Stock price response to marketing events in cosmetic industry : an event study

(1)

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)

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)

3

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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)

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 𝑅_𝑖𝑡 is total return of the firm i on each day t within estimation period, 𝑅_𝑚𝑡 is market return on each day t, 𝛼_𝑖 and 𝛽_𝑖 are the intercept and slope of this linear relationship between stock of firm i and the market returns respectively, and the 𝜀_𝑖𝑡 is error with E(𝜀_𝑖𝑡) = 0 and 𝛼_𝑖, 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)

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: 𝑁𝑅_𝑖𝑡 = 𝛼_𝑖 + 𝛽_𝑖𝑅_𝑚𝑡 (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.

𝐴𝐶𝐴𝑅_𝑒 =_𝑁1∑ 𝐶𝐴𝑅_𝑒 (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)

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)/ 𝐴𝐶𝐴𝑅_𝑒 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)

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)

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:

𝐶𝐴𝑅_𝑛 = 𝛽₀+ 𝛽₁× 𝑁𝑒𝑤𝑃𝑟𝑜𝑑𝑢𝑐𝑡_𝑛 + 𝛽₂× 𝐶𝑜𝑙𝑙𝑎𝑏𝑜𝑟𝑎𝑡𝑖𝑜𝑛_𝑛+ 𝛽₃× 𝑅𝑒𝑣𝑒𝑛𝑢𝑒𝐿𝑒𝑣𝑒𝑙_𝑛+ 𝛽₄× 𝑁𝑃𝑅𝐿_𝑛 + 𝛽₅× 𝐶𝑜𝑅𝐿_𝑛+ 𝜀_𝑛

where 𝐶𝐴𝑅_𝑛 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. 𝑅𝑒𝑣𝑒𝑛𝑢𝑒𝐿𝑒𝑣𝑒𝑙_𝑛 is dummy variable as well, which stands for the

(37)

37

revenue level of the firm in last financial year. 𝑁𝑃𝑅𝐿_𝑛 is an interacted variable that indicates the interaction between new products/ services introduction and the revenue level of the firm (𝑁𝑒𝑤𝑃𝑟𝑜𝑑𝑢𝑐𝑡_𝑛× 𝑅𝑒𝑣𝑒𝑛𝑢𝑒𝐿𝑒𝑣𝑒𝑙_𝑛). 𝐶𝑜𝑅𝐿_𝑛 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)

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)

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)

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)

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)

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)

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

Stock price response to marketing events in cosmetic industry : an event study