Statistics Canada said the main cause of an incorrect job report earlier this month was a failure by staff to understand that an update to data processing methods triggered wider and unanticipated changes in its flagship survey.
“This change was perceived as systems maintenance and the oversight and governance were not commensurate with the potential risk,” said the in-house report commissioned by Chief Statistician Wayne Smith and published today in Ottawa. “Communications among the team, labor analysts and senior management around this particular issue were inadequate.”
The review follows the Aug. 15 correction of the July employment report to show a gain of 41,700 instead of an initial estimate of 200 new positions. Since then, Finance Minister Joe Oliver and Bank of Canada Governor Stephen Poloz have expressed confidence in the agency’s work.
Statistics Canada said Aug. 15 the mistake was an “isolated incident” that was caused when staff failed to run a computer program that was part of a scheduled update to the survey. Canada’s dollar weakened after the initial job report and strengthened when the corrected figures were published.
The incorrect July figures passed a standard review and were presented to senior management before they were published, Statistics Canada said today.
The first time staff realized there could be a problem was Aug. 8, the morning the data were published, the agency said in its report. Statistics Canada analysts then re-ran their calculations and isolated the mistake.
“Once the error and its magnitude were confirmed to the satisfaction of the Chief Statistician on Tuesday August 12, 2014, a decision was taken, as per the agency’s Directive on Corrections to Daily Releases and Statistical Products, to remove the erroneous data, to notify the public of the error,” the agency said in the report.
The report doesn’t explicitly say why it took three days to publish a corrected report after the bad data was retracted.
“The bigger issue was not that a mistake was made, we all make mistakes, but that it was not caught by their process,” said Doug Porter, chief economist at BMO Capital Markets, by e-mail. It’s “incredibly unfortunate that the error just happened to occur in arguably their highest profile release.”
The report makes five recommendations to prevent a repeat of this kind of mistake, in areas of governance, testing protocols, diagnostics, documentation and communication.
“Only time will tell if the five recommendations properly address the fact that the error was not caught,” Toronto-based Porter said.
“No employee or group of employees acted inappropriately,” Statistics Canada spokesman Fabrice Mosseray said in an e-mail, responding to a question about whether any employees were disciplined. “The report concluded that it was the protocols and procedures that led to the error.”