I recently had the opportunity, and great pleasure to have lunch with Professor Yaari and his lovely wife Nurit, on a picturesque Friday afternoon at the Israel Museum in Jerusalem. Although the conversation started with pleasantries, politics, wine and the wonderful view, it eventually turned towards retirement economics. We discussed his varied career, illustrious students and current research interests. Why this great scholar should be the topic of this month’s Annuity Analytics, will soon become very clear.
Professor Yaari has collaborated with and supervised many well known Nobel Prize winning economists during the last 45 years of his career. He has written thought-provoking, fundamental research papers in the field of microeconomics and decision-making under uncertainty.
One of the hallmarks of his philosophy – which became evident from his very careful and deliberate replies to my questions – is that the economics profession should not rush to abandon the rationality of humans. According to him, many occurrences that might seem at odds with rational decision-making can be properly explained within a classical perspective. His spirited defense of the rational ‘economic man’ was quite refreshing, given the now daily bombardments of evidence professing to show how silly we all apparently behave with our money.
Needless to say, his writing has had a profound influence on my own thinking about economics, and more specifically about retirement income planning. Indeed, most practicing financial advisors and planners in North America might not have heard of Professor Yaari’s work, <but they should>.
Here is why.
More than 45 years ago, while still a doctoral candidate at Stanford University, he was the first economist to introduce annuities into the canonical life-cycle model. For those who want to look it up, his most famous research work was published under the title: “Uncertain Lifetime, Life Insurance and the Theory of the Consumer” and appeared in the <Review of Economic Studies>, in 1965. Fast-forward 45 years, and today every graduate student in economics and insurance is forced to read this paper, for very good reason. To put it simply, he introduced and then legitimized life annuities to all economists. He placed them in your portfolio.
You see, back in the 1960s academic economists hadn’t really given any thought to how lifetime uncertainty – the randomness of the length of retirement – affects financial planning, savings and investment behavior. I guess you can say that economists didn’t like to think about death, or its impact on economic activity. Sure, they had some vague notions that old age might make people cranky and impatient, but nothing concrete or formal.
Around the same time, modern portfolio theory – introduced by Professor Harry Markowitz – was just starting to catch on with academics (it would be decades before this reached Wall Street.) Yet, even Professor Markowitz and his noted contemporary Professor William Sharpe never addressed how the randomness of life might impact economic behavior and portfolio construction. The other giant names at the time, such as Milton Freidman, or Franco Modigliani – who first theorized that consumers like to smooth their standard of living over time, considering their lifetime resources when they do this – hadn’t said anything about mortality and longevity. In most of their models and papers, people died at a fixed and known time, denoted by the capital letter T. Now how unrealistic is that?
Enter a young Menahem Yaari writing his Ph.D. at Stanford University in the early 1960s. He started his famous paper with the following words:
<“…One need hardly be reminded that a consumer who makes plans for the future must, in one way or another, take account of the fact that he does not know how long he will live. Yet, few discussions of consumer allocation over time give this problem due consideration. Alfred Marshall and Irving Fisher were both aware of the uncertainty of survival, but for one reason or another they did not expound on how a consumer might be expected to react to this uncertainty if he is to behave rationally…”> (pg. 137)
Then, in a mathematical <tour de force>, he went on to describe how consumers would slowly spend down their wealth, in proportion to their survival probabilities and attitude to longevity risk, and gradually reduce their standard of living – rationally. But then, and here is where the light bulb went on, if you gave this same consumer the ability to purchase any type of annuities, they would not have to reduce their standard of living with age! They would, in fact, be able to hedge or insure against their longevity risk.
His 1965 paper then went one step further and derived the optimal “portfolio mix” between regular market-based instruments (e.g., mutual funds or ETFs) and their actuarial counterparts (life annuities), as a function of one’s preference for legacy vs. personal consumption. In modern terms, he introduced what I like to call “product allocation” just a few years after Professor Harry Markowitz introduced “asset allocation.”
As you can imagine, this paper has been cited thousands of times by economic scholars, in the 45 years since it was published. This, by the way, is the ultimate compliment in any scholarly field. Quite justifiably, some people refer to him as “the Markowitz” of the annuity world.
One of the most-quoted results attributed to him is that not only are life annuities an important component of a consumer’s portfolio, they should actually form the entirety of the portfolio in the absence of a bequest or legacy motive. He pointed out that the mortality credits are simply too valuable to ignore. And yet, as most readers of this column probably know already, very few people actively choose to annuitize any portion of their “nest egg.” I asked Professor Yaari why he thought this was the case, and I’ll get to his reply in a minute.
In a seminar that he gave at the IFID Centre at the Fields Institute to commemorate his work in the area, he mentioned that his 1965 work was originally intended to help resolve inconsistencies within neoclassical economics and the apparent low spend-down rate of assets around retirement. In that sense his paper was intended as “positive” (to explain observed behavior) as opposed to “normative” (to provide financial advice).
In other words, he never intended to write a manifesto on how people should behave in the face of lifetime uncertainty: namely, that they should hedge longevity risk by purchasing annuities. At the same time, he did acknowledge that this model can easily be inverted and used to offer guidance on how people should allocate their assets around retirement.
As with many great scholars, Professor Yaari laid his golden egg which established the value of annuitization for one’s nest egg, but never went back to elaborate or follow-up during the last 45 years of his career. Basically he introduced annuities to economists, and then moved on to other pursuits and problems. He endowed the field with what is now called “The Annuity Puzzle.” That is, why aren’t they valued and used more by consumers?
Professor Yaari stated to me that perhaps one of the reasons why many people don’t appreciate the value of life annuities is that personal tastes and preferences can change over time; and they know it. The current design of annuities might not allow retirees to adapt to changes in their own tastes. These changes might be in legacy preferences, or even for spending more now, versus later. There it was again, the defense of rationality. People know what they are doing. No need to nudge, or even shove them.
In contrast to Professor Yaari’s rationality school, the view offered by a growing number of economists in the behavioral school is quite different. Although both sides appreciate the value of annuities, the behaviorists argue that consumers’ aversion to annuities is an irrational bias that is due to psychological quirks outside the realm of classical models. In fact, many in this camp point to low voluntary annuitization as a failure of the rational model. This is precisely why some policymakers in the U.S. – citing the Yaari mantra – have gone so far as to suggest that people should be mandated or defaulted into annuitizing a portion of their wealth at retirement. In other words, paternalism defended by rational economic models.
The debate rages on. Are annuities shunned for good reason, or not? What can be done to make them more appealing? How do we reconcile evidence with Professor Yaari’s 1965 result?
Alas, for someone at the intellectual core of retirement planning, Professor Yaari is visibly humble and quite modest about the impact of his work on the field. In fact, when I mentioned to him that I had just returned from a research conference in which almost half the papers presented by graduate students were a derivative of his original annuity work, he was rather surprised and mildly amused. This was definitely one author who was not regularly checking the ranking of his book on Amazon.
As we finished desert, overlooking a magnificent Jerusalem city view that has waited centuries for peace, I asked him if he – after nearly 45 years — had any plans at all to revisit the topic and write another likely best-selling tome on annuities. “Might we ever see a sequel, after all these years”, I asked him. And, with a twinkle in his eyes, his reply was: “I have to think about it a little bit more…”
Moshe A. Milevsky, Ph.D. is an educator, author, consultant and entrepreneur based in Toronto, Canada.
