Download Video Advertising by Twitch Influencers* and more Lecture notes Marketing in PDF only on Docsity!
Video Advertising by Twitch Influencers*
Yufeng Huang
University of Rochester
Simon Business School
Ilya Morozov
Northwestern University
Kellogg School of Management
March 23, 2022
Abstract We study the effectiveness of influencer marketing in the video game industry. To this end, we construct a novel dataset on video game streaming on Twitch.tv, the largest video game streaming platform in the world, by monitoring live streams every 10 minutes for eight months. Leveraging these high-frequency data, we isolate plausibly exogenous variation in streamers’ daily schedules and use it to estimate the extent to which live streaming brings additional players into broadcasted games. We find that organic live streams only marginally increase the number of concurrent players in these games. We also find that sponsored streams solicited by game publishers are even less effective than organic streams, implying that sponsored streams generate, on average, negative return on investment (ROI). This result suggests that influencer promotions are less effective than previously thought. We then examine heterogeneous returns to streaming by estimating generalized random forests, and we find that sponsored streaming can significantly benefit games released by small publishers, inexpensive games, and “niche” games that strongly appeal to small groups of consumers. Therefore, despite the negative average ROI, influencer promotions may generate high returns when they promote games by lesser-known publishers or inform consumers about appealing game attributes.
*Contacts: Yufeng Huang, email: yufeng.huang@simon.rochester.edu; Ilya Morozov, email: ilya.morozov@kellogg.northwestern.edu. We thank seminar and conference participants at University of Delaware, HKUST, Marketing Science Conference, and University of Rochester for their helpful comments and suggestions. We also thank James Ryan for excellent research assisstance. All errors are our own.
1 Introduction
Over the last decade, influencer marketing has grown into a $14 billion industry.^1 This rapid growth is partly driven by the increasing popularity of major video content platforms, such as Twitch, YouTube, and TikTok, where influencers distribute their content and build their own fan communities. By 2022, YouTube has become the second most visited website in the world that attracts 30 million visitors per day, and Twitch has grown into a video game streaming giant that hosts over 100,000 live channels and attracts over 2.5 million viewers at any point in time.^2 These large audiences create a unique opportunity for companies to promote products and services by having video influencers showcase them during live streams. In fact, many practitioners believe that because influencers can better engage with their audiences and appear more trustworthy, in- fluencer promotions have greater return on investment (ROI) than traditional advertising.^3 This general excitement about influencer marketing, as it turns out, has little empirical support. When explaining why influencer promotions are effective, many practitioners cite anecdotes in which influencer marketing supposedly generated extraordinarily high ROI.^4 Nevertheless, indus- try observers caution against putting too much weight on these anecdotes, emphasizing that the industry has yet to devise reliable ways of measuring the effectiveness of influencer promotions.^5 Absent experimental variation, measuring the effects of influencer marketing poses a significant empirical challenge. Influencers can often choose which products to promote and when to pro- mote them. They may choose to promote high-quality products at a time when product sales are already trending up, which generates a simultaneity bias familiar from the literature on the effects of word-of-mouth (Seiler et al., 2018) and advertising (Shapiro et al., 2020). In fact, the prior work on influencer marketing emphasizes this simultaneity bias as the central empirical challenge (Li et al., 2021; Yang et al., 2021). Addressing this challenge would help researchers and practitioners understand whether influencer promotions are indeed a highly effective marketing channel. We study the effectiveness of influencer marketing in the video game industry, and we collect unique high-frequency data that enable us to address the main identification challenge. Specifically, we ask how video game streaming on Twitch affects the popularity of broadcasted games, defined as the number of their concurrent players. Twitch is the largest video game streaming platform in the world, hosting over 90% of all streaming content.^6 Streamers broadcast their gaming sessions (^1) Influencer marketing industry has grown to $13.8 billion by 2021 (Influencer Marketing Hub, 2021). (^2) YouTube statistics come from the official platform’s blog (www.blog.youtube), and Twitch statistics are taken from a third-party Twitch monitoring website (twitchtracker.com/statistics). (^3) See the 2019 industry report by MediaKix (Bailis, 2019). (^4) Nielsen Catalina Solutions conducted a correlational study, concluding that influencer marketing generates 11 times higher ROI than online banner ads. (^5) See https://influencermarketinghub.com/influencer-marketing-roi/. (^6) See the Streamlabs and StreamHatchet Quarterly Report for Q3 (2020).
streams and gaming are complementary leisure activities rather than substitutes. In this sense, our results reinforce those of Li et al. (2021), who empirically show that YouTube videos and e-sports tournaments serve as complements to gaming. We also find that sponsored streams are much less effective than organic streams at bringing additional players to the broadcasted games. Specifically, organic streams are four times as effec- tive as sponsored streams, with the estimated elasticities of 0.030 and 0.007. In fact, the estimated elasticity of sponsored streams is half the size of the median effect of TV advertising on sales estimated by Shapiro et al. (2020). We therefore do not find any evidence that influencer promo- tions are more effective than traditional advertising. To assess how profitable it is for publishers to sponsor live streams, we collect daily data on the subscription revenues of individual streamers and use it to proxy their hourly income. Combining this hourly income with the estimated streaming elasticities, we find that sponsored streams generate negative ROI for most games in our sample. Put another way, a typical game publisher in this industry would find that sponsoring influencer promotions is not worth the investment. Although Twitch influencer promotions generate negative average returns, they might still be effective at promoting relatively unknown games and at informing consumers about the quality and gameplay of these games. We enrich our analysis by allowing streaming elasticities to de- pend on game attributes that might be revealed during live broadcasts and on variables proxying which games are relatively unknown. We then estimate heterogeneous streaming elasticities using generalized random forests (Athey et al., 2019). This heterogeneity analysis helps us understand when and why influencer promotions work, distinguishing our paper from the prior literature that focuses on estimating the average effect of such promotions on product usage and sales (Li et al., 2021; Yang et al., 2021). Our results show substantial heterogeneity in streaming elasticities across games. We find that live streaming has large positive effects on games released by small publishers. Traditionally, these publishers have significantly lower marketing budgets than major video game conglomerates, and consumers might often be unaware of the games they release. By contrast, live streaming may increase this awareness by informing consumers about “indie” games released by these publishers. Such an effect is consistent with the prior work that documents the awareness effect of advertising (Honka et al., 2017; Tsai and Honka, 2021). Thus, one implication of our result is that the growth of live streaming might reduce the entry barriers in this industry by providing small publishers a new promotion channel. We also find that Twitch streaming has large positive effects on inexpensive games as well as “niche” games that strongly appeal to some consumers. This result suggests that, by watching several hours of gameplay on Twitch, consumers might learn about both vertical and horizontal game attributes. In this sense, our results suggest that one can view live streams as informative
advertising for video games (Grossman and Shapiro, 1984; Ackerberg, 2003). As a whole, our results show that despite the negative average ROI, Twitch influencer promotions may generate high positive returns when they are used to promote games by lesser-known publishers or to inform consumers about appealing game attributes. Our paper contributes to the literature on the effectiveness of marketing media. Much of the existing literature focuses on estimating the effects of TV advertising (Lodish et al., 1995; Li- aukonyte et al., 2015; Shapiro et al., 2020), online advertising (Johnson et al., 2017; Gordon et al., 2019), and word-of-mouth (Lovett and Staelin, 2016; Seiler et al., 2017). We contribute by study- ing the effects of influencer marketing, an emerging marketing channel that has gained momentum over the past few years. The closest related papers are Li et al. (2021) on the effect of YouTube influencer content about gaming and Yang et al. (2021) on the effect of TikTok influencer videos. Both papers use within-product variation in video uploads across days and examine the impact of these uploads on the usage and sales of promoted products. We attempt to improve upon their anal- ysis by constructing a novel dataset and developing an identification strategy that leverages high- frequency data. By leveraging within-day variation in streamers’ schedules, our approach brings the analysis to the level of granularity at which the simultaneity bias is less plausible.^8 Given the increasing availability of high-frequency data, this empirical strategy might prove helpful in future research on influencer marketing. Using our empirical strategy, we find that sponsored streams only marginally increase the popularity of broadcasted games. Our findings therefore question the conventional wisdom that influencer marketing generates much greater returns on investment (ROI) than that of traditional advertising.^9
2 Measuring Video Streaming on Twitch
2.1 What is Twitch?
Twitch.tv is an Amazon-owned video live streaming platform mostly dedicated to streaming video games and broadcasting esports competitions. The platform has experienced tremendous growth over the last decade. In 2012, it had only several thousand registered channels and 70,000 concur- rent viewers. By 2022, it grew into a video streaming giant that hosts over 100,000 live channels and attracts over 2.5 million viewers at any point in time.^10 This growth was mainly driven by (^8) In this sense, our paper resembles the prior work that measures the effectiveness of TV ads using high-frequency data and discontinuity-in-time research designs (Liaukonyte et al., 2015; He and Klein, 2019). 9 Outside of studying the promotional effect, Simonov et al. (2020); Rajaram and Manchanda (2020); Hwang et al. (2021); Ershov and Mitchell (2020) examine characteristics of influencer media and the extent to which these characteristics explain how viewers choose what to watch and whom to follow. (^10) These aggregate statistics are from twitchtracker.com, a third-party website that monitors the streaming activity on Twitch.tv and reports historical data going back to 2012.
notable examples). One such example is a streamer DrDisrespect whose vibrant personality and unique look – mullet hairstyle, 80s-style mustache, and polarized sunglasses – made him one of the most popular streamers on the platform.
2.2 Data on Twitch Streaming and Game Usage
To study how Twitch streaming affects the popularity of broadcasted games, we require a dataset that contains information about both live streaming and game usage. We construct a novel dataset by combining several data sources. First, we collect high-frequency data from Twitch by continu- ously monitoring live streaming and game viewership on the platform. These data describe when individual streamers go live, what games they stream, and how many viewers they attract at each point in time. Second, we also collect high-frequency data on the number of people currently play- ing each game. To obtain these data, we continuously monitor Steam, the largest online video game platform in the world. These data help us track changes in the popularity of several hundred games during periods when streamers broadcast these games on Twitch. Finally, we complement these two datasets with information on daily (self-reported) subscription counts of individual streamers, which helps us estimate their hourly income.
Video streaming data from Twitch. We monitored video game streaming on Twitch for almost eight months between May 11, 2021 and December 31, 2021. We first pre-selected a list of 60, streamers during a three-week pilot period. Each streamer was selected with the probability pro- portional to the total number of viewers they attracted on Twitch during this pilot period (see Ap- pendix A for details). Then, during the following eight months, we sent high-frequency requests to Twitch API to retrieve information about each streamer. Every 10 minutes, we requested the status of each streamer (online or offline), the concurrent number of viewers, the game they were streaming, and the title of each stream. Most streamers went live on Twitch at least once during the sample period, so our sample covers 96.8% of the streamers we attempted to track (58,060 out of 60,000). Tracking streamers at high-frequency enables us to record the exact times at which a streamer starts and ends broadcasting any game, which serves as the main input to our empirical strategy (see Section 3). Table 1 summarizes live streaming activity on Twitch, both overall and for the most popular streamers. Throughout the paper, we measure streamer popularity using the average number of concurrent viewers. In the last two rows, the table reports averages (1) across top 5% most popular streamers, and (2) across all 58,060 streamers. The average streamer attracts only around 150 viewers and streams about 5.4 hours per day conditional on working on that day (14.6 hours per week). By contrast, the top 15 streamers attract over 30,000 viewers at any given time and often
Table 1: Streaming activity on Twitch. Twitch Primary Game Average Maximum Daily Weekly Stream Stream
Streamer Streamed on Concur. Concur. Hours Hours Start Time End Time
Twitch.tv Viewers Viewers Stream Stream Avg (S.D.) Avg (S.D.) 1 AuronPlay GTA V 104,930 318,181 4.0 5.3 15:20 (0:50) 19:20 (0:50) 2 RanbooLive N/A 75,695 234,626 2.5 21.0 21:00 (2:50) 23:40 (3:10) 3 Ibai GTA V 75,513 1,538,645 4.9 10.0 16:30 (2:50) 21:20 (2:30) 4 Sapnap UNO 72,992 186,592 2.0 3.0 22:20 (3:20) 0:20 (2:40) 5 xQcOW GTA V 72,481 175,453 15.7 52.3 14:00 (7:20) 5:40 (4:30) 6 loud coringa GTA V 65,142 307,450 7.6 10.8 21:10 (5:30) 4:50 (3:10) 7 RocketLeague Rocket League 58,233 208,124 5.0 30.8 18:00 (3:30) 23:10 (3:40) 8 Flashpoint CS: GO 52,655 128,800 7.3 22.3 15:10 (1:50) 22:30 (3:30) 9 Asmongold Final Fantasy 52,135 135,042 7.4 18.3 15:10 (1:20) 22:40 (2:20) 10 thisisnotgeorge Among Us 50,243 113,707 3.5 3.2 21:40 (7:00) 1:10 (6:20) 11 MontanaBlack88 GTA V 48,447 159,731 6.3 5.5 15:20 (2:50) 21:40 (2:20) 12 Rubius GTA V 44,377 207,592 5.3 16.5 17:30 (1:00) 22:50 (1:50) 13 Mizkif Jump King 35,246 189,851 5.9 13.5 20:20 (2:30) 2:20 (3:10) 14 karlnetwork Golf w/Friends 33,127 93,412 5.5 3.3 20:40 (8:50) 2:10 (7:50) 15 shroud N/A 30,979 334,836 11.3 30.0 16:30 (5:20) 3:50 (5:10) Average (top 5% streamers) 2,375 17,988 8.2 22.0 15:20 (4:00) 23:40 (4:20) Average (all streamers) 148 1,277 5.4 14.6 17:40 (4:20) 23:00 (4:20)
This table summarizes live streaming activity on Twitch.tv, both overall and for the most popular streamers, across all games and non-game activities. We measure streamer popularity using the average number of concurrent viewers in our dataset. The primary game of each streamer is defined as a game that this streamer broadcasted for the largest number of hours and that attracted at least 25% of total viewers of this streamer. The last two columns show the average start and end times of live streams and report the standard deviations of start and end times in the parentheses. The time is in the UTC time zone (Coordinated Universal Time). The last two rows report the averages (1) across top 5% most popular streamers, and (2) across all 58,060 streamers.
attract hundreds of thousands of viewers in peak times. A Spanish streamer Ibai, for example, peaked at 1,538,645 active viewers by streaming a series of boxing matches that featured other Spanish streamers, establishing a viewership record in our sample. On average, top streamers work 8.2 hours a day and 22 hours a week (see the second-to-last row of Table 1). Column 3 reports the primary game of each streamer, which we define as a game that this streamer broadcasted for the largest number of hours and that attracted at least 25% of total viewers of this streamer. Although many top streamers broadcast hit games with international following, such as “GTA V” and “CS:GO,” several of them have risen to the top by broadcasting relatively unknown games like “Jump King” and “Golf with Friends.” Table 2 shows summary statistics for this sample of 599 games, describing how often these games are streamed and watched on Twitch. Panel A reports the total streaming and viewership
Table 2: Streaming activity, usage, and attributes of games. Mean Std.dev P5 P25 P50 P75 P Panel A. Streaming activity, viewership, and usage (total during the observation period) No. Streams (All Streamers) 9,079 42,206 14 178 959 4,481 31, No. Streams (Top Streamers) 666 3794 2 15 61 239 2093 Hours Streamed (thousands) 28.2 135.8 0 0.4 2.2 12.5 100. Hours Viewed (thousands) 6,842 52,482 5 72 394 1,689 17, Hours Played (thousands) 30,165 167,518 5 252 1,710 12,267 101,
Panel B. Streaming activity, viewership, and usage (per day) No. Streams (All Streamers) 32.8 152.4 0.1 0.6 3.5 16.2 113. No. Streams (Top Streamers) 2.4 13.7 0.0 0.1 0.2 0.9 7. Hours Streamed 101.7 490.3 0.1 1.3 8.0 45.0 363. Hours Viewed 24,700 189,467 19 262 1,423 6,097 61, Hours Played 130,585 725,187 23 1,092 7,406 53,105 439,
Panel C. Game attributes Publisher Size (no. games) 4.9 5.6 1.0 1.0 2.0 9.0 17. Years Since Release 3.9 4.2 0.1 0.8 2.7 5.5 12. Rating Metacritic Score 78.8 9.7 62.0 75.0 80.0 85.0 91. Regular Price (dollars) 21.1 16.3 0.0 10.0 20.0 30.0 60. Customer Rating St. Dev. 2.3 0.7 1.0 1.9 2.4 2.8 3.
This table shows streaming, viewership, and usage statistics for 599 Steam games in our sample. To estimate the number of people watching a stream at any given point in time, we multiply the number of current viewers obtained from Twitch API by 10 minutes (the frequency of data collection). We construct the streaming time statistics in a similar fashion. The variable “years since release” captures the number of years passed between the official game release and the first day of our data collection. The publisher size is the number of games a publisher released among 599 titles in our sample. The regular price is the 95-th percentile of the distribution of daily prices for a given game, which usually corresponds to the non-discounted price.
of broadcasting on Twitch. To this end, we gathered additional data on the number of active subscribers of each streamer. Because viewer subscriptions represent a significant share of stream- ers’ regular income, the estimate we obtain gives us an informative lower bound on the daily and hourly income of top streamers (see Appendix A.2 for details). Twitch does not publish any official data on subscriptions, so we instead obtained subscription data from a third-party website twitch- tracker.com, which tracks the number of active subscriptions for Twitch streamers who chose to publicly disclose this information. By tracking 10,000 most-subscribed streamers on a daily basis, we collected the current number of active subscriptions and its breakdown by subscription type (i.e., Tier 1, Tier 2, Tier 3, or Amazon Prime), each of which has a fixed dollar value per subscrip- tion. We then converted these subscription counts into daily and monthly income estimates. The resulting estimates capture pre-tax income after the streamer has paid the commission to Twitch.
The main limitation of these data is that streamers self-select into disclosing subscription counts, so we cannot assume our dataset on subscription revenues includes a random subsample of streamers. We find that the average streamer in our sample has 983 active subscriptions and earns around $3,800 per month in subscription revenues. In contrast, the average top 5% streamer earns about $20,000-30,000 in subscription revenues each month. Dividing their monthly subscription income on streaming hours, we obtain that the average top streamer earns about $144 per hour of live streaming. In Section 4.4, we use this number as an estimate of the hourly income of top streamers.
3 The Effect of Stream Viewership on Game Usage
3.1 Empirical strategy
We aim to estimate the causal effect of Twitch stream viewership on the broadcasted games’ pop- ularity, defined as the number of concurrent players. In an ideal experiment, we would make streamers broadcast random games at random times of the day and measure the corresponding lift in the number of concurrent players. Such an experiment is impossible to implement because nei- ther Twitch nor we can control when streamers go live and which games they broadcast. As Seiler et al. (2017) point out, this lack of experimental variation makes it difficult to measure the effect of organic content on product popularity. We adopt an instrumental variable strategy that mimics the ideal experiment and control for confounding factors using fixed effects. We leverage our high-frequency data and focus on vari- ation in the exact broadcast schedules of top streamers within a given day. Although streamers can strategically decide which days to work on and which games to broadcast, our main identify- ing assumption is that within a given day, their streaming schedules do not respond to real-time changes of game popularity. This assumption is reasonable in the context of Twitch streaming. Instead of working regular hours, many streamers start working whenever it is convenient for them given other demands on their time, such as university classes and part-time jobs. They may fin- ish broadcasting earlier than planned due to fatigue or later than planned if events that occur in the game lead to a longer gaming session. Because of these idiosyncratic decisions, many Twitch streamers follow irregular streaming schedules when broadcasting a given game. By leveraging this unpredictable variation in streaming schedules, we estimate how Twitch viewership affects game popularity. We now present model-free evidence that supports this empirical strategy. We first demonstrate that individual streamers follow irregular broadcast schedules. Figure 2 visualizes the variation in broadcast schedules of three streamers, randomly drawn from the pool of the 5% most popular streamers (“top streamers”). In each graph in Figure 2, a row corresponds to 24 hours of a given day, and square markers indicate whether the streamer was live on Twitch in
Table 3: Variance decomposition for start times, end times, and duration of streams. Variance decomposition (% of total variance) Std.dev Across games Across streamers (for a given game)
Across dates (for a given game-streamer) All streamers: Stream start time 6.08 2.0% 46.7% 51.3% Stream end time 6.16 1.9% 42.8% 55.3% Stream duration (hrs) 3.31 4.4% 35.2% 60.4% Top 5% streamers: Stream start time 6.06 3.3% 47.4% 49.3% Stream end time 5.92 4.7% 42.9% 52.3% Stream duration (hrs) 5.67 6.1% 34.7% 59.2% The table shows the standard deviations of start times, end times, and duration of streams (column 1) for the main 599 Steam games in our sample. We decompose this variance into three components: (a) variation across games (column 2); (b) variation across streamers within a game (column 3); and (c) variation across dates within a given streamer (column 4). The table reports statistics for all streamers in the upper panel and the same statistics for top 5% streamers in the lower panel.
and the number of players sharply increase. To isolate the effect of a single broadcast, we identify all days on which a given game was broadcasted by at most one top 5% streamer. We call the top streamer’s broadcast a focal stream. We select focal streams that do not overlap with any other stream of the same game—that is, no other top streamer broadcasts the same game within 10 hours before the start and 20 hours after the start of the focal stream. We then examine the change in the number of viewers and players of the broadcasted game during this 30-hour window. Although these selection criteria help us simplify visualization, we only use them for illustration purposes and relax them in the formal regression analysis.^13 Panel A in Figure 3 shows the number of viewers before and after the start of the focal stream. The figure plots the regression coefficients from a linear model that regresses the log number of viewers on a set of dummies, each capturing one-hour time periods for the 10 hours before and 20 hours after each stream. To account for systematic variation in game popularity that might correlate with broadcast activities, both across days and within a given day, the regression also controls for game-date, game-hour of the day, and time fixed effects—the same set of fixed effects as in our formal model, which we explain in detail below. As Panel A shows, the number of viewers remains roughly constant before the start of the focal stream, increases sharply after its start, remains high for two hours, and then gradually declines to its initial level before the stream. Panel B shows that (^13) This strategy of isolating discontinuous responses using high-frequency data is similar to that of He and Klein (2019) who estimate the effects of radio ads on the sales of national lottery tickets and Liaukonyte et al. (2015) who leverage exogenous variation in the timing of TV ads to estimate how advertising affects online purchases.
(A) Log Viewers (B) Log Players
0
1
2
3
Log Viewers
−8h −4h start +4h +8h +12h +16h +20h Time interval relative to the focal stream
Number of viewers before, during, and after the focal stream
−.
0
.
.^
.
Log Players
−8h −4h start +4h +8h +12h +16h +20h Time interval relative to the focal stream
Number of players before, during, and after the focal stream
Figure 3: The number of viewers and players before, during, and after a focal stream. Both graphs plot the estimated regression coefficients from a linear model that regresses the log number of viewers (left panel) or the log number of players (right panel) on a set of dummy variables for one-hour time intervals, which capture 10 hours before and 20 hours after each stream. The regression controls for game-date, time interval, and hour of the day fixed effects in order to account for predictable changes in game popularity both across days and within a given day. The average focal stream lasts for about four hours.
the number of players of the broadcasted game changes in a similar fashion, suggesting that streams indeed bring additional players into the broadcasted game. Notably, the number of players does not immediately peak after the start of the focal stream when the lift in viewership is the highest, and it slowly returns to its initial level in about eight hours. We interpret this apparent lagged response as preliminary evidence that the increase in stream viewership generates persistent effects on game usage. Motivated by this observation, we now specify a more formal model that allows for (but does not assume) the persistent effects of Twitch viewership.
3.2 Empirical model and estimation
We specify a model that captures both the immediate effect of Twitch viewership on game usage as well as its persistent effects in subsequent time periods. Let j index games, and let t index hour-long time periods. We model the number of players in game j in time t as follows:
log
1 + players (^) jt
= β log
1 +Vjt
- λ (^) j,d(t) + μ (^) j,h(t) + ηt + ε (^) jt (1)
where Vjt is the stock of total viewership explained below; λ (^) j,d(t) are game-date fixed effects, where d(t) is the date to which the time period t belongs; μ (^) j,h(t) are game-hour of the day fixed effects; ηt are time fixed effects; and ε (^) jt are idiosyncratic shocks. The game-date fixed effects λ (^) j,d(t) capture that games often become more or less popular over time, e.g., due to in-game events
one top streamer or not broadcasted by top streamers at all. The main identifying assumption is that, conditional on fixed effects, the broadcast decisions of top streamers in the past 12 hours are orthogonal to idiosyncratic shocks in game popularity, ε (^) jt , so that
E
[
ε (^) jt |Z (^) jt , λ (^) j,d(t), μ (^) j,h(t), ηt
]
One might worry that streamers strategically schedule their broadcasts to coincide with significant in-game events (e.g., new version releases or tournaments), which may create a correlation between the instruments Z (^) jt and shocks ε (^) jt even within a day. However, if this was the case, the number of players would increase even before a top streamer goes live, reflecting that the game is already trending up on Twitch by the time the live stream starts. By contrast, Figure 3 shows that the number of players remains relatively constant before the stream and increases sharply right after the start of the stream. We also demonstrate in Appendix C.1 that controlling for game-week instead of game-date fixed effects produces similar estimates of β and δ , suggesting that streamers do not strategically choose their broadcast days within a week based on game-specific demand shocks. By implication, it might be even less likely that they respond to these demand shocks when choosing the exact broadcasting time within a day. We therefore do not find any indication that within-day endogeneity poses a substantial threat to our empirical strategy. The identifying assumption in (3) implies the condition E
[
ε (^) jt Z (^) jt |λ (^) j,d(t), μ (^) j,h(t), ηt
]
= 0, which we use to estimate parameters β and δ. Specifically, we minimize the sum of squared interactions between residuals ε (^) jt and instruments Z (^) jt :
( βˆ , δˆ ) = arg min (β ,δ ) ∑ j^ ∑ t Z′jt Z (^) jt
log
1 + players (^) jt
− β log
1 +Vjt (δ )
− λ (^) j,d(t) − μ (^) j,h(t) − ηt
This objective function corresponds to GMM estimation with the identity weighting matrix, which under the assumption (3) yields consistent estimates of β and δ. To solve the minimization problem in (4), we perform a golden-section search for the parameter δ , and for a given guess of δ , we estimate parameter β using a closed form 2SLS formula (see Appendix B.2 for details). We obtain clustered standard errors via bootstrap by sampling game-date combinations with replacement.
3.3 The average effect of Twitch streaming
Table 4 presents parameter estimates from the model in (1). The first column shows the estimates from an OLS regression that assumes away persistent effects (i.e., setting δ = 0) and does not include any fixed effects. Because viewership and game usage are highly correlated, the OLS estimate without controls returns a high estimated elasticity of 0.606. The next two columns show
Table 4: The effect of Twitch viewership on video game usage Variable Parameter OLS IV IV Log Viewership Stock Vjt β 0.606*** 0.015*** 0.033*** (0.002) (0.002) (0.009) Effect Persistence δ 0.828*** (0.060)
Game-Date FE No Yes Yes Game-Hour of day FE No Yes Yes Time FE No Yes Yes Observations 3,277,728 3,277,728 3,277, Column 1 shows results from an OLS regression that fixes the persistence parameter to zero (δ = 0) and does not control for any fixed effects. Columns 2-3 shows results from our main specification in (1), without persistence (column 2) and with persistence (column 3). Bootstrap standard errors are clustered at the game- date level. *, **, and *** represent significance at the 10%, 5%, and 1% level.
IV estimates obtained using the GMM estimator in equation (4). Column 2 reports IV estimates from a specification that sets persistence to zero (δ = 0), which yields an elasticity estimate of 0 .015. This estimate is much lower than that from the OLS regression, consistent with the idea that including instruments Z (^) jt and fixed effects helps remove the simultaneity bias. But because this specification ignores persistent effects, it might still generate a biased estimate of the streaming elasticity β. For example, if the true effect persists several hours after a stream, this model will fail to attribute the elevated post-stream game usage to the effect of the broadcast and might bias the estimated β toward zero. Consistent with this observation, we find that allowing for persistent effects raises the estimated streaming elasticity from βˆ = 0 .015 to βˆ = 0 .033 (column 3 in Table 4). We also estimate the persistence parameter to be δˆ = 0 .828, suggesting that the initial effect becomes 17% weaker in every subsequent hour and dissipates to 15% of its initial magnitude within about ten hours. The estimated streaming elasticity of 0.033 is broadly in line with the previous work on the ef- fects of word-of-mouth and advertising. For example, Seiler et al. (2017) estimate the elasticity of 0.016 when studying how the number of organic comments on a microblogging platform increases the viewership of discussed TV shows. Their estimates can be compared to ours because they study the impact of word-of-mouth activity, which is analogous to Twitch viewership in our analysis, and they also adopt a measure of consumption (i.e., TV show audience) as the main outcome variable. Similarly, Shapiro et al. (2020) report the mean elasticity of 0.023 and the median elasticity of 0.014 when estimating the effect of TV advertising viewership on the sales of packaged goods.^14 (^14) Seiler et al. (2017) find no strong evidence of persistent effects. Since their estimation leverages one specific shock, i.e., the blocking of microblogging platform Sina Weibo in China, these results might reflect that the persistent effects are difficult to identify from their data. Shapiro et al. (2020) use a similar construction of viewership stock
3.4 Sponsored versus organic streams
Most of the streams in our data are organic in the sense that streamers broadcast their gaming sessions without getting paid by publishers. If a publisher were to sponsor streams, compensating a top streamer for the time spent broadcasting their game, the effect of such a sponsored stream might be substantially lower than that of organic streams. This might happen because sponsored streams tend to be less engaging or because viewers might negatively react to sponsored content (Ershov and Mitchell, 2020). To empirically assess this conjecture, we identify sponsored streams in our data and estimate the effect of these sponsored streams separately from the effect of organic streams. We identify sponsored streams in our data by searching for specific keywords in stream titles. Streamers are required to disclose the sponsored status of their streams due to several official requirements. First, the Federal Trade Commission requires all influencers to publicly disclose the sponsorship status (FTC, 2019).^15 Second, Twitch facilitates the match between streamers and game publishers via the internal platform “Bounty Board,” which also requires streamers to disclose the sponsored status to viewers. We can therefore identify sponsored streams in our data by searching for the keywords in stream titles that directly reveal the sponsorship status (e.g., #sponsored, #ad). Additionally, several publishers offer partnership programs to streamers who are willing to regularly broadcast their games. Because partnered streamers are also required to disclose their partner agreement on Twitch, we can identify partnered streams by locating stream titles that mention an official partnership program (e.g., #ApexLegendsPartner, #PubgPartner). Using these criteria, we find that around 3% of broadcasts by top streamers are sponsored, and 1% are partnered. Although we do not consider how prominent these disclosures are, the recent work of Li (2022) suggests that some Twitch streamers use long stream titles to make the sponsorship appear less prominent on the platform. We then separately estimate the streaming elasticities for sponsored and organic streams using the following model:
log
1 + players (^) jt
= β ns^ log
1 +V (^) jtns
1 +V (^) jts
- λ (^) j,d(t) + μ (^) j,h(t) + ηt + ε (^) jt. (5)
where V (^) jtns is the cumulative viewership stock of non-sponsored streams, V (^) jts is the viewership stock of sponsored streams, and both viewership stocks are constructed based on the previously estimated decay parameter δ = 0 .828. We interpret β ns^ and β s^ as streaming elasticities of organic and sponsored streams. We estimate equation (5) via two-stage least squares, using as instruments the current and lagged number of sponsored and organic streams by top 5% streamers. As Table (^15) Source: https://www.ftc.gov/news-events/press-releases/2019/11/ftc-releases-advertising-disclosures-guidance- online-influencers (accessed in February, 2022).
5 shows, organic streams are four times as effective as sponsored streams, with the estimated elasticities of 0.030 and 0.007. We find similar results when we consider sponsored streams funded by official partnership programs (estimated elasticities 0.032 vs 0.006). We draw several conclusions from these results. First, organic live streams have positive ef- fects on the short-term popularity of video games. This finding contradicts the belief of some practitioners in the video game industry that live streams divert consumers from playing games by providing an alternative source of entertainment and allowing them to experience the gameplay without paying (Johnson and Woodcock, 2019). In other words, we find that live streaming and gaming should be viewed as complementary leisure activities rather than substitutes. Second, our results show that sponsored streams are relatively ineffective at bringing additional players to the broadcasted games. Later in Section 4.4, we use our elasticity estimates to compute the return on investment (ROI) and find that paying for sponsored streams is not profitable for the average game. Our findings therefore question the conventional wisdom that influencer marketing generates much greater ROI than that of traditional advertising. Although Twitch influencer promotions generate negative returns on average, they might still be effective at promoting relatively unknown games. Given the long format of Twitch broadcasts, such promotions might also reveal rich information about the game’s price, quality, and gameplay. In the next section, we empirically explore these conjectures and study what kind of games might substantially benefit from being promoted in sponsored streams on Twitch.
4 Which Games Benefit the Most from Twitch Streaming?
4.1 Overview of potential mechanisms
To understand whether sponsored streams might be effective at promoting certain games, we need to first understand what kind of games are most likely to benefit from live broadcasts. To answer this question, we need to first understand what mechanisms drive the effect of streaming on game popularity. We hypothesize that Twitch streams either inform consumers about the existence of the broadcasted games or reveal information about their price, quality, or gameplay. First, consumers face an enormous choice set and might not be aware of all offered titles. On Steam alone, they encounter an assortment of more than 60,000 games, which grows with thousands of new titles introduced every year. Twitch streams can draw consumers’ attention to specific games, thus generating an awareness effect (Honka et al., 2017; Tsai and Honka, 2021). An example of such an awareness effect is the game “Among Us” by a small indie studio, which stayed dormant on Steam for almost two years and only became popular when consumers learned about it from Twitch broadcasts. Additionally, even if consumers are aware of a game, they might learn something about
