|
|
 |
| What is a bank's Bankability Composite Score? |
 FIRST MAJOR STEP - 2007 Q1 to Q2 We wanted to provide accurate and vital information to our audience when they search for high yield bank accounts. The best of such Key Performance Indicators (KPI) that came to our minds was the Asset:Liability Ratio. The idea is very simple, in order for a bank to meet its obligations, its assets (what it owns) must always exceed its liabilities (what it owes). It even turned out that no one ever discussed that idea of looking at a bank's asset:liability ratio. So that gave us the big confidence boost. As we started mining the historical data for banks beyond the high yield accounts we tracked, we were able to pin down a strong correlation between the Asset:Liability Ratio and interest rates. SECOND MAJOR STEP - 2008 Q2 to Q4 Banks started failing in 2008 Q2 after the dramatic collapse of IndyMac Bank. The once widely regarded as a joke type of ideas we worked on is suddenly something of interest to almost everyone. People started talking about the mortgage backed loans, sub-prime lending, loan defaults, and financial weapon of mass destructions (FWMD's) everywhere. We wanted to expand our KPI's to capture as much of the banks financial status as possible. Our databases got populated with more and more bank data. We studied the banks' historical data particular to Assets, Liabilities, Income, Asset and Loan Breakdowns. The most notable result was the development of 1911(TM), our bank search engine. More information about it can be found on this thread. 1911 was developed to give greater transparency into how each FDIC insured banks is doing. There were a couple big companies at the time offering similar reports (some free, some fee-based) but we were the only one with our system designed with the capacity to report on historical data instead of a single quarter. 1911 also compares the specific banks to others in the same Asset Concentration Hierarchy (specialization). THIRD MAJOR STEP - 2009 Q1 & Q2 Many more banks failed in the second half of 2008. Washington Mutual failed; Wachovia was taken over by Wells Fargo; TARP was injecting capital into major banks; Ken Lewis was under fire for Merrill Lynch; John Thain was under fire for buying the $87k area rug and $1,400 trash can for his office! (It was also funny to us that we started hearing all the talks about 2000+ banks expected to fail and people talking about this list of failing banks...one that we have always been curious to see for ourselves.) By this time, we have been drilling deeper into the bank data. So the first result we released in 2009 was the Bank Report Cards: Top 10 Roll Call for 2008. There we emphasized on the additional KPI's: Normalized Income, Delinquent Assets, Real Estate Loans, and Asset:Liability Ratio. We also had increased our pool of talents with the new hires from Cal Berkeley. That resource boost allowed us to carry out one another key piece of original research work: Failed Bank Reports. We were invited to the Economics Department at San Francisco State University and gave a seminar on How to predict if a bank is going to fail? FOURTH MAJOR STEP - 2009 Q2 to NOW While working on the individual KPI's (directional), we wanted to simplify that further to a single composite metric. The end of this is the Bankability Composite Score - a numerical representation of a bank's financial status. This score is designed to factor in the individual KPI's (directional) as well as other key data that are deemed vital in determining how well a bank is performing. Further, this scoring algorithm must obey the properties of monotonicity, transitivity, and consistency; good thing that some of us here had some good academic trainings . The results of this Bankability Composite Score has been good thus far. We applied this scoring algorithm to the historical bank data on banks that failed between 1/1/08 to 03/31/09 and it correctly predicted: 63 out of the 70 bank failures* 2 quarters prior to their failure; 66 out of the 70 bank failures* 1 quarter prior to their failure. We are working on including the KPI's and scores for all the banks in 1911. This will be available in the very near future of our V2 release. RELEVANT ITEMS See the Bankability Composite Score FAQ Thread for answers to commonly asked questions. We have published the list of banks rated very poor (score < 20) via their scores based on 2009 Q1 data. We have published the failed bank reports for banks failed since 2008 (updated as data permits). *having a score of < 40.
|
