APPLIED ECONOMICS

2024, VOL. 56, NO. 58, 8437–8446

https://doi.org/10.1080/00036846.2023.2290595

The complementary relationship between live performances and post-concert

streaming for top-performing artists
Eric Hogue
Department of Marketing Communications, Emerson College, Boston, MA, USA

ABSTRACT KEYWORDS
This article examines the complementary effects of live concerts on incremental post-concert Music; music industry;
music streams in 29 US cities. The work identifies that performers who have greater concert ticket music streaming;
demand, deeper hit-song catalogues and/or are in the middle of their careers experience stronger industry disruption
trends in their post-concert streams. While I found a positive complementary relationship between JEL CLASSIFICATION
concert and streaming that lasts up to 10 weeks after the event, this work also highlights the D12; D21; L82
negative effects when performers cancel shows. Using hand-collected data of concert ticket sales,
music streaming, and song rankings from the top 60 global performing artists, I utilized a panel
model with artist and market fixed-effects to identify post-concert decaying effects. This work will
help top-performing artists gain insight into the little-understood influence of live performance on
post-concert streaming of their recorded music.

I. Introduction I created a novel, hand-collected dataset from

March 2018 through June 2020 for the top-60 glo­
Musical performing artists are highly dependent bal touring artists, leveraging a series of industry
on income from live performance, which makes sources. The empirical approach utilizes a time
up 80% of their income (Krueger 2019). While series panel model with weekly and monthly
hit songs draw an audience, many artists have dummy variables to identify the magnitude and
been paying less attention to direct income from duration of concert streaming effects. The depen­
recorded music in recent years. For example, dent variable is the log of music streams by artist
bands such as Phish earned $34.7 million tour­ and market. Figure 1 visualizes the time series
ing in 2019 while earning $1.4 to $2.0 million sequence.
from streaming (Hogue 2023). Pre-concert effects were also included in the
In light of recent challenges in touring during analysis to control for promotional effects. While
COVID, performers are exploring complementary the effects of pre-concert recorded music on live
strategies via synchronizations used in program­ concerts have been thoroughly addressed as out­
ming and music streaming to monetize their lined above, they also highlight the need for their
music on digital music platforms. Several research­ inclusion as a control to accurately identify the
ers concluded that the cultivation of a catalogue of post-effects.
hit songs not only provides a stable source of Both pre- and post-concert effects highlight
income but also carries significant promotional the multifaceted behaviour that performers
effects on future concert revenue (Christensen amplify through their recorded music. Pre-
2022; Mortimer, Nosko, and Sorensen 2012; concert effects from a performer’s promotion
Papies and van Heerde 2017; Hogue 2023). This contribute to ticket sales and fans listening to
article explores post-concert effects at a local level. their music. Performers also benefit from post-
Its intent is to identify the threshold and duration concert effects that decay over time, as fans re-
of these effects. It will also measure its monetary engage with the performer’s music after attending
impact. their concert.

CONTACT Eric Hogue [email protected] Department of Marketing Communications, Emerson College, Boston, 120 Boylston Street, MA 02116,
USA
© 2023 Informa UK Limited, trading as Taylor & Francis Group
8438 E. HOGUE

concerts and festivals (Bracalente et al. 2011; Gazel

and Schwer 1997) on local economies. Unlike preced­
ing work, streaming carries a lower search cost (Datta,
Knox, and Bronnenberg 2018), so consumption of an
Figure 1. Time series categories. artist’s music is not limited by their willingness to pay
This article explores several research questions given subscriptions to DSPs are largely exogenous.
regarding the influence of concert dates on music Therefore, the marginal cost of listening to an artist’s
streaming. To what extent do concerts influence music approaches zero. So, a concert attendee’s post-
trends in intra-market post-concert music stream­ concert consumption is primarily based on their lis­
ing? Are post-concert trends heterogeneous across tening preferences and perhaps a desire to reminisce
artists with differing ticket demand, catalogue about their concert experience.
depth, and artists’ years of performing? And, what A fitting analogue would measure the effects
is the magnitude and duration of decaying post- of point-in-time events like the Super Bowl, the
concert trends for music streaming? Grammys, or the Academy Awards. Hartmann,
Creating a panel model with artist and market Klapper, and Super Bowl Ads (2016) imple­
fixed effects identifies post-concert effects of ele­ mented a panel model with market and time
vated streams that extend 10 weeks after the con­ effects to examine 20 weeks of post Super Bowl
cert date. Artists with higher ticket sales, more consumption effects. Their work identified that
weeks on the Billboard 100, and are in the middle Super Bowl advertising does not influence the
of their careers (17–29 years) see stronger stream­ consumption of beer and soft drinks during the
ing effects before and after their performance. By Super Bowl but instead creates
contrast, legacy artists who have played for 30+ a complementary relationship with sporting
years see less impact on their streaming activity in occasions by empirically demonstrating
markets where they perform. increased consumption during big sporting
This article makes a unique contribution to the events later in the spring, such as the NCAA
academic literature because it dimensionalizes the Tournament, MLB Opening Day, and NBA
contribution of live-concert performance on music Playoffs/Finals.
streaming for top-performing artists. While other Following Hartmann’s and Klappers’ approach,
researchers have examined the contribution of con­ a series of weekly dummies were created up to 13
certs on physical distribution and ownership, none weeks after the concert. Additionally, to control for
have examined the influence on subscription-based pre-concert promotion, 13 weekly dummies were
digital music streaming. Nor have the heteroge­ included to control for trends in music streaming
neous effects of demand for concert tickets, catalo­ before the concert.
gue depth, and years performing been examined. Table 1. Data source characteristics.
Source Characteristics
Pollstar Concert data includes Date, number of shows,
revenue, ticket sales, minimum/maximum/
II. Literature review average ticket price, venue, % of capacity
sold, city/state/country
Past research has identified the promotional effects of Music Connect Streaming data: All weekly total, audio, video,
and programmed digital streams by
audio and video music streams on concert revenue Spotify, Apple Music, Amazon, and all other
(Mortimer, Nosko, and Sorensen 2012; Papies and major digital streaming platforms for top-
29 concert markets in 2018 and 2019
van Heerde 2017; Christensen 2022; Hogue 2022). Billboard Rankings Peak rankings and weeks in Hot 100 Billboard
Only Papies and van Heerde (2017) identified concert (provided by Data. ranking of songs by week from 1959 to
World) 2019. Sample filtered for all songs
post-effects. However, their work primarily focused throughout career among the 2019 Pollstar
on concert impacts of physical media ownership (e.g. top-100 performing artists
US Census Incidence of gender age, race, ethnicity and
CDs, tapes, and vinyl) and downloads. This analysis population
Music Brainz.Com Album, E.P., Live Concert releases and
explores the impact of live concert performance on profile of artist gender, primary genre,
post-concert subscription-based streaming. Other and years playing professionally for each
top-60 performing artist
researchers have highlighted a spill-over effect of
 APPLIED ECONOMICS 8439

Table 2. Data frequencies at the artist week and market level. Table 4. Top-60 artists based on concert revenue.
N Mean min max Ed Sheerhan KISS Hootie & the Bryan Adams
Total streams per week 208,730 213,406 - 23,175,478 Blow Fish
Population by market 208,730 4,559,786 1,232,696 19,216,182 Pink Bob Seger & the Twenty-one JoJo Siwa
Cumulative weeks 208,805 119 0 998 Silver Bullet Band Pilots
in top 100 Metallica Garth Brooks Thomas Rhett Phish
Tickets sold per concert 1,549 17,043 280 205500 Elton John Fleetwood Mac Sandy & Junior
Panic! At the
Year: 2018 75,312 0.36 0 1 Disco
Year: 2019 89,007 0.43 0 1 BTS The Rolling Stones New Kids on the Florence and
Year: 2020 44,486 0.21 0 1 Block the Machine
Years performing 60 26 4 63 Bon Jovi Jonas Brothers Phil Collins Take That
Muse Florida Georgia Line Dave Matthews Chris Stapleton
Band
Shawn Andre Rieu Luke Combs Luis Miguel
Mendes
III. Data and empirical analysis Ariana Grande Iron Maiden Luke Bryan Breaking
Benjamin
Data Backstreet Eric Church John Mayer Bad Bunny
Boys
The data set includes a series of hand-collected Trans-Siberian Paul McCartney Billy Joel Eagles
Orchestra
data of the top-60 artists according to Pollstar’s Michael Buble Zach Brown Carrie Jennifer Lopez
2019 rankings based on ticket sales and music Underwood
Hugh Jackman Travis Scott Justin Pentatonix
streaming within the top 29 US markets. The Timberlake
data come from a number of sources (Table 1) Post Malone Cher Mark Knopfler Train
Mumford and Spice Girls Khalid Goo Goo Dolls
and covers the time period from 1 March 2018 Sons
through 30 June 2020. It includes weekly market- Source: Pollstar.
level music streaming and concert dates for each
artist from venues within a 90-minute drive of performed songs that spent an average of 119
the city centre for each market. Streaming data weeks in the Billboard Hot 100. The data set
include 208,730 observations. includes 1,843 concerts between 1 March 2018,
The data set also includes songs from the through 31 December 2019, of which 1,549
Billboard Top 100 rankings dating back to 1959, included ticket sales. Lastly, the 2020 music
market level demographic data, and artist tenure. streams’ primary role is to capture post-effects.
To facilitate time-series analysis, the sources have While artists performed through early March, the
been aggregated and aligned by week (e.g. weeks industry cancelled most concerts starting early
before, week of and weeks after the concert). March due to the COVID-19 pandemic.
Table 2 shows total streams per week per artist Table 4 highlights the artists included. The
per market average 213K with a peak of tens of top-60 artists span a number of genres, with the
millions in LA during early February 2019 when supermajority playing primarily Pop and Rock-
Ariana Grande released her Thank U, Next album. n-Roll.
Additionally, the markets included are large cities Music streams among the top 60 artists
with an average population of 4.6 million, ranging increased between 2018 and 2019 before level­
from New York at 19.2 million to Salt Lake City at ling off at lower levels during 2020 (Figure 2).
1.2 million (Table 3). Artists in the group have Seasonality of concerts also plays a factor with

Table 3. 29 markets included.

Market Population Market Population Market Population
New York 19,200,000 San Diego 3,338,330 Milwaukee 1,575,179
Los Angeles 13,200,000 Tampa 3,194,831 Raleigh 1,390,785
Chicago 9,457,867 Denver 2,967,239 New Orleans 1,270,530
DFW 7,573,136 St. Louis 2,801,423 Louisville 1,266,389
Houston 7,066,140 Charlotte 2,636,883 Salt Lake City 1,232,696
Washington 6,280,697 Pittsburg 2,317,600
Miami 6,166,488 Cincinnati 2,219,750
Philadephia 6,102,434 Kansas City 2,155,068
Atlanta 6,018,744 Columbus 2,122,271
Phoenix 4,948,203 Indianapolis 2,076,531
Boston 4,873,019 Cleveland 2,048,449
San Francsco 4,731,803 Nashville 1,933,860
8440 E. HOGUE

Figure 2. Streams and number of live concerts by month for top-60 performing artists.

more concerts performed during the summer by aligning the time-series data to the weeks in
months. Some of the peaks in streaming high­ the 13 weeks leading up to and the 13-weeks
light blockbuster releases by specific artists, such after the week of the concert. Concerts are dis­
as Arian Grande’s (Thank U Next album and aggregated by performance whereby multi-night
singles: ‘7 Rings’ and ‘Monopoly’) in Q1 2019, dates are each treated as unique concerts.
Khalid’s release of his Free Spirit album in Figure 3 shows the trend of music streams pre-
April 2019, and Post Malone’s single release of and-post concert. Consistent with the growth of
‘Writing on the Wall’ in September 2019. music streaming overall, streaming activity in
Additionally, pre-and-post concert effects play 2019 experienced double-digit growth versus
a role that will be controlled for in this work 2018. Weekly growth rates ranging from 0% to

Figure 3. Total stream (millions) of top artists 13 weeks before to 13 weeks after concerts and year-over-year percent growth within
market (below).
 APPLIED ECONOMICS 8441

34% indicate that while concert effects may be

similar, the trends are higher in 2019 compared
with 2018 data.

Panel model with artist and market fixed-effects

specification

While parsimonious, the approach used by

Specification 2 is created as a baseline for examin­
(Hartmann, Klapper, and Super Bowl Ads
ing the heterogeneity of artist characteristics. It
2016) effectively measures the weekly decay
limits the pre-and-post concert effects up to one
over time and is useful for measuring overall
month before (Concert during upcoming monthitm),
effects and examining heterogeneity in key
the week of the concert (week of concertitm) and up
cohorts via time series interactions (e.g. ticket
to one month after the concert (Concert within past
sales, depth of catalogue, and artists’ years of
monthitm).
experience interacting with time series dum­
mies). The cohort analysis consolidated the
time effects to one month before and one
month after the concert to capture heterogeneity
during a time frame most likely to experience
elevated streaming effects.
Using the pre-concert time dummies, the
approach controls for pre-concert promotion
sourced from ticket on-sale dates, radio promo­
tion, and local artist public relations. Notably, Specification 3 deconstructs the lead-lag period to
controlling for ticket on-sale dates is notoriously 13 weeks before the concert (weeks before the concert
hard to track and a fundamental limitation of dummiesitm), the week of the concert (week of
multiple works of research in this sector (Courty concertitm), and 13 weeks after the concert (weeks
and Pagliero 2011; Krueger 2005; Mortimer, after the concert dummiesitm). The purpose of doing
Nosko, and Sorensen 2012; Papies and van so is to examine the weekly lift preceding the concert
Heerde 2017). followed by the weekly trend decay after the concert.
To account for differences among performing
artists and cities, a panel model with artist and
market fixed-effects was chosen to examine the
influence of concert dates on pre- and post-
event trends in music streaming. The music
streaming outcome is denoted in the model spe­
cification as ln(Music streamsitm). Specifications
1 is designed to highlight the promotional
effects of upcoming concerts on streaming up Model results in Table 5 indicate strong significance
to three months before the concert (Concert dur­ in both pre-and-post-effects. Specification 1 coeffi­
ing upcoming 3 monthsitm), week of the concert cients outline a shape one would expect whereby the
(week of concertitm), and the post-effects up to coefficient is .147 up to three months before the con­
three months after the concert (Concert within certs, .407 the week of the concert, and decays to .077
past 3 monthsitm). All specifications include con­ during the three months after the concert.
trol variables for month and year ðδt Þ as well as Specification 2 plays the role of a baseline model to
artist and market fixed effects. be built upon with artist characteristics in the next
8442 E. HOGUE

Table 5. Panel model with artist and market fixed effects.

1 2 3

Ln(Streams)itm Coef. Sig Coef. Sig Coef. Sig

lead 0 to 3 months_concert_dateitm 0.147 (0.003)***
Week of _concert_dateitm 0.407 (0.010)***
lag 0 to 3 months_concert_dateitm 0.077 (0.003)***
lead 1 month_concert_dateitm 0.154 (0.011)***
Week of _concert_dateitm 0.379 (0.010)***
lag 1 month_concert_dateitm 0.108 (0.010)***
lead13_concert_dateitm 0.102 (0.011)***
lead12_concert_dateitm 0.087 (0.011)***
lead11_concert_dateitm 0.088 (0.011)***
lead10_concert_dateitm 0.098 (0.011)***
lead9_concert_dateitm 0.098 (0.011)***
lead8_concert_dateitm 0.097 (0.011)***
lead7_concert_dateitm 0.110 (0.011)***
lead6_concert_dateitm 0.123 (0.011)***
lead5_concert_dateitm 0.141 (0.011)***
lead4_concert_dateitm 0.164 (0.011)***
lead3_concert_dateitm 0.193 (0.011)***
lead2_concert_dateitm 0.234 (0.010)***
lead1_concert_dateitm 0.363 (0.010)***
Week of _concert_dateitm 0.381 (0.010)***
lag1_concert_dateitm 0.256 (0.010)***
lag2_concert_dateitm 0.188 (0.010)***
lag3_concert_dateitm 0.151 (0.010)***
lag4_concert_dateitm 0.123 (0.010)***
lag5_concert_dateitm 0.085 (0.010)***
lag6_concert_dateitm 0.056 (0.010)***
lag7_concert_dateitm 0.043 (0.010)***
lag8_concert_dateitm 0.030 (0.010)***
lag9_concert_dateitm 0.025 (0.010)**
lag10_concert_dateitm 0.018 (0.010)*
lag11_concert_dateitm 0.011
lag12_concert_dateitm 0.014
lag13_concert_dateitm 0.007
Constant 10.8 (0.005)*** 10.8 (0.005)*** 10.8 (0.005)***
R-squared 0.158 0.148 0.166
F-test 1292.204 1201.1 760.22
Prob > F 0 0 0
Number of obs 208711 208711 208711
***p < .01, **p < .05, *p < .1.

section. Isolated to a narrower timeframe, the model performance provides a sufficient timeframe for
is better suited to incorporate interactive effects of addressing the research questions.
artist characteristics. The results display a similar lift I also completed a counterfactual analysis to con­
and decay pattern of Specification 1. firm that when performers cancel concerts within
Specification 3 outlines the same trends with weeks of the event, the impact on music streaming is
more granularity where the effects lift steadily dur­ diminished. I was able to collect 96 concerts from
ing the 13-week pre-period, peaking the week of January through June 2020, of which 73 were can­
the concert, then declining to an insignificant level celled. Panel model results in Figure 4 highlight that
10 weeks after the concert. music streaming declined versus previous streaming
A robustness test on Specification 3 that added levels within 2 weeks of the concert and persisted until
times predictors for 14 and 26 weeks prior to the 4 weeks after the cancelled event. Further, I found
performance found that the streaming effects are a similar trend for Post Malone, who hosted 12 con­
significant up to 26 weeks prior to the performance. certs in January and February before having three
However, the net impacts were small ($65 per week concerts cancelled in March. The trend in his stream­
per artist). This confirmed that ±13 weeks of the ing mirrors the broader trend from the panel model.
 APPLIED ECONOMICS 8443

Figure 4. Cancelled concerts in 2020. Note: On Post Malone’s Concert data, the y-axis reflects the average index of streams for
Malone’s 2020 performances and cancellations where week (-4) = 100. The x-axis reflects weeks before, week of, and weeks after the
performance date.

Table 6. Panel model by ticket sale, weeks in top 100, and years playing cohorts with artist and market fixed effects.
4 5 6

Ln(Streams) itm Coef. Sig Coef. Sig Coef. Sig

Bottom tier sales
lead 1 month_concert_date itm 0.128 (0.010)***
Week of _concert_date itm 0.289 (0.021)***
lag 1 month_concert_date itm 0.066 (0.011)***
Middle tier sales
lead 1 month_concert_date itm 0.259 (0.011)***
Week of _concert_date itm 0.424 (0.023)***
lag 1 month_concert_date itm 0.17 (0.012)***
Top tier sales
lead 1 month_concert_date itm 0.35 (0.008)***
Week of _concert_date itm 0.526 (0.017)***
lag 1 month_concert_date itm 0.2 (0.009)***
1 to 70 weeks in top 100
lead 1 month_concert_date itm 0.088 (0.011)***
Week of _concert_date itm 0.216 (0.010)***
lag 1 month_concert_date itm −0.209 (0.005)***
71 to 270 weeks in top 100
lead 1 month_concert_date itm 0.111 (0.008)***
Week of _concert_date itm 0.155 (0.005)***
lag 1 month_concert_date itm 0.097 (0.004)***
271 weeks+ in top 100
lead 1 month_concert_dateitm 0.142 (0.009)***
Week of _concert_dateitm 0.015 (0.005)***
lag 1 month_concert_dateitm 0.214 (0.006)***
4 to 16 years performing
lead 1 month_concert_dateitm 0.143 (0.016)***
Week of _concert_dateitm 0.303 (0.016)***
lag 1 month_concert_dateitm 0.103 (0.016)***
17 to 29 years performing
lead 1 month_concert_dateitm 0.311 (0.019)***
Week of _concert_dateitm 0.602 (0.019)***
lag 1 month_concert_dateitm 0.225 (0.019)***
30+ years performing
lead 1 month_concert_dateitm −0.004
Week of _concert_dateitm 0.276 (0.020)***
lag 1 month_concert_dateitm −0.018
R-squared 0.161 0.171 0.15
F-test 1102.74 1188.65 1013.07
Prob > F 0 0 0
Number of obs 2,08,711 2,08,711 2,08,711
***p < .01, **p < .05, *p < .1.
8444 E. HOGUE

Panel model to identify heterogeneity of a more enduring lift post-concert. Specification 5

post-concert streaming effects by key cohorts identifies the artists with deeper hit-song catalo­
gues see strong pre-and-post concert effects. Artists
While identifying pre-and-post concert effects is
with 270+ and 71 to 270 weeks of hits in the top 100
meaningful, dimensionalizing the impact on key
supersede the lift of those with fewer weeks in the
characteristics of performing artists, their music,
top 100. In fact, those with 1 to 70 weeks in the top
and ticket sales holds promise for helping define
100 see a negative outcome in their music stream­
an artist’s strategy and expectations when they
ing after the concert. Specification 6 highlights an
tour.
interesting nuance. While artists who have per­
A panel model with year-month controls,
formed 17–29 years see a stronger pre-and-post
artist, and market fixed effects was created to
pattern than those with less playing experience,
test this hypothesis. Cohort groups were inter­
artists with 30+ years of playing experience have
acted with the one-month pre-and-post-
no significant pre-and-post concert effects.
effects.
Figures 5, 6, and 7 visually highlight the greater
Specification 4 interacts the one-month
coefficient trends for performers with stronger demand
effects with ticket sales tiers (bottom, middle,
for tickets, more weeks on the Billboard 100, and in the
and top tiers). Specification 5 interacts the
middle of their careers. Trends by gender were also
depth of catalogue using tiers of weeks in the
explored but were not significantly different.
Billboard Hot 100 (1 to 70, 71 to 270, and 271
Given censoring concerns among the 307 concerts
+). Specification 6 interacts tiers for those per­
with no data for ticket sales, a robustness check was
forming professionally for 4–16 years, 17–29
run to assess if these concerts experienced different
years, and 30+ years.
music streaming trends than the 1,549 concerts where
ticket sales were available. This analysis found no
difference in streaming trends between the groups.

IV. Discussion
The post-effects of live concerts on music stream­
ing are positive and significant. While other work
has identified pre-concert effects, this work identi­
fies the mirroring complementarity of live perfor­
mance on post-concert streaming.
The effects are heterogeneous, affecting some
cohorts more than others. Consistent with intuition,
performers who sell more tickets and/or have deeper
catalogues of hit songs are more likely to see positive
post-concert trends for their music. But there are limits
given that performers with longer tenure do not see
a post-concert increase in listenership. This could be
due in part to a perception that concert goers are
already deeply familiar with their catalogue. Many
who attend a Paul McCartney or Elton John concert
tend to be so familiar with their music that they have
already memorized the lyrics and don’t have as much
The specifications in Table 6 highlight the hetero­ preference for re-engaging with their music because it
geneity for each artist characteristic that aligns with is already so familiar to them.
expected complementary effects. Specification 4 Another implication of this analysis involves
shows an increase in the lift of ticket sales among the absolute financial impact on artists. While
the mid and top tiers over the bottom tier and specification 3 highlights a strong pre-concert
 APPLIED ECONOMICS 8445

with deep hit catalogues and/or strong concert

ticket sales?
The analysis includes pre-and-post periods of
up to three months to capture the fuller impact
in streams and dollar terms. Table 7 shows rela­
tively modest effects of concerts on music
streaming. With effects from ~$0 to $4,663 dur­
ing the three months before and after the con­
Figure 5. Month pre and post-effects of concert on streaming cert, performers stand to earn far more from the
(by tier of ticket sales). performance. This also aligns with the finding of
a similar analysis in Germany by (Papies and van
Heerde 2017).
While dollar amounts this small are unlikely to
impact an artist’s concert dates, it does provide
a meaningful signal for testing new songs. If an
artist plays a handful of unreleased songs, they
can offer first release rights to a DSP and monitor
streaming activity to gain early feedback.
Post-concert effects also show some indication
of turning negative if an artist cancels dates. This
Figure 6. Month pre and post-effects of concert on streaming highlights another of many ripple effects that hap­
(by tier of week in top 100).
pen when performers disappoint their fans.

V. Conclusions
Touring is costly and logistically complex for any
performing artist. Top 100 performers, in particular,
have become highly sophisticated road shows where
management leverages complementary revenue
sources to raise the value of all a performer’s assets.
Post-concert effects on streaming highlight another
Figure 7. Month pre and post-effects of concert on streaming source for monetizing a performer’s recording work.
(by tier years performing).
It also holds the possibility to add shows in current or
new markets to make their live performance easier to
effect and a post-concert decay of 10 weeks, attend. It also expands opportunities for exclusive
what does this mean in terms of streams and launches of new music and promotional tie-ins for
dollars? Likewise, what does it mean for artists DSPs.

Table 7. Streaming and revenue estimates.

Incremental streams (000) Estimated incremental revenue

Ln(Streams) itm Pre Week of Post Total Pre Week of Post Total
Bottom tier ticket sales 342.4 42.9 −108.5 276.8 $ 1,164 $ 146 $ (369) $ 941
Mid tier ticket sales 867.0 83.1 289.2 1239.2 $ 2,948 $ 282 $ 983 $ 4,213
Top tier ticket sales 1020.2 113.1 238.1 1371.5 $ 3,469 $ 385 $ 810 $ 4,663
1 to 70 cum weeks 186.7 104.3 −28.8 262.2 $ 635 $ 355 $ (98) $ 892
71 to 270 cum weeks 414.0 137.3 115.3 666.6 $ 1,408 $ 467 $ 392 $ 2,266
271 weeks+ cum weeks 566.7 112.7 285.1 964.5 $ 1,927 $ 383 $ 969 $ 3,279
4 to 16 years performing −106.1 69.1 −78.4 −115.4 $ (361) $ 235 $ (266) $ (392)
17 to 29 years performing 806.7 158.0 332.8 1297.5 $ 2,743 $ 537 $ 1,131 $ 4,411
30+ years performing −5.0 69.1 −78.4 −14.3 $ (17) $ 235 $ (266) $ (49)
Source for stream count: Music Connect. Revenue estimated from DMP median amount paid to artists, writers, publishers, and record labels based on 2018 data
(Pastukhov 2019).
8446 E. HOGUE

Disclosure statement Marketing Science 37 (1): 5–21. https://doi.org/10.1287/

mksc.2017.1051.
No potential conflict of interest was reported by the author(s). Gazel, R. C., and R. K. Schwer. 1997. “Beyond Rock and Roll:
the Economic Impact of the Grateful Dead on a Local
Economy.” Journal of Cultural Economics 21 (1): 41–55.
ORCID https://doi.org/10.1023/A:1007372721259.
Hartmann, W. R., D. Klapper, and Super Bowl Ads. 2016.
Eric Hogue http://orcid.org/0000-0003-2278-3256 Stanford University Graduate School of Business Research
Paper No. 15–16, Available at SSRN: https://ssrn.com/
abstract=2385058.
Hogue, E. 2023. “Promotional Effects of Recorded Music and
Compliance with ethical standards
Superstars on Concert Financial Outcomes.” Journal of
Stata code producing the attached results is available for Interdisciplinary Economics. publication in process.
review. However, several of the data sources are proprietary https://doi.org/10.1177/02601079231168750.
and cannot be shared. Krueger, A. 2005. “The Economics of Real Superstars: The
Market for Rock Concerts in the Material World.” Journal of
Labor Economics 23 (1): 1–30. https://doi.org/10.1086/425431.
Krueger, A. 2019. “Rockonamics. A Backstage Tour of What the
References
Music Industry Can Teach Us About Economics and Life.”
Bracalente, B., C. Chirieleison, M. Cossignani, L. Ferrucci, 181-194. 37:99–103.
M. Gigliotti, and M. G. Ranalli. 2011. “The Economic Mortimer, J. H., C. Nosko, and A. Sorensen. 2012. “Supply
Impact of Cultural Events: The Umbria Jazz Music Responses to Digital Distribution: Recorded Music and
Festival.” Tourism Economics 17 (6): 1235–1255. https:// Live Performances.” Information Economics and Policy
doi.org/10.5367/te.2011.0096. 24 (1): 3–14. ISSN 0167-6245. https://doi.org/10.1016/j.
Christensen, F. 2022. “Streaming Stimulates the Live Concert infoecopol.2012.01.007.
Industry: Evidence from YouTube.” International Journal Papies, D., and H. J. van Heerde. 2017. “The Dynamic
of Industrial Organization 85:102873. ISSN 0167-7187. Interplay Between Recorded Music and Live Concerts:
https://doi.org/10.1016/j.ijindorg.2022.102873. The Role of Piracy, Unbundling, and Artist
Courty, P., and M. Pagliero. 2011. “Does Responsive Pricing Characteristics.” Journal of Marketing 81 (4): 67–87.
Smooth Demand Shocks?” Applied Economics 43 (30): https://doi.org/10.1509/jm.14.0473.
4707–4721. https://doi.org/10.1080/00036846.2010.498350. Pastukhov, D. (2019). What Music Streaming Services Pay per
Datta, H., G. Knox, and J. Bronnenberg. 2018. “Changing Stream (And Why It Actually Doesn't Matter).
Their Tune: How Consumers’ Adoption of Online SoundCharts Blog. Accessed August 1, 2020. https://sound
Streaming Affects Music Consumption and Discovery.” charts.com/blog/music-streaming-rates-payouts.
Copyright of Applied Economics is the property of Routledge and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.