James Coe is Associate Editor for research and innovation at Wonkhe, and a partner at Counterculture

Clusters are at the centre of a significant number of major government R&D policies. They are this big idea that if you put lots of complementary research companies, institutions, universities, and other assets together, they will share expertise, grow, and in turn grow the economy.

The government’s R&D roadmap promises “place-focused innovation” to “provide holistic innovation and growth support to emerging clusters of innovative businesses.”

The response to the Nurse Review mentions clusters no fewer than 34 times with the goal to “improve prosperity and opportunities throughout the UK by nurturing innovation clusters.”

And the Science and Technology Framework promises “accelerated translation, commercialisation and knowledge exchange through targeted support for local innovation clusters.”

Going local

You can’t have a proper city region economic strategy without having something to say about clusters. The Liverpool City Region’s Knowledge Quarter is “creating cluster network effects in vital innovation”; the Tees Valley not only has clusters, it has “sub-clusters of companies in biological reagents, assistive technology, health informatics and pharmaceuticals”; while the West of England is full of “local dynamism and spill-over effects between various clusters.”

Everyone thinks clusters are important but very few people explain how they work. It might be because it’s assumed the clue is in the name. You cluster things together and then good things happen. But the nature of the activity is important.

Typical economic assumptions would hold that all things being equal a business will set up in the place where they can maximise their profits. This might mean a country with particularly low tax rates, somewhere with really skilled employees, or somewhere with access to cheap labour. Key to this assumption is that any geographical limitations like access to raw materials can be circumvented through the wonders of global capital.

The issue with this theory is that the world looks very different. If I owned a film company I might choose to set it up in Hollywood. If I’m a stockbroker or corporate lawyer I might go to the City of London. If I think I have the next Facebook or Apple I might go to Silicon Valley. Instinctively, these places feel somehow right and it’s because they have distinct specialties in a defined area of economic activity; or a cluster. Think of it another way – you rarely choose a holiday destination because of a single attraction but because it has a collection of things to do.

Setting up

Clearly, however, a cluster must start somewhere. Let’s take Liverpool’s life sciences research cluster.

Liverpool is home to a number of universities that are working in and around life sciences. Alongside the NHS the universities provide a publicly funded permanent body that has historically invested in all sorts of ideas into research in this area. The other benefit of a university within a cluster is that they supply a highly trained workforce which is attractive to companies seeking to expand. These staff sometimes create companies, they attract other companies to an area who want access to experts and ideas, and they invest in facilities like labs which benefit lots of companies. Once the cycle kicks into gear, other companies want access to established resources, further state funding becomes a more viable investment, and before we know it a cluster is born.

It is an enormously simplified version of how clusters come into being. Some clusters emerge primarily from the private sector whose companies, spin-outs, and employees encourage agglomerate benefits. Some clusters emerge because of advantageous natural assets like tourism businesses next to the beach.

Getting on

But the important lesson is how to get the right balance of assets, investments and activities to make a cluster viable.

Michael Porter – who along with Paul Krugman is one of the leading authorities on clusters – has meticulously set out some of their benefits. This includes increasing competition through proximity; reducing costs of access to skilled staff; better coordination of investment; simpler access for suppliers to multiple firms; greater utilisation of collective goods like labs; and the opportunities for small companies to access the expertise of larger companies.

If it is possible to broadly dissect what clusters and how they work, then it is also possible to reverse engineer some features of research clusters when establishing them, seeking investment, or expanding their footprint.

A crucial consideration is the complementarity of assets particularly where investment decisions are being coordinated through spending frameworks. For example, in the case of the freeports programme it would not make a lot of sense to encourage too broad a range of R&D businesses to relocate to a freeport. This would displace existing work with no guarantee the companies would have expertise to share. Instead, it would make more sense to consider how targeted investment and incentives could encourage complementary research organisations and businesses to set up within the freeport.

It is also crucial to consider how narrowly a cluster should be defined. A cluster cannot be about all research domains in a single place as it is unlikely there would be enough complementary assets to share ideas and resources. It also cannot be so specific that there are not enough companies or research assets to develop into a cluster. It has to be somewhere in the middle where there can be significant research and development activity within a domain specialism. Again returning to Liverpool, its focus on life sciences captures a range of research activity within a defined field. The developing Pandemic Institute is a more specific research cluster but still broad enough to attract a range of partners and businesses.

Making assumptions

The questions of how a cluster can be encouraged and what it does are important – but it’s also equally important to consider a cluster’s purpose. It will usually be to drive profit but in the case of some university institutes they will often work with companies, spin-outs, and people within their local proximity to access shared resources to improve their research outputs. This may not be appealing to firms that are primarily about short-term profit maximisation but might be appealing to firms that are more focussed on developing IP and long-term research benefits.

The point is that the assumption that clusters are inherently good needs to be unpicked a little bit more carefully. Having complementary research strengths in a single geography has lots of research benefits and by proxy tends to be really good for the local economy too. To take a step back if government policy is going to be more about place based research investment strategies for developing and establishing clusters needs careful consideration of their purpose, the complementarity of assets, the mixture of partners, the relative strength of research compared to other parts of the country, and their spillover benefits to the wider research community.

To develop a cluster is to choose an economic future for a place. In making the case for research organisations at the heart of clusters, we also make the case for research at the heart of our economies.

Leave a Reply