Default ratios, or the percentage of secondary loan prepayments that are attributable to defaults, can be considered a measurement of the health of small business in the U.S.  GLS, with default and borrower prepayment data supplied by Colson Services, has calculated default ratios for both SBA 7(a) and 504 loans since January, 2000.

Please note:  Unfortunately, the dataset we acquired does not supply the total dollar amount of outstanding loans, only the number and dollar amount of defaults and prepayments.  Therefore, we were unable to calculate actual default rates for each program, which would be very useful information.  However, we will continue to work to gather this information for future articles in the “CPR Report”.

Default Ratios, what do they mean?
The default ratio is calculated using the following formula: Defaults / (Defaults + Prepayments)

By definition, when the default ratio is increasing, defaults are increasing faster than borrower prepayments, suggesting a difficult business environment for small business, perhaps even recessionary conditions.  On the flip side, when the ratio is decreasing, either defaults are falling or borrower prepayments are outpacing defaults, each suggesting improving business conditions for small business.  Because the guaranteed investor is impacted equally by both types of prepayments, any decrease in overall SBA loan prepayments increases the return on their investments.

From the data, we notice three things:

1) The 7(a) data is more volatile than the 504 data.  This is not surprising since 504 borrowers tend to be higher quality credits then 7(a) borrowers.  The lower the credit quality, the more susceptible a company is to changes in economic circumstances.  
2) The 7(a) default ratio is greater than the 504 default ratio 77% of the time.  Again, this finding is not an epiphany since 504 loans have lower default rates than 7(a) loans.  
3) Both ratios have been trending higher since November, 2006.  The last time this happened was in late 2000 to early 2001, just before the 2001 recession.

Default Ratios and Economic Recessions
As mentioned previously, economic recessions should cause default ratios to rise.  Generally, economists believe that business defaults peak 12-18 months after a recession, as the damage they inflict finally takes its toll.  Fortunately for the US economy, there has only been one recession this decade (2001), so testing this hypothesis from our SBA data will be difficult, but not impossible.

The 2001 Recession
The 2001 recession lasted for approximately 5 quarters from April, 2000 to June, 2001.  While this time frame is at the beginning of our dataset, both sets of default ratios were increasing prior to the recession, albeit modestly.
With regard to 7(a) ratios, they generally increased throughout this 15 month period, but did not peak until February, 2003 or 20 months after the end of the recession.  This finding generally supports the general economic view on the timing of defaults caused by an economic recession.
The 504 experience during this timeframe is a bit more complicated.  While the 504 default ratio was generally increasing during this time period, the peak was reached in April, 2001 approximately two months before the end of the recession.  From that point, the default ratio generally decreased until bottoming out in mid-2005.  One possible explanation for this result is the dramatic growth in the 504 program this decade.
This lending growth would have kept the Weighted Average Loan Age (WALA) of all outstanding 504 loans low, as new originations outpaced run-off.  Since new loans tend to default and prepay at reduced rates when compared to older loans, the default rate would continue to stay low until the growth rate leveled off.  Unfortunately, without total outstanding balance data, this is only a theory.

Quarterly Default Ratios and US GDP Growth
To get a better sense of how economic growth and contraction can impact default ratios, we calculated quarterly default rates, to coincide with Gross Domestic Product (GDP) reporting periods. What was seen was that recently, GDP growth has begun to trend down, while the 7(a) default ratio has begun to trend higher.

Currently, economists place the likelihood of a recession in 2008 at between 50-70%.  If that is the case, we are likely to see the 7(a) default ratio continue to climb, as the economic damage takes its toll on small business.  Additionally, the current credit tightening in the US economy will lower the ability for SBA borrowers to refinance, which will also lead to a higher default ratio by decreasing the overall prepayment rate.

Looking at 504 default ratios versus GDP growth rates,  we saw the peak default ratio occurred during the 2001 recession.  As economic growth recovered, 504 default rates fell.  From the period of July, 2003 to April, 2007 504 default ratios stayed within a narrow band of approximately 9 – 13%.  Since then, the default ratio has exceeded 13% in all three reported quarters.  The average default ratio during the 2001 recession was 19.37%, suggesting that if we enter a period of recession in 2008, the default ratio has room to move higher.  Again, the WALA of all outstanding 504 loans will impact how high the ratio will go, but our dataset is too incomplete to make default predictions.  Quite simply, the growth in the 504 program clouds the default picture, when assessing default ratios.

However, that does not prevent us from offering some opinions as to “danger points” in default ratios.  In our opinion, a reading of 20% or greater on 7(a) default ratios and 15% or greater on 504 default ratios suggest economic weakness in these small business borrower groups.  

Conclusion
After reviewing the data, it seems that default ratios are impacted by economic conditions in the US economy.  With both ratios climbing, the data suggests that weak economic conditions in the US negatively impact both 7(a) and 504 borrowers.  One thing to note is that the impact has been greater on 7(a) loans as compared to 504 loans, likely due to the greater credit worthiness in the later group.