[QUOTE=Larry Swedroe]
not only have there been long periods of less than 4% real returns -there have been long periods of 0% real returns. For S & P 500 it underperformed one month cds from 66-84 and lg did so for 23 years from 66-88.Not exactly fair comparison since cds have some real return.But S & P 500 did have negative real returns from 66-82 and didnt get to 4% real return until 1991 -that’s 26 years.
Other periods of less than 4% real S & P returns from quick scan are 60-85 and 57-84 (28years)
[/QUOTE]
4% is a lot higher than the riskless rates that are available, even for long-term investments. Pay off the mortgage and you’ll reduce the possible fluctuations in your net worth. If you want to invest, you’ll need to find investments whose risk-adjusted returns outpaces the market by a percent or two, and presumably the only way to get those in an efficient market is to have better knowledge of what you’re investing in compared than the market as a whole. This is rather difficult for publicly traded securities, but maybe you know some business owners that are looking to expand their business or have another reason for a cash infusion, and you trust them with your money far more than a stranger would. So basically I’m telling you to either take the easy, safe return, or make a wildly risky investment, simply because any investment that isn’t wildly risky is likely to be priced correctly according to the market riskless rate, while your riskless rate is higher. Only those investments for which others cannot properly gauge the risk will be, on a risk-adjusted basis, a place to get a better return for your money than taking the above-market riskless rate available to you.
Of course, maybe you want more risk in your investments compared to an average investor, but it certainly doesn’t sound like it.
Also, if you pay it down, you can still access that liquidity through a HELOC, although the rate might not be as good as on a fixed-rate 30-year mortgage.
Personally, I would pay off the mortgage, and then immediately secure a home equity line of credit.
No debt, so you can go ahead and put income into retirement if you care to, and you have the security of being able to instantly get your hands on significant money.
There’s also the fact that 4.0% is a terrible rate for a mortgage and skews the equation. Even today with mortgage rates on the rise you can get a 10 year mortgage for 3.25%. If it was me I’d refinance with a 10 year mortgage. That’s what I did back in 2014 when I was able to get 2.875%.
The refinance for a shorter term might be your best bet.
One reason not to do that is that you need to come up with some serious money to cover the payments. One problem with having a mortgage and relying on investing the money to cover the payments is that with a mortgage part of your payment is amortizing the principal, which makes your payments a lot higher. So if your rate is 4%, you need to pay much more than 4% each year in order to cover the 4% interest plus the principal. So you can’t just invest the money at 4% and have the interest (or dividend) payments cover the mortgage payments.
If you take a shorter term, then you’re amortizing the principal at a much faster rate, so your payments will be a lot higher, and the amount you need to cough up over and above the return on your investments is going to be that much higher.
The idea is to see how for example in this case stock returns compare with mortgage rates as things vary in the market and the world. So in my underdeveloped study I take one case of the ‘30 yr’ mortgage rate on Date X and compare it to stock returns over the subsequent 18 yr* period. Your slightly less underdeveloped study takes Date X and Date X+1 day. So yours has twice as much information as mine? No, yours is the average of two return periods which are virtually the same 99.985% (1-1/(18*365.25)) the same daily returns which compound up to the 18 yr return. That tells you almost nothing about how mortgages rates and subsequent stock returns differ in varying circumstance which is the whole idea. The two answers have to be very close because they are overwhelmingly auto-correlated…by consisting of almost the exact same information.
And really, this point is too obvious to even find a quick link saying so. No proper academic study of comparative returns uses overlapping periods.
*due to repayment of principal the average life of a 30 yr mortgage is around 18 yrs at today’s rates.
On rereading your prior post in response to this one, I see that I misunderstood you. I was thinking of the stock market rate of return in absolute, while you were discussing the correlation of stock market returns to interest rates. You’re correct about your point. In retrospect your post is clearly saying what you describe, but I had earlier commented on using overlapping periods when looking at the stock market returns and was locked into that notion. My apologies.
That’s correct, I have no idea why Richard Parker specifically requested to compare real returns. But even if you look at nominal returns the S&P 500 had an average annualized return of just 3.07% for the 20-year period ended January 1st 1949 and a 3.51% return for the 15-year period ended January 1st 1921:
The definition of ‘sufficient emergency fund’ is not directly relevant to the key argument deployed in favor of keeping mortgages the investor has the cash to pay down: ‘stocks will make more than the mortgage rate’, implying the money would be put in stocks. But one should not rely on selling stocks to fund consumption in an emergency any more, or much more, than they’d count on selling or refinancing a home to raise cash in an emergency.
And you can’t direct excess cash to low risk assets instead of a mortgage pay down to make money, it loses money: 3.72% 18yr average life mortgage v 2.35% interpolated 18yr point on US treasury curve. Difference in tax treatment for some people and the refinance option on a mortgage will not generally turn that big negative spread into a profit.
The question IMO is what to do with the excess over ‘sufficient emergency fund’ as defined by each investor according to their circumstances (some people no longer need employment, others have relatively safe or shaky jobs etc). Subject to the investor having a view of the risk of too small an ‘emergency fund’ that’s reasonably consistent with their view of risk otherwise. But it’s not a point in favor of keeping mortgages one can afford to pay down just to say people disagree what’s a sufficient ‘emergency fund’.
The problem with nominal returns is you only have those going way back for one of the two things, stocks. There isn’t good data for 30 yr fixed mortgage rates prior to the 1970’s. And it’s not meaningful to compare a stock return from a mortgage-avg-life period of time in the distant past and compare it to today’s nominal mortgage rate. A simple illustration, the S&P return for 18 yrs from 1981 13.25%, excellent compared to today’s mortgage rate. But 30 yr mortgage rates in 1981 hit 18.5%.
You can’t take now’s lower mortgage rates out of context of the whole picture of rates and expected returns otherwise. Let’s hope the 3.72% 30yr mortgage rate now is a better deal relative to stock expected returns than 18.5% was in 1981. But you can’t assume it’s better point for point. IOW it’s not meaningful to compare today’s mortgage rate to past stock returns in environments where investors in stocks had the choice of putting their money into different, often much higher, ‘riskless’ rates than they do now.
So the reason one might quote real returns is to short cut trying to synthesize apples for apples 30 yr fixed rate mortgage data (‘long term’ mortgages used to be less 30 yrs in the distant past in the US) for pre 1970’s. And rather simply observe, broad side of barn, that real riskless rates were generally higher in the past than now, so more of the real return of the S&P was eaten up in the past by financing the position at a long term riskless rate (or the riskless rate plus mortgage-treasury spread, and that spread itself might have been higher, another missing data issue).
