Teaching arts policy this fall, I needed a two-page briefing to warn my students off using economic impact studies as an arts advocacy tool. Here’s the result:
What is an Economic Impact Study?
Definitions are hard to come by. I can tell you how a number is calculated, so let’s start there.
Pick a sector: it can be film production in a state, or nonprofit arts organizations, or cranberry farming, or anything.
First ask: What were total sales in that sector in a state over the course of a year? That should be easy to figure – every firm keeps track of its revenues. Suppose we arrive at a number, say $67 million.
Second: consider that the $67 million that was received in revenue was allocated somehow – to buying inputs, to paying wages, or retained as profits by the business owners (if it is a commercial firm and not a nonprofit). When inputs were purchased, that constituted revenue for a different company (say, a paint store). And the wages of employees were spent in different shops, giving those shops some revenue.
How much was spent in this second round in the state? It will be less than $67 million, since (1) some of the originally generated income will be put into savings, and (2) some of it will be spent on goods and services produced outside the state. But some of it is spent in state. Suppose it is $40 million.
Now, of that $40 million in “second round” revenue, some of that is spent in state, generating even more revenue. As we go through rounds, the numbers get smaller and smaller, but they add up.
There are input-output models of state economies that estimate how much revenue from any one sector ends up as revenue in a different sector.
So, let’s add up the original $67 million plus all the subsequent rounds of spending. We can call the $67 million direct impact, and the sum of all the other rounds indirect impact. Suppose we get a grand total of, say, $167.5 million. In this case we would say the spending multiplier is 2.5 – i.e. the sum total of all spending resulting from that initial $67 million is 2.5 times the initial spending.
People who do these studies would call $167.5 million the annual economic impact of the sector.
What the number means is a different question…
What is Wrong with the Economics of Economic Impact Studies?
Suppose the above study was of the cranberry farming sector in the state economy. An advocate for that sector might say “Cranberries mean $167.5 million in income in our state. If we were to lose our cranberry farms, that’s how much income we would lose.”
An economist would respond: “Hold on. Suppose this actually happened – a study came out that said cranberries were very, very bad for health and the demand for them collapsed. Then the land devoted to cranberries would go to something else – growing another crop, or residential or commercial development, or something. The people who worked on cranberry farms would work somewhere else – another farm, another business. The money that people used to spend on cranberries would be spent on other things.
“In our economy, industries are always rising and falling, people moving from one sector to another. It is not as if when any sector goes into a downturn that the income is lost forever, that people never get another job. Or that if the sector had not been created that some people would simply never have found work.”
The advocate quoted at the beginning on this section is assuming that the economy never adjusts, that there is never movement between sectors. But we know that is false.
Further, note that if land were to switch from cranberry farming to blueberry farming, then blueberry farming would also have all those “indirect” spending numbers – essentially, any job where people earn money has “indirect” effects. Cranberries aren’t special in that respect.
And so, to consider an arts example, suppose a mayor says “we should spend money building a new performing arts center. Construction costs would be $3 million, and the total economic impact of the construction would be $7.5 million”. An economist would say: “you could do a lot with $3 million: you could repair infrastructure, you could expand after school programs, you could lower taxes by $3 million and leave it to individuals to have more money to spend. Any of those options would also have ‘economic impact’. So ‘economic impact’ doesn’t justify spending on the performing arts center. What would justify a new PAC would be if the public benefits from using it exceeded the costs of building it and running it, i.e. a proper cost-benefit analysis. Building a PAC is a cost, it is not the benefit.”
Economists don’t like “Economic Impact” studies – they know that the conception of them is wrong, and they lead to bad reasoning.
What is wrong with the Politics of Economic Impact Studies?
A few things.
I sometimes hear from arts advocates that use EI studies that “well, it works – politicians listen to us when we have these numbers”. But…
First, I have never seen evidence that this is true. I’ve studied this subject for twenty-five years, and have never seen evidence that economic studies have informed decisions on public spending on the arts.
Second, the numbers don’t give any policy guidance. Suppose I were to tell you that the annual economic impact of the nonprofit arts sector in Bloomington is $73 million. If you were on city council, what would that tell you? That arts support should be increased? Or decreased? That this is a very big number? Or about what one would expect? That we should increase spending on arts program X but decrease it on arts program Y? I have never seen a policy decision where the economic impact number made a difference. (To see this, imagine that I told you “I’m sorry, I made a typing mistake, it’s not $73 million, it is $63 million”. How would that correction affect any arts policy decision?).
Third, remember that any advocacy activity involves an opportunity cost – if you stress this message, you have less time to press that message (Professional political campaign managers know this). So any effort in a media campaign directed at emphasizing economic impact means less effort on all the other reasons we might advocate for public support of the arts, all the other benefits.
Fourth, the whole point of arts advocacy is to make the case that the arts are different from other sectors. You might take the externalities / public goods approach of the economists, you might take the “our identities are tied to our cultural communities” approach of the communitarians, you might take the “we have an obligation to pass a rich cultural palate on to future generations” approach of Dworkin, but you’re saying the arts are special. But “economic impact” makes it no different from cranberry farming, stock car racing, petro-chemical production, or any other sector, since they all have “economic impact”.