Estimating the Impact of Live Simulcast on Theatre Attendance: An Application to London
[READ IN FULL AT SOURCE, the publisher PDF, including Table 1, the discussion of the anomaly, both appendices and the regression diagnostics. NOT A PEER-REVIEWED ARTICLE: this is a research-foundation working paper and must not be described as a journal study] Bakhshi, H., & Whitby, A. (2014). 'Estimating the Impact of Live Simulcast on Theatre Attendance: An Application to London's National Theatre.' Nesta Working Paper No. 14/04, June 2014. JEL Z11, O33. ISSN 2050-9820. BAKHSHI FIRST AUTHOR AND CORRESPONDING, Nesta. Data from The Audience Agency's Audience Finder, funded by Arts Council England; screening locations and dates supplied by the National Theatre. THE BEST MEASURED CASE FOR WHAT MEDIATION GAINS, AND IT ASKS THE ARTICLE'S EXACT QUESTION OF A CULTURAL PRODUCT. Their abstract: 'simulcasting may substitute for live attendance among existing audiences, but may also promote and engage new audiences.' DATA: about 16 million transaction records for 54 performing arts organisations across England representing 44 million tickets, from early 2009 when National Theatre Live launched to late 2013; after cleaning and artform coding, about 12 million records representing 28 million tickets, 32 per cent theatre; 21 of the 54 venues in London; a 30-kilometre catchment radius; aggregated to Medium Layer Super Output Areas giving N = 3,619 areas, M = 54 organisations, T = 5 periods. METHOD: ordinary least squares on log visits with two-way fixed effects for geographic area and for organisation by year, controls for population and Land Registry median house price, treatment coded as 1-kilometre distance bands from a cinema that showed a broadcast in the previous calendar year against a control region of 5 kilometres or more, standard errors clustered by area, residual diagnostics in an appendix. RESULT, TABLE 1 MODEL 1A, theatre visits: about 10 per cent higher within 1 kilometre, 7 per cent at 1 to 2 kilometres, 3 per cent at 2 to 3 kilometres, 0 per cent at 3 to 4 and minus 2 per cent at 4 to 5, with the 0 to 3 kilometre coefficients significant at the one per cent level. A single 3-kilometre dummy gives a 5.3 per cent average effect, 6.4 per cent in London and an insignificant minus 0.1 per cent outside London. House-price elasticity 0.253; distance coefficient minus 1.068, close to inverse-distance. THEIR CONCLUSION, verbatim: 'We find no evidence of cannibalisation on theatre attendance at a broad spread of English venues since the National Theatre Live programme was instituted in 2009. Indeed, as in that paper, we find some evidence of a complementarity between NT Live and in-person theatre attendance. Caution is, however, required in interpreting these results.' THEY STATE THE ANOMALY AGAINST THEMSELVES: their own Hypothesis 2.2 predicted a stronger effect on theatre than on other art forms, and non-theatre visits showed a similar pattern at 7 per cent within 1 kilometre; they offer two readings, that the broadcasts promote arts attendance generally, or that 'our analysis is suffering from some omitted variable which is correlated with the NT Live treatment variable', and they name the candidates, including cinemas self-selecting into the programme in anticipation of a growing high-cultural-participation demographic. They also warn against reading aggregate results as individual behaviour. TWO REACH FIGURES FROM THEIR INTRODUCTION: the National Theatre's first live broadcast, Racine's Phedre, on 25 June 2009, 'allowed it in one night to double the overall audience for the play over its three-month run at the National'; and on 27 February 2014, 120,000 people watched War Horse in United Kingdom cinemas against the 1,024-seat capacity of the theatre it was staged in. They cite Bakhshi and Throsby (2014), Journal of Cultural Economics 38(1), 1-8, as reaching the same conclusion for the National's own box office; THAT PAPER WAS NOT READ. WHAT MAY NOT BE SAID: that National Theatre Live caused the rise. It is an observational panel design, the authors report the unexpected non-theatre result themselves, and they name the omitted-variable candidates.
[READ IN FULL AT SOURCE, the publisher PDF, including Table 1, the discussion of the anomaly, both appendices and the regression diagnostics. NOT A PEER-REVIEWED ARTICLE: this is a research-foundation working paper and must not be described as a journal study] Bakhshi, H., & Whitby, A. (2014). 'Estimating the Impact of Live Simulcast on Theatre Attendance: An Application to London's National Theatre.' Nesta Working Paper No. 14/04, June 2014. JEL Z11, O33. ISSN 2050-9820. BAKHSHI FIRST AUTHOR AND CORRESPONDING, Nesta. Data from The Audience Agency's Audience Finder, funded by Arts Council England; screening locations and dates supplied by the National Theatre. THE BEST MEASURED CASE FOR WHAT MEDIATION GAINS, AND IT ASKS THE ARTICLE'S EXACT QUESTION OF A CULTURAL PRODUCT. Their abstract: 'simulcasting may substitute for live attendance among existing audiences, but may also promote and engage new audiences.' DATA: about 16 million transaction records for 54 performing arts organisations across England representing 44 million tickets, from early 2009 when National Theatre Live launched to late 2013; after cleaning and artform coding, about 12 million records representing 28 million tickets, 32 per cent theatre; 21 of the 54 venues in London; a 30-kilometre catchment radius; aggregated to Medium Layer Super Output Areas giving N = 3,619 areas, M = 54 organisations, T = 5 periods. METHOD: ordinary least squares on log visits with two-way fixed effects for geographic area and for organisation by year, controls for population and Land Registry median house price, treatment coded as 1-kilometre distance bands from a cinema that showed a broadcast in the previous calendar year against a control region of 5 kilometres or more, standard errors clustered by area, residual diagnostics in an appendix. RESULT, TABLE 1 MODEL 1A, theatre visits: about 10 per cent higher within 1 kilometre, 7 per cent at 1 to 2 kilometres, 3 per cent at 2 to 3 kilometres, 0 per cent at 3 to 4 and minus 2 per cent at 4 to 5, with the 0 to 3 kilometre coefficients significant at the one per cent level. A single 3-kilometre dummy gives a 5.3 per cent average effect, 6.4 per cent in London and an insignificant minus 0.1 per cent outside London. House-price elasticity 0.253; distance coefficient minus 1.068, close to inverse-distance. THEIR CONCLUSION, verbatim: 'We find no evidence of cannibalisation on theatre attendance at a broad spread of English venues since the National Theatre Live programme was instituted in 2009. Indeed, as in that paper, we find some evidence of a complementarity between NT Live and in-person theatre attendance. Caution is, however, required in interpreting these results.' THEY STATE THE ANOMALY AGAINST THEMSELVES: their own Hypothesis 2.2 predicted a stronger effect on theatre than on other art forms, and non-theatre visits showed a similar pattern at 7 per cent within 1 kilometre; they offer two readings, that the broadcasts promote arts attendance generally, or that 'our analysis is suffering from some omitted variable which is correlated with the NT Live treatment variable', and they name the candidates, including cinemas self-selecting into the programme in anticipation of a growing high-cultural-participation demographic. They also warn against reading aggregate results as individual behaviour. TWO REACH FIGURES FROM THEIR INTRODUCTION: the National Theatre's first live broadcast, Racine's Phedre, on 25 June 2009, 'allowed it in one night to double the overall audience for the play over its three-month run at the National'; and on 27 February 2014, 120,000 people watched War Horse in United Kingdom cinemas against the 1,024-seat capacity of the theatre it was staged in. They cite Bakhshi and Throsby (2014), Journal of Cultural Economics 38(1), 1-8, as reaching the same conclusion for the National's own box office; THAT PAPER WAS NOT READ. WHAT MAY NOT BE SAID: that National Theatre Live caused the rise. It is an observational panel design, the authors report the unexpected non-theatre result themselves, and they name the omitted-variable candidates.
Written from the rEAD IN FULL AT SOURCE, the publisher PDF, including Table 1, the discussion of the anomaly, both appendices and the regression diagnostics. NOT A PEER-REVIEWED ARTICLE: this is a research-foundation working paper and must not be described as a journal study.
What our sources record
Recorded when this source was checked, figures included, with page references and quotation marks removed.
s National Theatre.' Nesta Working Paper No. 14/04