Is Job Quality Really Improving?

The widely publicized CIBC Employment Quality Index would have us believe that average job quality has been improving this year even as the pace of job creation has slowed down, and even as the national unemployment rate has risen from 6.0% to 6.2%. http://research.cibcwm.com/economic_public/download/eqi-cda-20080716.pdf

I wonder if this is due to the realities of the job market, or due to the peculiarities of  the way in which this rather opaque index is constructed.

In point of fact, according to the Labour Force Survey, almost two-thirds of the net new jobs created between December, 2007 and June, 2008  – 83,500 of a total of 127,000 – were part-time jobs.  It’s true that self employment fell by 31,500 over the same period while paid employment rose by 158,000 – but its a mystery to me why a modest shift from self employment to paid jobs in a context of very disproportionate part -time job growth would register as an improvement in overall job quality.

The Index apparently shows that net job growth has been concentrated in relatively well-paid industrial sectors.  True to a point, but that’s  mainly because of  the relative health of spending on public and social services – employment in public administration and health care and social assistance combined is up 42,500 since December, representing more than one in five of the new jobs – plus continued job growth in construction (up 58,000 from December to June.)   However, the construction sector job boom seems to be coming to an end, with 16,000 jobs lost last month alone.

The key message from the Labour Force Survey data is that unemployment has been rising, job growth has been slowing, and new full-time jobs are very thin on the ground.  It takes a strange kind of index to turn that into a good news story.

One comment

  • Interesting that you post this story. I am in the midst of compiling a literature review on Employment quality indexes and job quality. Not much on available on the employment quality (more of a macro measure) but plenty on the second. (which ends up being more focused on the micro aspects in defining the quality of a specific job).

    My objective and I have been working on this for a bit is to develop a measure similar to the CIBC index, with one qualification: this index will strive to be provide a functionally accurate profile of employment quality.

    In achieving this objective, the index will strive to be cost effective and timely, i.e. use off the shelf, LFS and SEPH monthly data.

    In setting up this index I will have to perform a bit of historical research to set up many of the input variables.

    The index will be constructed from a range of input parameters and full documentation of the methodology will be released with the index. (unlike the CIBC methods, where Mr. Tal refuses to send in any documentation on how he creates this index.)

    Plenty of testing and calibrating will take place to try and determine the relationship between employment quality and other key variables.

    The need for such a measure has been recognized mainly within the progressive community. But for some reason, the CIBC is the only index I have seen in practise. Strange really given the discourse on the breakdown of the traditional the mass production era employment models.

    Given the realities of the continued and increasing precariousness of employment one would have thought that the public institutions such as Statcan, would have embarked upon such a project. Given the huge policy implications, it is like steering a boat without charts. My thought is, if they won’t build and legitimize it, then let’s see what we can do in the progressive community to counter a undocumented and unproven EPI such as the CIBC’s.

    The European Trade Union Institute for Research, has embarked upon a similar project. I will investigate further to examine their methodology and ideas. Anybody have a contact their?

    (see here

    http://www.eurofound.europa.eu/ewco/2008/06/EU0806019I.htm

    Paul

Leave a Reply

Your email address will not be published.