Professor Cole's Finance Blog

On March 4, 2009, the U.S. Treasury Department announced the Obama administration’s Making Home Affordable program, which included a “comprehensive $75 billion Home Affordable Modification Program now known simply as “HAMP.” HAMP was supposed to help 3 to 4 million at-risk homeowners avoid foreclosure by permanently modifying their mortgages in such a way as to reduce their monthly mortgage payment to a “sustainable amount.”

As the program reaches its one year anniversary, we must ask about how responsibly the Administration is spending $75 billion of our taxpayer dollars. Treasury released the most recent numbers on Feb. 16, which covers the program results through January 2010. According to this report, about 1.3 million borrowers have been extended an offer for a trial modification, 1.0 million borrowers have started a trial modification, of which 830 thousand are still active, 60 thousand have been canceled, and 117 thousand permanent modifications were started.

So, after 11 months, the program has “reached” more than a million of the targeted 3 to 4 million at-risk homeowners, but has produced permanent loan modifications for only a hundred thousand. That said, the January results were a vast improvement over November and December 2009, when only 31 thousand and 66 thousand permanent modifications had been started, respectively. So it appears that the program is beginning to hit its stride, but it will need to continue this progress if it is to make a real dent in the problem because the problem continues to grow.

On Feb. 19, 2010, the Mortgage Bankers Association released results from its National Delinquency Survey for the fourth quarter of 2009. This report showed that the combined percentage of loans in foreclosure or at least one payment past due reached 15.02%, the highest level on record, up from 14.41% in the third quarter of 2009. In the fourth quarter, the percentage of loans in foreclosure was 4.58% and the percentage past due was 10.44%, up from 4.47% and 9.94%, respectively, in the third quarter. This translates into an additional 335 thousand new delinquencies and an additional 60 thousand foreclosures during the fourth quarter.

On Feb. 23, 2010, First American CoreLogic released a report in which it estimated that, in the fourth quarter of 2009, 11.3 million residential properties representing 24% of all properties with mortgages were “under water,” meaning that the borrower had negative equity because the value of the property had fallen to where it was less than the outstanding mortgage balance. This was an increase of approximately 600 thousand properties from the third quarter of 2009. More than 10 percent of properties are estimated to be underwater by more than 25% of loan value. The report goes on to demonstrate a strong correlation between negative equity and foreclosure.

So we see that delinquencies are growing by 100 thousand per month, foreclosures by 20 thousand per month, and underwater mortgages by 200 thousand per month. Against this backdrop, HAMP has managed a total of only 120 thousand permanent modifications, although 50 thousand of these took place during January. In order to reach its goal of 3 million permanent modifications by 2012, HAMP will need to average 125 thousand permanent modifications per month. Good luck with that.

Unaddressed thus far is the knotty issue of re-defaults. Some researchers estimate that re-default rates on modified mortgages will exceed 50% within 12 months of modifications.

Posted by m1rac01 at 11:47 AM | Permalink | Comments (0) | TrackBacks (0)

Yesterday, the FDIC released its report on the condition of the U.S. banking industry based upon data from December 31, 2009. This report paints a dismal picture that bodes poorly for the future performance on the U.S. economy.

Forty-five banks failed during the fourth quarter of 2009, raising the total for the year to 140—the largest annual total since 1992. The number of “problem banks”—those that bank regulators consider to be in danger of failing—rose to 702 at the end of December 2009, up from 552 at the end of September 2009, and from 252 at the end of December 2008. The assets of problem banks rose to $403 billion at the end of December 2009 up from $346 billion at the end of September 2009 and from $159 billion at the end of December 2008.  Both the number and assets of problem banks are at the highest levels since 1993. Clearly, these numbers don’t include any of the four largest banks, in spite of stress tests indicating that these banks are insolvent. Too-Big-To-Fail is alive and well at the FDIC and other bank regulatory agencies.

Asset quality continued to deteriorate, as measured by loan losses, past-due and nonperforming loans and foreclosed real estate. Loan losses rose for the twelfth consecutive quarter to a 2.89 percent annual rate of net charge-offs, up from 1.95 percent from a year earlier. The 2.89 figure is the highest rate on record during the 26 years banks have reported this item.

Non-current loans and leases (past due 90+ days or in nonaccrual status) rose to $391 billion, or 5.36% of all loans and leases—the highest percentage in the 26 years that banks have reported these data. This includes the banking crisis years of 1985 – 1992, when more than 1,000 banks failed, which gives us an indication of the number of bank failures to expect during the coming years. Noncurrent loans were concentrated in the real estate sector, both residential and commercial: 9.3 percent of residential mortgages were non-current and 15.9 percent of construction & development loans were non-current. Surprisingly, only 1.8 percent of home equity loans were non-current, indicating that homeowners are staying current on second liens while defaulting on first liens. In total, banks hold $1.92 trillion in residential mortgages, $661 billion in home equity lines and $451 billion in construction & development loans. They also hold an additional $1.09 trillion in commercial real estate loans.

An additional $140 billion in loans and leases were past due by 30 – 89 days, so that the combined value of past due and nonaccrual loans is $531 billion, or 7.28 percent of total loans and leases. By sector, 3.2 percent of residential mortgages and 2.6 percent of construction & development loans were past due 30 – 89 days.

Other real estate owned (OREO), which consists primarily of real estate seized through foreclosure on previously delinquent loans, rose to $41.4 billion in December from $37.2 billion in the September. When combined with the $531 billion in bad loans, the total book value of nonperforming assets was $572 billion or 7.85% of total loans and leases. Banks hold reserves against loan losses of 3.12 percent of total loans and leases—the highest level in the history of the FDIC.

Total assets declined for the fourth consecutive quarter to $13.1 trillion, down 5.3% from a year earlier. Total loans and leases declined for the sixth consecutive quarter to $7.29 trillion down from $7.88 trillion a year earlier—a 7.5 percent year-over year decline that is the largest since the 1940s.

Earnings during the fourth quarter of 2009 were $914 million, which translates into a miserable return on assets of only 0.01 percent, but a dramatic improvement over the $37 billion loss recorde during the fourth quarter of 2008. The percentage of banks that were losing money rose to 29.5% in December, up from 24.8 percent in December 2008 and 12.1 percent in December 2007.

Posted by m1rac01 at 01:35 PM | Permalink | Comments (0) | TrackBacks (0)

Before I’ve even had the chance to see Tim Burton’s new rendition of the classic “Alice in Wonderland,” I have had the chance to peer down the rabbit’s hole, not into Wonderland, but into Obama-land.

Obama-land is a fantasy world where up is down and down is up; where a trillion-dollar health care bill is projected to reduce the natiotal debt; and where 2 million jobs were “saved or created” during the same period of time when almost 4 million U.S. workers lost their jobs. Yes, Obama-land is a magical place that is "long on fantasy" and "short on facts," just like Tim Burton's fantasy world.

This morning, President Obama and Vice-President Biden took a victory lap—celebrating the one-year anniversary of the Stimulus Act of 2009. In his prepared remarks, the president claimed that “the Recovery Act is responsible for the jobs of about 2 million Americans who would otherwise be unemployed” and that “it is largely thanks to the Recovery Act that as second depression is no longer a possibility.” Of course, the President provides no facts to support these statements, but instead refers to unnamed “nonpartisan economists across the spectrum.” Certainly, far-left-leaning Paul Krugman of the New York Times might be one of these economists, but I know of no right-leaning, or even center-leaning, economists who possibly might fit this description. Like the 2 million “saved jobs,” this is just another Obama fantasy. Into the "rabbit hole" we go. 

Let’s look at the facts instead of fantasy: According to the Bureau of Labor Statistics (“BLS”), 3.9 million U.S. workers have lost their jobs since Obama took office on January 20, 2009, and another million unemployed workers have given up looking for a job, thereby falling out of the official labor force. The number of chronic unemployed—those out of work for more than 26 weeks—has more than doubled from 2.7 million in January 2009 to 6.3 million in January 2010. Currently, there are 14.8 million unemployed U.S. workers, and another 10.5 million who have given up looking for work or been forced to take on part-time work when the would like to work full-time.

On Friday, the Bureau of Labor Statistics (“BLS”) released its monthly report “The Employment Situation,” which provides a summary and statistics on employment and unemployment during the previous month. The  headlines numbers for January were a loss of 20,000 payroll jobs but a decline in the unemployment rate from 10.0% in December 2009 to 9.7% in January 2010. Most people will scratch their head and ask how unemployment can decline when 20,000 jobs were lost.

The simple answer is that these contradictory numbers come from two different surveys conducted by the BLS each month. The payroll employment number comes from the “establishment survey,” which is a monthly survey of large businesses that employ most workers. The unemployment number comes from the “household survey,” which is a smaller monthly survey of households. The larger establishment survey is more accurate because of a larger sample size, but it cannot provide information on unemployment because it is a survey of employers, not workers. The smaller household survey is less accurate because of a smaller sample size, but enables the BLS to estimate how many workers are unemployed.

During January, the household survey showed that unemployment declined by 430,000 workers and that employment actually rose by an astounding 541,000 workers at the same time that the establishment survey showed employment falling by 20,000. One reason that the two surveys can be going in opposite directions is coverage. The establishment survey only covers workers on the payrolls of large established businesses, so it misses the self-employed, employees of small and newly established firms. This is critically important because research has shown that about two-thirds of employment growth takes place at small businesses, especially at newly established entrepreneurial firms. The half-million new jobs identified by the household survey may indicate that the economy has finally turned the corner and that small firms are on a hiring spree. I say “may” because of the large sampling error inherent in the design of the household survey. Just one month earlier, the household survey for December 2009 showed employment falling by 589,000, so we are still looking at a net loss of 48,000 jobs over two months. We can be hopeful, however, and await the February report for more information on which way the trend is really moving.

Now, more about the January report. Twenty thousand workers lose their jobs, but the BLS report states that “payroll employment was essentially unchanged.” Are these bureaucrats tone deaf or what? Also, the employment reports for November and December were revised, with November going from +4,000 to +64,000, but December going from -85,000 to -150,000. These revisions emphasize that the initial numbers reported each month are preliminary because not all establishments get their surveys in by the monthly deadlines. Instead, they get counted later and then the BLS releases revised numbers.

Also in January, construction lost 75,000 jobs but temporary workers gained 52,000 jobs. Now that’s change we can believe in, isn’t it! Of course, the federal government grew, adding 33,000 jobs. The average work week increased by 0.1 hour to 33.9  hours. This is another good sign, but increasing the number of hours worked by existing employees enables firms to avoid hiring new workers, keeping unemployment at elevated rates for a longer period. Average hourly earnings rose by 4 cents, up 2% over the past year.

U-6, the broadest measure of unemployment provided by the BLS, declined from 17.3% in Dec to 16.5% in Jan. on a seasonally adjusted basis. The 9.7% number is the U-3 measure of unemployment, which only includes unemployed members of the labor force. To be counted in the labor force, a worker has to actively look for a job during the previous four weeks; otherwise, they “drop out” of the labor force. Also excluded are employees working part-time even though they want to work full time. U-6 includes these discouraged and underemployed workers, but even U-6 excluded discouraged workers who have not actively looked for a job for more than one year. Some estimate that unemployment actually exceeds 20% when these long-term discouraged workers are counted as unemployed.

All of these numbers are “seasonally adjusted” by BLS staff using a statistical model that “smooths” out the month-to-month variations in employment numbers. For example, we know that employment jumps every June when high-school and college students look for summer jobs and then declines every September when they go back to school. The problem is that these “seasonal adjustments” allow for considerable mischief should unscrupulous or politically motivated staffers tweak their models to make things look better than the real numbers show. For example, U-6 actually rose from 17.1% in December to 18.0% in January on an unadjusted basis, so that the January unadjusted U-6 was of 9% higher than the seasonally adjusted U-6. Unadjusted, the U-3 measure of the unemployment rate would be 10.5%, not 9.7%. Such a large adjustment factor raised questions about the validity of the seasonally adjusted series. The January employment summary notifies us that the BLS just updated its seasonal adjustment model and its net birth/death model that is used to account for its undercoverage of small and newly established firms. Hmm.

Also in the January report is the result of its annual benchmarking process, whereby it reconciles the results from the establishment survey with actually unemployment records, which are not available on a timely basis, but are far more accurate. As a result of this process, payroll employment for March 2009 was revised downward by 930,000 and December 2009 employment was revised downward by 1,363,000!!! The numbers as originally reported and as revised cleary show that the establishment survey was consistently biased in the direction of higher employment. They appear in Table A of the BLS report.

The January report also notifies us that the household survey also was adjusted to reflect new population estimates from the U.S. Census and that this adjustment renders comparisons across time unreliable. So, how are we to ascertain if the newly announced decline in unemployment rate from 10.0% in Dec 2009 to 9.7% in Jan 2010 is actually an improvement, as it would appear.

Posted by m1rac01 at 10:40 AM | Permalink | Comments (1) | TrackBacks (0)