SQ published in January 2003 its annual industry rankings of the large acquirers, and The Nilson Report will be next. Many industry observers pay a lot of attention to these rankings. In this article, I want to provide some interpretations and give some thoughts on the meaning and value of these rankings and discuss other rankings that would provide valuable information to all of us.

For many years, The Nilson Report published the most complete annual rankings of industry acquirers. Others also have published annual rankings, but none has focused on all of the acquirers, including ISOs. Several years ago, Paul Green began to publish his GSQ list. Both of these lists are extremely useful. However, they are based on different information, some of which may not be 100% accurate. The result has been some confusion to many.

Both publishers ask major acquirers to report information on the honor system. In other words, the numbers published in these lists are self-reported and are as accurate as the information provided. No audits are performed on any of the reported data. While I believe that the majority of acquirers report the best information available to them, I learned long ago that some people count funny in this industry. For example, merchant locations or "sites" mean different things to different people.

I seriously doubt, for example, that the number of "sites" has grown from 4,265,358 in 2001 to 5,345,690 in 2002 (1,080,332 net new sites) or 25% as reported in the January 2003 GSQ. I am not doubting the accuracy of what was tabulated by GSQ but the accuracy of some of the information submitted! Some of the reporting acquirers count sites in a creative manner - sort of like Enron and Xerox were reporting their revenues awhile back.

Just for fun, divide the reported volume by the number of sites for some of the acquirers reported in either list. You will find portfolios with less than $1,300 per month per site! It's hard for many small businesses to survive that long. Methinks that way too many sites may have been reported by some acquirers. The result is that the number of sites reported isn't very reliable. Too bad for our entire industry that the honor system isn't taken more seriously by some of the reporting entities.

In gathering information, the two publishers ask very different questions, and the reported rankings are therefore different. The GSQ rankings are based upon the volume of transactions processed for the calendar year. The Nilson Report rankings are based upon the estimated annual volume of merchant contracts owned by the reporting acquirer at year end.

The GSQ rankings for 2002, for example, show that First Data Merchant Services (number 2) processed $162.7 billion of volume. This does not necessarily mean that FDC owns 100% of the merchant contracts for which it processes. It might just provide authorization and capture services, for example. Indeed, some of the acquirers in these rankings function as "servicers only," at least for a portion of their reported volume.

Most of the 31 acquirers ranked in the GSQ outsource 100% of their authorization, capture and settlement processing. However, all 31 report the volume as their "processing" volume. This means that some of the volume is double counted in the GSQ rankings. The word "process" means different things to different people.

Some acquirers outsource all of their transaction processing, all of their back-end processing and all of their sales. Yet they call themselves "processors." In fact, the only acquirers who might process all of their transactions are First Data, Concord EFS, Paymentech and Global Payments. (But then, all four of these true "end-to-end" processors outsource a very large portion of their sales to ISOs. There is nary a single 100% vertically integrated acquirer in this country today!)

Another point is that Bank of America, Certegy and the Sovereign Bank/First Data Alliance are not included in the 2002 GSQ rankings. Together these acquirers probably processed about $40 billion in 2002 (based upon published statistics). This brings the total processing volume of the top 34 acquirers to more than $1.2 trillion. If this is more than 100% of what Visa and MasterCard report, the above explanations partially account for the variance.

I am not trying to find fault with the GSQ rankings here. For goodness' sake, it is almost impossible for the stockholders of public acquirers to obtain accurate information about processing statistics. The public acquirers work hard to keep their processing data very private.

So in my view, despite the points I am making here, I think GSQ is doing a good job of reporting the best data the staff can garner from the acquirers. It is valuable to have the actual processing volumes reported - even if the data is not perfect.

Nilson asks for different data: the run rate of "owned" contracts as of Dec. 31, 2002. This is a whole new ballgame. Run rate is a snapshot of the portfolio at a specific point in time.

Again, this is useful information, but it is important to understand that "run rate" is a lot different than "actual processed volume" when reporting industry rankings.

For example, it has been reported that TransFirst aquired a several-billion-dollar subset of Bank of America's portfolio during 2002. It is likely that this several billion dollars of portfolio will be reported in the TransFirst numbers and subtracted from the Bank of America numbers in Nilson's 2002 rankings regardless of who actually processed the volume.

As indicated, the word "processing" means different things to different people. But the word "ownership" means different things to different people as well. Who really owns a merchant contract?

According to Visa and MasterCard rules, only the member bank has association rights and responsibilities relating to merchant contracts. That is why a member bank must be a party to every merchant contract. Member banks enter into contracts with acquirers and transfer their rights to the revenue stream after their fees and losses are paid.

In turn, thousands of alliance and ISO contracts with member banks define the "beneficial ownership rights" of the alliances and ISOs in the merchant contracts. Still more tens of thousands of contracts define "residual rights" between acquirers and ISOs and salespeople.

Merchant contracts are a lot like little oil wells. Lots of different parties have beneficial ownership in the profits of the contracts. So what does ownership mean as reported in the Nilson rankings?

Nilson contacts the major acquirers, and the major acquirers report their "owned" contracts. But if you are an ISO who is paying one of the outsource processors for services, you think you own the contract, right?

Twelve of the top 34 acquirers are non-banks, and 10 of these outsource most or all of their processing to the four processors mentioned above, who also are listed as the top eight acquirers. Most of the 22 member banks in this ranking also process with these same four processors.

Does this mean there is some double counting in the volume? I think that Nilson works very hard to make sure double counting does not occur within the alliances of First Data and U.S. Bank, for example. However, some of the reported data cannot be verified, and some errors slip into the mix.

To summarize, the rankings of acquirers provide useful information, but some of the information is double counted or reported by people who count funny. Hence, many of the totals reported are off and should not be taken as gospel.

As valuable as these rankings are to those of us who care about such things, I wish we could rank the players by different metrics. I think it is fair to say that most businesspeople are focused on creating stable, growing, profitable businesses.

Our industry has been one of the most attractive industries to public investors because of its recurring revenue model and the stable growth year over year in the top-line revenues.

As I indicated in a previous article in The Green Sheet, the acquiring industry generates approximately $3.5 billion of actual revenue annually in this country. Of this, approximately $1 billion goes for transaction processing and the other $2.5 billion goes for the rest of the acquiring functions.

Wouldn't it be great to see a ranking of the acquirers and processors by the amount of the $3.5 billion pie that each earns? Wouldn't it be great to see a ranking of how much of the $3.5 billion pie goes to the ISOs/sales organizations of the acquirers?

It would be interesting to see rankings based upon the "quality" (i.e., profitability) of the acquired volume in addition to the "quantity" of volume processed. What is the net revenue generated by the processed volume? What is the gross margin after ISO and sales commission payments? What is the EBITDA generated per dollar processed? What is the bottom line per dollar processed?

My point is that processing one $10 billion customer for 2 basis points of gross margin after commissions is qualitatively much different than processing $10 billion of volume for a portfolio of 50,000 merchants at 30 basis points. I believe that both of these extremes are realities in portfolios reported in recent rankings.

Another factor is that some acquirers process huge volumes of transactions for which they report "ownership" but in fact earn only the 5- to 12-cent transaction fee they are charging the ISO. These "acquirers" pay the rest of the revenues to the true owner of the beneficial interest in the contract - the ISO. Volumes processed in these cases are not relevant to the quality of the business of the "acquirer."

Some industry analysts confuse transaction processing with acquiring because all of the true processors except Vital are also acquirers. False conclusions are too often reached about the issues of commodity pricing and scale factors because the lines are now so blurred between true processors and the acquiring portion of their portfolios.

The key metrics I would love to see measured to clear up this confusion are:

• Net revenue
• Processing costs plus infrastructure depreciation
• ISO payments and commission expenses
• Gross margin after commissions
• EBITDA
• Amortization of intangible assets (mostly purchased portfolios)
• Net profit before taxes

These rankings would help all industry observers better understand who is creating what value for their companies with the transactions and volume they process or "own." If these qualitative factors gained visibility, a lot of arguments about scale and commodity would go away and more people would truly understand the power of the small-acquirer and ISO models. This level of reporting is not going to happen in my lifetime, but wouldn't we all love to see it!