Who’s afraid of a housing crash?

November 16, 2022

After the big jump in the stock market last week, everybody’s worries should be over, right? Well, maybe not. Real estate looks a bit shaky now! Prices have come down just a notch, but is there more to follow? Are the wheels coming off? Is the market going to crash? What’s the impact of the much higher mortgage interest rates? Are we going to see a replay of the 2008 housing crash? How did interest hikes impact the housing market back in the 1970s and 80s?

Lots of interesting questions! Let’s take a look…

What’s the damage so far?

The good news is that home prices have barely budged so far, see the chart of the Case-Shiller index below; the index is only about 1.3% below its all-time high in June. The bad news is that this index is published with a two-month delay, so we have data only up to August 2022.

Case-Shiller U.S. National Home Price Index. Source: S&P Dow Jones Indices, LLC and FRB St. Louis. Nominal/Not inflation-adjusted!

So, is this small dip in July and August just the start? In the chart below, I plot the Case-Shiller index adjusted for inflation as well as the CPI rental component (also CPI-adjusted, i.e., the rental inflation over and above average CPI inflation). I normalize the series to 100 in January 2004. I like that normalization because the blue and orange lines intersect around the times when the housing market was roughly in balance, neither wildly overvalued (like 2006) nor badly bruised (like 2011/12). So, for example, 2004 was about appropriately valued, as was the 2018 up to early 2020 before the pandemic hit. And the pre-GFC market bubble shows up as about as much overvalued as the 2012 trough. Pretty well calibrated!

CPI-adjusted Case-Shiller home price and CPI-Rent component. Sources: BLS, S&P Dow Jones Indices, LLC, and FRB St. Louis.

In any case, if we wanted to force the blue line back to the orange line, we’d be looking at another 22% drop. Ouch!

But it’s safe to assume that home prices have likely declined already since August 2022. To get more up-to-date figures, I’ve been tracking the monthly Zillow estimates of our single-family home in Camas, WA, which we bought in October 2018. Home prices in our area peaked earlier, in March 2022, and our home’s value has since declined by 8.4% nominal and 12% when adjusting for inflation. If 22% CPI-adjusted is the target drop of that blue line, we’re already more than halfway there.

But as a side note, since we bought our place four years ago, it still appreciated substantially, even in line with the stock market! In fact, if you consider that an owner-occupied home also pays you an implicit “dividend” in the form of not having to pay rent, our homeownership investment is even ahead of the S&P 500 total return index. In the chart below, I assumed a 6% rental yield, which is common in our area, and a 2.5% drag from expenditures (taxes, insurance, maintenance, repairs, etc.) for a net 3.5% extra housing yield to get an implicit housing total return. So, our current house has been a pretty good investment (just like our previous one). Provided you do the math right and don’t mix up price returns and total returns!

ERN residence: price vs. total cumulative return compared to the S&P500 total return since 10/31/2018. All returns are adjusted for CPI.

What’s the impact of higher mortgage rates?

Mortgage rates have increased from around 3% in late 2021/early 2022 to about 7% just a few weeks ago (though rates have come down since then already – probably standing at about 6-6.25% now). Bummer! Going from 3 to 7 is a 2.33x move equal to a 133% increase in the rate. (side note: please note the difference between the 4 percentage point increase and a 133 percent increase!) Does that mean that home prices now have to fall by more than 50% to make up for the more than doubling of the interest rate? Fortunately not! Nonlinear mathematics to the rescue! Because the mortgage amortization formula (e.g. PMT in Excel) is not precisely linear and proportional to the interest rate. A 133% increase in the interest rate increases the monthly mortgage payment by “only” 57.8%. Please see the example calculation below, for a sample $500,000 house, a $100,000 downpayment, and a 30-year mortgage with 3% and 7% rates, respectively:

Mortgage interest change from 3% to 7% and its effect on the monthly payment.

We can now also do the following thought experiment: By how much would the home price have to fall to keep the monthly mortgage payment equal to the previous amount of just under $1,700? It turns out, the mortgage would have to be about 36.6% smaller.

But the good news is that the property value has to decline by “only” 29.3%. That’s a little bit less than the full 36.6%.

Adjust the property price to make the buyer “whole”

In other words, if you still keep the $100,000 downpayment then you would then cushion the effect on the property price.

Provided, of course, you didn’t keep your downpayment fund at FTX or some other fraudulent crypto basement exchange. This reminds me, someone on Twitter dumbed down the whole FTX, FTT, Bitcoin, SBF, CZ, Binance, etc. saga for the crypto-uneducated folks like me. I found the two videos quite entertaining.

My favorites are the “day trader” robbing the vending machine in video 1 and the “buy high, sell low guy” in video 2! Hope you got a laugh out of those!

Jokes aside, back to the housing market. Notice that very leveraged borrowers, say, with a 3.5% downpayment and an FHA loan will feel essentially the entire 36.6% increase. On the other hand, if you had a much larger downpayment, say 40% instead of 20%, as is typical in some of the higher-end homes in my town, the drop in the property price would be lower, at “only” 22%. So, the 29.3% figure is only an average.

Also, keep in mind that this 29.3% loss in my thought experiment is even more severe than the drop in the (nominal) Case Shiller Index during the 2008-2012 housing bust! But also consider that there is no contractual guarantee for the buyer to get a property with the same PITI as in 2021 and all the loss is borne by the seller. More likely, buyers and sellers will have to share the pain. In other words, property prices can certainly fall from their insane and inflated 2022 peak, maybe by 10-20%. But buyers will also have to cough up 10-20% more if they missed the boat and didn’t buy a place when mortgage rates were at rock-bottom prices. Who will be closer to a 10% loss and who will be closer to a 20% loss needs to be determined. But I would suspect that some prospective homeowners who are currently renting might be willing to choke up some additional cash because the alternative – renting – is getting more expensive at an alarming rate, more than 9% annualized over the past 3 months! Or just move into your mom’s basement!

Attempting a forecast

Going back to the time series chart with the real home price vs. rental CPI prices, what would it take to bring those two lines back into balance again? A 22% drop overnight seems unlikely. Here’s a way to gauge a more likely path forward. Let’s do the following thought experiment:

• Between August 2022 (the last month for which the Cash-Shiller index is available) and August 2023, home prices decline by 10% in nominal terms. Then prices fall by another 2% during the next 12 months. After that, home prices start rising by 2% again going forward.
• Overall CPI increases by 4.4%, 2.8%, and 2.0% over the respective next three years, i.e., 4.4% between 10/2022 and 10/2023, then 2.8% until 10/2024, and 2.0% until 10/2025. This is my personal, proprietary, ERN-lab inflation forecast.
• Also, in light of higher interest rates and to make up for past underperformance, rental CPI will increase by 8%, 5%, and 2% over those three 12-month intervals.

If I extrapolate the three series, home prices, CPI, and the rental-CPI over the next 36 months, we get the following picture for CPI-adjusted home prices and rental-CPI, see below. That looks promising! We don’t even need a total collapse of the housing market to bring valuations back to normal. By early to mid-2024 we’ll be back at pre-pandemic valuations.

Extrapolate my CPI, CPI-rental and home price forecast to Oct 2025.

A caveat

Not every asset bubble deflates as nicely as in the chart above. Even if it’s true that the fair price isn’t even so far below today’s, it doesn’t necessarily mean that home prices approach their fair value in a nice smooth monotone way. Exhibit A is the housing crash in 2008-2012. The price could very well overshoot on the downside before recovering. So, if I assume that home prices decline at a 10% p.a. pace all the way to 2025 we get the picture below.

The same chart as above, but assumes a 10% annual decline in home prices until October 2025.

I hope we don’t end up in this scenario. A roughly 30% decline in home prices is even worse than during the Global Financial Crisis. Is it possible that the inflation and central bank uncertainty could cause a housing market meltdown worse than during the financial crisis? This brings me to the next point…

Any lessons from the 70s and early 80s?

If you think a 6-7% rate for a 30-year mortgage is expensive, recall that this was the mortgage rate in the early 1970s. Before rates shot up and reached a level of 18.63% in 1981! Please see the chart below:

Mortgage rates 1970-2022. Source: CNN

Did that 2.5x move in mortgage rates, even worse than our current 2.33x move, cause a housing crash? Amazingly, it didn’t, at least not according to the housing market data I have. The Case-Shiller index doesn’t go as far enough back, but looking at median sales prices, there didn’t seem to be much of a housing crash. There was one year-over-year 0.6% decline between Q1 1981 and Q 1982 but apart from that, prices were mostly up.

Real Estate values and median sales prices in the 70s/early-80s. Source: Federal Reserve, Census, HUD.

Also looking at the data from the Federal Reserve, Publication: Z.1 Financial Accounts of the United States, Table B.101, the total value of real estate on households’ balance sheets never even went down and neither did the home equity (=home values minus mortgages). Not even quarter-over-quarter! Of course, the values in Table B.101 are the aggregates (price times quantity), so they may slightly exaggerate the price index, but since the quantity of the total housing stock doesn’t move that fast, it’s a pretty good indicator that the uptrend in the median home price is legit and not due to some hidden selection bias in home sales.

How is it possible that during this bad mortgage mess prices didn’t decline? Well, of course, prices declined, if adjusted for inflation. But in nominal terms, everybody just muddled through. As I have pointed out previously on these pages, an inflation shock, as painful as it may be, is likely not as bad for the economy as a deflationary shock, like in the Great Depression or the 2008/9 Great Recession. Rising prices make it easier for leveraged agents – both households and corporations – to grow out of their debt burden. Disinflation and especially deflation often act like a much worse wet blanket on the economy than an inflation shock!

Conclusion

Circling back to the question in the title, am I am scared of a housing crash? Certainly not! Inflationary recessions don’t necessarily sink the housing market, see the 1970s and 80s. Home prices can just stagnate for a while or maybe fall moderately and inflation will bring real prices back into balance over time. Even if the market crashes again as in 2008, so what? I no longer depend on a paycheck and we don’t have a mortgage, so we can just sit out the potential misvaluation on the downside. But so far, I’m not even too worried about that crash. There are some positive signals that the Federal Reserve will soon reach the peak Fed Funds Rate at around 4.75-5.00% and we might just thread the needle and create a soft landing in the housing market. And hopefully economy-wide. I’ll post it here if my assessment changes, so stay tuned!

Thanks for visiting! Please leave your comments and suggestions below!

Picture credit: pixabay.com