[OpenSpending] Detroit and Chicago Default Risk as Measured by a Model

Wed Jul 24 21:47:24 UTC 2013

As many people on this list know, the City of Detroit declared bankruptcy last week.  Less well known is the fact that Moody’s – the major credit rating agency – downgraded the City of Chicago by three notches at about the same time.  Earlier this year, I used audited financial disclosures to estimate the risk of city bond defaults (which often accompany bankruptcies) in the state of California (the research was funded by a grant from the state, but its conclusions are mine and not those of any official agency). The goal was to see whether open data (collected from Comprehensive Annual Financial Reports that US governments typically file as PDFs) and open analytics can serve as an alternative to standard credit rating agency analysis.

The model is designed to calculate default probabilities – so higher scores are worse. Since 1940, the annual default rate for American cities was 0.10%. In 2012, 2 of 265 or 0.75% of California cities defaulted so that is the average score in the model. Scores substantially higher than 0.75% represent heightened credit risk.

A number of people have asked me how the scoring model would have treated Detroit and Chicago – which are in other US states.  Here is a my response.

A Google spreadsheet containing the model is available  <https://docs.google.com/spreadsheet/ccc?key=0AvdkMlz2NopEdGtKdnY2VTRPcmFLbmFPanNvc2pCdUE&usp=sharing&invite=CIDSpeED> here. It is modified version of our original model posted at  <http://www.publicsectorcredit.org/ca> http://www.publicsectorcredit.org/ca. I entered data from  <http://www.detroitmi.gov/Portals/0/docs/finance/CAFR/Final%202012%20Detroit%20Financial%20Statements.pdf> Detroit’s 2012 CAFR (which we found in mid-April and seems to have been published on December 28, 2012) and  <http://www.cityofchicago.org/content/dam/city/depts/fin/supp_info/CAFR/2012/CAFR_2012.pdf> Chicago’s 2012 CAFR.  Detroit’s probability score is 3.34% - worse than almost every California city in our survey.  Chicago’s score is also pretty bad: at 1.77% it is worse than Stockton, one of the two California cities that defaulted in 2012. 

The main driver of Detroit’s high DP score is its negative general fund balance.  The ratio of Detroit’s GF balance to GF expenditure is -27%.  As reported in our  <http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2258801> April working paper general fund exhaustion (very low or negative general fund balances) were associated with the Vallejo, Stockton and San Bernardino bankruptcies.  The Detroit situation provides further evidence that municipal bond investors and other stakeholders would benefit by monitoring this indicator.  

Although Chicago does not have a positive general fund balance, it had an annual general fund deficit and declining revenue – two of the other indicators that drive the score.  The Windy City also has a relatively high ratio of interest and pension costs to total governmental fund revenue. When these uncontrollable costs become relatively high, bankruptcy is harder to avoid.

Collecting, extracting and analyzing data from public financial disclosures can help us evaluate the credit risk of our local governments.  It is an important application of the OpenSpending concept.

Regards,

Marc Joffe

Public Sector Credit Solutions

P.S. – I know that some people on this list will be disappointed that the model is not in ODT format. Unfortunately, I used functionality that is not compatible with Libre Office. If anyone on the list would like to work with me to convert the workbook please contact me directly at marc[at]publicsectorcredit[dot]org.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/openspending/attachments/20130724/ae690096/attachment.html>