Arnold Kling  

Home Prices and Rents

PRINT
Discrimination in Academia: Le... Switch to Switchgrass?...

Fortune Magazine writes,


From 2000 to 2007 the nationwide P/R jumped from 15 to 24, an increase of 60%. The figure went from 12 to 21 in Tampa, 11 to 26 in Washington, D.C., and 28 to 51 in California's East Bay, an area that includes Oakland and the area east of the city.

The historical norm for the price-to-rent ratio is 15. But I have argued that it ought to be 25. My thinking is that if the real, after-tax, after-depreciation interest rate is 4 percent, then the P/R ratio is the inverse of that, or 25.

If you agree that the P/R is headed back to 15, then the article is correct to forecast major adjustments, with rents rising and prices falling. If you think that P/R can stay above 20, then the adjustment will be relatively minor.

Thanks to reader Prashant Kothari for the pointer.

UPDATE: lots of comments suggest that people aren't following my rent-buy calculations.

The basic formula is that you buy whenever

rental rate + appreciation - interest cost

is greater than zero. Otherwise, you rent.

What about taxes and maintenance? You can adjust "appreciation" for property taxes and maintenance, or you can put them in separately:

rental rate + appreciation - tax rate - maintenance - interest cost

The rental rate that sets everything to zero is the rental rate that makes "rent vs. buy" neutral. For example, if the mortgage rate is 6 percent, rents and prices on as-good-as-new residences go up 4 percent per year, but maintenance and taxes are 1 percent each, you have
0 = rental rate + 4 - 1 - 1 - 6

so that the rental rate is 4 percent.

If you think prices are only going up 2 percent as a long-term average, then the rental rate is 6 percent. But that means you are assuming that even if you maintain your house in as-new condition, it will only appreciate at 2 percent per year.


Comments and Sharing





COMMENTS (11 to date)
Jeff Hallman writes:

Either the market prior to the recent runup in house prices has been wrong for decades, or you are missing something. Are you a homeowner? I have been for about 18 years now, and I've found that maintenance (roofing, paint, HVAC repairs, broken water pipes, etc) runs at least 1 percent per year, and real estate taxes another 1 percent. To get a real return of 4%, then, annual rent has to be at least 6% of the house price. That would give a P/R ratio of 16.7, not far from the 15 you cite as the historical average.

Arnold Kling writes:

Jeff, I don't think you can figure the rate of return you need without saying something about the mortgage rate (after taxes), minus general inflation.

ErikR writes:

Why must the mortgage rate enter into the calculation? Since not all property for rent is mortgaged, rates on mortgages would seem a red herring.

As for inflation, rents would normally keep pace with inflation, so I don't see why that is an issue either.

I think Jeff's simple estimate is pretty good (assuming people do require a 4% return)

Dan Weber writes:
Why must the mortgage rate enter into the calculation? Since not all property for rent is mortgaged, rates on mortgages would seem a red herring.
It's the cost of capital. Keeping $300K tied up in a property has a significant opportunity cost.

Or just imagine that the property were to be sold to someone else who needs to finance it.

Maniakes writes:

Mortgage rates matter for the price-to-rent ratio because of cost of capital.

If you are buying a house to rent it out, then either you need enough income to service the mortgage, or you need enough cash to buy the house outright. If you buy it for cash, and the rental income would not be enough to service the mortgage, then you'd get a better rate of return on your cash by investing in mortgage-backed securities (or stocks or bonds, which in an efficient market would have the same risk-adjusted rate of return as mortgage-backed securities).

The same logic applies to primary residences. If you'd need a mortgage, you have a choice of making mortgage payments or paying rent. If you're paying cash, you're income on the value of your house is the rent you're not paying, which competes with the income you could get by selling your house and investing the procedes.

Lord writes:

Location, location, location. In large metro areas, high P/Rs are easily justified, in the countryside they are not. Why are things different today? Burgeoning congestion. Once congested rents rise faster than incomes, not with inflation.

John Thacker writes:

In large metro areas, high P/Rs are easily justified, in the countryside they are not. Why are things different today? Burgeoning congestion. Once congested rents rise faster than incomes, not with inflation.

Huh?
Higher prices are certainly somewhat justified (and also very much driven by restrictive zoning), but why exactly do think that higher price to rent ratios are easily justified? Many of the location factors that raise prices also raise rents in an attractive area.

And why are you arguing that high P/R ratios are "easily justified" in congested areas but simultaneously claiming that rents rise faster in those areas? I don't understand.

Dan Weber writes:

Maybe Lord is claiming that a higher P/R ratio is justified because R is expected to rise in the future. Much like a stock in which earnings are low today but expected to increase significantly tomorrow.

(I'm not justifying his theory, just trying to explain it.)

Floccina writes:

I think that p/r stays down blow 25 because the incentives are better to take care of a property you live in, if you own it.

Jeff Hallman writes:

Arnold, you're the one who posited a 4% real rate, not me. I simply pointed out that the cost of capital is not the only expense a landlord faces. If you add in those costs as well as the 4% real cost of capital, you get my numbers. What's so complicated about it?

Michael Sullivan writes:

The 4% real rate represents the cost of capital. If the cost of capital depends on mortgage rates, or any other rates, then those rates (or some function of them) should be substituted for the 4%.

Mortgage rates are only indicators of capital cost. If they are low enough to be greatly preferred to other ways of raising capital, that will affect the capital cost. Maybe it is (was) below 4% due to low mortgage rates?

The point Jeff H makes is that AK's ratio of 25 completely ignores every cost that would be borne by homeowners or landlords but not by renters *except* capital cost (for which he uses 4% as a rough guide).

I don't quibble with the 4%. It may not be perfect, but it's only there for the sake of argument. I, like Jeff, quibble with the idea that cost of capital is the only cost to consider. Homeowners pay real estate taxes and major maintenance costs that renters do not. That difference should be, and is, reflected in the price. If I could rent for no more than what I'd pay in interest for a mortgage, why would I ever sign up for taxes and maintenance?

Jeff Hallman's simple little calculation is just as detailed as AK's original calculation.

I'm thinking that AK needs to talk to some landlords.

OTOH, the general model is reasonable, and if returns on capital generally can be expected to be 3% rather than 4% going forward, that would imply a sustainable jump in the P/R.

Arnold, AFAICT, this is a serious error. The fact that it's been mentioned to you more than once in comments and you still persist in pulling this 25 P/R ratio out without commenting on these issues kinda gives me the willies about trusting you for economic information.

Comments for this entry have been closed
Return to top