Subscription services in the music industry : when do they result in the increased profits of the company?

(1)

Subscription services in the music industry - when do they result in the

increased profits of the company?

Bachelor Thesis, BSc Economics and Business

Iana Chernysh

11084839

31st January, 2018

University of Amsterdam

(2)

Statement of Originality

This document is written by Student Iana Chernysh who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

With the digitization of society, many entertainment industries experienced a need to pursue new business models in order to keep the business sustainable. One of the main reasons for this was an increase in digital piracy, which led to huge revenue drops in the entertainment industries. In the music industry, a subscription service was introduced in order to exploit the opportunities given my technological progress and decrease the number of illegal music downloads. This paper looks at what features assure that the subscription service is successful and how it influences overall revenues for the industry. It also takes Apple Music as an example of such service and analyzes how its introduction influenced the revenues of the company. The regression analyses are used to determine whether an introduction of a subscription service has a significant effect on revenues and how this effect may be explained. It was concluded that the introduction of Apple Music led to a structural break in revenues. At the same time variable of paid subscription was found to be insignificant. Although the main findings of this paper were controversial, they do give a base for future research.

(4)

With the digitization of the society, many industries have experienced serious changes in their business models. The size of digital markets of information goods has increased significantly (Hui, Yoo, Choudhary, & Tam, 2012). The music industry is not an exception and is now in a stage of a transformation process with the main changes being characterized by a drop in revenues and a significant growth in digital sales (Thomes, 2013, Wlomert & Papies, 2016). This created a need for new business models to meet the growing demand for digital music and renovation of the old ones (Wlomert & Papies, 2016; Cesareo & Pastore, 2014). Researchers also became more interested in the pricing of such experience goods, especially bundling and subscription services (Hui et al., 2012).

Digital music consumption has been present for many years and started becoming more popular during 1996 with the widespread of file-sharing websites (Kwong, Park, 2008). Such fast evolution in technology changed the way people perceive and retrieve music (Weijters, Goedertier, & Ver-streken, 2014). Moreover, new business models started appearing. For example, a big share of mu-sic nowadays can be purchased with unbundling, meaning there is no necessity to buy a whole al-bum, as one can purchase tracks individually (Elberse, 2010; Papies &Van Heerde, 2017). Current-ly, music streaming, which is the main focus of this research, is viewed as a new window in the in-dustry and is growing rapidly, making people listen to music more than in previous decades (Swan-son, 2013).

Another important reason for creating streaming services is a huge increase in music piracy (ille-gal music downloads). Growth in digital piracy significantly decreased the revenues of the music industry after the appearance of the first file-sharing services. (Small, 2012; Cesareo & Pastore, 2014). A huge drop can be seen in the period from 1999 to 2008, where the revenues in the industry dropped from 14.6 billion to 8.5 billion (Goel, Miesing, & Chandra, 2010). This was primarily caused by the appearance of the first file sharing service Napster. The service gave customers more power, changed the dynamics of distribution and cost (Scharf, 2011). Scharf (2011) also mentions that once consumers got an illegally downloaded song, they would not engage in downloading mu-sic legally anymore. Therefore, finding new sources of income from digital markets became an im-portant task of companies operating in the music industry (Thomes, 2013). Music streaming is con-sidered to be a solution to this problem. It is expected that streaming services will be a powerful in-strument in reducing illegal access to music and thus solve the issue of music piracy (Thomes, 2013; Borja & Dieringer, 2016).

(6)

The aim of this research is to find out to what extent and when it is profitable for a company to issue a subscription service in the music industry. The results of the research will help companies in making a decision on whether to pursue the subscription service strategy. The paper will also em-phasize what features are crucial for a successful implementation of the service. The structure is as follows: the review of previous researches on this topic, analysis of consumer demand and how to make a successful business model through subscription, as well as a comparison of subscriptions in the music industry to subscriptions in the video industry. Then, this paper will take one company that is offering music subscription services and analyze what its business model looks like. A re-gression will be made afterward to see whether the addition of this service significantly influences the profits of the company. Therefore, the research question will be answered by a combination of analysis of previous literature on the topic and new findings from the regression.

2. Literature review

Digital music services is a recently emerged industry, that is aimed at drawing consumers away from the illegal music downloading (Kwong & Park, 2008). The subscription-based music services (or on-demand streaming services) became a new and important channel for legal music distribution, also becoming one of the main sources of revenue for the industry (Cesareo & Pastore, 2014). As an alternative to other business models, music streaming is growing rapidly each year and increasing its revenues (Kim, Nam, & Ryu, 2017). For example, in 2014, worldwide music streaming services revenue grew by 39% (Sinclair & Tinson, 2017). The appearance of these services was not only an opportunity for technology companies (like Google or Apple) to have a new source of income, but also for the product right owners to experiment with new business models and revalue their propositions for the customers (Hampton-Sosa, 2017). A popularity of such services can also be explained by low costs in storing and distribution of the media, meaning low marginal costs in distributing digital music (Cesareo & Pastore, 2014; Elberse, 2010).

At the same time, the existence of these services is still a topic of controversy and ongoing debate (Wlomert & Papies, 2016; Swanson, 2013). The reason for this is the unclear effect of on-demand streaming on revenues of the industry. Free streaming (like Spotify) attracted new customers to the industry and decreased the number of illegal downloads. It is believed that the whole market can be expanded by these services (Kim et al., 2017; Wlomert & Papies, 2016). Moreover, as music is an experience good, such services are considered to be a discovery/sampling tool, which could stimulate digital music sales and the sales of complementary products (for example, concert tickets, official merchandise). It will also decrease the risk of unfulfilling customers expectations (Aguiar, 2017; Wlomert & Papies, 2016; Nguyen, Dejean, & Moreau,

(7)

2013). Reduced barriers to access music and the sampling effect of subscription services is expected to increase record music sales and consumers’ willingness to pay for such services (Lee, Choi, Cho, & Lee, 2016). Lee et al. (2016) made an analysis proving that streaming has a positive effect on offline record sales. There is also an effect that works the other way around: purchasing of licensed products increases the number of subscriptions, which in their turn help the company to grow in many ways (Wayne, 2017). On the other hand, it was also shown by several researchers (Mortimer, Nosko, & Sorensen, 2012; Papies & Van Heerde, 2017) that the development of digital music decreases album sales, but increases revenues from live performances. Moreover, Nguyen et al. (2013) found that streaming has no effect on overall music sales, but it does increase people’s willingness to attend live concerts.

With the appearance of music subscriptions, already existing customers started to switch to the new service which reduces their expenditure (Wlomert & Papies, 2016). Such cannibalization results in lowered profits of the industry. Even though streaming is considered to be a solution for music piracy, Borja & Dieringer, (2016) showed that piracy and streaming can co-exist and people using streaming services are more likely to engage in illegal music downloads. At the same time, Aguilar & Waldfogel (2017) found that streaming displaces both sales and piracy, which has an unclear effect on the industry and confronts with findings of Borja & Dieringer (2016).

2.1 Customer’s attitude

Customer’s willingness to pay changed significantly with the appearance of opportunities to

download music for free (Waldfogel, 2010). According to Waldfogel, 2010, it was hard for compa-nies to draw people away from illegal music downloads at first by providing a la carte services. He also stated that only in 2003, with the appearance of Apple’s iTunes, the share of legal music sales grew significantly followed by a drop in illegal music downloads. That is why it became crucial for the companies to analyze how to attract consumers and what customers appreciate the most about on-demand music streaming services (Kwong & Park, 2008). It was also proved that the revenues do depend on consumer preferences (Gamble, Brennan, & Mcadam, 2017).

As music is now an active part of people’s everyday lives, it is important for people to integrate it in the most useful and convenient way (Sinclair & Tinson, 2017). Several researchers (Kim et al., 2017; Weijters et al., 2014) found that people prefer services that offer both streaming and downloading, as well as different packages: free with advertisements and paid. It is also important for companies to provide an enjoyable experience for users by investing in service activities, like a creation of a mobile app or making a user-friendly interface (Kwong & Park, 2008; Cesareo & Pastore, 2014). Attaching social media accounts to the streaming service is also a feature that

(8)

attracts customers because it allows them to show their identity and express themselves (Sinclair & Tinson, 2017; Weijters et al., 2014). Providing an excellent service in this type of industry is crucial as the relationship between the provider and the consumer is usually repetitive and long-term, moreover, the company does not want any customers to cancel the subscription (Fruchter & Sigué, 2013). As well as this, if the service meets consumer’s needs and is more convenient than an illegal option, a consumer will stick to the legal service and, thus, reduce the overall amount of piracy (Management Science, 2017). Lastly, consumer’s willingness to pay for an Internet service largely depends on their attitude and the valuation of the service (Lu & Hsiao, 2010).

2.2 Streaming as a business model

As already mentioned, on-demand streaming is getting more and more popular among

compa-nies and their customers. For example, it already accounted for 42% of all digital sales in 2012 and this number is expected to expand further (Nguyen et al., 2014). This trend is also seen in other en-tertainment industries. The DVD sales increased in the period of 2000-2003 because of the increase in the online broadcasting of movies (Lee et al., 2016). Also, an article published by Jeunesse (2014) mentioned that offering digital content for teens TV-shows increased ratings for traditional TV-broadcasts as well. In general, many industries, such as software, movies and, games, already adopted the subscription model (Hampton-Sosa, 2017).

If one wants to analyze how successful of a business model issuing a music subscription service is, there is a need to look at costs and revenues. The music industry as a whole has high barriers to entry and incurs high initial costs (Goel et al., 2010). Moreover, the threat of substitution is strong as well as supplier bargaining power (Small, 2012). However, once the service is established and the issues with the music rights are solved, the subscription service is likely to provide high revenue. This happens, because the marginal costs are very low or even zero in some cases (Aguiar & Waldfogel, 2017; Goel et al., 2010). In general, the digitization of society decreases both the fixed costs and marginal costs, making it easier for the company to distribute their product (Management Science, 2017). It also happens that the variation in subscription sales is lower than in individual sales, which allows the company to maximize the surplus (Hui et al., 2012).

A research analyzing the profitability and competitiveness of on-demand streaming was done by Small (2012). It was proved that the paid on-demand service generates more profit than both the ad-supported free subscription and a-la-carte services. However, free-online streaming can still be a highly profitable model as long as advertising is not disturbing consumers (Ngyuen et al., 2014). Supporting these results, it was found by Swanson (2013), that an average consumer spends 60$ per

(9)

year on a-la-carte music downloads, whereas with using the subscription this amount doubles to 120$ per year.

In order for the service to be successful, it is important to offer a large catalog of content and great service. It is also crucial to make contracts which give enough revenue to the right holders, as not all artists are willing to sell their music to the streaming services (Small, 2012). Also, a compa-ny needs to work on its brand name, as some consumers do not want to go sampling each service offered and are going to go with the «most popular» option (Wayne, 2017). Already in 2012, it was offered by Small that Apple iTunes store is a perfect platform to issue a profitable subscription ser-vice.

2.3 Music industry and video industry comparison

The topic of subscription services is relatively new and becoming a more and more popular and crucial theme for research, especially considering how many various industries are involved in such services (Fruchter & Sigué, 2013). That being said, this paper will compare music industry to video industry in order to find similarities between subscription services adoption, as some findings from video industry researches can be applied to the music industry as well.

The two industries both offer entertainment experience goods and go through changes due to the digitization of society. In the video industry, same as in the music, the subscription service appeared as an alternative to physical goods and is considered to be more cost-effective (Alsmirat & Sarhan, 2010). Currently, both industries offer different types of subscription services: a la carte (pay per view), subscription-based, free with advertisement and hybrid (a mix between subscription-based and free with advertisement) (Al-Hadrusi & Sarhan, 2014). As mentioned above, the paid subscrip-tion does generate more revenues than free with advertisement for a music industry. However, in their paper Al-Hadrusi & Sarhan (2014) prove how using ads can generate profit and be beneficial for a subscription in the video industry. Firstly, ads do generate some revenues. Secondly, some ads are actually liked by the consumers, for example, movie trailers, meaning that ads can be used as a promotion of a similar good. In the music industry, it can be an advertisement of a new album. For example, in the video industry, some movies are now being released only online, as it is more prof-itable than a theatre release (Alsmirat & Sarhan, 2010). Lastly, ads allow the request to process faster, reduce waiting time for people and, thus, decrease the delivery cost for the company (Al-Hadrusi & Sarhan, 2014).

Other findings on this topic focused on consumer’s benefits. For example, a fixed-priced sub-scription assures that consumers do not get unexpectedly high bills (Arul & Shoufan, 2016). As well as this, they do not need to decide every time whether the benefit from watching a movie exceeds

(10)

the cost of purchasing it (Arul & Shoufan, 2016). At the same time, Arul & Shoufan (2016) men-tioned several negative sides of TV/movie subscription. One of them is that non-frequent users may find the price of the service too high, as they will not use it that often. On the other hand, large households would be willing to subscribe to several services in order to meet everyones’ interests, which makes their overall expenses very high (Arul & Shoufan, 2016).

One of the interesting new trends in the video industry is binge-watching, which implies watch-ing a whole season of a TV-series in a day, for example (Jenner, 2017). This trend appeared because of easier access to movies and series, partially thanks to subscription services (Jenner, 2017). More-over, people try to watch everything popular as soon as possible, because it allows them to keep up to date and engage in conversations (Jeunesse, 2014). This might also be applied to music industry. With easier access and a larger variety of music offered, people listen to music more than before and increase their overall consumption.

2.4 Previous findings

Previous researchers in the music industry focused on the perspective of the record labels, mainly on revenue models (Gamble, Brennan, & Mcadam, 2017).

Lee et al (2016) in their research focused on what influences sales of music records. They looked at such variables as previous sales, the number of online streamings, the audience rating of the record and price. They did find that the amount of streaming has a positive correlation with the number of record sales and thus increases the revenue of the record label. A similar research was done by Elberse (2010). In her paper, she looked at what drives the physical album sales and digital song sales in a bundle. A bundle - being a mix of physical and digital purchases of music. She looked at how the share of legal and illegal downloads of music in the household influences the sales, as well as the concentration of sales, availability of all songs from the album on iTunes, genre, and number of competitors in the market. The findings showed that downloading songs digitally positively influences the number of albums sold and vice verse. It was also shown that if some of the songs are not available on iTunes, it drives the sales for the album down. Another research conducted by Papies and Van Heerde (2017) looked directly at record label revenues. The influences of such variables were analyzed: advertising, amount of radio and video rotations, level of piracy and unbundling and google searches. The authors proved that piracy has a negative and significantly large effect on revenues. What is more interesting, is that the amount of radio and video rotations have a positive effect on companies revenues. Radio and video rotations can be seen as a tool similar to online streaming or illegal music sharing, as it has the same sampling effect. Thus, this research may expect to find a positive correlation between music subscription service and

(11)

revenues. Wlömert and Papies (2016) conducted a research comparing two business models: free streaming service supported by ads and a paid subscription services. The total revenue was found to be larger for a paid subscription, even though more users were engaged in the free streaming service. They also looked at cannibalization effect streaming service has on revenues from physical record sales. It was found that this effect is larger for a free streaming service than for a paid subscription.

A lot of papers looked at which types of consumers are more likely to become active users of an on-demand music streaming or engage in music piracy. Such a research was conducted by Nguyen et al (2014), who saw that being an active Internet user has a negative correlation with listening to music through streaming. Moreover, it was found that younger audience has a lower probability to purchase digital music legally. Among other variables, they looked at were education level, music taste and standard of living.

As mentioned above, most of the researchers did focus on the record label revenues or the factors influencing the likelihood of engaging in music piracy. Alternatively, this research will look at the perspective of the company issuing the on-demand streaming service. This topic was not discussed and analyzed before based on a real company, this may be explained by the lack of data and novelty of the topic. At the same time, some papers looked at hypothetical companies and built economic models that proved the profitability of the on-demand streaming. That is why the research done in this paper contributes to the previous researches made on this topic.

3. Methodology

Research in technology adoption became a common and highly popular topic several decades ago (Hampton-Sosa, 2017). As mentioned in the literature review, this research is focused on a current and very discussed topic and it will contribute to previous studies and provide a base for future ones. The main objective of this paper is to investigate the significance of adding a subscription service to a range of services and merchandise offered by the company. The absence of previous research on this topic together with the influence digital consumption has on entertainment industries highlight the importance of this research and subscription service as a variable of interest. However, there are several limitations to this research. As the chosen topic is very novel, the data sample is not large enough, which increases standard errors and makes estimators less efficient. At the same time, the error term becomes homoskedastic, which allows using t-test. Another limitation is that data is not divided in much detail, for example, there is no separate data on revenues from Apple Music and iTunes, they both fall under the service revenues category.

(12)

In order to answer the research question mentioned above, a regression analyses will be done. As an external company, Apple is taken. This company was chosen as a representative of the industry because it is one of the largest companies in the technology industry. Apple has a well-established brand name and loyalty of the customers, meaning the customers are willing to try new products and services issued by Apple. The company is also famous for making technological breakthroughs. Apple is a good example, as it already had a digital platform of iTunes, meaning it had resources (a customer base, rights to distribute the music, etc.) for issuing a subscription service. As analyzed by Small (2012), Apple had all necessary capabilities to become successful in this industry. Moreover, already in 2008 iTunes store was responsible for one-third of all digital music sales (Waldfogel, 2010), proving that a company had a plausible approach to the digital music market.

3.1 Data and Hypotheses

A regression will be built on dependent variable SerRev and TotalRev. The independent variables are: Year and PaidSub. The variables chosen differ from the ones used in previous researches. This is the case because previous papers focused on record label revenues and not on the company issuing the service.

SerRev: includes revenues from internet offered services (Apple Music, iTunes, Apple Care and Apple Pay). This is the main variable of interest, as it is supposedly directly influenced by the introduction of Apple Music.

TotalRev: a sum of all revenues of the company. An introduction of a new service can also influence other revenues, for example, revenues from selling of the merchandise, thus, it should be investigated whether total revenues change with an introduction of Apple Music.

PaidSub: dummy variable, which equals 1, when the company has a paid subscription. The variable which significance is tested in order to find the answer to the research question.

Year: includes quarter and year (for example, Q1 of 2013). This variable is important when investigating the presence of a structural break.

The linear regressions look as follows:

SerRev = β0 + βyear*Year + βpaidsub*PaidSub + e TotalRev = β0 + βyear*Year + βpaidsub*PaidSub + e

The Apple Music (paid on-demand streaming service) was issued in June 2015. The data was collected from the quarterly reports of Apple Inc. from 2013 till 2017, which gave a total n=20, with 10 observations before issuing of Apple Music and 10 observations after. The data was put in the excel sheet and then imported to STATA. In order to make the variable year continuous for STATA to recognize and have the effect of the quarter on revenues, the variable Year was recorded in the

(13)

following way: year.0 for the first quarter of the year, year.25 for the second, year.5 for the third and year.75 for the fourth quarter. Once the data has been imported into STATA, variable Year was changed from string variable to numeric.

In the first step of the analysis, a t-test will be done to test whether having a paid subscription has a significant impact on service revenues. This test will help to see the overall effect that implementing a new service has on revenues and will help to determine whether the effect is positive or negative. For that SerRev is regressed on Year and PaidSub. Moreover, to see what effect PaidSub has on total revenues, the regression of TotalRev on Year and PaidSub will be done as well.

H0: β(paidsub)=0 H1: β(paidsub) not = 0

The next step will be to see whether the company experiences a change in revenues after issuing of AppleMusic. This will be done by examining whether the structural break was present after June 2015. The test conducted for this analysis is Chow test, which measures whether the Beta stays the same across different periods of time, in this case before and after issuing of AppleMusic.

H0: Structural break does not exist H1: Structural break exists

The preparations for this test are done in STATA, however, the test itself is conducted manually. First, SerRev is regressed on Year, to get the SSR of the total period. Afterwards, SerRev is regressed on Year if PaidSub is equal to 0 (for the period before issuing of Apple Music) and if PaidSub is equal to 1 (for the period after). The SSRs are then put in the formula to find the F- statistic. F = ((SSRt - (SSRb + SSRa))/k)/((SSRb + SSRa)/n-2k) - where SSRt is total SSR, SSRb is SSR for the period before, SSRa is SSR for the period after, n is the number of observations and k is the number of total parameters.

4. Results and Analyses

4.1 Regression

In this part, the results of the conducted analyses, described in the previous part of the thesis, will be defined. In total there were 5 regressions conducted in this research, two being the linear regression for the t-test and 3 being the linear regressions for the Chow test.

Table 1 shows the results of the regression of SerRev on PaidSub and Year. The R-squared of the model shows that 91.05% variation in the dependent variable can be explained by the variation in the independent ones included in the model. It can be seen that Year does increase service revenues and its effect is statistically significant (p-value < 0.05). The coefficient for PaidSub is negative,

(14)

meaning that the introduction of the paid subscription decreases revenues. This may be explained by the fact that people did spend more money on a la carte service iTunes. Once Apple Music was issued, consumers switched to it, as it is a cheaper way of acquiring music. A more detailed reasoning for this can be found in the literature review. At the same time, the p-value for this coefficient is larger than 0.05, so the hypothesis of βpaidsub = 0 cannot be rejected, meaning that issuing of the subscription does not have an effect on service revenues. The reason for such a result may be a small sample and high standard errors.

Table 1

Regression of service revenues on Year and PaidSub

Table 2 shows the results of the regression of TotalRev on PaidSub and Year. As can be seen in the table, the p-values are much larger than 0.05, meaning that neither Year nor PaidSub have a significant effect on total revenues. With these results, a hypothesis of βpaidsub = 0 cannot be rejected. R-squared is very small as well, meaning that only 10.92% of the variation of the dependent variable is explained by the model. However, the coefficient to PaidSub is also negative here. The fact that Year influences service revenues and does not influence total revenues suggests

Source SS df MS Model 29687682.1 2 14843841 Residual 2918528.86 17 171678.168 Total 32606210.9 19 1716116.37 Number of obs 20 F (2,17) 86.46 Prob > F 0.0000 R-squared 0.9105 Adj R-squared 0.9000 Root MSE 414.34

SerRev Coef. Std. Err. t P> |t| [95% Conf. Interval]

Year 916.8606 129.0255 7.11 0,000 644.6406 1189.081

PaidSub -241.8515 371.9987 -0.65 0,524 -1026.7 542.9971

(15)

that even though the coefficient for PaidSub is not statistically significant, a structural break in service revenues may still be present.

Table 2

Regression of Total revenues on Year and PaidSub

Tables 3, 4 and 5 show the regressions used to find the SSRs for the Chow test. In all three regressions, R-squared is larger than 0.9, which states that the data is a good fit for the dependent variable. The F-statistic was estimated to be 32.81. The F-statistic is much larger than the critical value (3.00) meaning that H0 can be rejected and it can be concluded that the structural break is present in the service revenues after the Apple Music is issued. Together with the result of previous regressions, it can be interpreted that issuing the paid subscription did have an effect on service revenues and that effect is likely to be negative. The negative effect can be expected because of the estimated βpaidsub in both regressions (-241.8515 and -1286.376).

Source SS df MS Model 327007517 2 163503758 Residual 2.6664e+09 17 156845445 Total 2.9934e+09 19 157546320 Number of obs 20 F (2,17) 1.04 Prob > F 0.3741 R-squared 0.1092 Adj R-squared 0.0044 Root MSE 12524

TotalRev Coef. Std. Err. t P> |t| [95% Conf. Interval]

Year 3183.03 3899.905 0.82 0,426 -5045.049 11411.11

PaidSub -1286.376 11243.97 -0.11 0,910 -25009.09 22436.33

(16)

Table 3

Regression total period of time

Table 4

Regression period before issuing of Apple Music

Source SS df MS Model 1249618.94 1 1249618.94 Residual 106457.964 8 13307.2455 Total 1356076.9 9 150675.211 Number of obs 20 F (1, 18) 178.22 Prob > F 0.0000 R-squared 0.9083 Adj R-squared 0.9032 Root MSE 407.64

SerRev Coef. Std. Err. t P>|t| [95% Conf. Interval]

Year 844.1233 63.2307 13.35 0,000 711.2805 976.9661 _cons -1695806 127433.6 -13.31 0,000 -1963534 -1438078 Source SS df MS Model 1249618.94 1 1249618.94 Residual 106457.964 8 13307.2455 Total 1356076.9 9 150675.211 Number of obs 10 F (1, 8) 93.91 Prob > F 0.0000 R-squared 0.9215 Adj R-squared 0.9117 Root MSE 115.36

(17)

Table 5

Regression period after issuing of Apple Music

4.2 Data

This paper will also look at the pattern that the revenues form during given period of time. For that, some additional data has been found: the amount of merchandise revenues and other revenues for the period 2013-2017. Merchandise revenues include sales of iPhone, iPad, and MacBook; other revenues include money made from sales of accessories. Graphs 1 and 2 below present the movement of merchandise and other revenues over time.

Year 492.2909 50.80157 9.69 0,000 375.1423 609.4395 PaidSub 0 (ommited) _cons -987141.5 102320.7 -9.65 0,000 -1223094 -751189.5 Source SS df MS Model 9278338.05 1 9278338.05 Residual 953145.552 8 119143.194 Total 10231483.6 9 1136831.51 Number of obs 10 F (1, 8) 77.88 Prob > F 0.0000 R-squared 0.9068 Adj R-squared 0.8952 Root MSE 345.17

Year 1341.43 152.0084 8.82 0,000 990.8983 1691.962

PaidSub 0 (ommited)

(18)

Graph 1

Other revenues over time

Graph 2

Merchandise revenues over time

It can be seen that the revenues follow a pattern of waves, with having its peak in the first quarter of each year. Such pattern may be explained by «January effect» - a market anomaly, during which stock prices rise sharply all over the world, which may lead to increased revenues as well. There are no visible crucial changes after quarter 3 of 2015 when Apple Music was issued, so it is not possible to say which effect the introduction of a new service had on merchandise and other revenues.

(19)

Graph 3

Service revenues over time

The line has an upward sloping trend, meaning the revenues from services slowly increase over time. However, there can be seen a slight jump from quarter 4 of 2015 to quarter 1 of 2016. As Apple music was issued at the end of quarter 3 of 2015, this jump may be considered a sign that introduction of this service indeed influenced the revenues. This conclusion also goes along with the results of the Chow test, which showed that structural break is present in the amount of service revenues.  

5. Conclusion

The main goal of this paper was to analyze the opportunities that subscription services offer and their profitability for the company. Specifically, several papers were reviewed in order to give a theoretical overview and then a subscription service issued by Apple was chosen for the regression analyses. The data were collected quarterly for the period of 2013 till 2017, so two and a half years prior and after the launching of Apple Music were considered as a research period to test the impact of a paid subscription service on company’s revenues.

Based on the literature review, it can be summarized which aspects can allow the music service to actually be successful and profitable. First of all, a company has to have a high initial budget in order to maintain the relationship with the music right holders. It is also important that the subscrip-tion service offers a user-friendly interface and a mobile app. Moreover, a company is more likely to be successful if it has an established and popular brand name.

The regression analyses were based on the expectation that the introduction of the subscription service will have a significant impact on the revenues, meaning a structural break will be present. Also, two t-tests were conducted in order to see whether paid subscription has any influence on the revenues. Although the paid subscription was not found to be statistically significant for the

(20)

revenues, the results supported the hypothesis that a structural break was present. Controversial results can be explained by the limitations of this research, mainly, the lack of data. As Apple Music is a new service, launched only in July 2015, not enough observations can be found regarding the revenues. A small number of observations makes it difficult to make reliable claims, monthly data could have been used instead to make the regression more precise. Moreover, standard errors for each variable are big, which makes these results a concern for the reliability of this study.

As a suggestion, future researchers can repeat this analyses after several years to see whether the results will be different with a larger sample. They can also look at how Apple Music influenced the sales from iTunes store, as a lot of theories are built around the mutual influence of a la carte and subscription services. Another suggestion would be to look at several different companies. Even though a chosen company is a good representative of the industry, including several representatives will increase external validity and lead to more precise conclusions.

(21)

References

Aguiar, & Waldfogel. (2017). As streaming reaches flood stage, does it stimulate or depress music sales? International Journal of Industrial Organization, 1-30.

Aguiar, L. (2017). Let the music play? Free streaming and its effects on digital music consumption.

Information Economics and Policy, 41, 1-14.a

Al-Hadrusi, M., & Sarhan, S. (2014). A scalable delivery solution and a pricing model for commercial video-on-demand systems with video advertisements. Multimedia Tools and Ap

plications, 73(3), 1417-1443.

Alsmirat, M., & Sarhan, A. (2010). Detailed Performance and Waiting-Time Predictability Analysis of Scheduling Options in On-Demand Video Streaming. EURASIP Journal on Image and Video

Processing, 2010(1), 1-20.

Arul, & Shoufan. (2016). Subscription-free Pay-TV over IPTV. Journal of Systems Architecture, 64, 37-49.

Borja, & Dieringer. (2016). Streaming or stealing? The complementary features between music streaming and music piracy. Journal of Retailing and Consumer Services, 32, 86-95.

Cesareo, L., & Pastore, A. (2014). Consumers’ attitude and behavior towards online music piracy and subscription-based services. Journal of Consumer Marketing, 31(6/7), 515-525.

Elberse, Anita. (2010). Bye-bye bundles: The unbundling of music in digital channels.(Report).

Journal of Marketing, 74(3), 107-123.

Fruchter, & Sigué. (2013). Dynamic pricing for subscription services. Journal of Economic

Dynamics and Control, 37(11), 2180-2194.

Gamble, Brennan, & Mcadam. (2017). A rewarding experience? Exploring how crowdfunding is affecting music industry business models. Journal of Business Research, 70, 25-36.

Goel, S., Miesing, P., & Chandra, U. (2010). The Impact of Illegal Peer-to-Peer File Sharing on the Media Industry. California Management Review, 52(3), 6-33.

Hampton-Sosa, W. (2017). The impact of creativity and community facilitation on music streaming adoption and digital piracy. Computers in Human Behavior, 69, 444-453.

Hui, Yoo, Choudhary, & Tam. (2012). Sell by bundle or unit?: Pure bundling versus mixed bundling of information goods. Decision Support Systems, 53(3), 517-525.

Jenner, M. (2017). Binge-watching: Video-on-demand, quality TV and mainstreaming fandom. International Journal of Cultural Studies, 20(3), 304-320.

Kim, Nam, & Ryu. (2017). What do consumers prefer for music streaming services?: A comparative study between Korea and US. Telecommunications Policy, 41(4), 263-272.

(22)

Kwong, S., & Park, Jungkun. (2008). Digital music services consumer intention and adoption. The

Service Industries Journal, 28(9), 1461-1479.

Lee, Minhyung, Choi, Hanbyeol, Cho, Daegon, & Lee, Heeseok. (2016). Cannibalizing or Com- plementing?. The Impact of Online Streaming Services on Music Record Sales. Procedia

Com-puter Science, 91, 662-671.

Lu, & Hsiao. (2010). The influence of extro/introversion on the intention to pay for social networking sites. Information & Management, 47(3), 150-157.

Mortimer, Nosko, & Sorensen. (2012). Supply responses to digital distribution: Recorded music and live performances. Information Economics and Policy, 24(1), 3-14.

Nguyen, G., Dejean, D., & Moreau, S. (2014). On the complementarity between online and offline music consumption: The case of free streaming. Journal of Cultural Economics, 38(4), 315-330. Papies, D., & Van Heerde, H. (2017). The Dynamic Interplay Between Recorded Music and Live

Concerts: The Role of Piracy, Unbundling, and Artist Characteristics. Journal of Marketing, 81(4), 67-87.

Scharf, N. (2011). Napster's long shadow: Copyright and peer-to-peer technology. Journal of Intellectual Property Law & Practice, 6(11), 806-812.

Sinclair, & Tinson. (2017). Psychological ownership and music streaming consumption. Journal of

Business Research, 71, 1-9

Small, O. (2012). Reshaping The Music Distribution Model: An Itunes Opportunity. Journal of Me-

dia Business Studies, 9(4), 41-68.

Swanson, K. (2013). A Case Study on Spotify: Exploring Perceptions of the Music Streaming Ser- vice. MEIEA Journal, 13(1), 207-230.

The Effect of Subscription Video-on-Demand on Piracy: Evidence from a Household-Level Randomized Experiment. (n.d.). Management Science, Management Science, 2017. The Netflix Effect: Teens, Binge Watching, and On-Demand Digital Media Trends. (2014). Jeunesse: Young People, Texts, Cultures, 6(1), 119-138.

Thomes, T. (2013). An economic analysis of online streaming music services. Information

Economics and Policy, 25(2), 81-91.

Waldfogel, J. (2010). Music file sharing and sales displacement in the iTunes era. Information

Economics and Policy, 22(4), 306-314.

Wayne, M. (2017). Netflix, Amazon, and branded television content in subscription video on- demand portals. Media, Culture & Society, 016344371773611.

(23)

Technological Context: Putting the Influence of Ethics in Perspective. Journal of Business

Ethics, 124(4), 537-550.

Wlömert, & Papies. (2016). On-demand streaming services and music industry revenues — Insights from Spotify's market entry. International Journal of Research in Marketing, 33(2), 314-327.

Subscription services in the music industry : when do they result in the increased profits of the company?