|
In a word, “No.” It pays to know which merchants in a portfolio are attriting and why. In the weak retail sales environment we have today, a merchant acquirer has to know which merchants are attiring, how much revenue they produce and what is the margin that is lost. Working out effective strategies for keeping higher margin merchants can be the essential element of success.
In my work underwriting portfolios, I see lots of attrition estimates that are generalized industry standards or averages such as 1.5% or 2.0% per month attrition. This works fine for a small portfolio (defined as less than 1,000 merchants) where all the merchants are pretty much the same in terms of processing volume, revenue contribution and other characteristics. If the portfolio is larger than about 1,000 merchants, there’s bound to be a heterogeneous distribution of merchants – some big, some small, some high margin, some low margin. Each class of merchant has to be viewed separately and the attrition rate on each evaluated separately as well.
I looked at the performance of a single large portfolio of almost 18,000 merchants over an 18 month period from January 2008 through June 2009 to evaluate attrition and see the effect of uneven attrition across merchant classes. I excluded sales additions to this portfolio during this period. Here’s the basic data:
This looks like a pretty typical portfolio at first glance – large and therefore filled with a wide variety of merchants across a wide geographic and MCC code distribution. The average merchant processed $7,693/month in the first three months of 2008 making this predominantly a mom-and-pop type portfolio on the whole.
18 months later, the portfolio has shrunk and each merchant processed $6,180/month, a drop of almost 20% overall. Clearly a portfolio of small merchants became a portfolio of smaller merchants over time.
The interesting thing to note is the divergence of attrition rates – the merchant count attrited at -1.84% per month on average while residual revenue attrited at 2.46% per month, indicating that something is up.
The data gets more interesting when you think about the characteristics of the merchants that were lost over the 18 month period. The results of the lost merchants are in the third column on the right. The average processing volume of each lost merchant was 40% higher than the overall portfolio in the beginning and 75% higher than the ending processing volume. Further, the average number of transactions was higher by half and the average ticket was higher by $9 or 15%.
The net residual per merchant lost was 60% higher than the remaining merchants. One interesting note is that the basis point spread on the lost merchants was lower than the retained merchants indicating that they were priced a bit thinner on average in the beginning. If these merchants were taken by competitors (the most likely case) then their pricing went down from where it was in this portfolio – further evidence of margin compression in the marketplace. Price cutting appeals to merchants with higher volumes for the obvious reason that it reduces the absolute dollar cost of processing expense. With small volume merchants, the basis point spread is less of an issue since a downward revision doesn’t change the actual processing expense all that much. For example, if you can save a merchant $50 per month, it’s worth the pain to switch processors, but not for a savings of $10 per month.
There are several messages to draw from the data:
Large volume merchants attrit faster than smaller ones. Small reductions in processing price or per communication charges can yield a significant dollar savings to the merchant and give the low cost service provider a distinct advantage.
Price spread itself is less the issue for the merchant than the absolute dollar cost of processing services. Merchants care how much they are going to spend on processing and less about what the percentage price is.
Small merchants can be priced as small merchants when they are young. As they grow, they are vulnerable to poaching by the competition if their pricing is not adjusted for the size merchant they have become. Acquirers need to watch their small merchants and be prepared to adjust pricing to market rates applicable to the merchant’s current status or someone else will.
It’s not enough to talk about average attrition rates, acquirers must treat each class of merchant as a separate distinct business unit with their own behavioral characteristics. 
|
