Large Bank Technology: 142 Paradigms

This is a wrap-up of the three part series that started with As the Giant Banks Get Bigger, Their Technology Might Be Getting Weaker and includes quite valuable reader responses.

After 221 blogs, with a few duds thrown in, I was pleased to get so much feedback on this series. The responses included corrections, confirmations, cheers, requests for clarification, declarations of the truth, and happily, no insults. The responders were obviously guys in the trenches who know more about their turf than I do. I won't name them because if they wanted the exposure, they would have posted their comments on the blog platform for all to see.There are 142 large banks in the U.S. I define "large" as $8 billion in assets and above. It's not chipped in granite, but it is based on one IT characteristic that they all share-"We're different." The DOJ recently identified large as the top 50. Some vendors won't touch any bank if it's below the top 20. Where do you draw the line?

I concede to the idea that IT in the top 142 is different, but when one sees what comes out of the pipeline, it's all the same. In some cases, I can point to small banks that can deliver better results to a customer because of integration.

Large banks cannot claim integration. Their IT architecture is a patchwork quilt. That's not a cheap shot; they know it. They started to build their "systems" before packaged software was invented. As large banks grew, they fed their homegrown systems, some acquired best-of-breed applications, and some acquired integrated systems. So today, large banks use mostly homegrown software; 37 of the 142 big banks acquired integrated core products from FIS, Metavante, Fiserv and Jack Henry & Associates; some best-of-breed vendors include CSC-Hogan, AFS and Shaw Systems Associates; some even go so far back that names are meaningless today. For example: MSA, McCormack Dodge, D&B Software, and GEAC. Say what you will about the absence of elegance in this environment, but what CIOs will tell you is, "It works every day."

In one of my three blogs in this series, I made the statement that there hasn't been a large bank core conversion in decades. I was corrected, and not by the New York bank that did it recently or the vendor who sold it, but by an astute observer-an SVP at a Massachusetts S&L. Thanks, Russell, I need all the help I can get.

Only because I'm a numbers guy, I pay attention to ratios. So I will offer these ratios. Large banks use legacy technologies which are so outdated that they require "historians" to keep the applications current. I believe that's an expensive proposition. One of my ratios proves it. Large banks require 20 percent of the bank's total operating cost to drive technology. In a mid-tier and small bank that ratio is 7 percent to 10 percent. A blog reader from one of the four giants asked me to clarify a statement I made about spending money. The man was absolutely correct. I butchered the statement, and now I offer it to you, corrected.

I first said it to a group of bankers about 20 years ago: "The easiest IT problem to solve is one that requires writing a check." No one in the room liked that statement. I still believe it today. Would you rather write a check for a few million dollars, or spend months explaining to millions of customers why the bank couldn't post transactions for a week?

The point I'm trying to make is that there isn't any magic that will correct IT glitches with a wand. Spend the money for a real solution and you'll save more than the amount of the check.