Figuring it out:
Aha; I’ve found the source for the Oct 2005 DCMS claim that…

“Over the next decade 9.6 million jobs will be created in multimedia and software sectors across the EU” (words from Tessa Jowell’s speech).

The claim seems to arise from estimates in a MKW paper issued in 2001 by the European Union, titled ‘Exploitation and development of the job potential in the cultural sector’. It stated…

“There are currently approximately 1.5 million companies in the EU active in the areas of multimedia and software, representing a total of 12.4 million workers. Assuming a declining [my emphasis] annual growth rate over the next 10 years from 10 percent in 2001 to just 3 percent in 2011, we can estimate 22 million jobs in the year 2011. Thus, approximately 9.6 million new jobs will be created in multimedia and software in the next decade.”

But the report goes on to stress that…

“It goes without saying that not all of these 9.6 million new jobs are suited for cultural workers. Only those jobs that truly require creativity and content [creation] are of interest in this context, not those jobs representing the organisational and technical overhead that is necessary for the creation and and distribution of any product (e.g. secretary, cleaning personnel, pure technicians, etc.)”

… and the report suggests a guesstimate that 70 percent of these 9.6 million might be ‘creative’ jobs, to conclude there could be…

“6.8 million potential new jobs for creative workers.” from 2002-2011.

But MKW reveals that these are ‘weighted’ estimates (we’re not told how they have been weighted) and gives the source of the base data as: “WIMMEX Munich, March 2001”. WIMMEX Munich appears to have been a now-defunct consultancy that carried out a survey of 1,450 larger EU companies between August and December 2000. So there would seem to be three main objections to the DCMS peddling this “9.6 million” figure: 1) these are out-of-date estimates, taken from a relatively small sample whose methodology and weighting has not been fully published; 2) a few EU WIMMEX respondents in 2000 will have been standing in the rubble of the dot.com crash, but most would have been from EU countries in which the internet had not even really ‘arrived’ in 2000; and 3) the estimates were clearly for 2002-2011; which is hardly the “over the next decade” forecast (i.e.: 2005-2015) that was claimed by Tessa Jowell in her recent speech. What’s annoying about this sloppy use of statistics is that even the dubious “9.6 million” claim may now be an underestimate – but we just don’t know. Would it be beyond the wit of man (and the enormous EU budget) to discover just how many EU jobs in multimedia and software were actually created between 2001 and 2006? And from that derive a better estimate for growth in 2007-2017?

There’s also an interesting ‘sort-of’ source reference in another speech by Tessa Jowell this week. She said that:

“In the UK, KPMG predicts 46% employment growth and 136% output growth in the Creative Industries between 1995 and 2015”.

Her claim originated, according to the footnote, from a: “Creative London presentation to DCMS Strategy Seminar.” That’s hardly a sound academic reference, especially since regional creative industries networks are notorious for dodgy statistics. Surely the very well-paid DCMS researchers could have given us a footnote that referenced and dated the original KPMG document? Perhaps their reluctance arises from the thorny question of exactly how KPMG might “predict” for the years 1995-2005, years which are now in the past? I doubt even KPMG are so slick that they’ve invented time-travel. Creative London wouldn’t be peddling projected figures based on antique data from circa 1994-, would they? No, surely not. Or would KPMG be simply multiplying Creative London’s own “predicted growth rate” of a 4.5% annual CI growth in London by 10 years (=45%), and then assuming it relates directly to employment rates and can also be extrapolated across the whole of the UK and all CI sub-sectors? No, surely not. And that “136% output growth” productivity estimate wouldn’t have been artificially inflated because the Office of National Statistics have switched to a new methodology (‘chain-linked’), one that significantly boosts the headline CI productivity and output statistic? No, surely not. We need to be careful that the entire UK productivity baseline doesn’t become skewed by this sort of jiggery-pokery in telecoms and media.