Median Home Prices Compared to Median Income Since the 1960s | Hacker News Reader

Median Home Prices Compared to Median Income Since the 1960s

thesoundingline.com

52 pointsFour_Star68 comments

This seemed highly unpleasant but ignores the effect of falling interest rates over much of the study period. I put together a plot using FRED data to show the cost of monthly housing (assuming a US-style 30 year mortgage) as well as its annual cost as a percent of median US household income.

https://imgur.com/a/0F20SGF

It could be better - I'm quite envious of those who owned a home in the 1970's - but clearly there has been some secular change in US expectations of housing costs as a percent of income. There is an upward trend but nowhere near as dramatic as this post tries to show. In 1980 housing represented 21% of income, in 2016 it was 24%.

I think that is correct. While most people buy a house -- they budget to a payment. The last time we moved our realtor even tried to engage us in this direction. It is very similar to the way car dealers explicitly work -- "What do you want to spend a month?"

If interest rates rise dramatically, average home price will necessarily fall.

This recent Indicator podcast discussed this. Surprisingly, historical data does not support the notion that rising interest rates leads to lower housing prices. But maybe this time will be different. https://www.npr.org/sections/money/2018/05/15/611386612/risi... Here is the relevant paper by Laurie Goodman. https://www.urban.org/sites/default/files/publication/92541/...

If interest rates rise and then fall again over a relatively short term (several years or longer in the case of real estate) then housing prices might not fall much because owners will pull property off the market and wait for it to improve. The reduced supply (possibly greatly reduced) will tend to drive prices higher even as interest rates are driving them lower.

> (assuming a US-style 30 year mortgage)

this is a highly biased assumption (historically speaking) and we should be hard reversing the trend of longer and greater indebtedness not blindly normalizing it.

The U.S. economy's financialization has mostly only gone up since the 20s, with occasional crises like the Great Depression and Great Recession. How would one reverse a century-long process like this?

So... higher availability of credit / higher financialization -> higher prices?

That may be it. But interest rates are a big part of this.

Most people have to take a mortgage to buy a house. This means that the demand part of the "supply and demand" curve is set, not by the price of the house, but by the monthly payment for it. The interest rate affects the relationship between the two.

For example, I bought a house when interest rates were 9%. Two years later, interest rates dropped to 7%. In the same two years, the price of my house went up by 50%. If I bought the same house after those two years, my monthly payments would have been exactly the same.

But it's not just monthly payments. Down payment is the other thing that affects the demand curve. That's where "higher financialization" comes in - at least, if I understand the term correctly.

Well, I would think that low interest rates go together with financialization... By financialization I mean the stock of credit per-capita[0].

If you look at the ratio of house prices to years of work, these are things that have happened in that time: a) more mortgages, with longer terms, b) lower interest rates, c) higher financialization. I think all of these go together.

People wouldn't borrow so much to buy a house if they couldn't keep the payments low, for which they need longer-term mortgages and lower interest rates. Conversely, home prices couldn't rise as much if people couldn't afford them. Lower interest rates and longer mortgage terms stoke demand, but also, the fact that people can afford to pay more means that suppliers will demand more too. And since suppliers have debt too, I suspect that they could only have stayed profitable by increasing prices -- that merely increasing supply to match the higher demand without increasing prices was not an option for the industry as a whole.

You're right, I'm sure, that lower downpayment requirements are part of the story, so it's not just the monthly payment.

[0] https://en.wikipedia.org/wiki/Financialization

> Financialization is a term sometimes used to describe the development of financial capitalism during the period from 1980 until 2010, in which debt-to-equity ratios increased and financial services accounted for an increasing share of national income relative to other sectors.

I would argue that low interest rates in part caused this skyrocketing home value effect for the simple fact that low interest rates have convinced people over the last 30 or so years that borrowing is more profitable than saving (I'm grouping all interest rate regimes after 1981 into the low interest rate category). Would be nice if we could put to rest the notion of home ownership as a good forced savings plan, investment plan, etc.

Google cache http://webcache.googleusercontent.com/search?q=cache:NzYN98d...

So the ratio of median home prices to median family income has moved from about 3 to about 4.5 over the last 50 years.

This is because the many manufactured things have dropped in price over that period of time. This has meant that families have more money available to spend on things that are not manufactured. Land is the epitome of something that cannot be manufactured, so goes up in value.

Here's an article on Baumol's cost disease which explains this: https://www.vox.com/new-money/2017/5/4/15547364/baumol-cost-...

> Land is the epitome of something that cannot be manufactured, so goes up in value.

Lot sizes have gotten significantly smaller over the past 50 years. The difference is probably closer to 3 vs 20 if you include the amount of land purchased. There's $800,000 homes in my neighborhood on 4,000 sqft lots.

Have any data on this? I'm skeptical.

Zillow has published some work on it -- https://www.zillow.com/blog/bigger-homes-smaller-lots-184558... but it doesn't go back very far. -10% since 2000.

I agree. Where I live there are two types of homes being built: Dense condo developments, or huge houses on large lots (1 acre+).

Nobody is building the 1950s/1960s subdivision style 3-bedroom ranch on a quarter acre lot, and I don't really see many larger homes on small lots.

Anecdotal to be sure, however.

Also median homes have nearly doubled in size.

Yes but there are a lot of variables to consider there. Most homes today are just drywall, OSB, and carpet when in the past they were lath and plaster, wood planks, and wood flooring. On the other hand people today also pay for AC, more plumbing, and better insulation.

Homes built within the past ~40 years also don't last as long as they used to.

How about cell phones and computers? Robots? The Internet?

Surely their cost decreased but that didn't mean people spend less, no they spend MORE today on those things than 50 years before.

Ownership of land is not labor intensive, its not Baumol's cost disease.

The U.S. has had two decades of intense real estate price manipulation, first with credit pre-2008 and then with low interest rates post 2008. And then there is a global situation with china pushing up global demand.

There was a study I saw a few years ago on a large number of lots in the New York suburbs that tried to tease out how much of the cost of a house was the value of the land and how much was regulatory. They looked at the prices of lots with dilapidated houses that needed to be rebuilt but which could be rebuilt as of right by the new owner. And they compared them to the price of empty lots where you would need to get regulatory approval to build a house where none had existed before. They found that lots that come with the right to build cost several times what other empty lots cost.

Comparing home price to income is incorrect. The unit of measure of home price is dollars. The unit of measure of income is dollars per year. You should compare home payment, be it a rent, mortgage + maintenance costs to income per unit of time.

Home price highly depends on interest rates which vary a lot. Currently we have an ending low interest rate period and when rates will go up, situation might change.

If you want to see how it all works, see the chart for mortgage rates: http://www.fedprimerate.com/mortgage_rates_chart-graph.htm It was going down for almost the whole period discussed.

> Comparing home price to income is incorrect. The unit of measure of home price is dollars. The unit of measure of income is dollars per year.

That doesn't make the comparison incorrect, it just means that the result is a "work time" rather than a unitless factor. And you can compare differences/changes in "work time" just as you can compare differences in unitless factors.

IMO, from the article linked the author doesn't consider the result of the division as time. He uses the term ratio.

Also consider that the standard down payment required on a mortgage used to be 20%. That limited what you could buy.

That is true, but it is slightly more complicated.

When lenders were routinely offering 100% (or even higher) LTV mortgages, the relationship is as simple as you describe. Homeowners and renters alike pay out a portion of their income each month for the roof over their heads.

When lenders require a significant percentage as a down payment, then the rate of change of house prices and the rate of change of salaries are directly linked as well as via the monthly cost of a mortgage. If prices are rising faster than you can save up a deposit and incomes are stagnant, then it doesn't matter if interest rates are falling enough to keep the monthly payments stable, you still won't be able to buy.

Bear in mind that that scenario is a Landlord's dream, they can charge higher rents because everyone needs somewhere to live, which in turn prices more renters out of the homebuying market, whilst not deterring wealthy flippers.

The regular payment on a mortgage varies even more greatly depending on some of the decisions you take when you get the mortgage than the underlying interest rate. (e.g. rate of the mortgage, fix/discount/tracker/offset, length of initial term, length of total term, upfront fees, size of the deposit).

I disagree with the conspiranoic interpretation of interest rates. There are several reasons why land is expensive globally, but if we talk about places where median income is unable to buy property like SF, Buenos Aires, Vancouver, we are talking about highly regulated spots, with high productivity jobs.

Conspiranoic? I'm just pointing out that (as solomatov summarises) that these things are complicated, and that saying that one simple comparison is wrong, and another equally simple one is right does not work.

One factor that I suspect solomatov was originally hinting at, is that back in the 1960s when house prices were low against incomes, interest rates were so high that the monthly payments, as a proportion of their income, were roughly the same as they are today. However, that ignores the fact that saving up $SALARY * 2.4 / 5 for a 20% deposit in 1960, wouldn't have taken as long as saving up $SALARY * 4.2 / 5 would today. That only considers the US national medians mentioned in the article. The really unaffordable places are even worse. The longer you need in order to reach the savings target you originally set, the more you are affected by house price increases adding even more time to that period.

As you say, there are several reasons for the unaffordability of home ownership. My point is they are not all completely independent of each other and are even influenced by house prices themselves.

None of this is conspiracy, just market behaviour.

To summarize, everything is very complicated.

Median home price does not necessarily mean that houses are appreciating. People could be building larger, more expensive houses. Building codes may also add cost to a house. Homes didn't used to have to have smoke detectors all linked together, be grounded properly, have waterproofing, etc.

1. It's unlikely the effect on price of what you are describing would have the impact that is being shown.

2. Regardless of that fact, the area between the curves is the troubling part.

Sure but material science has progressed for cheaper materials, machines have reduced the required number of people needed... It cuts both ways.

It looks like you are right about that--looking around, it appears that inflation adjusted price per square foot has remained relatively stable. However, a larger home with price/sq ft remaining the same results in a higher price.

> We can see from the second chart that a median family home has been getting less and less affordable for the ‘median family.’ Whereas the median family could earn enough to purchase a home in just 2.4 years of work in 1970 (assuming 100% of their income went towards the home), it took the same median family over 4.3 years to buy a home in 2016, nearly twice the time and work.

This struck me as a slightly strange way to frame it, since the median buyer probably still has a standard 30 year loan. The ratio going up doesn’t mean it’s taking longer to buy, it means that people are no longer using the 1/3rd rule of thumb - buy something that takes no more than 1/3rd of your paycheck to pay down.

We definitely might be over leveraged as they suggest, but could this also be a change in the buying habits with the remainer of the paycheck? I.e., the other 2/3rds or one half... maybe on average the median buyer is on average buying small things like iPads rather than bigger things like boats or college tuition? Maybe the average median buyer is just saving less for the future, but believes that putting the money into their house is a form of saving (which is often true)?

As a homeowner I do like the routine appreciation in the value of my house (up $250K since I bought it in 2012) but I wonder when the other shoe will drop. I suppose to a large extent how this plays out will vary quite a bit by region, but it seems inevitable that there will be some form of correction in the future.

The "other shoe" is that your children (if any) will have to move away to buy homes in the future.

Or live in their parent's home with their spouse and children.

That doesn't sound bad at all.

In some countries you just build another floor each time more family moves in.

>> $250K since I bought it in 2012

That is really, really bad.

What if it's a hundred million dollar palace?

How do you figure? You have no idea what the initial home cost is. 250k on a 1.5 MM home.. yea sure maybe that's whatever. 250k on a 400k home is insanely good.

Rapidly appreciating housing values are an abject policy failure.

Great for homeowners, shit for everybody else.

Is that increase greater or smaller than investing in an S&P 500 tracker would have been? That's gone up 2.1x since mid 2012.

I live in the Cambridge, UK. Since 2012 it feels like very many things doubled in price: houses, FTSE shares, food, classic cars, software engineer wages. I guess what really happened is someone printed a lot of money and a pound sterling is worth about 60% as much as it was before.

In a sense, the other shoe may already be on the floor. The value of something you don't want to buy or sell is essentially irrelevant to you.

However, because the percentage increase applies broadly (though not uniformly) across the range of homes in your area, it is harder to upgrade (this may hit you from both sides, the house you want is getting further out of reach, and your house is getting further out of reach for the prospective buyers of your current house).

If you are already in the house you want to stay in until you leave feet-first, then the change in value is irrelevant.

If you are in an area with rocketing prices, and you want to move somewhere more stable, then you are on to a winner as long as the situation continues.

If you are in an area with rising prices and you want to upgrade within the same neighbourhood, then if the situation continues, you will eventually find yourself stuck.

Well isn't it normal? Food has become cheaper since these days. Stuff has become massively, incomparably cheaper. Travel has become cheaper. And Americans had enough of all these in 1950: it was a rich country already, very few went hungry or unable to buy clothes or household stuff, so as prices fell, possibility of increase of demand were limited, these markets, except travel, were already saturated.

So a lot of money released by these things getting cheaper had to be spent somewhere, and eventually it was spent on things which aren't so prone to technology-driven deflation, mainly real estate.

This would make sense if the housing were of higher quality than it used to be. I think the same house dated 50 years ago might be worth more of a median price today, even though its the same but older house.

House built 50 years ago is firstly, about half as big as today. Personal living space more than tripled in 100 years and more than doubled in 50 years.

But what is the price change of that same house? If the house of old is worth more today, its higher value for the same unit of space (and also stock decay).

Are old houses cheaper or more expensive today, taking into account house depreciation?

As is mentioned in the comments of the article itself, this initial analysis ignores the fact that the median square footage has also increased dramatically during the same time period.

Of housing or of new housing?

Based upon that older neighbourhoods should lose value as they age and they lose relative space.

It would be really cool to see the median home price to family income plotted as a topographical map.

Not that we don't all expect it but you'd basically get mountain ranges on the coasts. The really interesting thing would be to look at it over time and see when and how fast the "mountain ranges" develop. Obviously the trend it generally upward but it would be interesting to see when and how fast prices increase in various locales and see if that correlates with anything else.

4.5x would be a dream for Australians - Sydney and Melbourne are currently at 13x & 10.5x respectivly. Those two cities cover close to half the Australian population. Insanity.

Dont worry, it will go back to that eventually. They both have plenty of land, im sure that its a momentary phenomena.

It’s been about 25 years so far, with no real downwards correction. Definitely a major bubble, but ive given up trying to guess when it will burst. Everyone acts like it will go on forever, but that would mean >$50mil median homes in another 20 years which is surely impossible (but what do I know)

Then don't buy at all, and work in selling houses :)

It's not surprising that home prices have gone through the roof, since many government policies are intended to keep home prices rising.

I'd be curious to see how this relates to the increased urbanization of our culture. It would make sense for property value to rise if more and more people are trying ot fit into a fixed amount of city space.

Reply on news.ycombinator.com