August 25, 2010
Todd Zywicki on Hawkins' Regulating on the Fringe
Posted by Christine Hurt

Overall, this is one of the most enlightening and insightful articles by a law professor on consumer credit that I have read in some time.  I especially like Jim’s methodology: rather than starting with the all-too-common assumption that fringe lending products are “bad” and then essentially launching an investigation to document all the ways in which it is bad, Jim actually disaggregates the claims about the various defects in fringe lending to consider their truth.  Certainly fringe lending products are expensive.  It also seems clear that borrowers who use fringe lending do so because they believe that they lack more attractive alternatives and that the alternatives are more expensive (such as bounced checks and overdraft fees).  (See my papers on auto title lending and payday lending for a summary of the theory and evidence).

Jim’s analysis focuses on a particular discrete question: even if those who use fringe lending believe themselves to be made better off, is fringe lending structured in such a manner so that it tends to produce economic distress?  The hypothesis is potentially plausible: fringe lending often has high costs (measured at least by APR measures) and consumers appear to often exhibit use patterns that seem baffling to upper middle-class law professors and bureaucrats who observe their behavior. 

A researcher could address this question in two ways.  First, one could examine the experience of fringe lending customers to see what they think of the products and how they use them.  As noted, fringe lending customers themselves overwhelmingly believe that access to fringe lending products improves their welfare and makes it easier for them to manage their finances.

But Hawkins’s paper takes a novel, and extremely useful approach: he asks not about the borrowers but examines the loan products themselves.  And he argues—compellingly in my view—that the structure of fringe lending products are unlikely to be a substantial source of financial distress even for low-income borrowers.

His reasoning is straightforward: although each of these loans are often expensive, the overall risk exposure that they present for borrowers is inherently constrained.  All of these loan products are capped either by their small size (as with payday loans) or by the value of any collateral posted for the loan (as with auto title lending or pawnshops).  Certainly, a borrower can confront hardship from losing personal property from an unredeemed pawn or a repeatedly-revolved payday loan.  But it is highly unlikely that in all but the rarest cases the total amounts of these loans can be ruinous.

Supporting Hawkins’s hypothesis, Elliehausen found that 40 percent of bankruptcy filers who list payday loans on their bankruptcy schedules have only one payday loan and the average amount of those payday loans was $350, which was 1.3 percent of the bankrupt’s unsecured debt and the cost of servicing the debt was only about 2.4 percent of the bankrupt’s average net monthly income. Moreover, as Hawkins notes, those who find a connection between financial distress and fringe lending often mistake causation for correlation: it may be that many of those who turn to fringe lending do so as a last resort, thus these loans may be a manifestation of financial distress, not a cause.  Thus, for example, absent access to fringe loans it is plausible that studies of bankrupts would reveal an unusually high level of bounced checks or utility shut-offs in the period preceding bankruptcy.

But this is only the tip of the iceberg.  As I note in my papers referenced above, in evaluating the costs and benefits of fringe lending and its contribution to financial distress, it is also necessary to consider financial distress that could be caused by the absence of payday lending.  For example, Morgan and Strain find that when Georgia banned payday lending, chapter 7 bankruptcy filings increased.  Donald R. Morgan and Michael R. Strain, Payday Holiday: How Households Fare after Payday Credit Bans (Fed. Res. Bank of NY Staff Report no. 309) (Feb. 2008).  By depriving consumers of access to short-term loans that can bridge liquidity problems, banning short-term loans can convert a short-term liquidity problem into a solvency problem.  Similarly, auto title loans can enable a borrower to liquify wealth contained in household durables—again, making it impossible to access that wealth can convert a short-term liquidity problem into a solvency and bankruptcy problem.

Finally, banning fringe lending products can induce borrowers to use credit cards and cash advances on credit cards that can be more likely to precipitate financial breakdown.  For most consumers credit cards are a preferred source of personal credit relative to payday and title loans.  Those who use payday or title loans, however, do so because credit cards are either unavailable or would turn out to be more expensive than fringe products.  Thus, for example, when those who would otherwise prefer to use payday loans are forced instead to use credit cards, the result is a tendency to have higher levels of delinquency and to trigger behavior-based fees more frequently, resulting in an effective cost that is higher than fringe lending.  Thus again, eliminating access to fringe products is more likely to accelerate rather than prevent financial breakdown.

And, of course, that doesn’t even account for illegal lending, which tends to be more prominent in places that restrict access to fringe lending.

On the central point of the paper, therefore, I think that Hawkins has made a compelling case on his central thesis.

In the interests of completeness I note a few other passing observations:


The paper contains a very interesting discussion of how to measure the concept of financial distress—noting in particular that bankruptcy is not necessarily the ideal manner.  Hawkins appears to accept the claim that the “best measure of a typical middle class family’s financial distress is its debt-to-income ratio.”  But that contention is simply untenable.  As I have noted previously, far from being the “best measure” of financial distress, debt-to-income ratio is almost certainly one of the worst measures.  Debt-to-income measure measures a stock variable—debt—against a flow variable—income.  It is thus no more sensible than the alternative would be—the current debt to assets ratio.  For a household, the best measure of financial distress is probably the debt-service ratio (or alternatively the financial obligations ratio), which is analogous to the concept of “equity insolvency” in corporate law.  Alternatively, balance-sheet insolvency (net wealth) is also plausible, but probably less useful for reasons I have discussed elsewhere.

The reason, of course, is that debt-to-income ratio takes no account of either the interest rate on debt or the maturity period of debt.  Thus, the broad secular drop in interest rates—which is largely responsible for the growth in the stock of debt for American households over the past 25 years—is lost in the debt-to-income ratio.

One could raise other issues about other elements of Hawkins’s article (such as his acceptance of the thesis that credit cards are generally a unique cause of financial distress), but many of those issues are sidelights to his central thesis about the purported connection between fringe lending and financial distress.  On that central point his argument is persuasive and his cautions about regulation are well-advised.

This is an important paper for those results alone.  But this is an important paper for a larger reason—Hawkins’s approach here is an excellent model for future scholars investigating the welfare effects of fringe lending and its regulation.  One hopes that it will spur similarly enlightening research in the future.

Junior Scholars | Bookmark

TrackBacks (0)

TrackBack URL for this entry:

Links to weblogs that reference Todd Zywicki on Hawkins' Regulating on the Fringe:

Recent Comments
Popular Threads
Search The Glom
The Glom on Twitter
Archives by Topic
Archives by Date
January 2019
Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    
Miscellaneous Links