Horror Stories
EuSpRIG Horror Stories
Spreadsheet mistakes – news stories
Public reports of spreadsheet errors have been sought out on behalf of EuSpRIG by Patrick O’Beirne of Systems Modelling for many years. There are very many reports of spreadsheet related errors and they seem to appear in the global media at a fairly consistent rate.
These stories illustrate common problems that occur with the uncontrolled use of spreadsheets. In many cases, we identify the area of risk involved and then say how we think the problem might have been avoided.
Stories are identified by those who kindly collated and sorted them:
POB: Patrick O’Beirne, Eusprig chair
FH: Felienne Hermans (winner of the 2011 David Chadwick student prize and now an assistant professor at Delft University of Technology).
NS: Tie Cheng, a EuSpRIG committee member.
MPC: Mary Pat Campbell, an actuary, trainer, and a member of the EuSpRIG Discussion group
| Identifier: | POB2001 |
| Title: | Data not controlled, 16000 UK Covid-19 test results lost for a week |
| Source: | https://www.bbc.co.uk/news/technology-54423988 |
| Release Date: | 08 October 2020 |
| Risk: | Lives put at risk because the contact-tracing process had been delayed |
| Discrepancy: | 16,000 test cases in a week |
Excel: Why using Microsoft’s tool caused Covid-19 results to be lost
“The badly thought-out use of Microsoft’s Excel software was the reason nearly 16,000 coronavirus cases went unreported in England. [The labs] filed their [result logs] results in the form of text-based lists – known as CSV files – without issue. PHE had set up an automatic process to pull this data together into Excel templates so that it could then be uploaded to a central system. The problem is that [Public Health England] PHE’s own developers picked an old file format to do this – known as XLS. As a consequence, each template could handle only about 65,000 rows of data rather than the one million-plus rows that Excel is actually capable of. And since each test result created several rows of data, in practice it meant that each template was limited to about 1,400 cases. When that total was reached, further cases were simply left off. To handle the problem, PHE is now breaking down the test result data into smaller batches to create a larger number of Excel templates.”
[POB] I can correct the headline to:
Why a lack of basic data controls caused Covid-19 results to be lost
Of course they should not have chosen a file format with a size limit to process results. Nonetheless, whatever technology they used, anywhere data is exchanged between systems there must be checks and controls that reconcile the output of a transformation stage to its input, such as record counts and hash totals. Even if the upload process would only accept XLS files, batch total controls could have been imposed there too.
| Identifier: | POB2003 |
| Title: | Scientists rename human genes to stop Microsoft Excel from misreading them as dates |
| Source: | https://www.theverge.com/2020/8/6/21355674/human-genes-rename-microsoft-excel-misreading-dates |
| Release Date: | 06/08/2020 |
| Risk: | Inconvenience |
| Discrepancy: | none |
The bioinformatics community decided it was easier to change gene symbols than changing peoples’ habits.’Módos, whose job involves analyzing freshly sequenced genetic data, says Excel errors happen all the time, simply because the software is often the first thing to hand when scientists process numerical data. “It’s a widespread tool and if you are a bit computationally illiterate you will use it,” he says.’
| Identifier: | POB2002 |
| Title: | Spreadsheet error led to Edinburgh hospital opening delay |
| Source: | https://www.bbc.com/news/uk-scotland-edinburgh-east-fife-53893101 |
| Release Date: | 26/08/2020 |
| Risk: | Financial loss |
| Discrepancy: | GBP16M |
An NHS Lothian-commissioned review found a “human error” in a 2012 spreadsheet with the specifications for air flow in critical care rooms. It was only when the hospital had been handed over to NHS Lothian, and £1.4m monthly repayments had started, that independent checks found the critical care rooms were operating with the wrong air flow. Remedial work worth £16m has since been carried out and the new Sick Kids building started hosting outpatient appointments in July. An independent tester appointed by both sides “did not identify the non-compliance with the guidance within critical care”, according to the report
| Identifier: | POB1906 |
| Title: | State fund for jobs loses €750k due to ‘human error’ |
| Source: | https://www.rte.ie/news/business/2019/1126/1095303-human-error-cost-ntma-750-000-on-an-investment/ |
| Release Date: | 26/11/2019 |
| Risk: | Financial loss |
| Discrepancy: | EUR750K |
The National Treasury Management Agency explained that the agency purchased a fund in dollars, but it was not designated or marked on the spreadsheet record as such. It was recorded as a euro fund. “Subsequently, when the error was discovered, the dollar exchange rate had moved against the NTMA and the investment return was down €750,000.” The Public Accounts Committee (PAC) questioned the agency about the “control weakness”.
| Identifier: | POB1905 |
| Title: | “The error constituted maladministration on the part of the actuary” |
| Source: | https://www.iol.co.za/personal-finance/retirement/who-has-the-final-say-in-pension-fund-disputes-34927606 |
| Release Date: | 04/10/2019 |
| Risk: | Penalties and court appearances |
| Discrepancy: | R40 million loss |
The retirement fund had initially referred the matter to the adjudicator, which ruled that the actuary alone was responsible for R40m in overpayments made to exiting members of the fund. The overpayments, made over a number of years, stemmed from a “hard-coding error” in the spreadsheet system the actuary had been using. The adjudicator ruled that the error constituted maladministration on the part of the actuary, who should pay damages.
| Identifier: | POB1904 |
| Title: | AG: State overpaid Stroudsburg nearly $500K |
| Source: | https://www.poconorecord.com/news/20190729/ag-state-overpaid-stroudsburg-nearly-500k |
| Release Date: | 29/07/2019 |
| Risk: | Financial loss |
| Discrepancy: | US$500K |
The screw up was caused by the district using cumulative mileage totals rather than running calculations on a sample average for vehicles, which resulted in the district significantly over reporting total mileage data, causing the subsidy overpayments. The audit found the errors were due to a clerical error on the spreadsheet the district was using. In some cases, DePasquale said, the spreadsheet was double counting total days for some of the activity runs. The audit also found that the district lacked written procedures
| Identifier: | POB1903 |
| Title: | NYCLU apologizes for misstating racial disparity in Schenectady pot arrests |
| Source: | https://www.timesunion.com/news/article/NYCLU-apologizes-for-mistating-racial-disparity-14069700.php |
| Release Date: | 03/07/2019 |
| Risk: | Mistake harms trust department has worked to build |
| Discrepancy: | Loss of trust |
Released in April amid discussions in the state Legislature to legalize the drug in New York, the analysis erroneously indicated that black residents in Schenectady were 74 times more likely to be arrested for a marijuana-related offense than white people, despite making up just 12 percent of the population. The error, which placed Schenectady atop a list of 10 counties with the worst racial gaps in cannabis arrests, was a result of a “misaligned spreadsheet” which improperly merged Saratoga County U.S. Census data with Schenectady arrest numbers from the state Division of Criminal Justice Services, researchers said. The numbers, while accurate, were imported incorrectly due to a spelling discrepancy.
| Identifier: | POB1902 |
| Title: | A company submitted a bid nearly $3 million lower than it should have been |
| Source: | https://brookingsregister.com/article/district-seeking-new-bids-on-next-phase-of-mms-construction |
| Release Date: | 27/03/2019 |
| Risk: | Bid rejection |
| Discrepancy: | US$3M |
“The rescinded bid was for $6.5 million. As Willert explained, a spreadsheet error caused the company to submit a bid nearly $3 million lower than it should have been, and due to the nature of the error, the district would not have been able to make the company stick to its submitted bid.”
| Identifier: | POB1901 |
| Title: | A week after posting earnings, Canopy Growth restates EBITDA after calculation error |
| Source: | https://www.benzinga.com/markets/cannabis/19/02/13228789/canopy-growth-refiles-md-a-after-spreadsheet-formula-error |
| Release Date: | 21/02/2019 |
| Risk: | Loss of value |
| Discrepancy: | CA$48M |
Canopy Growth , the cannabis industry’s largest company by market value, refiled its third-quarter management discussion and analysis, or MD&A, on Thursday due to a spreadsheet formula error. Originally, Canopy’s adjusted EBITDA loss for the nine months ending Dec. 31, 2018 was CA$52 million ($69 million). Under the refiled MD&A submitted to the Canadian Securities Administration, the EBITDA loss amounted to CA$155 million ($117.5 million). Canopy’s stock fell 3.78 percent.
| Identifier: | POB1802 |
| Title: | Buskers Fest lost money last year due to ‘spreadsheet error’ |
| Source: | https://www.stuff.co.nz/the-press/news/102889835/buskers-fest-lost-money-last-year-due-to-spreadsheet-error |
| Release Date: | 18/04/2018 |
| Risk: | Financial loss |
| Discrepancy: | NZ$150000 |
World Buskers Festival organisers spotted a $150,000 “spreadsheet error” just a month before the event began. “The budget showed a surplus of $56,798, however a spreadsheet error overstated net door donations by $157,413 and the real bottom line was a deficit of $100,614,” the report states.” The trust was dissolved at the end of last month.
| Identifier: | POB1801 |
| Title: | Drinks company £5.2M spreadsheet arithmetic error |
| Source: | https://www.theguardian.com/business/2018/mar/21/bargain-booze-owner-conviviality-must-raise-125m-to-halt-bankruptcy |
| Release Date: | 21 March 2018 |
| Risk: | Bankruptcy after financial mismanagement |
| Discrepancy: | £5.2M |
At first the retailer said profits would be 20% lower than the £70m expected by the City, with £5.2m of the £14m hole that had opened up in its forecast down to a spreadsheet error, while the remainder was a reflection of weakening profit margins. On 21 March the Guardian (UK) reported “Firm issues third profits warning; says it will meet investors to raise funds via a share placing. The company, in a stock exchange announcement, said it was holding meetings with investors to raise £125m via a share placing that would help it pay a £30m tax bill due at the end of the month, fund overdue payments to creditors and repay a £30m loan.The company blamed the first shock profit warning on a spreadsheet arithmetic error made by a member of its finance team and weakening profit margins, and then admitted it had not budgeted for the £30m tax bill due this month.”
| Identifier: | POB1706 |
| Title: | Eskom made R1.5bn spreadsheet error in Optimum fine |
| Source: | https://www.fin24.com/Economy/Eskom/live-statecapture-ex-finance-director-at-eskom-tsholofelo-molefe-testifies-20171108 |
| Release Date: | 08 Nov 2017 |
| Risk: | Reputation loss |
| Discrepancy: | € 100 million |
The original fine had been R2.2bn (which saw the firm go into business rescue), but when Daniels met with a committee to discuss this, they discovered there had been an error in the spreadsheet.
“I was shocked at the answer,” she said. “There was an error in the spreadsheet. At that point, I was beyond furious. We have gone out in the media to say it was R2.2bn.”
“The final analysis of the claim showed the fine should have been R722m. I was quite pleased to hear Mr Marsden (Optimum business rescue practitioner) estimated it to be around R700m.”
| Identifier: | POB1705 |
| Title: | SLO County budget surplus now a deficit after ‘human error’ |
| Source: | http://www.ksby.com/story/36685143/slo-county-budget-surplus-now-deficit-after-human-error |
| Release Date: | Oct 26, 2017 |
| Risk: | Reputation |
| Discrepancy: | $8 Million |
“It’s common, it’s human, everyone does that. I’m glad I’m not responsible for it though,” said Bobby Gallow, San Luis Obispo resident.
The county says recently approved salary increases were “inadvertently omitted in the calculations.” Basically, someone forgot to include those numbers in the spreadsheet.
| Identifier: | POB1704 |
| Title: | Risks of using a spreadsheet for election calculation |
| Source: | Malaysia news |
| Release Date: | July 2017 |
| Risk: | Mistake or fraud? |
| Discrepancy: | Voting |
“As a result of an error in the Excel spreadsheet, the votes for Candidates number 31 to 38 were exactly the same as those of Candidates number 61 to 68. This can be observed from the table below:”
http://www.malaysia-today.net/karpal-tried-to-cover-up-the-fraudulent-dap-party-election/
“They needed a Malay face so they switched Vincent Wu with Zairil and blamed it on Excel. Karpal, however, knew that it was not a computer glitch but vote rigging. ”
| Identifier: | POB1703 |
| Title: | Emailed spreadsheet contained private data in ‘hidden’ columns |
| Source: | https://www.theregister.co.uk |
| Release Date: | 22 Feb, 2017 |
| Risk: | Data leakage |
| Discrepancy: | unknown |
Blundering Boeing bod blabbed spreadsheet of 36,000 coworkers’ personal details in email.
Global aerospace firm Boeing earlier this month sent a notification to Washington State Attorney General Bob Ferguson, as required by law, about a company employee who mistakenly emailed a spreadsheet full of employee personal data to his spouse in November, 2016.
The spreadsheet, sent to provide the employee’s spouse with a formatting template, contained the personal information of roughly 36,000 other Boeing employees, including Social Security numbers and dates of birth, in hidden columns. Some 7,288 of the affected employees resided in Washington State.
| Identifier: | POB1702 |
| Title: | Spreadsheet errors cost Clallam $494,157 |
| Source: | http://www.peninsuladailynews.com/news/spreadsheet-errors-cost-clallam-494157/ |
| Release Date: | 17 Feb, 2017 |
| Risk: | Financial loss |
| Discrepancy: | $494,157 |
A $331,074 Excel spreadsheet calculation error was made on a sheriff’s operations salary worksheet during the annual budget cycle last fall, commissioners learned Monday. A $163,083 spreadsheet error was made on a sheriff’s jail salary worksheet, officials said. “We did some forensics,” Benedict said Wednesday. “The spreadsheets were emailed back and forth between us and [County Administrator] Jim Jones’ office. Because of some cutting and pasting, not all the formulas were pasted correctly. It was an unintended error.”
| Identifier: | POB1701 |
| Title: | Unofficial spreadsheets land Italian pharma plant with regulatory warning |
| Source: | http://www.in-pharmatechnologist.com/Regulatory-Safety/Shredded-data-unofficial-spreadsheets-land-Italian-plant-with-warning |
| Release Date: | 31 Jan, 2017 |
| Risk: | Uncontrolled electronic spreadsheet |
| Discrepancy: | Regulatory sanction |
Data integrity issues have landed Italian drugmaker FACTA Farmaceutici with a US FDA warning letter.
The letter published today was sent to the firm earlier this month following an inspection at the manufacturing facility in San Nicolo a Tordino, Teramo in January 2016. The US Food and Drug Administration (FDA) observed a number of GMP violations at the plant, including discrepancies in data surrounding multiple sterile drug product lots.
http://www.fda.gov/ICECI/EnforcementActions/WarningLetters/2017/ucm538068.htm
“You stored original data in an ‘unofficial’ and uncontrolled electronic spreadsheet on a shared computer network drive,” the Agency said. “Your analyst stated that original data was first recorded in this ‘unofficial’ spreadsheet and transcribed later to an ‘official’ form. This spreadsheet showed failing results above the limits you established in your procedure, PCH 035 Visible Particle Determination in use prior to September 1, 2014.”
Search FDA for ‘spreadsheet’ to find other relevant documents.
| Identifier: | POB1607 |
| Title: | Wisconsin presidential recount will cost $3.9 million |
| Source: | http://www.wrn.com/2016/11/wisconsin-presidential-recount-will-cost-3-5-million/ |
| Release Date: | 28 Nov, 2016 |
| Risk: | Loss of financial control |
| Discrepancy: | $400,000 |
The Wisconsin Elections Commission announced Tuesday afternoon that a spreadsheet error caused them to incorrectly calculate the total cost of the recount. The revised cost estimate has climbed from $3.5 million to $3.9 million.
| Identifier: | POB1606 |
| Title: | SolarCity adviser Lazard made mistake in Tesla deal analysist |
| Source: | http://www.reuters.com/article/us-solarcity-lazard-idUSKCN11635K |
| Release Date: | Sep 1, 2016 |
| Risk: | Loss of reputation for advisors |
| Discrepancy: | $400 million |
Lazard Ltd (LAZ.N), the investment bank that advised SolarCity Corp (SCTY.O) on its $2.6 billion sale to Tesla Motors Inc (TSLA.O), made an error in its analysis that discounted the value of the U.S. solar energy company by $400 million, a regulatory filing by Tesla showed on Wednesday. This was the result of a computational error “in certain SolarCity spreadsheets setting forth SolarCity’s financial information that Lazard used in its discounted cash flow valuation analyses,” according to the filing.
| Identifier: | POB1605 |
| Title: | Gene name errors are widespread in the scientific literature |
| Source: | http://genomebiology.biomedcentral.com/articles/10.1186/s13059-016-1044-7 |
| Release Date: | 23 August 2016 |
| Risk: | Misinterpretation of scientific data |
“Gene name errors are widespread in the scientific literature.” Mark Ziemann, YotamEren and Assam El-Osta, Genome Biology201617:177 DOI: 10.1186/s13059-016-1044-7© The Author(s). 2016
They provide their data sources. They wrote scripts to trawl through published papers on genetics which had data files attached, and checked those files for data errors. They screened 35,175 supplementary Excel files and confirmed gene name errors in 987 supplementary files from 704 published articles in 18 journals. Linear-regression estimates show gene name errors in supplementary files have increased at an annual rate of 15 % over the past five years, outpacing the increase in published papers (3.8 % per year). “In conclusion, we show that inadvertent gene name conversion errors persist in the scientific literature, but these should be easy to avoid if researchers, reviewers, editorial staff and database curators remain vigilant.”
The root problem is that the researchers who uploaded those files NEVER checked them.
There is a paper about the problem from 2004 which also shows the problem with Gene names being turned into dates. See https://bmcbioinformatics.biomedcentral.com/articles/10.1186/1471-2105-5-80 . So 12 years after the paper was published people geneticists still have not learned how to solve the problem.
| Identifier: | POB1604 |
| Title: | M&S results hit by spreadsheet mishap |
| Source: | https://www.accountingweb.co.uk/tech/excel/ms-results-hit-by-spreadsheet-mishap |
| Release Date: | 15 July, 2016 |
| Risk: | Loss of reputation. restatement |
Marks & Spencer chief financial officer Helen Weir told the Financial Times last week she was shocked after discovering that a spreadsheet summing error forced the retailer to issue a correction to its quarterly trading statement. The original statement issued at 7am reported group sales had grown 1.3%. But at 1:31pm a correction was released showing that group sales had in fact fallen 0.4%.
| Identifier: | POB1603 |
| Title: | Blackpool Teaching Hospitals fined by Data Protection Commissioner |
| Source: | https://ico.org.uk/media/action-weve-taken/mpns/1624118/blackpool-nhs-trust-monetary-penalty-notice.pdf |
| Release Date: | 28 April 2016 |
| Risk: | Penalty £185,000 |
The Information Commissioner has imposed a penalty of £185,000 on Blackpool Teaching Hospitals NHS trust, who posted spreadsheets summarising their equal opportunities performance on-line. The WHOLE of each spreadsheet was posted, including the pivot tables containing sensitive personal information about every employee. These were on line for 10 months before anyone noticed, despite the Commissioner penalising two other trusts for similar disclosures, and blogging to highlight the risks. No checks were made before uploading, “nobody knew that Excel could do things like that”
| Identifier: | POB1602 |
| Title: | Calculation errors in the Babylonian tablet Plimpton 322 |
| Source: | http://www.math.ubc.ca/~cass/courses/m446-03/pl322/pl322.html#errors |
| Release Date: | 1800 BC |
| Risk: | Unknown |
This is alleged to be the oldest “spreadsheet” error in the world. There are older tables of numbers, such as a ruled papyrus from 4000 BC, but no mention of errors in that one!
https://theodox.quora.com/The-worlds-oldest-spreadsheet
| Identifier: | POB1601 |
| Title: | BlackEnergy .XLS Dropper |
| Source: | http://blog.didierstevens.com/2016/01/07/blackenergy-xls-dropper/ |
| Release Date: | 7 Jan, 2016 |
| Risk: | Shutdown of power plants |
Didier Stevens : “Analysing Malicious Documents“.
“I analyzed the spreadsheet (97b7577d13cf5e3bf39cbe6d3f0a7732) used in the recent BlackEnergy attacks against Ukrainian news media and electric industry.”
(For more on Office vulnerabilities, follow www.Decalage.info )
| Identifier: | POB1507 |
| Title: | Spreadsheet error overstated oil production from Fort Berthold Reservation |
| Source: | http://bismarcktribune.com/bakken/spreadsheet-error-overstated-oil-production-from-fort-berthold-reservation/article_ff7d4887-8085-5ca1-8963-01652bcceef7.html |
| Release Date: | Dec 16, 2015 |
| Risk: | Loss of credibility |
Oil production on the Fort Berthold Reservation had been grossly overstated as far back as 2012. Department spokeswoman Alison Ritter said an internal audit in October discovered an error in a formula used in an Excel spreadsheet to break out production figures for the reservation.
| Identifier: | POB1506 |
| Title: | Spreadsheet error helps Snyder County budget |
| Source: | http://www.dailyitem.com/news/spreadsheet-error-helps-snyder-county-budget/article_0cdb6014-a39b-11e5-a378-f75ee561a50e.html |
| Release Date: | Dec 15, 2015 |
| Risk: | Misestimation of budget |
| Discrepancy: | Actually a gain this time! |
Board Chairman Joe Kantz said Tuesday he found an error in a spreadsheet that revealed the county anticipates an additional $330,000 in revenue from the state for Children and Youth Services next year. The extra revenue means the expected shortfall in the $18.2 million budget for 2016 is now at $1.42 million.
| Identifier: | POB1505 |
| Title: | Tax flub raises budget questions |
| Source: | http://www.fortsaskatchewanrecord.com/2015/12/10/tax-flub-raises-budget-questions |
| Release Date: | Dec 10, 2015 |
| Risk: | Distrust of officials |
A major accounting error discovered in the 2016 city budget has decreased the overall tax rate by more than three per cent. A spreadsheet error made by the city’s finance department doubled the Dow Centennial Centre’s annual debt, instead of counting it once. The $1.27-million overestimation on city taxes left some city councillors second guessing the overall budget deliberations and its outcomes. “I can’t support this budget because I have no trust in the budget or its processes, especially after this error.”
| Identifier: | POB1504 |
| Title: | Censured after pointing out errors |
| Source: | https://www.hometownsource.com/stillwater_gazette/ |
| Release Date: | Nov 15, 2015 |
| Risk: | Shooting the messenger |
“There are some linkage and formula issues,” Schroeder said. “I want to say that, to me when I analyze it, (they) appear to happen after the year 2021.” In the past, the spreadsheet had been vetted by Moody’s and by Northland Securities up until 2021, Schroeder said. In 2015, another nine years was added to the pro forma, and Schroeder said the problems with the document happen after that date. “There was a non-patternized amount there, I would say, and it stands out,” Other problems within the spreadsheet relate to manual or automatic calculations, Schroder said.“I find this spreadsheet to be overly complicated”
| Identifier: | POB1503 |
| Title: | SFMTA Retracts Report of 651% Jump in Bike Traffic on San Jose Avenue |
| Source: | http://sf.streetsblog.org/2015/06/22/sfmta-retracts-report-of-651-jump-in-bike-traffic-on-san-jose-avenue/ |
| Release Date: | http://sf.streetsblog.org/2015/06/22/sfmta-retracts-report-of-651-jump-in-bike-traffic-on-san-jose-avenue/ |
| Risk: | Embarrassment |
The actual increase in morning peak-hour bike counts was 14 percent, said SFMTA spokesperson Ben Jose. In the evening peak hour, the reported 221 percent was actually 62 percent.“A second analysis of the underlying raw data revealed a spreadsheet error overstating the bicycling increase,” Jose said in a statement.
| Identifier: | POB1502 |
| Title: | Local Government Association says it will issue refunds after mathematical error led to councils being overcharged more than $70k |
| Source: | http://www.heraldsun.com.au |
| Release Date: | May 25, 2015 |
| Risk: | Reputation loss |
| Discrepancy: | $70,000 |
THE Local Government Association has admitted it botched membership fees for almost a decade and will have to refund about $70,000 to eight metropolitan councils. LGA communications director Chris Russell said Marion paid more because of a flaw in the spreadsheet, created 10 years ago to calculate fees. He said the LGA hired a consultant, who he would not name, to build the spreadsheet.
| Identifier: | POB1501 |
| Title: | Officials: Whistle-blower in Simon Property Group theft case found dead |
| Source: | http://www.orlandosentinel.com/news/breaking-news/os-simon-property-group-theft-ring-20150521-story.html#page=1 |
| Release Date: | May 23, 2015 |
| Risk: | To criminal proceedings |
Days after authorities announced arrests in a theft ring that allegedly bilked $20 million from America’s largest commercial-real-estate company, officials discovered only $4 million is allegedly missing and the lead witness in the case is now dead. “We used an Excel spreadsheet to document all the individual thefts because they were so numerous. … There was a calculator error within the spreadsheet that inaccurately summed the total,” according to Special Agent in Charge Danny Banks, head of FDLE’s Central Florida office. “Although the sum was totally inaccurate, it does not affect any of the outcome. The criminal charges are the same.”
| Identifier: | MPC04 |
| Title: | Spreadsheet Error Costs Tibco Shareholders $100M |
| Source: | http://ww2.cfo.com/spreadsheets/2014/10/spreadsheet-error-costs-tibco-shareholders-100m/ |
| Release Date: | 17 Oct, 2014 |
| Risk: | Regulatory sanction |
| Discrepancy: | $100 million |
Tibco Software shareholders will be getting $100 million less than originally anticipated from the company’s more than $4 billion sale to Vista Equity Partners as a result of a spreadsheet error that overstated Tibco’s equity value.According to a regulatory filing, Goldman Sachs, which is advising Tibco on the deal, used the spreadsheet in calculating that Tibco’s implied equity value was about $4.2 billion. The merger agreement, reflecting that number, was announced Sept. 29.
| Identifier: | MPC03 |
| Title: | Brownback budget director apologizes for incorrect figures |
| Source: | http://cjonline.com/news/2013-02-18/brownback-budget-director-apologizes-incorrect-figures |
| Release Date: | 18 Feb 2013 |
| Risk: | Budgeting error |
| Discrepancy: | $2 billion |
Quote from story: “Kansas Gov. Sam Brownback’s budget director apologized publicly Monday for supplying him with an incorrect figure that led the Republican governor to make erroneous claims about total state spending under his Democratic predecessor. Budget Director Steve Anderson said in a statement that the mistake occurred in entering data on an internal administration spreadsheet. The information was passed along to Brownback’s office, and he used the material in recent presentations to groups about his budget and tax proposals.”
| Identifier: | MPC02 |
| Title: | UPDATE 2-UBS trader discussed “slush account” with Adoboli |
| Source: | http://uk.reuters.com/article/2012/10/08/ubs-trial-idUKL6E8L8LKJ20121008 |
| Release Date: | 8 Oct, 2012 |
| Risk: | Deliberate fraud |
| Discrepancy: | $2.3 billion |
Using spreadsheets to perpetrate fraud — person on trial was being accused of running hidden trading book, masking true positions, going beyond risk limits set by his clients/institution. Convicted November 2012, sentenced to 7 years in prison — called largest fraud.
| Identifier: | MPC01 |
| Title: | Calculation error leaves school short by $25 million |
| Source: | https://www.deseretnews.com/article/865555955/Lawmakers-examine-cause-solutions-to-25M-education-budget-error.html |
| Release Date: | 12 April, 2012 |
| Risk: | Mispointed reference |
| Discrepancy: | $25 M |
“Error was <1% of state’s education budget, but two people resigned (and had resignations accepted) as a result of this error. Not clear if it was a real error, as this seems to have been a budget item based on an estimate. Was politically important enough to cause two resignations, and an investigation — looks like a formula was pointing to a wrong cell for making an estimate of number of students in the future. Quote “Shumway said the error was the result of a mathematical formula in an Excel spreadsheet referencing the wrong cell to calculate the state’s weighted pupil units. That inaccurate figure was then used for a per-pupil funding estimate that resulted in public education being underfunded by the Utah Legislature….”Why are we relying on Excel?”” Shumway said. “”It’s as close to being free as you can get.””
| Identifier: | POB1301 |
| Title: | Report identifies lack of spreadsheet controls, pressure to approve |
| Source: | http://files.shareholder.com/downloads/ONE/2261602328x0x628656/ 4cb574a0-0bf5-4728-9582-625e4519b5ab/Task_Force_Report.pdf |
| Organization: | JP Morgan |
| Region: | EU |
| Release Date: | 18 January 2013 |
| Risk: | Lowering estimate of VaR in Basel II models |
| Tags: | Financial |
| Spreadsheet Causes: | Logic not reviewed, manual copy/paste operations |
See pages 131-132 of the JP Morgan Task Force Report “…further errors were discovered in the Basel II.5 model, including, most significantly, an operational error in the calculation of the relative changes in hazard rates and correlation estimates. Specifically, after subtracting the old rate from the new rate, the spreadsheet divided by their sum instead of their average, as the modeler had intended. This error likely had the effect of muting volatility by a factor of two and of lowering the VaR”
As reported in “A tempest in a spreadsheet” http://ftalphaville.ft.com/2013/01/17/1342082/a-tempest-in-a-spreadsheet/? Lisa Pollack comments “On a number of occasions, he asked the trader to whom he reported for additional resources to support his work on the VaR model, but he did not receive any. Also it appears that he (had to?) cut a number of corners, which resulted increased operational risk and artificially low volatility numbers … pressure was put on the reviewers to get on with approving the model”
| Identifier: | POB1302 |
| Title: | Researchers embarrassed after student finds error in their published paper |
| Source: | https://www.peri.umass.edu/publication/item/526-does-high-public-debt-consistently-stifle-economic-growth-a-critique-of-reinhart-and-rogoff |
| Organization: | Harvard University / University of Massachusetts Amherst |
| Region: | USA |
| Release Date: | 16 April 2013 |
| Risk: | Academic discredit; possible damage or discredit to economic policies that cited their paper as justification of austerity |
| Tags: | Financial, economics |
| Spreadsheet Causes: | Formula omits cells in an average, logic not peer reviewed |
“In a paper, ‘Does High Public Debt Consistently Stifle Economic Growth? A Critique of Reinhart and Rogoff,’ Thomas Herndon, Michael Ash, and Robert Pollin of the University of Massachusetts, Amherst criticise a 2010 paper by Harvard economists Carmen Reinhart and Kenneth Rogoff, ‘Growth in a Time of Debt.’ They find three main issues: First, Reinhart and Rogoff selectively exclude years of high debt and average growth. Second, they use a debatable method to weight the countries. Third, there also appears to be a coding error that excludes high-debt and average-growth countries. All three bias in favor of their result, and without them you don’t get their controversial result.”
At one point R&R set cell L51 equal to AVERAGE(L30:L44) when the correct procedure was AVERAGE(L30:L49). They left Denmark, Canada, Belgium, Austria, and Australia out of the average. When you fix the Excel error, a -0.1 percent growth rate turns into 0.2 percent growth. See the article for a picture of the error.
| Identifier: | FH1203 |
| Title: | AstraZeneca spreadsheet slip-up forces reiteration of mid-term financial forecasts |
| Source: | http://uk.reuters.com/article/2012/01/09/uk-astrazeneca-idUKTRE8080BX20120109 |
| Organization: | AstraZeneca |
| Region: | EU |
| Release Date: | 09 January 2012 |
| Risk: | Firm’s shares down 0.4 percent |
| Tags: | Confidentiality |
| Spreadsheet Causes: | Confidential data embedded in template |
Britain’s second largest drugmaker AstraZeneca (AZN.L) was forced to reiterate its 2011 and mid-term financial forecasts on Monday after inadvertently releasing confidential company information to analysts. The Anglo-Swedish firm, …. described the released details as “out of date planning information” which it said did not represent its forecast for the full year for 2011 or for future periods.
AstraZeneca …. said the mistaken information had slipped out during “a routine consensus collection process.”
“Confidential company information was inadvertently embedded in a spreadsheet template sent to the sell-side analyst community that follows the company,” its statement said.
| Identifier: | FH1204 |
| Title: | Spreadsheets behind Olympic data misentry |
| Source: | http://www.telegraph.co.uk/sport/olympics/8992490/London-2012-Olympics-lucky-few-to-get-100m-final-tickets-after-synchronised-swimming-was-overbooked-by-10000.html |
| Organization: | London 2012 Olympics |
| Region: | EU |
| Release Date: | 04 January 2012 |
| Tags: | DataEntry |
| Spreadsheet Causes: | Manual data entry |
The London 2012 organising committee (Locog) confirmed on Wednesday that a decidedly unsynchronised error in its ticketing process had led to four synchronised swimming sessions being oversold by 10,000 tickets.
Locog and its ticket agent Ticketmaster spent the Christmas period contacting ticket holders and offering them alternatives, starting with tickets for sports they had applied for but been unsuccessful.
Locog said the error occurred in the summer, between the first and second round of ticket sales, when a member of staff made a single keystroke mistake and entered ‘20,000’ into a spreadsheet rather than the correct figure of 10,000 remaining tickets. The error was discovered when Locog reconciled the number of tickets sold against the final layouts and seating configurations for venues, and began contacting ticket holders before Christmas.
| Identifier: | FH1218 |
| Title: | Interest calculation low by $400,000 |
| Source: | http://www.wiscnews.com/baraboonewsrepublic/news/local/article_7672b6c6-22d5-11e1-8398-001871e3ce6c.html |
| Organization: | West Baraboo (Municipality) |
| Region: | USA |
| Release Date: | 09 December 2011 |
| Discrepancies: | 400,000 |
| Risk: | Underestimation of costs |
| Tags: | Government |
| Spreadsheet Causes: | Bad formula |
Due to a calculating error by their financial advisors, West Baraboo officials learned Thursday they will be paying about $400,0000 more over the lifetime of their most recent 10-year borrowing plan than originally projected.
During its regular December meeting, the West Baraboo Village Board looked back over last month’s decision to sell $1.1 million in general obligation bonds to cover a variety of village projects, said Village Clerk Mary Klingenmeyer. The review was required after the board received a letter from its financial advisory firm, Ehelers of Brookfield.
Ehlers advisor James Mann said “operator error” resulted in a spreadsheet underestimating the total cost of the 10-year bond.
“When we re-looked at the numbers we discovered a cell was not across the line being added correctly,” he said. “So it understated the impact.”
| Identifier: | FH1219 |
| Title: | Test errors impede History applications |
| Source: | http://www.cherwell.org/news/academic/2011/12/07/test-errors-impede-history- |
| Organization: | Oxford University Faculty of History |
| Region: | EU |
| Release Date: | 07 December 2011 |
| Risk: | False examination results |
| Tags: | Education |
| Spreadsheet Causes: | Bad formula |
Prospective History students were late in being told whether they had interviews this year, after the History Aptitude Test examiners mixed up marks and candidate numbers.
The History Aptitude Test (HAT) forms part of the entrance procedure for History undergraduate courses at Oxford University. Andrea Hopkins, a representative for the History Faculty, explained to Cherwell that “errors [had] occurred where markers had incorrectly transcribed the candidate’s number,” upon entering each candidate number and corresponding mark onto a spreadsheet. Each examiner’s spreadsheet was then compiled to create a larger spreadsheet for all of the 1613 applicants.
| Identifier: | FH1202 |
| Title: | Bad spreadsheet link: $6 million error and $12,500 audit fee |
| Source: | http://www.knoxnews.com/news/2011/dec/03/trustees-office-mistake-to-cost-taxpayers-12500/ |
| Organization: | Knox County Trustee’s Office |
| Region: | USA |
| Release Date: | 03 December 2011 |
| Discrepancies: | 6,000,000 |
| Risk: | Additional audit costs |
| Tags: | Government; OveraMillion |
| Spreadsheet Causes: | Bad link |
The auditor in late October discovered what officials called a $6 million accounting error. KPMG officials also said the Trustee’s Office took too long — four weeks — to compile information that makes up the office’s portion of the county’s financial statements and that it “was significantly out of balance.” The auditor did not respond Friday to an interview request.
Knox County Trustee John Duncan III initially said the error was the result of a software problem. Then he said it was human error and a “simple mistake.”
He said the reports his office sent to KPMG didn’t include one account, which threw off his department’s financial statements. The total cash on hand for June 30 that the office should have reported was $128.9 million, he said, and not $122.7 million, which was what was sent to the auditor. It occurred when one account wasn’t correctly linked into an Excel spreadsheet.
| Identifier: | FH1213 |
| Title: | County overlooks, then finds, taxable property worth $1.26 billion |
| Source: | http://www.bakersfield.com/news/business/economy/x986931070/County-overlooks-then-finds-taxable-property-worth-1-26-billion |
| Organization: | Kern County |
| Region: | USA |
| Release Date: | 25 November 2011 |
| Discrepancies: | 12,000,000 |
| Risk: | Underestimation of tax |
| Tags: | Government |
| Spreadsheet Causes: | Wrong spreadsheet |
County overlooks, then finds, taxable property worth $1.26 billion.
Twenty-five thousand acres seems like a lot of property for the county tax assessor’s office to have overlooked last summer. And this wasn’t just any land. It was one of Kern’s most valuable oil fields, generating about $12 million a year for taxpayers. Fortunately, the oversight was spotted in time and corrected. County Assessor-Recorder Jim Fitch blamed the mistake on a clerical error by an experienced staffer who used the wrong spreadsheet to calculate Occidental Petroleum Corp.’s tax bill for property it owns in Elk Hills.
| Identifier: | FH1201 |
| Title: | Securities and Exchange committee itself relies on spreadsheets. |
| Source: | http://www.gao.gov/htext/d12219.html |
| Organization: | Securities and Exchange Commission |
| Region: | USA |
| Release Date: | 16 November 2011 |
| Risk: | Systemic Financial Collapse |
| Tags: | Government |
| Spreadsheet Causes: | Systemic Overdependence |
In a recent audit by the Government Accountability Office, they noted the following: “As we have reported in prior audits, SEC’s general ledger system and certain software applications and configurations are not designed to provide the accurate, complete, and timely transaction-level data needed to accumulate and readily report reliable financial information. Instead, the initiation and recording of significant transactions is accomplished through the use of spreadsheets, databases, manual workarounds, and data handling that rely on significant analysis, reconciliation, and review to calculate amounts for the general ledger posting of transactions.” Their financial reporting is still heavily based on spreadsheets: “This process area continues to have a significant deficiency in FY 2011. Many of the agency’s financial reporting processes are still manual in nature and reliant on spreadsheets and databases to both initiate transactions and perform key control functions. ” And some of the spreadsheets were even linked to errors that were found: “In addition, we noted that a spreadsheet error resulted in incorrect amounts for SEC’s legal liability and lease liability disclosed in the notes to SEC’s interim financial statements [….] These and other errors we found point to a lack of effective review procedures, which are even more critical given SEC’s heavy dependence on spreadsheets and error-prone manual data entry for its financial reporting and accounting processes. ” Their CFO promises to improve this in the coming year: “…redesigning and implementing controls over the spreadsheets and databases used by the agency for material financial reporting-related transactions and key operational management decisions, based on risk; ”
| Identifier: | FH1220 |
| Title: | $350,000 fine for a spreadsheet error |
| Organization: | Arizona Portland Cement Corporation (cement factory) |
| Region: | USA |
| Release Date: | 08 November 2011 |
| Discrepancies: | 350,000 |
| Risk: | Money Loss |
| Tags: | Commercial |
In 2006, the company paid the state $300,000 to settle a case involving emitting air pollution, failing to submit test reports, and not installing temperature monitors.
In January, the EPA fined Arizona Portland $350,000 for an expansion plan that would have exceeded emissions for particulate matter. The plant was ordered to reduce expected annual emissions by 80,000 pounds by adding $6.7 million of control equipment to the plan.
Although the EPA still lists the company as a high-priority violator, there never was a physical violation and the entire episode was sparked by a math error, said Steve Regis, vice president of engineering for CalPortland, Arizona Portland’s parent company.
“Our consultant had an error in his spreadsheet, and we submitted the permit application using his data. We are in compliance. We were always in compliance,” Regis said.
| Identifier: | FH1217 |
| Title: | 1,791 voters inflated to 4,870 |
| Source: | http://www.chieftain.com/news/local/ortiz-scales-down-inactive-voter-count/article_abcb94c0-06a4-11e1-8668-001cc4c03286.html |
| Organization: | Pueblo County |
| Region: | USA |
| Release Date: | 04 November 2011 |
| Risk: | False Elections |
| Tags: | Government |
Thursday, Ortiz ultimately agreed with Gessler that the state database showed 1,791 inactive ballots were returned in Pueblo County. Ortiz blamed simple counting errors in his office for his inaccurate number. The state computer data is cumbersome to read, so Ortiz’s office converted its display to a different form of spreadsheet. That’s where the counting error occurred, the clerk said. Anytime there is a dispute over tallying votes, it always raises suspicions about possible voter fraud.
| Identifier: | FH1210 |
| Title: | $1M went missing as staff managed “monstrous spreadsheets.” |
| Source: | http://www.metrowestdailynews.com/top_stories/x1876834739/Framingham-discovers-1-5-million-error |
| Organization: | Town of Framingham (municipality) |
| Region: | USA |
| Release Date: | 15 October 2011 |
| Discrepancies: | 12,000,000 |
| Risk: | Money Loss |
| Tags: | Government |
The town mistakenly reckoned it had $1.5 million more in this year’s budget than it actually has and must now use $600,000 in unexpected state aid to help fill that gap, officials said yesterday.
Chief Financial Officer Mary Ellen Kelley said she takes responsibility for the mistake, which she found Wednesday night on the debt services line item in the $208.6 million fiscal 2012 operating budget. She said a figure went missing as staff managed “monstrous spreadsheets.”
“It’s frustrating,” Kelley said yesterday. “I hate when we make mistakes. People are human and they do make mistakes, but I hate it.”
| Identifier: | FH1216 |
| Title: | £4.3M spreadsheet error leads to resignation of its chief executive |
| Source: | http://www.express.co.uk/posts/view/276053/Mouchel-profits-blow |
| Organization: | Mouchel (Outsourcing specialist) |
| Region: | EU |
| Release Date: | 07 October 2011 |
| Discrepancies: | 4,300,000 |
| Risk: | Too high income expectation |
| Tags: | Commercial |
An accounting error yesterday forced outsourcing specialist Mouchel into a major profits warning and sparked the resignation of its chief executive.
Richard Cuthbert stood down as analysts warned the firm was in danger of breaking banking agreements on debt after it had to reduce full-year profits by more than £8.5million to below £6million.
The mistake was not Cuthbert’s, however, but an outside firm of actuaries. It told the company on Wednesday that a spreadsheet error meant a pension fund deficit had been wrongly valued. As a result, Mouchel had to write down profits by £4.3million.
| Identifier: | FH1221 |
| Title: | State Auditor of Kentucky: bad smells from Sanitation District #1 |
| Source: | http://boonecounty.fox19.com/news/business/75526-audit-sd1-reports-unbid-work-missing-info-incomplete-board-guidelines |
| Organization: | SD1 (Sanitary District #1, Kentucky Sewage Water treatment) |
| Region: | USA |
| Release Date: | 17 August 2011 |
| Risk: | Overestimation of costs |
| Tags: | Government |
| Spreadsheet Causes: | Bad formula |
During the course of the examination, auditors noted several instances of accounting errors and apparent lax accounting controls and oversight. The specific instances involved errors in a spreadsheet allocating labor costs to projects, significant accounting entries made without proper review or documentation, and errors in the rate setting model.
SD1 calculated the labor costs to be charged to a particular project using an Excel spreadsheet containing a formula which multiplied the hours by the applicable rates. The formula was in error resulting in 151 projects being charged excessive labor costs. SD1’s failure to recognize such a significant error in labor allocation calls into question the quality and frequency of accounting review and oversight.
| Identifier: | FH1207 |
| Title: | Clallam County cashier hides rows in a spreadsheet to cover up theft |
| Source: | http://www.peninsuladailynews.com/article/20110721/NEWS/307219992/-866-or-617467-theft-amount-issue-in-trial-of-former |
| Organization: | Clallam County |
| Region: | USA |
| Release Date: | 20 July 2011 |
| Discrepancies: | 617,467 |
| Risk: | Money Loss |
| Tags: | Government;Trail;Theft |
| Spreadsheet Causes: | Hidden Rows |
Betts, ….., is charged with first-degree theft, money laundering and 19 counts of filing false or fraudulent tax returns with the state Department of Revenue. Betts is accused of stealing the funds by exchanging checks for real estate excise taxes with money from the office’s cash drawer.
The actual amount taken cannot be determined, state Auditor’s Office fraud investigator Jim Brittain said in his report, but he cited “questionable transactions” during the audit period of Feb. 1, 2004, to May 19, 2009, of $793,595. The amount taken was placed in hidden rows in her spreadsheets that weren’t discovered until Betts allegedly confessed to supervisors May 19, 2009, of stealing about $1,200.
| Identifier: | FH1215 |
| Title: | MI5 makes 1,061 bugging errors |
| Source: | http://www.governmentcomputing.com/news/2011/jul/01/mi5-data-collection-errors |
| Organization: | MI5 (secret intelligence) |
| Region: | EU |
| Release Date: | 01 July 2011 |
| Risk: | False information |
| Tags: | Government |
| Spreadsheet Causes: | wrong cell format |
MI5 wrongly collected subscriber data on 134 telephone numbers as a result of a software error, according to interception of communications commissioner Sir Paul Kennedy’s annual report.
A spreadsheet formatting error caused the service to apply for data on the identity of telephone numbers ending in 000, rather than the actual last three digits. “The subscriber data acquired had no connection or relevance to any investigation or operation being undertaken by the Security Service,” writes Kennedy.
| Identifier: | FH1228 |
| Title: | Major Blunder By Norwegian Central Bank Gives People Totally Wrong Impression About Interest Rates |
| Source: | http://www.businessinsider.com/norwegian-central-bank-blunder-2011-6 |
| Organization: | Norwegian Central Bank |
| Region: | EU |
| Release Date: | 28 June 2011 |
| Risk: | False rent percentage |
| Tags: | Bank |
| Spreadsheet Causes: | Typo |
It turns out that last week, the Norwegian central bank put out a spreadsheet indicating a 60-70% change of rate hikes in either August or September. Turns out that was a sloppy mistake.
According to this translated article from e24.no, the mistake was identified, and the number was changed to 100%, without a big press release or any other notification that such an error had been made.
| Identifier: | FH1227 |
| Title: | King’s Fund apologise for “error” in figures on health spending in Wales |
| Source: | http://www.leftfootforward.org/2011/05/kings-fund-apologise-for-error-on-health-spending-in-wales/ |
| Organization: | The King’s Fund (health charity) |
| Region: | EU |
| Release Date: | 27 May 2011 |
| Risk: | Overestimation of costs |
| Tags: | Charity |
Respected health think tank, The King’s Fund has been forced to write to the Welsh government to apologise for errors in figures it produced on health spending in Wales. On Wednesday, media outlets declared that Wales faced deeper spending cuts to the health service than the rest of the UK, with King’s Fund chief economist John Appleby, in an article for the British Medical Journal, arguing the health service budget in Wales faced cuts of up to 11 per cent over three years. However, in a letter to Welsh health minister Lesley Griffiths, Mr Appleby retracted the figure, writing:
“Dear Ms Griffiths
British Medical Journal Data Briefing: Correction
As you will be aware, there has been some media interest following a data briefing I wrote for the BMJ on NHS spending in the four UK nations. I am writing to let you know of a correction to the article.
“The article originally stated that Wales would receive a 10.7 per cent real terms cut in NHS spending over the next four years. It should have stated that this is over three years. This was corrected on the BMJ web site. Unfortunately, a spreadsheet error which was only noticed after publication also exaggerated the real change in Wales’ NHS spending. The correct figure is -8.3% by 2013/14 (compared with 2010/11) and not -10.7%.
The BMJ will publish a correction to my article but I also wanted to set the record straight with you and apologise for any confusion this has caused.
Yours sincerely,
John Appleby, Chief Economist, The King’s Fund”
| Identifier: | FH1226 |
| Title: | Computer errs in tally |
| Source: | http://www.dailymail.com/ap/ApTopStories/201104290624 |
| Organization: | Morgantown (municipality) |
| Region: | USA |
| Release Date: | 29 April 2011 |
| Risk: | False Elections |
| Tags: | Government |
| Spreadsheet Causes: | Bad formula |
Morgantown City Clerk Linda Little said a computer error led to a discrepancy in vote totals for council candidate George Papandreas.
The candidate — who was defeated by incumbent Ron Bane — actually received 99 more votes than was originally indicated in the total listed on the city’s unofficial results. Papandreas said he had no issue with the error. The error was corrected by Thursday morning, after The Dominion Post contacted Little about it Wednesday night. The city clerk’s office had originally reported Papandreas, a 1st Ward candidate, received 935 votes in the city’s election. The report, a spreadsheet which included vote totals for each of the city’s 17 precincts, was given to The Dominion Post Wednesday.
The Dominion Post added the totals for each precinct and determined he had 1,034 votes.
Little said election officials had put the total number of votes for each precinct into a Microsoft Excel spreadsheet. The computer program skipped over an entry when it added the totals, she said.
| Identifier: | FH1225 |
| Title: | AXA Rosenberg paying $242M to settle case on code |
| Source: | http://www.businessweek.com/ap/financialnews/D9L5I5I00.htm |
| Organization: | AXA Rosenberg |
| Region: | USA |
| Release Date: | 03 February 2011 |
| Discrepancies: | 242,000,000 |
| Risk: | Overestimation of losses |
| Tags: | Commercial |
An investment firm is paying $242 million to settle civil fraud charges that it hid an error in its computer code that resulted in losses for its clients, federal regulators said Thursday.
The settlement with AXA Rosenberg Group announced by the Securities and Exchange Commission is among the largest for the agency and is the first of its kind.
The SEC said senior managers at two AXA Rosenberg affiliates knew about an error in the code for the quantitative investment model used to manage clients’ assets. The agency says the error disabled a key element of the model for managing risk. The agency said senior managers ordered others to conceal the error and failed to fix it, causing $217 million in client losses.
AXA Rosenberg, based in Orinda, Calif., is owned by French insurance company AXA SA. AXA Rosenberg, neither admitted nor denied wrongdoing in settling the case but did agree to refrain from future violations of the securities laws.
The use of quantitative models has proliferated on Wall Street.
“To protect trade secrets, quantitative investment managers often isolate their complex computer models from the firm’s compliance and risk-management functions and leave oversight to a few sophisticated programmers,” SEC Enforcement Director Robert Khuzami said in a statement. “The secretive structure and lack of oversight of quantitative investment models, as this case demonstrates, cannot be used to conceal errors and betray investors.”
The SEC said the error came into the computer model in April 2007. It was eventually corrected, but the CEO of AXA Rosenberg Group wasn’t told about it until November 2009, the agency said. The firm did its own investigation and told SEC examiners about the error in March 2010. The firm’s clients were informed on April 15.
AXA Rosenberg agreed in the settlement to pay investors the $217 million they were said to have lost, as well as a $25 million civil fine. The firm also agreed to hire an independent consultant in quantitative investment techniques to review its disclosures.
“We deeply regret that the coding error adversely impacted many of our clients,” AXA Rosenberg’s chairman, Dominique Carrel-Billiard, said in a statement. “The exhaustive review that we undertook of this matter reflects our commitment to regaining our clients’ confidence and restoring trust.”
The firm noted that it has made significant changes in recent months in its organization, named new executives to several top positions and put in additional measures to control risks.
| Identifier: | FH1212 |
| Title: | Provision for income taxes understated by $36.4 million |
| Source: | http://www.gurufocus.com/news/114917/georgia-gulf-corp-reports-operating-results-10qa |
| Organization: | Georgia Gulf Corporation (chemicals company) |
| Region: | USA |
| Release Date: | 22 November 2010 |
| Discrepancies: | 36,400,000 |
| Risk: | Too high income expectation |
| Tags: | Commercial |
| Spreadsheet Causes: | Manual data entry |
In …. 2010, we engaged a different third-party firm of tax professionals to assist us with the preparation of our 2009 U.S. federal income tax return. During the preparation of that tax return we, with the support of our tax advisors, identified certain issues that caused us to re-evaluate the application of the relevant provisions of the IRC relating to the 2009 financial restructuring activities. Consequently, we determined that a manual input error to a spreadsheet used in the tax calculations relating to the tax impact of our 2009 financial restructuring activities had been made and that certain applications of the relevant provisions of the IRC were incorrect. As a result, the reduction in various tax attributes resulting from the CODI we recognized in 2009 was understated. This error caused our provision for income taxes to be understated by $36.4 million and our net income to be overstated by $36.4 million, each for the year ended December 31, 2009. This adjustment did not, however, result in any additional tax liability payable by us to tax authorities in respect of 2009 or earlier periods.
| Identifier: | FH1211 |
| Title: | Council ‘loses’ £21m spreadsheet error |
| Source: | http://www.denbighshirefreepress.co.uk/news/94441/council-loses-21m.aspx |
| Organization: | Denbigshire County Council |
| Region: | EU |
| Release Date: | 14 October 2010 |
| Discrepancies: | 21,000,000 |
| Risk: | Underestimation of assets |
| Tags: | Government |
| Spreadsheet Causes: | Bad formula |
The clanger was discovered by Denbighshire County Council staff just before its annual audit.
In a report considered by the authority’s corporate governance committee on Monday (September 27) the head of finance and assets Roger Parry said the audit was virtually complete but a revised set of accounts had been prepared. “A significant error was made in the valuation of the council’s assets,” he said. “Along with other adjustments this has meant a £21m (3.5 per cent) reduction in the assets shown on the council’s balance sheet. “The error was picked up internally prior to the audit commencing but after the draft accounts had been produced. “Now that the valuations department has moved into the finance department a full review of all the procedures is being undertaken and a significantly improved process will be implemented for the 2010-2011 accounts production.”
A council spokeswoman explained later that the discrepancy had occurred because of an error on a spreadsheet where some assets had been double-counted.
| Identifier: | FH1224 |
| Title: | Light Dept. Memo Defends Substation Team, Cost Increases |
| Source: | http://belmont.patch.com/articles/light-dept-memo-defends-substation-team-cost-increases |
| Organization: | Belmont Municipal Light Department |
| Region: | USA |
| Release Date: | 22 September 2010 |
| Risk: | Underestimation of costs |
| Tags: | Government |
| Spreadsheet Causes: | Manual data entry |
The head of the town’s electrical utility is the subject of serious accusations of mismanagement and stonewalling information on a nearly 20 percent increase in the cost of the 115 kV transmission substation the BMLD has fought long to bring online….
in the memo – referencing the “Recent 115kV Cost Impact” – sent after the Sept. 8 meeting of the Advisory Board but before his meeting with the independent group on Sept. 13, Richardson acknowledged that unspecific cost increases was going to ratchet up the total cost for the new substation to $47 million. That would be a $9 million increase from the long accepted price tag of $38 million the BMLD announced in June …..
In addition to accepting an incorrect amount, Selectman and Light Board member Ralph Jones said for the error to be caught after an initial $38 million cost estimate was made public in June, “shows that they failed on quality control.” Ralph said it should be standard procedure for spreadsheets to be reviewed to find just these sorts of “miscalculations.”
| Identifier: | FH1223 |
| Title: | Millions undisclosed by TX congressman |
| Source: | http://www.wwlp.com/dpps/news/national/south/millions-undisclosed-by-tx-congressman-_3581532 |
| Organization: | Personal Financial disclosure report of congresmember Michael McCaul |
| Region: | USA |
| Release Date: | 15 September 2010 |
| Risk: | Bad Publicity |
| Tags: | Government |
| Spreadsheet Causes: | Manual data entry |
Rep. Michael McCaul, R-Texas, the ranking Republican on a U.S. House ethics subcommittee, is admitting his accounting firm made “clerical errors” worth millions, thus failing to fully disclose his finances. Roll Call, a Capitol Hill newspaper, detailed missing information about dozens of stock transactions made by McCaul’s wife. McCaul has said a spreadsheet his accountants, Millan & Company, prepared actually cut off stocks that began with the letter “E,” all the way to the letter “Z.” The full electronic statement was actually disclosed to the ethics subcommittee prior to the firm’s printing. The firm released a statement, saying the error was “inadvertent and unintentional.”
| Identifier: | FH1229 |
| Title: | Blatant Data Error At The Federal Reserve |
| Source: | http://www.zerohedge.com/article/blatant-data-error-federal-reserve |
| Organization: | Federal Reserve (USA central bank) |
| Region: | USA |
| Release Date: | 01 August 2010 |
| Risk: | False information |
| Tags: | Bank |
| Spreadsheet Causes: | Copy paste |
A vigilant reader, who combed through the backup of today’s Consumer Credit G.19 statement points out a flagrant and obvious error in the Fed’s data. While luckily the data impact is not major (at most $4 billion, which in our day and age is a pithy 50% of Goldman’s FICC trading desk bonus), the implication that the Fed does not check its work in something as critical as one of the core data series (or at least it used to be until a few machines took over the market, to whom, as today indicated, a record credit contraction somehow ended up being a positive event) is very, very troubling.
The original, Fed-hosted excel file with the backing data of the actual G.19 statement can be found here. We welcome all readers to compare cells AC 804 through AC 809, which is the data for “Consumer Revolving Credit Owned by Nonfinancial Business, Not Seasonally Adjusted” for the months June through November of 2009, and to compare it with data in cells AC 792 through AC 797, which is comparable data for the months June through November of 2008. These are identical and very much wrong! So, dear Fed auditors, while you obviously are very highly overpaid for your error-proofing work, can you please tell us what the real Consumer Credit number for November is?
| Identifier: | FH1222 |
| Title: | Hungary’s Central Services Directorate fined for misleading spreadsheet |
| Source: | http://bbj.hu/economy/kdb-fines-central-services-directorate-for-misleading-spreadsheet_53252 |
| Organization: | Central Services Directorate (an organisation which manages procurement for ministeries and other governmental organisations) |
| Region: | EU |
| Release Date: | 28 June 2010 |
| Discrepancies: | 2,000,000 |
| Risk: | Money Loss |
| Tags: | Government |
| Spreadsheet Causes: | Misleading sheet |
Hungary’s Public Procurement Arbitration Board (KDB) has invalidated the results of a tender called by the Central Services Directorate and ordered it to pay HUF 2 million because of a misleading spreadsheet provided to bidders.
Bidders received a spreadsheet with two columns and were to fill in the first one, while the second was filled in automatically. However, 25 of the 28 bidders misunderstood the spreadsheet and manually filled in the second column. The directorate must evaluate the bids again.
| Identifier: | FH1234 |
| Title: | Ex-county workers consider suit over reduced retirement |
| Source: | http://www2.dailyprogress.com/news/cdp-news-local/2010/mar/25/ex-county_workers_consider_suit_over_reduced_retir-ar-75157/ |
| Organization: | Albemarle County |
| Region: | USA |
| Release Date: | 25 March 2010 |
| Discrepancies: | 66,000 |
| Risk: | Money Loss |
| Tags: | Government |
Former Albemarle employees who say they’ve been cheated out of money by the county are considering taking legal action against the local government, according to former police Officer Ron Kesner. The county has offered employees monthly stipends in exchange for retiring early as a way to reduce personnel expenses. However, officials say a miscalculation caused them to offer 42 employees more money than intended. While the county is not making the employees pay back money they’ve received, the retirees are receiving less than they had been promised.
“All I want is what’s fair,” said Juanita Irvine, a former senior family support worker for the county’s social services department. “I made my decision to retire based on their figures.” Kesner said that affected retirees are hoping to resolve the discrepancy with county officials and are considering legal action only as a “last resort.”
Officials have estimated that the mistake cost the county $66,000 in overpayments; had the error not been corrected, the county would have been out $360,000. Questions about precisely how the stipends were miscalculated and who was to blame remain unanswered.
Kimberly Suyes, director of the county’s Human Resources Department, said there was a “miscalculation in the spreadsheet. There’s not much more.” An employee in Human Resources made the error. Suyes declined to say whether the employee who made the error was fired.
| Identifier: | FH1235 |
| Title: | NAMB corrects missionary count data to GCR Task Force |
| Source: | http://www.gofbw.com/news.asp?ID=11451 |
| Organization: | North American Mission Board |
| Region: | USA |
| Release Date: | 09 March 2010 |
| Risk: | False information |
| Spreadsheet Causes: | Bad formula |
The North American Mission Board has informed the Great Commission Task Force that data it provided last summer regarding the geographic distribution of missionaries in North America contained a significant error. The NAMB error resulted in incorrect data being published….
The error occurred when NAMB double-counted the states of Texas and Virginia. Both states have two, independently operated state Baptist conventions with which Southern Baptists are affiliated. The data spreadsheet used listed each of those state conventions with their individual subtotals and also an overall total from the state. Both the subtotals and the totals from those states were added when the overall missionary count was computed, leading to a double-count for the two states.
| Identifier: | FH1209 |
| Title: | Spreadsheet error costs time and money, yet again |
| Source: | http://andyjko.com/2010/03/05/spreadsheet-error-costs-time-and-money-yet-again/ |
| Organization: | ISTA (gas water electricity provider) |
| Region: | USA |
| Release Date: | 05 March 2010 |
| Discrepancies: | 270 |
| Risk: | Too high customer bills |
| Tags: | DataEntry; Commercial |
| Spreadsheet Causes: | Manual data entry |
Back in November, I got my first water, sewage, and gas bill from a company called ISTA. My apartment management had taken a while to set up the billing after the previous billing company went of business (or dropped the contract, I don’t remember exactly what happened). So I hadn’t actually paid water and sewage for two or three months.
So when the bill came for $173, I wasn’t too surprised. I didn’t really remember what I’d paid the previous year, but this seemed reasonable for a few months of water, sewage and gas. I wrote the check, and forgot about ISTA.
Forty five days later, I got the next bill, but this time something seemed wrong: $463 and it had only been a month and a half. What the hell was going on? I looked back on my old bills and noticed that my average 30 day bill was about $30, even in the winter. Either the company was trying to extort money from me or somebody had made an accounting error.
I looked more closely at the bill, which had three columns: previous usage, current usage, and usage. The difference between the first two columns was exactly 1000. The value in the third column was 10,000. Was there some hidden multiplier I didn’t understand? Maybe there was some rate that just happened to be 10, and I had just kept my apartment and showers really warm this winter.
So I called ISTA and disputed the bill. They immediately escalated it to their dispute manager, who called me back after a few days. They said that there had been a misread meter and that they had corrected the reading, and that after the bill was now only about $270, after they had applied the credit. When I got the call, I had a meeting to be at, so I didn’t think about it much.
After I got home that night, it still didn’t seem right. $270 for 45 days? What happened to the rates? They must have gone up by a factor of 10! So the next day, I called ISTA back, and spoke to a nice lady about my problem. Rather than call the dispute manager again, she told me she was opening my spreadsheet. She proceeded to walk through the calculations with me, describing the rates and the formulas. I jotted them down on paper as we went. Finally, we got to the final total calculation, and she said, “so this times the multiplier is … wait, it shouldn’t be.” She immediately put me on hold.
A few minutes later, she came back, saying that she needed to have the accounting department look at my spreadsheet. My spreadsheet, implying that every customer has their own. She said that the dispute manager would, yet again, call me back in a few days.
Four days later, the dispute manager called me back and explained that there was some sort of disagreement between billing and accounting, regarding the cause of the problem. Billing thought it was the spreadsheet and accounting thought it was the meter readings. She said she’d call back in a few more business days, after they’d worked out their differences.
When she did call back, she leveled with me: accounting was wrong, there was an error in the spreadsheet, and after fixing the multiplier cell, my bill was reduced by a factor of 10. After the credit calculators, they determined that I had overpaid from the previous bill by about $100, and that I probably wouldn’t have a bill for the next two cycles. She apologized for how long it took to resolve the issue, but reassured me that it wouldn’t happen to me again.
But I wasn’t thinking about me at this point. I was thinking about all of the other customers, whose spreadsheets probably had the same error. Would the accountants audit all of the spreadsheets that copied the error? How many customers would call about the bills? How many would insist, like I did, that there was a spreadsheet error, and demand that it be properly diagnosed? And how much of this feedback would ever make it to the accountants writing the buggy spreadsheets?
| Identifier: | FH1208 |
| Title: | Flintshire County Council school cash blunder ‘down to spreadsheet error’ |
| Source: | Flintshire County Council school cash blunder ‘down to spreadsheet error’ Source: http://www.flintshirechronicle.co.uk/flintshire-news/local-flintshire-news/2010/02/18/flintshire-county-council-school-cash-blunder-down-to-spreadsheet-error-51352-25856321/ |
| Organization: | Flintshire County |
| Region: | EU |
| Release Date: | 18 February 2010 |
| Discrepancies: | 1,000,000 |
| Risk: | Money Loss |
| Tags: | Government |
| Spreadsheet Causes: | Manual data entry |
A £1M council cash ‘cock-up’ which prompted calls for the resignation of a top Tory councillor was down to a spreadsheet error, it has been claimed.
County Hall officials have not published details of an internal investigation carried out in the wake of the blunder, which left Flintshire Council £1m out of pocket after they over-allocated money cash to the county’s 12 high schools last year.
But the Chronicle understands the mix-up occurred because someone ‘put money in the wrong column’ – and no one spotted it before the money was handed out.
| Identifier: | FH1206 |
| Title: | $153 million error in Department of Energy Tax spreadsheet leaves reporter searching for the truth |
| Source: | http://www.pv-tech.org/chip_shots_blog/for_the_record_manufacturing_investment_tax_credit_typo_pushes_rec_solar_to |
| Organization: | Department of Energy |
| Region: | USA |
| Release Date: | 13 January 2010 |
| Discrepancies: | 153,300,000 |
| Tags: | Government |
Whose information should you trust more, an agency of the Federal government or a major solar company? In the case of the recently announced U.S. Advanced Energy Manufacturing Investment Tax Credits, the advantage goes to the private sector. Because of an error in the Department of Energy’s spreadsheet of credit recipients, a $153.3 million discrepancy has propelled REC Silicon from well down the list of tax credit recipients to the top of the charts.
| Identifier: | FH1205 |
| Title: | $15 million mistake: that representative doesn’t work for the company anymore |
| Source: | http://www.tmcnet.com/usubmit/2009/11/21/4494996.htm |
| Organization: | Monroe County |
| Region: | USA |
| Release Date: | 21 November 2009 |
| Discrepancies: | 15,000,000 |
| Risk: | Too high income expectation |
| Tags: | Government |
| Spreadsheet Causes: | Bad Link |
The financial discrepancy came to light earlier this month during a county council work session. At that time, Smith described it as a reporting error that happened when two employees switched jobs, after which the new employee in her office began using a different spreadsheet that contained a broken data link. A single entry error compounded over the past few months to make the apparent discrepancy larger, she said at the time, eventually leaving the county’s books showing $15 million more than the bank said it had.
| Identifier: | FH1231 |
| Title: | Foutje in spreadsheet kost waterschap 2,7 miljoen |
| Source: | http://binnenland.nieuws.nl/566978# |
| Organization: | Waterschap Veluwe |
| Region: | EU |
| Release Date: | 01 September 2009 |
| Discrepancies: | 2,700,000 |
| Risk: | Money Loss |
| Tags: | Government |
Het Waterschap Veluwe moet 2,7 miljoen euro interen op zijn reserves omdat het een te lage heffing heeft opgelegd aan eigenaren van woningen en bedrijfspanden. Dat komt door een foutje, meldt het waterschap. In een spreadsheet zijn per ongeluk twee velden omgewisseld, waardoor de aanslag zo’n vijf procent lager uitvalt.
De gemiddelde eigenaar betaalt hierdoor tien tot negentien euro minder waterschapslasten over dit jaar. Huurders hebben geen voordeel van het foutje, doordat de heffing voor hen niet geldt. Het waterschap heeft ervoor gekozen het misgelopen geld niet later alsnog in rekening te brengen. “Volgend jaar vragen we weer het normale bedrag, en we gooien er geen schepje bovenop.”
Het geld wordt uit de zogenoemde egalisatiereserve geput. “Dat potje moet voorkomen dat we de waterschapslasten opeens flink opschroeven als we bijvoorbeeld nieuwe gemalen nodig hebben”, legt de zegsman uit.
In een folder bij de belastingaanslag wordt uitgelegd dat de lasten volgend jaar stijgen. Het foutje heeft niet tot ontslagen geleid. “Het bestuur heeft maandag besloten dat dit nou eenmaal kan gebeuren. Wel gaan we na hoe we herhaling kunnen voorkomen.”
| Identifier: | NS01 |
| Source: | C&C Group admit to mistake in revenue results |
| Organization: | C&C |
| Release Date: | Jul-09 |
| Discrepancies: | 11% of sales for the four months to the end June |
Shares in C&C fell 15 per cent after it said total revenue in the four months to end-June had not risen 3 per cent as reported, but had dropped 5 per cent. C&C said cider revenues in the UK had fallen 12 per cent, not 1 per cent, while cider revenues in Ireland were flat instead of up 7 per cent as reported last week. C&C’s group finance director and COO said the error in last week’s announcement occurred after data were incorrectly transferred from an accounting system used for internal guidance to a spreadsheet used to produce the trading statement. “It was basically human error… there’s nothing wrong with our accounting systems,”
| Identifier: | NS02 |
| Source: | FSA fines Credit Suisse £5.6m |
| Organization: | Credit Suisse |
| Release Date: | Aug-08 |
“2.33.3. The booking structure relied upon by the UK operations of Credit Suisse for the CDO trading business was complex and overly reliant on large spreadsheets with multiple entries. This resulted in a lack of transparency and inhibited the effective supervision, risk management and control of the SCG {Structured Credit Group}”
| Identifier: | NS03 |
| Source: | Budget discrepancies attributed to computer error |
| Organization: | Nevada City |
| Release Date: | Jan-06 |
| Discrepancies: | $5 million |
| Risk: | Delay in city budgeting process, embarrassment |
| Avoidance: | Change tracking, comparison, validation, publishing a PDF of printed version on the web |
The Nevada city budget spreadsheet apparently worked correctly until sometime in late December 2005 when, city finance director Ron Chandler says, it developed a problem, causing the 2006 budget to show a $5 million deficit in the water and sewer fund. Chandler said that it took him most of the day Wednesday to fix the problem. While he was working on it he found some other errors in the spreadsheet that needed to be corrected. “Once it’s a PDF it can’t change,” Chandler said.
| Identifier: | NS04 |
| Source: | Westpac jumps the gun on profit |
| Organization: | Westpac |
| Release Date: | Nov-05 |
| Discrepancies: | Around $10 million |
| Risk: | Public embarrassment, loss of investor confidence |
| Avoidance: | User training on how to remove data properly |
Westpac was forced to halt trading on its shares and deliver its annual profit briefing a day early after it accidentally sent its results by email to research analysts. Details of the $2.818 billion record profit result for the 12 months to September 30, were embedded in a template of last year’s results and were accessible with minor manipulation of the spreadsheet. (Some reports indicated an employee had thought that a black cell background fill would hide black text.) Mr Chronican said “It is not just one error, it is a compounding of two or three errors … We will obviously be conducting a full inquiry to make sure it doesn’t happen again.”
| Identifier: | NS05 |
| Source: | Kodak |
| Organization: | Kodak |
| Release Date: | Nov-05 |
| Discrepancies: | Around $10 million |
| Risk: | Loss of share value, investor confidence, career damage |
| Avoidance: | Data quality control |
SAN FRANCISCO (MarketWatch) – The hefty $11 million severance error was traced to a faulty spreadsheet. Kodak spokesman Gerard Meuchner said “There were too many zeros added to the employee’s accrued severance. But it was an accrual. There was never a payment,” he said. Robert Brust, Kodak’s chief financial officer, called it ” an internal control deficiency that constitutes a material weakness that impacted the accounting for restructurings.”
| Identifier: | NS06 |
| Source: | Go figure: ‘Think-and-do tank’ flubs the math |
| Organization: | Virginia Tech |
| Release Date: | May-05 |
| Risk: | Public embarrassment, adverse press coverage |
| Avoidance: | Spreadsheet review |
The Center for Regional Strategies recently confirmed that a researcher’s errant cut-and-paste from a spreadsheet caused one measure of the region’s level of educational attainment to appear a lot worse than it is. The Center for Regional Strategies, a self-described “think-and-do tank” housed at Virginia Tech, reported that a dismal 11 percent of the region’s population older than 25 had bachelor’s degrees or higher. That number should have been 20 percent. Stuart Mease, a spokesman for the Center for Regional Strategies, said “It was just a simple cut-and-paste error,” he said. “I don’t know how it happened, but it did. We apologize for our mistake and want to correct it.”
| Identifier: | NS07 |
| Source: | RedEnvelope skids on loss forecast |
| Organization: | RedEnvelope |
| Release Date: | Mar-05 |
| Risk: | Loss of share value, investor confidence, career damage |
| Avoidance: | Data quality control |
Shares of RedEnvelope Inc. tumbled more than 25 percent Tuesday after the online retailer of specialty gifts drastically reduced its fourth-quarter outlook and said its chief financial officer will resign in April. Stanford Group analyst Rebecca Jones Kujawa said in an interview. “…they were underestimating the cost of goods sold….it is likely CFO Eric Wong is being pushed out because of this error, which could demonstrate a material weakness in controls over financial reporting, an issue that usually leads to a lengthy review of accounting practices.” RedEnvelope spokeswoman Jordan Goldstein said the budgeting error was simply due to a number mis-recorded in one cell of a spreadsheet that then threw off the cost forecast and was unrelated to the CFO change.
| Identifier: | NS08 |
| Source: | University of Toledo loses $2.4M in projected revenue |
| Organization: | University of Toledo |
| Release Date: | May-04 |
| Discrepancies: | $2.4 million |
| Risk: | Loss of budget |
| Avoidance: | User training and certification |
University of Toledo loses $2.4M in projected revenue. “While official UT projections call for a 10 percent decline in graduate student enrollment, an increase mistakenly was shown in a spreadsheet formula that led officials to overestimate enrollment and therefore revenue, Mr. Decatur said.” “Dr. Johnson said no job action will be taken against the employee who made the mistake, who has a good performance record. Officials will, however, pursue systemic changes to provide more safeguards in the future. “We have very competent people,” Dr. Johnson said. “I do think that the continuing fiscal pressures on universities have forced us to a level of staff support where there is little or no redundancy in the process.”
| Identifier: | NS09 |
| Source: | Fannie Mae Corrects Mistakes In Results |
| Organization: | Fannie Mae |
| Release Date: | Oct-03 |
| Discrepancies: | $1.2 billion |
| Risk: | Loss of shareholder confidence |
| Avoidance: | Peer review |
Fannie Mae filed a Form 8-K/A with the SEC amending their third quarter press release to correct computational errors in that release. “There were honest mistakes made in a spreadsheet used in the implementation of a new accounting standard. The bottom line is that the correction has no impact on our income statement, but resulted in increases to unrealized gains on securities, accumulated other comprehensive income, and total shareholder equity (of $1.279 billion, $1.136 billion, and $1.136 billion, respectively)… the correction had to do with a computational error in performing complicated calculations required in the implementation of FAS 149.”
| Identifier: | NS10 |
| Source: | TransAlta Says Clerical Snafu Costs It $24 Million |
| Organization: | TransAlta |
| Release Date: | Jun-03 |
| Discrepancies: | $24 million |
| Risk: | Huge financial loss |
| Avoidance: | Peer review, double-checking outputs |
TransAlta Corp. said on Tuesday it will take a $24 million charge to earnings after a bidding snafu landed it more U.S. power transmission hedging contracts than it bargained for, at higher prices than it wanted to pay. […] the company’s computer spreadsheet contained mismatched bids for the contracts, it said. “It was literally a cut-and-paste error in an Excel spreadsheet that we did not detect when we did our final sorting and ranking bids prior to submission,” TransAlta chief executive Steve Snyder said in a conference call. “I am clearly disappointed over this event. The important thing is to learn from it, which we’ve done.” As New York ISO rules did not allow for a reversal of the bids, the contracts went ahead.
| Identifier: | NS11 |
| Source: | Allied Irish Banks / Allfirst Currency Trading Fraud |
| Organization: | Allied Irish Banks / Allfirst |
| Release Date: | Mar-02 |
| Discrepancies: | $700 million |
| Risk: | Fraud, huge financial loss |
| Avoidance: | Peer review, double-checking outputs |
Allfirst “Would not pay the US$ 10,000 for a direct data feed from Reuters to the risk control section”. Instead, they got Rusnak to download his Reuters feed into a spreadsheet. He then substituted links to his private manipulated spreadsheet. The total losses hidden by the fraud were almost US$700M. Rusnak exaggerated bonuses by over half a million dollars. Ray Butler points out in the article above that “One error in a spreadsheet will subvert all the controls in all the systems feeding into it”. An auditing tool (SpACE) would have found the external links in the key spreadsheet.
| Identifier: | NS12 |
| Source: | Jamaican Commercial Bank Failures |
| Organization: | Jamaican Commercial Bank |
| Release Date: | late 1990’s |
“Poor control over spreadsheets at Jamaican indigenous banks contributed to management information and external reporting problems (i.e., P&L distortions) that contributed to the banks’ management and external regulators losing sight of the banks’ true positions and exposures. This problem fed a downward spiral into liquidity crisis”[Lemieux, 2005] “As the Jamaican financial crisis unfolded, the government also recognised that fraud and corruption had contributed to the collapse of indigenous banks. To address these allegations, it established a team of foreign and local forensic auditors to work with the police fraud squad to identify and take action on instances of fraud. Inaccessibility of source documents, however, seriously hampered the auditors’ work. The work of reconstructing what in some cases were very convoluted financial transactions was made extremely difficult by the fact that critical records, many in spreadsheet form, were missing”. [Lemieux, 2005]
| Identifier: | NS13 |
| Source: | Computing error at Fidelity’s Magellan fund |
| Organization: | Fidelity’s Magellan fund |
| Release Date: | Jan-95 |
| Discrepancies: | $1.3 billion |
| Risk: | loss of share value |
| Avoidance: | independent review and comparison of source with output |
“There was a big flap recently over Fidelity’s Magellan fund estimating in November that they would make a $4.32/share distribution at the end of year, and then not doing so. A letter of explanation was sent to the shareholders (including me) from J. Gary Burkhead, the President of Fidelity, including the following pertinent items: “During the estimating process, a tax accountant is required to transcribe the net realized gain or loss from the fund’s financial records (which were correct at all times) to a separate spreadsheet, where additional calculations are performed. The error occurred when the accountant omitted the minus sign on a net capital loss of $1.3 billion and incorrectly treated it as a net capital gain on this separate spreadsheet. This meant that the dividend estimate spreadsheet was off by $2.6 billion…
EuSpRIG Original Horror Stories
The original list, developed by Patrick O’Beirne, is still [here]
Copyright acknowledgement
News items are the copyright of the original authors as cited by the link to the story source.
Editing and comments copyright (c) 2003, 2004 Patrick O’Beirne, European Spreadsheet Risk Interest Group. The contents of this page may not be copied without attribution. We encourage a link to us with this description:
Spreadsheet mistakes – news stories collated by the European Spreadsheet Risks Interest Group. They illustrate common problems and errors that occur with the uncontrolled use of spreadsheets, with comments on the risk and possible avoidance action.
If you would like to contribute to this, email Stories at eusprig dot org with your story. You may find more by searching the web for “Spreadsheet” and these keywords (and their variants such as -ed, -ing, -s and -ly):
anomaly, audit finding, bug, bungled, chaos, cock-up, corrupt, damage, debug, defect, deficiency, disaster, dubious, embarrassment, error, erroneous, Excel, exposed, failed, failure, false, falsify, fault, faulty, fiasco, flaw, foul-up, fraud, fraudulent, fubar, garbage, GIGO, glitch, hell, horror story, inaccurate, inadequate, inadvertent, incorrect, insecure, invalid, invalidated, issue, lies, loss, manipulation, miscalculate, miscalculation, mislead, missing, mistake, missing, misunderstood, nightmare, non-compliance, not correct, oops, overestimate, overstatement, problem, risk, rogue, serious, slip, snafu, suspect, suspicious, trap, typo, unchecked, uncontrolled, uncorrected, uncovered, underestimate, understatement, undetected, untested, violation, weak, weakness, wrong