Erroneous Excel: Part I
Abrigo is in the business of convincing bankers to replace Excel spreadsheets with software designed specifically for estimating their ALLL and by extension, managing loan portfolio risk. But what’s wrong with Excel? By making it easy for almost anyone to do the kind of number crunching once reserved for accountants, it has achieved status as, some statisticians suggest, the most used business program ever.
“Microsoft Excel is one of the greatest, most powerful, most important software applications of all time,” writes James Kwak for The Baseline Scenario. “It provides enormous capacity to do quantitative analysis, letting you do anything from statistical analyses of databases with hundreds of thousands of records to complex estimation tools with user-friendly front ends. And unlike traditional statistical programs, it provides an intuitive interface that lets you see what happens to the data as you manipulate them.”
Stuart Leung of Salesforce adds: “Spreadsheet software provides an easy-to-use data entry platform for students and professionals, particularly in the marketing, sales, business, and finance fields.”
Still, for all its usefulness, Excel presents problems to bankers, in particular, when it is used for the most critical quarterly calculation, the ALLL.Numerous studies over the past ten years have uncovered errors in spreadsheets used every day to manage trillions of dollars. Fact finders have determined that nearly nine out of ten spreadsheets (88 percent to be exact) contain errors. So why isn’t there a mass exodus from software so flawed? Because almost none of those errors is caused by the software itself.
“A majority of errors reported were caused by human error—meaning they could have been completely avoidable mistakes,” states Leung.
If you have used Excel or other spreadsheet software, you are likely familiar with how easy it is to alter a calculation inadvertently, copy and paste incorrect data, improperly use any one of the automated functions, etc. Ease of use is Excel’s strong suit. It can also be its weakness in an institution that relies on an Excel without a quality control system of thorough and ongoing accuracy checks. And the problem is exacerbated when spreadsheets are passed from individual to individual, from department to department, for updates and additions.
“In programming, we have learned to follow strict development disciplines to eliminate most errors,” cites an often-quoted study, “What We Know About Spreadsheet Errors,” by Raymond R. Panko of the University of Hawaii. “Surveys of spreadsheet developers indicate that spreadsheet creation, in contrast, is informal, and few organizations have comprehensive policies for spreadsheet development. . . . there has long been ample evidence that errors in spreadsheets are pandemic”
Pandemic? Too strong a term? Maybe not, considering how many years spreadsheet software has permeated a multitude of business, research and education applications. Panko observes that concerns over error rates in spreadsheets are as old as spreadsheets themselves, because as user-developed systems, “end-users seldom implement the disciplines long known to be necessary within the development of programming languages.”
Even a protocol of ongoing checks of balances might not solve the inherent accuracy issue.
While Excel “. . . is reasonably robust, the spreadsheets that people create with Excel are incredibly fragile. There is no way to trace where your data come from, there’s no audit trail (so you can overtype numbers and not know it), and there’s no easy way to test spreadsheets, for starters,” writes Kwak. “The biggest problem is that anyone can create Excel spreadsheets — badly.”
But what’s unique about Excel? Doesn’t any software run the risk of a high error rate since all software is created by people, the beings that cause Excel spreadsheet errors?
“While all software breaks occasionally, Excel spreadsheets break all the time. But they don’t tell you when they break; they just give you the wrong number,” Kwak says.
Kwak admits that software developers make mistakes. Yet the development languages and tools make software easier to test, to find and correct errors. Plus, the process begins, presumably, with someone who possesses the aptitude for building software.
Research suggests that efficiencies provided by tools like Excel often lead to increased job responsibilities and requirements (think multitasking). And the cognitive ability that allows us to be “fast and flexible” also lends itself to a mistake every now and then. To err is human.
The biggest problem with Excel is that institutions fail to account for its associated risks.
“It’s all become so complex and it’s handled in such a slapdash manner that no one is really on top of it anymore,” wrote Tim Worstall in an article for Forbes Magazine. “Microsoft’s Excel Might Be The Most Dangerous Software On The Planet.”
It seems hopeless, a problem without any reasonable, cost-effective solution. But that statement may itself be erroneous.
88% of spreadsheets have errors, Jeremy Olshan, MarketWatch, 4/30/13
Sorry, Your Spreadsheet Has Errors, Stuart Leung, Salesforce, Forbes, 9/13/14
The Importance of Excel, James Kwak, The Baseline Scenario, 2/9/13
What We Know About Spreadsheet Errors, Raymond R. Panko, University of Hawaii College of Business Administration, Spring 1998, revised May 2008
Microsoft’s Excel Might Be The Most Dangerous Software On The Planet, Tim Worstall, Forbes, 2/13/13