Symphony Concert Demand: Does Programming Matter?

Abstract

We examine factors that determine attendance for major U.S. symphony orchestras over seven seasons. We find that performing more of the standard repertoire increases attendance but the choice of a composition that has a particular solo instrument does not affect attendance. Other management options that increase attendance include spending on advertising and broadcasting concerts. Own-price elasticity of demand and income elasticity are inelastic.

Keywords:

• elasticity
• programming
• standard repertoire
• symphony concert demand

Notes

1. We use the League of American Orchestras (LAO) definition of major U.S. SOs, which is determined by budget size and artistic quality. This includes all group 1 and 2 SOs which have budgets greater than $12 and $5 million, respectively.

2. As mentioned above, SOs are dependent on donations to cover operating costs.

3. An anonymous reviewer noted that SOs with three of the five highest RCI values faced serious financial difficulties during the period from 2001 to 2007: Columbus, Honolulu, and San Antonio; a fourth—San Diego—was recovering during this period. Although is possible that struggling SOs would be more likely to revert to the standard repertoire, there are other factors that could explain the RCI value. Pompe, et al. (2011) find that major SOs are more likely than nonmajor SOs to perform nonstandard repertoire. In addition, increased levels of revenue from ticket sales, endowments, and local government encourage innovative programming.

4. Often, actual attendance will be less than paid attendance because some season subscribers may not attend a concert that has less popular compositions.

5. The Boston Pops Orchestra, which is included in our data set, would be the only exception to this expected relationship. We tested the models without the Boston Pops observations, but found no significant change to the results.

6. Generally, the average price of a ticket is used in models that estimate the demand for the performing arts, although those sitting in better-quality seats which are more expensive would most likely have more inelastic demand. Seaman (2005) explains the advantages and disadvantages of using average price as well as discussing the varied results of several studies that use less aggregated prices.

7. Most non-major SOs are community orchestras that are semi-professional. In our data set only one location (Minneapolis-St.Paul) has more than one SO that is classified as a major SO.

8. We expect no simultaneity bias because of the recursive nature of SO decisions. We estimated simultaneous equations, but the results were very similar to those reported in Table 5.

9. The Sidak adjusted values are significant at the 5 percent level for YR1, YR2, and YR3. Including YEAR variables were insignificant and did not have much impact on the regressions, except that the coefficients on RCI were much higher when YEARS were included. In model (4), we omit the year dummy variables and the impact of a 10 percent increase in RCI falls significantly to an 11.9 percent increase in attendance.

10. The decrease in the R² is negligible in the remaining models when SUBSCR is omitted.

11. Indeed, anecdotal evidence indicates that concerts that include a violin or piano soloist are often better attended. TWOPLUS, which is positively correlated with ATTEND, INC, QUAL, and AD, and negatively correlated with POPS, appears to be an indicator of a better-quality SO.

12. It is possible that the income elasticity is more elastic but is dampened by the rising cost of leisure.