On July 28th Statistics Canada released the Survey of Employment Earnings and Hours for May. It shows that BC had 1.662 million paid employees in May, a gain of 66,565, or 4.2%, since May 2001. By comparison, in the three year period before the New Era began, June 1998 to June 2001, BC gained 81,608 paid employees, 5.4%. That may make you wonder what Premier Campbell is talking about when he says the economy is improving. Wouldn't "improving" mean bigger employment gains than in the previous three years?

Looking at just the past 12 months, BC's paid employment grew by 1.3%. Using the seasonally adjusted employment data from the Survey we can look at similar same month year to year comparisons as is done in the accompanying graph. In provides the annual growth rates for paid employment from May 1995 through May 2004. There were low points with job loses in June 1996, January 1999 and December 2001. The average annual growth rate since the last election is 1.37%. The average annual growth rate in the three years prior to the last election is 1.77%. Not only is growth in paid employment lower now than it was prior to the last election, it is more than a full percentage point lower than it was just one year ago. If that's "turning the corner", it's a wrong turn.

The Survey makes it possible to look at paid employment by industry. Over half of BC's gain in paid employment over the past three years, 53.7%, has been in retail and wholesale trade. Construction is the next big leader, making up a further 24.3%. Forestry lost 17,733 employees over the three years, and "accommodation and food services", a key component of the tourism industry, lost 2,851 jobs. Premier Campbell has been boasting recently about a turnaround for mining. It is true that higher mineral prices, not to mention soaring natural gas prices, have been a stimulus, but the Survey shows a gain of only 893 paid employees over the past three years in "mining, and oil and gas extraction".

According to forecasts, yet to be tested against reality, BC's economy will show a slight improvement this year with 3.0% real GDP growth compared to 2.2% last year (down from 2.4% in 2002). That is good, but how good depends a lot on where you live and what you do. The key role played by construction and trade suggests that the growth is very sensitive to interest rates. Rising interest rates could be bad news.

The good economic news that is part of the Campbell song sheet doesn't talk about a gain of just over 66,000 jobs over the past three years; it speaks of a gain of over 167,800 jobs since December 2001. What accounts for the difference of over 100,000? Picking December 2001 rather than the election date matters a lot since BC lost jobs during the recession of 2001. Most of the rest of the difference is accounted for by self employment. Gains in self employment are important; we just don't know much about the nature of those jobs. Most doctors, lawyers and dentists are self-employed, so are some gardeners, couriers and cleaners. Some people are self employed because they want to be, others because they have no choice. While the Labour Force Survey doesn't publish data on the nature of the self employed jobs, it does report that in May 2001 there were 356.2 thousand self employed in BC; in May 2004 there were 396.0 thousand. About 60% of the self employed report that they work 35 or more hours per week.

The Campbell government cites data from the Labour Force Survey rather than from the Survey of Employment Earnings and Hours. The Survey of Employment Earnings and Hours is described by Statistics Canada as "Canada's only source of detailed information on the total number of paid employees, payrolls, hours at detailed industrial, provincial and territorial levels." It does not include those primarily involved in agriculture, fishing and trapping, private household services, religious organizations and military personnel of defence services. It uses information from employers, both the Business Payroll Survey results (a sample of 11,000 employers) and the payroll deductions administrative data received from Canada Customs and Revenue Agency, to measure the number of paid employees. The Labour Force Survey, which is more often cited, includes the self employed. It samples 53,500 households using computer assisted interviewing. It is arguable that the Survey of Employment Earnings and Hours provides a more accurate measure of paid employees. Proving the existence of a substantial economic improvement shouldn't depend on which survey you use.