In his column Paul Krugman discusses the reasons he is worried that recent efforts by the Federal Reserve to stabilize financial markets are falling short. Of particular interest is the following quote:
"First, we had an enormous housing bubble in the middle of this decade. To restore a historically normal ratio of housing prices to rents or incomes, average home prices would have to fall about 30 percent from their current levels."
The ratio of housing prices to rents is a way of trying to sort out whether housing prices are overvalued. What you’re doing is comparing prices to an estimate of the rents a homeowner could earn if instead of living in their home they rented it out. It’s calculated in the same spirit as a price-earnings ratio on a stock. The higher the ratio is relative to its long run average the more you worry that housing prices are unsustainably high. We can calculate a version of the housing price to rent ratio for the Philadelphia Metropolitan Division (PA Portion).
The ratio has hovered around 1.61 since the 2nd quarter 2006 a historic peak. Over the whole period from 1983 to 2007 the ratio averaged 1.17. The Krugman quote above which argues that national housing prices might have to fall by as much as 30% is based on the assumption that the national housing price to rent ratio will fall back to its own long run average.
To think this through for the Philadelphia Metropolitan Division it might be good to start with the past. The last peak in the housing price to rent ratio in Philadelphia was in the 2nd quarter of 1988 when it reached 1.28. So what happened after that?
Housing prices in the region began to grow more slowly while rents kept growing briskly at about 6 percent a year, so the ratio fell back in the neighborhood of the long run average of 1.17 in the first half of 1991 (as you can see in the figure above it would actually keep falling until 1997). This meant that housing prices between 1988 and 1991 still managed to grow by 8% while rents grew by 20%. Consumer prices excluding shelter grew by 15% over this period.
Where We Are Now
So the ratio now stands at 1.61. Let’s assume that the housing price to rent ratio in Philadelphia gets back to its long run average of 1.17 just as quickly as it did in the early 90s, which would be by the end of 2010.
Let’s assume that rents will grow in the region by 6% a year (they are actually growing by about 4.5% recently).
Making all of those assumptions housing prices in the region have to fall by 13% and that is before factoring in inflation, add that in at of 3% a year and that means housing prices fall by 21% by 2010.
Eeek!
All of this is assuming the past is a predictor of the future which is often not the case. And perhaps most important who says 1.17 is the normal level of the ratio? Maybe the relationship between rents and housing prices in the Philadelphia metro have permanently changed? Maybe housing prices will grow very slowly over a longer period of time? There are also some new wildcards which make the current situation different than in the past like the prevalence of subprime loans and the related risk of increased foreclosure rates.
Given the state of mortgage lending and the clear evidence that housing prices are growing more slowly in the region, it seems like people ought to start thinking about how these trends are going to impact the City, its quality of life and its budget.
For starters it’s a good time to pay more attention to the good people over at the Philadelphia Unemployment Project!
--Mark Price
Notes:
Housing price data is for the Philadelphia Metropolitan Division which includes only the Pennsylvania counties of Bucks, Chester, Delaware, Montgomery, and Philadelphia. Data on rents is for the Philadelphia-Wilmington-Atlantic City, PA-NJ-DE-MD Metropolitan area. It is unknown what the difference in geographic coverage between the measure of housing prices and the measure of rents makes in the pattern of growth in the housing price to rent ratio.
Data on rents is the Consumer Price Index-Owners' Equivalent Rent estimated by the Bureau of Labor Statistics.
Rubber meets...more analysis required.
If you are using housing prices to rents as comparable to the valuation of assets like shares of public company stocks, let's put some things out there...
From Finance 101, we know that we can dissect your "Price to Rent Ratio" like we would a lizard in high school biology into three pieces:
(1.) the expected growth of earnings from the asset in the future - in this case rents;
(2.) real interest rates - in this case real interest rates; and
(3.) the risk premium we demand to account for risk and other costs - in this case we could throw in our confidence in housing price increases, how quickly we think we could sell the house, etc.
So...(1.) + (2.) + (3.) = 3-story Manayunk row a steal @ $249,999.99?
Now (and you can just take my word for this, I've had way too many delicious Manayunk beers to go hunting for sources), changes in company valuations have statistically relatively little to do with changes in dividends (something most entrepreneurs know intrinsically, of course), so let's take that same logic and apply it to your analysis and say that (expected) changes in rents have little impact on changes in housing prices. Good idea? I think so too.
Using our dissection above, that would leave real interest rates and our risk premium to explain increases in your Housing Price to Rent Ratio. The general trend in real interest rates has been downward (particularly post-2000), which would, like a bond, put significant upward pressure on housing prices and thus impact your Ratio (explanation number one?). Much can be explained, therefore, in what will happen to housing prices by what happens to real interest rates in the future (and I'm no economist, but I would hesitate to agree with your 13% decrease in housing prices by 2010).
If you don't agree with my first explanation, you can turn your attention to component #3, which would be our risk premium. Based on the variances between the Philadelphia Metro and the U.S., holding interest rates constant across the country, there have been times (most notably, recently) where we've been much more upbeat than the nation as a whole about our housing market. In so far as there have been rational reasons for such exuberance (and our metro's fundamentals, at least overall, have been relatively strong), the impact of such feelings on lowering risk premiums and thus allowing for strong housing values is good and should continue.
Public Policy To Do's
Price says:
As much fun as feeling depressed about the impact of slower housing growth on our quality of life might be, I would first take steps to better understand how our region's housing prices have reacted to changes in interest rates, and perhaps ask what other variables have accounted for differences in the performance of our market against that of, say, Chicago or Detroit. It's not my understanding that such analysis has been completed.
With little control over interest rates, I would argue that our response "for starters" lies not within paying more attention to the Philadelphia Unemployment Project, which in terms of foreclosures represents a relatively small share of our metro housing market, but rather in focusing on item #3 of our equation: our expected risk premium associated with housing, the one variable where smart public policy can play a role in sustaining strong housing values.
What can we do to alleviate liquidity concerns regarding our housing market?
What can we do to lower prospective investors' notion of risk in entering our market? Being known as one of the most corrupt, murderous, and least business-friendly cities in the country probably doesn't help.
Can public policy lower transaction costs, such as our transfer tax, with respect to housing?
What impact will full value assessments (and subsequent millage adjustment) have?
Would continuing abatements indefinitely help?
Just some thoughts.
===================================
:( society.
Francis, you are so sweet.
Francis let me just begin by saying I have missed you, where have you been?
You do bring up some really interesting points so thank you.
You could have applied the same logic to the dot-com bubble. Prices kept going up relative to earnings and then suddenly the music stopped and investors had their Wile E. Coyote moment. The ratio doesn’t tell you when the music will stop it only tells you that you’re looking at trends that might be unsustainable.
I agree with you. There is no question that cheap and I would add easy access to credit played an important role in appreciation. But there is also a point when appreciation itself becomes an engine of further growth.
If I wasn’t clear I should be clear, it’s a guess based on the performance of the housing price to rent ratio in the past. The relationship between prices and rents in the region may have fundamentally changed meaning that prices will not fall at all. And it is also possible that ratio can get back to some more normal level by the slow appreciation of prices over a long period of time.
Ok, let’s call it animal spirits, and assuming I follow you, I would agree they are important.
Obviously for the people being foreclosed this is a series issue. In addition there is good evidence that foreclosures lower prices not just on the homes foreclosed but in the immediate neighborhood, so if you have more foreclosures than in the past then that puts downward pressure on prices and hurts neighborhoods in tangible ways which we might put under the rubric of “blight”. PUP has long been worried about foreclosures but there is good reason to believe that this problem is going to be larger in the immediate future than it has been in the past.
The rest of your questions are interesting, I would add to that list the following:
Has access to credit in the region been reduced?
Not unrelated is credit relatively more expensive than in the past in the region?
Are there groups of buyers who were in the market before because of the presence of the subprime loans who are not in the market now and how big an impact will their absence now have?
Has the stock of homes for sale and their time on the market increased relative to the past?
Has new residential construction fallen off?
How many homeowners in the region might have negative equity if prices really do start to decline?
Interesting post, Price: a question
My immediate reaction was a bit similar to Francis' although not quite as erudite. You present interesting data, but there is a big piece missing: what could possibly be a causal link between home prices and rents, or could there be a third phenomenon that affects both home prices and rental rates similarly? It would seem that if anything, credit rates might be such a phenomenon. At least most pundits seem to explain credit rates as being the driving force behind changes in the housing market.
So, what new are you presenting here? Don't we already know that the unique credit rate circumstances that led to irresponsible lending and borrowing has changed - and has led to dropping home prices and increased foreclosures? Don't we already know that we can expect the bottom falling out of sub-prime lending to have negative ramifications extending well into the future?
Are you just presenting another way to predict how far housing prices might drop, as an alternative to trying to predict how long and how deep will be the impact of the sub-prime fallout based on historic changes in credit rates? Or, are there some more specific implications to what you're presenting - for example, increased support for programs aimed at providing assistance to low-income renters as opposed to low-income home owners?
always has to be a point to it all.
D.E. II, so typical of you, there has to be a point to it all.
The opportunity cost of living in your home is the rent you could earn if you rented it out to somebody else. If the growth in rents falls behind the growth in housing prices it could suggest that housing prices are overvalued. That is a big “could” - (I mean I'm not sure this is the case).
So I’m worrying out loud about the possibility of price declines. A very bearish reading of the trends in the ratio is that prices are overvalued and do for an actual downward correction. If that happens it heightens the need to worry about increased foreclosures and I think makes PUP’s current efforts in that department even more urgent.
Now it also may be that prices just grow very slowly over the coming years.
And whatever the outcome perhaps it would be a good time to start to think through the implications of slower growth in housing prices for the city? Does it mean fewer home sales and less revenue for the city?
Why buy a house?
There are all sorts of reasons why home ownership in the city of Philadelphia (I'll leave the metro part alone for a moment) has become more attractive in the past decade, even relative to renting (which has also become more attractive).
1) Speculation on future value, which whether by homeowners or investors is the principal cause of a property bubble;
2) Low interest rates;
3) Tax abatements;
4) Relative value -- i.e., Philadelphia looks more attractive compared to its neighbors than it did a decade ago, so it benefits from surplus demand;
5) Change in the overall economic picture, so that more money is chasing fewer goods;
6) Greater appeal to the specific market.
What do I mean by 5? Well, while the rental and purchase markets have some overlap -- plenty of people do decide to purchase rather than rent, because of 1-4, or vice versa, when 1-4 aren't sufficient or are negative -- for the most part, they're fairly distinct. Many, perhaps most, renters rent because they don't have a lot of capital, have lower income, worse credit, or are in impermanent employment or household arrangements. In other words, a lot of younger unmarrieds and poorer families.
Now circumstances can change for this group -- you might have an influx of hipsters or creative class types snatching up all the rental properties, or you might have a loss of the overall number of rental vacancies, which lets owners charge more in rent. But for the most part, they don't change that much.
If, on the other hand, you have a steep increase in middle-class jobs, in increase in the safety or perceived safety of middle-class neighborhoods, and in general a greater perception of prosperity and stability -- then you start to attract homeowners who are increasingly willing to spend more and more on housing purchases in those neighborhoods. Even people who would rent will now buy if their employment, household, and community feel more stable.
In other words, there are lots of other dynamics driving the housing market in Philadelphia than the ones that are affecting the city nationwide. The bottom has dropped out of the big bubbles, and the speculators have lost their lunch. The market in Philadelphia is a lot saner (that is, unless you're desparate to sell) than it was three years ago, and property values haven't fallen.
The best predictor of the future isn't the past but the present. Philadelphia's market has been corrected -- but some of the facts on the ground have fundamentally changed, and continue to change, for everyone.