"Lies, Damn Lies and Government Statistics"


Location Date: 
November 15, 2011
Content: 

It was Mark Twain who coined the phrase, “Lies, damn lies and statistics”. Without a doubt, statistics are often used as a form of deception because the general public can be ignorant of their construction and true meaning. After all, how many readers remember the enjoyable hours studying advanced statistics? Okay – maybe a few actuaries, but not many! I ended up taking courses in advanced econometrics in school and am very happy to have those days long behind me.

So why the post? One of my good friends, John Cook from BPI Financial days and current President of Greenchip Financial www.greenchipfinancial.com and a Managing Director of Investeco Capital Corp. www.investeco.com dropped me a line the other day. He was concerned that Canadian Finance Minister Jim Flaherty was going to start working on a more ‘specific’ mandate for the Bank of Canada. Why? Well, sceptics will say it is because reported core CPI is now at 2.2% and over the 2% target. In fact, the overall headline CPI in Canada rose to 3.2% this September – this is the one you and I have to live with in our daily lives. John also flipped me an old article from Harper’s magazine (click here to read) that tied political expediency and economic statistics.

The concern is that whenever an ‘issue’ presents itself as a political hot potato, politicians will do their best to bury it for the next guy to work out or “kick the can” further again. In fact, if we look at the key statistics to measure economic performance, our minds inevitably turn to CPI, GDP and unemployment. And this is where a problem starts – the way we measure these statistics has been substantially changed over the past 30 years. The most authoritative voice in the U.S. on the subject is John Williams (no, not the Star Wars composer!). His website www.shadowstats.com offers up adjusted figures over time on all three measures that paint a truly different picture from that presented in the financial media. We will get to Mr. Williams in a minute. 

So is this some grand conspiracy? Not really – this is a story of “Pollyanna creep” or consistently changing the rules subtly and, in most instances, at the behest of the highest political figures. In the case of why we look at “core” versus “headline” inflation now, we have to go back to Richard Nixon. He asked then-Fed Chair Arthur Burns to get rid of the “volatile” components of CPI (food and energy) in the mid 70’s, as they were politically embarrassing. 

The CPI figures continued to be distorted. In 1987 the BLS (Bureau of Labour Statistics) started substituting homeowner costs with something called “Owner Equivalent Rent” to calculate CPI. Why? The BLS thought rising costs were overstating the CPI!! This move lowered CPI substantially and some have linked this change as a factor in the creation of the housing bubble in U.S... In the 90s, we had three other downward adjustments in the CPI calculation and all dubious. Quoting Kevin Phillips in Harper’s magazine:

“product substitution (if flank steak gets too expensive, people are assumed to shift to hamburger, but nobody is assumed to move up to filet mignon), geometric weighting (goods and services in which costs are rising most rapidly get a lower weighting for a presumed reduction in consumption), and, most bizarrely, hedonic adjustment, an unusual computation by which additional quality is attributed to a product or service”. 

Of course, the big beneficiary of these changes is the government itself – social security outlays, pension indexing, etc. are linked to the CPI. Additionally, if nominal rates are lower, the burden of government debt (and all other debt for that matter) is reduced, so a larger amount of debt can be carried and financed. Exactly the situation the U.S. and others find themselves today. 

Unemployment is a bit more ‘clean’ but still plenty of distortions. President Clinton took his opportunity to fudge the employment data. Quoting Phillips again: 

“in 1994, the Bureau of Labor Statistics redefined the workforce to include only that small percentage of the discouraged who had been seeking work for less than a year. The longer-term discouraged—some 4 million U.S.adults—fell out of the main monthly tally. Some now call them the “hidden unemployed.” For its last four years, the Clinton Administration also thinned the monthly household economic sampling by one sixth, from 60,000 to 50,000, and a disproportionate number of the dropped households were in the inner cities; the reduced sample (and a new adjustment formula) is believed to have reduced black unemployment estimates and eased worsening poverty figures”.

The BLS has 6 different measures of unemployment – U1 to U6. The media and government talk about U3, but the economic reality is U6. But even here there are issues that are too complicated to get into for this piece. One example is the 'Birth – Death' model. Many commentators simply add/subtract the implied birth/death figures from the headline numbers to get net unemployment, which is wrong. So even the 'experts' are confused! Our view would be to look at the widest measures of unemployment possible as they reflect the “true” reality.

Back to Williams. Feel free to visit his website as a good part of it is free to browse. I have posted below his versions of where CPI, unemployment and GDP are in his view based on a consistent measurement over time. Needless to say, the picture is considerably bleaker than even the headlines.
 

 

 

Jim McGovern

P.S. For statistical comic relief - 

Steven Wright: "And remember 47.3% of all statistics are made up on the spot."

and George Carlin: "Just think of how stupid the average person is, and then realize half of them are even stupider!"