I do not remember how and when I discovered HonestMath.com, it must have been on Twitter.
Talking about Twitter, I fool myself all the time by thinking I can use it like a sort of RSS feed: someone posts a nice link, I go and read it. It happens very rarely. This is why, while I saw many (well, at least before Elon fu*&ed up everything) tweets from HM, I only went recently to their website to check their research.
They produce a report called “Mortgage Math” where they challenge the napkin math of comparing mortgage rates against an average or projected investment return: “if the amount of your expected investment return exceeds the interest rate on your debt, then investing the money provides a simple and profitable arbitrage opportunity”.
This blog was not born to talk only about leverage but I find myself circling around this topic more and more lately. As someone who tried multiple times to transform, at least partially, his mortgage into an interest-only debt to further delay the eventual repayment, I was really interested in their counter-argument.
This comparison is more straightforward than the other mainstream debate, owning vs renting, where many idiosyncratic elements can tilt the dial in one sense or the other. Here there are some tax implications to consider but not for me because:
- I do not receive any relief for my debt (maybe I will in Switzerland? If I do, it is even more of a reason to have debt)
- I do not pay capital gain taxes on my investments (that was true also in the UK via ISA)
Their analysis excludes taxes as well.
The Math
“Our approach to simulation analysis uses fatter tails, multiple time series models, and forward-looking capital market expectations. We also allow users to evaluate the impact of untimely “Black Swan” events, and emphasize that future events are not limited to historical precedent.“
I like their approach in general but that “forward-looking capital market expectations” deserves some words. It is true that using past bond returns, when rates went from double-digits to zero, as a ‘blind’ expectation for their future performance is wrong: at maturity, a bond will have a nominal return equal to its coupon (if fixed and if it does not default before). There is your expectation.
But when it comes to stocks, those expectations have to pull a looooot of weight. GMO anyone? If I have a cent for every person that tried to time the market because their CAPE-or-whatever model said that it was ‘expensive’ I won’t be here writing this blog. That’s what we are talking about, lowering/increasing future expected returns and acting on it means that you can improve your returns, which means that you can beat the market.
Here is Ben Carlson opinion on the matter:
On a 30-year horizon, stock returns are remarkably stable. Sure, there are some positive and negative spikes and we are Dollar Cost Averaging in this exercise, so the average balance will be invested for a shorter period, but still…
It makes a lot of sense, in theory, that future returns should depend on how much the stock market is cheap or expensive today, but in reality, it is close to impossible to profitably act on this intuition.
Why does this matter? Because in the Crunching The Number section of their paper, HM keep constant the interest rate on the mortgage (3.5%) while testing scenarios with different expected returns taken from shops known for their ability to generate alpha from macro calls: Vanguard, JPMorgan and BlackRock. Srsly, if it was up to me I would have kept the expected returns fixed at a level that I would consider long-term ‘fair’, 9% -ish?, and tested different mortgage costs. Conceptually, I would consider relevant the element I can control, when to take debt and at what rate, rather than scrapping the bottom of my espresso to divine where the market might go.
My ideal break-even analysis would be to start from an even more conservative expected return, 8.5%, and see which rate brings the frequency of success to 50%. I invest because I believe that over the long run, stocks will perform in line with the past. If I would not have this conviction, then there would be no point in investing, leverage or not leverage.
Historical Analysis
For the above reasons, I found their historical analysis…curious. I understand that the past is the past, those were the numbers at that time, and I am probably too influenced by the fact that in Europe we had very low rates for a very long time, but…who df would take a 8+% debt and think “ok, let me use this to buy more stocks”?!?
I find the premise not realistic at all. Sure, on the napkin you have 8.43% < 10% (?) so I guess the preliminary test is…passed? Plus, again by chance, Investor B starts investing aggressively in the stock market very close to that cycle bottom, 2009, after having repaid in haste a somehow expensive (?) debt. All combined, it feels to me that the historical scenario is pretty close to a ooopsie scenario, from a cost AND sequence of returns perspective, for Investor A and yet…they are still neck-to-neck with the other guy?
Feels like a win to me
The historical analysis and this:
look like a very compelling case to actually conclude that the investor SHOULD leverage.
A turn of events
A day after I read the paper, I found myself discussing with HM on Twitter and the discussion helped me to understand their point.
From a pure financial return point of view, Investor A vs Investor B death match, whoever has more at the end survives, they are right: the “napkin math” fails to consider the impacts of leverage, volatility, and sequence of portfolio returns.
But this is not a death match. This is not a Safe Withdrawal Rate type of exercise. Investor A does not LOSE MONEY, they simply have a lower return compared to a different portfolio. Actually, I would not criticise someone that says that the dumb napkin math is a good rule of thumb to solve a complex problem involving the impacts of leverage, volatility and sequence of returns. You do not win every time but, especially if you apply a bit of margin of safety, the odds of getting ahead are pretty good.
Other Factors
The main reason I debated with HM was that, while reading their paper, I had other features of having cheap debt pinned in my mind and I evaluated the paper’s conclusion along with them. In our discussion, HM pointed me to the fact that they also considered those features in this passage, included in the Executive Summary: “The decision of whether to be a cash buyer, mortgage prepayer, or full-term mortgager might best be served by prioritizing factors other than perceived arbitrage investment opportunities, such as budgetary constraints, special or unusual tax implications, lifestyle preferences, risk aversion, and personal goals.”
As I explained on this blog in the past, there are some critical advantages of having cheap debt that I consider pivotal in such a conversation. Mortgages do not hold a special place in my heart, they are simply the (almost) only way to achieve cheap leverage for a common lad. Even when strictly pondered as debt-linked-to-a-house, mortgages provide two benefits:
- flexibility: in stress-testing parlance, there is a crucial difference between capital and liquidity (that’s what I am dealing with these days so, take it). If you use your cash to repay a mortgage, your capital position does not change, you are exchanging one asset for another. On the other side, your liquidity position deteriorates massively. If your car breaks down and you need to buy a new one, good luck asking the bank for your money back (in Europe we do not have HELOCs). This flexibility has a cost and on this blog, you can find portfolio ideas that should provide high-enough returns with low-enough volatility. How aggressively you want to repay your debt is a function of the utility you get from flexibility.
- diversification: for the majority of us, a 20% down payment to buy a house represents a massive portion of our Net Worth. Aside from the illiquid nature of the asset, I do not understand why any risk-conscious person would dig themselves more into this risk, committing additional capital than what is strictly necessary. Whatever study you read about house prices, your very very very personal situation will never match that (yes, you can get higher returns than the average). 45% of my NW is stuck in an apartment in London; I do not care that I love the apartment, I do not care that it is cashflow positive and will be more so because I was a genius extremely lucky in negotiating my mortgage. I would sell half of the package in a heartbeat if only I had the opportunity.
Sequence of Returns
HM correctly highlights that mortgage leverage is not constant, thus escalating the investor’s exposure to a sequence of return risk. But investors are already exposed to sequence of return risk, as explained in the paper linked in this post. Buying a house is a significant market timing call on the housing and stock markets (if this person cares about holding stocks, obv). In the investor’s overall context, delaying the mortgage repayment also reduces their sequence of return risks because they can better spread through time their exposure to stocks.
Take again the three scenarios in the Crunching the Numbers sections: the investor does not know if the future will shape as predicted by Vanguard, JPM or BlackRock. But the market call is actually refraining from being exposed to stocks, the 15-year Borrower scenario, not the reverse.
The key element is the mortgage rate. I do not think anyone could draw a definitive line in the sand like: if the mortgage rate is >7% -> repay the mortgage, otherwise invest in stocks. But it is only around that figure that I would start to ‘hedge my bet’, i.e. consider a shorter mortgage length. I would up the duration if given the possibility of lower rates; for sure, I would not go below 50% LTV to preserve diversification.
Strange World
[rant /on]
I do not like to live in a world where Ben Felix has to caveat covered call strategies are sub-optimal with but if you have an “I like golden showers” utility function then they are fine. The more I think about it, the more I realise I am in the wrong, it’s their money and they are free to do whatever they want. But then, this is my blog, so here you have it. I do not like it at all.
In the mortgage world, the equivalent is the person that rushes into repaying it because “I do not feel comfortable being in debt”. Are you worried about being in debt? What about having all your savings stuck in a single, very illiquid asset? Ain’t that scary or are you that have a weird sense of fear? I guess that when things go wrong, they can always go on the news and blame the government or the immigrants or globalisation.
More flexibility is strictly better than less flexibility. That’s why option prices do not go negative. But maybe one day we will have that as well.
“I would like to have less flexibility AND lower returns, please“. Sure, let me offer you this grand Private Credit Fund monsieur, and for the wee-wee, sorry I have to finish this beer first.
[rant /off]
I do not know if I will even own a property here in Switzerland but if I do, it looks like I landed in the perfect place. Swiss mortgages are unusual in that they are divided into two mortgages.
The first mortgage will typically:
- Cover around 60–65% of the purchase price
- Have an indefinite repayment period
The second mortgage will typically:
- Cover around 15% of the purchase price
- Have a fixed repayment period, usually around 15 years. So you will pay off about 1% of the mortgage each year for the first 15 years.
The remaining balance is the 20% mandatory downpayment. So the focus is on paying off the second mortgage rather than the first. Many Swiss homeowners never fully pay off their first mortgage…and by the look of it, the country is still doing just fine.
What I am reading now:
