Modelling the economic impacts of cultural policies

In Policies for boosting arts demand I build on my argument that Australian cultural policy needs more demand-side policies. The case is based on the observation of recent data on the Australian cultural sector and some simple economic supply and demand modelling. This post supplements these articles by explaining the modelling process and reasoning that underpin them. This is a rough draft – I welcome comments or suggestions.

Recent data suggests that the Australian arts sector is grossly ‘oversupplied’. In the first decade of the century, Australian participation in creative arts work increased dramatically – in some arts activities it more than doubled or tripled! At the same time, artists’ relative incomes have declined. The Census, for example, shows that the full time incomes of artists dropped by $4,000 relative to other professionals. The figure below puts the two trends together to illustrate how dramatic the inverse relationship has been between involvement in creative arts practice and artists’ incomes.[1]

These are classic signs of a sector straining under the weight of labour supply: increases in labour supply tend to reduce wage rates, which is likely to be reflected in declining incomes. Australia’s cultural policies predominantly work on stimulating supply, and so are likely to have made matters worse.

In my Culture360 article I compare demand-side and supply-side policies across a range of dimensions. One of those is the economic dimension, where I argue that demand-side policies increase arts production and put ‘upward pressure’ on arts incomes. More policies  to stimulate demand could, therefore, bring some much-needed balance to the sector.

These conclusions come right out of the most basic form of economic supply and demand modelling. If art is assumed to have demand and supply curves similar to any other goods or services, then the arts sector, or arts industry, can be modelled like any other market using standard supply and demand curves and the technique of ‘static dynamics’. This post runs through the model to explain how the policy impacts in Policies for boosting arts demand were derived.

Modelling the arts sector as a market
The most basic way economics models a sector is by using supply and demand curves. These curves provide a way of thinking about the sector as a ‘market’. Though limited and very basic, this modelling technique is a useful starting point when thinking about the impact of government policies.

The model is based on a Cartesian space, such as that in the figure to the right, with price on the y axis and quantity demanded and quantity supplied on the x axis. Economics recognises that quantity demanded and supplied will depend on a lot of things, not just price, but price holds a special fascination for economists. Modelling price can bring a lot of insights into the analysis of a sector and the likely impacts of policy intervention.

The key aspects of the model are:

1. The demand curve
The demand curve shows the quantity of art demanded at each price level. As with other goods, the art demand curve slopes downward: few people are willing to buy art at higher prices, more people are willing to purchase art at lower prices. The peculiar nature of some art means that there may be caveats and exceptions to this rule, but on the whole I believe it’s a worthy way to model the bulk of exchangeable artistic and cultural experiences. If in doubt, think about the market for theatre performances and literature: When theatres use different prices for different audience segments, they are demonstrating that the performing arts have a standard downward-sloping demand curve; and the hullaballoo that often surrounds book taxes and the parallel importing of books demonstrates the same for literature.

2. The supply curve
Supply works in the opposite to demand. The higher the price of art, the more producers are willing to produce and sell art. They are not so willing at low prices. The supply curve slopes upward.

3. Equilibrium
The two curves meet at the price at which quantity demanded equals quantity supplied, at the holy grail of economics – market equilibrium.

4. Total revenue
At equilibrium, there is q₁ amount of art exchanged at price p₁. The total amount spent on art, or the arts sector’s revenue, can be calculated as q₁ x p₁. This is represented by the area in blue.

The model is obviously a major simplification of the ‘real world’. Perhaps it requires too much of a leap of faith for some. If you don’t buy this modelling concept, read no further. Bear in mind, though, that there could still be benefit in suspending disbelief for just a moment. As with all models, these are tools to think about the real world. They are not supposed to be perfect replications of the real world: models that did that would be impossibly complex and would miss the point of modelling.

Below I run through the simple analysis of what happens to the ‘arts market’ under two scenarios: supply-side policy intervention; and demand-side policy intervention.

Scenario 1: demand-side policy intervention
Demand-side policies act to encourage demand for art. Culture vouchers, for example, allow consumers to attend a performance or buy an art work at a price they were previously not able to buy at. The effect of demand-side policies is, therefore, to increase the amount of art demanded at each and every price. This equates to a shifting of the demand curve outward, as in the figure to the right.

Note that, if supply remains stable:

• The art price increases
  (from p₁ to p₂)
• The quantity of art demanded and supplied increases
  (from q₁ to q₂)
• Total revenue of the arts sector increases (an expansion of the revenue rectangle).

A demand-side policy leads to an expansion of the sector, with both quantity demanded and quantity supplied increasing, and total sector revenue going up. If there is no change in the number of artists and arts workers, there is more revenue to share across artists and arts workers, so demand-side policies are likely to place upward pressure on arts incomes.

Scenario 2: Supply-side policy intervention
Supply-side policies act to encourage art supply. Grants, for example, allow producers to produce art at prices they were previously not able to produce at. In other words, supply-side policies increase the amount of art able to be supplied at each and every price. This equates to a shifting of the supply curve outward, as in the figure.

Note that, if demand remains stable:

• The price of art drops
  (from p₁ to p₂)
• The quantity of art supplied and demanded increase
  (from q₁ to q₂)
• Total revenue of the arts sector changes (from the blue rectangle to the red rectangle).

The supply-side policy leads to an expansion of the sector – with quantity supplied and demanded both increasing – but an ambiguous impact on total sector revenue.

Whether total revenue goes up or down depends on the relative sizes of the blue and red rectangles. It is not always obvious which is bigger, and this depends on some seemingly random matters such as how the demand and supply curves are drawn, and how far the shift in the supply curve is drawn. In Economics, these are determined by things called elasticities. Elasticities determine whether total revenue will go up or down when the supply curve moves, and how far the curve will move in response to policy intervention. There has been quite a bit of research done in cultural economics about the elasticities associated with arts goods and services. (See Bruce Seaman’s exhaustive review of elasticities in the performing arts, Attendance and Public Participation in the Performing Arts: A Review of the Empirical Literature). The results of this highly technical research are inconclusive, although the majority of studies have found performing arts to be price inelastic, which means that total revenue will drop with a decrease in price. Inelastic arts demand also seems to fit well with the expectations of Cultural Economists.

A drop in total revenue following a drop in price, which research and preconception suggest is likely, means that there will be less money in the sector with which to remunerate artists and arts workers. If there is no change in the number of people in the arts workforce, then, mathematically, average incomes will drop: A decline in arts revenue will put downward pressure on arts incomes. (It is important to note that the number of arts workers and their average incomes are not part of the supply and demand curve models, and are not shown in the graphs)

The data presented at the top of this post suggest that expansion in the Australian arts sector, as highlighted by increasing number of people involved in arts work, has indeed led to declining relative incomes.

The data are consistent with a supply-side expansion causing downward pressure on artist incomes. Although policies are unlikely to be wholly responsible for declining incomes, the trend does follow after a substantial expansion in Australian government supply-side support through the ‘Howard years’, so supply-side policies will certainly not have helped.

It is probably the greatest – and possibly least understood – paradoxes of contemporary arts policy: that the very policies used to support the arts also put financial pressure on arts incomes.

Time and the perils of static dynamics
The comparison of supply and demand curves between two points in time is known as ‘comparative statics’ analysis. This rather clumsy technique is necessary in standard economic supply and demand curve analysis because time is not included in the models. But not including time in the models can lead to some difficulties in interpretation. Long-term impacts of policies are likely to differ substantially from short-term impacts. It is not easy to decide which is most important or how long the changes modelled will last.

The analysis above has already struggled with temporality: in predicting income effects, for example, it was assumed that the number of artists didn’t increase or decrease. But this is a rather strict condition that is unlikely to equate to the real world, where people can enter or leave the arts labour market. If increased revenue leads to increased arts incomes, more people will likely be attracted into the arts labour market, and this could drive average incomes down. In the long-run, higher average arts incomes could be competed away by new entrants.

There are good reasons to believe that this effect is very real in the arts labour market. Arts occupations tend to have low barriers to entry. Anyone can, in effect, call themselves an artist. Artistic success or failure is measured more by talent or the quality of art works than by the holding of a qualification or professional memberships (unlike other professions, such as law and medicine, in which formal documentation is a prerequisite for practice – no coincidence that these professions also tend to have higher incomes).

Two other aspects of arts work would facilitate the ease of entry into artist professions: first, people with an artistic ‘bent’ have a profound preference for working in the arts – a tendency formally recognised in David Throsby’s ‘work preference’ model of the artist labour market; and second, the multiple job-holding and contingent employment patterns of arts workers mean that there is a ready ‘pool’ of potential entrants into the arts labour market, many of them with high levels of experience  and ability.

With such ease of entry and exit into and out of the arts labour market, supply of arts labour will move relatively quickly – ‘elasticly’  – with changes in arts income levels.

Conclusion
Modelling policy impacts with demand and supply curves suggests that supply-side policies will expand the arts sector but lower the arts price. If arts demand remains unchanged and the demand for art is price inelastic, supply-side policies can put downward pressure on arts incomes. Data shows this occurring recently in Australia: when creative arts practice expanded dramatically between 2001 and 2007, artists’ relative incomes declined. Modelling suggests that Australia’s predominantly supply-side cultural policies will have done little to alleviate, and may have even accentuated, the decline in artists’ relative incomes. It also suggests that demand-side policies could alleviate some of the downward pressure on artists’ incomes by expanding arts sector revenue.

Postscript
11 February 2011

Ross Gittins, Economics Editor at the Sydney Morning Herald, has recently published some interesting articles on economic modelling. Gittins goes on the offensive against poor modelling, and, in the process, explains in layperson’s terms some of the more obscure methods and assumptions underpinning economic models. In The very model of a future based on guesswork (Sydney Morning Herald, 4 February 2012) he draws particular attention to the importance of a modeller’s ‘time horizon’ and concludes that the models ‘should be used only in private by consenting economists.’

See also Damned lies and economic modelling (Sydney Morning Herald, 1 February 2012), in which Gittins looks at some of the ways that advocates and lobbyists misuse economic modelling, many of which I raise in the context of art and cultural policy in Using economic’ impact studies in arts and cultural advocacy: a cautionary note, Media International Australia Incorporating Culture and Policy, no.98 (February), 2001.

Gittins’s articles seem to have been prompted by the release of an interesting – though rather shoddily presented – research report released by the Australia Institute, The use and abuse of economic modelling.

---

[1] To view the data, see Australia’s Creative revolution and section 3.2.3, page 70, of Compendium of data on artists’ work and employment in Australia.