Tontines vs life annuities

The case for tontines: part 2

Why this Medieval risk-sharing concept beats annuities

Getty Images

Getty Images

Last week in this space we looked at a Medieval risk-sharing concept called the tontine, as reprised by finance professor Moshe Milevsky in his recently published book, King William’s Tontine: Why the Retirement Annuity of the Future Should Resemble its Past. In brief, a tontine is a seemingly morbid concept that involves the pooling of investments such that those who die end up subsidizing those who keep living. 

The prolific Milevsky, in concert with certified financial planner Alexandra MacQueen, also recently published the revised international edition of Pensionize Your Nest Egg. A must-read for anyone lacking a Defined Benefit (DB) pension, the new edition is aimed not just at Canadians but also English speakers in the United States, the United Kingdom, Australia and New Zealand.

We’ll look at this other book in more detail in an upcoming column but suffice it to say for now that Milevsky makes a distinction between a real pension—the DB pensions on offer by employers and also government benefits like CPP and Old Age Security (OAS)—and capital-appreciation vehicles like RRSPs, TFSAs and even Defined Contribution pensions. The latter he does not consider true pensions, which must have an element of a guarantee of lifetime income, or at least “highly predictable” income for life.

The book chiefly describes how these capital appreciation vehicles can be converted to true pensions, making it useful for anyone who doesn’t have a gold-plated, inflation-indexed employer-sponsored DB plan. And the main vehicle he mentions is life annuities.

But what about tontines? Strangely, despite his writing an entire book about tontines, the word does not appear in either edition of Pensionize Your Nest Egg. And what exactly is the difference between annuities and tontines?

I pressed the busy Milevsky for his definition of the difference and here’s his reply, via email:

“The difference between the tontine and the annuity is that with a life annuity there is someone standing behind the ‘deal’ guaranteeing the payments. This requires capital and is costly. For example, if people live longer than anticipated the sponsor has to make extra payments. On the other hand, with a tontine the group itself shares in the risk. If people live longer, the payment is reduced to the entire group. So, while this introduces a certain element of uncertainty to the group as a whole, the reduction in cost—due to the reduced need for capital and reserves—makes it a viable competitor to the annuity.”

Milevsky also elaborated on the difference in a recent presentation for the Individual Finance and Insurance Decisions (IFID) Centre, entitled “Designing longevity insurance for the 21st Century.”

There he said that with a tontine, the total interest paid each year to those within the group stays constant, while with a life annuity, as people die, the total paid out to the group declines over time.

A second distinction is that with a tontine, payments to survivors increase—and the longer they live the more they get! By contrast, with a life annuity, payments and income to survivors stay constant.

Citing Scottish social philosopher and political economist Adam Smith, Milevsky says the tontine “appeals to the gambling instinct”—the same confidence that frequent buyers of lottery tickets seem to exhibit, despite the poor odds of winning. As Smith said, the tontine annuity “generally sells for something more than it is worth.”

In most of the world today tontines are no longer permitted, although Milevsky’s book makes the case for their revival in the 21st century. There he also notes that there are several countries that utilize the “tontine principle” in their pension plans, but don’t actually call them tontines. In fact, he adds, one of the largest pension systems in the world—the New York City-based TIAA-CREF—uses a tontine-like system to adjust payouts.

Back in the 17th century, old age pensioners often had the choice of a tontine or a life annuity. Even though the tontine in England in 1693 paid only 8% and a life annuity paid 14%, the promise of a bigger payoff if you outlived everyone else still caused some to choose the tontine, since like a lottery someone had to win. The winning nominee in King William’s Tontine of 1693 was a 100-year old female.

But what about 2015? Milevsky asks us to imagine a tontine-like product in which expected cash-flows increase at the rate of inflation, are capped at advanced ages, are homogeneous across demographic cohorts, and that require no capital. This means you may be able to get 10% or 15% more income on average, as long as you’re willing to share some demographic risk with your neighbour. “Remember, this isn’t stock market risk!”

His question for you, as well as the financial industry and its aging customers, is whether you would choose to “allocate” such a product as an alternative to a life annuity.

Jonathan Chevreau runs the Financial Independence Hub and can be reached at [email protected].