I would however like to see more concrete examples of Contingent Value/Willingness to Pay arts value studies. Has it been done before? And what are the numbers that came out? I know you mention the Australian study (which has numbers), but it only measured the value of the overall state expenditures. Is there a study out there that measured the value attached to a given museum or symphony, or is CV/WTP only useful for the total amount of public spending in the arts?

A sub-question: if you can measure the public value of specific art disciplines/venues/artists, and find out that people are willing to see public money go to, say, Céline Dion, should legislators go that way?

The real problem here, as Alan acknowledges, is the absence (to my knowledge at least) of relevant, robust and sufficiently contemporary evidence.

If it were possible to wave a wand and fund just two properly resourced pieces of research right now, I would nominate the following:

1. following the Work Foundation’s 2007 report “Staying Ahead”, a systematic analysis of the economic “spill-overs” between the subsidised performing arts and the wider commercial, cultural and creative industries -especially, though not exclusively, the copyrightable content sectors (film, music, theatre and so on) so that we can have an informed discussion about aggregate returns on public investment in the arts: and
2. a UK study of the business and economic terrain covered by Prof. Richard Caves in his book on “Creative Industries: contracts between Art and Commerce” (Alan alludes to this). Caves’ unique and penetrating analysis is based almost exclusively on American evidence, and, having been published in 2000, is largely pre-digital in scope. This is a very serious qualification because of the huge impact of digitisation on value chains in media and creative markets. Notwithstanding this reservation, and this is a measure of how far behind I think we are in the UK, Caves’ study makes it possible for us to understand the economics of creative businesses, and the commercial dimension of the creative economy, in a way that is essential to getting the balance right between “instrumentalist” and “intrinsic” approaches to policy-making.

PS Maybe someone out there has a wand?

First, every research method can be manipulated or used badly. Scandals regularly erupt in the natural sciences – witness the Lysenko case. If CVM is seen as the latest ‘toy for getting money’ it will be discredited and we will have failed. It should be done well – hence, for example, the idea of a DCMS centre of Excellence.

Second, CVM has come of age. There is a ‘right’ way to do it, and instruments to protect against the ‘wrong’ uses Martin describes. It should be applied.

Most tellingly, however, I’m far from convinced that ‘instrumental’ measures offer anything better. After six years in the business of getting the numbers I would say we probably know what they are, and can even guess at what they mean – but we don’t know how they work. We know the creative industries are linked to innovation, employment, or growth – somehow. But the devil is, we don’t know how. Beyond the tacit knowledge of creative entrepreneurs, and the derived expertise of business-focussed studies like Richard Caves, no proven method exists to find out what really goes on, economically, when money goes into a new creative venture.

I’m going to stick my neck out (for discussion) and suggest that arts policy-making based on instrumental benefits is under these circumstances pseudo-objective. It looks great on paper, but no sound causal reasoning supports it.

That’s not to say instrumental benefits don’t exist. DCMS and NESTA research into finding them remains vital. A model of what should be aimed at is our growing understanding of the mechanisms of climate change. Basically, as the Stern report showed, we know enough to put a number on the human cost of getting emissions wrong. But we can’t do the same for the arts. In our present state of knowledge, we cannot reliably predict whether this-or-that arts investment will produce that-or-this instrumental benefit, or even if it will succeed.

We can however get a reliable answer to the much simpler question ‘what is the value of art?’ It’s not a perfect answer, but it is, actually, more objectively-grounded than the instrumental answer. So why not use it?

Three initial comments. First, most contingent valuation/stated preference surveys that I have come across were intended to promote, even commissioned in order to promote, a pre-conceived outcome. Such valuations do not in my experience constitute the pure, unadulterated milk of objectivity that, or so the authors imply, would give us empirically unassailable bases for policy-making.

In short, such surveys can be manipulated. This is not necessarily the case of course, and larger measures of objectivity are achievable, but designing in the required methodological checks and balances is expensive (and often unwelcome!). Jim Barrett is therefore right to say that “it’s naïve to assume you can ever divorce research from advocacy in the policy arena”.

Secondly, I find the authors’ enthusiasm for “commensurability”both counter-intuitive and somewhat disturbing. I know that economists adore data, but do the arts want to end up in such a utilitarian place? I’m not comfortable with this way of thinking unless it is heavily qualified – as, to be fair, it sometimes appears to be in the text of the pamphlet.

Finally, and here I do of course reveal conclusively that I am not a “lover” of economics, for which I can only apologise, I know of no economic theory that comes remotely close to expressing the “intrinsic” value of a great performance of Bach’s “St Matthew Passion”, or for that matter of Bernstein’s “West Side Story”, or their capacity for enriching, even changing lives. That gives me a nagging problem with the use of the word “intrinsic”.