Innovation Necessity and a Tale of Two Industries

May 29, 2013

Milos Dunjic – CTO, Cardis International


This paper compares the need for technological transformation in two segments of the financial industry, both exhibiting similar phases of development but at different points in time. It presents the experiences from the rapid transformation of the electronic trading industry and relating these to the current challenges and pressures experienced in the evolving electronic payment industry. The paper offers a clear path forward for enabling substantial growth potential in electronic payment transaction volumes.

Equity Trading Industry

Before I made a transition to the electronic payments industry five years ago, I had spent significant time in capital markets. Here I managed technology teams responsible for delivering electronic stock trading systems. The core trading software built by one of those teams still fuels one of the major stock exchanges in North America. Since 2000, the average exploitation lifespan of the trading engine system in that particular stock exchange was less than four years before someone ordered its replacement with the new one.

Why was this mad tempo imposed on everybody inside the organization? In order to understand that, one has to realize that the stock exchange revenue comes from

1. Trades they generate during the business day (exchanges are not paid by the number of orders they process)

2. Selling of market data feeds

3. Listing fees that companies would pay for the privilege and opportunity to list on the exchange

These three business drivers above are very much interconnected and interdependent. It is a simple business model based on liquidity of the order flow and matching at its core. Early stock exchanges had been based on manual trading floors. These were replaced with automated computerized trading engine systems in the late 80s and early 90s. Initially and for certain period of time the ‘order to trade ratio’ on electronic exchanges was very low and very efficient – 2:1, 3:1, … It meant that for every two, three or at most five orders entered, the system usually generated a single trade.

The rate of the order flow was pretty predictable, because it took time for human traders to make their trading decisions. Sudden huge swings in the order flow volumes were extremely rare. Frequent upgrades of the electronic trading engine systems were not required. The functionality, reliability and stability of the system were the most important goals and exchanges used very expensive systems with built-in reliability. The cost of trading was not a major concern.

Fast-forward to the early 2000s. With the invention and introduction of algorithmic trading (i.e. using computers to generate order flow, rather than human traders) the situation significantly changed. Stock exchanges now had to cope at short notice with the rapid increases in order flow. Order to trade ratios have been raised to 50:1 even 100:1, since those computer-generated orders did not materialize into a significantly higher volume of trades. It meant that the exchanges were facing the need for constant upgrades to computer hardware without the guarantee that the additional revenue from additional trades would cover the costs of those upgrades. They had to keep upgrading just to survive, because the new computer order generators were now hooked to multiple exchanges and able to quickly redirect the order flow to another place if the particular exchange could not cope. Losing the order flow for the stock exchanges posed a nightmare scenario.

The only way to cope with this was by reducing the transaction processing costs. It meant a gradual abandoning of the expensive hardware. It also meant a switch (as quickly as possible) to cheap commodity Linux platforms.

“Faster, Better, Cheaper” suddenly became the new mantra preached by the capital markets’ business and technology executives. In less than two years every major exchange in the world had basically replaced their expensive legacy trading and market data distribution platforms with much faster and considerably cheaper computing platforms. Of course this involved some high implementation risks, since new commodity platforms didn’t come with built-in reliability of Tandem or IBM mainframes. The exchanges then had to rapidly innovate to achieve the same reliability as before. Previous innovation procrastinators became the biggest innovators out of pure necessity.

Electronic Payments Industry

Let’s take a look at what’s happening within the present card payments industry. One may wonder how the previous capital markets experience would relate to what’s happening in card payments. The challenges faced by the main players in these two distinct industries, may indeed look quite different. However, the ultimate resolution of the challenges in both cases requires properly addressing the transaction economics, i.e. by reducing the cost per transaction.

The electronic payment industry is today facing several major challenges:

1. Mounting pressures from continued efforts in trying to move the low value transactions from cash to electronic payments, despite already significant investments and marketing of the contactless payment infrastructure

2. Mounting pressures from the regulators which are imposing significant reduction / limits on the amount that the payment industry is able to charge merchants for each transaction

3. Mounting competitive pressures from a host of new entrants with innovative solutions – some offering lower transaction processing fees as loss leaders.

Current Card Transaction Economics and Cash Replacement Opportunity/Challenge

The Payment Industry is finding itself at a similar juncture today as the electronic trading community did about a decade and a half ago. The major payment schemes (like Visa, MasterCard) control the technology standards, operating rules, transaction routing / switching, etc. Payment card transactions originating at the merchant POS terminals ride the rails of the infrastructure that was designed in the 1970s for higher value credit card transactions.

The processing of each card transaction carries a direct variable cost independent of the transaction amount. This is due to transaction authorization processing, transaction clearing and settlement processes with extensive reconciliations between parties, fraud detection systems / processes.

These costs incurred by card issuers and merchant acquirers are too high for LVP – there is either a loss for issuers/acquirers, or fees are too high for the retailers. Numerous reports and statistics confirm the fact that consumers want to use their smart card or smart phone to pay for everyday small purchases – something which is becoming increasingly possible with contactless and NFC products. However, it simply being a different form factor riding the same debit/credit card rails, contactless and NFC solutions are facing the same economic issues, as they do not change the cost structure. The zero sum game issue between service providers and retailers remains intact and this is stalling wide acceptance. Changing the interchange fee level does not solve the issue, as it merely redistributes costs and the loss in the system.

Regulatory Pressures

Card issuers and payment schemes are also facing increasing pressure from regulatory bodies imposing significant restrictions on the level of the interchange fees on debit and credit card transactions they are allowed to charge merchant acquirers (and ultimately the merchants). This, of course, limits the payment industry’s ability to cross-subsidize losses from processing low value transactions by imposing higher fees for high value transactions. Increased volumes of low value payments processed through the current infrastructure, which is heavily regulated and unable to shift the cost toward merchants, would make the issuers’ losses even more pronounced.

New Entrants

The regulators are opening the market to new entrants, especially in the LVP segment. The new players include online giants and telcos that are entering the electronic payments business. Companies like Square, iZettle, Intuit and others are introducing mobile POS solutions mainly targeting small merchants with very aggressive processing fees, often below their cost. These rapid changes put significant competitive pressure on the traditional electronic payment industry players – issuers, acquirers and payment schemes alike – that must find innovative ways to protect their positions in the market.

Critical moment for innovation

The payment industry, as in the case of the capital markets industry earlier in the century, is rapidly moving towards its own ‘critical moment’ for innovation. As the electronic trading industry had to address the costs of its own transaction processing in order to survive the introduction of algorithmic / high-frequency trading, so the electronic payments industry must address the cost of processing small-value transactions. The market is ready for innovative low-value-payment solutions that provide for positive disruption in the payment processing value chain by fundamentally changing the way they are processed. Without such a change, the many opportunities offered by e-payment systems may be lost.