25 times my annual spending needs is enough? If an investor bails out into a LMP in her 50s anticipating working to 65, but loses her job, the LMP is suddenly not enough for a longer retirement.

One side note most Canadians would agree with: the U.S. has “a dysfunctional health-care system and an inadequate safety net.”

On the subject of safe spending from a portfolio: “Below the age of 65, a 2% spending rate is bulletproof, 3% is probably safe, and 4% is taking chances. Above 5%, you’re taking an increasingly serious risk of dying poor. (For each five years above 65, add perhaps half a percentage point to those numbers.)”

It has become popular to plan for reduced spending throughout retirement. I’ve written many times about flaws in this idea. Bernstein takes this further by explaining why we might want to spend more in the future, even in inflation-adjusted terms. “Worker productivity, wages, and per capita GDP all grow at a real rate of about 2% per year. So will your expectations. Would you be happy with a 1960s standard of living? When everyone else has or will soon have an iPad, could you stand to live without one?”

Overall, I recommend this book for its critical thinking about life-cycle investing. However, I take issue with some of the specific recommendations.


  1. Apart from the longevity risk issue with a ladder of inflation protected bonds, it is even possible in Canada to build such a ladder? I have read that even in the US, some years TIPS were not issued. I think I prefer the traditional 3-4% portfolio withdrawals.

    1. @Grant: It wouldn't look like a traditional bond ladder, so maybe that's not the best term to describe it. You could split your money among RRBs maturing every 5 years far out into the future. The idea is that each investment provides spending money for 5 years. So, when RRBs mature, instead of reinvesting, you'd just spend the money for 5 years. If RRBs don't have long enough durations for your full retirement, you might have to reinvest part of the longest-duration RRB when it matures. I'm not advocating this approach, but it's the best I can think of to follow Bernstein's advice.

  2. To provide a real life example of this challenges of planning for old age. My parents saved $1.5 million for retirement (very small work pension) and own their own condo in Markham. They were quite comfortable until health issued began in their mid 70s. In order to avoid moving to a government nursing home (forget about private nursing care - $120,000-$150,000/ year depending of level of care needed), they are spending $85000/year of home care. Add to this the cost of food, condo fees, etc, and their yearly expenditures are close to $110,000/year or 7.3% of their savings. Thank goodness for CPP and OAS which lowers their spending rate to 5.7% which is still above your recommendation. But what's the alternative? Government nursing home?

    1. @Larry C.: I recently went through this helping my aunt. At first she needed only a couple of hours of help each day, which was covered by a provincial program in Ontario (LHIN). As her needs escalated, LHIN limits forced me to find a for-pay service that my aunt couldn't afford (she had much less money than your parents). Eventually, she needed the for-pay service 12 hours a day ($160k/year). As I was considering hiring someone live-in for less money and setting up a reverse mortgage on her house, she ended up in hospital and never left. The whole experience was very difficult, despite the best efforts of the LHIN people. My aunt's lack of assets complicated matters greatly and I ended up spending 5 figures of my own money.

      A generation earlier, elderly members of my family ended up living with their adult children. This has some financial advantages, but it can be a crushing burden to deal with an old person with dementia in diapers.

      I don't have any good answers for you, but money definitely helps, as long as an honest family member is watching the money to make sure the elderly people aren't getting fleeced somehow.

  3. I've mentioned to several people how easy it would be for me to make off with large sums of money undetected. That leads me to believe it's probably a common occurrence. The other dilemma is the asset mix - 90% common stock concentrated in 6 names. The dividends are great but they would have to pay huge capital gains taxes if they were sold off and replaced with stock and bond index funds. I sold off their TFSA stocks and moved that money to GICs but I'm hesitant to lose 10% of the portfolio to taxes.

    I also have nothing but good things to say about our local LHIN. They have provided much support to my parents, but there are limits to what can be provided.

    It's a challenge but what in life isn't? Thank you for your writing.

    1. @Larry C.: It seems unavoidable that we have to trust someone else when we're no longer able to protect our assets ourselves.

      Another of my family members has a concentrated portfolio. I've been selling off a little each year to contribute to a TFSA. It helps that income is low enough that these sales don't trigger any income taxes.

      Glad you enjoy my blog.

  4. Michael,

    Do you have an opinion on the results generated from this online calculator provided by morneau shepell re: safe spending rates in retirement. When I run the numbers for both my family and my parents, they show significantly higher spending available vs. the 2% safe, 3% probably safe, 4% taking chances. If the math makes senses (as best as can be determined with actually seeing their algorithms), could it be a US vs. Canada thing? Thanks.

    1. @Larry C.: I suspect the difference comes from underlying assumptions. Frederick Vettese has written multiple times that he believes we will naturally want to spend less as we age. I don't believe this. It's true that many studies show people do spend less as they age, but this comes from averaging in data from people who overspend early and are forced to spend less later. The Morneau Shepell calculator is likely based on Vettese's assumptions.

      Bernstein observes that incomes actually grow slightly faster than inflation (because society gets wealthier over time), and that our spending desires will actually grow with time. So, there is about a 2-3% per year gap between the pace of increased spending assumed by Vettese vs. Bernstein. Over decades, this gap makes a huge difference.

      My own assumptions are between the two, but closer to Bernstein than Vettese.

  5. No free lunch. The results only go out to age 80 so it's impossible to know how they expect spending to change after that. In my parents' case, spending went way down at first when minor health issues started to occur but then jumped when they needed help.

    The more research I do on government nursing homes in Ontario, the more confident I am that they will be able to find a quality affordable option after the 3-4 years waiting time that exists for the homes with the best reputation. The key is to have financial resources to manage this wait well.

    And when they are finally settled, extra monies can be directed to providing additional support in the home that makes their situation more comfortable.

    Thanks again.