This note was originally published April 25, 2013 at 09:23 in Financials
Long in the Tooth, but OK for Now ...
Both seasonally adjusted and non-seasonally adjusted initial claims posted sharp improvement in the latest week. In the past we've shown how initial claims are the best leading indicator for forward delinquency metrics at credit card lenders, on a 13-week lead/lag. Clearly, card results were better than expected in the first quarter of 2013 and the claims data we're seeing currently suggests further progress in second quarter from Discover Financial Services (DFS) and Capital One (COF), and American Express (AXP) to a lesser extent.
We've argued in the past that renormalized delinquency levels, i.e. reaching DQ levels consistent with 2005/2006 averages, would herald an end to reserve release tailwinds, and, by extension, usher in earnings growth headwinds. This ongoing positive trend in claims, however, coupled with Krugman's "Jobless Trap" - the segment of the population now perpetually long-term unemployed - has enabled a continued fall in DQ rates since this former core subprime cohort is now "non-underwriteable".
This curious confluence of positive selection (vs. adverse), fast turning assets and secular credit tailwinds has pushed operators to take down coverage further than we would have thought prudent or possible. While this sets the stage for ongoing improvement in the short/intermediate term, we continue to fret about the long-term implications of making all-time lower low DQ rates in a low-growth, highly competitive asset class. Auto is perhaps an even more glaring example, as we noticed in the weekend paper a local credit union offering 0.99% on terms up to 48 months. How that will even cover normalized credit losses in this newly emerging environment of falling used car collateral values I have no idea. But for now, everything's good.
Prior to revision, initial jobless claims fell 13k to 339k from 352k week-over-week, as the prior week's number was revised up by 3k to 355k.
The headline (unrevised) number shows claims were lower by 16k week-over-week. Meanwhile, the 4-week rolling average of seasonally-adjusted claims fell -4.5k week-over-week to 358k.
The 4-week rolling average of non-seasonally adjusted claims, which we consider a more accurate representation of the underlying labor market trend, was -6.3% lower year-over-year, which is a sequential improvement versus the previous week's year-over-year change of -3.8%