How Much Damage Will Rising Rates Do To the Housing Recovery?

From Marketwatch:

Even a spike in mortgage costs since late spring — interest rates have jumped nearly a full percentage point — doesn’t appear to have stanched the flow of would-be buyers. It’s possible some buyers moved up their purchases to lock in attractive rates before the rose, analysts say

Housing has been one bright spot in the US economy over the last year, as sales are now in uptrends for both new and existing homes.  However, over the last few months we've seen a drop in bond prices (and a corresponding increase in bond yields).  Consider this chart of the TLT -- the long end of the Treasury market:

Prices have fallen through the 113/114 price level which was providing short term support.  Now prices are in the 106-110 level and appear to be finding a short-term bottom.  There's an uptick in momentum and a newly positive CMF implying at least a stabilization.  In addition, consider the yield on the 10 year is now fluctuating around 2.5%, which isn't a bad rate of return in a low inflation environment.  Put another way, I think we've seen the big bump up in rates that would be expected after the recent Fed announcement regarding tapering its bond buying program.

However, consider this scenario.  Interest rates are clearly rising.  Rate increases will be contained to a certain extent as purchasers such as foreign central banks re-balance their portfolios to take into account the rising rate scenario.  But make no mistake; rates are clearly moving higher.  At the same time, US unemployment is still 7.6%, meaning there is little to no upward wage pressure.  This explains why median incomes have been stagnant for the last 10 years.  At the same time, we've seen a recent spike in oil prices that are also eating into incomes.  And then there is the increase in the median price of new homes:

At some point the combination of rising rates, higher oil prices, stagnant incomes and higher home prices may start to slow home sales. 

Mortgage Rates Rise; Mortgage Bonds Drop

Mortgage backed bonds have dropped below key support at the 108 price level.  Other parts of the chart add to the bearish interpretation.  Prices are below the 100 day EMA with the smaller EMAs moving lower.  The 10 and 20 day EMA have moved through the 20 day EMA and momentum is negative.

The Financial Times outlines the the potential problems:

The change could affect the US economy in two ways: by making new loans less affordable, it could damp the recent recovery in house prices; and it could reduce the number of Americans refinancing into cheaper mortgages, eliminating savings that have boosted consumer spending. 

Keith Gumbinger, vice-president of mortgage research company HSH, said: “The rise in the rate is enough to slow down the refinancing market, as even a small bump up in rates can put homeowners on the fence.”
Refinancing applications had already begun to decrease, falling 12 per cent in the week to May 24 according to the Mortgage Bankers’ Association – the largest single-week drop this year and back to their lowest level since December 2012.

However, other analysis disagrees.  This is from Wonkblog, citing Goldman research:

It also shows an initial reason for some optimism. At a 30-year fixed-rate mortgage rate of about 3.8 percent, the typical American homebuyer can afford a $279,000 house. That’s 45 percent more than the current price of houses. That suggests that affordability isn’t the thing holding Americans back from buying houses (instead, it may be such factors as tight credit standards, difficulty building up a down payment  or lack of confidence in future job prospects). It also implies that slight increases in the mortgage rate shouldn’t completely undermine the improvement in the housing market; the thing to watch is not rates per se, but what happens on those other factors that are drags on would-be homeowners.

And that bodes particularly well:  As we wrote last week, the rise in rates over the past month appears to be driven primarily by improving economic prospects. If that’s the case, even as homes become a bit more expensive, they will be doing so at the same time those other restraining factors dissipate. So rising mortgage rates, if they’re rising for good reasons, could actually be net positives for the housing market if they result from more people having jobs and being confident in their prospects.

How affordable is housing?

- by New Deal democrat

There is a commenter who always asks me whether stabilization in housing prices is really a good thing if the median family income can't afford the median priced house. That's a fair point and I've been promising to look into the issue. So, what is the state of housing affordability? There is no one single answer, but the results may surprise you.

In determining if the median American family can afford the median house, there are at least two separate factors to consider: (1) how affordable is the typical down payment? and (2) how affordable is the mortgage? Let's look at each of them in succession.

The down payment, of course, is a percentage of the sale price of the house. So if, for example, we want to know how affordable a 20% down payment is, all we have to know is the sales price of the typical house, because then we can just divide it by five.

There are several ways of measuring historical house prices. For example, here is a graph from the blog jparsons.net, which measures house prices since 1970, first by using FHFA data, and then segueing to Case-Shiller data beginning in 1987. The blue line is nominal pricing, and the red line represents prices adjusted by the CPI:



Notice that even the red line is gradually increasing over time, probably reflecting increasing population vs. the fact that "God isn't making any more land."

Since housing prices themselves are a large part of the CPI, however ("owner's equivalent rent"), I think a better way to measure the price of housing for a typical family is to divide it by average hourly earnings, which is what the following graph does for both Case-Shiller series (the 10 city series began in 1987, and the 20 city series in 2000):



This shows that, using either series, the housing bubble has been entirely deflated. The 20 city series is lower than it was at its inception, the 10 city series slightly above.

But even that does not entirely answer our question. For that, let's look at the long-term graph comparing median family income from Nowandfutures with the sales price of a median house, which is what the following graph does from 1967 to mid 2011:



Note that the long term median price is 2.78 times median family income (I have seen averages of 2.7 elsewhere).

So where do we stand now? For that I need to do a little explaining of terms. "Median family income" is a subset of "median household income." The latter includes. e.g., households consisting of a single adult. "Median household income" reached its peak in real terms in 1999. At mid-year 2011, two researchers measured it at $49,909. In September, the Census Bureau estimated it at $49,445. The study by the two researchers mentioned above estimated that median household income had actually fallen by about 7% since the bottom of the recession in 2009.

"Median family income", by contrast, was $60,395 when last measured by the Census Bureau for 2010. 2.78 times that income is $167,900. 2.7 times that income is $164,100.

And what was the median sales price of a house in December 2011, the last month for which we have a measurement? According to the National Association of Realtors, it was $164,500. By contrast, at the peak of the bubble, the median sales price was $221,900. If "median family income" did not fall between 2010 and 2011, according to this measurement, right now we have just reached the point where the median family income can reaasonably purchase the median house sold.

Several caveats are in order. First of all, in 2009 median family income was $61,030. Like median household income, it fell between then and 2010, although by only 1.1%, At least a slight further decline between then and December 2011 probably did occur. Additionally, we know that the median asking price for homes nationwide, as estimated by Housing Tracker, is slightly over $200,000. This tells us that it is the lower-priced houses for sale which are actually being sold, while higher priced houses are sitting unsold. It is reasonable to ask if those prices are going to fall further. If they do, then we will "overshoot" the long term median average to the downside -- i.e., house prices will become even more affordable than the long term norm.

Next, let's turn to mortgage payments. Mortgage rates have fallen from 20% in 1980 to under 4% now. The below graph shows this from 1987:



Another way of looking at the same data is to consider affordability, i.e., how much house can I buy for the same mortgage payment. As this next graph shows, that price has more than doubled since 1987:



But house prices obviously haven't remained stagnant either. So this next graph factors in average house prices as measured by the Case-Shiller 10 city series. It tells us how affordable the mortgage payment for a typically priced house has been at the prevailing mortgage rate since 1987, and here the story looks quite different:



The mortgage payment it took to buy the typically priced house reached its least affordable at the peak of the housing bubble, where at a typical interest rate, the average mortgage would only purchase less than 40% of the house it could in 1987. Only in the last 12 months has it finally approached its 1987 level.

But once again, even that does not tell the whole story. For that, we need to ask how affordable was the typical mortgage payment for the typically priced house for a family earning the average hourly wage? That is what this next graph shows us:



The above graph tells us that a family earning the average hourly rate, at a typical mortgage interest rate, could afford only 70% of the typical mortgage for a typically priced house at the peak of the housing bubble compared with 1987. Despite lackluster wage growth, the renewed decrease in mortgage rates, together with the steep decline in housing prices, have made mortgage payments far more affordable than at any point in the last 25 years.

The same trend is shown in the National Association of Realtor's "housing affordability index," shown below:



So, there are at least two answers to the question "how affordable is housing?" The first answer is that, in terms of down payments, house prices are probably very slightly -- as in 5% or less -- above their long term norm. Since it is mainly lower priced housing that is selling, it is possible that higher priced housing will continue to decline in price and if so, there will be an overshoot to the downside (which is good for home buyers). The second answer is, that at least in terms of mortgage payments, now really is a good time to buy a house. We are probably at or near generational lows in affordable mortgage payments, particularly after factoring in family income.

This is the beginning of the end of the housing bust

- by New Deal democrat

A month ago when housing starts and permits were reported, I wrote: Psst: Is this the beginning of the end of the housing bust? With this morning's report of 681,000 permits issued, I believe we can affirmatively answer, YES.

As I said last month, housing construction is a long leading indicator, indeed along with interest rates probably the most important one. So those commentators who say that we won't get housing improvement until we have job improvement have causation exactly backwards. Rather, it is much more likely that we won't get more meaningful job improvement until we have more meaningful housing improvement. Further, the decline in housing starts and permits after the expiration of the $8000 housing credit was probably an important factor in the slowdown in GDP earlier this year, and as I wrote last week, probably plays a role in ECRI's recession call.

This morning's report of 681,000 housing permits is with the exception of the month when the $8000 housing credit ended, the highest in 3 years [note: the St. Louis FRED hasn't updated its graphs yet, so the 4 graphs below do not include today's report]:


In the past it has typically taken an improvement of 200,000 housing starts from the bottom to signal that a housing-led expansion has begun:



With November's report we are 85% of the way there from the March 2009 bottom of 513,000.

Another way of looking at housing and expansions is to measure the YoY improvement in the raw numbers. Typically in expansions there have been sustained periods of 200,000+ growth YoY:



With today's number we are 60% of the way there, for the first time without help from the housing credit:



Additionally, per Bill McBride a/k/a Calculated Risk, there is a strong leading relationship between housing starts and the unemployment rate, so a confirmed uptrend should mean a reduction in the unemployment rate.

While by no means are we at the end of the housing bust, today's number serves as confirmation that we are at least at the beginning of the end.

November housing roundup

; - by New Deal democrat

Two months ago I wrote that It's time to admit that house prices have stopped falling. November's housing price data rendered a split decision on that forecast, as asking prices from Housing Tracker confirmed a bottom, while data of repeat house sales in Case-Shiller contrarily did - finally - make a new low under their March 2011 level.

First, here is the 20 city Case Shiller index (from October 2008 to emphasize the post-crash reports, including the effects of the $8000 housing credit):



While the Case Shiller indexes report comparable sales prices, Housing Tracker reports asking prices in 54 metropolitan areas. I follow these because they showed the peak in the housing market at the beginning of 2006 before the Case Shiller sales prices did, and because since then they have led trend changes in the Case Shiller reports by about 4 to 6 months -- which makes sense, since houses are put on the market with asking prices months before they go under contract.

Comparing the Case Shiller graph above with this graph of asking prices for the 75th (yellow), 50th (green), and 25th (blue) percentiles for the last 5+ years at Housing Tracker shows that asking prices have been bottoming out for a year:



Indeed, Housing Tracker's data for all of November showed the YoY rate of declines on a monthly basis is now only -0.7% YoY. Here's the updated chart:

Month	2007	2008	2009	2010	2011
January	---	-7.5%	-11.5%	-5.8%	-8.7%
February	---	-7.8%	-12.0%	-5.2%	-8.4%
March	---	-8.3%	-10.9%	-5.0%	-7.3%
April	-2.7%	-8.6%	-9.6%	-5.0%	-6.8%
May	-3.5%	-9.1%	-8.1%	-5.0%	-5.6%
June	-5.0%	-9.8%	-7.0%	-5.0%	-4.4%
July	-5.4%	-10.4%	-6.1%	-5.1%	-4.2%
August	-6.0%	-10.6%	-5.5%	-6.1%	-2.8%
September	-6.2%	-11.1%	-5.1%	-6.6%	-1.7%
October	-6.7%	-11.4%	-4.5%	-7.0%	-0.9%
November	-6.6%	-11.7%	-4.5%	-6.7%	-0.7%
December	-7.2%	-11.4%	-5.6%	-7.8%	---

Housing Tracker's weekly YoY comparisons for November are instructive: in order, they were -0.9%, -0.5%, -0.3%, and +0.1%. December will tell, but it appears that the turning point has been reached.

Note that the YoY% decline in asking prices bottomed in January of this year. Case Shiller data shows that the change in YoY% decline in sales prices bottomed in May and June, and has been receding since, as shown in this graph:



Keep in mind that even after a nominal bottom in housing prices, there is likely to be an extended period where they are still declining in real, inflation-adjusted terms. Prof. Shiller himself believes another 15% or more decline in real terms is likely, and I do not have any reason to disagree with that.

Indeed, adjusting the 10 and 20 city Case Shiller reports for average wages paid (where the beginning of the 20 city index in January 2000 = 100%), shows that in real terms, the average wage buys more house than at any point since that series began, and is only about 10% above the 10 city's average in the late 1980s and 1990s:



The discrepancy between the Case Shiller and Housing Tracker reports will get resolved one way or the other. I continue to expect Housing Tracker to be the leading series. In that regard, Bill McBride a/k/a Calculated Risk has noted that typically price inflection points are reached at about 6 months' housing supply -- more than that means declining prices ahead, less than that means rising prices ahead. As of the October housing sales reports, new home supply is at 6.3 months and has been declining at a rate of 2 months per year for the last several years. Existing home supply is at 8.0 months, and while the trend was distorted much more by the housing credit, recently inventory has been declining at a -10% YoY rate, which has also coincided with a rate of decline of 2 months' inventory per year. If those trends continue, then within a year both will be under 6.0 months and under CR's metric, we should expect housing prices to be increasing.

Psssst: is this the beginning of the end of the housing bust?

- by New Deal democrat

Housing construction is a long leading indicator, indeed along with interest rates probably the most important one. So those commentators who say that we won't get housing improvement until we have job improvement have causation exactly backwards. Rather, it is much more likely that we won't get more meaningful job improvement until we have more meaningful housing improvement. Further, the decline in housing starts and permits after the expiration of the $8000 housing credit was probably an important factor in the slowdown in GDP earlier this year, and probably plays a role in ECRI's recession call.

With that in mind, housing permits coming in at 653,000 on Wednesday is a significant positive. It confirms the uptrend we've been seeing in housing permits this year, and is the first reading over 650,000 in a year and a half:



This time we were led out of the bottom of the recession by manufacturing and exports, but obviously they need help. In the past it has typically taken an improvement of 200,000 housing starts from the bottom to signal that a housing-led expansion has begun:



With October's report we are 70% of the way there from the March 2009 bottom of 513,000.

Another way of looking at housing and expansions is to measure the YoY improvement in the raw numbers. Typically in expansions there have been sustained periods of 200,000+ growth YoY:



With Wednesday's number we are half the way there, for the first time without help from the housing credit:



It is worth keeping in mind that Bill McBride a/k/a Calculated Risk has shown the strong leading relationship between housing starts and the unemployment rate, so a confirmed uptrend should mean at least a small reduction in the unemployment rate.

As Bonddad described earlier this morning, in the past week we have had a raft of very good economic reports. Housing permits was probably the most significant of them all. While by no means are we at the end of the housing bust (although the bottom was probably put in two years ago in terms of permits and starts), Wednesday's number was significant. If the trend continues, we may have 200,000+ improvement off the bottom within 6 to 12 months. In short, we may be at least at the beginning of the end.

October housing prices: more evidence of a bottom

- by New Deal democrat

Last month I discussed house prices with the provocative title It's time to admit that house prices have stopped falling October's data only adds further confirmation to that trend. The Case Shiller 10 and 20 city indexes both remained higher than their March lows. Here is the 20 city index (from March 2009 to emphasize the post-crash results, including the effects of the $8000 housing credit):



You may notice the "kinks" in the March 2010 and 2011 data. It's possible that there is a slight problem with the seasonal adjustment for that month. Deleting March, here is how the same graph looks:



With that adjustment, it looks like the trend is that the fall in prices, while not over, is probably within 1% of its bottom.

While the Case Shiller indexes report comparable sales prices, Housing Tracker reports asking prices in 54 metropolitan areas. I follow these because they showed the peak in the housing market at the beginning of 2006 before the Case Shiller sales prices did, and because since then they have led trend changes in the Case Shiller reports by about 4 to 6 months -- which makes sense, since houses are put on the market with asking prices months before they go under contract.

Housing Tracker's data for October is in, and it continues to show that the YoY rate of declines has decreased, now only -0.9% YoY. Here's the updated chart:

Month	2007	2008	2009	2010	2011
January	---	-7.5%	-11.5%	-5.8%	-8.7%
February	---	-7.8%	-12.0%	-5.2%	-8.4%
March	---	-8.3%	-10.9%	-5.0%	-7.3%
April	-2.7%	-8.6%	-9.6%	-5.0%	-6.8%
May	-3.5%	-9.1%	-8.1%	-5.0%	-5.6%
June	-5.0%	-9.8%	-7.0%	-5.0%	-4.4%
July	-5.4%	-10.4%	-6.1%	-5.1%	-4.2%
August	-6.0%	-10.6%	-5.5%	-6.1%	-2.8%
September	-6.2%	-11.1%	-5.1%	-6.6%	-1.7%
October	-6.7%	-11.4%	-4.5%	-7.0%	-0.9%
November	-6.6%	-11.7%	-4.5%	-6.7%	---
December	-7.2%	-11.4%	-5.6%	-7.8%	---

Note that the YoY% decline in asking prices bottomed in January of this year. A YoY graph of the Case Shiller data shows that the change in YoY% decline in sales prices bottomed in May and June:



It is important to note that YoY data will not turn positive (or negative) until months after the bottom (or top). For example, nonfarm payrolls bottomed in February 2010. The YoY number did not turn positive until 5 months later, in July. Since Housing Tracker's numbers aren't seasonally adjusted, we have to go with the YoY result, but it is a virtual certainty that at some point in the last 12 months, an actual bottom has already been established.

Why do I track housing prices? In addition to the fact that conventional wisdom, while solidly entrenched, appears to be wrong, it makes a huge difference as to whether any further declines in the housing market are nominal, or only occur after adjusting for inflation (Shiller himself believes another 15% or more decline in real terms is likely, and I do not have any reason to disagree with that).

Sellers who bought a house for $250,000 10 years ago, saw the value increase to $500,000 and then fall all the way back to $250,000 in nominal terms can, even if inflation continues to erode the value of their house by 20% in real terms, still sell their house without bringing cash to the table. On the other hand, sellers whose further depreciation of 20% is nominal (to $200,000), must bring $50,000 to the closing table. You can see that the two scenarios make an enormous difference in the ability of sellers to move, and the likely increase (or not) of foreclosures.

Indeed, comparing nominal Case Shiller values with "real" values adjusting for average wages paid since the beginning of 2000, shows that in real terms, house prices are still falling by about 3% a year - meaning that the average wage earner can afford about 3% more house on an annual basis. Note also that the average wage will buy you more house than at any point since the series began:



Another 4 years of similar real declines will get us to Shiller's 15%+ number -- without any further nominal decline in prices at all.

Rethinking mortgage applications (slightly)

- by New Deal democrat

Another component of the Weekly Indicators, besides money supply which I discussed yesterday, is the Mortgage Bankers' Association's report on purchase mortgage applications. This gives us a high frequency read on housing sales. I've been reporting this in terms of week over week and YoY %age moves.

In the last few weeks, I've been very concerned that it had turned down again, coming in not just lower week over week (to the tune of over 10% in just two weeks) but also turning negative YoY again on the order of more than -5%. On further thought, however, I think I've been a little too pessimistic because I haven't taken into account the absolute level of such applications.

To show you why, here's a five year graph of purchase mortgage applications from Mortgage News Daily:



The above graph is based at zero and each horizontal line is worth 100 on the scale. As you can see, at its peak the value of purchase mortgage application was between 450 and 500. In the three years between mid 2007 and mid 2010, the index fell by about 80 or 90 points a year. Since May of 2010, however, it has been varying between 160 and 200 - even with the recent decline.

In other words, a 10% decline now, for example, is only equivalent to a 3.3% decline when the index was at its peak. This in the graph above, the 5%+ YoY decline of the last month really does not violate the essential flatness in applications for the last 16 months. Put another way, while the decline is certainly not good, it doesn't seem to warrant quite the pessimism it has contributed to my outlook in the last few weeks. But I'd still like purchase applications - and monthly car sales, that other big consumer durable - to recover to their springtime level to be less concerned about the trend in the economy.

Is the bottom in housing prices staring us in the face?

- by New Deal democrat

Many if not most times you will see me cite year-over-year trends in economic data. This is because prior research has been based on YoY trends, and also because many, many data series have marked seasonality. So, for example, with rail traffic you simply can't compare wintertime post-Christmas carloads with late summer or autumn carloads. Similar issues exist for tax withholding and for state tax receipts. Thus the only valid way to see a trend is YoY.

But where there is no seasonality, or where the data has already been seasonally adjusted, YoY comparisons lag turning points. For example, weekly initial jobless claims are seasonally adjusted by the BLS. Waiting for a YoY change there in 2009 would have completely missed the turning point. YoY initial claims did not turn positive (i.e, lower) until November 19, 2009, even though the bottom was made on the week of March 28, 2009.

Which brings us to housing prices. Note: I'm discussing nominal prices here, not "real" inflation adjusted prices. Yesterday the Case-Shiller housing indices for June were released. Here's a graph of the Case-Shiller 20 city index for the last five years:



The Case-Shiller data in the graph above IS seasonally adjusted. So we don't have to wait for YoY changes to make valid statements about the data. Well, here is the data for the 20 city index shown above for the last 6 months:

2011-01-01 141.75
2011-02-01 141.31
2011-03-01 140.30
2011-04-01 140.94
2011-05-01 140.84
2011-06-01 140.76

As I pointed out yesterday, the variation in the index over the last 6 months is less than 1%, and on a seasonally adjusted basis we have not made a new low in 3 months. In fact we are less than 1% below where this index stood in mid- 2009 (before most of the $8000 housing credit distortions kicked in).

Yes, it's only nominal and not "real," and by no means is it clear that March will prove to have been the absolute bottom, but even if there is some further deterioration, we are probably near the bottom in nominal (not "real") housing prices. In fact the bottom in housing prices may be staring us right in the face.

----------
P.S.: It may be tempting to think that the overhang of foreclosures will drive house prices further down. But consider that the same overhang existed in January. It existed in February. It existed in March. And April, May, and June, too -- but it did not drive prices down in those six months. Is anything different about the foreclosure overhang now than during the last half year?

Housing price declines lessen further in August

- by New Deal democrat

Housing Tracker's final report of asking prices in 54 metropolitan areas for August is in, and it shows that the YoY rate of declines continues to lessen, now only -2.8% YoY. Here's the updated chart:

Month	2007	2008	2009	2010	2011
January	---	-7.5%	-11.5%	-5.8%	-8.7%
February	---	-7.8%	-12.0%	-5.2%	-8.4%
March	---	-8.3%	-10.9%	-5.0%	-7.3%
April	-2.7%	-8.6%	-9.6%	-5.0%	-6.8%
May	-3.5%	-9.1%	-8.1%	-5.0%	-5.6%
June	-5.0%	-9.8%	-7.0%	-5.0%	-4.4%
July	-5.4%	-10.4%	-6.1%	-5.1%	-4.2%
August	-6.0%	-10.6%	-5.5%	-6.1%	-2.8%
September	-6.2%	-11.1%	-5.1%	-6.6%	---
October	-6.7%	-11.4%	-4.5%	-7.0%	---
November	-6.6%	-11.7%	-4.5%	-6.7%	---
December	-7.2%	-11.4%	-5.6%	-7.8%	---

If this rate of second derivative improvement continues, we could see a YoY increase in asking prices nationwide before the end of the year. If so, that would mean the nominal bottom in housing prices has already occurred -- probably last January (because of the strong seasonality in housing prices)!

Additionally, as Calculated Risk notes, Housing Tracker's updates continue to show that inventory is also declining.

Note that Housing Tracker is current through last week, vs. this morning's Case-Shiller report, which is an average of April, May, and June. Because of the distortions resulting from the $8000 tax credit that expired in spring 2010, it is interesting to compare 2011 YoY vs. 2009 Case-Shiller index as well as 2011 YoY vs. 2010. Here are the numbers - the first column is vs. 2010, the second vs. 2009:

February: -3.5% -2.7%
March : -3.9% -1.4%
April : -4.2% -0.6%
May : -4.5% 0.0%
June: -4.5% -0.6%

With May and June's data, I suspect the YoY% declines in the Case-Shiller index have made a trough. In the next month or two, with the end of the YoY housing credit distortions, I expect Case Shiller to join Housing Tracker (which is current through last week) in reporting "less worse" declines.

Additionally, with today's data we can see that the seasonally adjusted Case Shiller 20 city index has meandered within a range of less than 1% in the first six months of this year. I bet that little fact isn't getting mentioned on other blogs, is it?

----------
BTW, Dante Atkins a/k/a thereisnospoon has a piece up at Digby's Hullbaloo, purporting to show that present housing prices are still way too high. While I very much appreciate his political commentary, this economic piece is badly misleading.

The "roller coaster ride" of prices is not adjusted for YoY% changes in housing prices For example, there was a 10%+ decline in real housing prices between 1926 and 1933. Even more glaring, there was a 35%+ decline in housing prices between 1912 and 1921 -- almost identical to the declines since the beginning of 2006 till now. Go back and watch the roller coaster ride and pay attention to the markers that tell you when you are in the 1910's, 1920's and the Great Depression. The "roller coaster" stays almost completely flat -- during two of the three worst downturns in the last 100 years!

Thereisnospoon suggests that housing has much, much further to fall, but in fact, this series has never been at a level under 110 since 1949. It is presently at about 132, and thus is only about 15% higher than its lowest reading in 62 years. I expect almost all of the remaining adjustment to take place via inflation rather than a continuing decline in nominal prices.

Housing and the double-dip

- by New Deal democrat

Suppose my analysis is wrong, and we actually have a "double-dip" recession? How bad is it likely to be?

I wanted to put up a much longer discussion on this, but frankly I don't have the time right now. One important point, however, is to look at housing. Housing historically has been a long leading indicator ( 6 to 12 or even 15 months ahead of the economy as a whole). Bill McBride a/k/a Calculated Risk has pointed out a number of times that he did not expect the unemployment rate to drop that much, because it correlates quite well with housing starts - with a lag - and housing starts have gone more or less sideways for two years. The relationship going back 50 years is shown in this graph:



(Note: unemployment rate is inverted, right scale, better to show the relationship)

As I have previously pointed out, housing starts also correlated very well with job growth in the 1920s and 1930s as well. But if the correlation holds true, than the inverse also holds true also. That is, if housing starts have not declined, then we should not expect the unemployment rate to grow very much in any double-dip, either. A ceiling of about 10% or so in the next 6 - 12 months on the unemployment rate based on housing present status is certainly a valid estimate.

And the housing market shows no sign of any meaningful accelerated downturn. To the contrary, new home sales data for July showed that the months of supply of houses on the market has dropped to 6.6 months. With the exception of a very brief period during the artificial support to housing of the $8000 tax credit, this is the lowest months' of housing for sale since the onset of the recession almost 4 years ago:



(courtesy Calculated Risk)

With no further goverment stimulus at all, this metric has declined by 2.5 months during the last year. It is not unreasonable at all to believe it could drop below the long-term average of 6.0 months during the next year.

This week Housing Tracker reported that YoY asking prices are only down -2.3%. One quarter of all metro areas tracked are now reporting YoY increases in asking prices.

In my opinion the bottom in nominal housing prices (as opposed to inflation-adjusted prices) is much closer than almost all analysts suspect. Once we start to get an upward trend in housing, then according to the correlation with the unemployment rate, there should be good fundamental support for job growth. Housing simply does not support a substantial double-dip.

June YoY house price declines least since May 2007

- by New Deal democrat

Consider two different scenarios by which house prices could still decline 25% in real, inflation-adjusted terms, from the present:

(1) nominal prices decline 5% a year for 5 years, and inflation is 0% over that time.
(2) nominal prices do not decline at all over the next 5 years, but inflation is 5% a year.

Both of those give us "real" price declines of 25%, but with very different results, and very different amounts of pain.

In the first scenario, more and more homeowners are "underwater" with houses not worth what they paid for them, and not even worth the outstanding mortgage amount. They are unable to sell, since they can't bring cash with them to the closing table to make up the difference. More and more allow their houses to slide into foreclosure, thus increasing the shadow inventory of bank-owned houses.

In the second scenario, however, no more homeowners whatsoever are "underwater." Even better, mortgage payments - and purchases - are made with inflated currency. There is no further incentive to hand in the keys and walk away from the house, and shadow inventory is worked off.

Needless to say, the second scenario is a lot less painful than the first one and yet accomplishes the same result.

Over the last several weeks, I've read more and more commentary suggesting that at least nominal prices in the housing market might start to make a bottom. Now that we have the most recent Case-Shiller report (from April, but really a February, March, and April average), and the final June asking price data from Housing Tracker, which accurately showed the turn at the top of the market in 2006, let's take a look at whether housing prices may better fit the first or second scenario.

Here is the updated chart of YoY% change in asking prices from Housing Tracker's 1,000,000+ home database:

Month	2007	2008	2009	2010	2011
January	---	-7.5%	-11.5%	-5.8%	-8.7%
February	---	-7.8%	-12.0%	-5.2%	-8.4%
March	---	-8.3%	-10.9%	-5.0%	-7.3%
April	-2.7%	-8.6%	-9.6%	-5.0%	-6.8%
May	-3.5%	-9.1%	-8.1%	-5.0%	-5.6%
June	-5.0%	-9.8%	-7.0%	-5.0%	-4.4%
July	-5.4%	-10.4%	-6.1%	-5.1%	---
August	-6.0%	-10.6%	-5.5%	-6.1%	---
September	-6.2%	-11.1%	-5.1%	-6.6%	---
October	-6.7%	-11.4%	-4.5%	-7.0%	---
November	-6.6%	-11.7%	-4.5%	-6.7%	---
December	-7.2%	-11.4%	-5.6%	-7.8%	---

The YoY -4.4% decline in June is the smallest YoY decline since May 2007. Here is the same information (through 2 weeks ago) shown graphically (h/t Silver Oz):


In comparison, here is the YoY% change up through the February - April average in the Case-Shiller 20 city index:



Over the time period of comparison, the Housing Tracker trend in asking prices has appeared to run 1 to 4 months ahead of the Case-Shiller sales data. Keep in mind that the most recent Case-Shiller data compares sales from a period when buyers and sellers wanted to close quickly to take advantage of the $8000 housing credit, with a period one year later where there is no credit. Even so, it looks like the YoY% change may be close to bottoming.

I fully expect the Case-Shiller series to bottom out in YoY% terms within the next several months, and to mirror the more current better YoY comparisons in the Housing Tracker database. In other words, it continues to look like the second scenario set forth above is going to be closer to the truth, and among the possibilities, that's good news.

New aggregate Housing Tracker data tears down conventional wisdom

- by New Deal democrat

Yesterday provided some perfect examples of the conventional wisdom about the housing market - namely, that recent new declines in the Case-Shiller Index show that we are spiraling down into the double-dip of Doom.

Jeff Cox of CNBC wrote:

It's official: The housing crisis that began in 2006 and has recently entered a double dip is now worse than the Great Depression.

Prices have fallen some 33 percent since the market began its collapse, greater than the 31 percent fall that began in the late 1920s and culminated in the early 1930s, according to Case-Shiller data.

Writing at the Huffington Post, David Paul of the Fiscal Strategies Group was even more pessimistic, writing that Those Waiting for a Housing Market Rebound May Have a Long Wait Ahead:

Even with the sharp contraction in prices that we have realized since the unraveling of the mortgage bond market almost three years ago, housing prices remain high by historical standards ....

[W]ith the expiration of the first-time homebuyer tax credit and looming end to the quantitative easing ..., the housing market is only now being left to face the brunt of post-crisis market forces without those two forms of federal support. As that support falls away, it is hard to be optimistic about the direction of the housing market ....

... S&P projects that when all is said and done, home prices will fall 35% from their pre-crisis peak, meaning that we have another 15% downside to go. Yet by historical standards, this will not produce "cheap" prices as S&P suggests, but rather return prices to within historical norms. But whatever bottom is ultimately reached in housing prices, a resurgent housing market will also require an environment that provides reasonable expectations for asset price stability, if not appreciation. And this appears unlikely to happen any time soon.

(my emphasis)

To begin with, let's take a look at the Case-Shiller 20 City index and see what all the hubbub is about. Essentially Cox is saying that the small downturn in the below graph (showing prices as a percent of the April 2006 peak) for the last 9 months, putting its value slightly below its 2009 lows, was just enough to make the difference between a skosh above Great Depression declines, to a skosh below them:

To the contrary, I submit that the above graph shows that there is no descent into the abyss discernable in house prices over the last 2 years.

Furthermore, certainly as to the Case Shiller 20 city index (red in the graph below vs. the 10 city index in blue), real inflation-adjusted prices have fallen to year 2000 levels and are certainly in their pre-bubble normal range for the last 24 years since either series was begun:



This doesn't mean that real prices can't fall further. But what it does mean is that most if not all of the remaining decline might not be in nominal prices. Nominal prices could easily stabilize while inflation means that the "real" value of housing continues to decline, just as it did in the earlier 1990's.

And that brings me back to the 1,000,000+ asking price database from Housing Tracker, which accurately showed the turn in the market in 2006 - a fact I know because I documented it at the time. On Monday, using an unweighted average from the 54 metro areas tracked in the database, I showed that the YoY% decline in asking prices was rapidly abating, and that if the trend continued at the same velocity, a bottom in prices could be reached as early as next winter.

As chance would have it, over the weekend for the first time the site added aggregate information for the entire US from April 2006 to the present. This means that I can easily calculate the YoY% changes in asking prices nationwide since the peak of the market, right up until last week (marked with an asterisk in the chart below):

Month	2007	2008	2009	2010	2011
January	---	-7.5%	-11.5%	-5.8%	-8.7%
February	---	-7.8%	-12.0%	-5.2%	-8.4%
March	---	-8.3%	-10.9%	-5.0%	-7.3%
April	-2.7%	-8.6%	-9.6%	-5.0%	-6.8%
May	-3.5%	-9.1%	-8.1%	-5.0%	-5.6%
June	-5.0%	-9.8%	-7.0%	-5.0%	-4.5%*
July	-5.4%	-10.4%	-6.1%	-5.1%	---
August	-6.0%	-10.6%	-5.5%	-6.1%	---
September	-6.2%	-11.1%	-5.1%	-6.6%	---
October	-6.7%	-11.4%	-4.5%	-7.0%	---
November	-6.6%	-11.7%	-4.5%	-6.7%	---
December	-7.2%	-11.4%	-5.6%	-7.8%	---

The weighted average shows a similar abatement in YoY% declines this year as did the unweighted average I calculated on Monday. Generally we can see that the declines got worse and worse until the $8000 credit was enacted. The declines rapidly abated, and then increased again as soon as the credit expired. (Also, Calfiornia had a temporary, additional $10,000 credit which enhanced the effect there). But the declines have swiftly abated again this year as "real" prices approach their historical norms and inventories continue to decline to lows not seen since 2006. In fact, the 4.5% YoY decline reported for last week is equal to the smallest decline in 4 years.

Here is the same information from the chart above shown graphically (h/t Silver Oz):


This is a series that, even without artificial government support, wants to continue to trend towards the zero threshold.

In comparison, here is the YoY% change for the same period of time in the Case-Shiller 20 city index:

The Housing Tracker trend in asking prices appears to run 1 to 4 months ahead of the Case-Shiller sales data, which as of the moment is only available through March. Even in the Case-Shiller series, it looks like the YoY% change may be close to bottoming.

To put it bluntly, an examination of the newly available aggregate US asking price data from Housing Tracker contradicts the conventional wisdom above that, as "the housing market is only now being left to face the brunt of post-crisis market forces, it is hard to be optimistic about the direction of the housing market," that "we have another 15% downside to go," and that "reasonable expectations for asset price stability, if not appreciation [a]ppears unlikely to happen any time soon."

To the contrary, housing prices have already "faced the brunt of market forces" without support for a full year, as a result of which they have been falling closer and closer to equilibrium, the rate of decline is abating, and actual real time data shows that nominal if not inflation adjusted stability may indeed be reached as soon as early next year.

Economic Idiocy on the front page of Daily Kos

- by New Deal democrat

It's bad enough that the only diarist left who hits the Wreck list there is the Pied Piper of Doom, but now the economic lunacy has hit the front page as well.

Last night Joan McCarter put up a piece called Housing news remains bad, but is likely to get worse. Relying on a story by Kevin Drum, who is talking about housing *prices*, she says:

This ... is really not good news for the nation's economy.
Home values posted the largest decline in the first quarter since late 2008

and continues:

....Of course, ongoing 9% and higher unemployment makes that picture even more grim, as does the news that more than 28% of homeowners are currently underwater on their mortgages. Foreclosures are spreading [and] home values continue to plummet....

Between high unemployment, stagnant wages, and rampant ongoing fraud and mortgage problems, who is going to be in the market to get a mortgage and buy a house?

(my emphasis)

I should start by saying I don't know Ms. McCarter at all, and have no personal ill will towards her. But the above is pure nonsense.

To begin with, foreclosures peaked over a year ago, in March 2010.

More to the point, on the very same day that a DK front pager says "who is going to be in the market to get a mortgage and buy a house?" the blog Housing Wire reports that:

Florida existing home sales followed the national trend upward in the first quarter, jumping 13% compared to the year prior to more than 44,500 homes statewide.

Existing condo sales rose 29% compared to the first quarter of 2010, reported Florida Realtors. The organization said all but one Florida metropolitan area witnessed annual gains in condo sales including Tampa-St. Petersburg-Clearwater, Fla., which saw a 27% increase in condo sales. Only condo sales in Tallahassee, Fla. dropped compared to 2010, down 20%.

In total, 23,375 condos sold around the Sunshine State, up 29% from little more than 18,000 in the first quarter of 2010.

And via Data Quick, we learn that:

Phoenix-area home sales surged in March, rising more than usual from February and posting a year-over-year gain for the third consecutive month as sales to absentee buyers hit a new high.... A total of 10,352 new and resale houses and condos closed escrow in March in the combined Maricopa-Pinal counties metropolitan area. That was up 44.3 percent from the month before and up 7.5 percent from a year earlier, according to San Diego-based DataQuick, which tracks real estate trends nationally via public property records.

A sharp rise in sales from February to March is normal for the season, although this year’s jump was larger than usual.... March’s sales were the highest for that month since 10,712 homes sold in March 2007. This March’s sales tally was just 1.0 percent short of the average number of sales for the month of March since 1994.

and further that:

Las Vegas region home sales held at a five-year high in March amid unusually strong activity among investors, cash buyers and others targeting sub-$100,000 homes.... In March, 4,953 new and resale houses and condos closed escrow in the Las Vegas-Paradise metro area (Clark County) – the highest sales tally for any March since 2006, when 8,486 sold.

March sales were up 27.3 percent from February and up 1.6 percent from March 2010, according to San Diego-based DataQuick.... On average, the region’s sales have risen 29.0 percent between February and March since 1994, when DataQuick’s complete Las Vegas region statistics begin. March’s sales total was 0.1 percent higher than the average number sold in March since 1994.

Memo to Daily Kos front pagers: when you make something cheaper, and the interest rates to pay for it go down, sooner or later it turns into a real bargain, and more people will buy it. Yes, prices are off an all time record from their highs -- which is awful if you are a seller, but is terrific if you are a buyer, especially a first time buyer. Which is most certainly not "bad news for the nation's economy." As I put it in March, describing this situation, Lo and behold, supply and demand works!

Are home sellers in 2011 throwing in the towel?

- by New Deal democrat

There is a really interesting article by Scott Sambucci today, Housing: Why are new listing [prices] down so sharply?.

It's a really interesting articles with great explanatory graphs. I recommend you go over and read the whole thing. His basic conclusion: unlike prior years since the onset of the housing bust, people (or banks) listing houses for sale for the first time this year are underpricing houses already on the market. In other words, they are throwing in the towel.

House Price declines mean the Biggest Deflation since 1950

- by New Deal democrat

Last week Zillow made headlines with a report stating that Home Value Declines Surpass those of the Great Depression, specifically that:

November brought continued declines in home values. In fact, the Zillow Home Value Index has now fallen 26% since its peak in June 2006. That’s more than the 25.9% decline in the Depression-era years between 1928 and 1933.

Since there was no consumer price sub-index for housing in existence until 1935, Zillow either had to be comparing house prices to the overall inflation rate, or else using a different source. Since then, Zillow has confirmed that its source was Robert Shiller's book, Irrational Exhuberance, in which Shiller says in a footnote that house prices declined 25.9% from 1929 to 1933.

The problem is, like our own era, housing did not peak in 1929. In fact house prices peaked in 1925 according to tables derived by Shiller. Thus Zillow is comparing prices from our peak, but not the peak in the 1920's. For an apples to apples comparison, we need to calculate from peak values. Originally I thought that would negate Zillow's conclusion. The truth turned out to be just the opposite.

Using the data from Shiller's price table linked to above, we see the price index declined from a peak of 6.34 in 1925 to 4.41 in 1933, or 30.4% (note: Shiller's table only calculates on an annual basis for that era). In our era, the quarterly index peaked at 190.44 in Q1 2006 and has declined to a trough - so far, at least - of 131.39 in Q1 2009, for a decline of 33.0%. (Since then, it has rebounded slightly to 133.22).

But what about in "real" terms? Officially, we have inflation now, vs. about 25% deflation between 1929 and 1933. But of course back then CPI wasn't calculated using "owner's equivalent rent." Which means that doing a Case-Shiller price index adjusted CPI (or CS - CPI) comparison of cumulative price changes is the best method of comparison.

According to the BLS, the weight given to owner's equivalent rent is 23.830. For our purposes we can round to 25%.

The BLS calculation is that between April 2006 and March 2009, overall inflation was +6.5%. Owner's equivalent rent's 1/4 share of that, believe it or not, increased 7.8%. Thus replacing +7.8% with -33.0%, and giving that a 1/4 weighting of the inflation index gives us CS-CPI decrease of -3.4%.

In other words, not only have house prices declined slightly more than they did during the Great Depression, but the best apples to apples comparison of inflation from April 2006 through March 2009 means that we experienced a deflationary episode that is the worst since the -4.1% of 1948-50 (the 1938 Recession's was -5.5%). And while they haven't yet hit their 2009 bottom yet, since the expiration of the $8000 housing credit last May, house prices have started to decline again.

Housing: the Great Recession vs. the Great Depression

- by New Deal democrat

Several commentators, notably Calculated Risk, believe that residential investment will probably increase - finally - late in 2011, thus going from a drag on the economy to a contributor to GDP growth. Since the housing bubble of the first part of last decade helped lead the economy into the "Great Recession", there has been consternation as to whether it could lead the economy out of the recession -- or at least finally contribute.

Several years ago, Prof. David Leamer helpfully noted that there was a housing bubble in the 1920s as well - that burst several years before the actual nonset of the Great Depression, and bottomed in 1933. Following up on that, I was able to find the data for "non farm housing starts" in the Statistical Abstract of the United States. I posted the results a year and a half ago in Housing during the Roaring Twenties and Great Depression. Included then was a chart that converted the equivalent housing starts to our time, by multiplying the data proportionate to the population for each year compared with our modern population of 300 million.

Adjusting for population makes a significant difference in how one looks at the data. Building 2 million new houses a year means something quite different in 2005 with a 300 million population, than in 1960 with a 160 million population. So adjusted, the 2000's housing bubble does not look so extreme:



Adjusting for population alone doesn't tell the whole story. There remain the issues of demographics and immigation -- in other words, one must consider the factors behind household formation. I discussed this more thoroughly here, but to summarize for this post, young people begin to move out of their parents' homes at about age 20. Those who marry very young form their own household, while young singles live with roommates. By age 25, new household formation rapidly accelerates, and continues through age 35, by which time close to the maximum percentage of people are living in their own or spouse's household.

Thus Boomer household formation coincided with the 1970's-1980's peaks, and the smaller cohort of Gen X coincides with 1990's lower numbers of housing starts. Gen Y, or the "Millenials" have already begun to form households, just as the housing bust hit.

With another year's passage of time and additional data, I am republishing the chart from my original post about housing during the Roaring Twenties and Great Depression, but I have expanded it. Not only does it adjust the 1920's figures to the current era by population, but also I am comparing the equivalent year from the Roaring Twenties and Great Depression with the last decade of housing starts, normed to the peak year, which are averaged on an annual basis to make an apples to apples comparison. All data is in 1000's:

Year	1920's-1930's Nonfarm housing starts (actual)	1920's-1930's adjusted for population	2000's housing starts adjusted for population
1920/2000	247	699	1689
1921/2001	449	1236	1701
1922/2002	716	1953	1799
1923/2003	371	994	1932
1924/2004	893	2350	2014
1925/2005	937	2423 (high)	2121 (high)
1926/2006	849	2177	1837
1927/2007	810	2042	1347
1928/2008	753	1867	895
1929/2009	509	1252	546
1930/2010	330	805	592
1931	254	615	---
1932	134	322	---
1933	93	221	---
1934	126	300	---
1935	221	522	---
1936	319	748	---
1937	336	781	---
1938	406	937	---
1939	515	1179	---
1940	602	1368	---

In reviewing this chart, keep in mind that in 1930, 25% of the US population lived on farms, compared with only 2% today. Also be mindful that back then, the typical mortgage was a 5 year note that frequently included a balloon payment at the end.

The data shows that if anything, the 1920s boom was much more of a bubble than the recent one, and the ensuing eight year decline from 1925-1933 was more gradual but also sank to a lower level before bottoming.

At this point, according to recent Case-Shiller data, some housing markets are at near quarter century lows in prices. For example, here are Charlotte (green), Las Vegas (red), and Cleveland (purple) (CPI inflation in blue):



In all three of these cases, prices are actually lower now in real terms than they were 23 years ago or at any time since.

All of the above doesn't tell us that housing starts *will* increase this year, but it certainly suggests that if not this year, then either next year or the year after will see a significant increase. Further, the odds appear favorable that in at least some metropolitan areas, we will see an increase in building at some point this year.

The Housing Bust: Is it time to start looking for the end?

- by New Deal democrat

Way back in 2004 (if not earlier), Ben Jones' Housing Bubble Blog started, dedicated to the proposition that there was a housing bubble that would eventually burst, with very bad consequences. Although the peak of the bubble did not occur until the beginning of 2006, the signs were there to see for the discerning.

The other day, with the release of the most recent Case-Shiller housing indexes, I decided to take a look at what the "real", inflation-adjusted cost of housing was in the cities surveyed. Some of the results were stunning. Prices in about half of the 20 cities surveyed are at the lowest they have been, in "real" terms, in nearly a quarter century -- and that's before taking into account mortgage interest rates.

Below are the results in graphic form. Since housing itself is a component of inflation, rather than use the CPI, I have compared housing prices with the growth of average hourly earnings in private industry since 1987 (blue line in each of the graphs below).

This compares average hourly earnings with the overall 10-city Case Shiller housing index:

As you can easily see, in the aggregate housing prices are still a good 20% above their long term average, suggesting that prices as a nationwide average have a substantial way to fall. Based on that, upon release of the most recent Case Shiller data a few days ago, Standard and Poors opined that they expect house prices to fall another 7 to 10% through the end of 2011.

On a national scale, that is probably close to the mark. But look what happens when we start looking at different cities.

Here are three cities at the epicenter of the housing bubble: Tampa, Miami, and Phoenix:


These three cases all had smaller housing booms in the late 1980s that did not bottom out until the mid-1990s. Housing prices in each locale now are as low as they were at that mid-1990s nadir in real terms.

Here are three more cities -- Las Vegas, Charlotte, and Cleveland -- that did not have mini-busts in the 1990s, but in the case of Las Vegas, was another epicenter of the housing bubble:


In all three of these cases, prices are actually lower now in real terms than they were 23 years ago or at any time since.

Atlanta, Minneapolis, and Detroit likewise have the lower prices now than at any time since this series started:

Detroit, of course, more than any other large city, has suffered from de-industrialization.

So, where are prices still elevated?

It is good to be the seat of the financial or political Empire, respectively. Here are New York City and Washington, DC:


No regression to the mean in either of those two cities. Nor is there any along the West Coast:


Los Angeles, San Francisco, Portland, and Seattle still show highly elevated "real" prices. Although I haven't graphed it, so does San Diego which is also part of the Case Shiller sample.

In short, fully half of the cities examined in the Case Shiller index have house prices now that are equal to or lower than any other prices in real terms in the last 20+ years. The bubble has fully burst in those cities.

The low absolute home prices are only amplified by mortgage rates, which were about 11% in 1987, 7% or so in the mid-1990s, but only a little above 4% now.


So the monthly mortgage payment now is only about 40% of what it would have been in real terms back in 1987.

Of course, the fact that housing prices in the above referenced 9 locales are at quarter-century lows does not mean that they won't go lower. Some of them have been low for several years, and yet continued to decline.

Courtesy of Ben Engebreth's Housing Tracker, here are housing inventories for 5 of the metropolitan areas referenced above.

Here's Tampa:


Here is Phoenix:


Here is Atlanta:


Here is Cleveland:


Finally, here is Washington, D.C.:


In all five of the above metropolitan areas, inventory of houses for sale has declined from peak. But in none of the cases is the inventory back where it was in 2005 when the market in the most bubblicious areas started to turn. It is when this inventory goes back down to its 2005 levels or lower that we can expect a bottom in house prices. My best guess is that the national low for the Housing Bust is still several years away.

Yet the low real prices and declining inventory in some of the most severely impacted cities argues that we should start to look for a bottom in at least some of those locales within the next 6 to 12 months.