Wednesday, December 08, 2010
the hidden message of the consumer credit statistical release
but there's a fly in the ointment, it would seem. one of the sticking points of the release to me was the massive growth in "student loans" -- running at over 80% year-on-year, an expansion of a very material $120bn. this is a stock, not a flow, and the idea that the amount of outstanding sallie mae loans had nearly doubled in a year is silly. so the question became what was driving the anomaly?
with the help of @Alea_ i managed to track down the difference. via firedoglake:
The Federal Family Education Loan program (FFEL) allows private financial institutions to provide students with loans, but the government assumes the risk of default, and pays the financial fees while the student attends college. This amounts to privatized gains combined with socialized loans. The CBO found that compared to the William D. Ford Federal Direct Loan program, an alternative system in which the government just directly provides students with loans, FFEL loans cost the government 10 to 20 percent more (PDF). Eliminating the unjustified subsidies and government financial backing for the FFEL program by expanding the existing direct loan program is projected to save $67 billion over the next decade.
Under the FFEL program, financial institutions like Sallie Mae, Bank of America, National Education Loan Network (NELNET) JPMorgan Chase, Wachovia, and Wells Fargo would originate these FFEL loans with students, and then sell them on the secondary credit market. In 2008, the credit market dried up, and the private lenders had nowhere to sell these government guaranteed loans. So, the government stepped in to buy up these loans and protect a program that was already a massively wasteful corporate boondoggle.
The bailout was authorized with HR 5715 Ensuring Continued Access to Student Loans Act (ECASLA). The bill allowed for the Department of Eduction to produce three different programs, the Loan Purchase Commitment Program, the Loan Participation Purchase Program, and a buyer-of-last-resort Asset-Backed Commercial Paper Conduit.
this purchase program -- which amounted to the department of education buying privately-originated student loans that were intended to be securitized but now could not be -- was radically expanded in 2009 and 2010, with a purchase amount target of about (you guessed it) $120bn. (the reporting of the actual purchases is here.)
in other words, what is being included in the g.19 as an expansion of student loans (and thereby consumer credit) is really in fact a bailout of several large banks and finance companies stuck with immovable loans.
so what does the picture of consumer credit look like with the influence of these asset purchases removed? as you can see, there's been very little letup in the pace of consumer deleveraging. the blue columns show the YoY change of unadjusted consumer credit as reported -- obviously a big collapse, but more recently an attenuated contraction that resembles recovery. the red columns show the same less the line items including student loans; clearly, there's not only no attenuation but year-on-year contraction in still picking up pace.
also of interest i think is the monthly sequential version of this same chart. we can easily see the seasonal waves of consumer credit expansion leading up to the holidays and contraction in the winter months. looking at the blue as-reported columns, late 2008 was extraordinary for its lack of holiday credit expansion, which of course shows up as the massive year-on-year contraction is was in the previous chart. ever since there's been contraction for the most part, but (as noted) attenuating through to the recent months which have shown three months of actual sequential consumer credit growth.
what's also apparent in the red columns, however, is how the ECASLA loan repurchase program has in 2010 completely distorted the true picture of consumer credit.
i've previously noted here the softening of leading economic indicators since april of 2010. i've also highlighted how the employment ratio derived of the household survey of the employment report has deteriorated markedly over roughly the same period (along with other LEI, such as CFNAI and the philly fed ADS). now we can also see that consumer debt deleveraging, after dissipating through the middle of 2009, has begun to reintensify powerfully in the last few months despite being "hidden from view" by the ECASLA bailout.
meanwhile, via paper economy (h/t @merrillmatter), we've also seen what appears to be a reintensification of deleveraging in the commercial paper market. and we're seeing evidence of renewed downturns in both commercial real estate pricing and residential house prices.
i've mentioned many times before steve keen and his three-party stock-flow model for banking economies (demonstrated here), one of the conclusions of which is that the change in aggregate demand flow is equal to the sum of the change in GDP flow plus the second derivative (or acceleration) in debt -- this is also known as the "credit impulse" formulation of aggregate demand. what the g.19 is giving evidence of is a renewed deceleration in consumer debt levels (that is, faster deleveraging) which should negatively impact aggregate demand. and that impact is, i expect, going hand-in-hand with contracting LEI and deterioration in EMRATIO.
it is also, i must imagine, behind the recent government move to initiate a second round of quantitative easing and an attempted extension of some economic stimulus measures. the administration must know as well as anyone that the rolloff of stimulus measures will be subtracting from GDP in coming quarters if they are not sustained; and that, if that rolloff is paired with a renewed deleveraging impulse, a powerful recession is all too likely. from a starting point economy that, as measured by per capita real final sales, hasn't really recovered at all, such an outcome could be politically, economically and spiritually devastating.
Nice to see you blogging again! (I don't do twitter).
Thanks for delving into this, I've been wondering about the 'federal government' line growth for a few G.19 releases also.
But how sure are you that ECASLA is actually boosting the top line non-revolving credit numbers versus just shifting the composition of its components? Looking at nonrevolving credit totals in the G.19 report it seems like there is an ongoing shift from finance companies (and to a lesser extent commercial banks) to federal government. That would show a noisy but roughly flat trend in total non-revolving credit (still a shift from the pre-2008 rapid growth).
By subtracting the portion held by government you are assuming that its increase came from "elsewhere", i.e., numbers that weren't previously accounted for in the G.19. That might be true but on my quick reading of all this isn't immediately obvious to me. (Let me know if I missed something).
But either way, a shift to direct government lending could enable a decision to lower lending standards and boost borrowing (and spending) activity. That may be partially sustained for as long as the government wants it that way, but would partially change the deleveraging story if true (at least if people take on that debt with the intent to service it vs planned abuse/defaults).
------ ------- ------
------ ------- ------
------ ------- ------
Post a Comment