Include These 3 Factors In Your Bank C&I Credit Review to Better Predict Defaults

Predictive Analytics and Bank Credit Underwriting

What if there were a set of three easy to distinguish factors that, if all present, could predict the future performance of your customer and reduce the probability of default to half that of their respective industry?  Would you do anything differently?  Would you price lower?  Would you extend more credit?  Would you change your sales or marketing process at all to go after those accounts?

 

If the answer is no, then this article isn’t for you. However, if you are like us, and you are on a constant quest to find ways to make your underwriting more efficient and accurate, then these three factors are important to capture.

 

Most of the information in a credit package is noise, which not only takes time to gather and analyze, but may lead to the wrong conclusions. However, we recently reviewed a dataset collected by The Principal Financial Group that looked at 20 years of data from over 100,000 mid-sized businesses that have between 85 and 1,000 employees. The finding helps augment our earlier research that points to the fact that not only is cash flow important, but determining the quality of cash flow is even more important. The more free cash flow generated by core operations, the more likely the company is to make its debt obligations. In addition, we have found that other factors such as having an independent board of directors, having audited financials and producing consistent growth above 12% all point to lower than industry average probabilities of defaults and the greater usage of banking services.

 

With this new research we will start to track three items: 1) Top line revenue growth; 2) Head count growth; and, 3) the presence of some sort of formalized employee equity plan or profit sharing with employees. These three factors, when taken together over a three to five year period of time, point to a company that will perform in the top two percent of their respective industry.

 

One key finding from the data set is that not all growth is the same. A company that has huge, double digit revenue growth above 30%, for example, may be lighting the world on fire, but it also may point to a higher probability of default than average. Companies, the research reveals, grow in common patterns with many growing fast then plateauing. However, it is the consistent growth companies that point to higher quality management and not just fast growth industries.

 

Further to that point, it turns out that it is the superior management that can hire and train correctly, in an orderly fashion and continue to produce consistent growth that points to better corporate performance. This is the heart of the finding and makes sense, as many companies after experiencing rapid growth may hire a large number of employees, but the lack of applied management talent combined with weak infrastructure creates separate challenges on their own. Thus, it is the combination of growth and the ability to hire, and then grow some more that is a statistical marker for outperformance.  

 

Finally, management that is willing to share the profits or the upside of equity statistically tends to either be better managers or create an environment for more productive employees. We are not sure from the research what is the factor for causation, but it doesn’t really matter, as either way the presence of a profit sharing/equity plan combined with consistent revenue and employee growth should make your lending more accurate.

 

Many banks armed with this information may do nothing. However, banks that wish to experiment may start capturing growth rate, hiring rate and profit/equity sharing as additional inputs to loan write ups. Over time, this data will be extremely valuable as banks can then have the basis for adjusting capital, pricing and resources to make their lending more accurate and more efficient. 

Add new comment

Filtered HTML

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.