Walter John Williams '71 Tu '72, a consulting economist who co-founded the website Shadow Government Statistics in 2004 and has recently been cited in several national news outlets, has been a longstanding critic of U.S. economic statistics, which he said understates the U.S. rate of inflation and the real unemployment rate, while overstating economic growth.
Williams, who has been a private consulting economist for 30 years, said he discovered early in his career that in order to make meaningful forecasts about the economy, he had to fully understand how the government's statistics were being reported and any "unusual features" they may have.
He began researching the quality of government statistics in the 1980s, when changes made by the U.S. government in calculating the gross national product "eventually made the underlying data worthless" for one of his early clients, according to the Shadow Government Statistics website. He talked to key individuals who had been involved in the system since 1928, when Herbert Hoover was the head of the Commerce Department.
Approximately 70 percent of the business economists he surveyed thought that the statistics' quality was a severe problem. Changes were introduced and concepts were redefined to make economic growth seem larger and inflation seem lower, he said.
"The crux of the issue is that I found over the years that common perceptions with what an average guy was looking at with inflation or unemployment or growth of the economy have increasing variants with what the government was reporting," Williams said.
Williams is now trying to adjust the economic statistics to a level that reflects public perception of the economy because, in general, "the average person has a much better sense of what's going on," he said.
He highlighted two major issues he found with the methodology adopted by the Bureau of Labor Statistics, which he said misrepresent the level of inflation.
The first adjustment is substitution, which means that if the price of a good goes up, consumers will substitute it with a cheaper product, thereby decreasing the inflation rate.
The second adjustment is hedonic, or quality, adjustments, which account for the release of higher quality products. If an updated product is sold for the same price as an older version, the government statistics record the price as lower because consumers get a better product for the same amount of money. Statistics take into account items that cannot be directly quantified and thus misrepresent the level of inflation, he said.
If the Bureau of Labor Statistics had not changed its methodology, inflation would be increasing by approximately 7 percent annually, causing the price level to quadruple in just over a decade, Williams told Bloomberg News. This 7 percent discrepancy means social security payments should be "almost double" what they are today, he said.
Real unemployment, which was 8.2 percent in March 2012, should be approximately 22 percent when those who have given up on searching for jobs are added into the equation, Williams told Bloomberg News.
Williams said that those on Wall Street and in the corporate world "know what's going on," but they paint a better picture that stands in contrast to the reality people are facing.
The Bureau of Labor Statistics addressed both the issue of substitution and quality adjustments in a 2008 article in the Monthly Labor Review co-authored by Bureau of Labor Statistics research economist John Greenlees and Bureau of Labor Statistics Division Chief Robert McClelland.
Greenlees and McClelland said that the Bureau of Labor Statistics "does not assume that consumers substitute hamburger for steak," which is a common argument against the substitution adjustment. The Bureau of Labor Statistics uses a formula that "implicitly assumes a degree of substitution among the close substitutes within an item-area component of the index," which Greenlees and McClelland consider a noticeable occurrence in everyday life, the article said.
The Bureau of Labor Statistics adjusts for quality change because "the number and types of goods and services found in the market are constantly changing," according to the article. Serious upward or downward Consumer Price Index biases would occur if the statistics treated all new products the same as the older versions, the article said.
Many past critics of the CPI claim that inflation was being overstated as opposed to understated. A Congressional review committee formed in 1995 and headed by Stanford economics professor Michael Boskin found that the flaws in CPI cause it to be overstated by 1.1 percent annually, according to Bloomberg News.
Williams said that much of his way of thinking "came out of the conversations I had with the people up there in Hanover."
