網路免費行銷之系統動態模擬－以台灣流行音樂產業為例

全文

(1)國立高雄大學亞太工商管理學系碩士班碩士論文. 網路免費行銷之系統動態模擬－以台灣流行音樂產業為例 Using System Dynamics on the Internet Free Concept Marketing: An Example of Pop Music Industry in Taiwan. 研究生：許峻源撰指導教授：李亭林博士. 中華民國一百年七月 I .

(2) 謝誌碩士的兩年時光過得特別的快、特別的充實，感覺把多年才能完成的事情在兩年的時間裡完成，所學的知識、所遇到的人、所完成的成績單，現在回想起來特別的懷念。感謝老爸和老媽營造了這個優良的環境，讓我可以自由地嘗試我所感興趣的東西。有太多的人要感謝，沒寫到的就讓我用行動來感謝吧。首先先感謝指導我的李亭林老師，總時能提供我需多的意見，幫助我繼續在各個領域學習，兩位口委蕭乃沂老師、王俊賢老師的意見，也使這本論文更加的完善。還有大學時期的專題指導李博志老師在碩士時期的關心並給我許多建議。另外，劉信賢老師、郭英峰老師、楊書成老師等許多的老師們，感謝老師們提供給我許多寶貴的意見，不論是在人生或是在這本論文上，非常的感謝您。 EMBA 學長也給予我許多的幫助，吳文進大哥、王家盛大哥讓我更加了解不同領域與不同產業的狀況。特別感謝遠在越南的萬尚杰大哥，還有 Eddie 哥與陳哥，在越南實習期間所經歷的事情與大哥們的教導使我學習到許多課業外的知識，真的非常感謝您們的照顧，還有聖權還陪我一起去越南流浪。在研究所時期，特別感謝曾莉媖大姊，沒有她我的研究所生涯所學到、看到的領域不會那麼寬廣，雖然無法一同從亞太系畢業是個遺憾，希望之後能再見面。感謝我的室友建臺、聖權陪伴我最辛苦的一年，忍受我的白目跟任性，在家還可以有一起討論的對象真的非常的幸福啊！感謝同門的淑涓和同班六年的郁倫，一同經歷因為論文哀嚎的日子。感謝丁丁大大讓我在論文寫不出來的時候一直鬧他，高勇思可以讓我欺負，三國團隊真的是一群無聊的人，燿名、阿璊、梅君、阿謀、阿胖等學弟妹讓研究室更加有活力。在與各位同學們經歷這兩年研究所的競爭合作下，在此感謝大家的照顧與恭喜大家都順利畢業了。許峻源謹誌於國立高雄大學中華民國一百年七月 I .

(3) 摘要網路的出現改變了過去市場銷售的習性，行銷經理在面對環境改變也應做出適當的決策改變。免費，一直以來便是一種好用的銷售手段，他可以快速的吸引消費者的注意，並使產品快速的進入市場。在網路世界中，豐富的數位資源使每個產品的邊際成本趨近於零，使免費行銷可以真正的應用。當然，最終還是有人需為免費買單。本研究先以文獻探討證實免費行銷的效用，之後探討現今音樂產業面對數位科技興起所面臨的改變，並對該產業環境狀況做出改善建議，最終以系統動態學進行該產業環境模型建制與模擬分析導入改善後之結果。. 本研究因果關係圖顯示，盜版並不會造成音樂產業收益下降，反而會使其收益增加，而網路外部性的效果才是使音樂產業收益減少的主要原因。此外，系統動態學的模擬結果顯示，免費運用在音樂產業上是可行的，另外若免費吸引到更多的音樂愛好者或是使售票演場會舉辦場次增加的狀況下，免費更有實際運用的可行性。以此結果撰寫本研究發現與建議。. 關鍵字：免費、系統動態學、網路、音樂產業. II .

(4) Abstract Internet has changed the habits of the past consuming. In a changing environmental, marketing manager should also be changed to make an appropriate strategy. “Free” has been a useful marketing tool. It can quickly acquire the attention of consumers, and make new products entering market rapidly. In the world of internet, an abundance of digital resources make the marginal cost of each product close to zero, and the free concept marketing can implement really. But someone has to pay the bill to the end. First, this research confirms the utility of free by literature review. Second, it will explore the changes in the development of digital technology and conduct the free concept to make the improvement suggestion in Taiwan music industry. Finally, based on the music industrial environment, a system dynamics model will be constructed and the free concept will be inducted by way of simulation.. The causal loop shows that piracy will not decrease music industry revenue, but will increase revenue. And the “Network Externality” is the major factor to make music industry loss. Furthermore, the result of simulation shows “Free” concept practices in music industry is feasible. And if “Free” can make more population to listen to music or more booking concerts, “Free” has more feasibility to actual practice. The results above, the author wrote the finds and suggestions.. Keyword: Free, System Dynamics, Internet, Music Industry.. III .

(5) Index 謝誌 ............................................................................................................................... I 摘要 .............................................................................................................................. II Abstract...................................................................................................................... III Index ........................................................................................................................... IV Figure ........................................................................................................................ VII Table ........................................................................................................................... IX Chapter One Introduction .......................................................................................... 1 1.1 Research Background ...................................................................................................... 1 1.2 Research Motivation........................................................................................................ 2 1.3 Research Objects and Research Questions ...................................................................... 4 1.4 Research Process ............................................................................................................. 4. Chapter Two Literature Review ................................................................................ 7 2.1 The theoretical background of “Free” concept ................................................................ 7 2.1.1 The stimulus of Price .............................................................................................. 10 2.1.2 Theories of consuming risk .................................................................................... 11 2.1.3 Remarks .................................................................................................................. 14 2.2 Introduction of music industry ...................................................................................... 14 2.2.1 The challenges of music industry ........................................................................... 14 2.2.2 Boundary of music industry ................................................................................... 18 2.3 E-commerce in Music Industry ..................................................................................... 23. Chapter Three Research methodology .................................................................... 28 3.1 Concept Research Structure .......................................................................................... 28 3.2 Systems dynamics ......................................................................................................... 29 IV .

(6) 3.2.1 The Basic Theory of System Dynamics ................................................................. 30 3.2.2 The Modeling Process of System Dynamic ........................................................... 35 3.3 Problem Articulation (boundary selection) ................................................................... 37 3.3.1 Marketing model .................................................................................................... 38 3.3.2 Pirates music........................................................................................................... 39 3.3.3 Purchase intention and behavior ............................................................................. 40 3.3.4 Revenue and cost .................................................................................................... 40 3.4 Data Collected ............................................................................................................... 41. Chapter Four Dynamic Hypothesis in System Dynamics ...................................... 43 4.1 Casual loop .................................................................................................................... 43 4.2 Formulation of dynamic hypothesis .............................................................................. 45 4.2.1 Experience population by traditional channel ........................................................ 46 4.2.2 Experience population by digital music ................................................................. 48 4.2.3 Purchase intention and behavior ............................................................................. 52 4.2.4 Accumulated revenue ............................................................................................. 54 4.2.5 Population of piracy ............................................................................................... 56 4.3 Model test and basic simulation .................................................................................... 59 4.3.1 Dimensional Consistency Test ............................................................................... 59 4.3.2 Structure-verification Test ...................................................................................... 59 4.3.3 Extreme-condition Test .......................................................................................... 60 4.3.4 Basic simulation ..................................................................................................... 61 4.4 Policy simulation - The Free Condition ........................................................................ 68 4.4.1 The effect of free .................................................................................................... 68 4.4.2 Increased booking concerts .................................................................................... 70 V .

(7) 4.4.3 Remark ................................................................................................................... 71. Chapter 5 Findings and Suggestions ........................................................................ 73 5.1 Findings ......................................................................................................................... 73 5.1.1 Digital music is good for music industry................................................................ 73 5.1.2 “Free” - the future trend in Internet ........................................................................ 74 5.1.3 “Free” use of different situation in the others industry........................................... 74 5.1.4 Music industry should “Free” in Internet ............................................................... 75 5.2 Suggestion ..................................................................................................................... 77 5.3 Limitations..................................................................................................................... 79 5.3.1 Data collection:....................................................................................................... 79 5.3.2 Model Structure: ..................................................................................................... 80 5.4 Future Research ............................................................................................................. 80. Reference .................................................................................................................... 81 APPENDIX A ............................................................................................................. 87 APPENDIX B ............................................................................................................. 90 APPENDIX C ............................................................................................................. 98. VI .

(8) Figure Figure 1-1 The Flow Chart of This Research ................................................................ 6 Figure 2-1 The model of stimulus – reaction ................................................................. 8 Figure 2-2 General model of consumer behavior .......................................................... 9 Figure 2-3 The value function of prospect theory........................................................ 13 Figure 2-4 Pop Music Record Sales............................................................................. 15 Figure 2-5 The structure of Taiwan music industry ..................................................... 19 Figure 2-6 The situation of operation in music industry.............................................. 19 Figure 2-7 Revenue structure of Audio Production Industry....................................... 21 Figure 2-8 Three income streams model of the recording industry ............................. 21 Figure 2-9 The Evolution Chart of Music Storage Media ........................................... 27 Figure 3-1 Conceptual Research Structure .................................................................. 29 Figure 3-2 Causal loop diagram notation..................................................................... 31 Figure 3-3 Four equivalent representations of stock and flow structure ..................... 32 Figure 3-4 Auxiliary variables ..................................................................................... 33 Figure 3-5 Material Delay Structure ............................................................................ 34 Figure 3-6 Information Delay Structure. The Perceived Value of The Input Adjusts to The Actual Input in Proportion to The Size of The Error in Belief. The Adjustment Time Determines How Rapidly Beliefs Respond to Error. ... 34 Figure 3-7 Modeling is embedded in the dynamics of the system .............................. 37 Figure 4-1 Casual loop diagram of this study .............................................................. 45 Figure 4-2 The model of experience population by traditional channel ...................... 48 Figure 4-3 The model of experience population by digital music ............................... 52 Figure 4-4 The model of purchase intention and behavior .......................................... 54 VII .

(9) Figure 4-5 The model of accumulated revenue ........................................................... 56 Figure 4-6 The model of population of piracy ............................................................. 57 Figure 4-7 System dynamics model ............................................................................. 58 Figure 4-8 The result of extreme-condition test........................................................... 61 Figure 4-9 The simulation result of Purchase behavior ............................................... 63 Figure 4-10 The simulation result of consumer average purchase volume ................. 63 Figure 4-11 The simulation result of music platform member .................................... 64 Figure 4-12 The simulation result of accumulated revenue......................................... 65 Figure 4-13 The simulation result of purchase intention by traditional channel ......... 65 Figure 4-14 The simulation result of purchase intention by music platform ............... 66 Figure 4-15 The simulation result of purchase intention by digital piracy .................. 67 Figure 4-16 The simulation result of population of piracy .......................................... 67 Figure 4-17 The policy simulation result of the effect of free ..................................... 70 Figure 4-18 The result of increased booking concerts ................................................. 71 Figure 5-1 Reason for purchase original album........................................................... 77. VIII .

(10) Table Table 2-1 The relevant studies of the digital music ..................................................... 16 Table 2-2 The statistic of Taiwan records sale volume ................................................ 22 Table 2-3 The percentage of copyright section ............................................................ 23 Table 2-4 The digital music platforms in Taiwan ........................................................ 25 Table 3-1 Steps of the modeling process ..................................................................... 35. IX .

(11) Chapter One Introduction. 1.1 Research Background In the past, a common marketing strategy is the promotion of products through discount to attract consumers. “Free” offers can attract consumer’s attention effectively, but it was not truly “Free” (Anderson, 2009). Only when the “Internet become a major new medium can businesses offer truly free promotions” made the “Free” truly. The resources are scarce and limited in the real world, but not on the Internet. Digital products have very little manufacturing and delivery cost. The statistics of Taiwan Network Information Center (TWNIC) showed that over half population of Taiwan have used ADSL in 2010, at 13,590,123. This finding showed that people are inseparable with the Internet, and increasing Internet usage is the trend of the future in Taiwan. By the way, the age level of album buyers are between age 15 to 25 (Deng, 2000), that have the consistence with the master age level of network user in Taiwan.. One of the most important theories in computer science is the Moore’s Law (1965). It states that the amount of computer processing power for given cost doubles every two years. But impressively, digital storage technologies and bandwidth costs are improving at an even faster rate. This Law highlights the exponential pace of technological development. Even if the growth of technological explosion slows, it is understood that the cost of manufacturing technological products will continue to decrease in the foreseeable future.. 1 .

(12) The development of computer’s calculating speed, cost of storage (server), and bandwidth of the Internet are the primary factors that drive vigorous development. People are glad to share the information and resources with other people in the network. After all, the Internet culture was built on norms of collaboration, cooperation, and a willingness to share resources when others requested them (Rheingold, 2000). And the most important thing is: it’s almost free to use the Internet. People are willing to share information or files with strangers around the world. This selfless act has made many industries suffer. And people also think that everything on the Internet can be obtained for free. For example, in the Taiwanese music industry, the number of original albums sales continued to decline since 1998. Consumers would calculate utility and cost between buying legal copies or pirating, before they make a decision (Rochelandet & Le Guel, 2005). Of course, in the instance above, if consumers download music from a legitimate website, the music industry would not suffer the destructive impact.. 1.2 Research Motivation Music industry has a large impact on the overall economy. The merging of digital music changed traditional music industry sales channels but over 90% was illegal (Shau, 2009). Music industry faces an unfair competition that was confirmed by numerous studies. The concept of “Free” is the reason that drives people to illegally download. In 2010, new surveys confirmed this trend in Sweden (GfK), Australia (CCi Digital Futures), the UK (Harris Interactive) and China (The Nielsen Company) (IFPI, 2011). A multitude of industry reports postulate a cannibalization effect of file sharing on music industry revenues, but the results of academic studies 2 .

(13) are inconclusive. Deng (2000) discovered that consumers who visited MP3 website tends to have the positive impact on the behavior of albums purchasing. And some researchers agreed that file sharing does not damage the music industry and provide a channel to experience new albums (Oberholzer-Gee & Strumpf, 2007). In the past of digital music relevant research, there is no definitive result to explain the current situation of music industry. And most of the research is quantitative analysis, which analyze music industry by collecting secondary data or questionnaires. And no scholars consider the whole music industry to research the relationship factors among music industry. So, this article entered the factors to system dynamics model by system thinking. Expect the results can explain the impact of digital music in music industry.. As mentioned previously, the development of digital technology made the “Free” concept true. Not only that, it even made “Free” the only natural way for some people to obtain products. According to the theory of path independency, when the event was on the way, it is often impossible to change the results that occur in such a situation (Sterman, 2000). If we want to change this situation, compulsory legislation to forbid unlicensed file sharing is necessary. Music companies actively liberated the copyright of music on the Internet fare better. In real life, resources are limited, but resources are infinite in the virtual world. People trade scarcity thinking for abundance thinking. Maybe that can provide music industry a method to break through the difficult position.. 3 .

(14) 1.3 Research Objects and Research Questions The main object of the research presented in this article is to examine and simulate the concept of “Free.” The second object is to confirm in the usability and validity of the Internet marketing of “Free,” which could shed some light for Taiwan music industry. in light of these concerns, this article has three purpose: first, this research is going to examine of the “Free” concept has the enormous influence on the consumer behavior reported in the literature. Second, the author establishes a dynamic model based on the present status of music industry by using system dynamics. Then this study is going to simulate a “Free” condition and test whether it can change the current situation. Finally, to the findings will be applied to other industries.. Three points above seem to be helpful in attempting to explicate the purpose. The research questions include:. 1. What are the challenges in music industry in recent years? How to break the difficult position? Is the concept of “free” a strong stimulus which leads to behavior changing of consumers? How was the concept of “Free” implemented in marketing? 2. Can the concept of “Free” be modeled in music industry? What are the key factors that drive music industry to develop? Can the concept of “Free” be applied to another industry?. 1.4 Research Process For these objectives to be achieved and the research questions answered, this 4 .

(15) study is structured as follows (Figure 1-1). The first section makes sure these questions by definition meet the objects of this research. The second section deals with the theoretical foundations for the development of the research then determine the operation definition in this article. The third section is the data collection from literature review and related reports. A music industry dynamics models will be constructed, and after that, simulation and analysis are going to be conducted.. 5 .

(16) Figure 1-1 The Flow Chart of This Research Source: This Study. 6 .

(17) Chapter Two Literature Review. 2.1 The theoretical background of “Free” concept Psychology is the subject that researches and explains human’s mental processes and behaviors (United-States-Department-of-Labor, 2009). For long, Psychologists attempt to understand the role of mental functions in individual and social behavior, while also exploring the physiological and neurological processes that underlie certain functions and behaviors.. In 1913, John Broadus Watson the psychologist in American published in Psychological Review the article "Psychology as the Behaviorist Views It,” that is often called the manifesto of the behaviorist movement. Psychology as the behaviorist views it is a purely objective experimental branch of natural science. Its theoretical goal is the prediction and control of behavior (Watson, 1913). Behavioristic psychology has as its goal to be able, given the stimulus, to predict the response, or seeing the reaction to identify the stimulus that called out the reaction. And four claims have been proposed by Behaviorism: (1) The researchers of natural science is able to objectively observe and measure human’s external behavior. (2) The basis of the individual response is constituted Behavior, set the number of responses to known acts of the whole behavior. (3) Individual behavior is not innate or determined by heredity, that is affected to learning by environmental factors changing. (4) Through experimental research on animals and children, the principles of behavior can explain generally behavior of human (Watson, 1997).. 7 .

(18) Figure 2-1 The model of stimulus – reaction Souse: Waston (1913). Consumer behavior official became an academic subject in 1960. It blends elements from psychology, sociology, social anthropology and economics. It attempts to understand the buyer decision making process, both individually and in groups. The model in Figure 2-2 cleanly represents the variables and relationships that can affect consumer behavior (Mullen & Johnson, 1990). And it also presents the relationship with behaviorism psychology, cleanly.. After the subject of consumer behavior emerged, a new subject in explaining human behavior being announced in 1994, that been named behavioral economics. Behavioral economics is the combination of psychology and economics that investigates what happens in markets in which some of the agents display human limitations and complications (Mullainathan, 2000). Those are much similarity between two subjects. In general terms, researchers view or model an individual as a collection of decision rules (rules that dictate the action to be taken in given situations) and a set of preferences used to evaluate the outcomes arising from particular situation-action combinations (Lucas Jr, 1986). What different between two subject?. The nature of consumer behavior, focusing on when, why, how, and where people do or do not buy products (Sandhusen, 2000). It focus on study “consumers” 8 .

(19) mental thought. Behavioral economics doesn’t be targeted any particular groups. It research generally human behavior in specific situation. It challenges the traditional economic theory assumes rational behavior. Most of people make decisions should under the condition of irrational or limited rational (Ariely & Jones, 2008; Becker, 1962).. Figure 2-2 General model of consumer behavior Source: Mullen and Johnson (1990). It should be concluded, from what has been said above, that people suffer the stimulus of external or internal environment, and then change people’s behavior response. Of course, the conclusion can apply in consumer’s behavior. This research takes the risk of consuming and the price the external stimulation to discuss “free” concept. Confirms “free” concept can affect the consumer behavior effectively, and gets the fast dissemination the advertisement information.. 9 .

(20) 2.1.1 The stimulus of Price Although most of consumers don’t know how much about retail prices (Grocer, 1964), the consumer decision-making mostly is influenced by the price information. The price is a kind of stimulation, will cause the different response. For example, retail advertiser often presents some comparative price information, that have many form: “was $X, now $Y,” “list price $X, our price $Y,” “save $X,” “save X%.” The comparative information presents or implies a reference price against which the offering price can be judged (Biswas & Blair, 1991). Consumer can fast make the response to the price change.. Promotion often offers the user a discount on the purchase of a product. A coupon is a commonly used sales promotion device. The past studied have proved price promotions have a positive attraction effect on store performance and indicate price promotion can improve performance by attracting more people in the front traffic to enter the store and converting more visitors in shoppers (Lam, Vandenbosch, Hulland, & Pearce, 2001). The past studied also have proved a coupon can tempt the consumer to change purchase the discount brand (Cotton & Babb, 1978; Dodson, Tybout, & Sternthal, 1978; Shoemaker & Shoaf, 1977).. Promotion scope can enhance the promotion effective. The higher the coupon value, the higher the incentive to purchase (Bawa & Shoemaker, 1987). And when alternate sources of price information were present did deep discounts lead to higher purchase intentions (Raghubir, 2004). Lam, et al. (2001) also discovered greater promotion scope increases not only the attraction effect, but the unit- and dollar-spending effects of price promotion. Judging from the above, it can be noted 10 .

(21) that more discount scope, the product price is lower, and attract more people to purchases. And what kind of price promotion scope can deeper than sell product by free?. Ariely & Jones (2008) took one experiment, studied human behavior when they were facing “free.” He set up a table at a large public building and offered two kinds of chocolates – high-quality and general chocolates. And had a large sign above the table that read, “One chocolate per customer.” At first, he set the price of a high-quality chocolate at 15 cents (almost the half of market price) and a general chocolate at one cent. That could be expected consumers tended to choice the high-quality chocolate. About 73 percent of them chose the high-quality chocolate and 27 percent chose the general. After that, he changes the situation to see how “free”. He lowered the price of both kinds of chocolate by one cent, the high-quality chocolate at 14 cents and the general chocolate free. The result has the huge change, about 69 percent of consumers chose the “free.” What is it about “free?” Most transactions have an upside and a downside, but when something is “free!” we forget the downside (Ariely & Jones, 2008).. 2.1.2 Theories of consuming risk Consider the example quoted above. Why consumers prefer to choice the “free” goods and forget the high-quality goods? Because people have the loss aversion behavior, they afraid to make a wrong decision then cause the loss that is the risk consumers have to suffer. And consumer loss aversion behavior is the powerful reason of the attractive of “free.” When people choice the free goods, there is no visible 11 .

(22) possibility of loss.. Over the past years a considerable number of studies have been made on risk (Deering & Jacoby, 1972; Dowling, 1986; Edwards, 1962; Kahneman & Tversky, 1979; Sinn, 1985; Taylor, 1974). The concept of risk is one of the most pervasive in the theories of human choice (Dowling, 1986). Taylor (1986) attempted to construct a comprehensive and operational theory of risk taking in consumer behavior. He felt the outcome of a choice can only be known in the future, the consumer is forced to deal with uncertainty, or risk.. 1. Prospect Theory:. Kahneman & Tversky (1979) has published a model that presents a critique of expected utility theory as a descriptive model of decision making under risk, and develops an alternative model, called prospect theory. The result of the research was that: risk seeking is observed for losses while risk aversion is observed for gains, and outcomes received with certainty are overweighed relative to uncertain outcomes (Thaler, 1980). Figure 2-3 is the other contribution of this research shows the value function and as the s-shaped asymmetry implies, given the same variation in absolute value, there is a bigger impact of losses than of gains (loss aversion). It means the aggravation that one experiences in losing a sum of money appears to be greater than the pleasure associated with gaining the same amount (Kahneman & Tversky, 1979).. 12 .

(23) Figure 2-3 The value function of prospect theory Source: Kahneman & Tversky (1979). 2. Mental Accounting. Thalar (1985) developed a new model of consumer behavior, called “mental accounting”. The development of the model starts with the mental coding of combinations of gains and losses using the prospect theory value function. Summarize the main points that have been made in mental accounting, it researched and evidenced consumers mentally track the costs and benefits of a transaction (Prelec & Loewenstein, 1998). When consumers would like to purchase something, they open a “mental account” and start to subjectively frame a transaction in their mind will determine the utility they receive or expert. According to prospect theory, consumers have loss aversion, they more sensitivity in loss than gain. So it would be possible to agree that consumers glad to close the mental account and take the enjoyment in the “free” situation.. 13 .

(24) 2.1.3 Remarks From this section has been said above, that “free” is a strong stimulus both in price or risk that the information of consumers suffered. Therefore, it can reasonable to conclude “free” has its formidable attraction regarding the consumer, and may cause the consumer to make the response. It also explains “free” can be the strategy for the enterprise in the emerging marketing place – internet.. 2.2 Introduction of music industry Building on the foundations laid by the information and technology industry, the cultural and creative industries (CCI) have become seen as the “fourth wave” of economic progress in recent years. In 2007, the Government Information Office (GIO) commissioned the Taiwan Institute of Economic Research (TIER) to conduct the Audio Production Industry Survey. In addition to describing structural changes affecting the industry, the report began to outline the industry chain. In 2009, the GIO, furthermore, extended the research range and scope, published “Survey of Taiwan’s Music Industry.” As a result of the survey, enterprises are able to apply new modes of thinking under the twin pressures of globalization and digitalization.. 2.2.1 The challenges of music industry The record product sales volume slid largely since 1997, many people blamed on the technology of P2P file-sharing and digital music files (mostly in MP3 format)(see Figure 2-4). File-sharing technologies pioneered by Napster and for a while dominated by Kazaa have become popular among certain online communities and 14 .

(25) target for legal prosecution by record companies(Peitz & Waelbroeck, 2006). 14 12. NT$ (billion). 10 8 Sales 6 4 2 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Year. Figure 2-4 Pop Music Record Sales Source: (TIER, 2010). The internet provides a number of different opportunities for publishing music to the public. Although, that almost is illegal and hard to control the quality of music file and the purpose of downloader, but lots of Internet site offering free downloads of MP3 sites from strangers may be doing so in order to promote the artists’ CDs (Throsby, 2002). That also provides a means of cutting costs in the propaganda of traditional formats and replacing it with online broadcast with all information about the singer. Further, that has online retailing with customized delivery to consumer, if consumer would collect the music.. Digital music has also lead to new business models. Digitalized music files can be reproduced, stored, and transferred at almost zero-cost (Chen, et al., 2009). Whatever that is opportunity or threat for music companies. That is the issue they have to discuss. Many scholars published a lot of research in the past. The related. literatures about digital music are integrated as Table 2-1. 15 .

(26) Table 2-1 The relevant studies of the digital music Scholars. Title. Conclusions. Chen, Shang, & The intention to download music files in a Lin (2009) P2P environment: consumption value, fashion, and ethical decision perspectives.. 1. Found that moral reasoning moderate the relationships among fashion involvement, consumption value, and behavioral intention to download music. 2. Consumers are essentially value maximizes when choosing between buying and downloading music.. Liebowitz (2005). Pitfalls in measuring the impact of file-sharing on the sound recording market. Posited some explanations for the variations in results and conclude that the evidence strongly suggests that file- sharing harms the sound recording industry.. Peitz & Waelbroeck (2004). The effect of internet piracy on CD sales: cross-section evidence. 1. Internet piracy has played a significant role in the CD market in 2001. Different factors can explain the current downward trend in CD sales. 2. Labels may be able to save on marketing and promotion costs, by letting consumers search for their most preferred music.. Zentner (2005). File Sharing and International Sales of Copyrighted Music: An Empirical Analysis with a Panel of Countries. Find some evidence that file sharing may explain a change in the composition of sales by types of repertoires, with a higher reduction of sales of types of music that are being shared more heavily.. Bounie, Bourreau, & Waelbroeck (2005). Pirates or Explorers? Analysis of Music Consumption in French Graduate Schools. Finds that there are two populations: explorers and pirates. 1. Explorers: MP3 sharing through physical contacts has a positive impact on CD consumption. 2. Pirates: anonymous file-sharing on the internet and a large number of MP3 files have a negative effect on CD consumption. 16 . .

(27) Deng (2000). Gopal & Sanders (2006). Effect of MP3 Use on CD Consumption Behavior. There are relevant between consumers visited MP3 website and purchase behavior. The frequency of visiting website is positively to affect the frequency and quantity of purchase behavior.. Do Artists Benefit from Online Music Sharing?. 1. File-sharing technologies erode the superstar phenomenon widely prevalent in the music business. 2. Put sample music online that decreasing sampling costs not only lead more potential consumers to sample unknown music items but also lead more consumers to buy the music items that they have sampled. 3. File-sharing has a positive effect on their purchasing intentions. 4. For higher valued songs, online search and sampling capabilities have a beneficial impact on sales. Lower valued music items are pirated more than higher valued items, ceteris paribus, and consequently sales of those suffer. 5. The economic implications of this analysis are generally applicable to other similar experience goods.. Oberholzer-Gee The effect of file sharing on record sales& Strumpf An empirical analysis (2007). 1. Found that file sharing has no statistically significant effect on purchases of the average album, and rejected the hypothesis that file sharing is responsible for the majority of lost sales. 2. If we are correct in arguing that downloading has had little effect on the production of music and ignoring possible price effects for complements such as concerts. Source: This study 17 .

(28) As the Table 2-1 above, that showed that opinions are divided among scholars on this subject. Some studies show a negative effect of file-sharing on music sales (Chen, Shang, & Lin, 2009; Liebowitz, 2005; Peitz & Waelbroeck, 2004; Zentner, 2005), whereas others find a positive effect (Bounie, Bourreau, & Waelbroeck, 2005; Deng, 2000; Gopal & Sanders, 2006; Oberholzer-Gee & Strumpf, 2007).. That showed two reasons why the author research for the digital music. The first, recent empirical analyses indicated to ambiguous result. The author constructed a virtual model to simulate the real life by system dynamics. The ultimate object is to find an answer that can explain the ambiguous result. The second, all the studies above admitted file-sharing can accelerate the diffusion of music that is consistent with the “Free” concept.. 2.2.2 Boundary of music industry Although the business model that has music companies at the music industry’s core is failing. They still can play a key role in collaboration with other industries and investing in talented musicians. Figure 2-5 shows the structure of Taiwan music industry. And Figure 2-6 shows the situation of operation in music industry, it shows the main business activities. Of course, those are the main revenue source.. 18 .

(29) Figure 2-5 The structure of Taiwan music industry Source: (TIER, 2010). Figure 2-6 The situation of operation in music industry Source: (TIER, 2010). 19 .

(30) In the economic, music belongs to the empirical product, it means consumer recognizes the value of products after they actual experience. When new product to publish, the music company would conduct lots of propaganda activity for the product popularity and catching consumer’s eye. Unfortunately, after the period of propaganda (one to three months), the memory of product gradually disappear caused the rapidly information delivery. And the music company starts the next product propaganda plan. Nevertheless, the music is not the end in the period of propaganda elapsing. Each album, even every song has the application intellectual property right, and the period of music right is 50 years. In music industry, hence, only 20 percent of all acts on a label will break even, about 80 percent will lose money, while only the top 5 percent will make enough income through sales to support the entire label(Cusic, 1996). If establish a platform, that make music can extend rapidly and low-cost. That is good for reducing the cost of music company, after all, the marketing and music production are occupy high percent of operating expenses.. The past survey shows the revenue structure of music industry (Figure 2-7). This industry generated revenue structure in 2008, records (albums) revenue were still the main source (49.93%). The rest of items of annual revenue were 12.78% for copying right, 8.84% for management of artists, 8.32% for authorizing use of KTV, and 2.31% for live entertainment. On the other hand, the five items occupy the 82.18% revenue, except management of artist doesn’t have direct relationship with consumer and copying right and authorizing use of KTV should below the same section “Copyright,” so we can define the three sections (records sale, copyright and live entertainment) are the most valuable sections in music industry. Hull (1998) took a similar view (Figure 2-8). American pop music industry is segmented three sections: 20 .

(31) Live appearances (live entertainment), Recordings and Songwriters (Copyright). From the above, this article will focus the three sections to research.. Figure 2-7 Revenue structure of Audio Production Industry Source: (TIER, 2008). Figure 2-8 Three income streams model of the recording industry Source: Lam, et al. (2001, p24). 21 .

(32) 1. Music Record. In Taiwan music industry, there was a rapid drop whether the records sale volume or revenue from 1997 (see Table 2-2). Although, it is weak and drop largely, it still occupies revenue nearly 50 percent. Table 2-2 The statistic of Taiwan records sale volume Years. Sale Revenue. Grow Rate. Sale Volume. Grow Rate. 1997. 12,333.3. 1%. 47.6. 8%. 1998. 10,688.0. -17%. 39.6. -13%. 1999. 9,886.7. -12%. 35.1. -7%. 2000. 7,493.7. -24%. 26.8. -24%. 2001. 5,775.7. -32%. 18.3. -23%. 2002. 4,977.7. -7%. 17. -14%. 2003. 4,487.1. -10%. 15.3. -10%. 2004. 4,455.1. -1%. 15. -2%. 2005. 3,208.7. -28%. 10.9. -8%. 2006. 2,126.6. -34%. 7. -36%. 2007. 1,962.0. -8%. 6.5. -7%. 2008. 1,329.9. -32.2%. 5.2. -21.2%. 2009. 1,296.0. -3.0%. 5.0. -3.0%. Source: (TIER, 2008). 2. Copyright. So long as is used in “the public broadcast performance” or “each kind of sale behavior,” both of activities concern with the copyright, after audio products are published. Copyright is the second source of revenue in record company. Table 2-3 shows the percentage of copyright section. Digital music and KTV are main revenue origin. 22 .

(33) Table 2-3 The percentage of copyright section Digital Music Authorizing use of KTV Percentage. 35.92%. 43.76%. Others 20.23%. Source: (TIER, 2008). 3. Live Entertainment. The main sources of revenue today for the pop music industry are the “performance” and “talent agent” segments (see Figure 2-6). Both concerts and other commercial performances (such as Lunar New Year banquet performances, New Year activities, and music festivals, as well as artists’ are being featured in advertising campaigns) have gradually become an important source of revenue. More than 80 concerts were held in Taiwan in 2008. (This figure covers mainly those of artists performing in a Sinitic language.) Over 375,000 people attended these shows, and ticket sales garnered NT$830 million. Despite difficulties in ascertaining the true size of a market, the report of GIO 2010 suggested that at least twice that amount remains untapped (TIER, 2010).. 2.3 E-commerce in Music Industry For digital music service, recording companies almost had their official website (59.46%), but few of them provided digital music service to visitors (TIER, 2008). In Taiwan, digital music companies are independent operation, and their E-commerce business can be segmented three sections: Online music, Mobile music, and Others (Table 2-3). 23 .

(34) (1) Online music. There are 7 digital platform companies in Taiwan (TIER, 2008), the top three platform have the most member population are KKBOX, ezPeer+, and KURO. But after the lawsuit between IFPI and KURO in 2006, KURO was no longer a major player in music platform.. In music platform there are two types of online music platform operation ways, the first type is “download” by online P2P file sharing. Consumers search the music they want to download the file into their computer from the platform. The strength is consumers can own the song forever in their computer or other media (it can’t broadcast on public). And the weakness is you have to wait for the download time, it’s hard to listen to music immediately, and even you hold the best bandwidth (ADSL).. The other type is “streaming” that is use Client-Server architecture to set up platform. In the “streaming” pattern, consumers only listen to music online, and it can’t own the music file. If consumers can’t line to internet, it can’t enter the platform and choice the music they want. The strength of this pattern is it unnecessary to use the spare of consumers’ hard discs and also pretest music companies’ ownership of the file. And the weakness is consumers have to hold ADSL, if they want to listen to music smoothly.. (2) Mobile music. There are 5 companies offering mobile music download service in Taiwan (TIER, 2008). They make consumers can download their favor music anywhere by 24 .

(35) mobile phone, and charge NT$15 to 30 per file. But mobile music has not developed complete, so this article won’t further investigate.. (3) Others. It offers philharmonic a platform for original sound (music) and independent music publishing and selling. The most important is all the music doesn’t apply the protection of intellectual property right. It means visitors can download without any burden.. Table 2-4 The digital music platforms in Taiwan Name. Website. Start The number Business of songs. Members/ Payment members. Charge. Content. Online Music KKBOX. http://www.kkbox.com.tw/fu nky/index.html. EZPEER http://web.ezpeer.com/ + http://ecd.kuro.com.tw/exec/i KURO ndex.php Hinet http://himusic.hinet.net/himu music sic/index.jsp http://www.inmusic.com.tw/i iNmusic n/Default.aspx iKala. http://www.iKala.TV. RealMa http://www.realma.com/. 4,500,000/ 240,000 2000,000/ 200,000. 2004. 2,000,000. 2006. 1,000,000. 2006. 1,000,000. 500,000/NA. 2007. 500,000. NA. 2008. 100,000. NA. 2007. NA. NA. 2004. 600,000. NA. NT149 / month NT149 / month Single charge NT149 / month NT149 / month NT 99 / month free. Pop music Pop music Pop music Pop music Pop music Pop music Pop music. Mobile Music Hitto. http://www.hitto.com.tw/. iNDIEV http://www.indievox.com/ OX YOYOR http://www.yoyorock.com.tw OCK /html-index.jsp http://www.e7play.com/v2/d e7play efault.aspx http://www.starring.com.tw/s Star ring tarring/index.php. 2001. 5,852. NA. 2008. 7,000. NA. 2001. NA. 250,000/ 8,000. 2001. NA. NA. 2001. NA. NA. 25 . NT15~30 / single NT15~20 /single NT69~29 9/month NT15~30 /single NT15~30 /single. Pop music Indie Pop music Pop music Pop music.

(36) Others eTunes http://www.etunes.com.tw/ Lagio. A platform for digital original sounds (music) publishing.. http://www.lagio.cc/user/expl A community offers a platform for music writers ore interact.. Original sound/ Indie Original sound. Source: (TIER, 2008). When Yahoo was keeping to profit from charging premium E-mail version in 2004. Google announced a new Web mail service, called Gmail, and it still inherited google’s traditional character “free.” In that time, Yahoo was the largest Web mail provider (it still is now), it with around 125 million users. Most people used the free version, which only offered ten megabytes of storage. And if you want more storage spare, you could pay for the premium version that offered 25 megabytes to 100 megabytes. But while Google started the E-mail service, it offered 1000 megabytes of storage free – one hundred times what Yahoo offered (Anderson, 2009).. Although Google had no E-mail customers, so it only needed to offer few thousand customers’ storage spare that just a few servers should handle. However, Yahoo had to consider around 125 million users might have to buy a ware house of serves, that made E-mail service had a large gap between Google and Yahoo. For reasons mentioned above can reasonable explain the successful of Gmail, but the most important successful factor was Google hold the future tendency. Digital storage technologies follow the Moore’s law, even overachievement. Not only digital storage technologies but computer processing power and internet bandwidth are improving at an even faster rate, whether cost or efficiency. It causes “free” can practical application in internet.. 26 .

(37) There is absolutely no doubt that the development of technology has a deep impact to culture products. Like music player from phonograph, long-playing record, tapes, compact disc (CD) to digital file, the each change have brought a considerable impact (TIER, 2008). The music industry is also evolution along with time (Figure 2-9). From entity market to virtual market, the media for music from phonograph, tape, compact disc to computer, MP3 player and mobile phone. Music industry face this huge transformation has to make some strategy. This article would suggest a direction of development by system dynamics, help to change the industry environment.. Figure 2-9 The Evolution Chart of Music Storage Media Source: MIC (Market Intelligence & Consulting Institute), 2005. 27 .

(38) Chapter Three Research methodology This chapter represents this research’s concept structure and introduces how to establish the dynamic assumption for modeling research dynamic model construct.. 3.1 Concept Research Structure According to the research motivations, research objectives, and the literature. reviews, the author constitutes the concept research structure as shown in Figure 3-1. After understanding the music industry and the “Free” concept by literature review, the author constructed the marketing model of music industry, which includes traditional and digital promotional channels, by system dynamics. The model can show the relationship between revenue and marketing in music industry clearly. The marketing model in music industry can attract potential consumers by traditional and digital promotion, further to gain revenue. Beside, this study added “Free” concept in policy test to test whether “Free” can affect the revenue or not.. 28 .

(39) Figure 3-1 Conceptual Research Structure. 3.2 Systems dynamics The theory of System Dynamics originated in the late 1950s and early 1960s and was pioneered by J. W. Forrester of the Massachusetts Institute of Technology. His particular attention was devoted to the relationship between operations research and the developing ideas of information science, and also in the use of modeling and simulation as a new approach to industrial management and education (Lee, 2002). Forrester realized that the name of ‘Industrial Dynamics’ was inhibiting the notion that the approach could be applied to complex systems in several disciplines. The new name he applied to this discipline was thus System Dynamics. It could be very useful in describing a wide range of dynamic systems, ranging from industrial organizations to social, political, and ecological systems (Mandinach & Cline, 1994).. The author uses system dynamics to confirm whether free online music can 29 .

(40) improve the result of album sales. After all, system dynamics is good for understanding the dynamic and complicated system, and the dynamic model is used for determining the behavior character of a system. It can effectively examine models in the context of how different policies within the model change the nature of ongoing behavior (Forrester, 1980).. 3.2.1 The Basic Theory of System Dynamics System dynamics is based on the theory of nonlinear dynamics and feedback control which was developed in mathematics, physics and engineering. System dynamics also applies computer simulation to help to learn about dynamic complexity and the sources of policy resistance, and provide more effective policy suggestions (Coyle, 1977; Sterman, 2000). Through analyzing the processes of information feedback inside organizations and by computer simulation, System dynamics shows how structure, policies, decisions and time delays within organizations and business systems are interrelated and influence growth and stability. Managerial policies affect dynamic behavior, and dynamic behavior results from the way in which management responds to external variations. System dynamics is also a problem-solving approach that involves models and simulations that examine the varying relationships among the components of any dynamic process (Mandinach & Cline, 1994).. System dynamics contains many tools to help users present the boundary and causal relationships of the model, such as model boundary chart, subsystem diagrams, causal loop diagrams, stock and flow maps, and policy structure diagrams (Sterman, 2000). Causal loop diagrams and stock and flow maps are utilized frequently to 30 .

(41) capture the structure of the investigated problem. Causal loop diagrams contain the variables and arrows for presenting the causal effects among them (Sterman, 2000). Figure 3-2 is an example of causal loop diagrams.. Figure 3-2 Causal loop diagram notation Source: Sterman (2000, p.138). In a stock and flow map, there are many distinctions drawn between stocks and flows across a range of disciplines. The four kind of basic terminologies of stock and flow map: stock, flow, auxiliary and delay, that have to be known before construct model.. (1) Stock. Stocks (or levels) are quantities of material or other accumulations. They are the states of the systems. They are usually easily recognized as the results of accumulation, or of averaging. They are also the variables whose values would not drop to zero if all the flows in the system were stopped (Sterman, 2000). 31 .

(42) (2) Flow. Flows are the rates at which these system states change. They cannot be observed and their values used except by accumulation or averaging. Rates can only flow into and out of levels, except when they feed a delay which involves hidden internal levels (Sterman, 2000). Figure 3-3 shows the stock and flow diagram.. Figure 3-3 Four equivalent representations of stock and flow structure Source: Sterman (2000, p.194). (3) Auxiliaries. Auxiliaries are components of the processes by which the rates of flow are to be determined by the decision system. They can be substituted into the rate equations if we wish to do so. Thus, mathematically, there is no real difference between rates and auxiliaries, and the only advantage of auxiliaries is that they break cumbersome rate equations into simpler components and make the model easier to understand (Coyle, 1977). Figure 3-4 shows the correct and incorrect auxiliary variables in dynamic model. 32 .

(43) Figure 3-4 Auxiliary variables Source: Sterman (2000, p.205). (4) Delay. A delay is a process in which output lags behind input. When the input to a delay changes, by definition the output lags behind and continues at the old rate for some time. There are two types of delay: one is called the material delay, since it captures the physical flow of material through a delay process (see Figure 3-5). The other type of delay is an information delay; many delays represent the gradual adjustment of perceptions or beliefs. Stated clearly, there is a delay between the receipt of new information and the updating of beliefs. Though there is no physical flow of material, information delays still concern stocks. Figure 3-6 is the information delay structure, and it shows the feedback structure of adaptive expectations. In adaptive expectations the belief or perceived value of the input, Xˆ , is a stock. The rate of change in the belief is proportional to the gap between the current value of the input, X, and the perceived value. 33 .

(44) Figure 3-5 Material Delay Structure Source: Sterman (2000, p.416). Figure 3-6 Information Delay Structure. The Perceived Value of The Input Adjusts to The Actual Input in Proportion to The Size of The Error in Belief. The Adjustment Time Determines How Rapidly Beliefs Respond to Error. Source: Sterman (2000, p.428). 34 .

(45) 3.2.2 The Modeling Process of System Dynamic For the purpose of choosing an appropriate problem, achieving a useful perspective on the problem, and succeeding in capturing the essentials of a complex and real-world phenomenon during the construction of a model, a guideline of enough information for modeling the process is quite important. Sterman (2000) suggests five very similar steps of the modeling process (see Table 3-1).. Table 3-1 Steps of the modeling process 1.. Problem Articulation(Boundary Selection) Theme selection : What is the problem? Why is it a problem? Key variables: What are the key variable and concepts we must consider? Time horizon: How far in the future should we consider? How far back in the past lie the roots of the problem? Dynamic problem definition(reference models): What is the historical behavior of the key concepts and variables? What might their behavior be in the future?. 2.. Formulation of Dynamic Hypothesis Initial hypothesis generation: What are current theories of the problematic behavior? Endogenous focus: Formulate a dynamic hypothesis that explains the dynamics as endogenous consequences of the feedback structure. Mapping: Develop maps of causal structure based on initial hypotheses, key variables, reference modes, and other available data, using tools such as. 3.. z. Model boundary diagrams,. z. Subsystem diagrams,. z. Causal loop diagrams,. z. Stock and flow maps,. z. Policy structure diagrams,. z. Other facilitation tools.. Formulation of a Simulation Model Specification of structure, decision rules. Estimation of parameters, behavioral relationships, and initial conditions. Tests for consistency with the purpose and boundary. 35 .

(46) 4.. Testing Comparison to reference models: Does the model reproduce the problem behavior adequately for your purpose? Robustness under extreme conditions: Does the model behave realistically when stressed by extreme conditions? Sensitivity: How does the model behave given uncertainty in parameters, initial conditions, model boundary, and aggregation? …Many other tests.. 5.. Policy Design and Evaluation Scenario specification: What environmental conditions might arise? Policy design: What new decision rules, strategies, and structures might be tried in the real world? How can they be represented in the model? “What if…” analysis: What are the effects of the policies? Sensitivity analysis: How robust are the policy recommendations under different scenarios and given uncertainties? Interactions of policies: Do the policies interact? Are there synergies or compensatory responses?. Source: Sterman (2000, p.86). Figure 3-7 bases on the five steps of modeling process. It shows the modeling process embedded in single- and double-feedback learning loops. Iteration can occur from any step to any other step (indicated by the interconnections in the center of the diagram). Simulation models are informed by our mental models and by information gathered from the real world. Effective modeling involves constant iteration between experiments and learning in the virtual world and in the real world.. 36 .

(47) Figure 3-7 Modeling is embedded in the dynamics of the system Source: Sterman (2000, p.88). 3.3 Problem Articulation (boundary selection) In the real world, there are many factors that will affect the results. But most of the factors only have minor effects. Therefore, it is important to find the key factors to construct a system dynamics model. To build the model, the key factors must be identified and the boundary selected first. 37 .

(48) 3.3.1 Marketing model Marketing plays an important role in the music industry. It can affect the reputation of artists, albums sales and even the music industry’s revenue. The two types of marketing models are:. 1.. Traditional channel. Music is an empirical product in economics, which means that the consumers know the value by experience (TIER, 2008). In order to attract consumers’ attention, music companies utilize many forms of media to broadcast a series of propagandas (like concert/autograph session, TV programs), with the final goal of making consumers buying the music albums. To achieve this goal, music companies must invest in advertisement in various media channels. This industry’s revenue is tied closely to its marketing efforts (TIER, 2008).. 2. Digital channel. The Internet is the newest propaganda channel. There are two types of music platforms: “download” and “streaming”. The “download” pattern gives files mobility (meaning that the users can own the files), and strength of utility gives this pattern a good market share, in both legal and illegal file-sharing platforms (Hennig-Thurau, Henning, & Sattler, 2007).. The “streaming” pattern, on the other hand, means the users can’t own the files. It can best protect the copyright of music companies. In the past, it was hard for “streaming” because Internet bandwidth was limited. Recently, with increase in 38 .

(49) average bandwidth and widespread adaptation of Internet access (even governments in many countries are developing wifi-cities), the weakness of the “streaming” pattern is decreasing. For example, in Korea, most people have a high bandwidth, and all of the digital music platforms are of the “streaming” pattern, without any download sites (Shiue, 2004).. In this study, the author assumed the online-music platforms were of the “streaming” pattern. Accept online music platform is the promotional channel. We knew there is a large part of digital music is piracy by literature review. And digital file-sharing can also accelerate the diffusion of music. The further discussion will be presented in the next section.. 3.3.2 Pirates music Piracy has been considered is the major cause for the drop in music sales (IFPI, 2011; Shang-Jyh Liu, 2009; TIER, 2008; TIER, 2010). Although this study was unable to find the related variable to prove that piracy had direct or indirect influence on albums sale by literature review, the author still put “piracy” in the system model for comparing with result of other variables and discovering their relatedness. There are two kinds of piracy: “entity” and “digital”.. Before 2006, in Taiwan, piracy was mainly composed of trading physical copies of pirated albums, and digital piracy (file-sharing, download) only occupied the minority. This “entity piracy” was the most rampant in 2001, but soon the “peer-to-peer” network applications appeared and invaded the space of entity piracy. Piracy shifted from criminal organizations which seek to make a profit to individual 39 .

(50) network sharings, which is decentralized and non-profit. Statistics indicated that of all music files shared on the Internet, only 5%-10% was authorized legitimately. The other 90-95% was “illegal piracy” (Shau, 2009).. As mentioned above, digital piracy will be the factor to accumulate consumers who have purchase intention in the dynamic model. And this study ignored the purchase intention which was accumulated by entity piracy.. 3.3.3 Purchase intention and behavior The author used the theory of reasoned action (TRA) concept to link marketing and revenue. TRA states that, before people make some behavior, they must have some intention first (Fishbein & Ajzen, 1975). This applies to both the traditional and digital promotion channels. When consumers hear the music, they will more likely develop purchase intentions, which is what music companies want.. 3.3.4 Revenue and cost In music industry’s dynamics system, the stock “accumulated revenue” represents the revenue among each variable. The source of income is composed of albums sales, music platform revenues and booking concerts. Music companies hold the main status in Taiwan music industry. And the major sources of revenue were “albums sale”, “copyright”, and “live entertainment” (Lam, et al., 2001), and most of revenue of “live entertainment” was rewarded for entertainer and marketing companies, while the other sources are music companies’ revenue.. 40 .

(51) The cost of this system included “cost of recording an album”, “staff expense”, and “marketing budget” that make up about 61% of music companies’ budgets (TIER, 2008). “Marketing budget” was the most important variable in this system, and was influenced by yearly income. To music companies, the revenue from research and development of related products of entertainers was not as important and direct than album sales. Therefore, they needed to invest in propaganda for increasing market; on the other hand, the marketing budget affected the market growth ratio (Lee, 2001).. 3.4 Data Collected For effective simulation of the dynamic model of the music industry, this study referred to many reports for building a valid model. The data was collected from “Government Information Office (GIO)”, “Taiwan Institute of Economic Research (TIER)”, “Taiwan Network Information Center (TWNIC)”, “International Federation of the Phonographic Industry (IFPI)”,. and “Recording Industry Foundation in. Taiwan (RIT)”, and related thesis review. But explicit statistical data could not be obtained for many variables (e.g. TV rating, concert/autograph session participators), so the author combined online analysis data with literature review and online survey result for reasonable estimations.. Some data in this research. was collected by online questionnaires. The survey. was collected online for a 5-week period from 13 March to 16 April, 2011. In survey, 189 people participated, and all of them completed the initial demographic and related questions. The participants lived in each region of Taiwan, 64% were 12 to 24 years old and 70% were students that were the key demographic for the music industry. 41 .

(52) These data was used to construct the system dynamics model. The detailed data distribution is in Appendix A and B.. The data in the proposed model covers the period from 2000 to 2009. Simulation time is expected to 2019. Related data collection, estimate and sources have been ordered in Appendix C.. Although not all the statistical data was explicit, it can still get the result of simulation by system dynamics. System dynamics model was based on the system structure that not only estimated the statistical and the parameter value, but also the relationship between factors. The modeling focused on understanding the interactive characteristic and the change tendency of system dynamic behavior. Therefore, whether the statistical data explicit or not was not important (Huang, 2006).. 42 .

(53) Chapter Four Dynamic Hypothesis in System Dynamics. 4.1 Casual loop Figure 4-1 shows the casual loop diagram of this study. The casual loop includes three positive and one negative feedback loops. The first positive loop at the right side showed how the music industry makes revenue. As soon as a new album published, the music company must prepare marketing budget to hold various activities. The consistent goal is to attract consumers to listen to the album. The music company increases more marketing budget, and the population of listening will also increase. After consumers listened to the new album, some of them will have the purchase intention. Furthermore, some of the people who have purchase intention will have actual purchase behavior (buy the album). The more people have ever listened to the album, the more people will actually buy the album. Music industry’s revenue will increase along with the population. Then, the revenue increases, and the music company can invest more marketing budget to promote the next albums.. The second positive loop at the left side showed the situation of music spreads, especially on the Internet. When the music starts to be published on the Internet, the Network Externality will happen. People are glad to share the information and the files with other people on the Internet. As long as more people access to the relevant information, the Network Externality effect will be more intensive. And the Network Externality will make payment users decrease. As long as the payment users decrease, the piracy users must increase. The piracy is also a promotional method to increase the population to listen to music. 43 .

(54) The third positive loop showed the situation that piracy can combine with the first positive loop to become a big positive loop. The dig positive loop showed the music industry’s revenue will increase continuously. After all, people who listen to piracy music will have the purchase intention, too.. And the negative loop showed the Network Externality directly decreases the payment users and indirectly make the music industry’s revenue decreases. That can explain the problem about the ambiguous result in the past researches. Piracy is not the factor to make the music industry’s revenue decreases; on the opposite side, that will make gains. The Network Externality is the truly factor which causes music industry’s revenue decreases.. “Network Externality effect is not shown directly in the model.” It is hidden in the section of experience of population by digital music, and the full discussion will be presented in next section. Most Internet users in the model got music by file-sharing, not music platform. That means when more people use the Internet, more people and more piracy files will be used. That is “Network Externality” effect.. 44 .

(55) Non Music Listen Network Externality. Marketing Budget +. + Payment Users. 2. + Population of listening to music (Fans) +. Music Piracy. Revenue of Music Industry + +. 1. + 3. Purchase Intention +. Purchase Behavior +. Figure 4-1 Casual loop diagram of this study Source: This study. 4.2 Formulation of dynamic hypothesis The system dynamics model was constructed by using Vensim DSS, version 5.6a. The stock and flow diagrams were used thoroughly and this study was copied directly from the Vensim model. Detailed account of the model equation is given below. In the system equation of simulation model, the level is labeled as L; the rest can be deduced in the same way with R = rate, N = initial value, A = auxiliary, C = constant, G = graphic function. Each factor will mark its unit, and some data which are assumed by the author will be marked after the units.. The model contains five major sectors: experience population by traditional 45 .

(56) channel, experience population by digital music, purchase intention and behavior, accumulated revenue and population of piracy.. 4.2.1 Experience population by traditional channel Music industry often held concert/autograph sessions and joined the TV program for promoting new published albums (Chiang, 2002). The population that participated concert/autograph session was calculated by three parameters, “session hold of an album,” “average participators” and “average published albums” (see equation 1.A). That represented the population who experienced a new album by concert/autograph session every year. The other section “population by TV channel” was calculated by Nielsen audience measurement (see equation 5.A). The formulation is as follows:. The market growth rate was limited by the marketing budget which the music company invested in pop music industry and the album propagandizing expense was between 5 million and 30 million (Chiang, 2002). However, according to another article from literature review, I found out the actual expenditure of marketing expense that only occupied total revenue 10-15% (TIER, 2008). The author assumed that if “marketing budget” was the same or higher than “average published albums” multiplied “average album propagandizes;” then the sufficient marketing budget would make the market more prosperous. If marketing budget was only 10% to 15% 46 .