by Anonymous on 7/13/2007 08:34:00 AM

From the Census Bureau, "Advance Monthly Sales for Retail Trade and Food Services":

The U.S. Census Bureau announced today that advance estimates of U.S. retail and food services sales for June, adjusted for seasonal variation and holiday and trading-day differences, but not for price changes, were $373.9 billion, a decrease of 0.9 percent (±0.7%) from the previous month, but 3.8 percent (±0.7%) above June 2006. Total sales for the April through June 2007 period were up 3.9 percent (±0.5%) from the same period a year ago. The April to May 2007 percent change was revised from +1.4 percent (± 0.7%) to +1.5 percent (± 0.3%).

Retail trade sales were down 1.0 percent (±0.7%) from May 2007, but were 3.5 percent (±0.8%) above last year. Nonstore retailers were up 9.5 percent (±4.5%) from June 2006 and sales of health and personal care stores were up 7.0 percent (±1.7%) from last year.

You don't get the pretty updated chart you usually get because CR is spending a long weekend getting high climbing mountains and he doesn't let me mess with his charts. Last month's chart is here.

by Anonymous on 7/13/2007 07:41:00 AM

Yes, I mean utter worthlessness. I have seen one or two unhelpful Fed Letters in my time, but this one, "Comparing the Prime and Subprime Mortgage Markets," brings a remarkably shallow grasp of the subject to bear on a perfectly vacuous thesis in order to produce three and a half pages of tripe. If this is the sort of advice the Fed Banks are taking from "economists," then no bloody wonder we're in trouble.

Here's how it starts:

We show that the subprime mortgage market is facing substantial problems, as measured by delinquency rates, while the prime mortgage market is experiencing more typical delinquency rates, i.e., at historical averages (see figure 1). Within the subprime mortgage market, we observe a substantial increase in delinquency rates, mostly for adjustable-rate mortgages (ARMs). Since the subprime ARM market is less than 7.5% of the overall mortgage market and a vast majority of subprime loans are performing well, we believe that the subprime mortgage problems are not likely to spill over to the rest of the mortgage market or the broader economy.

That's it. In three and a half pages, the authors demonstrate that subprime ARMs are the segment performing at worse than "historical" rates of delinquency, and that they are indeed 7.5% of the outstanding mortgage book. What we never get is a statement of what it might mean for "subprime problems" to "spill over to the rest of the mortgage market or the broader economy." What, exactly, does that mean? Is the implication that the "subprime problem" could "spill over" if it were only a bigger piece of the mortgage pie? What, exactly, would be the mechanism of this "spill over"? If you guessed that it might have something to do with declining MEW and downward spirals of home values due to subprime foreclosure waves, you might have a point, but neither of these issues is raised in the Fed Letter. Apparently, on Chicago Fed's planet, things just "spill over" when the bucket gets full enough of delinquencies.

The whole thing is littered with irritating mischaracterizations of both mortgage lending practices and recent so-called "bailout" initiatives, the latter of which involves throwing around some numbers ($20 billion from Freddie Mac, a billion from Citi and BoA) without offering any analysis of how far that goes to sort out the $1.5 trillion in subprime loans on the books. You get this level of analysis on Bloomberg (and you get it more timely).

Dear Chicago Fed: we're all really tired of rehashing sound-bites. How about you using the institutional smarts you have to answer some interesting questions for a change? Such as: how did we get so many subprime borrowers in the first place? Do states like Michigan and Indiana have high rates of delinquency and default because of employment woes, or high rates subprime originations because of employment woes? Is it possible that the "historically" low rate of prime mortgage default exists because subprime has been available to accept the "spill over"? What happens when the lender of last resort goes away?

Look, there are two reasons why the default rate on prime loans is as low as it is. The first, obviously, is that decent credit standards in the first place limit the number of loans that experience delinquency. The second is that, historically, there has been somewhere for those loans that do experience delinquency to go: either a voluntary sale of the home that covers the loan amount or a refinance into a subprime loan. The existing home market and the subprime refi market are the "spill overs" of the prime mortgage market. To discuss the question of "prime contaigon" without reference to existing home sales or refinance opportunities (by price and by credit standard tightening perspectives) is, indeed, utterly worthless.

by Anonymous on 7/12/2007 07:19:00 PM

I thought you might be interested in this little tidbit from Moody's. According to Moody's, the average serious delinquency rate for 2006-vintage subprime securities at 10 months of seasoning is 9.2%. But some originators are more average than others:

S&P also released information on actions taken on the 612 securities it put on "watch negative" on July 10:

Regarding the July 10, 2007, CreditWatch actions affecting 612 classes of RMBS backed by first-lien subprime mortgage collateral, 498 classes were downgraded, 26 classes remain on CreditWatch, and the ratings on 74 classes were affirmed and removed from CreditWatch. Additionally, the ratings on nine other classes were affirmed and removed from CreditWatch because they involve Alternative A mortgage collateral and were not intended to be included in July 10, 2007, action. These nine classes are from the following deals: GSAA Home Equity Loan Trust 2006-5, Lehman XS Trust 2006-7, and Luminent Mortgage Trust 2005-1, and will be addressed when Standard & Poor's reviews transactions backed by Alternative A mortgage collateral.

The ratings on 26 classes remain on CreditWatch because the issuer has appealed the decision based on the presence of mortgage insurance in those transactions. We are currently reviewing this appeal. In addition, the ratings on five other classes remain on CreditWatch because they are backed by closed-end second-lien mortgage collateral and will be addressed when Standard & Poor's reviews transactions backed by closed-end second-lien mortgage collateral.

Regarding the 70 classes placed on CreditWatch before July 10, 2007, 64 were downgraded and six remain on CreditWatch. Three classes remain on CreditWatch because the issuer is appealing the decision based on the presence of mortgage insurance and we are reviewing this appeal. Three classes remain on CreditWatch because they were placed on CreditWatch before July 10, 2007, and involve either closed-end second-lien or Alternative A mortgage collateral. They will be addressed when Standard & Poor's reviews transactions backed by closed-end second-lien and Alternative A mortgage collateral.

Ooops. We didn't notice the mortgage insurance coverage? We can't tell the difference between subprime and Alt-A? Some second liens snuck in when we weren't looking? It has not been a stellar week for S&P.

Of the 612 classes placed on CreditWatch on July 10, 2007, the 498 downgraded classes total approximately $5.69 billion in rated securities, which represents 1.01% of the $565.3 billion in U.S. RMBS first-lien subprime mortgage collateral rated by Standard & Poor's between the fourth quarter of 2005 and the fourth quarter of 2006. The 64 downgraded classes that were placed on CreditWatch before July 10, 2007, total approximately $700.9 million, which represents 0.12% in RMBS first-lien subprime mortgage collateral rated between the fourth quarter of 2005 and the fourth quarter of 2006. The combined impact of these 562 downgrades total approximately $6.39 billion in rated securities, or 1.13% of all RMBS first-lien subprime mortgage
collateral rated by Standard & Poor's between the fourth quarter of 2005 and the fourth quarter of 2006. The ratings associated with the downgraded classes, as a percentage of the total $6.39 billion in downgraded securities, are as follows:

Rating Percent
-- AA 0.07%
-- AA- 0.22%
-- A+ 1.66%
-- A 4.61%
-- A- 6.79%
-- BBB+ 14.01%
-- BBB 17.96%
-- BBB- 24.49%
-- BB+ 16.58%
-- BB 11.24%
-- BB- 1.06%
-- B 1.31%

And finally, Fitch is getting into the game:

Fitch Ratings-New York-12 July 2007: Following its monthly surveillance review, Fitch Ratings identified 170 U.S. subprime transactions among its $428 billion rated universe of subprime transactions as 'Under Analysis', indicating that Fitch will be issuing a rating action over the next several weeks. The total amount of bonds rated in the BBB category and below, which are the ones most likely to face rating actions, is $7.1 billion, representing 1.7% of Fitch's rated subprime portfolio.

by Anonymous on 7/12/2007 10:22:00 AM

Remember the big deal S&P announcement from Tuesday, announcing negative ratings watch on 612 classes of subprime securities? This is what S&P said on July 10:

The affected classes total approximately $12.078 billion in rated securities, which represents 2.13% of the $565.3 billion in U.S. RMBS rated by Standard & Poor's between the fourth quarter of 2005 and the fourth quarter of 2006.

This is what the revised version, dated July 11, says:

The affected classes total approximately $7.35 billion in rated securities, which represents 1.3% of the $565.3 billion in U.S. subprime RMBS rated by Standard & Poor's between the fourth quarter of 2005 and the fourth quarter of 2006.

No explanation is given for the difference between $12.01 billion and $7.35 billion. I'd certainly like to hear S&P verify that $7.35 billion is the original balance of these securities. If it's the current balance, then I'd like to know what the current balance is of the $565.3 billion it is being compared to. Perhaps an enterprising Real Reporter (you know who you are) will delve into this question.

by Calculated Risk on 7/12/2007 09:05:00 AM

The Census Bureau reported today for May 2007:

"in a goods and services deficit of $60.0 billion, compared with $58.7 billion in April"

Click on graph for larger image.

The red line is the trade deficit excluding petroleum products. (Blue is the total deficit, and black is the petroleum deficit).

The increase in the trade deficit was primarily due to oil imports, and mostly because of price increases.

Looking at the trade balance, excluding petroleum products, it appears the deficit peaked at about the same time as Mortgage Equity Withdrawal in the U.S. This is an interesting correlation (but not does imply causation). I had more on MEW vs. the trade deficit last month.

by Calculated Risk on 7/12/2007 08:55:00 AM

From Bloomberg: U.S. Foreclosures Jump 87 Percent as Lending Practices Tighten

There were 164,644 loan default notices, scheduled auctions and bank repossessions in June, led by filings in California, Florida, Ohio and Michigan that together accounted for half the total, according to RealtyTrac, a seller of foreclosure data.

The June foreclosure figure was 7 percent lower than that in May, when filings reached a 30-month high ...
...
An estimated 58 percent of properties in the foreclosure process are linked to borrowers with subprime loans, and RealtyTrac expects U.S. foreclosures to reach 1.8 million by year's end ...