Copyright 2004 National Post, All Rights Reserved National Post (Canada) February 2, 2004 Monday National Edition SECTION: Editorials; Pg. A11 LENGTH: 216 words HEADLINE: 'Lying' statistics SOURCE: National Post BYLINE: Thomas Fox BODY: Re: How Efficient Are Our Governments? Not Very. William Watson, Jan. 29 Mr. Watson should take a refresher course in mathematics before quoting statistics to prove the thesis in his title. He uses "performance per unit of public spending" to measure government efficiency when it measures nothing of the kind. Consider the following model: Suppose in country A, people spend $100 each on health care, paid directly to service providers (doctors, hospitals, etc.). In country B, people pay $100 in taxes to the government, which then pays for all health care. If the health care delivered is the same, Mr. Watson would say the government of country B was horribly inefficient because it spends a portion of the GDP on health care while government A spends 0% of GDP on health care. And if the people of country B get better service, Mr. Watson still says government B is inefficient -- "there isn't much bang for the big-government buck." How can better service for the same cost to the public be a sign of inefficiency? There are, of course, reasonable and accurate ways to measure government efficiency, but this isn't it. As Disraeli said "There are three kinds of lies: lies, damned lies and statistics." Dr. Thomas Fox professor of mathematics Dawson College, Montreal.