Actually true for most workers in “defined contribution” plans
In my last blog post I started with the basic idea that a report on retirement plan accumulations would be a lot more meaningful to the recipient if, instead of simply reporting on the accumulated assets under current market conditions, it also reported on the proportion of current salary that those assets would replace at retirement, under current market conditions. And I came to the (at first sight ridiculous) conclusion that such a report, for my hypothetical 45-year-old worker, would have revealed almost a doubling of that replacement percentage between the end of 2021 and the end of 2022 – apparently ridiculous because, as we all know, both the stock and bond markets plunged in 2022. So, how on earth did that replacement ratio go up so much? (Or at all, for that matter.)
The explanation is actually very simple, yet most unusual because people don’t think of this perspective. (At the end I’ll also explain other ways to calculate the replacement ratio, but you’ll see that they don’t affect the big conclusion that, from this worker’s perspective, 2022 was a great year for the prospect of ultimate retirement.)
***
OK, here’s the explanation.
The big economic thing that affected those retirement assets in 2022 was the huge rise in interest rates: the yield on 10-year US Treasury bonds, for example, rose from 1.52% to 3.88%. (That came about because inflation was high, and interest rates needed to rise to offer some protection against future high inflation.) That huge rise in interest rates had a big impact on both bonds and stocks. Both plunged in value. Let me explain why that was perfectly reasonable.
Start with bonds. Why did bond prices decline? Because if you owned the right to a stream of future payments equal to 1.52% of your assets, and interest rates went to 3.88%, that means that new money could buy a stream of future payments equal to 3.88% of your assets; and so nobody would pay you face value for your 1.52% stream that your bond holdings represent. That face value would go down a lot. And yes, the values of the existing bonds in the portfolio fell, by 15% to 20% of their previous year-end value.
The same thing with stocks. Whatever the profit or dividend stream you projected at the end of 2021, that stream would be worth much less if new money could automatically earn a higher interest rate. And so stocks also fell in value in 2022. Even after receiving stock dividends, the MSCI World Index stock return was roughly (-)18% for 2022.
And that’s why the return for the hypothetical 70/30 mix of our worker’s portfolio was reported as (-)17.76% for 2022.
All perfectly true and understandable. And misery-inducing.
But wait! Interest rates were now 3.88%, at the end of 2022. Remember that these retirement projections always assume no change in current conditions, meaning that the stock market’s value doesn’t increase in the future (actually, that the stocks are reinvested in bonds) and interest rates don’t change. Well, think about the consequences of that “interest rates don’t change” scenario. It means that the depleted assets will earn a higher rate of interest in the future, and at the assumed retirement age of 66 you can buy an annuity at that higher interest rate.
And for our 45-year-old, the impact of 21 future years of earning 3.88% a year instead of 1.52% a year, and then buying a much cheaper annuity at age 66 because then the annuity purchase price would build in a future (for about 25 years) interest rate at that higher 3.88% – well, together they combine to make the salary replacement ratio much, much higher than at the previous year end, when interest rates were only 1.52% and the assumption was (as always in these projections) that they wouldn’t change.
You get it? The rise in interest rates more than compensated for the huge fall in asset values. And so 2022 hugely improved the salary replacement ratio for our worker. What a great year for retirement investing!
Looking backward (“here’s what happened to your assets”) produced an extremely negative perspective. Adding a forward perspective made the world seem a much healthier place. (Rather like: “Over this past year, most of the time I wasn’t very well. But now, my future health looks stronger even than before my illness.”)
We’re totally used to looking only at a backward perspective, not only in retirement reports but in newspapers, in what market commentators say, and pretty well everything you hear. Maybe these two blog posts will convince you that that’s only half of a valid perspective, and that the forward look should always also be included. It’s not just “What happened to my assets because interest rates rose?” More completely, it’s “Now that my assets have fallen in value with the rise in interest rates, will they actually replace more or less of my salary when I retire?”
[Note for techies. If you’ve followed my writing over the years, you’ll recognize that I’m stressing the concept I call my Personal Funded Ratio as being far superior to just the asset value, because the Personal Funded Ratio uses not only the asset value but also the value required to finance a given income stream, and it’s that required value that has fallen so much in 2022 with the rise in interest rates.]
***
I noticed that, in the 15 years I used for my hypothetical worker, 10-year US Treasury bond interest rates also rose a lot (specifically, I’m thinking of increases of more than 1%) before, in 2009 and 2013. Those were also years of negative returns on the retirement portfolio. And sure enough, they were also years when the replacement ratio rose steeply.
And in fact the years when interest rates fell, and investment returns were high, were typically years in which the replacement ratio either fell or didn’t go up much. In other words, for our hypothetical worker, who is saving in order to replace some large proportion of salary at retirement, the message delivered by those reports was in the opposite direction to that which a more sensible measure (“How much of my salary are my retirement assets projected to replace, if market conditions don’t change?”) would have induced.
***
Naturally I wanted to check, at least approximately, whether my hypothetical 45-year-old was typical of all workers, in receiving this miserable-but-it-should-really-be-positive message after 2022. For example, a 25-year old would surely have received an even happier 2022 message than the 45-year-old in salary-replacement terms. Why? Because the 25-year-old would have had 41 more years of interest at 3.88% before the annuity purchase, whereas the 45-year-old only had 21 more years.
Fair enough. But then the natural question is: how about workers older than 45? How about, for example, at the extreme, a 66-year-old active worker about to retire and (hypothetically) purchase an annuity? Answer: close to break-even, but still a positive message. The 66-year-old would also have lost between 17% and 18% in 2022 (as it happens, regardless of the fact that at 66 surely the glide path asset allocation wouldn’t have had nearly as much as 70% in stocks), but would have been able (because of the more favorable annuity purchase rates at the higher interest rate) to overcome that and purchase 4% more income than at the end of 2021.
In other words, all active workers, in my hypothetical example, would have had a favorable investment year in 2022!
***
Finally, let’s mention some of the things you might change in my example. They’re technical things. You don’t have to use 10-year US Treasury bond yields as defining the current set of interest rates; you could use 20-year Treasuries or a full year-by-year yield curve. You don’t have to use a joint-and-50%-survivor annuity to reflect the income capable of being purchased at retirement; you could use a single life annuity or a joint-and-100%-survivor annuity. You don’t have to assume the purchase of a level-dollar annuity; you could use an inflation-protected annuity instead (though you won’t find one in the US or Canada, or most countries in the world).
And so on. None of these changes would change the conclusion that 2022 was a favorable, rather than a disastrous, year for workers saving for retirement.
Won’t interest rates change again in the future? Of course they will. So too will the value of equities. That’s what is implied by taking risk. That’s what causes these projections to fluctuate over time rather than remain stable: the world changes, and we want to see whether the changes have improved our prospects or made them worse. That’s why we need these reports. (I hope we don’t need them monthly or daily – surely that would be obsessing over short periods in what is essentially a long-term commitment.)
But please, in these reports, help workers to look forward, not just backward. 2022 is a year that illustrates what a huge difference there is between those two perspectives.
***
Takeaway
When interest rates change, bond values, stock values and annuity purchase prices also change. Remember to take them all into account in retirement reports, because the traditional assets-only retirement report often produces exactly the opposite perspective from what a report on salary replacement produces.
2 Comments
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.
Don
Does your calculation take into account the higher inflation?
Having a fixed nominal salary is no good when inflation is higher which is why rates have increased. The pattern will still be there, younger workers benefit from the higher discount rates, but the break-even point (age) won’t be as high for an inflation-adjusted salary/pension.
Across the lifecycle there will be this pattern of price changes affecting savers of different ages differently. Young people who have a lot of asset purchasing to do over their lifetime will want prices to be lower before they purchase their investment assets. Older people, including retirees, want prices to keep going up so that they get the most for their investments when they sell them. An odd observation of this is that there will be an age where these benefits offset: when markets fall, the loss on the assets will be offset by the reduction (discounted) liabilities. This age will vary with different market adjustments, but it is always there.
(1) No, it doesn’t take the higher inflation into account. It should, of course. You’ll be glad to know that you’re in good company, in questioning that angle. So, permit me to deal with it in the next blog post. My only reason for not dealing with it in that blog post is that I wanted to keep the post focused just on the angle that the replacement ratio is not just informative in a very useful (indeed, essential) dimension, but that it also reveals a totally different perspective from just looking backwards on the asset values.
(2) Your point about price changes having different implications for buyers and sellers is an outstanding one, and (your “odd observation”) the fact that there is always some age at which a price change makes no difference over a lifetime is a profound one, something that had never occurred to me. So I draw it to the attention of our readers. Thanks very much! Your students are lucky to have you teaching them.