One of the idiosyncrasies of the ERN family early retirement plan is that it involves a relocation. It’s not that we don’t like our current location. But even with our nest egg solidly in the seven figures we likely couldn’t afford to retire here comfortably because of the insanely high housing costs. The state income tax rates are also unpleasantly high. So, if everything goes well we will relocate to another state with low or no income tax and lower housing costs.
The options we consider:
- Own a house, mortgage-free
- Own a house, plus mortgage. But what term: 30-years or 15-years?
- Rent a house or apartment, long-term
- Nomadic lifestyle: have no fixed residence, move from place to place with light luggage
Ok, I have to admit, I threw in that last option just for fun. Some people can pull it off (GoCurryCracker), but I doubt that the nomadic lifestyle is for us. I like to have a home base! The way I can tell is that as much as we love to travel, it’s always nice to come back home to sleep in our own bed. Even if I know I have to head back to the office the next day. Seriously!
Quantifying the tradeoffs
We can write as much as we want about the pros and cons of renting vs. owning, but in the end, it all boils down to the numerical assumptions, especially the rental yield (annual rent divided by purchase price):
- If we can rent a house for only 5% p.a. of the purchase price or less it’s likely a no-brainer to rent. The opportunity cost of our money tied up in a house plus the depreciation and taxes would be too large. Unless, of course, we factor in huge property appreciation. But our baseline assumption is that property values appreciate with the rate of inflation. The last time folks were budgeting outsized returns in housing it didn’t end so well, remember 2008/9? So, renting can be much smarter than owning, see some examples at 10!Rocks and Millenial Revolution.
- If the annual rent is 10% or more of the purchase price, it’s almost a slam dunk to buy.
Somewhere in between has to be the sweet spot. Let’s check where’s that crossover point in the rental yield!
Here are the assumptions:
- Purchase price: $300,000, which buys us a 3 to 4 bedroom, 2 bathroom house in a nice middle-class neighborhood. We stay away from the big McMansions when practicing Stealth Wealth!
- Rent per month: We look at six different rental yields: 5, 6, 7, 8, 9, and 10% of the purchase price in annual rent. Per month this implies a rent of between $1,250 and $2,500 in steps of $250.
- Mortgage rates: 15-year: 3%, 30-year: 3.75% (did anybody else notice the recent rise in mortgage rates? Oh my!!!)
- The cost of owning: Some of the cost assumptions and estimates are borrowed from the excellent posts on Slowly Sipping Coffe and MoneySmartsBlog to account for repairs, taxes, and insurance.
- Repairs and maintenance: $7,500 p.a. (2.5% of the value)
- Taxes: $3,000 p.a. (1.0% of the value)
- Insurance: $1,500 p.a. (0.5% of the value). This value might be a bit high because the renter would also need renter’s insurance and general liability and the $1,500 figure would be in addition to the few hundred dollars a year in renters insurance.
- Inflation: 2% p.a. (=0.165% compounded monthly). The following variables grow at this monthly inflation rate: home value, rent, repairs, taxes, insurance
- Transaction costs for home purchase: 1% of the purchase price upfront, 7% of the home value on the backend.
- No impact on taxes from either mortgage interest or capital income.
- We assume we use the standard deduction in retirement. There is no tax reduction from the mortgage interest paid!
- All dividends are qualified and capital gains are long-term taxed at a zero rate in the first two federal income tax brackets.
- Equity investment expected return assumptions:
- The mean expected return (nominal): a conservative 7% p.a. (=0.565% compounded per month). This is the nominal return, so we target a real equity return of around 5% p.a.
- The standard deviation of equity returns (for Monte Carlo simulations): 16% p.a. (=4.34% monthly)
- Notice that all of our numbers are scalable. If you wonder how this exercise would look like for a $240,000 house, then the crossover rental yields are the same and all the $ values would be scaled by a factor of 240/300=0.80.
Let’s look at how the cash flows compare across the different housing choices. The 7% rental return translates into a monthly rent of $1,750 (adjusted for inflation every month, just for simplicity). After 10 years the final rent is well in excess of $2,000. Buying a house with cash sets us back $303,000 (purchase price plus 1% fees). The monthly costs are only around $1,000 (maintenance, taxes, insurance). The value of the house, net of 7% transaction cost and the final month’s maintenance cost amounts to only slightly under $339,000. With an 80% mortgage, the upfront cost is $240,000 lower but the monthly costs are much higher, too, because of the mortgage payments. Paying off the remaining mortgage balance after 10 years when selling the house yields net proceeds of around $245,000 (15-year mortgage) and $150,000 (30-year mortgage), respectively.
How do we compare which option is best? Renting costs less upfront but yields no payoff at the end of the 10 year period. So, what if we were to invest the incremental cash flow of owning over renting in our go-to investment vehicle, i.e., an equity index fund? Let’s look at the numerical results below!
Numerical results Part 1: Assume a deterministic 7% p.a. equity return
As a function of the house holding period and the initial monthly rent, here’s the incremental return over renting. We do this for the 10-year horizon, as well as shorter holding periods of 3 and 5 years. The first result that jumps at me is that paying for a house with cash seems quite unattractive. But it’s not too surprising: You tie up $300k in a house that’s a depreciating asset (4% carry cost, only 2% gains p.a.) and incurs a total of 8% transaction cost. The renter, on the other hand, can generate around $21k in annual capital gains and dividend income, in a very liquid investment.
Buying a house with 20% down and financing the rest with a 30-year mortgage looks more attractive than buying and paying with cash. Over short horizons, the real estate transaction costs of around 8% are going to hurt the homeowner. But over a 10-year horizon, the rent crossover point is somewhere between $1,750 and $2,000 monthly rent. So probably around 7.5% yield is the cutoff.
Numerical results Part 2: Assume a random equity return (Monte Carlo Simulations)
Nobody can guarantee exactly 7% on their investments. How much uncertainty is there in the estimates and how much of an advantage or disadvantage over renting can we expect for different returns over the 10 year period? Let’s get the computer warmed up and run 1,000 Monte Carlo simulations with 120 random draws of monthly returns with 7% p.. expected returns and 16% annualized risk.
Let’s look at the table below for the distribution of incremental performance over the rental, both for owning+pay cash (top portion) and owning with a 30-year mortgage (bottom portion). The median incremental return is roughly the same as under the fixed 7% return assumption above. Paying cash already becomes attractive at a 9.5% rental yield. But at $2,000 (8% rental yield) you still have a 38.6% chance of coming out ahead of the rental and even at $1,750 rent you beat the rental with a probability of almost 30%. With a mortgage, you beat the rental of $1,750 with a probability of over 36%.
Let’s look in more detail at the $1,750 rent assumption and see what’s the origin of the large range of possible incremental returns. Several $100k in both cases!
In the scatterplot below, I plot the incremental return of owning+pay cash as a function of the average equity return over the 10 years. Large shortfalls occur when the equity market is doing really well. That makes perfect sense: The renter can invest a cool $303,000 more into the stock market upfront. The homeowner without a mortgage has slightly lower cash flow needs than the renter throughout the 10 years but has no chance of ever catching up with the renter if the equity market is on a roll and returns 10%+. Likewise, it doesn’t take much of an equity return disappointment to beat the renter. Average returns of about 4 maybe 4.5% and lower and we’ll likely beat the rental. One could interpret the home purchase as an insurance against a bear market. It will cost us around $118,000 over ten years but pays off handsomely if the equity market return assumptions don’t cooperate.
Same for the 30-year mortgage scenario, see chart below. True, you’re most likely to fall behind the rental but the house plus mortgage is a hedge against poor equity market performance. When the renter beats us by six-figures or more (equity return of 10%+), we wouldn’t be too distraught, knowing that our seven-figure equity portfolio also went through the roof.
Numerical results Part 3: 15-year vs. 30-year mortgage
Despite using a significantly lower equity expected return (5% real vs. 8.4% in FL’s case), we get the same result. The 15-year mortgage comes in a bit behind the 30-year mortgage. But we don’t want to dwell on that and rather point out the fact that going with the 15-year mortgage over the 30-year mortgage has that same equity hedge feature we saw before. With more noise, that’s for sure, but there is definitely a negative correlation with the equity return. As crazy as it sounds but we may entertain a 15-year mortgage despite giving up a little bit of expected return. Or at the very least, use the mortgage choice as an equity market timing mechanism: If equities seem expensive (CAPE-ratio) then go for the 15-year mortgage. If equities seem cheap we’d go for the 30-year mortgage.
Other factors to consider
- We’ll be stuck in the new location for 10 years to make homeownership work. A lot of things can go wrong during that time. Renters have the option to just pack up and leave without incurring a 7% loss when selling the property. That’s worth a lot!
- Rental inflation could be higher than overall inflation! Two recessions and a housing crash haven’t prevented rental inflation from running hot since 2000. Specifically, since 2000, rental inflation was 3% p.a., compared to only about 2% for overall inflation, see chart below. Higher mortgage rates will only fuel more rental inflation going forward!
- In retirement, we will have no more regular employment income. Unless we finalize the mortgage application before I hand in my resignation letter we will have a tough time securing a mortgage with a competitive interest rate.
- Tax bracket mayhem: the extra cash flow needed for either the mortgage or rent, may push us into the 15% tax bracket for ordinary income and/or the 15% bracket for capital gains. Suddenly an 8% rental yield becomes 0.08/0.85=9.4%. Buying a house with cash could look attractive again.
Conclusions
- The financial aspect of the rent vs. buy decision in early retirement (or any other time, for that matter) boils down to how long we want to stay in one location and what’s the prevailing rental yield.
- With a conservative equity market expected return of 7%, 4% carrying cost for the house and just under 4% mortgage interest for the 30-Year, it looks like the crossover point is somewhere at 7.5% rental yield. Buying a house with cash seems unattractive for rental yields under 9.5%.
- Use caution when looking exclusively at point forecasts and crossover points! Having a house, whether paid for with cash or with a mortgage offers a hedge against bad equity returns. Homeownership beats renting exactly when the stock market performs poorly. Likewise, if the stock market were to go up by more than 7%, true, we’d fall back behind the rental but with our 7-figure equity portfolio going through the roof, we’d still be happy
campers,uhm, homeowners. We may probably shave off another 0.5% or 1.0% from the rental yield to account for this hedging feature, so 6.5-7.0% minimum rental yield when buying with a mortgage and 8.5-9.0% minimum rental yield when buying with cash. - It would be foolish to tie such a profound decision to just monetary factors. We’ll have to weigh some purely emotional factors, like the pride of homeownership vs. being stuck in one location.