Accrued Savings of the Year 2000 Computer Date Problem
Leon Kappelman & Phil Scott
Anxious about trying to explain to your CEO the impending challenge of fixing the year 2000 computer date problem? Concerned that he or she might respond? with a "How did this happen?" or a "Who's responsible for this?" Rather than find yourself on the defensive, a better approach may be to explain the economic and business reasons for "Why it happened?" Not just in terms of the physical limitations on record sizes and computer memory 30 years ago, but with dollar approximations of what your organization saved in storage costs in the 30-odd years since those decisions were made.
In the late 50's and early 60's, when many of today's general accounting and core business systems were designed and programmed, a major constraint was the number of characters which could be recorded on an 80-column punched card. This physical limitation led us to record dates with only 2 digits representing the year to save precious space. This limitation was eliminated by disk storage when it became widely available in the early 60's.
Disk storage devices greatly enhanced the usefulness of computers because of their ability to access data randomly, but this new type of storage was very expensive. In 1996, as we upgrade and/or replace systems to accommodate the millennium change, the lease price of mainframe hard disk storage is under 10 cents per megabyte (MB) per month. But in 1963, 1 MB of mainframe storage cost about $175 a month to lease (that's about $10,600 per year per MB in 1995 dollars or nearly 1 million percent more than the 1996 cost). So, to save money we continued to use 2-digit year fields. And economic reasons were why we continued this space saving practice.
By 1972 the price of mainframe storage was down to $36 per MB per month, but that's still over $1,600 per year per MB in 1995 dollars (or nearly 150,000 percent more than current cost)! By 1983 the price was just under $22 per year per MB in 1995 dollars (but still 2000% more than the 1996 cost). So just how much have we saved our organizations over the years?
Depending on the industry and application, about 3% to 6% of the data stored in organizational databases is dates. Using 4-digit, rather than 2-digit year fields would have required 33% more storage for dates (MMDDYYYY versus MMDDYY). Using the lower 3% figure for date density, we conservatively estimate that 1% more disk storage would have been required through the years if we had stored all 4 digits of the year field.
One percent doesn't seem to be very much until you consider several important factors: the large volumes of data stored by organizations, the historically high cost of storage, and the rate of inflation. So for every gigabyte (GB) of organizational data stored, a 1% savings represents 10 MB. Designing systems to save 1% of storage costs since 1963 has saved between $1.2 and $2 million per GB (in 1995 dollars) depending on how quickly storage equipment was upgraded.
Then, there's the cost of capital, as well as the opportunity cost had those precious dollars been spent on storage rather than invested in other ventures. Using only a 10% internal rate of return (IRR) on capital, not an unreasonable figure considering the much higher 15% to 25% IRRs most companies try to earn on their capital, $16 to 24 million (in 1995 dollars) has been saved per GB of total storage in the 30 years from 1963 through 1992. This is a more realistic indication of the economic benefits of our date storage decisions.
In light of these savings, the $10 to $50 million estimated cost organizations may incur to correct the "millennium bug" pales in terms of benefits accrued. One organization, currently storing 12 terrabytes of mainframe data and about 700 GB 15 years ago, estimates that if their average storage over the 30 years from 1963 through 1992 were only 10 GB, they saved over $100 million in 1995 dollars. They estimate that their actual savings may be 5 to 10 times that size!
An economic analysis of the impact of the decision to not store those 2 digits of the date indicates that very good business judgement was indeed used! We have no reason to go "hat in hand" to our CEO to request the funds to accommodate the turn of century. We can present a strong business case for why this happened, and show that the cost to upgrade is reasonable in light of the many millions of dollars our decision saved our organizations during the past 35 years.
October 10, 1996
Leon A. Kappelman is an Associate Professor of Business Computer Information Systems Department and Associate Director of the Center of Quality and Productivity at the University of North Texas in Denton. He has dozens of journal articles and co-chairs the Year 2000 Working Group of the Society for Information Management (SIM). He can be reached at kapp@unt.edu and 817-565-3110.
Phil Scott, retired information systems consulting partner with Ernst & Young LLP, is now marketing director for Harvard, Porter & Associates, a firm dedicated solely to preparing its clients for the 21st century. He can be reached at 214-248-1750.
A version of this study now appears in the November 25th, 1996 issue of Computerworld.
(c) 1996 Leon A. Kappelman. All Rights Reserved
Copyright Notice: The not-for-profit distribution and use of this manuscript is authorized when done to raise awareness of, or to help solve, the year 2000 computer date problem, so long as credit for authorship is duly noted and the following ownership and contact information included: "(c) 1996 Leon A. Kappelman. All rights reserved. Box 13677, Denton, Texas 76203; Phone: 817-565-3110; Email: kapp@unt.edu" All other rights reserved.