This is the summary of a paper that was included in a National Endowment for the Arts/Brookings Institution volume published last year. We just uploaded it to the Nesta working paper series - you can download the paper here.

In recent years, academics and consultants have argued that the arts and cultural sector can boost productivity in other sectors of the local economy, through two main mechanisms.

1. By creating urban environments that attract professionals with high levels of human capital and their innovative, high-growth employers.
2. By supplying other parts of the local economy – in particular, commercial creative firms – with new ideas and skills that enhance innovation.

Although these arguments have justified policies for creative place making, urban branding, and public investments in signature buildings and dedicated cultural districts, the evidence base underpinning them is still sparse, and mostly confined to the US. Additionally, little is known about the relative importance of different mechanisms and types of cultural clustering (occupational or industrial) in boosting local productivity.
 
In Capital of Culture, we seek to address these gaps in the literature by building an econometric model exploring the impact of cultural clusters on the productivity of English cities. In doing this, we draw on a well-established body of literature on urban wage premiums and human capital externalities.

Our model tests the impact of cultural agglomeration on worker wages (which act as a proxy for productivity) at the city level. We use three measures of cultural clustering (cultural sector employment, cultural occupations and cultural institutions), constructed from official labour force and business registry survey data, and a unique dataset of almost 5,000 UK cultural institutions from Culture 24. We control for important individual and city level characteristics.

What are our findings?

There is evidence that skilled workers sacrifice higher wages to locate in areas with strong cultural clustering

Our findings confirm that there is a positive relationship between cultural clustering and average wages in English cities: English cities in the 90th percentile of cultural employment clustering have average hourly wages of £12.48, £1.11 higher than the average wage for cities in the 10th percentile. However, once we control for individual characteristics (particularly skills as proxied by an individual’s qualifications), the coefficients for two out of our three measures of cultural clustering in our wage equations (cultural employment and institutions) become significantly negative, while the cultural occupations coefficient becomes insignificant.

This ‘negative cultural urban wage premium’ is consistent with there being a compensating differential. In other words, workers may, other things equal, be willing to take a wage cut to reside in cities with relatively more cultural amenities, as these contribute to its quality of life – its ‘livability’, and ‘lovability’.

Creative cities seem to be more productive

We also use our econometric model to examine the relationship between worker wages and measures of creative clustering (focusing on employment and occupations in commercial creative industries as compared to the arts and culture). In this case, we find evidence of a positive wage premium in ‘creative cities’ even after controlling for individual skills – this is particularly the case for cities with strong creative occupational clustering. Although caution is advised in the interpretation of this finding given the obvious potential for reverse causality (affluent cities attract creative industries), it lends support to those who advocate targeting occupations instead of industies to support urban development.

There is evidence of innovation spillovers from cultural clusters into the commercial creative economy

Finally, we test the impact of cultural clustering on the wages of workers in the local ‘commercial’ creative industries, bearing in mind the literature’s emphasis on knowledge spillovers across related – rather than distant – industrial domains. Here, we find some evidence that creative workers in cities with high levels of cultural clustering enjoy a wage premium, which suggests that not-for-profit arts and cultural sectors may generate knowledge spillovers for the commercial creative economy. Once again, these results should be seen as indicative at best, as the causality could work in the opposite direction (a vibrant arts and cultural scene may emerge in places with more productive creative clusters).

Our conclusions 

The preliminary conclusion from our analysis is that, although English data support the view that there is a relationship between cultural clustering and urban development, that relationship appears to be subtler than is generally acknowledged. In particular, the economic impact of public investments in urban arts and cultural infrastructure may be manifest in improvements in the productivity (and wages) of creative professionals, and may not be associated with higher wages in the wider economy if cultural activities serve as a compensating differential.

Image credit: Handover of the European Capital of Culture from Liverpool to Vilnius and Linz via Eric The Fish at Flickr.