The post-war Welsh economy was initially based upon agriculture, coal and metal manufacturing and processing. We then had a growth of light manufacturing and, by the late 1980s, Wales was attracting more foreign direct investment than any other region or nation in the UK. This flowed particularly into light manufacturing.

A combination of proximity to markets, cheap labour, and often direct Welsh Development Agency support, meant that suppliers looking to establish new plants found Wales offered a compelling package for investment. Unfortunately, many of these branch factories closed when either markets reduced or it became cheaper to relocate elsewhere. Whole plants moved from Wales as manufacturers shifted to cheaper locations. One hundred and seventy one plants closed between 1998 and 2008, with job losses amounting to 31,000.

It would clearly be a mistake for Wales to fixate on winning new flows of overseas investment. Some may, of course, be attracted, but it is extremely unlikely that anything approaching the golden age of the 1980s and 1990s could, or should, be replicated—attempts to do so by either reducing wages or reducing environmental or safety laws are entering a race that Wales cannot, and I believe should not, enter.

Using the most recent business register and employment surveys that I was able to access shows that the Welsh economy employs substantially fewer people in the following areas classified by the Office for National Statistics: information and communications; finance and insurance; professional, scientific and technical; and business administration, including support services. I think that if you were to make a list of those areas that paid the most in the economy you'd have most of those on it. There are structural weaknesses in the Welsh economy—if there weren't, our gross value added would be higher. Some solutions to Welsh economic problems have to be reached. Firstly, there needs to be an upskilling of the economy department. We cannot have another Kancoat, where those assessing the scheme cannot see that colour coating is not advanced manufacturing, and do not know that two colour-coating plants had previously failed in south Wales and that Shotton colour coats steel.

Secondly, we need to learn from what previously worked and worked well. In the 1990s, Wales experienced a golden age of animation production. S4C was integral in helping to produce a number of popular animated shows, which I mentioned a couple of weeks ago: SuperTed, Fireman Sam, The Gogs. These were high profile and they actually got translated into English. S4C rapidly gained a reputation, nationally and internationally, for their commercially successful and award-winning animation programmes. As S4C expanded, a whole host of other popular characters came: Wil Cwac Cwac, Toucan 'Tecs, Funnybones. This is an example of where money put into the demand side generates new industry, but needs initial demand. Far too often, we fixate on putting money into the supply side and then we oversupply—we need the demand side to be up there as well.

On a further thing: it must seem to you all rather odd that the world's biggest-selling video game, Grand Theft Auto, started life in Dundee, but there's an economic logic to the way that, in 20 years, the Scottish city has become a notable cluster for the world's video games industry. Abertay University in Dundee found political and financial backing to establish a department offering the first computer games degree in the world in 1997. There are now a clutch of related degrees, including games design and production management. There's also a large computer games industry based in Dundee. Is anybody surprised at the fact—money has been put in, it was there at the beginning—that it now has a very successful industry in an area that pays very well and that is growing? This outlines three points that I continue to make—and I'll probably make them again this time next year—first, a university can be a driver of the economy and the local economy; secondly, that industries tend to cluster; and, third, geography is less important than skill availability. If you're ever picking a place to locate—no disrespect to Dundee, but it isn't top of anybody's list of geographically the most central.

Thirdly, like Gordon Brown, I'm a believer in endogenous growth theory, which holds that investment in human capital, innovation and knowledge are significant contributors to economic growth. The theory also focuses on positive externalities and spillover effects of a knowledge-based economy, which will lead to economic growth. The growth theory primarily holds that the long-run growth rates of an economy depend on policy measures—for example, subsidies for research and development or education increase the growth rate, using the endogenous growth model, by increasing the incentive to innovate. And that's what we lack, really. We talk around it, but what we have not got is enough highly paid jobs, we haven't got enough innovation and we haven't got clusters in high-paid industries. If we can accept what that problem is, then we have to work out how we're going to address it.

I've tried to outline, briefly, some ways of improving the Welsh economy, and I'm, of course, available to discuss these in greater detail with the Welsh Government or anybody else who wants to talk to me about it, but we really have to become innovators—we really have to get into the high-value parts of the economy. You're not going to become rich by having lots of employment with low pay. We can knock our unemployment rate down, we can increase our employment rate, but it doesn't do much for our GVA until we start getting some highly paid jobs in there.