Arnold Kling  

House Prices

PRINT
The Two Employment Surveys... The Free Trade Case...

Brad DeLong says that in the housing market, one person's capital gain is another person's capital loss.


Yes, many people who have refinanced have now boosted their own consumption spending because they feel (and are) richer. But why haven't those who will buy your house in thirty years and their parents cut back on spending by an equal amount as they strive to accumulate the bigger nest egg that they will need?

What Brad is saying is that there are people who are "short" in the housing market. Children and others who will own a house one day are in this short position now. As home prices go up, they lose, but they do not reduce consumption to compensate.

This weekend, I became curious enough about high home prices in our area to check them against a rule of thumb. My rule of thumb is that the ratio of a home's price to one month's rent should be roughly 300. The rule of 300 is based on a price-earnings ratio of 25, where earnings are the annual rents on a home. Multiplying annual rent by 12 months gets you to 25x12 = 300.

Anyway, I used the ads in the Washington Post to find a home for sale and a home for rent in our area that are roughly comparable. The rental was for $1750 per month, and the asking price on the home for sale was $549,000. That is within the ballpark of the rule of 300. My conclusion is that home prices are not a bubble in our area. If the ratio of home price to monthly rent were close to 400, that would have me very concerned.

One little-remarked fact about the 1982-2000 stock market climb is that housing values did relatively poorly over that period, so that in many locations the ratio of prices to rents fell below the rule of the 300. While the run-up in house prices since 2000 is significant, it has done little more than catch up for the previous poor performance.

As of now, I think that there is little or no margin of safety in buying a house, at least in the Washington, DC area. My guess is that if you buy now, you might see paper gains for another couple of years if prices overshoot their fair value, but over the long haul your house will not outperform other assets in terms of risk-adjusted return.

For Discussion. Try testing the rule of 300 in your area. Remember that you need to find a home for rent that is comparable to the home for sale against which you are benchmarking. Use the comments section to report your results.


Comments and Sharing





COMMENTS (27 to date)

That is an awfully high P/E ratio for a rather utlity like asset as a house.

There is also the possiblity that housing, in general, is overpriced - and both the earnings and therefore the underlying assets are too hight.

Larry writes:

How did you come to your rule of thumb? Is 25 the long-term historical average of home P/E?

Kerf writes:

Question: Why the 25 p/e ratio as a rule of thumb? I dug up the following from an article in Fortune from December 2003 suggesting that a p/e ratio of 20 is steep:

One way to look at the economics of your home is to use what we'll call a price-to-annual-rent ratio (or P/R), the rough equivalent of a price/earnings ratio for a stock. Like stock P/Es, they hew toward long-term averages. But also like P/Es, P/Rs can defy gravity and reason. Prices sometimes jump far faster than underlying rents, driving P/R multiples to extremely high levels. Danger looms when future rents look as if they can't grow fast enough to satisfy the market's Brobdingnagian expectations.

That's precisely what has happened. The P/R multiple is extremely high both by historical standards and compared with ratios for apartment buildings. Over the years P/Rs for both single-family dwellings and larger rental structures have averaged between 11 and 12. Since the beginning of the boom in '98, however, when both stood near 11, P/Rs for houses have blown ahead of those for apartment buildings. Rental P/Rs rose sharply with the strong economy, then fell with the recession. As rents dropped or leveled off, apartment prices cooled but kept the gains they'd registered from 1998 to 2001. Their multiples rose from 11 to around 13 in most markets.

By contrast, houses totally broke loose from the pull of falling rents. Prices just kept skyrocketing. Since 1998, P/Rs in the 20 hottest markets, including San Diego, Los Angeles, Denver, Boston, and New York City, have jumped, on average, from 11 to 20—an incredible 82%.

To be sure, housing prices had plenty of solid, logical reasons to soar. When interest rates plummeted, investors rightly decided that they could pay a lot more for houses and still do better than the meager returns on government bonds. What's more, once-onerous transaction costs for purchasing homes have shriveled. In the early 1990s banks charged two points to process the average mortgage; today the number is 0.4. And towns have been shrinking the supply of buildable land by restricting development. That's making it difficult for builders to meet the brisk demand for houses.

Those factors justify at least part of the explosion in housing P/Rs. The rub is that the main reason they've jumped—the steep fall in carrying costs—is already reversing. Once again, think of your house as an investment. If the expenses rise, the net rent, or cash flow, to investors shrinks. That effectively lowers the value of the building. Say a home in a hot market—New York, Denver, or suburban Washington, D.C.—is worth $800,000 today and carries a mortgage of $600,000 at 6%. The mortgage payment is $36,000 a year. In addition, the owner pays $22,500 a year in property taxes, not an incredible figure in the New York or Virginia suburbs. Total carrying costs are $58,500.

My question: What is the right answer?

Arnold Kling writes:

If the right P/E for houses is 11, then the risk-adjusted interest real interest rate for houses is about 9 percent. I find that really hard to believe.

I can believe that the historical average for the P/E ratio for houses was something like 11, but I think that reflects houses being underpriced. Just as for most of the 20th century, stocks were underpriced.

Brad Hutchings writes:

Wow. I am surprised that your number is so high. My SoCal condo had a comparable sale for about $380K last month, yet the realtors tell me I could rent it out for $1800 - $2000. Perhaps that means that there is still a lot of room for prices to climb?

I do see more people who are buying more stretched than ever. Arnold, do you have any sense that the business model of lenders may allow for higher than historical default rates? I.e., are lenders making more risky loans because they've figured out how to make more money on defaults?

-Brad

Eric Krieg writes:

I agree with DeLong (as painful as that is for me). I feel house poor right now, even though I have owned it for 5 years.

Yeah, I have had some great gains in the price of my home. But everything around me has gone up in price as well. So, I am not really any better off, because I can't afford to move (at least in the Chicago area).

Tapping into that capital gain for the purposes of consumption would be financial suicide. Anyone who does that is a moron.

One area that I am better off in is in my ability to improve my property. I can tap into that equity to add square footage and ammenities. For example, I knocked down a 1 car garage and built a 2 car, and I expanded a half bath into a full bath, both of which raised the value of my house more than the cost of the projects.

My long term plans include an extension that would expand my small kitchen, improve the quality of my family roon, add another half bath, and a master bedroom/ master bath.

Judging from my neighborhood, I am not alone in my home imrpovements. We even got our first knockdown, where an outdated ranch home was kocked down and a McMansion is being put up in its place.

Eric Krieg writes:

I'd like to see more research into asset inflation. While the last twenty years have seen inflation conquered, it seems that assets like homes, stocks, and (maybe this is a stretch) education have gone through the roof.

Someone coming out of college has to contend with student loans, buying a home, and funding their own retirement. That eats up a great deal of income, if the person isn't completely discouraged from saving in the first place.

We now have no-money down, interest only loans. If young people STILL can't afford homes, where can the housing market possibly go from here?

Bernard Yomtov writes:

I've always thought of buying a house as a hedge against rising rental costs - an inflation hedge. So I would expect the price/rental ratio to move with expected inflation. If you think rents are going to increase at 10% a year you will pay more for a house than if yoy think they will go up only 5%.

Bob DObalina writes:

I knocked down a 1 car garage and built a 2 car, and I expanded a half bath into a full bath, both of which raised the value of my house more than the cost of the projects.

Did you perform the labor yourself, or is the market really that inefficient?

Eric Krieg writes:

>>Did you perform the labor yourself, or is the market really that inefficient?

It's supply and demand, right? There is no necessary 1:1 relation between cost and price. In fact, most home improvements won't increase your home value more than the cost of the improvement.

It just so happens that 2 bath homes with 2 car garages are scarce in my neighborhood. And in fact there are many people who will not even consider 1-1/2 bath, 1 car garage homes. I have expanded the potential number of buyers.

I did the half bath myself. That is where I get the most bang for my buck, because I spent less than $1000 to do it, and it would have cost $5000 to have it done.

JT writes:

Arnold writes: "If the right P/E for houses is 11, then the risk-adjusted interest real interest rate for houses is about 9 percent. I find that really hard to believe."

Can you outline how you arrived at that number, Arnold? Thanks.

Arnold Kling writes:

Assuming no inflation and no risk premium, then Pr=E. That is price times the interest rate should equal earnings (or rent). So, if P/E=11, then r=1/11 = .09

Albert writes:

I think that a better seat of the pants calculation is Pr(1-T) = E, where P = Price, r = interest rate, T = tax rate, and E = rent (earnings). Then, if the federal plus state taxes is 35%, this gets r = 1/11 * .65 = 5.85%, which isn't too far from the going rate.

Albert

JT writes:

Thanks to both Arnold and Albert. Two comments:

1) Why the real interest rate? Rents have not been moving with inflation in the last couple of years, I believe. In any case, the homeowner is paying the nominal rate and needs compensation for that. Please explain the logic here in more detail.

2) Albert: I think that assumption of a 35% tax rate may be incorrect. I'm not a tax expert but my understanding is that the tax write-offs available to owners of rental properties make the investment substantially more attractive. That 35% number is likely too high.

Eric Krieg writes:

Taxes make the calculation very difficult.

Interest on the mortgage is deductable for the majority of people, especially for someone in the 35% tax bracket.

Property taxes are paid by homeowners, but they are also paid by landlords. Conceivably, the cost of property taxes are rolled into the rent. They probably should be subtracted out of the rent when doing the P/E calculation.

And property taxes are deductable off income taxes for homeowners as well.

I don't think it skews things much to simply ignore taxes.

Frank DeWith writes:

This discussion seems to be equating monthly rent with earnings. It does not take into account expenses... periods of vacancy, cleaning, advertising, maintenance, repairs, property taxes, arbitration, uncollectable rent... I'm not a landlord and don't know what percentage of rent is actual profits, but if it is 50% then the ratio of house price to rental earnings is doubled.

Frank

Eric Krieg writes:

Okay, if we're going to play this game...

You can't just look at realtor.com and figure out real prices from asking prices. You need to find actual selling prices.

And you need to take into account that the realtor gets 6% of the selling price as their fee.

Brad Hutchings writes:

Bob asks: Did you perform the labor yourself, or is the market really that inefficient?

Sometimes, we do projects we can afford to have others do simply because we enjoy the charcter building aspect of the project. Sometimes, we can't afford to outsource and have the added benefit of getting to enjoy the character buidling aspects of the project!

-Brad

Bernard Yomtov writes:

Did you perform the labor yourself, or is the market really that inefficient?

It would only be inefficient if you didn't assign a cost to the hassle of getting the work done, to go with the actual bill. Eric says it would have cost $5K to have it done. This could reasonably lead to an increase of, say, $8K in the value of the house if it's worth $3K to a buyer to avoid having to deal with the construction.

Just a thought.

rvman writes:

I think the inefficiency question has to do with the "increase value more than cost" part. The housing market, to answer, is not, and cannot, be perfectly efficient. If the price of housing with certain characteristics close-in to downtown rises, it is not necessarily possible to build more with those characteristics.

Besides, the efficiency argument is predicated on the idea that, if something were profitable to do, somebody would do it. In this case, "somebody" is Eric. It's the old "$20 bill" joke. Two Rat-Ex economists are walking down the street. One sees a $20 bill lying on the sidewalk. The other says "It can't be $20. If it were, someone would have picked it up by now."

Ba-Dum-Bump Crash.

Matt Young writes:

If the rising costs of housing is normal when computed on a PE basis, then inflation must be underestimated. Rents rising fast to keep up with the PE would indicate fast rising wages.

JK writes:

You do have to look at the net yield. Houses around here (upper end Bay Area suburb) cost about $900,000 and rents seem to be stuck at about $3,000 a month.

But if you buy that $900,000 house today, property taxes are going to be $1,000 a month. Subtract another $400 a month for maintenance and your net pretax yield is only $1,600 a month. That's about a 2%p.a. pretax cash yield on a long term asset. Either you're going to make an additional capital gain or a bond would be better.

This suggests to me that the current level of house prices is supported only by the anticipated further rise in the level of house prices, which is pretty much the definition of a bubble.

Housing bubbles can get larger than this. The Economist currently estimates the extent of housing price overvaluation as follows:

US 23%
Australia 33%
UK 50%
Spain 68%

The other three are still rising strongly as I write this!

Eric Krieg writes:

$1000 per month for property taxes?!? Yikes, that's high. That's more than my mortgage.

dsquared writes:

Arnold, your PER calculation has to be calculated on the basis of the net rental income, not the gross (that's more like a price/sales ratio). Your 25x PER is equivalent to a rental yield of 4% (sorry for the inversion, but I work faster in yields 'cos I know them) Deduct 1.5% for maintenance costs and voids and the owner's getting 2.5% on his money, so the actual PER is more like 40.

To get a 4% yield net of costs (and 1.5% is a low estimate of voids & maintenance, btw, and I believe that a US investor ought to be factoring taxes in somehow but I don't know the details), you'd need a gross yield of 5.5%. That's more like a PER of 18 and a "rule of 220".


The

PC writes:

This is an interesting site with all the discussions, but I found this because I was searching for some info about the relationship between inflation and housing prices. Is there anyone who can explain to me why our reported annual inflation rate is down around 3% while the home prices are increasing at around double digit rate? Is the house price not a component of inflation? If not, it should be. After all, salary adjustments are somewhat tied to the inflation (for those who are lucky to have a job and a salary adjustment).

I find it hard to believe in the reported inflation numbers. It now costs around $10 to go to a movie theater, over $2 per gallon for gas, and it always clean me out when I go to Longs Drug for a little bit of medicinal supply. How about the cost of higher education. That's also growing at double digits. My niece is going to college next year, and her parents are figuring it will cost them about $150,000 for the next four years. Ouch!

I wonder if it would be prudent to take out a home equity line of credit to finance the purchase of a smaller rental property in a less expensive neighborhood. Is this a good way to hedge against inflation? What do you think? Too risky or too much trouble?

Would anyone consider it crazy to take out home equity and put it in REIT's? Does anyone have some insight to how REIT's will perform over the next decade or two against inflation? Lastly, does anyone know the relationship between inflation and prime rate (which is index many home equity line of credit is tied to)?

Katie writes:

But what about the costs of not having a home? I live in the SF bay area and still rent. I wanted to toss my tax papers out the window, thinking about how much I could save if I simply had a home. I'm actually considering buying this fall, simply because I'm throwing away such giant gobs of money on rent! Of course, I don't even know if I could afford a house out here now, much less in the fall. Or if house prices will keep going up. All I know is that my rent is awfully high, and since I'm so "rich" I don't even qualify for some tax credits. (Rich! HAH. Not for out here.)

Steve Goldfarb writes:

Background:
Our 60+ year old split level home in a nice town in Central NJ needs a lot of work...new gutters, probably a roof, inside paint, rugs, front stoop, and the list goes on. We refinanced a few months ago and the house was valued at around $350,000. A nice Colonial across the street built around 6 years ago was sold for over $600,000 just last summer. Spouse and I both in our 50s making professional incomes. We have a son and daughter in college right now, one a freshman and one graduating this year. We also have a lovely retarded 18 year who will be with us for a while.

Question:
What are the economic plusses/minusses of razing our house and building a new one like the one across the street? We would expect to sell the house and move 10 years from now.

Thanks,
Steve

Comments for this entry have been closed
Return to top