Technology policy — never satisfied
11/12/2012 § 2 Comments
It doesn’t seem like the government can win with technology policy. When it spends on technology R&D and nothing happens, people complain about the waste of money. When a company with a government grant looks like it’s going to be bought out for US$130m, people complain about the waste of money. Damned if you do, damned if you don’t.
I have two contradictory thoughts on this latest tweet-spat. The first is that evaluating tech policy with headline case studies is fruitless. The second is that these people have a point.
The first problem results from policy design meeting sample selection bias. The specific grants in question — grants to businesses to help them develop R&D — are intended to help businesses become more commercially successful. The idea is that (a) people under-invest in R&D because of public-good aspects of research, and (b) people in business have a better sense of what might be commercially viable than academics or bureaucrats, so (c) government should provide money and businesses should provide direction (and more money) and the economy will be better off.
R&D is a chancy thing. It might work, it might not. As a result, many initiatives fail, some muddle through, and the odd one does really well. When you give the grant, or better yet, when you establish the funding policy, you don’t know what’s going to work. It’s easy to look back and say, ‘given what we know now about the value of the company, maybe it didn’t need that grant after all.’ It’s harder to make that decision at the time the grant is given. In fact, the more grant recipients are successful, the more it calls into question the grants themselves. Perverse, eh?
As it happens, I’ve done quite a few case study assessments of technology programmes and companies. You can learn a lot from case studies, because they allow you to delve into how programmes develop over time. One of the key tasks is selecting the right ones to analyse. Obviously, the ‘right’ one depends on what you are trying to achieve. You always have to remember that there is a selection bias — it isn’t a random sample.
My second thought, though, is that Selwyn Pellett has a point, and we should keep debating our tech policies. One of the main reasons to provide late-stage R&D funding is to build New Zealand’s technology capacity. We are trying to create self-propelling agglomerations of technology companies, and these grants are supposed to be getting the process started. If, in fact, they are just funding single companies until they can be sold offshore, then we aren’t meeting our goals. And, given the lack of capital gains tax, we aren’t even getting our money back.
I’m not sure that taking an ownership stake is necessarily the solution, either. That could get quite complicated, and I’m not sure that the government has the skill to be a good shareholder in a bunch of small tech companies.
We do need to know whether these grant policies are ‘value for money’. Some analysis has been done: one on the Research for Industry Fund (pdf) comes to mind. Most analysis, though, is piecemeal — this specific institution, that particular research programme. Analysis that is more sustained and comprehensive would greatly inform the debate.