I just don’t get it. How can running a system in a leveraged environment cost more than operating the exact same software in your own bank?
One of my roles at Cornerstone Advisors is negotiating contracts for our clients. In that role, I review and negotiate terms for both in-house and service bureau offerings. Every vendor and every type of delivery are contained in the myriad of contracts I see. Contracts for both bank and credit union clients with assets as low as $1 billion and reaching to $30 billion continue to show the same troubling pattern: delivery of service bureau services compared to the same system run in-house will be higher. In some cases, the five year cost is as much as 60 percent higher. Why is that so? If someone can ’splain that to me, I wish they would.
I first observed and commented on this issue in a prior GonzoBanker article titled “Core Outsourcers are Thrashing Their Best Customers.” It just seems like a professional organization running a large data center leveraging hardware, software and staff should be more cost efficient than a stand-alone center. But let me digress…
The Cornerstone Report: Benchmarks and Best Practices for Mid-Size Banks 2005 will be published in the next few weeks. Reduced spending for core systems is still occurring. Below is core system spending in the last four surveys expressed as the ratio of core spending to average assets. The trend of core spending by banks is decidedly downward. Not only is it downward but it is trending at an increasing rate.
While core spending has been steadily declining, total spending for information technology has been frozen at 26 basis points to average assets in every study. Each year spending for new technology initiatives has been funded by the reductions in core spending. Sure makes a person wonder why the core vendors would want to spend much money making their core products better.
So what? Well, Bucko, a bank that is spending more on core systems than its peers will have less money to spend on the really important initiatives that can differentiate its products and services and drive profits to the bottom line. In making this statement, I am assuming that the inherent nature of bankers to only spend 26 basis points for technology will continue into the future.
Now let’s get back to why the core vendors’ pricing models yield the upside-down results. When we compare one delivery cost versus another, it is the total cost over a period of time, usually five years. It has been our practice to build these models to demonstrate what happens to costs when a bank grows. Occasionally an attractive offer from an aggressively priced proposal will yield serious bottom line erosion if a merger of equals occurs. Comparing the models will clearly show which proposal or delivery method is the most cost effective.
Every attempt is made to make the comparison as “apples to apples” as possible. In-house pricing will have the appropriate number of staff allocated for operations, etc. In-house models also reflect depreciation of the original hardware and license costs. For in-house spending, the disparity is much larger after the initial investment in hardware, conversion, software and professional services is run off. This disparity is even more surprising since our models generally only cover five years, the usual amortization period for initial licensing, etc.
Let me give you a few recent examples of the differences:
Now let’s make some assumptions:
- One computer is capable of running more than one bank.
- Larger computers are less expensive per computing unit than smaller computers.
- The same number of operations staff can operate computer(s) for more than one bank. (OK, at some point additional staff will have to be added, but it is not a linear function.)
- Floor space, networks, etc., can all be shared among one or more banks running at the same site.
- Banks will continue to spend a total of 26 basis points of average assets for information technology.
- Core system spending will continue to decline over time.
A comparison of the assumptions against the table above it indicates either the assumptions or the table are wrong. It just seems backwards to me. Running an in-house system for one bank should be more expensive for a bank because it is unable to leverage staff, hardware, software, floor space, networks, etc.
For most banks caught in this trap, it means they make up the difference in other areas of technology spending. Perhaps they put off infrastructure upgrades, undertake fewer projects or go cheap on their networks. Can they be high performers?
I have challenged every vendor when I spot the disparity. Most often, the two alternatives come from two different internal vendor organizations, and they simply shrug their shoulders. I have never seen a vendor change the relationship between the two delivery methods. They may change the amounts, but service bureau five-year pricing will always be more expensive over five years.
It looks as if this would drive more banks to in-house solutions rather than service bureau solutions. Perhaps this is the ulterior motive of the vendors. I don’t think so, since most vendor CFOs really like five to eight year service bureau contracts. The revenue stream is nice and even and stretches over a long period of time.
Therefore my only conclusion is that this is a really fat cash cow for the vendors, and as long as banks are willing to step up to the trough, vendors are going to keep dishing out the service bureau chowder. So what should banks do when faced with this issue?
First of all, regardless of delivery, the best practice bank will make sure its total spending for technology is in line with peers. Second, it will make sure core spending is also in line with peers so that it is not necessary to rob Peter to pay Paul. Banks should take a hard look at alternatives for delivery. If they are dramatically different, banks should get in their vendors’ faces and have them explain why this is true. The vendors should be shown the trend of core spending so that banks can be sure they are being provided with the best product using the delivery that supports their strategies and executive comfort levels.
When the vendors start pushing back, best practice banks will tell them they read GonzoBanker and they “aren’t going to take it any more!”