How Bad is The Bad Debt Situation?

Paul Krugman, the Princeton Economist and NY Times Columnist, recently wrote a piece called "After the Money's Gone" that I think is spot on and, echoes themes I have been talking about regarding the sub-prime debt crisis. So I'm going to steal his analogy from the piece. He writes about a hypothetical single bank, which is rumored to have made a big bad loan and could go under. All the depositors line up to get their funds. Thankfully, the Federal Reserve steps in, allows the bank (which has not really made a bad loan) to borrow as much as needed to forestall the short-term "liquidity crisis" and the bank is saved. Which begs the question, What happens when the bank did make the bad loan? The answer is, it depends if it eats up enough of the bank's capital to make the bank insolvent or unable to pay back all depositors even after getting back all the remaining value of its loans. In practice the Feds would put strict controls on a bank headed in this direction, well in advance of a full out implosion.

Right now banks are loath to lend to other banks, because they don't know how many bad loans the other guy (bank) has made. What they do know is that everyone has made some. So this brings us to the question: How bad is the bad debt situation really? While it truly appears to be the "not knowing" that is impacting inter-bank lending today, let's hope that knowing won't make it even worse. As you will see below, the Blogosphere is filled with articles on the proliferating bad debt problem and yesterday both Capital One and American Express released negative information about mounting credit card and home equity loan losses. It also appears that people who can no longer tap their homes for quick cash are turning to their credit cards, with credit card debt spiking 11.3% percent in the November, according to Federal Reserve numbers released recently (a similar phenomenon happened leading into the 2001 recession). While revolving credit tends to be about 36 to 37% of total consumer debt this has been ticking up as access to home equity has fallen.

So let's define some terms and then crack open the history books. First, let's briefly discuss the process by which banks deal with their inevitable bad loans (they make some in both good times and bad; it's part of the business). The first thing that happens is that a loan becomes "delinquent." According to the Federal Reserve Board of San Francisco, "delinquent loans and leases are those past due thirty days or more and still accruing interest as well as those in nonaccrual status." This is for purposes of the historical data they keep, which we will be examining to get a feel for how bad the bad debt issue really is. After 90 days or more the loan becomes non-performing and when a bank has concluded that it will not be able to collect on a non-performing loan, it "charges it off," this write-off of the value of the loan is subsequently revised down to reflect whatever value the bank is able to get in a recovery - sale of the underlying property in foreclosure. This is why historically, after a major recession, charge-offs can go negative as some of the charged off value is actually re-captured. The Federal Reserve Bank of San Francisco defines charge-offs as "the value of loans removed from the books and charged against loss reserves, measured net of recoveries as a percentage of average loans and annualized."

So where are we historically with regard to bank loan delinquencies and charge offs? Please note that the following do not include the massive "write-offs" by big banks that you have seen publicized in the newspapers recently, as the data available is only through Q3 2007 and these write-offs are a different animal than "bad loans." What the banks have "written down" are the carrying values of sub-prime and other mortgage-backed securities. These are not loans, but rather securities backed by the payments from loans. As securities I believe that they are not actually carried on bank's books in the loan category (someone who has expertise here correct me if I'm wrong) and won't show up in the data we will be discussing.

Presumably, banks only chose to keep the very best loans, in geographic markets they know well, on their books. I make this presumption because in recent years banks could have easily sold off all but their very best loans to the "secondary market" and gotten the liability off their books, while collecting riskless fees. It is therefore likely that the data of more recent vintages somewhat understates the severity of the bad debt problem, as we are looking at problems with the very best loans. O.K. so here goes.

Residential Loans

Delinquent residential bank loans peaked in Q3 1991 at 3.36% of total bank loans, and they have been under 3% since Q4 1992. Note that they only started collecting data in Q1 1991 - when a crisis was already in full swing - typical! At Q3 2007 delinquent residential loans were at 2.74% of all loans, higher than at any time since late 1993. Charge-offs of residential loans peaked in Q4 1992 at 0.27% of loans. Note that charge-offs peaked well after delinquencies (Q3 1991) and of course are much lower, as much of the value of a delinquent loan can usually be re-couped in a work out deal with the borrower or a foreclosure and sale of the underlying property. Interestingly, as of Q3 2007, charge-offs of residential loans were only a hair's breadth lower than the peak of the early 1990s crisis at 0.25% This argues that either despite a lower number of delinquencies, the ability to salvage value on a delinquent loan is much lower today than in the early 1990s, or something else is going on. Frankly I have only one speculation on what it could be, which I will share with anyone who cares....but if you are a bank examiner and have some insight please enlighten us.

Commercial Real Estate Loans

Commercial real estate delinquencies peaked at a ridiculous 12.07% of all loans in Q1 1991 (the first period for which data are available). Obviously, this was where the real problem was back in the old days, when a perverse tax incentive structure and bank corruption drove massive over-building of commercial real estate. In Q3 2007 commercial delinquencies were 1.94% of all loans, up from historic lows of 1.02% in Q1 2006, and above recessionary levels of Q3 and Q4 2001. But they were nowhere near the absurd levels of the early 1990s and the high levels that persisted through 1998. Commercial charge-offs peaked in Q3 1992 at 2.54%. The Q3 07 level of .10% is well off the 0.03% recent low set in Q4 of 2005, but still below the recession-driven peak level of 0.17% in Q4 2001.

Credit Cards
Credit card delinquency rates peaked at 5.45% in Q2 1991. They actually hit the 5% level again in Q3 2001 driven by the dot com/September 11 recession. Credit card delinquencies were 4.29% in Q3 2007, up significantly from the trough of 3.54% in Q4 2005, but still below the higher "normal" levels seen in the late 90s. Interestingly, credit card charge-offs' historic peak was 7.67% in Q1 2002. This was driven by the recession and likely by the much larger sub prime credit card market that existed versus the early 90s, coupled with a greater societal acceptance of bankruptcy as an option for expunging debt. As of Q3 2007, credit card charge-offs were 4.00%, well off the 2.98% low of Q1 2006, but still well within the recent historically "normal" range.

All Bank Loans

Delinquencies for all bank loans peaked at 6.15% of total loans in Q2 1991; the interim high of 2.75% in the dot com/September 11 recession wasn't even close. Delinquencies have risen from the 1.51% trough of Q1 and Q2 2006 to 2.12% at Q3 2007, but they are still in very good shape historically at this point.

So the very good news is that as in Krugman's example, so far the banks don't have a big enough bad loan problem to be insolvent and the current issue really looks like its the liquidity crisis and bank's willingness to lend, rather than their ability to lend. I will be watching these delinquency and charge off numbers closely, and I am very confident the street will as well, to see if the slowing economy puts significant upward pressure on these numbers and hence banks' ability to lend. The Q4 numbers should be out around March 1.

Please note that the data referred to above is for all US Commercial banks and does not include non-federally chartered S&Ls, credit unions etc.

Various Bad Debt Issues From The Blogosphere

Defaults on Insured Mortgages Hit Record

New Wave of Defaults in the Offing for 2008

Defaults Moving beyond Sub-Prime

Defaults To Rise As Economy Slows

Amount of Unpaid Credit Card Bills is Rising

Shares Battered By Possible Student-Loan Defaults

• Add to del.icio.us
• Digg this!
• Add to Technorati
• Add to Reddit
• Add to Newsvine