Research shows that even with small merchants, new portfolio additions are bringing in less revenue than the ones that left. Offsetting this alarming trend is something called lifetime value management.
Negative substitution in acquiring is a concept in which an acquirer’s new merchants are added at price points lower than the price points of the merchants that are attriting, causing the portfolio average pricing to fall, other things equal. This is an easy thing to say and quite another thing to measure and analyze, but is clearly of increasing importance in U.S. acquiring.
It is perfectly clear in the upper mid-market and large-corporate spaces that negative substitution is a huge factor for acquirers, driven by the rate of decline in pricing in these segments in the U.S. The single most predictive variable for an acquirer’s portfolio average pricing relative to market in the mid-market and large- corporate segments is the average tenure of the merchants.
An acquirer whose portfolio skews to a newer merchant--a lower average tenure--is going to have lower average pricing no matter how skilled the acquirer is in selling value and no matter how strong the relationship is with the merchant. This is because the pricing in these segments has fallen so substantially over the past 3 to 5 years.
‘Ticking Time Bomb’
Historically, this phenomenon has been specific to the ultra-competitive mid-market and large-corporate space, but our hypothesis is that this concept is more central in the small-merchant market, as well. It makes sense that an acquirer would lose its higher-priced merchants and sign new merchants at relatively lower rates, but we set out to measure this phenomenon’s impact.
First Annapolis analyzed nearly 650,000 new and attrited merchants at 20 U.S. acquirers over the last several years, well distributed over types of acquirers. For 17 of 20 acquirers, the pricing for new merchants was below the pricing of the merchants the acquirers had lost, and the order of magnitude of the difference was sobering: new merchants were some 35% or 44 basis points below attrited merchants.
Some 12 of the 20 acquirers also had a negative volume variance, meaning that the volume produced by the new merchants was less than the volume produced by the attrited merchants. On average, volume for new merchants was 25% below attrited merchants for our study acquirers.
(This is not a zero sum game--some merchants go out of business, and newly established merchants skew to smaller average merchant sizes. However, it is true that some acquirers are the beneficiaries of competitive churn and tend to pick up the larger merchants that other acquirers are losing.)
So, if this were the only factor at play, it would be evidence of a ticking time bomb in acquiring, a looming unmitigated disaster. The higher the volume attrition at a given acquirer, the more significant the phenomenon. New merchants signed at pricing 35% less than attrited merchants and at 25% lower volumes with industry average revenue attrition in the low double digits is a fast train to a tight spot. The high-attrition acquirer would be, in the vernacular, hosed.
However, there is another critically important variable--lifetime value management.
It is a data-intensive analysis, and we do not have as much data as for the negative- substitution analysis, but we conducted a cohort analysis to see how acquirers managed pricing over time. Specifically, we looked at pricing at merchants signed in a base year and then measured pricing at the same merchants at base plus three years to see if pricing, broadly defined, had increase or decreased.
We concluded that on average, across the acquirers for whom we could develop the measure, pricing was 22% higher in year three than in the year the merchant was signed. Whereas the negative-substitution effect applies to the subset of merchants lost or signed in a given year--a sizable minority of merchants for acquirers--this increase in cohort pricing applies to retained merchants, which is a half-again larger population of merchants for nearly all acquirers. This is direct mitigation of the negative-substitution effect.
Caveats
The caveats and explanations are too numerous to count. In our negative- substitution analysis, we deal in very large samples, but in our cohort analysis, we are dealing in small samples.
The increases in pricing for a cohort definitely reflects a pervasive strategy in the U.S. acquiring market in which acquirers use “teaser rates” (though few acquirers intellectualize and describe their pricing like that)--low introductory pricing replaced by higher “go-to” pricing over time. (It is also true that there are more ethical and more dubious ways of implementing such a strategy in terms of fair dealing with and disclosure to merchants.)
However, it’s the wrong impression to think that this cohort phenomenon is solely due to acquirers jacking up pricing. The improvement in revenues reflects both pricing increases as well as cross-sale concepts as acquirers modify their offerings dynamically over time.
By its nature, our cohort analysis went back further in time than the negative- substitution analysis, and it is probably true that the pricing increases and cross- sale opportunities for the 2006 and 2007 cohorts, for example, are not the same as more recent cohorts.
For example, the heavy introduction of PCI compliance-, non-compliance-, and warranty-related offerings (which have been pretty uniformly introduced at high price points and substantial margins), were a one-time phenomenon in the context of this analysis.
In other words, the 2010 cohort will have been charged these prices in its early years if not its first year. Whereas PCI-related fees generated huge upward movements in all-in pricing in the 2006 cohort, which did not generally have PCI program pricing early on, the same will not be true in the 2010 cohort.
Therefore, PCI-related revenue will not offset the negative-substitution effect we see in the 2010 cohort. It is unclear to what degree acquirers will be constrained in pursuing this leg of portfolio performance over time, as revenue enhancements will not be available forever.
One of the most important caveats, however, is that acquirers in our sample had huge differences from one another in how successfully they increased revenues from merchants. Several of the acquirers we examined had declining revenues for their cohorts; several others had improvements well above the 22% average. For many of these acquirers, lifetime value management more than neutralized the negative-substitution effect, at least looking backwards in time.
It is a key question how long acquirers can keep up this shuffle--improve performance of the base even as the intersection of new and attrited merchants produces an unattractive combination. Acquirers have different levels of focus on lifetime value management and have made different levels of investment in related portfolio-management tactics.
To state the obvious, these investments and capabilities are likely to be decisive over the coming (not-too-many) years.
Chris Sanson is an analyst in the acquiring practice at First Annapolis. Marc Abbey is the managing partner for First Annapolis responsible for the acquiring practice.