Posts Tagged ‘economic development’

Innovation, evidence and industrial strategy

September 1, 2017


[A What Works Centre post that’s good here too.]

Industrial strategy is one of the big issues for the What Works Centre and its local partners and innovation is one of the main themes of industrial strategies in the UK, and around the world.

Public policy plays a number of important roles in supporting innovation — see thisdebate between Mariana Mazzucato and Stian Westlake for a good intro. And as I wrote back in January, it’s equally important that we understand what the most effective tools are.

The good news for the UK is thatwe are — slowly — building an evidence base on what works for promoting innovation, as well as other pillars of industrial policy. What’s more, what we have suggests some current UK programmes work pretty well.


Our latest case study summarises Innovate UK’s programmes of support for microbusinesses and SMEs: mainly grants but also loans, awarded on a competitive basis, either to individual firms, or to promote partnerships with other companies or with universities.

Using standard UK administrative data, evaluators were able to set supported firms alongside similar non-supported companies, then compare how the two groups did. This ‘difference in difference’ approach is one of the methods we endorse, as it meets our minimum standards for good evaluation.

Encouragingly, Innovate UK’s programmes seem to have raised treated firms’ survival prospects (by 14 percentage points), employment (an extra 32 staff on average), and possibly sales too (although this result is less robust). These positive effects are biggest for 2–5 year old companies and those aged 6–19 years old. That is, these programmes seem to have helped innovative firms to scale.


This is another helpful piece of the industrial strategy puzzle, for several reasons.

First, in our innovation evidence review back in 2015, we found lots of evidence that these kinds of programmes raised firms’ R&D — but rather less evidence on growth impacts further down the line. Now we have good UK evidence of those growth and scaling impacts.

Second, we already know that the UK’s R&D tax credit system is pretty effective in stimulating firms’ patenting. We can now add good evidence on grants and loans alongside that.

Third, we can set these innovation findings alongside other evidence on business support programmes — where again, we have a decent stock of UK evidence, with several programmes (e.g. on export support) showing positive impacts.

Finally, it’s reassuring to see that evidence for these types of innovation support programmes in the UK broadly lines up with what we’ve found for OECD countries as a whole. We’ve had a number of conversations with policymakers worried that innovation programmes are very context-specific, so results from one country won’t generalise to others. This may be true in some cases. But for grants, loans and tax credits, what we know suggests that what works across the OECD also works in the UK.


Originally published here on 17 August 2017.

Doing broadband better

June 30, 2017


A What Works Centre post I thought would be good here too.


Our new broadband toolkit takes a look at how policymakers (nationally and locally) can increase broadband takeup.

Why do we care about this? From a social point of view, it’s increasingly clear that decent internet access is a citizen right (especially as many public services are being shifted online). From an economic angle, evidence from our broadband review shows that household broadband takeup can have positive effects on house prices, female labour market participation, employment, firm growth, and economic growth. (Household adoption is also strongly linked to firm adoption.)

With that in mind, the toolkit looks at three different areas where public policy can step into the market.


The first toolkit looks at direct public and public-private provision. EU State Aid rules limit what member governments can do here, but even so there is room to provide networks in rural areas. And if Brexit means these rules no longer apply, the UK can (potentially) make big investments.

We find that direct public provision raises broadband takeup, even in countries like the US where the private sector is already active. We also found a study of an Italian programme that successfully raises takeup in rural areas. Crucially, though, the US evidence suggests that provision on its own is only half as effective as provision combined with info/education on broadband’s benefits.

Frustratingly, it’s hard for us to say much on cost-effectiveness on the basis of the available evidence; although it’s clear that there are huge variations in programme costs across countries. Some of this reflects physical ruggedness (Norway is tougher to pipe than the Netherlands), but there may also be potential to deliver schemes more smartly than the UK currently does.


The second toolkit looks at ‘local loop unbundling’ – essentially, opening up the last mile of broadband networks to all-comers. Currently only BT is obliged to do this in the UK, but in theory, national government could bring cable providers into the scope of the legislation. There’s also the question of access fees and other arrangements.

The argument against LLU is a simple one: firm X has invested in a very expensive network, so why should others get the benefit, and why should X invest more in future if they are forced to open up access? (In BT’s case this is complicated by the company’s history as a state monopoly.) The counter-argument is also simple: the wider benefits to society (via higher broadband takeup) outweigh losses to a single firm, which can in any case be compensated through charging for access.

The available evidence shows a pretty clear impact of LLU on household broadband takeup (it’s less clear for firms). Across the EU, between 2000 and 2010, LLU raised household broadband adoption by 15%. Perhaps surprisingly, the majority of studies also find no evidence that LLU crowds out future investment by the network owner, and in one case, may even lead to upgrading. This seems partly explained by relatively high access charges in most countries.

We’d urge some caution here though. For the UK, the positive effects of LLU have decreased over time, as broadband rollout covers more and more of the population. That might change if future technology gets a lot better, but this seems unlikely any time soon. Setting lower access charges would also bump up the adoption effect, but also risks discouraging future investment.


The third toolkit looks across other tools that Governments have used to nudge providers and users: incentives (loans, subsidies, tax breaks), information campaigns, and demand-side measures such as buying clubs and bulk purchasing.

Sometimes policy works directly with providers (ISPs); at other times the nudges are applied to users (firms and households). Does it matter which? Not from the point of view of take-up: we find evidence that both kinds of measure can be effective. Specifically, we find that loans to ISPs and demand aggregation measures like buying clubs have the effect of raising downstream takeup. For direct-to-user policies, subsidies, training and providing computers are all effective.

For providers, a cross-OECD study finds that a bundle of producer incentives have the effect of raising fibre uptake by 10% (from slower copper networks). On the user side, a US study shows that a partial loans scheme for farms raised broadband takeup by 13-14% points.

Again, it’s not easy to figure out cost-effectiveness, but we estimate that typical household programmes cost £1.1-1.3k per extra connection, with the farms scheme above costing around £3-4k per farm connected.

The UK has strongly pushed programmes like this, notably the last government’s SME broadband voucher scheme. Sadly, this has had no proper evaluation done (something we’d like to see change for future schemes) – although a user survey was published which (perhaps not surprisingly) found that voucher recipients were very happy with £3k off the cost of a fast broadband line …


Originally posted here on 23 June 2017.

Future chat

February 23, 2012

I’m giving a couple of seminars in the next few weeks.

First up, on 5th March, I’m at LSE London talking about cultural diversity and the London economy. This will draw on some of my PHD research on the economics of diversity. It turns out diversity and co-ethnic networks are good for London businesses; the broader effects of immigration on labour markets more mixed. I’ll talk about these results and the policy lessons. Details here.

Next, on 7 March I’ll be giving a talk on Tech City at UCL’s Bartlett School of Planning. I’ll be discussing emerging findings from a project on the East London technology cluster, which I’m leading at the Centre for London. Only a few people are looking at this stuff, so I’m excited to be working on it. I’ll be looking at international experience, what the numbers tell us, and feeding back some of our conversations with local firms. Details here.

If you’re around, come and say hello!

%d bloggers like this: