Low interest rates are bad for households with assets, but good for households with liabilities (loans).
For example, I was able to recently refinance my mortgage. So although the asset part of the balance sheet isn’t expected to do so well, I at least have a pretty sweet interest rate on my mortgage.
The balance sheets of retirees are dominated by assets, however. And while debt is on the rise among retirees, it’s the assets that fund retirement.
This suggests bond investors are going to have less money 10 years from now, in today’s dollars, than they do today. Plus, this doesn’t even factor taxes and fees, which will make things even worse (call it at least a -1% real return over the period).
Not exactly exciting prospects.
I’ve done quite a bit of research on the implications of low bond yields (and likely low equity returns) with Michael Finke and Wade Pfau. As a result, I think it’s always important to remind advisors, retirees and pretty much anyone who will listen that most research on optimal retirement income strategies has been based on historical U.S. returns, and the historical yield on bonds has been way higher than it is now.
The average yield on 10-year government bonds going back to January 1870 is around 4.5%. If we start in 1926, which is when the Ibbotson SBBI dataset begins (a common dataset among academics) it’s slightly higher at around 4.9%.
That means the current yield on government bonds is about 4% lower than the historical long-term average. Tack on investment fees, which used to be significant (at least 1% of assets), which lots of studies and advisors ignore, and you’ve got a very different investing environment today than we’ve seen historically.
Despite this fact, many financial advisors are still using historical long-term average values in financial plans. I think that’s beyond lazy, it’s almost malfeasance.
While I get it’s impossible to know what future returns are going to be (if I knew I sure wouldn’t be telling you!), they need to be reasonable — since these calculations form the basis of savings, retirement age and other expectations. In other words, I don’t think implicitly assuming 10-year government bonds yield ~4.5% (the long-term historical average) is reasonable.
When I wrote some of the original research on the impact of low yields on retirement, I remember some advisors (and reporters) complaining that the research was capturing a short-term anomaly and that low rates wouldn’t stay low for long.
Here’s the thing, though, interest rates are lower today than they were back then when we did the research. That begs the question: How low will rates go (and how long will they stay there)?
I’ve got some bad news for you — the historical trend isn’t all that promising. There was some fascinating research released earlier this year by the Bank of England, which many of you may have seen, that contains data on the historical trend in interest rates from 1311–2018.
It shows that the historical trend for interest rates has been to decrease over time. When the piece was released the actual real yield noted (in 2018) was positive (0.51%), which was above the trend-implied value of .19%.
As of July 31, 2020, the real yield was -1.0%, which is below the implied value but generally consistent with the longer-term implied relation.
Now that I’ve refinanced my mortgage, I certainly hope that rates rise again (so I can earn a higher rate of return on my assets!), but I’m not convinced they are going to (or that they should).
A Long Road Ahead
Only time will tell what’s going to happen, but I think it’s time we start preparing investors for the possibility of a harsh investing reality that could last for awhile.
For lots of people this is going to mean working longer, saving more, spending less in retirement. For someone who has already retired the options are even more bleak: spend less.
That’s why it’s so important to start setting expectations as early as possible, because it gives people more options.
I’ve seen retirement projection tools that use 100,000 runs, which is a bit absurd, but are based on purely historical returns. That’s ridiculous.
You could do a billion runs using historical returns, and I would contend that it would be less useful than one that does 100 runs but is based on more realistic forward-looking returns.
In summary, interest rates are low today, and it’s not clear how low they are going to go and how long they’ll stay there. But if we don’t start preparing our clients for the possibility this isn’t a short-term anomaly, lots of people are going to be significantly unprepared for retirement and have relatively few options to make up for lost time when they get there.
David Blanchett is head of retirement research for Morningstar’s Investment Management Group. His comments on ThinkAdvisor are expressed independently from the research firm.