1
Stock Price Response to Marketing Events in
Cosmetic Industry: An Event Study
University of Amsterdam Faculty of Economics and Business
MSc. in Business Administration – Marketing Track
Thesis Author: Linye Li (11371560) First Supervisor: Dr. J.Y. Guyt
2
Statement of originality
This document is written by Student Linye Li who declares to take full responsibility for the
contents of this document.
I declare that the text and the work presented in this document is 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
3
Table of contents
Statement of originality ... 2 Table of contents ... 3 Abstract ... 5 Acknowledgements ... 6 1. Introduction ... 7 2. Literature Review ... 112.1. Relationship between selected events and stock returns ... 12
2.1.1 New product introduction... 12
2.1.2. Collaboration ... 15
2.1.3. CSR activities ... 17
2.2. Event study ... 20
2.2.1. Stock returns as financial measurement ... 20
2.2.2. Feasibility in marketing research ... 22
3. Conceptual Model and Hypothesis ... 23
3.1. Main question ... 23
3.2. Conceptual model ... 24
3.3. Hypothesis ... 25
3.3.1. New product introduction... 25
4
3.3.3. CSR activities ... 26
4. Methodology ... 27
4.1. Prerequisites ... 27
4.2. Population identification and Data collection ... 28
4.3. Variables ... 29
4.3.1. Independent variables: Three types of events ... 29
4.3.2. Dependent variable: Stock returns ... 30
4.4. Event study ... 31
5. Result and discussion ... 33
5.1. Statistic description ... 34
5.2. Multiple regressions analyses... 36
6. Discussion ... 39
6.1. Theoretical and practical implications ... 40
6.2. Limitations and further research ... 46
7. Conclusions ... 48
References ... 50
Appendix 1: The information of selected firms ... 59
5
Abstract
Marketing activities contribute significantly to the corporate financial performance of the
cosmetic industry. Previous literature focused on the financial effects of a single event and did
not limit it to one industry. Thus, it is important to figure out the different financial effects of
three types of events on firms‘ stock returns. This research chooses new product introduction,
collaboration and corporate social responsibility related activities as research objectives and
collects more than 350 relevant events from thirty world leading cosmetic companies. In this
paper, the event study methodology is employed to get the cumulative abnormal return which
is the dependent variable in the regression models. Two regression analyses are followed to
examine whether the occurrence of the marketing events has a linear and positive relationship
with the stock returns. The revenue level of companies in the last financial year as a control
variable enters the second model in order to explain more variance of CAR. The results suggest
that the first two types of events have significant positive effects on the stock returns but the
third type of events does not have any significant effects. After adding the control variable, the
interaction will weaken the positive relationships. And the interaction between CSR-related
activities and the independent variable is significant. The findings of this paper result in good
marketing strategy choices and more rational management decisions.
Keywords: Event study, cosmetic industry, new product introduction, collaboration,
6
Acknowledgements
The master thesis is my final assignment before completing all the requirements of MSc
Business Administration Marketing track in University of Amsterdam. This thesis focuses on
the marketing activities in the cosmetic industry, which combines both my learning contents
and the past internship experience. It employs event study methodology and regression
analyses to examine the relation between selected types of marketing events and firms‘ stock
returns.
I would like to write down my gratitude to all the people who have helped and encouraged me
during the master thesis writing in this section. First and foremost, I would like to thank Jonne
Guyt, my supervisor who guided me and gave me the feedback in this tense half of year. From
the topic choosing to the model improving, Jonne inspired me for many times and made a clear
schedule of our meetings. My friends also gave me great encouragement to pull through
difficulty times when I struggled in how to balance the thesis writing and other courses. I
would like to thank everyone involved in my thesis and supported me in this process.
Linye Li
7
1. Introduction
In this highly informational age, the cosmetic industry has benefited from increasing number of
social media channels, such as Facebook and YouTube. There is no doubt that the cosmetic
industry has striking impact on the diversification of social life,
which contributes to the growth of social value. The amount of beauty-related content views on
YouTube has grown from 200 to 5000 million per month between September, 2009 and June,
2015 (see Table 1). Cosmetic products have entered consumers‘ daily lives and play an
important role in personal care consumption for the female and even the male. People keep
looking for the proper makeup and skincare goods to make themselves look flawless and the
global annual expenditures on cosmetics is estimated at 18 billion U.S. dollars (Khraim, 2011).
What is more, the revenue of cosmetic industry climbed to estimated 244.8 billion U.S. dollars
(Source: EY report, 2014). In the last decade, cosmetic industry has expanded further. This
traditional industry has entered a new era that attracts more focus and amount of investment
than ever before. Its global market grew by more than 3.5 percent year by year since 2009 (Source: L‘Oreal‘s annual report 2015). Furthermore, the value of the worldwide beauty market
8
was forecasted to increase by approximately 62.9 billion U.S. dollars from 2014 to 2019
(Source: Euromonitor, 2014), which indicates the huge profit space still exists in today‘s
hyper-competitive environment. The market potential in cosmetic industry indeed catches many researchers‘ attention. Researchers referred to that the cosmetic industry is
environmentally sensitive, which means consumers‘ purchasing intention and behaviors react
strongly and immediately for marketing events, due to lack of brand loyalty and less related
knowledge (Usman, Rida, Madiha & Mohsin, 2012). Nowadays, consumers can find
multitudes of substitutes among different brands or categories. In this way, it is difficult to
follow one certain brand and they will be affected by marketing events easily. This
representative relationship between companies and customers in cosmetic industry results in
the observable market response to various announcements (Papista & Dimitriadis, 2012). It
will be of interest and value to concentrate on the influence of marketing events on corporate
performance for managers in cosmetic industry.
Numerous previous studies pay close attention to the question that which type of events
marketers should invest more in to maximize the benefits (Smyth & Lecoeuvre, 2015; Huang,
2015; Srinivasan & Hanssens, 2009). Marketers are able to allocate budget in product-related
activities, non-product-related activities or both. Parker and Brey (2015) advocate the
importance of investment in new product development and collaboration, which affect
corporate financial return directly. Consumers regard the new products and partnerships as a
competitive advantage that expands the extent of products and services. On the other hand,
9
However, the potential influence on consumers‘ purchasing intention and investors‘ attitude
leads a better access to financial performance. Especially in cosmetic industry where substantial similar goods exist, it is difficult for enterprises to control consumers‘ buying
behaviors totally by the products themselves (Wu & Lee, 2016). Cosmetic product buyers can
be affected by the actual value and the perceived value simultaneously. In order to compare the
contribution of product-related activities and non-product-related activities to the firm‘s
financial performance, this paper selects three types of marketing events that are adopted
frequently in strategy. First one is new product introduction which is risky yet increasingly crucial to develop company‘s competitive advantages, notably the corporate growth. Because
of the explosive growth of new products launched every year, the importance of new products
introduction cannot be ignored. New product introduction is associated with the firm‘s creating
ability and manufacturing ability as the foundation to increase corporate assets and revenue. It
also can become a chance for expansion and growth of their business so as to satisfy
consumers‘ changing needs and expectations. The stock price is affected the announcements of
new products in an obvious way. For instance, Apple Inc. introduced iPhone 6 officially as an
innovative product for the first time on September 2014, which caught worldwide attention and
raised its stock price by 3.7% to 101 U.S. dollars (Hao & Wang, 2015). The second type of
events is collaboration, which includes the partnerships with other communities in product,
activities, technology or brand level. Except for boosting financial performance, the main goal
of collaboration is adding extra associations to strengthen the brand and corporate identity, further contributing to customers‘ favorable perception (Fyrberg, 2008). Moreover, in light of
10
closed relationship with their customers since the strong relationship results in their repeated
purchase behaviors (Keller, 2001). Besides, collaboration activities have a significant effect on
broadening the extent of target consumers and improving firm‘s awareness. So, its overall
impact on both tangible assets and intangible assets should cause incremental attention from
marketers and researchers. Corporate social responsibility (CSR) activity is the only financial
unrelated event in this paper which is viewed as a tool to improve reputations and emerge good
opinion among customers (Chernev & Blair, 2015). According to a survey of over 300 CFOs,
CSR experts and investment analysts, majority of respondents suggest that improving
reputation is a key component of value creation by marketing activities (Quarterly, 2009). So,
from practical viewpoint, marketers feel eager to figure out the specific association between
CSR-related activities and the corporate financial performance. In other words, it‘s important to
assess whether the investment in CSR activities has been transferred to stakeholders‘ value.
This paper will utilize stock returns as a measurement for company performance which has
been proven appropriate and efficient in previous study (Conchar, Crask & Zinkhan, 2005).
Furthermore, Markovitch and Steckel (2012) stated that the stock market‘s short-term response
can evaluate the uncertain-outcome events and financial impact on events, which also
contributes to investigating the factors of company profitability. However, the stock price
response to marketing events in cosmetic industry has drawn no attention from the existing
studies thus far. This study will examine this question in quantitative method and the following
questions can help to build a deeper view. What types of event has positive or negative
11
many relevant studies about different event types and the indicators of stock price change will
be studied. This research chooses three types of event that affect stock price in efficient way.
The main research objectives are:
a. To find out the relationship between marketing events in cosmetic industry and stock returns
b. Compare the financial impact of different types of events and explain the reason why the
difference exists
This paper focuses on the different stock response to new product introductions, collaboration
and CSR activities in cosmetic firms by event study methodology. With the purpose of
attending the research objectives, this research will be structured in seven chapters. The paper
will start from literature review that describes the existing findings including the cosmetic
industry environment analysis and the efficiency of the event study method. The research gap
among previous researches will be analyzed clearly as well. Secondly, the conceptual
framework will include the conceptual model and event definition. The hypotheses will be
introduced in the third chapter. Next, the methodology will elaborate how the data will be
collected and analyzed. To conclude, the last chapter will present a conclusion of this study
with explanation of the academic and managerial contribution and the key limitation.
2. Literature Review
This chapter reviews the previous findings concerning the three selected marketing events
12
representative types of marketing events and stock returns and the potential reasons for the
difference among existing studies. Furthermore, the feasibility of the event study method
applied in marketing research will be analyzed fully. The goal is to figure out the research gap
and the setting for this paper.
2.1. Relationship between selected events and stock returns
2.1.1 New product introduction
New product introduction plays a crucial role in retaining and strengthening its competitive
advantages of a company (Hao & Wang, 2015). Chan and IP (2011) also indicate that new
product introduction can help corporations to survive and succeed in competitive market as a
powerful weapon. Launching new products for a brand is a common but significant strategy
because it is able to boost brand development and generate revenues within a long duration
(Aaker, 1990). Chaney, Devinney and Winer (1991) demonstrate that new product introduction
is a key underlying component of expected future cash flow — in other words, it yields
revenue for stakeholders directly. Therefore, new product introduction becomes an important
issue that affects most of the stock investors (Wang, Chen & Chang, 2011).
New product introduction has been a current research hotspot over the past decade (Shane &
13
announcements of new products launching affect investment activities. In their study, this
psychological process is defined as a multi-stage ―infection‖ model. It starts from that the
signal of introducing new products is observed and perceived by potential investors and
stakeholders, which will further affect their evaluation on whether to invest or not. Then, they
make the investment decision and take actions, thereby having a significant influnce on the
stock price. Based on this theory, numerous existing literature stresses the relationship between
new product announcement and stock response. Some scholars suggest new product introductions are positively correlated to shareholders‘ value. Sood and Tellis (2009) agree with
this view and concentrate on stock response to the announcement of different stages of new
products. They believe innovation is one of the most crucial factors in influencing this positive
relationship. Woodman and Sawyer (1993) suggest the creation of new products and service
resulting from innovation adds advantages to organizational creativity. Moreover, innovative
behavior is regarded as the engine of firm development as well as growth because the new
product innovation is highly important to the firms‘ continued existence (Chaney, Devinney &
Winer, 1991). In addition, Borah and Tellis (2014) also argue that the stock market is
associated positively with the announcement of making new products independently due to
higher evaluation of corporate creative ability. In the real business environment, the
information of new product introduction is likely to be disclosed prior to the launch and under
this circumstance stock investors still keep positive reaction to new product preannouncements
(Sorescu, Shankar & Kushwaha, 2007). New product introductions linked with marketing
resources as a moderator lead to a result that new products tend to be accepted more easily
14
product announcement do not have significant impact on stakeholder value. They demonstrate
investors do not react to the announcement immediately because the overall new product
policy actually affects their investment decisions in the general evaluation process. But
limitation indeed exists in their findings. The main thing is that authors select raw data across
several industries, so the dataset choice weakens their results to a certain degree. Besides, they
examine the hypothesis by monthly rather than daily stock price data so that the effect of new
product introduction on stock price change may be more confusing with other events by
overlapping effect. Furthermore, Lee and Chen (2009) suggest investors have favorable
attitude toward companies to conduct new product launch activities in general. But they find a
U-shape relationship between new product announcement and stock returns of firms moderated
by research and development resources (R&D). When the cost of R&D does not outweigh the
expected earnings of investors, the stock price is still positively correlated with new launch
actions. Once companies make a commitment to spend plenty of money in new products which
cannot surprise investors, they possibly cut, even stop their investment behaviors. According to
the earlier literature, the complex relationship between new product introduction and stock
price change is unclear, especially in one certain industry.
Particularly, products in cosmetic industry belong to fast-moving consumer goods which are
defined as products that are sold and consumed rapidly (Dibb et al., 2006, p.298). For this
category, the products are always manufactured extensively and can be substituted simply by alternatives when some news interrupts consumers‘ mind (Van Elzakker, Zondervan, Raikar,
15
Hoogland & Grossman, 2014). So, the new product introductions attract more attention in
other industries.
2.1.2. Collaboration
To enhance the influence of products and other activities, marketers are likely to adopt a
collaboration strategy at company, brand, activity or product level. The finding of Keller (2003,
p.360) shows that co-branding called brand alliance comes up when two or more brands are
combined into a conjunct product or in other forms. It is important to note that, partnership
with other strong brands can be deemed as both product competitiveness and marketing ―creatives‖ which would utilize the public‘s growing interest in collaborative contents.
Likewise, the company can seize the opportunity to enter into new markets and share the
producing resources and related knowledge with each other (Cao & Sorescu, 2013). Scholars
have researched many levels of brand collaboration, such as joint sales, co-advertising and
promotion, and dual-branding in order to raise the core value of products or services (Kapferer,
2008). However, the effect of collaboration is still uncertain because the positive stock reaction
is not always observed. Cao and Sorescu (2013) indicate that stock investors are likely to
respond more positively to collaborated products compared to single-branded products. The main reason for this result is consumers‘ favorable attitudes towards collaborated products,
which is examined by the consumer research (Walchli, 2007). New collaborated products
convey a high-quality signal to consumer and build a positive association to customers (Rao,
16
positively to collaboration announcement but only if brands are coupled with positive customer
attitudes as well as high familiarity. Moreover, marketing alliance is a key component in
collaboration-related studies. Swaminathan and Moorman (2009) stress that marketing alliance
announcements have a positive impact on stakeholder value in terms of increasing firm value.
Several scholars figured out different reasons for firm value growth. First, firms can gain
access to new market segments through marketing alliance (Bucklin & Sengupta, 1993). For
instance, by forming an alliance with men grooming firm, a women skincare firm can cover a
completely new market segment. Second, a marketing alliance has potential to supply an
access to products, brands or services for companies (Kalaignanam, Shankar & Varadarajan,
2007). This access contributes to enriching entire offerings, which is beneficial to consumer
acquisition, satisfaction and retention. Third, a firm can benefit from building a marketing
alliance through an approach to new skills and knowledge, which helps to save the cost of
development and research (Rindfleisch & Heide, 1997). All of these merits have great potential
to increase a firm‘s future cash flow and affect investors‘ decision making directly
(Swaminathan & Moorman, 2009). Cosmetic companies generally choose to collaborate with
biological laboratory to develop new technology and products and the effect of technology
elicits the positive reaction in stock market (Kalaignanam, Shankar & Varadarajan 2007). For
horizontal collaboration, the stock market regards it as a more valuable strategy than others
(Wu, Luo, Slotegraaf & Aspara, 2015). The key reason is that horizontal collaborations create
strong advantages of absorptive ability and communication effectiveness (Sivadas & Dwyer,
2000). Gleason, Mathur and Wiggins (2003) believe horizontal collaborations lead to positive
17
partner can obtain more on late-stage codevelopment (Fang, Lee & Yang, 2015). Most of
cosmetic companies locate in late stage of the whole supply chain as manufacturers and
retailers. Das, Sen, and Sengupta (1998) make a different standpoint from above that the
marketing collaboration has no significant effect on stakeholder value, possibly because the
selection of samples includes only 18 industries instead of a specific industry. Without control
of industries, the conclusion of quantitative research can be blurry (Bass, Cattin & Wittink,
1978). Another reason emerges from the long span of samples which is over 20 years and the
contribution of collaboration to companies has changed dramatically. Wu, Luo, Slotegraaf and
Aspara (2015) argue that non-technical collaboration has a null effect on stock returns.
2.1.3. CSR activities
Corporate social responsibility (CSR) has been an increasingly crucial issue of interest for
managers and for academics in the last three decades. The definition of CSR is ―actions that
appear to further some social good, beyond the interests of the firm and that which is required by law‖ (McWilliams & Siegel, 2000). Nowadays, people don't merely lay their sight on the
product-related attributes but also the brand concept, so brand building should satisfy customers‘ functional and emotional needs. The essence of CSR is aligned with customers‘
underlying perception that society is the important playground for their daily lives as well as
business activities (Banerjee, 2007). Since companies recognized their obligation to the society,
they devote the executional ability to realize the combination of economic, altruistic and ethical
18
reports to improve information transparency in order to satisfy investors‘ and customers‘
demand (Kim, Park & Wier, 2012). Banerjee (2007) introduces the statistical proof of
responsible organization that in 2005 all the Fortune 500 companies and a significant number
of small and medium-sized enterprises (SMEs) have some kinds of statement of CSR in
different types of reports. What is also found that in 160 out of the 250 largest cross-national
companies released a CSR report (Porter & Kramer, 2006). Based on stakeholder theory,
managers cannot determine a specific cut goal that contributes to assessing company
performance, thus leading to a harder process of management evaluation and higher costs
(Jensen, 2002, p. 242). This stakeholder view assumes that firms value the stakeholders‘
assessment, where CSR plays a crucial role in managing these relations (Friede, Busch &
Bassen, 2015).
The relationship between CSR activities financial performance existing or not has become a
popular issue in numerous previous literature over a long time (Crifo & Forget, 2015).
Tsoutsoura (2004) advocates the mainstream viewpoint that the relationship is positive through
the analysis including 422 companies from 1996 to 2000 and using DSI 400 index as an
indicator. After this, Friede, Busch and Bassen (2015) also analyze 2200 different researches
covering a period from the early 1970s to today and conclude a similar result. Investment in
CSR activities are regarded as a necessary input due to low visibility according to conventional
wisdom; however, Oh, Bae, Currim, Lim, and Zhang (2016) indicate CSR strengths developed in line with stakeholders‘ high expectation of CSR strengths, which will lead to an increasing
19
measure the strength of CSR more accurately, they choose adjusted CSR score as a statistical
instrument to weaken the positive upward bias to correlation between CSR and stock returns,
but the result of a positive correlation is still not being changed. Fatemi, Fooladi and Tehranian
(2015) agree with the positive sides of CSR, meanwhile emphasizing its role of the cost, but
CSR can create advantages to companies in regards to both reputation and financial
performance in the long-term. However, Friedman (1970) values the negative sides of CSR and
suggests that companies should not use the corporate wealth to achieve their social goals since
this action will diminish stakeholder returns directly by the use of what can be essentially
regarded as their money. Thus, when a company executes some CSR decisions, it will abandon
part of the profit it makes to act in a way that is considered to be socially responsible (Bénabou
& Tirole, 2010). As such, the viewpoint of being responsible to society is supposed to be left to
the shareholders of the firm (Fatemi, Fooladi & Tehranian, 2015). However, Yuwita and
Kalanjati (2017) point out that the CSR-related activities have no significant effect on the stock
returns as this study found that CSR index fails to boost company‘s financial performance in
form of stock return in the cigarette industry from 2007 to 2014. Brine, Brown and Hackett
(2007) conduct a quantitative research in the Australian context covering 277 firms for the year
2005 but also do not find a specific correlation between CSR and the stock market. Verbeeten,
Gamerschlag, and Möller (2016) believe that CSR disclosures are value relevant, but the
relevance differs from the type of information provided. There is no association between
environmental disclosures and stock returns and the social disclosures are positively related to
stock returns. Excessive environmental disclosures are related to corporate assets and liabilities,
20
lack of association. On the other hand, the linkage between social disclosures and the higher
capability of companies‘ human capital produces a potential account for the positive
association. The result in the finding of Mishra and Modi (2016) provide evidence that CSR
will affect stakeholder value when marketing capability is noticed. According to the marketing
contribution to company relationship with the stakeholders (Kumar, Jones, Venkatesan &
Leone, 2011), marketing capability has become a crucial component in managing CSR
activities. The reason for this is that investors‘ high evaluation to firms derives from high
marketing ability. In other words, when the marketing capability is high, investors are likely to
believe that the value of CSR can be translated to stakeholders successfully, resulting in higher
stock returns.
2.2. Event study
This section will introduce the applicability of the event study methodology from the financial
field to the marketing field.
2.2.1. Stock returns as financial measurement
Scholars have considered all the time about what indicators can more accurately reflect the
effects of corporate events, thus conducting a methodology called event study, which is applied
in economics and finance studies originally (MacKinlay, 1997). Previous studies have proven
21
performance (Conchar, Crask & Zinkhan, 2005). Stock return is independent through time in
statistical model of event study (MacKinlay, 1997). The strength of stock returns is that it is
presented in a continual shape which is convenient to trace back to the historical data. The
other advantage is its immediate reactions to events at that moment (Dmitri & Joel, 2009).
Stock price also reflect the earning of investors directly and clearly. Previously, other metrics
are also regarded as an effective financial measurement of event performance. Barber and John
(1996) argue the firm‘s return on assets (ROA) well explains the operating performance of
firms, further contributing to firm‘s stock market evaluation. Higher ROA means a firm yield
greater value market-based assets (Hitt & Ireland, 1985), so investor can estimate the expected value by firm‘s previous operating performance. Compared with stock returns, ROA is just
available at low frequencies which cannot reflect the event-induced change timely. The
announcement of marketing events is instantaneous, so this delayed indicator will include other
impact of other events in a longer period. Gruca and Rego (2005) demonstrate that market
share also is an important indicator to affect the ability that the market-based assets are
translated into future cash flow. The definition of market share is the proportion of a firm‘s
sales revenue accounted for the whole industry sales profit (Tuli, Bharadwaj & Kohli, 2010).
Market share emphasizes the sales impact on firms so that it has certain limitation to perform the overall influence, including the influence on firm‘s operating ability, reputation,
relationship with customers and other aspects. To conclude, stock returns focus on the
aggregation of the impact from various aspects due to events on the complex role of the
company and has advantage to figure out the specific relationship with time based on the daily
22 2.2.2. Feasibility in marketing research
The event study methodology was created by finance researchers but in the last decades has
already been adopted extensively to several fields, including marketing (Sorescu, Warren &
Ertekin, 20017). Marketers in the past commonly studied the events issued by firms, such as
acquisition and merger but recently they perceived the importance in marketing events.
Expected for the selected events in this paper, scholars have studied a great number of events in
marketing activities. Jarrell and Peltzman (1985) point out a negative response of stock returns
to product recalls in drug and automobile companies. Horsky and Swyngedouw (1987) found a mean excess return of 0.61% for corporation‘s name change and the magnitude of the excess
return relies on the firm type. Industrial firms have a higher magnitude yet financial institutions
have a lower one. Agrawal and Kamakura (1995) indicate that stock returns react positively to
110 tested celebrity endorsement announcements because the celebrity endorsement contracts
can be realized as the investment in advertising. Geyskens, Gielens and Dekimpe (2002)
suggest that the net influence on expanding an online selling channel is positively associated
with firm‘s stock returns on average and the results are varied by channel power. Previous
literature has tested the feasibility of this financial methodology applied in the marketing field,
which is beneficial to the quantitative studies.
Although a great deal of work on the application of event study still remains in marketing
23
(Rao, Agarwal & Dahlhoff, 2004). However, Srinivasan and Bharadwaj (2004) stress that
―despite the considerable potential of the event study method to relate marketing strategy
initiatives to changes in shareholder wealth, event studies have been underutilized in marketing.‖ According to their findings, this paper aims to focus directly on investor‘s benefits
and build a guide to support markets‘ investment decisions.
3. Conceptual Model and Hypothesis
In this chapter, I elaborate on the relationship between selected marketing events and related
stock returns. I begin from the main question and the process that related to the independent
variables.
3.1. Main question
Current event study literature in the field of marketing have examined the linkage between
marketing activity announcements and stock return change of firms, but the application in
cosmetic industry has not yet entered scholars‘ vision. This research will utilize a statistic
approach by considering different events that have a significant impact on corporate financial
performance. Three selected events, new product introduction, collaboration and CSR activities,
are viewed as study objects. This paper aims to add knowledge to event study literature by
24
“Do the three types of marketing events affect stock return positively or not in cosmetic industry? – If so, which one event contributes most significantly to stakeholder value?”
3.2. Conceptual model
The main question is visualized in the following framework and it can help to understand the
researching process.
H1
H2
H3
The model shows that three selected types of marketing events are expected to have a direct
impact on the financial performance of firms which is measured by stock return. This
relationship has been proven in substantial studies (Hao, & Wang, 2015; Wu, Luo, Slotegraaf
& Aspara, 2015; McWilliams & Siegel, 2000). However, these studies concentrate on the
single type of event, thus not considering the varying degrees of influence of different events
on stock return. This model doesn‘t involve moderating effects of industry diversity and firm
size by limiting data population in the cosmetic industry and selecting data sample from
worldwide top 30 companies in sales from 2014-2016. Event study method is applied IV (i): New Product Introduction
IV (ii): Collaboration
IV (iii): CSR Activities
DV:
Stock
25
appropriately in this research to figure out the explicit relationship between independent
variables and dependent variable and the quantitative output is comparable.
3.3. Hypothesis
This chapter briefly explains and lists all the hypotheses. The more concrete introduction of the
subjects has been elaborated in the literature review.
3.3.1. New product introduction
Hao and Wang (2015) indicate new product introductions can actually add advantage to investors‘ profit by affecting the customers‘ evaluation process. Sood and Tellis (2009) regard
new product introduction as an innovation that contributes significantly to firm‘s both financial
and opreating performance. Moreover, Chan and IP (2011) suggest this marketing activity is
crucial for companies to improve competitiveness. From a brand perspective, Aaker (1990)
believe new product introduction can increase the brand value and has huge potential to be
translated into revenue in a long period. In contrast, Eddy and Saunders (1980) don‘t find any
specific correlation between new product introduction and firm‘s stock return. So, the
following hypotheses are formulated:
26 3.3.2. Collaboration
Cao and Sorescu (2013) assume that stakeholders reactive more positively to collaborated
products than single-brand products due to consumers‘ increasing trust. Walchli (2007) states
that several forms of collaboration lead to investors‘ favorable attitudes towards firms, which
encourages them to make investment decisions. Swaminathan and Moorman (2009) find the
positive response of the stock market to collaboration announcements under consumers‘
favorable attitudes as well. Previous researchers stress that the broad impact of collaboration on
firms contributes to their stock returns ((Bucklin & Sengupta, 1993; Kalaignanam, Shankar &
Varadarajan, 2007; Rindfleisch & Heide, 1997). But Das, Sen, and Sengupta (1998) indicate
there is no significant impact of collaboration on stakeholder value.
H2: Collaboration has a positive impact on the firm’s stock return.
3.3.3. CSR activities
Friede, Busch and Bassen (2015) advocate the important role of CSR activities in stakeholders‘
assessment to firm value. Tsoutsoura (2004) finds the positive relationship between CSR activities and stock price change by analyzing 422 companies‘ correlative activities.
Furthermore, based on the finding of Bae, Currim, Lim, and Zhang (2016), investors‘ confidence in firm‘s success is positively correlated to the leakage in terms of CSR information.
27
investment in social goal reduces stakeholder value and deviate firms from original profitable
goal.
H3: CSR activity has a positive impact on the firm’s stock return.
4. Methodology
Previous researchers have spent much time in understanding what events drive the company
profitability (Jensen & Ruback, 1983; Malatesta, 1983) and the stock price is an effective
measurement which reflects the change of company performance immediately (Sorescu,
Warren & Ertekin, 2017). However, the effect of events in the cosmetic industry is still
unknown with little researches. So, this paper examines which type of events has the greatest
effect on company performance in cosmetic industry through event study method.
4.1. Prerequisites
Before examining the hypotheses by event study methodology, several conditions are supposed
to be satisfied. First of all, a crucial precondition is a semi-strong form of market efficiency,
which is a class of Efficient Market Hypothesis (EMH) and includes two underlying
assumptions (Fama, 1991). One states that publicly available information as a whole is
calculated into stock returns. The other indicates that stock prices will change when the
28
should be left to formal releases. Information leakage possibly happens prior to official
announcements from insiders or partners, but the effect of information leakage is limited to a
small range. It is possible that few investors are aware of leakages and believe the veracity.
However, taking this possibility in consideration, the start point of event window is set two
trading days before official announcements. Moreover, the four trading days after official
announcements are included in this research to reflect the dispersion of information across
stakeholders. The last condition is about the overlapping impact of two events occur in close
proximity on financial performance. Sorescu, Warren and Ertekin (2017) suggest that
overlapping events will not lead to the confounded observation in short-term event studies.
Based on the choice of short estimation period and event window, it is assumed that
eliminating the overlapping impact is unnecessary.
4.2. Population identification and Data collection
My sample population comes from main cosmetic companies located in the U.S., Japan, the
U.K., Germany and France those listed in different exchanges (NYSE, NASDAQ, TSE, LSE,
FWB, PAR) within the period from 2012 to 2016 and the sources of data of stock price are the
Center for Research in Security Prices (CRSP) Database and Google Finance (The information
of selected firms is exhibited in Appendix 1). The data of NYSE and NASDAQ comes from
CRSP and the other data is collected from Google Finance. I extract all the event dates from
Factiva.com which classifies a large amount of business and news publication into different
29
Partnerships/ Collaborations and Corporate Social Responsibility, are aligned with the researching objects in this research.
All listed cosmetic companies are sorted by overall sales of 2014 - 2016 (from Statista.com)
and the leading 30 companies headquartered in these five representative countries are selected
given that: (a) they were listed in same stock exchanges between 2012 and 2016, (b) did not
merged or acquired by other companies over this period. The companies fulfilled these criteria
are typical in the worldwide cosmetic industry with high revenue and market capitalization.
Specially, among these 30 companies, Coty Inc. was listed in NYSE since 23rd June 2013 so that its events are selected from this day. These 5 countries were chosen for the following
reasons: (a) the market sizes of these countries are leading globally (Ana-Maria, Constantin,
Catalina, Alina, 2011), which means the consumption in this range accounts for majority of
worldwide cosmetic market, (b) they locate in North America, Europe and Asia where the
leading companies are headquartered and major exchanges are located as well. Selecting the
broad area eliminates the mistakes caused by regional difference.
4.3. Variables
4.3.1. Independent variables: Three types of events
30
Dmitri and Joel (2009) found the s short-term response of the stock market to new product
introduction news is observable and valuable to analyze from a range of different categories.
This type of event is closely related to firm profitability.
b. Collaboration
Kenneth and Nigel (2011) generated a research about the correlation of collaboration between
marketing and sales with operating performance. The collaboration in product categories and
campaigns with other brands or artists is common in the cosmetic industry, which also
contributes to stock return.
c. Corporate social responsibility activities
CSR activities always don‘t lead to economic profit directly, but have a significant impact on
consumers‘ attitude towards the brand. Consumers‘ favorable attitudes towards CSR activities
are likely to be transferred to the attitude towards the company or the brand (Chernev & Blair,
2015).
4.3.2. Dependent variable: Stock returns
Stock return is defined as the logarithm of ending stock price minus the original price of the
stock. In this research, the researching object is daily stock price, so the calculating period is
one trading day.
Total Stock Return = ln(𝑃1/𝑃0) (1) 𝑃0 = 𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑠𝑡𝑜𝑐𝑘 𝑝𝑟𝑖𝑐𝑒 (A trading day before) 𝑃1 = 𝐸𝑛𝑑𝑖𝑛𝑔 𝑠𝑡𝑜𝑐𝑘 𝑝𝑟𝑖𝑐𝑒
31 4.4. Event study (Sorescu, Warren & Ertekin, 2017)
The main reason needed to conduct an event study is marketing events whose new information
is likely to affect the future expected cash flow. This methodology aims to calculate abnormal
stock return and calculated abnormal return so as to represent the effect on financial
performance. The studying steps are following:
a. Event definition:
New product introduction is defined as that companies launch new products or services
independently to differentiate from collaboration. Collaboration is defined as that firms
collaborate with other corporations, brands or artists with profit purpose in different forms,
such as co-branding products, joint development and promotion events. CSR activity is defined
as all the actions contributing to the environmental or social well-being, like public donations,
sponsorship of NGO and the environmental advocacy activities.
b. Selection criteria:
Selection criteria have been elaborated in population identification (4.2.).
c. Estimation period and event window:
This research chooses short estimation period and event window due to the high frequency of
marketing events in the cosmetic industry. For instance, the brand under Estee Launder Companies Inc. introduced at least one new product to meet existing customers‘ expectation
32
products are fast moving consumer goods, which indicate that the short product cycle life
elicits repeatable purchasing behaviors in short term. According to this fact, the estimation
period only covers a length of 28 calendar days before event window, including 20 trading days.
The event window starts on 2 trading days before event announcement date and ends on 4
trading days later. In one word, the estimation period is day (-22, -2) and event window is day
(-2, 4).
𝑇1 𝑇2/𝑡1 𝑡0(𝑒𝑣𝑒𝑛𝑡) 𝑡2
Estimation period event window
d. Decomposition of total return (𝑅𝑖𝑡) into normal (𝑁𝑅𝑖𝑡) and abnormal (𝐴𝑅𝑖𝑡) returns: 𝑅𝑖𝑡 = 𝑁𝑅𝑖𝑡+ 𝐴𝑅𝑖𝑡 (2)
If the event did not take place, market model is applied in this research to calculate the normal
stock return. This model assumes that the ―beta‖ accounts for the difference among the stocks
and does not assume the ―beta‖ of each stock is equal to one. Therefore, abnormal return is
better to be defined as the residuals of the market model, 𝑅𝑖𝑡 = 𝛼𝑖 + 𝛽𝑖𝑅𝑚𝑡+ 𝜀𝑖𝑡 (3)
where 𝑅𝑖𝑡 is total return of the firm i on each day t within estimation period, 𝑅𝑚𝑡 is market return on each day t, 𝛼𝑖 and 𝛽𝑖 are the intercept and slope of this linear relationship between stock of firm i and the market returns respectively, and the 𝜀𝑖𝑡 is error with E(𝜀𝑖𝑡) = 0 and 𝛼𝑖, 𝛽𝑖 and 𝜀𝑖𝑡 are estimated through the ordinary least squares (OLS) regression on data within the estimation period. Market returns vary according to the exchanges the companies list in.
33
S&P 500 index is used for NYSE and NASDAQ; DAX index is used for FWB; FTSE 100
index is used for LSE; Nikki 300 index is used for TSE; CAC 40 index is used for PAR. So, the
expected normal return is calculated using the following formula: 𝑁𝑅𝑖𝑡 = 𝛼𝑖 + 𝛽𝑖𝑅𝑚𝑡 (4)
Then, the abnormal return of firm i on day t is defined as the predicted error of market model: 𝐴𝑅𝑖𝑡 = 𝑅𝑖𝑡 − 𝛼𝑖 − 𝛽𝑖𝑅𝑚𝑡 (5)
The abnormal returns of each firm i over event window (-2, 4) are aggregated and the
accumulative abnormal return (CAR) will be worked out. CAR𝑖(𝑡1, 𝑡2) = ∑𝑡𝑡= 𝑡2 1𝐴𝑅𝑖𝑡 (6)
Finally, the cumulative abnormal returns of each type e of events within researching period are
aggregated and averaged to ACAR.
𝐴𝐶𝐴𝑅𝑒 =𝑁1∑ 𝐶𝐴𝑅𝑒 (7)
5. Result and discussion
In this chapter, the result of the statistical analysis will be introduced. First, the general
characteristics of the dependent variable and correlation between any two variables are
34 5.1. Statistic description
This research uses CAR that calculated by event study methodology as the dependent variable
in the following analyses. The characteristics of CAR of different selected events are
summarized in Table 2. The samples size of each group is greater than 100. A successful
marketing event is defined when CAR is positive. For all different types of events, each ACAR
across firms is slightly positive, which demonstrates that the cumulative abnormal returns of
three types of events are significant from statistic perspective over the event window. Totally
more than 80% CARs are positive, but the performance of CSR activities (success rate =
60.78%) is not as good as the other two.
Table 2. Characteristics of CAR
Event Type E(CAR)/ 𝐴𝐶𝐴𝑅𝑒 Std.dev. Positive Success Rate Event Number
New Product .0378 .0472 133 90.48% 147
Collaboration .0347 .0498 98 89.91% 109
CSR .0113 .0590 62 60.78% 102
Total .0293 .0527 293 81.84% 358
To analyze the relationship between independent variables and the dependent variable in
different conditions, a control variable – revenue level is added in this research. Actually, all
the independent variables and the control variable are set as dummy variables to represent
which type of event happened and this company has higher-level revenue or lower-level
35
more than 10,000 million U.S. dollars and 0 if less. Variable NewProduct is 1 if a company
formally announces the release of a new product or new service to market and 0 if the company
does not make this announcement. Similarly, variable collaboration is 1 if a company formally
announces the collaboration with other organizations or individuals and 0 if the company does
not. Variable CSR is 1 if a company formally announces a CSR-related activity and 0 if the company doesn‘t. And the dependent variable CAR is the only numeric variable.
Before researching the statistical relationship between three selected types of events and the
stock returns, it is crucial to check the correlation between them. The correlation matrix is
conducted to check for the significant correlations between any two variables. Based on the
values in Table 3, CSR shows a negative correlation with CAR with a Pearson correlation
coefficient of r = -.216 and the significance value < 0.01. This correlation is not expected in the
hypothesis since according to this result investment in CSR activities is likely to take less
financial benefits. Likewise, the significant correlation between NewProduct and CAR is less
strong but positive (r = .135, p < 0.05). Collaboration also shows moderately positive
correlated with CAR, but it is not of significance (r = .068, p = .099). These positive
correlations meet the expectation that the stock returns will grow when companies introduce
new products or establish partnerships. There is slightly positive correlation between
RevenueLevel and CAR, where significance does not reach 0.05 level (r = -0.008, p = .439). Higher revenue level does not lead to a significant increase of total CAR.
36
Variables Mean Std.dev. 1 2 3 4 5
1. Revenue Level .634 .482 -
2. New Product .410 .493 .068** -
3. Collaboration .300 .461 -.077 -.552** -
4. CSR .285 .452 .004 -.527** -.418** -
5. CAR .029 .053 -.008 .135* .068 -.216** -
** Correlation is significant at the 0.01 level (2-tailed)
* Correlation is significant at the 0.05 level (2-tailed)
5.2. Multiple regressions analyses
To further examine the impact of three selected types of events on abnormal return, two
multiple regression analyses are done on CAR of event window (-2, 4). The results of
regression analyses are able to provide more insights in hypotheses testing. Two regression
models are formulated as the following:
Model 1:
𝐶𝐴𝑅𝑛 = 𝛽0+ 𝛽1× 𝑁𝑒𝑤𝑃𝑟𝑜𝑑𝑢𝑐𝑡𝑛+ 𝛽2× 𝐶𝑜𝑙𝑙𝑎𝑏𝑜𝑟𝑎𝑡𝑖𝑜𝑛𝑛+ 𝛽3× 𝑅𝑒𝑣𝑒𝑛𝑢𝑒𝐿𝑒𝑣𝑒𝑙𝑛+ 𝜀𝑛 Model 2:
𝐶𝐴𝑅𝑛 = 𝛽0+ 𝛽1× 𝑁𝑒𝑤𝑃𝑟𝑜𝑑𝑢𝑐𝑡𝑛 + 𝛽2× 𝐶𝑜𝑙𝑙𝑎𝑏𝑜𝑟𝑎𝑡𝑖𝑜𝑛𝑛+ 𝛽3× 𝑅𝑒𝑣𝑒𝑛𝑢𝑒𝐿𝑒𝑣𝑒𝑙𝑛+ 𝛽4× 𝑁𝑃𝑅𝐿𝑛 + 𝛽5× 𝐶𝑜𝑅𝐿𝑛+ 𝜀𝑛
where 𝐶𝐴𝑅𝑛 is the cumulative abnormal return for the event n and is the dependent variable in the regression models. 𝑁𝑒𝑤𝑃𝑟𝑜𝑑𝑢𝑐𝑡𝑛 and 𝐶𝑜𝑙𝑙𝑎𝑏𝑜𝑟𝑎𝑡𝑖𝑜𝑛𝑛 are dummy variables, which stand for the type of event n. 𝑅𝑒𝑣𝑒𝑛𝑢𝑒𝐿𝑒𝑣𝑒𝑙𝑛 is dummy variable as well, which stands for the
37
revenue level of the firm in last financial year. 𝑁𝑃𝑅𝐿𝑛 is an interacted variable that indicates the interaction between new products/ services introduction and the revenue level of the firm (𝑁𝑒𝑤𝑃𝑟𝑜𝑑𝑢𝑐𝑡𝑛× 𝑅𝑒𝑣𝑒𝑛𝑢𝑒𝐿𝑒𝑣𝑒𝑙𝑛). 𝐶𝑜𝑅𝐿𝑛 is also an interacted variable that stands for the interaction between collaboration activities and the revenue level of the firm
( 𝐶𝑜𝑙𝑙𝑎𝑏𝑜𝑟𝑎𝑡𝑖𝑜𝑛𝑛× 𝑅𝑒𝑣𝑒𝑛𝑢𝑒𝐿𝑒𝑣𝑒𝑙𝑛 ). For individual event n, values of variable 𝑁𝑒𝑤𝑃𝑟𝑜𝑑𝑢𝑐𝑡𝑛 and 𝐶𝑜𝑙𝑙𝑎𝑏𝑜𝑟𝑎𝑡𝑖𝑜𝑛𝑛 must be 0 if the type is CSR activity. Therefore, the constant variable 𝛽0 can reflect the effect of CSR activities and the variable CSR is omitted in the models. Model 1 does not include the interacted effect of control variable so as to explore
the independent impact of independent variables and the control variable on cumulative
abnormal return. After two interacted variables entering, the model 2 is formulated to figure out the contribution of combined effect of events and firms‘ different revenue level to the
dependent variable. The result of two models is exhibited in Table 4.
Table 4. Multiple regressions of CAR
Model 1 Model 2
R .218 .295
R square .048** .087***
Adjusted R square .039** .074***
Std.Error .052 .051
B Std.Error Beta t B Std.Error Beta t
NewProduct .027*** .007 .248 3.990 .0507*** .011 .474 4.574 Collaboration .023** .007 .205 3.287 .0571*** .011 .499 5.096 RevenueLevel -.001 .006 -.009 -.179 .031** .010 .288 3.009
38 NPRL -.038** .014 -.320 -2.740 CoRL -.055*** .014 -.397 -3.827 * Significance is at 0.05 level **Significance is at 0.01 level ***Significance is at 0.001 level
The result of the regression model 1 is statistically significant and explains 4.8% variance of CAR (𝑅2 = .048, F (3, 354) = 5.886, p < .01), indicating that most variances of the dependent variable are explained by other variables. Two of the four variables including the constant
variable are statistically significant, with NewProduct recording a higher coefficient (B = .027,
p < .001) than Collaboration (B = .023, p < .01). In other words, if a company launches a new
product in the market or establishes a partnership with other organizations or individuals, the
cumulative abnormal return will increase by 2.7% or 2.3%. As indicated by the positive
coefficients on the two variables, H1 and H2 are thus supported. The relation between CSR
activities and CAR is positive but not significant (B = .012, p = .058). Therefore, H3 is rejected.
The effect of the control variable is negative and lacks of significance (B = .001, p = .858).
Furthermore, the result of the regression model 2 is statistically significant with the explained variance of CAR increasing by 3.9% (𝑅2 = 0.087, F (5, 352) = 6.704, p < 0.001). There is a difference between the values of the unstandardized coefficient in the two models in terms of
these two independent variables due to the interactive effect of the revenue level. The
interactive effect of NewProduct and ReveuneLevel is significant with a negative coefficient (B
39
the relationship between new product introduction and CAR. Because the coefficient of
NewProduct grows to .0571 (p < .001), it can be considered that when a company with a lower level of revenue introduces a new product or service, its CAR will increase by 5.07%. On the
other hand, according to the interacted variable NPRL (B = -.038. p < .01), if a company with a
higher level of revenue launches a new product or service, its CAR will only increase by only
1.27%. Similarly, the interactive effect between Collaboration and RevenueLevel is significant
with a negative coefficient (B = -.055, p < .001). As the coefficient of Collaboration rises
to .0571 (p < .001), it can be considered that when a company with a lower level of revenue
builds a partnership with other organizations or individuals, its CAR will increase by 5.71%. In
addition, if a company with a higher level of revenue builds a partnership, its CAR will
increase slightly by 0.21% based on the variable CoRL (B = -.055, p < .001). The interactive
effect between CSR and RevenueLevel shows significance (B = .031, p < .01), which indicates
that compared to the company with a lower level of revenue, the CAR of the company with
higher level of revenue will increase by an additional 3.1%. However, the constant variable that
stands for CSR has a negative but insignificant relationship with the dependent variable (B =
- .009, p = .294).
40 6.1. Theoretical and practical implications
In line with previous studies, this research provides evidence that there are positive relations
between the three selected types of marketing events and a company‘s financial performance.
Recent studies attempted to understand the more specific linkage between marketing strategies
and financial performance. This research adds its contribution by examining the change of an
abnormal stock market return when a leading cosmetic company officially announces any of
the following: the launch of a new product, collaboration or CSR-related activities. Statistical
analysis used in comparing common marketing strategies increases the depth of research in this
marketing field. This study further elucidates our understanding of event study applied to the
cosmetic industry. Model 1 does not take into account the interaction with a company‘s
revenue level, so its result tends to explain the situation of the cosmetic industry as a whole.
Three positive coefficients indicate that the selected events contribute to a positive abnormal
return; however, the impact of CSR-related events is not significant.
The result of Hypothesis 1 supports the theory of Chaney, Devinney and Winer (1991) that new
product introduction does raise shareholders‘ expectations, leading to a projected increase of in
cash flow. When stakeholders or potential investors are faced with a new product
announcement, their investment decisions are likely to be positive (Hao & Wang, 2015). The
conceptualization of new products as innovation to boost the growth and development of
business, as introduced by Woodman and Sawyer (1993), is also supported by the highest
41
to possible pre-announcements, which corroborates the findings of Sorescu, Shankar and
Kushwaha (2007). The opposing standpoint that investors cannot be immediately affected by
new product announcements does not appear to be borne out in the cosmetic industry (Eddy &
Saunders, 1980). Cosmetic companies have such strong power in both the conventional and
social media that information reaches investors quickly. Furthermore, the result of Hypothesis
2 is a statistical corollary to current research that stock investors will respond positively to
collaboration with other corporations or brands. (Cao & Sorescu, 2013). Partnerships can be
regarded as the competitive advantages to expanding the new market and enriching the
productive resources that build an access to related skills and knowledge (Bucklin & Sengupta,
1993). It is easy for stock investors to be sensitive to the high-quality signal based on such
collaboration. In this case, positive associations will be attached to the products themselves.
Thus customers‘ and investors‘ evaluation of the products or the company will be significantly
enhanced (Rao, Qu & Ruekert, 1999).
With particular regard to cosmetic companies, who play such a key role in the supply chain as
both manufacturers and retailers, collaboration is able to create greater stock returns. This is
reflected in the second highest coefficient in the result. Hypothesis 3 is rejected in the
analysis of model 1. This result concurs with the view of Yuwita and Kalanjati (2017) and
Brine, Brown and Hackett (2007), who find no significant effect of CSR-related activities on
the stock returns in the cigarette industry and the Australian context respectively. However,
numerous studies across industries stress that CSR-related activities contribute positively to a company‘s financial performance (Tsoutsoura, 2004; Friede, Busch & Bassen, 2015; Oh, Bae,
42
Currim, Lim & Zhang, 2016). In my own opinion, stakeholders in different industries tend to
assess CSR-related activities from different perspectives. A few investors believe that
investment in CSR is an unnecessary cost which cannot directly lead to revenue increase
within a short period (Jensen, 2002, p. 242). However, Fatemi, Fooladi and Tehranian (2015)
still see the positive value of CSR but suggest that investors will not change their expectation
and reactions overnight. Cosmetic products as fast-moving consumer goods aim to expand
their market share rapidly to increase their revenue. This goal leads the companies to place
greater attention on product development and marketing promotion rather than any long-term
vision. Thus, the short event window in this study has its limitations in yielding positive results.
This comparison among three types of marketing activities also demonstrates that the effects of
product-related activities are greater and more significant than the non-product-related activities on the firms‘ financial performance.
The result of model 2 expands the relationship between marketing events and stock market
returns by adding an extra factor, that is, the differing performances of companies in higher or
lower revenue levels. This result adds further insight into existing studies through dividing the
impact of one type of marketing event into two groups according to different levels of revenue
in the last financial year. All the interactions among the three selected types of events and
revenue levels of firms are significant in the result, which demonstrates the influence of the
level of revenue factor on stock returns. For companies with a higher level of revenue, the higher value of the significant coefficients comes from new product introduction (∆r = .0127, p
43
The financial contribution of CSR-related activities in higher-revenue companies is greater
than in lower-revenue companies. For the latter group of companies, collaboration has a greater
positive impact than new product introduction on stock returns. The reason for the difference is
that investors in lower-level companies have a higher expectation of collaboration, while for
investors in higher-level companies, new product introduction has greater value. Cosmetic
companies with lower profitability are likely to have a lower market share, with a relatively
limited number of their target customers. Therefore, their common business goal would be to
enhance their brand power and thereby reach out to a wider market. Collaboration with other
organizations or individuals would provide access to the new customer groups and even new
market segments. The companies in this group are therefore willing to take a shortcut to secure
a competitive advantage.
On the other hand, companies with higher profitability due to their large customer base would
tend to put more emphasis on new product development in order to increase the buying depth
of existing customers. When their loyal customers perceive new products or services in the
offing, there is an excellent probability of repeating their purchasing behaviors, and this
becomes the source of short-term revenue. Their investors do not regard collaboration as an
efficient means to grow the income rapidly and hardly make an obviously positive response at
the moment. Corporate social responsibility efforts affect the stakeholders‘ evaluation of a company‘s long-term performance and sustainability (Mishra & Modi, 2016). Investors are
likely to expect CSR-related activities to have indirect consequences on a company‘s long-term