Tuesday, July 28, 2009

Latest US house price data

OK the Case Shiller house price data is out today (for May). Last months data I pointed out an emerging trend.

"5 cities that did not have a big boom, the change in prices was basically 0 over the last 2 months (with a range of -2% to +1%, while in the boom cities the average was -3.8% (range -1% to -7%). Clearly a significant difference, both statistically and practically speaking. So the recent trend is that in the non boom cities prices are almost stable despite the rising unemployment while in the cities which had booms the price falls are slowing but still there. "

Let's see now looking at the last 2 months if this trend is becoming clearer or was merely a statistical blip. 5 cities that did not have a big boom, the change in prices averaged 1% up over the last 2 months (with a range of -1% to +3%) while in the boom cities the average was -1.7% (range 1% to -7%). So for non boom cities prices do appear to have stabilised while for the boom cities prices continue to fall but that rate of price decrease has slowed significantly compared to 3-4 months ago.

So what does this mean? Well it's probably not too bad a time to buy a house in the non boom cities. However most banks exposures to bad loans are in the boom cities and these are still falling though at a slower rate.

There has been some discussion amongst bloggers about problematic Option ARM loans and when they are likely to default. This discussion has centred on recast and rest dates when payments or interest rates change for these loans. See http://www.calculatedriskblog.com/search/label/Option%20ARM for a summary of Option ARM posts on the excellent calculated risk blog.

As I pointed out in my last post default decision are likely in most cases to depend upon changes in personal circumstances. So defaults are likely to be spread over time rather than bunched according to rest or recast dates. This is what appears to be happening now see http://www.calculatedriskblog.com/search/label/Option%20ARM. US banks face a long hard road ahead.

Wednesday, July 1, 2009

Forclosures - a short or long term problem?

Over at http://oldprof.typepad.com/a_dash_of_insight/2009/06/interpreting-housing-indicators.html Jeffrey A. Miller has a good post on seasonality in data and on house prices in particular.

On the latest Case-Shiller US housing price data the most interesting thing I found (based on the more reliable seasonally adjusted data) was that for the 5 cities that did not have a big boom, the change in prices was basically 0 over the last 2 months (with a range of -2% to +1%) while in the boom cities the average was -3.8% (range -1% to -7%). Clearly a significant difference, both statistically and practically speaking. So the recent trend is that in the non boom cities prices are almost stable despite the rising unemployment while in the cities which had booms the price falls are slowing but still there.


Looking to the future I look at two questions.
1. How are things going to pan out in the next 3-4 years and beyond for house prices in these various cities in the US?
2. How about foreclosures over this 3-4 year period?

OK house prices. Generally the futures markets and economic forecasters suggest some more falls in the boom markets over the next year or two (say 10-15%) then some flattening out. This seems a reasonable guess to me though of course there's a fair margin of error and clearly different cities and different market sectors are going to have different outcomes. e.g. the high priced end of the market in boom cities may have further to fall as price differentials have widened with the collapse at the lower end and there is a lack of "move up" buyers.

Now turning to the second question what is going to happen regarding defaults and the results modification, foreclosures, short sales etc. i.e. problems for banks. Most commentators give the impression that the foreclosure problem is related largely to sub prime and that as we have worked through a lot of this the problem will be largely gone over the next year as the worst of these loans would have already defaulted. On this basis they see bank losses related to the residential housing market as not being a problem beyond the next 12-18 months. I disagree for two reasons.
1. The recent data.
2. My hunches about why people will default and how often this will happen.

The recent data on delinquent loans (i.e loans where payments are more than 60 days late) show a steady and significant increase in each of the last 4 quarters in the proportion of prime loans going delinquent. Prime loans are two thirds total loans and are the "best quality" loans. So the data suggests a growing and broad problem not related to sub prime that is not going to go away when the majority of sub prime loans have defaulted. Sure you might say "but it's the recession dummy - its' people loosing their jobs and not being able to keep up with the mortgage payments". That's part of it but it's far from the whole story as I outline below.

Let's turn now to the reasons why people default and consider whether this is likely to continue over the next 3-4 years (and possibly much longer).

My hunches are that the main triggers for default are:

1. The pure economic motive. Some people default when they see the value of their property is much less then the value of their loan and they don't expect their house value to improve significantly anytime soon. I presume investors and many who bought near the peak of the boom are in this category. Basically many of the people in this category who were going to default have already defaulted so while this will be a continuing issue I don't see it being the main issue in the longer term unless it is teamed with the circumstances outlined below.

2. Defaults because the homeowner can't afford the repayments anymore. Clearly triggers here include: decreased income due to job loss, a reduction in working hours or being forced to take a worse paying job are going to be triggers. Also clearly this is going to be a big issue over the next year at least. However even when unemployment has stabilised there are still vast numbers of jobs lost and created in any year in a "normal" economy. Coming out of past recessions this has not mattered much to banks as for almost all homeowners their loan value was less then their home value. So the homeowner maybe forced to sell the house (but this caused the bank no loss), or take a second mortgage with the equity they had in the house, or borrow from relatives etc. However for the foreseeable future, in all the housing markets that had the big boom bust cycle, these options are either impossible or no longer in many homeowners best interests. It's in their interest to default. This means even in the years following the end of the recession there will still be substantial defaults where banks will loose substantial money on the foreclosure sale when people temporarily can't afford the repayments.

3. The same dynamic exists as point 2 for situations where some other change in life circumstances triggers a home sale (and where the loan value is greater than the house value). Examples of these circumstances include: moving to another area for job or family reasons, life cycle events that trigger a desire or need to move/downsize/upsize (divorce, marriage, death, birth etc). These are the normal factors that trigger most sales. For many of these people the new reality in "post recession" economic times will be that they will be better off financially by defaulting on their loan.

These three points suggest that, in the markets that had the house price boom bust cycle (i.e most metro markets), defaults and foreclosures are going to be commonplace long beyond the end of the recession. This has four significant implications.
In the post boom housing markets it will mean:
1. Foreclosures and short sales which will dampen any medium term rebound in house prices.
2. A negative impact on the profitability of new construction in those markets as house prices will remain subdued.
3. Ongoing losses over to those lenders that made (even prime) loans in these markets even after the recession passes.
4. Residential housing investment is unlikely to rebound as strongly as it usually does at the end a recession. This will tend to make economic recovery sluggish - especially in post boom markets (like the sunbelt) where the economic situation is already worse than average.

It is point 3 that I have yet to see mentioned in commentary.