Note: A new version of this post with major updates was posted in January, 2021.
How have 30 years of falling mortgage rates changed the economics of homeownership? The answer shocked me and changed my view of how the real estate market really works.
At the end we’ll compare how it affected Baby Boomers versus Millennials, but first let’s start with the conclusion and then I’ll explain how I got there step by step.
Real Mortgage Payment-Adjusted House Prices Since 1990
- House Price Index
- Inflation-Adjusted House Price Index
- Two Prices for Every House!
- Interest Rates & Monthly Mortgage Payments
- Monthly Mortgage Payment-Adjusted House Price Index
- Monthly Mortgage Payment-Adjusted & Inflation-Adjusted House Price Index
- Los Angeles – Theoretically
- Real Monthly Mortgage Payment Index
- Total Mortgage Debt
- Millennials vs. Boomers
- Millennial Inflation?
- OK Boomers and Millennials, Remember Gen-X
- Equation for Calculating Monthly Mortgage Payments
Take the Case-Shiller House Price Index
- Adjust It for Inflation
- Adjust It for Mortgage Interest Rates
Keep in mind that in January 1990, the Savings & Loan real estate bubble was peaking or had already peaked in many metropolitan areas, especially in the Northeast and on the West Coast and 1990 wasn’t a typical year at all.
House Price Index
Now it’s 2020, so let’s see what’s happened to nominal house prices since 1990.
A few different house price indexes are out there. The S&P CoreLogic Case-Shiller Home Price Index is probably the most well known. It’s my favorite. They publish data on house price appreciation for 20 cities and the USA every month. The Case-Shiller Index is based on nominal house prices.
Case-Shiller uses a price index instead of actual dollar prices which has advantages and disadvantages but it does make it easier to see price changes over time and, especially, to compare price changes over time between the 20 cities they cover.
The index doesn’t capture changes in prices due to houses getting bigger. It’s based on how the prices of individual houses change over time.
Their baseline year is 2000 but using a little math we can shift the baseline year back to 1990 so it’s easier to see the whole story of house prices unfold over the last 30 years. Case-Shiller only has data back to 1990 for 17 of its 20 cities, so when we shift the baseline year back to 1990, Atlanta, Dallas and Detroit drop out.
Inflation-Adjusted House Price Index
The Case-Shiller Index is based on changes in nominal prices. We want to see real, inflation-adjusted house prices over the last 30 years.
There are a few ways we can adjust for inflation. Probably, the most common way is to use the Consumer Price Index (CPI). Among the consumer price indices, probably the most commonly used is the CPI-U, the Consumer Price Index For All Urban Consumers.
I applied the CPI-U to the Case-Shiller Index to get an inflation-adjusted house price index. I much prefer to see the real, inflation-adjusted index whenever I’m looking at house price changes over more than a few years time.
First, in the graph below, let’s compare the nominal Case-Shiller Index to the Inflation-Adjusted Case-Shiller Index.
Next, let’s compare the Inflation-Adjusted Case-Shiller Index for all the metros they follow back to 1990.
The shocking truth is that real, inflation-adjusted house prices haven’t skyrocketed in all metro areas.
Adjusting house prices for inflation makes the graph a lot more realistic but we’re still missing a huge factor.
Mortgage interest rates have fallen a ton since 1990 and that greatly reduced the monthly mortgage payment price of houses… at least in the short and medium term.
Two Prices for Every House!
It’s like we have two different prices for every house.
- Cash Price
- Monthly Mortgage Payment Price
Cash Price. If you’re paying all cash when you buy a house, then obviously the sales price is the price you’re paying for the house. The graph above of the inflation-adjusted price index does a great job of showing you how the real cash sales prices have evolved since 1990.
Monthly Mortgage Payment Price. But when you’re borrowing money to buy a house like most buyers – especially first-time homebuyers – then the price you’re paying is less about the Cash Price of the house and a lot more about the Monthly Mortgage Payment Price of the house.
Most house buyers pay a mix of the Cash Price and the Monthly Mortgage Payment Price. If you’re putting 5% down as a down payment, then you’re only paying 5% of the price of the house at the Cash Price and 95% of the price at the Monthly Mortgage Payment Price.
To see what house prices looked like to people who are borrowing the money to buy their houses, I needed a graph that converted the Cash Price into the Monthly Mortgage Payment Price.
Interest Rates & Monthly Mortgage Payments
When mortgage rates fall, monthly mortgage payments fall even if the cash price doesn’t. How much do monthly mortgage payments fall when mortgage rates fall?
I converted mortgage rates into an index of monthly mortgage principal and interest (P&I) payments. The base of the index is the first week of 1990, therefore, all the readings show what the monthly P&I payment per dollar of mortgage amount would be relative to the same mortgage amount in January 1990.
The 30-year fixed-rate mortgage rate was 9.8% the first week of January 1990 (although the rate was above 10.0% for most of 1990). For the first week of January 2020, it was 3.7%. What does that difference mean for your monthly mortgage payment?
Hover over the index line for January 2020 and you’ll see that – dollar for dollar – the monthly principal and interest mortgage payment in January 2020 was only 53% of what it was in January 1990. That’s how much mortgage rates going from 9.8% to 3.7% affected the Monthly Mortgage Payment Prices of houses per dollar of mortgage amount borrowed.
(Also, notice an interesting quirk in the graph. Mortgage payments track mortgage rates very closely but as rates get closer to zero, lower rates have less impact on monthly mortgage payments.)
Monthly Mortgage Payment-Adjusted House Price Index
Now let’s adjust the Case-Shiller Index for changes in monthly mortgage payments due to changes in average mortgage rates so we can see how the Monthly Mortgage Payment Price Index has changed over the last 30 years.
I used the 30-year mortgage rate for every week since 1990 to calculate what the monthly payment would be per dollar borrowed and then created an index of monthly mortgage payments with the baseline being the first week of 1990.
Nice. But the graph doesn’t take into account inflation and we saw earlier the huge impact inflation has on the nominal cash price of houses over time.
Monthly Mortgage Payment-Adjusted & Inflation-Adjusted House Price Index
To see what’s really happening to the monthly mortgage payment price of houses, let’s adjust the Case-Shiller Home Price Index for BOTH inflation AND mortgage rates to get a mortgage payment adjusted real house price index.
Take the Case-Shiller House Price Index
- Adjust It for Inflation
- Adjust It for Mortgage Interest Rates
Get a view of real house prices through the lens of monthly mortgage payments for all those homebuyers who borrow most of the money to buy their houses.
Los Angeles – Theoretically
Using round, theoretical numbers, let’s say a house in Los Angeles sold for $100,000 in the first week of 1990 when the 30-year fixed-rate mortgage rate was 9.8%. The monthly mortgage principal and interest payment on that $100,000 would be $863 per month. Adjusting for inflation, $863 per month in 1990 would be equivalent to $1748 per month in 2020.
The Case-Shiller Index suggests our theoretical Los Angeles house by the first week of 2020 would have appreciated to about $290,000 and the 30-year fixed rate mortgage rate had fallen to 3.7% by then. The monthly mortgage principal and interest payment on $290,000 would be $1335 per month in 2020.
Therefore, the monthly mortgage payment for the house in 1990, after adjusting for inflation, was a lot higher than it would be now in 2020.
In round numbers, the Real Monthly Mortgage Payment is almost 25% LESS for the 2020 buyer in Los Angeles compared to the peak of the S&L real estate bubble in 1990. But the total amount of real, inflation-adjusted debt is more than 40% HIGHER for the 2020 buyer in Los Angeles compared to the peak of the S&L bubble in 1990.
Finally, let’s compare the Real Monthly Mortgage Payment House Price Index for different metros on the same graph.
Real Monthly Mortgage Payment Price
The shocking truth is that Real Monthly Mortgage Payment-Adjusted House Prices are lower today than in 1990 for the U.S. and in most metro areas.
Remember, if house prices were crazy high in a metro in January 1990 (for example, in California), then an index value of 100 equals crazy high house prices. In the three years leading up to January 1990, home prices increased almost 70% in Los Angeles, according to the Case-Shiller Index.
Also remember the index doesn’t reflect changes to house prices caused by increases in the average size of houses from 1990 to 2020. The index is based on how prices of individual houses have changed over time.
Visually, the Real Monthly Mortgage Payment House Price Index seems to me to explain the housing market better than the unadjusted Case-Shiller Index.
Total Mortgage Debt
Lower rates, however, eventually lead to higher house prices and more future mortgage debt.
Total U.S. mortgage debt on single-family houses is quadruple today what it was in 1990. Adjusted for inflation, it’s double. The U.S. financial sector is far larger today than in 1990 due in part to the growth of U.S. mortgage debt.
Millennials vs. Boomers
Who had it better, Baby Boomers 30 years ago with much lower house prices and much higher mortgage rates, or Millennials today with much higher house prices and much lower interest rates?
Looking at the Real Cash Price, Baby Boomers had it much better but looking at the Real Monthly Mortgage Payment Price, Millennials have it much better today in many places.
One thing for certain about buying a house in 2020 versus 1990 is that mortgage rates aren’t going to fall like they have the last 30 years.
The decades of falling mortgage rates helped Baby Boomer homeowners in two different ways;
- The falling mortgage rates caused house prices to increase a lot which caused Boomer wealth to increase a lot, and at the same time,
- Boomers could refinance into the lower rates, lowering their monthly mortgage payments.
Increasing wealth and decreasing expenses! Nice combination, all thanks to lower mortgage rates!
But as rates fell, people could — and eventually did in most metros — pay a lot more for houses. The lower mortgage rates got monetized into higher house prices.
The shocking truth is Millennials in 2020 have lower Real Monthly Mortgage Payments than Boomers did in 1990 for the equivalent house in many areas. The Real Cash Price, however, is much higher in 2020 and Millennial mortgage debt is shockingly high. Because they have so much more debt, fewer Millennials will be able to pay off their mortgages early. Millennials have more debt, longer.
And Millennials definitely won’t have the Boomers’ tailwind of decades of falling mortgage rates. Millennials won’t see decades of increasing family home equity wealth and decreasing monthly mortgage payments due to falling mortgage rates like the Boomers did.
It doesn’t seem likely but it’s not inconceivable, however, that Millennials could see a different kind of tailwind. The first-time homebuyers of the World War II generation, for example, benefited in a way that the Baby Boomers didn’t. The rocketing inflation of the 1970s was, in one way, a big break for World War II generation homeowners. The high inflation caused their fixed monthly mortgage payments to shrink fast in real, inflation-adjusted dollars. The U.S. dollar in 1980 was about half the value of the U.S. dollar in 1970.
Of course, the high 1970s inflation that helped the World War II generation homeowners ended up forcing the Baby Boomers to pay crazy high mortgage interest rates in the 1980s and beyond.
If, perhaps, inflation increases more than expected in the 2020s, Millennial homeowners could end up following in their grandparents’ footsteps a little bit while leaving Gen-Z or the next generation to cope with the higher mortgage interest rates that would follow.
OK, Boomer… and Millennial, Remember Gen-X
Boomers and Millennials weren’t the generation hardest hit by the Great Real Estate Bubble. Most Boomers had already bought their first houses long before at lower prices and most Millennials were still in high school or college.
The generation hurt the most by the Great Real Estate Bubble was by far Gen-X. Many Gen-Xers were set back many years financially by the real estate bust and most did nothing wrong. Their only mistake was hitting first-time home buyer age at the wrong time.
But the Gen-X story is depressing so after the real estate bust their story was buried. Sad stories are bad for business. The real estate industry instead moved on and talked non-stop about the positive impact Millennials would bring to the market when they hit first-time homebuyer age.
Since the bust, how many real estate stories have you read about Millennials? How many have you read about Gen-Xers? In real estate, Gen-X is the forgotten generation.
Equation for Calculating Monthly Mortgage Payments
Equation for calculating the monthly principal and interest mortgage payment on a 30-year fixed-rate mortgage.
P = M* [ (r/12)* ((1+(r/12))^360) / (((1+(r/12))^360)-1) ]
P = Monthly principal and interest mortgage payment
M = Mortgage amount borrowed
r = Annual mortgage interest rate
360 is the total number of monthly payments for a 30-year mortgage.
Example. If the amount borrowed was $100,000 and the annual interest rate was 4.0%, then the calculation would look like this:
P = $100,000* [ (0.04/12)* ((1+(0.04/12))^360) / (((1+(0.04/12))^360)-1) ]
P = $477 would be the monthly principal & interest mortgage payment on a 30-year, fixed-rate mortgage of $100,000 that had a 4% annual interest rate.
29 Responses to The Shocking Truth About House Prices After You Adjust for Inflation AND Interest Rates
Interesting stuff. Trying to figure how to share this
Thanks! FYI, it looks a lot better on screens bigger than phones because of the graphs.
I was able to share it in Facebook. Thanks John. Looked at it on my Pad.
I’d be curious to see how the other factors in mortgage payments, insurance and taxes, have evolved over that time period. Also curious as to whether you used the overall national CPI or used the city-specific ones, and if you used inflation rates for the economy overall or housing-specific inflation rates.
Hi Mark! I was thinking I could add those in the future. How much did rising real taxes and insurance impact monthly mortgage payments. I’m thinking the change in interest rates dwarfed the change in taxes and insurance but I’m guessing. Wow! You’re coming up with a lot of good ideas! I used the national CPI-U for all the deflating. (The Fed, I believe, usually uses Personal Consumption Expenditures as their deflator.) The next improvement I want to make would be to include changes to real household income. After that I could check out taxes and insurance (I wonder if that data is available). Mark, thanks for your great comments!
Mark, thinking about your comment, I might go back sometime and use “CPI-U Less Shelter” instead of CPI-U as the inflation corrector. House prices would end up being higher after adjusting for inflation less shelter. The Inflation-Adjusted House Price Index for USA would go from 138 to 148. On using city-specific CPI, I didn’t know that existed so that was a good tip! Looking into it, however, they don’t go back to 1990 for 4 of the 17 cities and then depending on the city it changes from monthly to quarterly, which is an additional complication. I might play with fine tuning it with city-specific inflation rates and see if it helps explains things better than CPI-U or CPI-U Less Shelter.
John- Thanks for the detailed analysis! I’ve just read it once, and I think it’s going to take a couple of times to absorb.
Let be know what you think after it sinks in. How can it be improved?
Thanks for this. I’m one of those Gen-Xers that got burned and sold our little starter condo 9 years after we bought it for the same price despite lots of upgrades. Since then we’ve been renting and have built up plenty of savings with the prospect of buying again, but our history has made me skittish.
This post was helpful in calibrating the factors at play and what we need to take into consideration as we debate buying vs continuing to rent. The takeaway I’m seeing is that housing is expensive if you’re paying cash, but not if you’re using debt. So if we do elect to buy, it probably makes sense to put no more than 20% down and to make minimum payments, letting future inflation reduce the debt burden and using the extra money to invest elsewhere. Given the money that has been printed in response to coronavirus, your prediction of inflation helping out the millenials (or us Gen-Xers who are restarting) is looking more likely. The paradox is that my first experience has given me an extreme distaste for taking on debt, but relying on debt seems to be the most economical option in the current environment. Maybe I just need to plug my nose and follow the numbers.
Hi John. Thanks for addressing my questions (and so much more). To clarify, the question about building possibly being relatively cheaper was not with regard to the purchase of new construction, but rather to the home building process itself, where the costs of materials and labor should be tracking inflation independent of interest rates. Certainly sellers of new construction will raise prices in accordance with demand, but if one’s build cost is based only on materials and labor (setting aside the cost of land), it seems like the low interest rate environment with high cash price housing will make building relatively attractive. I think I may be seeing signs that this is holding up as I’m starting to investigate the possibility of building. I would also expect that to stimulate new construction, but as you noted, these things take time.
Thanks Travis! I recently read somewhere that individuals buying a lot or a tear-down and building their own house is MUCH for common in many other countries. In Phoenix, I associate building your own house with luxury homes. Had a client buy a small 1950s house on a large lot in Paradise Valley, the most expensive area in Arizona, and they built their brand new dream house. The wife was super creative – in clothing design – so she LOVED the process. However, I think the final price ended up being like 20% more than the builder first pitched to them. So, you have a lot more price uncertainty when building your own versus buying existing.
John, this is a really awesome analysis. Thank you for sharing. As a millenial with some savings and getting ready to buy, I have this sinking feeling that the “obvious choice” of taking a big mortgage at record low rates is going to end up badly in the long run. Everyone is worried about inflation (particularly with the Corona stimulus), but as you point out the demographic curve is in favor of flattening and eventually dropping demand as boomers die off. My inner contrarian is blaring alarm bells but I’m struggling to NOT pull the trigger, especially as I’m witnessing the massive surge in real estate activity in my locale. Just based on mortgage affordability alone, prices should be up by about 9% given that rates have dropped from 3.5% to 2.8%.
What’s your best resource on the demographic trends that you talked about (expected death rates and home-buying rates for boomers and millenials, respectively)?
Narayana, Thanks for the comment. I don’t have a favorite resource for demographic trends. If you find a good one, please let me know! I would love to have a go-to website about future demographics.
About buying a house, you’re unlikely to see the tailwinds Boomers had unless you get a lot of inflation which eats away at those 30-year fixed mortgage payments (but houses prices would fall as inflation and interest rates go up). I’d just say buy a 10-year house – a house that you would be happy living in for at least 10 years – then you’re probably okay in most of the worst-case scenarios.
Reminds me, I helped my son buy their first house last year. When looking at homes, it wasn’t unusual to tour houses selling for about the same price the sellers paid when they bought the houses in 2005-2008. That would hurt. But I don’t see us having a bust like that at all. However, to be fair, I don’t see it last time, either.
It’s crazy that buying a house in the wrong year could set those buyers back a decade or more financially. The system is insane.
This analysis continues to be on my mind and I was wondering if you have any thoughts about how this relates to builders. One of the options we’re considering is building, but so far the information I’ve found makes it still look relatively expensive compared to buying when just looking at the cash prices. But when I think of it in terms of your analysis, I would expect the cost to build (except land) to roughly track inflation. If much of the current real estate gains can be attributed to interest rates, then this should favor building (relative to the past), where the material and labor costs are not tied to interest rates. Theoretically you would also expect this to stimulate the supply of new construction, but it doesn’t seem like that’s happening. How do you see your analysis applying to home building? Thanks in advance.
I’ve been planning to respond to your comment for days. Sorry about the delay. I’m experimenting with adding a forum to the website where I can respond with text or video. A framework that, hopefully, makes it easy to discuss ideas. In a nutshell, however, I’d suggest buying sooner rather than later because the lower rates now and in coming months will likely be capitalized into higher house prices over the next 2 or 3 years (or least preventing them from falling). But do NOT expect to see Boomer levels of appreciation over the longer term. Starting around 2025, the number of people hitting first-time buyer age starts to decline slightly and the Boomers start to die off. Add that to interest rates not falling anymore and the key to home equity in the future could very well be paying down your mortgage, not house value appreciation. Related to your first comment, I want to talk about, also, what happened in Texas in the early 1980s, inflation, interest rates and house prices.
Thanks John. Any thoughts on how this fits into the buy vs build question? Does it make sense that the relative economics of building should be better when the cash price index is driven higher by interest rates, since the build costs should generally track inflation independent of interest rates? I haven’t found any evidence that this is actually playing out, but I don’t really know where to look.
You have a promotional youtube video at the part of the chart that is 2020 that cant be moved? Out of all places to put a self promotional youtube video. You choose to put it where the information is.
Thanks for the tip. Fixed it.
Well done! This is an incredibly useful data set and a thoughtful analysis. It might be interesting to carry this analysis back to the 60s to see how a rising rate environment plays out, as that is likely what will happen at some point in the future. Also, an attempt to normalize around median income or unemployment in these various market segments might be interesting and may explain some of the variance we are seeing between these various locations.
Thanks, Mike! We’re on the same wavelength. I’ve been wanting to add income to the mix someday but it’s not very current that I can find. I hadn’t thought of unemployment but that dataset is super current so that might be a good proxy for income. Thanks for the tip!
The Case Shiller index is an INDEX (it does not represent actual prices) and is inherently adjusted for inflation, so it is odd that you added in an “inflation-adjusted” graph. The index value should be 100 if the relative value of the home stays constant (relative to other factors such as CPI and income levels). With your graph, you’re basically double-adjusting for inflation.
The Case-Shiller Index is based on nominal/current prices. The Case-Shiller Home Price Index measures house price inflation and is similar in concept to the government’s Consumer Price Indexes which measure consumer price inflation.
I’m trying to use this to estimate the current value of a home that was appraised at $375,000 in 2013 – what would be worth today?
Cool idea. Maybe I should create a graph where you can put in the last sales price & year and it would estimate what the current value of the house would be if it appreciated the same as houses in the metro area as a whole.
I take several issues with this analysis. First and foremost, is one of the golden rules of finance: it is impossible to price an investment without taking into account the total risk of the investment. This risk can include inflation risk, liquidity risk, transaction costs, interest rate risk, market risk, labor market risks and long term stability of incomes and expenses, and the fundamental credit risk that comes with high leverage.
1) First you need to compare things to median incomes, not inflation rates. If incomes have not kept pace with inflation then your indexed numbers (compared to 1990 base rates) cannot be interpreted as changes in unleveraged affordability because they would be equating increases in income and increases in debt ratios.
2) … and again measuring risk is paramount. You can not compare simple real median incomes in 1990 to real median incomes today without taking into account changes in the stability of incomes, the prevalence of pensions, the power of unions, etc.
3) … and today people simply have more debt to service. This is more risk. I would much rather pay a higher interest and a larger percent down payment on a moderate loan than leverage my life savings 20-to-1 on a giant loan even at a low interest, especially for long term loans. On an investment, by the way, which (according to your 3rd chart) has only returned 1.2% annually in the past 30 years post inflation. This aversion to leverage is even greater if you consider the fact that many people today have lots of college debt on to pay off. To a lender everything is credit risk, but to a borrower it’s liquidity risk and this is the ultimate form of liquidity risk because the borrower is forced to make these payments for decades or else incur the high transaction costs of the real estate market.
3) … and finally you have to take into account the fact that people in their late 20s-early 30s today MAY be forced into retirement before they will want to. What does it mean to leverage 20-to-1 on a 30 year mortgage when you won’t even be able to work for 30 years? I have spoken to many people in my neighborhood who were pushed out of the job market in their 50s in the last 5 years. Regardless of where they are with their finances, they simply can’t find jobs that are not tantamount to exploitation. Up to statistical error, no one should be banking on working in their late 60s to pay off a mortgage.
Great comment! I’ll respond tomorrow.
Edgar, Great points. On #1, I’m in the process of incorporating income, per capita income, because I can get it for the metro. Unfortunately, it’s only annual and the latest number is for 2019. 2020 data won’t be available until September 2021. It’s be interesting to see the impact of incorporating income. Household income would have been better than per capita income but I can’t find household income at the MSA level, just the state level.
On #2, I’m not even trying to quantify risk.
On #3, Good point. Reminds me that falling rates for 30 years means that the investment returns will be higher than in a flat interest rate environment. That is, that risk-taking was a lot more profitable in the last 30 years, given falling rates and increasing asset prices, than it probably will be in the next 30.
On #4, I don’t know about the retirement age falling, involuntarily. If been seeing the percent of older people working increase. But to your point about the 30-year mortgage, I agree. Especially, without falling rates for the next 30-years, a ton more home equity is going to be generated by paying off the mortgage, not through house price appreciation.
Thanks again for the comment.
Thanks for the reply John. I’ll only address your reply to point #2. My response is what I started my first comment with: it is impossible to price an investment without taking into account the total risk of the investment.
Comments are closed.