May 26, 2009
House Prices Continue to Slide
House prices continued to tumble in March, according to the Case-Shiller index. Time to see what the futures say (keeping in mind the forecasting capacity of the Case Shiller futures are not well known).
Figure 1: Case-Shiller 10 city price index, (blue line), CME futures prices, 26 May 2009 (red triangle), and CME futures prices, 21 Sep 2008 (green diamond). NBER-defined recessions shaded gray, and start date dashed gray line Source: Standard and Poors' [xls], ino.com, St. Louis FRED II, NBER, and author's calculations.
So as of March, the 10 city is 40% lower than its May 2006 peak, in log terms (37%, in percent terms). The CME futures indicate that the 10 city index will be 52.7% lower by May 2010 (41% in percent terms). Compared to last September, the trough has moved up (the trough back then was slated to be in May 2011). However, one doesn't want to make too much of these long horizon indications, since the futures prices for these dates (November 2011 onward) have not changed since, for instance, the February 25 futures (shown in this post).
Of course, futures prices incorporate both expectations and risk preferences. In addition, these markets -- particularly at longer horizons -- are not likely to be particularly thick, so one has to be careful about using these prices as forecasts, even if one were to assume risk neutral agents. For an alternative, one can use the forecasts of the old OFHEO indices and convert to implied Case-Shiller (see this post for example).
More commentary at Calculated Risk.
[Update: 10am Pacific 5/27/09]
Here's a graph of CPI deflated Case-Shller index. The decline has been more pronounced in real terms.
Figure 2: Log Case-Shiller 10 city price index, (blue line), log CME futures prices, 26 May 2009 (dark blue triangle), and log CPI-all deflated Case-Shiller 10 city price index (thick red line). NBER-defined recessions shaded gray, and start date dashed gray line Source: Standard and Poors' [xls], ino.com, St. Louis FRED II, NBER, and author's calculations.
Posted by Menzie Chinn at May 26, 2009 08:53 AMdigg this | reddit
C/S bottoms at 130 in May 2010? That's plausible. A lot depends on the course of inflation.
Posted by: Steve Kopits at May 26, 2009 03:33 PM
Beyond short-term shocks to confidence, this isn't a bad thing.
Houses are finally cheap for first-time buyers.
Plus, if you want to prop up housing prices, the government merely has to issue greencards to about 1 million highly-educated immigrants who are already in the US, but on H-1 or student visas. They already live here, and either have jobs here or are in graduate school here.
Give them 1 million greencards, and they will settle here, and many of them will buy homes. They have a higher income than the national average, as most are engineers, doctors, MBAs, or PhD scientists.
Posted by: GK at May 26, 2009 04:51 PM
It is important to note that the majority of the losses are concentrated in just 4 states : CA, NV, AZ, FL.
The 80% of the US population that does not live in these states, has not seen that much of a fall. Texas, for example.
Posted by: GK at May 26, 2009 04:56 PM
Really? Let's do some math. How can we make a metric out of "not much of a fall?" Let's be very, very conservative and define it as "within two standard deviations of the mean." Case-Shiller seasonally adjusted is www2.standardandpoors.com/spf/pdf/index/SA_CSHomePrice_History_052619.xls. Calculate monthly returns. Calculate means and sd's. Subtract two of the latter from one of the former. Count how many of the last 24 months are more than two standard deviations below the mean.
The best of the twenty metro areas here come in at 4 months of such falls -- Dallas and Boston. Every other one has seen at least six. The median is 12. And that is measuring "not much of a fall" as "less than a two-standard deviation drop." Make the metric one sd -- which is way closer to your assertion -- and only Dallas is not looking at 12+ months out of 24 of steep drops. And even Dallas is at 9.
And Dallas, last I heard, is not 80% of the US. Except, maybe, if you're George W. Bush. Say -- do you think the other 20% live in Midland?
Posted by: wcw at May 26, 2009 05:40 PM
Exclude those 4 states, and the rest had a far more modest percentage fall, like under 10%. Period.
I see you clumsily avoided talking in percentages of price declines, instead clinging to the false metric of 'number of months of declines'.
Ah, yes, Bush Derangement Syndrome. You couldn't keep restraint for more than 2 paragraphs, could you? No wonder you are incoherent.
So you are a bad investor, and lost your pants. Deal with it.
Posted by: GK at May 26, 2009 08:34 PM
GK, making fun of irrational Texan exceptionalism is good clean fun, period. W. Bush is just the cherry on top.
Yes, if you define a 10% drop off peak in the housing market as small, then most markets have only seen a small fall. Thing is, a 10% drop in residential housing is not small, it is huge. Seven of nine census divisions, including the West South Central containing Texas, were showing year-over-year declines of greater than 2 sd below the mean at 08Q4.
The truth is that 80% of the country's regions have their housing markets getting crushed.
As for my investing, I am not proud of my twelve months through April. Still, I don't think -5% qualifies as losing my pants. After all, I did beat world equity markets by roughly 35%.
I find knowing the facts helps.
Posted by: wcw at May 26, 2009 10:55 PM
I find knowing the facts helps.
Then why haven't you put in any efforts to learn facts?
80% of the US population (those living outside of CA, NV, AZ, FL), have seen an average fall of less than 10%. That is not huge at all.
There is no simpler way to put it than that.
Posted by: GK at May 27, 2009 09:04 AM
GK: I will make two observations. First, the real (CPI deflated) decline from peak has been larger than the nominal, by about 5 percentage points (log terms). Second, while I have focused in the post on measuring relative to peak, for thinking about economic behavior one wants to focus on how the trajectory of house-stock-based wealth has shifted. In that context, first derivatives of expected housing values matter as much as the level for consumption behavior.
Posted by: Menzie Chinn at May 27, 2009 09:55 AM
an average fall of less than 10%... is not huge at all.
gk, in an unlevered investment this would be true. but no property held by american households is more levered than the house.
i live in suburban chicago -- a place like many midwestern cities very far from the big four bubble states and yet still very affected by declining house prices -- so i'll take that as archetype. the house i rent was bought for 460 in 2005. asking in my neighborhood is now 370, closing (if you can sell) even less -- a decline of about 20%. but the loss to the owner is not 20%; it is 100% of his downpayment.
this leverage is an extremely serious issue in a nation with $15tn in mortgage debt outstanding on what was at the peak $24tn in housing stock. recall too that one-third of homes are unmortgaged; this means $15tn is secured by something more like $18-19tn of housing stock as considered at peak bubble valuations.
that is to say it's very likely that the entire aggregate equity position of american households with mortgaged homes will be wiped out in this crisis (in fact has effectively already been), and furthermore just possible that the peak-to-trough move will wipe out most of the net aggregate housing equity in the united states.
even if we wishfully ignore the major population centers which are more dramatically affected in favor of the scattered few plainsmen denizens of rural nebraska -- and i cannot imagine why we would bother, as we are all one nation with one wildly overlevered private sector balance sheet, one damaged banking system and one collapsed securitization engine of capital importation -- there is no way to credibly minimize this event, and we shouldn't try to. it is a epochal household balance sheet shock, and its effects are being felt as resoundingly in the flyover sticks as elsewhere. my family, for example, is in part a farm family -- is anyone aware that arable land prices are collapsing? the recent chicago fed's ag letter severely underestimates the wreckage we're seeing anecdotally. for many of the same basic reasons as are at work in residential and commercial property, expect to see plenty of overborrowed farmers shut up shop and liquidate over the next year or two.
Posted by: gaius marius at May 27, 2009 08:34 PM
MC, thanks for redirecting our intrepid interlocutor. I don't know what else to tell him, except perhaps to ask when we last saw nominal 10% drawdowns nationwide in house prices.
gm, thanks also for reminding us that a 10% drop in a levered market has pernicious effects. Check the latest flow of funds report to turn your anecdote into data: owners equity as percentage of household real estate at 08Q4 was down to 43%. It had been 58.5% in 2005. Ouch.
BK, try engaging Menzie or gaius, please.
Posted by: wcw at May 27, 2009 10:00 PM
Speaking of the Chicago suburbs, in 2006 the number of home sales recorded for Arlington Heights as of April 22 was 359; in 2009 it is 115. In 2006 home sales at or above $600k totaled 109; in 2009 the number as of April 22 was 5 (yes, five). As the number of homes listed on the MLS at $600k or above is 82 (87 if you include those at or above $599k), one might characterize the market as being a bit slow.
Posted by: jm at May 27, 2009 11:50 PM
Case-Shiller cities outside of CA, AZ, FL, NV with declines greater than -20%:
That brings the total to 14 of the most heavily populated states. I know places like Des Moines IA and Omaha NE are experiencing a housing bust as well. It is just in a different form. In those places prices never skyrocketed because you could just build new. So their housing bust damages their economy more fundamentally because they became overly reliant on new home construction.
Posted by: MinniRenter at May 28, 2009 06:37 AM
I know places like Des Moines IA and Omaha NE are experiencing a housing bust as well.-MinniRenter
I'm in Madison, WI, another city under the Case-Shiller radar. We have a lot of empty homes, esp condos, for sale. I know of one ghost neighborhood of oddly empty condos. There are probably more.
It feels like we are where the coasts were a few years ago. Rents are declining and now house prices are too, but slowly.
I agree with MinniRenter's point. There are many people affected who don't live in those four states. Many of them don't even live in a city covered by an index.
Posted by: Charles Gervasi at May 28, 2009 10:59 AM
jm, where do you get local data? MLS?
i'm mostly working from anecdote in my neighborhood, but the volume slowdown is really thick now and prices have come unanchored in my middle-to-upper-middle-class tier.
and then we get this dislocation in the bond market to boot. amazing.
Posted by: gaius marius at May 28, 2009 06:54 PM
With 10% unemployment looking how can you even begin to talk of a bottom?
The bubble was driven by people selling their equity to buy bigger and better. That supports a normal market but at a more realistic pace.
With so much negative equity that's not going to happen for years.
Unless of course Obama and crew restore lost equity as a new part of the stimulies program.
Posted by: Anonymous at May 29, 2009 12:19 PM
Things are going to get rougher...for those who lost their jobs(and homes), the first group is about to loose their unemployment benefits. When they do, they cannot reclaim again until they post earnings over a 4 month quarter...since they cannot find jobs....no more unemployment either...so who is suppose to buy all these houses....Of those who can find jobs, you hear from the GM worker who makes $20 some dollars an hour...do I have to take a job that pays $7-$10 per hour....can those people buy a median priced home in our area which currently goes for $215,000...I think not....It is time to figure out how to get people back to work instead of trying to get them to continue with mortgages they simply cannot afford and to let the price of houses meet the wage of the market...
Posted by: Jerry at May 30, 2009 10:05 AM
The prices are going to continue to slide for a while. :(. Forbes predicted that the market won't return across the country until Q2 2010.
Posted by: Alethea SEO at May 30, 2009 10:46 AM
I heard a number that 37 billion dollars of foreclosed homes on the market. Why doesn't the fed buy those homes and take them off the market?
Use them for low income property dwelling, or demolish them, taking them off the market?
Posted by: Darryl at June 16, 2009 11:15 AM