Life After Full-time Work Blog

Learn about preparing for life after full-time work through posts from Don's upcoming book.

#158 Inflation-Protected Lifetime Income Streams

If only they were available!


Following my post on longevity insurance, my friend Dr Zvi Bodie (the Norman and Adele Barron Professor of Management at Boston University until his retirement) mentioned to me that, while a few years ago there was one US insurance company that wrote inflation-linked annuities, today there are none. That’s something we both regret. (I understand that in Australia, Challenger Life writes them – a rare stand-out globally.)

In this blog post I’ll explain how an inflation-protected annuity works, why I think the inflation protection is a valuable feature, and why most people tend not to appreciate the value of that link.


First, how it works.

Very simply. It’s a cash flow stream bought from, and guaranteed by, an insurance company. The company makes periodic (yearly, quarterly, monthly, whatever) payments to the purchaser, for a specified period, which could be for a fixed term or for as long as the beneficiary is alive or a “joint and last survivor” (the longer of two lives) . Periodically (typically, once a year) the payment is adjusted to reflect the amount of inflation since the previous adjustment, according to a specified index.

That takes a lot of words to describe, but think of it this way: payments (typically for life) increase with inflation every year.


Second, why the inflation link is a valuable feature.

Again, very simple. Wouldn’t you like to receive an income stream that increases every year by enough to combat the effects of inflation? Of course you would! Whatever purchasing power you have in the first year, stays with you for as long as you need it. You don’t need to complain, “Prices have gone up by 7% (or whatever) since a year ago, so I have to pay 7% more for the same stuff, and whatever I paid last year no longer buys the same amount of stuff as before.” No more “Boo hoo” attitude.

So, obviously the inflation protection is a very valuable and desirable feature. And it’s feasible for the insurance company to guarantee it, as they can buy TIPS (Treasury Inflation-Protected Securities – bonds for which the periodic interest payments and the returned capital are increased to match inflation) from the US government, and similar securities in other countries. Clearly, then, the risk of keeping up with inflation is taken by the government rather than by the insurance company or the purchaser of the income stream.


Now to the crux of the matter. Why don’t people clamor for this sort of payment stream?

Is there a catch to it? In a way, yes. For a given amount of capital (the “purchase price,” as the insurance company calls it) the income stream starts out at a lower amount than would a stream that has the same purchase price and does not increase with inflation.

Zvi provides a numerical example.

Suppose you have a purchase price of $100,000 and you want a 20-year payment stream. (Ignore the lifetime feature, for the purpose of this illustration.) And suppose interest rates are 2% a year, and “real” interest rates (as offered on TIPS) are 0%. “Real” here means “in addition to inflation,” so obviously, for the buyers of these bonds, the expected rate of inflation is 2% a year.

If you buy a 20-year unvarying income stream, with payments made at the start of each year, you’d get $5,995.76 each year for your $100,000.

This is the stream that doesn’t change with inflation, so each year it would buy you less. How much less? Well, that depends on what inflation turns out to be, of course. But let’s say that it turns out to be exactly the 2% a year that those government bond buyers expect. Then, by the 20th and final year of the contract, the $5,995.76 you receive would only buy you $4,115.67 worth of what you could buy in the first year. That’s what inflation does: the same amount of dollars buys you less than before.

Suppose, instead, you bought an inflation-protected income stream with your $100,000. Then you’d get $5,000 in the first year, and each year after that you’d get $5,000 increased by the aggregate inflation since the start of the first year. (For example, at the start of Year 20 you’d get $7,284.06.) So your purchasing power would stay at whatever $5,000 would have bought you in the first year.

That’s your choice. You can get a fixed $5,995.67 every year, buying you less every year (eventually buying you, at the start of Year 20, the equivalent of $4,115.67 in the first year), or $5,000 worth of purchasing power every year. The two streams have the same present value, of $100,000.

If inflation (which is unknowable in advance) turns out to be less than 2% a year, the $5,995.76 income stream will turn out to have greater purchasing power over the 20 years. If inflation turns out to be more than 2% a year, the “real $5,000” income stream will turn out to have greater purchasing power over the 20 years.


To make it easier for me to understand the difference, I did a thought experiment, changing the context of the two income streams.

I imagine two countries, the first with the dollar as its currency, the second with a currency called the StableUnit. It’s very easy to exchange the currencies, at no cost. (Of course, this is my thought experiment, so I can specify any conditions I like!)

The two countries have identical goods and services for me to buy. In the first country, inflation increases the prices of the available goods and services. In the second country, all StableUnit prices are fixed, and never change.

Today the exchange rate is $1 for 1 StableUnit. Of course, it can’t stay that way when inflation comes! It’ll cost more than $1 to buy 1 StableUnit in the future, so each dollar will buy less and less of the goods and services I want. Inflation is, of course, identical in the two countries, in my thought experiment.

I like to visit the StableUnit country for the first month of every year. I enjoy the country, and spend a lot of money there, having fun. (Again, this is my thought experiment!)

I have $100,000, which of course today I could convert to 100,000 StableUnits. And I plan to visit every year for 20 years, and buy an income stream with my $100,000. My question is: should I buy the income stream here, or buy it in the StableUnit country?

If I buy it here, I’ll get $5,995.76 a year for 20 years, and I’ll take it to the StableUnit country and spend it there. Of course, it’ll buy less and less every year.

If I buy it in the StableUnit country, I’ll get 5,000 StableUnits a year, and it’ll buy the same goods and services every year.

What should I do?

There’s no right answer, in the sense that every other answer is wrong. It depends on what I think inflation will be, and my tolerance for living with inflation risk. And that’s where this thought experiment becomes interesting.

I don’t know what inflation will be, and that’s crucial. If I’m pretty sure it’s going to be much more than 2% a year, I’ll obviously go for the StableUnits; and if I think it’s going to be much less than 2% a year, I’ll go for the dollars. But what do I know about the future? Nothing! So my expectations don’t incline me one way or the other.

What I do know is that inflation is uncertain. And I don’t like uncertainty. So I’d rather buy the inflation-proofed income stream, that is, the stream of StableUnits, because inflation-risk-avoidance is, to my mind, a good thing, and in this thought experiment I can get it at no cost.

It’s not just me who feels that way. Risk-avoidance is inherent in the psychological make-up of human beings. If I’m offered a prospective reward for taking inflation risk – let’s say, the dollar stream is priced lower than the StableUnit stream today – then maybe I’ll be tempted. But with no reward, I’ll choose the risk-avoidance stream over the risky stream. Quite simply, it would be pointless to take the risk, because it gains me nothing.

The irony is that in fact the dollar stream is the risky one, even though it appears to give me the certainty of knowing exactly how many dollars I’ll get each year. In reality it’s the uncertain stream – in StableUnits – that actually protects my purchasing power, and therefore is the riskless one.

So, what this thought experiment tells me is that, given this choice, and other things being equal (whatever that implies – I’m not trying to sneak in other conditions) I’d prefer the StableUnit income stream.


And yet the vast majority of potential purchasers make exactly the opposite decision in practice. Given a choice between a nominal income stream (the $5,995.76) and an inflation-proofed income stream (the 5,000 StableUnits), they go for the nominal income stream.

The demand for StableUnit (inflation-proofed, inflation-adjusted, inflation-linked, call it what you will) income streams is so small that only Challenger Life, in Australia, offers it.

To me, this is counter-intuitive, and requires an explanation.

I don’t know for certain what that explanation is. But I can make a pretty good guess. And here’s my guess. It’s the way our minds work. And there are three aspects that I think are relevant.

The first is that we don’t understand compound interest, which describes the way purchasing power decreases over time. It’s what makes the $5,995.76 able to purchase only $4,115.67 worth of goods and services at the start of the 20th year. And we never see that $4,115.67 number. We still get what looks like $5,995.76. And we have no idea how much less it purchases in Year 20: we just know things cost a lot more.

So we see $5,995.76, and it’s bigger than $5,000 (or 5,000 StableUnits) today, and we never see the decline and crossing over of the value of the two income streams explicitly. We have no way of knowing that the 5,000 StableUnits at the start of Year 20 will then be worth $7,284.06 in terms of Year-20 dollars.

So we prefer the $5,995.76.

The second reason is that we tend to underestimate our life expectancy. I explained why in an early blog post. Essentially it’s because our average survival age increases, the longer we’re alive. For example, we may remember something like: US average life expectancy is 76 years for men. What we don’t realize is that that 76-year number is held down by those who will pass away early. For those who survive to, let’s say, age 65, the average survival age is a further 18 years, that is, to age 83. So when we reach age 65, our actual future life expectancy is several years longer than the 76 years we think it’ll be.

And so, when we see an income stream that starts off higher, we think we’re unlikely to get to the crossover point, and we prefer the dollar stream.

The third reason is that we see the two streams and make an instantaneous and instinctive decision. Nobel Prize winner Dr Daniel Kahneman wrote a wonderful book called Thinking, Fast and Slow, in which he explained that the emotional part of our brain thinks instinctively and totally dominates the much-slower-thinking rational part of our brain. In fact, it takes a large amount of will-power for us to say something like, “Wait, if I have time, perhaps I should think this through and not just go with my gut feeling.” And in this case, as I explained earlier, the rational risk-avoidance strategy is to buy the StableUnit stream. And even then, we’d need to understand the points I’ve made in this blog post to reach the rational decision! The gut-feeling dollar stream wins.

No wonder there’s so little demand for inflation-indexed income streams. I don’t know what the demand is in Australia, but I’m glad Challenger Life offers them, and I hope that, if you’re in the market for a lifetime income stream you’ll think about this, and send it to anyone you know who might also be in the market. And perhaps together we’ll be able to persuade insurance companies both inside and outside Australia that there really is a demand that they should consider satisfying.


Just to say this explicitly: I have no interest, financial or other, in any insurance company anywhere. This post simply represents my views on a subject that Zvi and I got to discussing following my earlier post on longevity insurance that happened to mention inflation-linked streams in passing, and we believe this subject will become increasingly important, not only in days of high inflation, but at all times.



Inflation-protected lifetime income streams are both feasible and desirable.


I have written about retirement planning before and some of that material also relates to topics or issues that are being discussed here. Where relevant I draw on material from three sources: The Retirement Plan Solution (co-authored with Bob Collie and Matt Smith, published by John Wiley & Sons, Inc., 2009), my foreword to Someday Rich (by Timothy Noonan and Matt Smith, also published by Wiley, 2012), and my occasional column The Art of Investment in the FT Money supplement of The Financial Times, published in the UK. I am grateful to the other authors and to The Financial Times for permission to use the material here.

4 Responses to “#158 Inflation-Protected Lifetime Income Streams”

  1. leonard wissner says:

    Such an instrument could be created synthetically using Treasury strips and inflation swaps issued in the U.S.
    In the 21 century rates were quite attractive almost 3% real at the beginning of the century. They are unattractive now but with the Fed abandoning its quantitative easing program they might become attractive again. I would certainly go for a real yield of 3-4% per annum.

    • Don Ezra says:

      Thanks for this angle and for the link. I confess I am not familiar with the world of inflation swaps. My interpretation is that setting up a portfolio of the required kind is feasible if there are counterparties in the market willing to take on the inflation risk.

  2. Paul Owens says:

    Don: great article. Very clear and logical. Fits perfectly well with my experience in plan administration, unfortunately.
    2 examples:
    1. Pension income options at retirement:
    a) pension life only = $1,000/ month
    b) pension guaranteed 10 years = $ 950/month
    c) pension 60% joint and survivor = $900/month
    Member invariably chooses the $1,000. Why? Because it is the biggest number

    2. Termination options upon termination of plan membership at say age 45:
    a) deferred pension payable at 65 = $1,000 a month
    b) lump sum commuted value transfer to LIRA/LIRRSP = $50,000
    Again, member usually chooses the $50,000 transfer. Why? Because it is the bigger number

    The pension industry as a whole has made an erroneous assumption that their client/member understands as much as the plan sponsor/service provider does. Very few members understand any of the following:
    a. life expectancy
    b. interest rates
    c. present value
    d. optional forms
    e. actuarial equivalents
    And we allow them to make uninformed life altering retirement income decisions.
    The pension industry made a serious mistake in offering options without ensuring the audience understood what they were choosing. In retrospect, the pension industry and legislation should have provided only deferred pensions payable at age 65. Married members should have been able to select a pension with a minimum survivor pension of 60%.
    Paternalistic? Yes and deliberately so. Had we done that there would be more retirement wealth payable to members and survivors – many of whom waived their entitlement to survivor pensions and only learned the consequences of their actions when their spouse passed away and the life only pension that was chosen stopped.

    • Don Ezra says:

      There speaks the voice of experience — thanks for your wisdom! In effect, I see you suggesting a default option for decumulation — something that is desperately needed. Members can override the default, but at least a default tends to do better than most individuals are likely to do without the expertise built into the default.

Leave your question or comment