Seattle Bubble

Yesterday’s P-I had a big front-page story from Aubrey Cohen that seems to be at least partly designed to frighten renters and encourage landlords: Rent at an all-time high — if you can find a place

“There are a lot of apartments available, but the desirable ones seem to rent very quickly,” said Kilbourne, a University of Washington student who must move out of his dormitory by mid-June and is about to start graduate school.

“Seems like you have to find a place within the first day of it becoming available, or it’s gone.”

A new report affirms that apartments are about as hard to find now as they have been any time in the past three decades and rents are on the rise.
…
“Even though rents have increased significantly over the past couple of years, there is still room for more increases because rents have not kept pace with consumer incomes,” Dupre + Scott wrote. “Rents today are still 10 percent to 15 percent below what renters can afford to pay.”

My main problem with these studies released by Dupre+Scott is that they only look at rental properties with 20 or more units. This totally overlooks the market that consists of individual homes and condos being rented out by individuals (as opposed to businesses or institutions). Also worth noting is that as with most real estate articles in the P-I, the graphs and most of the numbers in the article focus on Seattle proper, where rents are naturally higher and vacancy rates lower.

Aubrey does spend some time talking about what the report doesn’t tell us, and exploring why things aren’t shaping up as badly for renters as some were predicting a year ago. I definitely give him credit for not making the piece as one-sided as the headline writer made it out to be.

“Vacancies weren’t lower very often in the past 27 years,” says the report, by Dupre + Scott Apartment Advisors, a Seattle company that tracks the rental market. “Even so, we expected vacancies would be lower by now.”

In March, 4.1 percent of King County apartments and 3.1 percent of those in Seattle were vacant, according to Dupre + Scott.

That’s much tighter than the rates of a few years ago, but up from March 2007, when vacancies were at 3.9 percent in the county and 2.8 percent in Seattle.

One big reason vacancies have started increasing is that the housing slump has all but ended conversion of apartments into condominiums.
…
Inventory also has grown as homeowners rent out homes they have been unable to sell, said Dean Foggitt, a broker at Brink Property Management, a Bellevue company that manages about 600 rentals in and around King County. He said his portfolio of rental houses is up 15 percent to 20 percent from a year ago.

“As far as tenant demand, we haven’t seen that huge increase that we thought we would have, given the slowdown in the sales market,” he said. “What we’ve seen more is people staying put, less tenants giving notice.”

So, as the housing market stalls, rental inventory is actually increasing, which is preventing rents from rising as quickly as expected. That sounds awfully familiar… maybe because it’s exactly what we have been predicting would happen, at least as far back as 2006:

As flippers become unable to sell, and 100%-financed families find themselves unable to afford their homes, it would seem that individual units are likely to come onto the rental market in greater numbers.

King County vacancy rates are at 4.1% (according to the report), which isn’t a super-tight market, but does favor landlords slightly (5% is considered a “balanced” rental market). What I think is also interesting is that vacancy rates have been slowly increasing since late last year. In October the county-wide vacancy was at 3.8%. Even Dupre + Scott is predicting that the vacancy rate will continue rising, up to over 5% by 2010.

Unfortunately, I don’t have enough money burning a hole in my pocket that I’m willing to spend the $130 for a Dupre + Scott subscription, so I don’t have access to the full historical data on local apartment vacancies. However, I was able to find this pdf from the Seattle Times, which shows the vacancy rates for King, Snohomish, Pierce, and Kitsap back through 1997. Here’s a reproduction of the vacancy rate graph (with 2008 data added for King and Snohomish):

As you can see, King County has had lower vacancy rates in six of the last twelve years. Only from 2002 to 2005 were vacancy rates significantly higher than they are now. I’d hardly describe the current rental market as a panic frenzy to find a place that you can barely afford, which is what the P-I headline makes it out to be.

I’d also like to point out Aubrey’s response to some criticism he received for the article over on his blog:

I knew it would happen. I arrived at work this morning to an e-mail from a reader angry about my story on rising apartment rents.

“Gee thanks! I wonder how much more money I will be paying for rent this year because of this article,” he wrote. “I’m sure you’ve made many landlords very happy today.”

I’ve gotten similar responses to previous stories about rising rents in the Seattle area. Some renters have told me landlords included copies of my stories with notices of rent increases.

I know that many real estate professionals blame the media for much of the current sideline sitting among potential buyers.

I hope that stories do have some sort of impact. After all, what’s the point of journalism otherwise, right?

That said, I’m not after any particular outcome. I try to put the best information out there and let readers decide what to do with it.

He has some additional thoughts on the subject, and is soliciting feedback from his readers, so head over there and leave a comment if you’ve got strong feelings on that particular subject.

(Aubrey Cohen, Seattle P-I, 04.21.2008)

One of the arguments often discussed with respect to the housing bubble is the fact that the ratio of prices to rents has been fairly consistent on a historical basis, but that this ratio has been blown out in the past few years as home prices have shot up. Indeed, this can be seen in the chart below, which compares King County median home prices to annual rents for a typical two-bedroom apartment for the past ~20 years. In this chart, you can see that home prices have typically hovered in the range of 20 to 25 times rent. But since 2001 this ratio has steadily climbed to the point where it stood at 38 times rent at the end of 2006.

The strong historic relationship between rents and home prices is also validated in a recent academic paper on historic price to rent ratios, which shows that on a national basis, the same pattern has existed – with the price:rent ratio averaging about 20x - over the past ~45 years.

More recently, there have been several articles in the local press discussing the rate of rent increases. According to Dupre + Scott, Seattle rents have risen 8.6% in the past year – versus only 2.8% in total for the first six years of the century. Many who are bullish on local housing have commented on this as proof positive that it will be an increase in rents, not a decrease in home values that will bring the relationship back to its historic balance. The argument is basically that rents fallen behind incomes since 2001 - when the local economy slowed and the many people moved away after the dot-com boom. There is some merit to this. Rents have indeed been flat since the start of the new century. On the flip side, home prices have increased so rapidly during the same period it is hard to believe low rents are all of the cause.

So which is it? Rents too low? Home prices too high? Or some combination of both?

Recently, I came across a great data set from Conway Pedersen Economics, Inc, a local economics consulting firm - that provided enough history that I thought it would be possible to run some comparisons on rent increases vs. home prices using local data – and possibly shed further light on what is going on and where we are headed.

My basic premise is that rents and housing prices should track to each other at a fairly consistent ratio over time, but also and more importantly, that the rate of increase in both of these is governed by increases in income over time (as discussed here and here by Tim and myself, and in the aforementioned articles).

Using Conway-Pedersen’s annual data for housing costs and income back to 1985 – we can see that both rents and home prices have indeed closely tracked incomes over this time period. The correlation between the data series is quite strong.

• Rents and income - as represented by the blue diamonds – have a nearly linear relationship. The correlation between these two time series is about 99%. During the time period for which I have data, incomes rose at a 4.8% CAGR, and rents rose at a 4.2% CAGR. You can see that at the end of the period, the current actual rent level appears to be below the trend line. Using a simple linear regression shows that rents were about 4% under where one might expect them to be based on income growth. But with an 8.6% increase in 2007 (the data only goes to 2006) it’s likely that the same analysis today would show them to be spot on.
• The story for home prices and income – as represented by the red squares - is similar, but only through 2002. From 1985 to 2001, the time series for income and home prices were also about 99% correlated. As a matter of fact, you can go back to 1970 and find a 99.5% correlation. But in 2002, the two series diverged. Adding the last five years of data (the green triangles) drops the correlation to 97%, as home prices clearly move far above the 1985-2002 trend line. Using the function derived from a linear regression of 1985 to 2001 data to predict where prices should be relative to incomes shows that home prices at the end of 2006 were about 34% above what the long-term relationship would indicate.

So what is one to take away from this? Is it definitive? Hardly. It has all the usual caveats about sample size, my limited grasp of statistics, etc. But it is interesting that we do appear to have had a clear divergence from our long-term “steady state” relationship between housing costs for both rents and home prices.

My thoughts based on this data:

• Given rent increases in the past year, is likely that rents are now back in line with income levels. Rent increases of ~5% per year are probably to be expected - vs. 2.8% for the first six years of this century. Expecting rent increases that vastly exceed income growth is probably wishful thinking (or paranoia, depending on whether you are writing or receiving checks) as rents have only been greater than +/- 5% of the income-predicted trend line for 2 of 21 years, and then only to the low side (under 8.3% in 1985 and by 6.2% in 1986).
• The bulk of the diversion from the historical mean in price:rent ratio has been driven by home prices, all of which has occurred in the past 5 years.
• If rent increases do indeed continue to track incomes at about 5% per year, it will take until 2015 to get back to a price:rent ratio of 25x if home prices just stay flat!

My best guess: we meet in the middle and are back in synch by 2011-12 – which would imply a 12-16% decrease in nominal home prices over the next three or four years.

What do you think? Comment away!

I should at least mention the local news reports about the latest Dupre + Scott rental survey. Here’s the meat of the report, courtesy of Aubrey Cohen at the P-I:

The King County vacancy rate hit 3.8 percent this month, down from 3.9 percent in March and 4.2 percent in September 2006, according to Dupre + Scott Apartment Advisors, a Seattle company that tracks the rental market.

“That’s not the lowest rate we’ve ever seen, but vacancies weren’t lower very often in the past 20 years,” report co-author Patty Dupre said in a statement accompanying the report.

The average apartment rent in the county is $1,001, up 6 percent from the spring and 9 percent from last fall.

Elizabeth Rhodes also had an article on the report, but it’s not even worth quoting, as the entire thing is nothing more than the most extreme cherry-picked statistics all strung together.

As I said when last quarter’s report came out, the current trend in rising rents is entirely predictable and to be expected. As home ownership took off, rents have been artificially low. One would expect them to correct up as the ability to commit financial suicide diminishes, thus keeping people renting and sending never-should-have-been home “owners” back into the rental market.

It’s possible that with the current turmoil going on in housing, rents will slightly over-correct, but in general I still expect the 8-9% YOY increases to taper off to a more sustainable 3-5% by this time next year. As I’ve stated before, rent increases are necessarily tied to incomes (i.e. - you can’t use exotic financing to pay your rent), and therefore while they can certainly rise more slowly than incomes, they can’t rapidly rise out of control like home prices have.

It should also be kept in mind that the Dupre + Scott report polls only apartment complexes of 20 units or more, and therefore does not account for the ever-increasing market of rentals created by condo and house “investors” (or “owners” that must move) who are unable to sell in the down market, and instead resort to renting out their unit to cover at least part of the mortgage.

For the first time that I’m aware of in the local media, Mr. Cohen actually addresses this at the end of his article:

[Epic Asset Management VPO Bill] Austin speculated that problems in the mortgage market would force more people back into rentals in coming months, but said that could be balanced by condominiums coming back on the market as rentals, a slowdown in condo conversion and change of some planned new buildings from condos to apartments.

Speaking of condos converting to apartments, it looks like the “repartment” trend that’s sweeping the nation has finally come to the Seattle area:

A reader brought this to my attention. At first I was a little skeptical but have since confirmed that the Max Condo in Greenwood has pulled the plug and is reverting back to apartments.

It seems sales during the first three weeks did not meet expectations. Which is a bit surprising as Greenwood area condos sell very well.

Thanks to Lumpeninvestor for pointing that one out in the comments.

(Aubrey Cohen, Seattle P-I, 09.27.2007)
(Elizabeth Rhodes, Seattle Times, 09.29.2007)
(Ben_Kakimoto, Seattle Condos and Lofts, 10.01.2007)

Finally some semi-useful information out of Dupre + Scott. Mr. Cohen reports on the “1-19 Unit Report” just released by Dupre + Scott in this terrifying tale of skyrocketing rents and plummeting vacancies:

Rental houses get scarce, expensive

The good news for people who cannot afford to buy a house in Seattle is that it may make economic sense to rent for the moment.

The bad news? Rental houses are a lot harder to find and pricier than they were a year ago.

The typical Seattle rental house now costs $1,604 a month, up 4.6 percent from a year ago, according to data that Dupre + Scott Apartment Advisors released Monday. House rents for all of King and Snohomish counties were up 6.5 percent from a year ago.

Dupre + Scott did not release vacancy rates just for houses, but among all King County rental buildings with one to 19 units, including houses, the rate declined from 3.7 percent a year ago to 3.1 percent now. The house trend is similar to that for apartments in general, according to Dupre + Scott.

It should be noted that even this report excludes homes or condos rented out by individuals. According to the Dupre + Scott website, the 1-19 Unit Report is derived from a survey of “apartment owners, professional property managers, and on-site property managers.” Given the logistical complexity of such a task, you can’t really blame them for not surveying individual owners, but it’s important to keep in mind what this data is and is not telling us.

Dupre + Scott’s April apartment report asserts that people would make more money renting, and investing the money they save elsewhere, than paying current prices to buy a home.

“There are a lot of reasons to own a home,” Dupre + Scott co-owner Mike Scott said. “From a purely financial perspective I’d say no, it doesn’t make sense. But if home prices go up 10 or 15 percent in the next year, I’ll have to eat those words.”

That makes about as much sense as the following: “There are a lot of reasons to visit a casino. From a purely financial perspective I’d say no, it doesn’t make sense. But if you make 10 to 15 percent profit on your next visit, I’ll have to eat those words.”

No words will in fact need to be eaten, because even if home prices go up another 10 to 15 percent next year, it still doesn’t make financial sense to buy a home right now. To put it another way… Just because a reckless decision paid off in the short term doesn’t mean that it wasn’t reckless, it just means that you were lucky.

Bob Melvey, assistant manager of Windermere Real Estate’s Ballard office, said renting and investing the savings might pay off “if someone is very, very disciplined and they truly do put that difference in the stock market and they do a good job of managing their stock portfolio.”

But most people would end up spending the savings on other things, Melvey said. “Owning your own home is, to a degree, a forced savings plan.”

A “forced savings plan” where 80% of the money you “deposit” in the first five years simply vanishes (the interest portion of the payment), and from which you can only withdraw money by paying a 6-9% fee (not on the amount you have “saved” mind you, but on the total sale price) and relocating. Wait, that doesn’t sound anything like a savings plan.

To give these figures a bit of perspective, let’s say that the average single-family home at $1,583 (King/Sno) is comparable to the median home, which sold for $454,950 last month (King), resulting in an approximate PITI payment of $3,400. We will assume the homebuyer had 20% (almost $100,000) to put down, and we will ignore maintenance, HOA, utilities, and tax deductions in order to keep this simple. Right off the bat, the renter’s monthly payment just 46.6% of the buyer.

It would take 13 years of 6.5% rent hikes before the renter’s payment exceeded the homeowner’s. Of course, during that time, the renter’s $90,000 in liquid investments will likely have doubled or tripled, and that doesn’t even consider that they’re adding the monthly savings to the investment. If they bank the difference, their $90,000 could quite easily become $500-$600k. Also, if you think that rents will increase steadily at 6.5% per year for the next 13 years, I don’t think you have a very good grasp of history.

Suffice it to say that despite the scary language in the headline of Mr. Cohen’s latest volley, renting still beats buying in the Seattle area hands down (financially speaking).

(Aubrey Cohen, Seattle P-I, 04.30.2007)