New research on arts participation and economics has been released by the National Endowment for the Arts. I won’t try to summarize everything there, just a few comments:
- Two of the reports are on participation: one asks about who participates in what, the other asks people about why they participated. The data will be interesting for researchers willing to dig: what relationships are buried in those survey numbers beyond what is on the surface? I see each of these reports as a spark to future studies. For example, although the participation data is broken down into four-or-five state regions, it would be really interesting to compare by city or metro, and ask ‘all things equal, how does participation vary with assorted urban environments?’
- I have more confidence in the ‘who participates’ data than the ‘why participate’ data. The former asks a question with a straightforward answer – ‘did you go to a live performance last year or not?’ – there will be some error in the data, people are forgetful after all, but it should be fairly accurate. But the latter asks about motives, whether ‘why did you go to the museum?’, or, in a more complex set of questions, for those who said they were ‘interested’ in some exhibition or performance but in the end did not attend, ‘why not?’. Here the reported results need to be recognized for what they are, with people able to give answers that they think ‘sound right’, but are not based upon actions that either happened or didn’t. Intuitively I trust the result that the interested-but-did-not-attend people were more turned off by not having the time rather than not having the money (a simple look at relative prices, valuing the time cost at the person’s wage rate, would confirm that for the majority of people interested in the arts the time cost is likely to exceed the ticket cost). But I don’t want to say the survey is accurate just because it confirms my prior beliefs!
- The third study is an interesting partnership with the Bureau of Economic Analysis, looking at the arts contribution to the economy as a whole. Note that the ‘arts’ here is more broadly defined – broadcasting, movies, advertising, publishing and retail trade are all included, with broadcasting and film being much larger than traditional arts. Still, good to begin to consider the arts sector along side all the others, for comparisons on growth rates over time.
- Et tu, NEA? ‘Economic impact’ and ‘multipliers’? Argh. Here’s the thing: the numbers on the arts as % of GDP, employment, and capital investments have a logic to them – we know what the terms mean (even if we disagree about who ought to count as an ’employee’ in the arts and cultural sector, we understand what an ’employee’ is). But when the report claims ‘every $1 increase in the demand for arts and culture generates $1.69 in total output; for every job created from new demand for the arts, an additional 1.62 positions are also created‘, I don’t know what that means. Where does the initial ‘increase in the demand for arts and culture’ come from? A reduction in demand for other goods and services? What happened in those sectors as a consequence? An increase in government funding for the arts? Where did the government funds come from? Who paid the taxes, and how did it affect them? ‘Economic impact’ numbers move us from generally accepted, clearly defined economic variables into a fog. The new work with the BEA is a great new thing; next time, let’s skip the multipliers.
Hi Michael. Thanks very much for your comments. We’ve tried to address some of them. First, the SPPA does capture arts participation for 11 distinct metro areas; they are available on our SPPA Arts Data Profile Page (http://arts.gov/artistic-fields/research-analysis/arts-data-profiles/arts-data-profile-5/arts-data-profile-5).
About the arts and cultural satellite account, we hope the following answers will help inform this topic:
Where does the initial increase in demand for arts and culture come from?
The following are examples of changes in final demand: investments in new construction, equipment, and software; government purchases; purchases made by outside consumers (exports); and purchases made by household consumers.
Therefore, new demand for the arts and culture can include the construction of new performing arts centers; increased government spending on the arts; more exports of movies and TV shows; or more demand for the performing arts resulting from the media or a successful advertising campaign.
A reduction in demand for other goods and services? What happened in those sectors as a consequence?
These comments relate to two key assumptions about BEA multipliers. First, if an increase in demand for one firm results in a decrease in demand for another firm, only net changes in final demand should be applied to the multiplier. If a new performing arts center’s performances draw audiences away from an existing center, then only the net change in ticket sales should be used to estimate impact.
Second, BEA multipliers derive from fixed-price models that incorporate no supply constraints. The model assumes firms are not operating at full capacity.
The most recent year captured by the ACPSA is 2012, a year marked by an 8.1 percent unemployment rate. However, should the ACPSA continue for several years, it may come to reflect a stronger U.S. economy—one that is at or near full employment. At full employment, new demand for arts and culture may result in inflation.
An increase in government funding for the arts? Where did the government funds come from? Who paid the taxes, and how did it affect them?
The BEA literature does not explicitly discuss budget constraints for government. However, it does note two studies that have examined this issue. First, a 1993 article published by the Cato Institute, and authored by Edwin Mills, criticizes state and local government interpretation and use of commercially prepared multipliers.
And second, and perhaps more relevant, is a 2009 article published by the National Bureau of Economic Analysis (NBER), authored by Robert Hall. That article reports that government multipliers are greater when nominal interest rates are low (as they have been in recent years), and when employment is “elastic” enough to respond to increased demand, among other factors.
Thanks, again, for your discussion on this.
Bonnie, thanks for your reply. I wasn’t aware of the SPPA data on different metro areas, and I will be sure to check it out – it should provide a rich source for future studies.
Thanks also for your detailed reply on ‘economic impact’. It does rather prove my point though: the methodology behind estimating value-added in the satellite account, and the ‘% of GDP’ figures, have an accepted methodology, and a clear definition. The ‘economic impact’ numbers do not – they rely on hidden, and questionable (to say the least), assumptions. The inclusion in the BEA accounts of the creative sector is an excellent thing, to be applauded. The ‘economic impact multipliers’ were ill-advised.