London briefing/Chris Johns: Much of what happens to economies can simply be put down to chance and coincidence. Just how much influence we have over our economic future is still a matter of furious controversy.

Here in the UK there is a multiplicity of government agencies all acting to try to affect economic outcomes in regions and specific industries. European countries typically have even more people than does the UK who spend their working lives trying to make a better future for the rest of us. But how much can they really achieve?

A stark example of how sheer luck - good and bad - can affect a region is to be found in the South Wales valleys. The Rhondda and surrounding valleys used to comprise a thriving community based on the production of coal. Output of coal peaked, we think, just before the first World War, and there has been a steady decline to the point today where there is just one small deep mine left operating.

It is tempting to think that such natural resource-based economic developments happen simply because something is found (oil, gas, gold etc) and is then exploited until the stuff runs out. But there is still plenty of coal left in South Wales. So what went wrong?

The answer is to be found in terms of geology and the use to which the coal is put. South Wales coal produces a lot of heat "per lump" and was very good for ships that used to burn coal. As soon as the British Navy switched to oil, that was the beginning of the end for the valleys. Power stations, the only other consumer of coal that matters, don't need the "steam coal" produced by the Welsh mines.

The history of the valleys is a kind of mini-Klondike story. Before coal started to be mined, relatively few people lived there. Population peaked roughly at the same time as coal production and there are now many towns and villages - but by no means all - whose sad condition reflects decades of economic decline and outward migration.

Attempts to slow or even reverse the decline have been only partially successful. Some jobs have moved into the area, for which everyone is grateful but the problems that remain are large ones - and are not just economic. With high unemployment comes the inevitable list of social ills that always seems to follow.

One bleak conclusion reached by some people is that it is inevitable that the valleys will ultimately revert back to their pre-coal state and that government interventions (including EU money) to try to stimulate the regional economy are merely delaying the inevitable. Similar debates surround many other relatively deprived regions across Britain and the rest of Europe.

I am convinced that fate has nothing to do with the future, economic or otherwise. There is nothing inevitable about the ultimate demise of the valley communities. Luck might come to the rescue: Cardiff might become an even bigger boom town than it is already and people might decide that the staggering natural beauty of the valleys makes them attractive locations for commuter towns.

That will require, among other things, even more investment in transport links between the valleys and the coast, but that is where policy has a role.

Decline is possible or even probable, perhaps, but there is some evidence to suggest that well-designed policies can make a difference. Unfortunately, there is also plenty of evidence, globally, that clearly demonstrates that badly designed policy can make no difference at all. And there is, sadly, much more of the "bad policy" type evidence than there is of the good stuff.

The debate over whether economic policy can work often ignores the "quality" dimension. To observe that bad policy always fails is not the same thing as saying that policy doesn't work. The issue of policy design is critical here. We have far too many doctrinal debates instead of simply asking, what works and what doesn't?

Recent revisions to UK economic data paint a truly remarkable picture. The economy has been doing even better than we thought and there is evidence to suggest that the long awaited rebalancing of activity (away from the consumer) is also under way.

Despite my many criticisms of Gordon Brown I have to acknowledge that something is working. Certainly, people are working. Unemployment remains low and even in South Wales there are encouraging signs of activity rates picking up. The UK is beginning a cyclical expansion at a time of astonishingly low unemployment. Perhaps that, more than anything, should give hope to the poorer regions of Britain. They may be last in the queue for jobs but that line-up is looking very short.