It is not surprising that Finance Minister Gary Collins was quick to issue a news release boasting about the October Labour Force Survey (LFS). On a seasonally adjusted basis BC gained 30,100 jobs in October, the biggest gain since the survey began in January 1976 (although January 1999 and June 2002 came close). The increase is so remarkable that it deserves close scrutiny. The first point to note is that on a seasonally unadjusted basis BC gained 20,600 jobs in October, which is less than the 22,800 jobs which were lost in September (see graph below). Nevertheless, the seasonally adjusted increase should be closely examined.

At a cost of $9.00 plus GST, you can purchase the monthly Statistics Canada publication titled "Labour Force Information" which provides more details than the numbers published without charge in what they title "The Daily". The detailed data show that despite the one month record job gain, BC lost 3,400 jobs in trade, an area where job growth would be expected to occur if the economy was growing. While 4,200 jobs were added in construction this October, even after that increase there are 7,000 fewer construction jobs than there were October 2002.

Three primarily public sector industrial categories account for 40% of the October job growth, although they account for only 20% of total employment. Education services (2,500 new jobs), health care and social assistance (3,000) and public administration (6,700) together had 12,200 new jobs. It might strain most people's imagination to find that the public sector was leading job growth! When October 2003 is compared to October 2002, education services had 4,800 more jobs and public administration had 12,600 more but health and social services had 14,300 less. In the private sector, the two categories of "transportation and warehousing" and "finance, insurance, real estate and leasing" each added 5,900 jobs in October (11,800 combined). The three primarily public sector categories and those two private sector categories accounted for 73% of the seasonally adjusted October job gain. Self-employment accounted for 9,700 additional jobs in October. Some of those jobs could be in the five industrial sectors discussed here, but to the extent they are not, they explain most of the rest of the job gain.

Statistics Canada also publishes "standard error of change in the two consecutive months". For an increase to be statistically significant at the 99% level, it should be greater than 1.96 standard errors. When job growth is analyzed at the level of industrial categories, the increase in public administration was the only category that was statistically significant. It is possible that the increase was related to fighting forest fires. If that is true, no one should think that the jobs are permanent or that they represent an economic recovery.

Statistics Canada also provides data on employment by age and sex. Young people under age 25 accounted for 15,000 of the new jobs in October. Men age 25 and over got 8,100 of the jobs and women age 25 and over got 7,000 - as a percentage increase for the month in employment, that's 5.2% for youth (0.9% for men and 0.9% for women).

It will be great for BC if the October seasonally adjusted employment numbers are more than a statistical quirk; whether they have staying power and are part of a trend cannot be determined for another month or two. The November Labour Force Survey will be released on December 5th. It will be all the more interesting in light of the surprising numbers in October. By the end of November, before the release of the November LFS, Collins must release his Second Quarter Financial report which will include a revised forecast and economic update. Observers will be watching to see if that document is as optimistic as last Friday's news release. The graph below shows the difficulty in projecting October's results six months ahead - only time will tell. The unadjusted data will not change, but the seasonally adjusted data will be revised by Statistics Canada early next year when it looks back with the benefit of more information.