The Crisis of 2008 - Betting the Bank. [Did we also bet the credit union?]

Remember the good old days when marketers were the business people you couldn’t trust?  They would say anything to make a sale.  Then came Enron, and accountants were not to be trusted.  They would make your books say anything - for the right price.  Both of them now pale in comparison to bankers, who since the financial crisis of 2008 have been seen as wrecking the entire economy – not just of the United States, but of the world.  Should such shame be aimed just at bankers or did credit unions also take on “excessive risks” taking us ever closer to financial Armageddon?

Did Both Banks and Credit Unions take on Excessive Risk?

To help answer that question we start by looking at some basic statistics.  First let’s look at how much risk appears to have been borne by the two insurance funds:  FDIC and NCUSIF.  Both websites post the names of institutions closed since the beginning of 2009.  In the graph below, we compare the insured deposits of closed financial institutions reported the quarter before the closing to all insured deposits in all insured institutions.  For our purposes here, we decided to exclude the closed corporate credit union in the fourth quarter of 2010.  It is also worth noting that most of the risk that shows up for credit unions in these graphs came with just one very large credit union closure (Kern Central Credit Union) in the 1st quarter of 2010.  The higher the percentage, the higher the risk to the insurance fund those failed institutions are.  It can be seen that FDIC insured institutions by this measure appear far riskier than NCUSIF insured institutions.

But these measures are almost certainly comparing apples and oranges.  The insurance funds operate differently.  The markets are in fact rather different.  Banks and thrifts are profit and value maximizing organizations.  Institutions cannot be enticed to take on a failing institution unless it is in the best interest of their shareholders.  Therefore, most bank closures were followed by asset acquisition.  In the case of credit unions, they are cooperatives.  And NCUSIF is set up in more of a cooperative environment[i].  All institutions suffer if institutions fail, since insurance fund shortfalls are made up by all surviving institutions.  Therefore, it is in the best interest of healthy institutions to acquire failing institutions before they fail.  This tendency has been noted in research as the likely reason why credit union mergers don’t tend to benefit the acquiring firm.

As a partial adjustment for the cooperative nature of credit union deposit insurance, we rework the above graph to include in the closed credit unions’ insured deposits the deposits of all merged credit unions approved for the same quarter, where the reason given in the Monthly Activity Report for the merger was “Poor Financial Condition”.  The restated percentages are displayed in the graph below.  Note that even allowing for these adjustments, credit unions have risked less as a percentage of total insured deposits than banks[ii].  The one exception appears to be early in the crisis.  But as happened in the 1980s, this might simply be NCUSIF working to resolve potential problems as quickly as possible.  Note in this instance that means that troubled credit unions were merged with healthy ones.

Recent research has pointed to stock options in managerial compensation packages as one reason banks seemed to have taken on excessive risks.  It is argued that managers of for profit (and for that matter publicly traded) banks are given incentives to take on more risks.  The evidence seems to suggest that when managers are given stock options, they do tend to take on more risks in an effort to make more money[iii].  But most of these incentives do not hold for credit unions.

Credit union managers by contrast can’t be given stock based compensation.  Although their salaries may be based on performance, what is important to members is not necessarily what is important to the shareholders of the bank.  For instance, members want lower loan rates and higher deposit rates.  That would mean that credit union members want managers to produce lower net interest margins, whereas bank shareholders want managers to produce higher net interest margins.  A similar logic could be used to come to a similar conclusion concerning non-interest income and non-interest expense. 

Did Both Banks and Credit Unions take on Risky Assets?

Because credit union managers are answerable to all members (customers) they cannot take advantage of one set of loan customers at the expense of another.  And as was shown in the previous paragraph, credit unions are not seeking to maximize profit margins anyway.  As such they do not benefit in an obvious way from taking on riskier loans.  In research I presented last year at a conference, I showed that when compared to banks, credit unions differed in loan risk preferred.  Although credit unions do tend to seek return on assets (so that they can maintain sufficient capital as the institution grows), higher ROA was associated with lower levels of bad debt expense.  Banks on the other hand tended to take on riskier loans in an attempt to increase their ROA.  Bad debt expense as a percent of assets was positively correlated with ROA for them.  The data set I used in this research ran through 2007 – just before the crash.

The fact that banks took more risks on the quality of loans can be observed from their allowance for loan losses, and net charge-offs (net of recoveries) as a percent of total loans.  The graph below shows the allowances made for loan losses as a percent of total loans and leases.  It is interesting with this measure how far above credit unions the banks started out in 2007.  As the crisis developed the allowance for loan losses grew.  And the two lines have yet to return to pre-crisis levels.

But to me the most graphic example of the difference in loan quality is the net charge offs.  These measures were essentially equal in first quarter of 2007.  Both charged off approximately 0.11% of the value of their loan portfolios that quarter.  The charge-offs on the part of the banks just exploded to 0.74% in 4^th quarter 2009, whereas credit unions had only grown to 0.33%.  In third quarter 2010, banks were down to 0.58%, but credit unions are still below half of that at 0.27%.

Some may argue that banks and credit unions have very different loan portfolios, and therefore defy comparison.  As I have pointed out in earlier writings, the two are beginning to converge in the financial products and services offered.  The research is beginning to suggest that the two seek different levels of risk.  Based on the last two financial crises, credit unions appear to prefer fewer risky assets.  Although part of that risk preference is due to a difference in managerial compensation, it is more likely that the real difference is due to a difference in the preferences of the owners.  With a stockholder run bank, the stockholder earns more as the income on a highly leveraged and regulated industry increases.  With a credit union, the owner prefers a fair return to debt (both deposits and loans are debt).  Debt markets tend to be less risk seeking. 

---

[i] A really good academic piece compared the risk to the taxpayer of both of these insurance funds during the last financial crisis (in the 1980s).  It was very complimentary of the NCUSIF. Kane, Edward J; Hendershott, Robert; 1996.  “The federal deposit insurance fund that didn't put a bite on U.S. taxpayers.”  Journal of Banking and Finance, Vol. 20, Issue 8 (September), pp. 1305-1327.

[ii] I am not completely convinced that the NCUA is disclosing all failed institutions in their list of “closed” institutions.  In the 2011 Statistical Abstract of the United States, the NCUA reports 31 failed institutions in 2009.  On their website, they list only 15 “closed” institutions for that same year.  I can’t reconcile the other 16 institutions.  There were 50 mergers approved by NCUA in 2009 due to “Poor Financial Performance” so that would not account for the 16 missing institutions. If I have a reader from NCUA, I would appreciate some insight here.

[iii] Two papers currently in print make this case.  One is an academic paper:  Zhiyong Dong, Cong Wang, Fei Xie, 2010. “Do executive stock options induce excessive risk taking?” Journal of Banking & Finance, Vol. 34, Iss. 10 (October); pp. 2518-2529.  

     The other paper is more of a professional – easier to read paper:  Jian Cai, Kent Cherny, Todd Milbourn, 2010. “Compensation and Risk Incentives in Banking and Finance.”Federal Reserve Bank of Cleveland. Economic Commentary. Cleveland: Sep 14, 2010. , Iss. 2010-13; pg. 1, 5 pgs.