December 12, 2013
Late last month former Bank of Canada governor and recently-appointed Bank of England governor Mark Carney surprised many by criticizing the quality of the economic data being provided by the UK’s Office for National Statistics. Measurement of investment in Britain was one source of discomfort for the former Goldman Sachs insider, who said that “We’re not putting full weight on that data and it has to be said that it doesn’t entirely feel right that investment is, as measured, falling at a time when we see continued strengthening investment intention.” He added that “a lot of work” would have to be done to bring Britain’s flow of funds data up to international standards.
To be clear, what was surprising was not that the ONS data is incomplete and misleading; everyone with an ounce of economic understanding already knew this. What is surprising is that a high-ranking bank official was admitting that the data they and the market in general is relying on is so unreliable.
But perhaps this shouldn’t be surprising. Back in 2007, Li Keqiang admitted that China’s GDP data is “manmade” and “unreliable.” Then head of the Communist Party in Liaoning province, Keqiang reportedly told U.S. Ambassador Clark Randt that China’s GDP numbers were for “reference only.” Keqiang is now the Premier of China, but oddly has clammed up about the country’s fudged statistics since taking the office.
More common than blatant admissions that the data is made up are “revisions” to the data that take place weeks or even months after their original release. That happened earlier this week when the Japanese government quietly revised their 3rd quarter GDP growth figure to 1.1%. Last month, they touted a 1.9% growth rate, widely reported at the time as a respectable figure for the perpetually moribund Japanese economy. In the end, most people only ever see politicians gloating over the latest, greatest stats as they are released. Few if any notice the revised figure weeks later when it is buried in the back of the newspaper.
Perhaps the most obvious example of this phenomenon in the American context is in the unemployment figures. By manipulating the criteria for being considered “unemployed,” the Bureau of Labor Statistics is able to keep the numbers artificially low. ShadowStats.com analyst John Williams calculates that if the government was still using the same standards it was using in 1994 to come up with their employment numbers, the current unemployment rate would be over 23%. This is most clearly seen in the labor participation rate, the percentage of workers actually in the workforce, which reached 35 year lows earlier this year even as the government’s cooked unemployment number continues to drop.
Keep this in mind the next time a panel of talking heads try to convince you of how wonderful the economy is based on the latest government-supplied official statistics.