The marketing mix effectiveness premium for market leaders in the fast moving consumer goods sector

(1)

The marketing mix effectiveness premium for

market leaders in the fast moving consumer goods

sector

Master Thesis Sven Luijmers

Supervisors

Dr. Ir. Maarten J. Gijsenberg Roelof P. Hars MA MSc

University of Groningen Faculty of Economics and Business

Master of Science in Marketing Marketing Intelligence

(2)

The marketing mix effectiveness premium for market

leaders in the fast moving consumer goods sector

Master Thesis Sven Luijmers Helperzoom 203 9722 BL Groningen +31620528134 s.luijmers@student.rug.nl svenluijmers@gmail.com Student number: s3010414

First supervisor: Dr. Ir. Maarten J. Gijsenberg Second supervisor: Roelof P. Hars, MA MSc

University of Groningen

Faculty of Economics and Business MSc Marketing Intelligence

(3)

ABSTRACT

Higher market power, a price premium and scale economies make that market leaders are doing economically better over their smaller competitors. Nevertheless, the distinction regarding the effectiveness of the marketing mix tools between market leaders and followers have not been made. Therefore the main focus of this research is on finding asymmetries in the effectiveness of the marketing mix tools between market leaders and followers. An error correction model estimated short-term and long-term effects for mass advertising expenditures, feature advertising, display advertising, price increases and price decreases on the volume sales of the brand. The model is estimated one time for market leaders and one time for followers. For the proposed model we use four years of weekly data from the three largest Dutch supermarkets in the fast moving consumer goods sector. The research included three product classes with each three product categories, with a total of 33 brands. Market leader advantages are found concerning the short-term and long-term price increase elasticities and the short-term display elasticity. Counterintuitive to our expectations followers gain an advantage on the short-term mass advertising expenditures elasticity, the long-term display advertising elasticity, and the short-term and long-term price decrease elasticities. The term mass advertising expenditures elasticity, the short-term and long-term feature advertising elasticities are equal for market leaders and followers.

(4)

PREFACE

So, there I am writing my final words for my master thesis and I have to admit it has been an interesting journey.

When I started the pre-Master Marketing at the University of Groningen I had no idea on choosing the Marketing Management or Marketing Intelligence track. During the pre-Master I gained more and more interest for the Intelligence track, since I had the feeling the Management track would develop my general marketing knowledge and the Intelligence track would be my last opportunity to develop a special skill at academic level: analyzing and modeling big data. At the same time, this analyzing and modeling made me questioning myself if I would have enough capacity to capture these skills. Now I (almost) finished the Intelligence track I am happy to say I made the right choice and really gained analyzing and modeling skills that could help me somewhere in my future career. At this moment I am not sure if I will become a hardcore data analyst, since the MSc program made me realize I really like to work with people as well and I need more than only “dry” analyzing challenges. Now, I am proud to present my master thesis. I developed my analyzing skills by learning about the error correction model and gained increasing fun during the MSc thesis process. However, this was not possible without the help from others. First I would like to thank my supervisor Maarten Gijsenberg who always knew what to say to challenge me and get me on the right track. The combination of his positive social approach, critical comments and quick feedback made it very pleasant to have him as my graduation supervisor. Besides, I want to thank my fellow thesis group students for the exchange of ideas and collaboration over the last months. Last but not least I want to thank my girlfriend, family and friends for their unconditional support.

It was tough and sometimes a little frustrating, but overall a very challenging, educational and rewarding process. I hope and think this thesis forms a worthy finalization of my studies.

Sven Luijmers,

(5)

1. INTRODUCTION

Already in the seventies researchers indicated that “under most circumstances, enterprises that have achieved a high share of the markets they serve are considerably more profitable than their smaller-share rivals” (Buzzell, Gale, & Sultan, 1975). The meta-analysis of 276 prior studies from Szymanski, Bharadwaj, and Varadarajan (1993) found that, on average, market share has a positive effect on business profitability. However, they concluded that the magnitude of the market share-profitability relationship is moderated and therefore attempts at increasing market share alone may not yield greater profitability. The complex quest to explain these performance differences is an fundamental issue for companies. According to Caves and Porter (1977), the positioning of market leaders in industries where they can take advantage of favorable competitive forces, such as barriers to entry or mobility, enhances performance. So, winning the battle for market leadership is at the heart of competitive strategy of companies. Many companies try to overtake the market leader, which is previously defined by Smith, Ferrier, and Grimm (2001) as the "dethronement of leaders”. Some companies try this by imitating the market leader and although perfect imitation of the market leader is difficult, “imperfect imitation, from time to time, enables follower firms to surpass superior firms” (Posen, Lee, & Yi, 2013).

(7)

kind of Matthew effect: “to those who have, more shall be given” (Merton, 1968). This research therefore focusses on finding asymmetries in the effectiveness of the marketing mix tools between market leaders and followers. It is interesting to see if market leaders gain an effectiveness premium in their marketing mix tools, and if present, which tools should be used to gain maximum advantages regarding the volume sales. The possible double jeopardy of being a smaller follower firm, and possibly more difficulties for those followers to gain market share because they are less effective in their marketing mix, is addressed.

Since Dorfman and Steiner (1954) investigated the joint effects of the marketing mix tools pricing and advertising, many researchers followed and extended on investigating the relationship between advertising and pricing strategies (e.g. Chintagunta, Kadiyali, & Vilcassim, 2006; Gupta & Di Benedetto, 2007; Kadiyali, 1997; Karray & Amin, 2015). Recently Lu, Gou, Tang, and Zhang (2016) stated that pricing and advertising are considered as most important factors affecting consumer demands, and therefore are effective marketing tools to improve firm’s revenue. Opposing, Ataman, Van Heerde, and Mela (2010) found higher short-term and long-term sales elasticities for product and distribution, and smaller elasticities for advertising and discounting. Overall, research found negative short-term and long-term price elasticities (e.g. Gijsenberg, 2017; Greenleaf, 1995; Pauwels, 2007; van Heerde, Gijsenberg, Dekimpe, & Steenkamp, 2013) and small positive significant short-term and long-term advertising elasticities (e.g. Allenby & Hanssens, 2004; Gijsenberg, 2017; Sethuraman, Tellis, & Briesch, 2011). Since prior research suggests the importance of including both short-term and long-term advertising and pricing effects (e.g. Assmus, Farley, & Lehmann, 1984; Givon & Horsky, 1990; Gotlieb, Scholl, Ridout, Goldstein, & Shah, 2017; Palda, 1965), this research will extend on the research concerning both the short-term and long-term elasticities.

This particular research focusses on three product classes with each three product categories from the three largest Dutch supermarkets in the fast moving consumer goods sector (from now: FMCG sector). The classes beverages, food and personal care together represent a typical supermarket assortment. With including several product categories per class we try to make the findings in this study broadly generalizable.

(8)

(9)

2. RESEARCH FRAMEWORK AND CONCEPTS

In this research we focus on finding asymmetries in the effectiveness of the marketing mix tools between market leaders and followers. The possible double jeopardy of being a smaller follower firm, and possibly more difficulties to gain market share because they are less effective in their marketing mix, is addressed. This is considered as kind of a Matthew effect (Merton, 1968), for the companies who have a lot, more is given (the market leader has a higher marketing mix effectiveness) and for the companies who have less, less is given (smaller follower firms have a lower marketing mix effectiveness). Figure 1 provides the reader with the framework of the conceptual model.

Figure 1

Framework conceptual model

The model indicates that the five single marketing mix actions have an effect on sales, on the short-term as well as the long term. These relations are dependent by being the brand leader or a follower firm. By focusing the study on different product classes we represent a typical supermarket assortment. With including several product categories per class we try to make the findings in this study broadly generalizable.

2.1 Marketing mix: pricing and advertising

According to many researchers pricing and advertising are considered as the most important factors affecting consumer demands, and therefore are effective marketing tools to improve

Sales:

• Short-term • Long-term Marketing mix tools:

• Price increases • Price decreases

(10)

firm’s revenue (Lu et al., 2016). The pricing and advertising effects on the firm’s profit were already discussed in the fifties by Dorfman and Steiner (1954), when they investigated the joint effects. Building on this research several studies extended on investigating the relationship between advertising and pricing strategies (e.g. Chintagunta, Kadiyali, & Vilcassim, 2006; Kadiyali, 1997). According to them, advertising may lead to a higher or lower sales price, depending on the effect of advertising on the sales price. Since then authors focused among other things on joint pricing and advertising models for a new product facing potential competitive entry (Gupta & Di Benedetto, 2007), a model of cooperative advertising and pricing decisions for aligning manufacturer and retailer decisions with uncertainty demand (He, Prasad, & Sethi, 2009), and assessed the effects of cooperative advertising between competing retailers considering both advertising and pricing variables (Karray & Amin, 2015). Besides all the research that focused on pricing and advertising, Ataman et al., (2010) found higher short-term and long-term sales elasticities for product and distribution, and smaller elasticities for advertising and discounting. However, the scope of this research is only on pricing and advertising and will extend on the aforementioned research.

Pricing

Pricing is one of the most important issues in marketing (Gijsbrechts, 1993) and has grown in importance as a determinant of sales since the 1950s (Bijmolt, van Heerde, & Pieters, 2005). Therefore it is essential to understand pricing issues and to include pricing as a marketing mix tool in this study. In the past, many researchers studied on the effect of price discrepancies on sales, but those studies did not agree on the results. The prospect theory from Kahneman and Tversky (1979) implied that buyers will react asymmetrically to the price discrepancy depending on whether this discrepancy is perceived as a gain or a loss. This is supported by more recent research from Bruno, Che, and Dutta (2012), who concluded that “business customers react asymmetrically to price increases and price decreases”. Some researchers argued that price increases have stronger effects than price reductions (e.g. Kahneman, & Tversky, 1979; Kalyanaram & Winer, 1995), while others showed that consumers react more strongly to price decreases in the short run (e.g. Greenleaf, 1995; Pauwels, 2007), with similar results on the long-term.

(11)

price-to-sales elasticity of -2.6, and over time discounting has become more effective in boosting sales, while raising prices has an increasingly negative impact on sales. Jackson and Wilcox (2000) stated that price reductions may help companies to increase short-term sales, but it is arguably harmful to them in the long run. Gijsenberg (2017) and Kahneman and Tversky (1979) concluded that long-term increase elasticities are stronger than price-decrease elasticities. Adding on this, this paper examines if there are any differences in pricing effects between market leaders and followers. It might be the case that market leaders should worry less about price increases, compared to followers.

Because of the mixed results in prior research, it is unknown if price decreases or price increases will have a stronger impact on the volume sales. Regarding the size of the effects we expect stronger price elasticities on the short-term, and smaller price elasticities on the long-term.

Advertising

Advertising, next to pricing the second essential factor, is used to send product messages to consumers and induce purchases. Firms can pay to distribute advertising messages via some new media (e.g. websites, social media) or via traditional media (e.g. newspaper, magazines, displays) which is always fixed over a longer period (Lu et al., 2016). This study focusses on three advertising variables, namely mass advertising expenditures, feature advertising and display advertising.

Mass advertising expenditures

(12)

their metanalysis. This expectation is supported in prior research, which stated that lower elasticities are common in CPG categories (e.g. Allenby & Hanssens, 2004).

So, in this paper we consider mass advertising as an out-of-store brand-building tool. Mass advertising expenditures affect sales positively both in the short-term and long-term. We expect stronger reactions to mass advertising expenditures in the long-term, and smaller reactions to mass advertising in the short-term.

Feature and Display advertising

Next to mass advertising expenditures, companies often place in-store displays to highlight a certain product (from now: display advertising) and feature advertisements in their sales promotions programs (from now: feature advertising). Although these advertising variables can be deployed as single marketing actions, they are often combined with price reductions. “The beliefs are that the synergy of these marketing tools create greater sales than the sum of the single effects when each promotion is separately offered” (Zhang, 2006). Therefore marketing researchers have to be aware of observing strong multicollinearity between these advertising variables and price reductions in sales data.

Prior research indicated that display and feature ads can significantly increase sales, even when the price discount effect is controlled for (e.g. S. Gupta, 1988; Papatla, 1996; Srinivasan & Grover, 1992). Two behavioral explanations on the mechanisms, through which the effects take place, have been suggested by the marketing literature. First, Inman et al. (1991) proposed that individuals with low need for cognition react to the simple presence of a promotion signal, high need for cognition individuals react to a promotion signal only when it is accompanied by a substantive price reduction. Second, other researchers proposed that displays and feature advertisements can be utilized to form consideration sets (e.g. Andrews & Srinivasan, 1995; Fader & McAlister, 1990; Mehta, Rajiv, & Srinivasan, 2003). The brand is more prominent to customers and so the probability of being chosen is increased.

(13)

2.2 Short-term and long-term effects

We already addressed that some marketing mix tools encounter short-term oriented effects, and other tools are inducing long-term effects. “One of the early observations of researchers was that the effect of advertising in one period may be carried over, at least partially, into future periods” (Givon & Horsky, 1990). As result of the marketing mix decisions both the current sales (short-term) and the lagged-sales (long-term) are taken into account.

Thus, to make sure we encounter a complete model we focus on both the short-term and long-term sales effects of the marketing mix tools In this paper the short-long-term and long-long-term definitions are derived from Mela, Gupta, and Lehmann (1997).

The short-term effect: This is the current immediate (weekly) effect of the marketing mix tools on the sales of a brand.

The long-term effect: This is the one-week lagged (weekly) effect of the marketing mix tools on the sales of a brand.

As indicated it is important to include both the short-term and the long-term sales effects to make sure we present a complete model. “Misspecification by exclusion of a carryover term can result in a model which fits less well” (Assmus et al., 1984). Already in the sixties Palda (1965) showed, by applying a model with and without carryover effects, that not including carryover effects lead to a model that fits less well. In the recent years, not many researchers did address the topic of carryover marketing effects. Yet, Gotlieb, Scholl, Ridout, Goldstein, and Shah (2017) studied the short-term and long-term effects of exposure to campaign advertising for politicians. They found some direct effects, but concluded that the majority of effects are “sustained or sleeper effects, emerging after the campaign has ended” (Gotlieb et al., 2017). This implies that carryover effects might still hold nowadays and it is very important to include the long-term effects for marketing mix tools, even when they are aimed at direct short-term actions.

2.3 Market leaders and follower firms

(14)

with the highest market share may become isolated in competitive interactions and enjoy the use of established routines (Derfus et al., 2016). According to Katila, Chen, and Piezunka (2012) large market leaders tend to maintain their dominance by competing based on conservative moves. Of the top three leading brands in 1923, more failed than remained market leaders by 1997 (Golder, 2000). They found that nondurable goods (that include FMCG) have higher leadership persistence and lower failure rate than durable goods. Although difficult, overtaking the market leader is possible. A follower firm that takes advantage of the opportunities in changing environment, can dethrone the market leader (Smith et al., 2001).

In this section we propose three mechanisms that imply a difference in effectiveness between market leaders and followers. These mechanisms show that market leaders gain effectiveness premiums compared to their followers. The premiums possibly transfer to a difference in the effectiveness of the marketing mix which creates even more difficulties for followers to gain market share and overtake the market leader.

Price premium

(15)

favor of the market leader. Kotler and Keller (2006) indicate that a better (perceived) “product quality will maintain a high level of customer satisfaction, which encourages customers to make their next purchases”. This statement is in line with more recent research from Yuen and Chan (2010), who indicated that product quality is positively related to customer loyalty. For this research we expect that the higher perceived product quality transfers to brand loyalty, which is followed by better perceived marketing actions for market leaders, since the consumer will follow the brand regardless of the marketing action. This implies a marketing mix effectiveness premium for market leaders.

Market power of companies

(16)

Economies of scales in marketing actions

(17)

3. METHODOLOGY

Given the research objectives stated in the previous section, several modeling challenges arise. First, this study aims to find differences in the effectiveness of the marketing mix tools between market leaders and followers, and thus the modeling approach should allow for differences between brands. Besides the research focusses on both the effect of the marketing mix variables on the short-term (immediate effects) as the delayed effects on the long-term. Therefore the model should distinguish between short- and long-term effects, for both market leaders as followers. Next, the study aims for to provide generalizable insights for different product classes and categories. So the model should allow for insights for the classes as well as the individual categories.

3.1 Error correction model

To assess the impact of the aforementioned marketing mix variables on sales, we propose a new time-varying error-correction model (ECM) that separately estimates short- and long-run elasticities. This model accounts for all the above-mentioned modelling challenges.

∆𝑙𝑛𝑆_𝑏𝑡 = 𝛼_𝑏0+ 𝛽1_𝑏𝑠𝑡∆𝑙𝑛𝑀𝐴𝐸_𝑏𝑡 + 𝛽2_𝑏𝑠𝑡∆𝑙𝑛𝐹𝐴_𝑏𝑡 + 𝛽3_𝑏𝑠𝑡∆𝑙𝑛𝐷𝐴_𝑏𝑡+ 𝛽4_𝑏𝑠𝑡∆𝑙𝑛+𝑃_𝑏𝑡 + 𝛽5_𝑏𝑠𝑡∆𝑙𝑛−_𝑃 𝑏𝑡 + ∏ [𝑙𝑛𝑆𝑏𝑡−1− ( 𝛽6_𝑏𝑙𝑡𝑙𝑛𝑀𝐴𝐸_𝑏𝑡−1+ 𝛽7_𝑏𝑙𝑡𝑙𝑛𝐹𝐴_𝑏𝑡−1 + 𝛽8_𝑏𝑙𝑡𝑙𝑛𝐷𝐴_𝑏𝑡−1+ 𝛽9_𝑏𝑙𝑡 𝑙𝑛+𝑃_𝑏𝑡−1 + 𝛽10_𝑏𝑙𝑡 𝑙𝑛−_𝑃 𝑏𝑡−1 )] 𝑏 + 𝛽11𝑄1_𝑏𝑡 + 𝛽12𝑄2_𝑏𝑡+ 𝛽13𝑄3_𝑏𝑡+ ε_𝑏𝑡 where

∆ First difference operator

Sbt Volume sales of brand b in week t αb0 Intercept of brand b

MAbt Mass Advertising expenditures of brand b in week t FAbt Feature Advertising of brand b in week t

DAbt Display Advertising of brand b in week t +_P

(18)

Because volume sales, mass advertising expenditures, feature advertising, display advertising, price increase and price decrease are specified in natural logarithms, effects can be interpreted as elasticities1. The elements of 𝛽_𝑏𝑠𝑡 are the instantaneous or short-run elasticities. The 𝛽_𝑏𝑙𝑡 parameters give the marginal effect of a permanent change on long term effects in sales, and also the cumulative effects of the temporary changes. As such, the 𝛽_𝑏𝑙𝑡 parameters describe the long-run equilibrium relationship between the marketing mix tools and sales. The πb represents the speed with which the adjustment to long-term equilibrium occurs. In order to capture the marketing mix effects for both market leaders and followers the model has been estimated twice: once to assess the effects of the marketing mix on the volume sales for market leaders and once to assess the effects of the marketing mix on the volume sales for followers.

3.2 Stationarity

Prior to model estimation, we tested whether all variables are stationary. Recent studies show that tests based on individual series lack power compared with panel-based unit-root tests. Therefore both the Levin, Lin, and Chu (2002) and Im, Pesaran, and Shin (2003) panel unit-root tests are conducted on the (log transformed) variables. When the series are stationary, one can not only interpret the long-term effects as the permanent effects of permanent changes in the advertising variables, but also as the cumulative effects of temporary changes (Gijsenberg, 2014). The results of both tests show that for mass advertising expenditures, feature advertising, display advertising, pricing and volume sales the null hypothesis should be rejected at the 5% level, thus indicates that all series are (trend)stationary.

3.3 Model estimation procedure

First an Ordinary Least Squares (OLS) regression is conducted for all separate brands to obtain the initial estimates of the ECM (3.1 Error correction model). To obtain the long-term effects for the brands, the initial estimates for the lagged (long-term) variables are divided by

−π to obtain the long-term advertising and pricing elasticities (e.g. 𝛽𝑏𝑙𝑡

−π ). Next, the standard errors and t-values of the long-term advertising parameters have been calculated using the delta method (Greene, 2003, p. 175).

In order to assess the effects of the marketing mix variables on the volume sales across all brands, for both short- and long-term, they are evaluated by means of the added Z-method

(19)

(Rosenthal, 1991)2. The added Z-method gives us the opportunity to draw empirical generalizations over the ECM estimations for leaders and followers, since it allows for the combination of the p-values across the different brands. With use of this method each effect of the 9 brands considered as market leader are therefore taken together for one estimation, as well the effects for the 24 followers are combined for one estimation.

(20)

4. DATA DESCRIPTION

For the proposed model we use four years of weekly data (week 29, 1994 through week 28, 1998) from the FMCG sector. Data are available for the three largest supermarket chains in the Netherlands (based on a sample of ±350 stores). The dataset covers a wide range of food, beverages and personal care categories and thus provides us with a good sample of a typical supermarket assortment. Table 1 provides the reader an overview of the product classes and categories used in this study. To make sure there are no misunderstandings on these categories we provide the reader with a definition. Derived from prior literature definitions, we concluded on the following definition that is used for this study: “a product category is a set of competing brands that consumers perceive as close substitutes”.

4.1 Included brands and thresholds

(21)

Table 1

Overview of included product classes and categories

Product classes Categories Number of brands per category

Example brands

Beverages Cola 4 Coca Cola, Pepsi

Pilsners 4 Heineken, Grolsch

Ground coffee 4 Douwe Egberts, Kanis & Gunnink

Food Chips 3 Smiths, Croky

Candy bars 4 Mars, Twix, Milkyway

Dry soup 3 Honig, Knorr

Personal care Deodorant 4 Sanex, Nivea, AXE

Toothpaste 4 Aquafresh, Prodent

Diapers 3 Pampers, Huggies

The aforementioned thresholds together made us concluding on a dataset with 3 product classes with each 3 product categories. The market leader (9 brands in total) for each of the categories is determined as the brand with the largest market share over the four-year period of the dataset. This market share is computed with the volume sales of each brand in the category. The included followers (24 brands) for each category are the two or three companies with the largest market share after the market leader. The smaller competitors are not included in this research because “companies are affected by the firms that are similar in size and resources” (Debruyne & Reibstein, 2005). Smaller firms are therefore not considered as direct competition of the market leader. By focusing the study on as well beverages, food and personal care we represent a typical supermarket assortment. By including several categories per class we try to make the findings in this study broadly generalizable.

4.2 Model-free insights

(22)

Table 2

General descriptive statistics

Product classes Average market share Advertising frequency (% of weeks) Feature frequency (% of weeks) Display frequency (% of weeks) Overall Leaders Followers 15.51% 31.58% 9.49% 50.93% 53.26% 50.06% 57.94% 65.33% 55.17% 84.67% 92.31% 81.81% Beverages Leaders Followers 18.09% 40.34% 10.68% 74.52% 85.90% 70.73% 74.76% 84.29% 71.58% 92.75% 99.20% 90.60% Food Leader Followers 18.65% 37.22% 10.70% 36.88% 47.60% 32.28% 56.44% 72.12% 49.73% 89.23% 99.36% 84.89% Personal care Leaders Followers 9.84% 17.16% 7.09% 37.98% 26.28% 42.37% 40.95% 39.58% 41.47% 71.72% 78.37% 69.23%

(23)

market leader. Regarding the display advertising frequency it is again the market leader that advertises more often and so is in line with the ratio of all the other numbers.

4.3 Volume sales for market leaders and followers

Figure 2 shows the indexed volume sales evolution for all brands over the full dataset length of 208 weeks. The figure distinguishes between the sales evolution for market leaders and follower firms. The evolution appears as an index relative to the average level of the brands over the whole period. Because the volume units are different for different types of products (e.g. grams, liters), sales evolutions were indexed on a brand-per-brand basis, after which the average per week for both market leaders and followers is computed.

Figure 2

Indexed volume sales evolution: market leaders compared to followers

(24)

5. MODEL-BASED INSIGHTS

5.1 Quality of the model

Before discussing the results of the analysis, the fit of the model is discussed. Since this study makes the distinction between market leaders and followers, the model is estimated twice, once for market leaders and once for followers (table 3).

Table 3

Model fit

ECM Model R² Adjusted R²

Market leader 0.5475 0.5143

Follower 0.6244 0.5969

For both the R² as the adjusted R² we computed an average from all the single brands. The R² is telling us how much of the variance in the volume sales is been explained by the independent variables from our model. When looking at the numbers we see that the follower model is doing slightly better than the market leader model, respectively 54.8% and 62.4%. The adjusted R², which penalizes for adding variables, also indicates that the follower model is doing better than the market leader model with respectively 51.4% and 59.7%.

5.2 Model results

Table 4 and 5 show the parameters estimates for market leaders and followers together with the associated added Z-scores (Rosenthal, 1991).

5.2.1 Market leader model results

(25)

Table 4

Model results for market leaders

Variable Weighted β Z-score p-value Intercept 𝛼 5.2475 12.1262 0.0000*** ST Advertising expenditures 𝛽1𝑠𝑡 _0.0005 _1.1403 _0.1271 ST Feature Advertising 𝛽2𝑠𝑡 _0.0087 _8.2976 _0.0000*** ST Display Advertising 𝛽3𝑠𝑡 0.0123 5.6247 0.0000*** ST Price Increase 𝛽4𝑠𝑡 -0.7900 -6.0799 0.0000*** ST Price Decrease 𝛽5𝑠𝑡 -1.9837 -14.8858 0.0000*** LT Advertising 𝛽6𝑙𝑡 0.0024 2.4600 0.0069*** LT Feature Advertising 𝛽7𝑙𝑡 0.0125 4.6032 0.0000*** LT Display Advertising 𝛽8𝑙𝑡 0.0049 1.1930 0.1164 LT Price Increase 𝛽9𝑙𝑡 -0.3695 -3.2311 0.0006*** LT Price Decrease 𝛽10𝑙𝑡 -0.3896 -3.3731 0.0004*** Q1 𝛽11 -0.0218 -2.6303 0.0043*** Q2 𝛽12 -0.0215 -2.4942 0.0063*** Q3 𝛽13 -0.0129 -1.4864 0.0686* Adjustment π -0.4010 -20.8429 0.0000*** * p < 0.10. ** p < 0.05. *** p < 0.01.

When looking at the advertising effects for market leaders we find significant short-term effects for feature advertising and display advertising, and significant long-term effects for mass advertising expenditures and feature advertising.

For mass advertising expenditures we find a non-significant short-term elasticity (𝛽1𝑠𝑡_{) and a} significant positive long-term elasticity (𝛽6𝑙𝑡) of 0.0024 (p<0.01). We expected to find positive elasticities on both term and long-term. Because of the non-significant short-term effect our expectation is partially true. Besides we expected stronger elasticities for the long-term, and weaker elasticities for the short-term. This expectation is true since the weighted β of the long-term effect is larger than the non-significant short-term effect.

(26)

For the pricing effects for market leaders we find significant short-term effects for price increases and price decreases, and also for the long-term both price increases and price decreases have a significant effect on the volume sales. The negative short-term price increase elasticity (𝛽4𝑠𝑡) is -0.7900 and significant (p<0.01), also the short-term price decrease (𝛽5𝑠𝑡) elasticity is negative -1.9837 and significant (p<0.01). For the long-term pricing effects the price increase (𝛽9𝑙𝑡) elasticity is -0.3695 and significant (p<0.01), the long-term price decrease (𝛽10𝑙𝑡_{) elasticity is -0.3896 and significant (p<0.01) as well. The negative elasticity} for the price decrease seems to be in the wrong direction. But the negative value for the price decrease in the dataset, combined with the negative β, results in a positive sales effect and therefore is in the direction we expected it to be. Because of mixed results in prior research, it was unknown if we had to expect stronger price increase or price decrease elasticities. Conforming the results of Greenleaf (1995) and Pauwels (2007) the results from the analysis show that the effects of price decreases are stronger than price increases on both the term and long-term volume sales. Besides we expected stronger price elasticities on the short-term compared to the long-short-term. This expectation turned out to be true, since the weighted β’s are more extreme values for the short-term for both the price increase elasticity as the price decrease elasticity.

5.2.2 Follower model results

(27)

Table 5

Model results for followers

Variable Weighted β Z-score p-value Intercept 𝛼 5.6362 20.7495 0.0000*** ST Advertising expenditures 𝛽1𝑠𝑡 _0.0007 _2.3927 _0.0084*** ST Feature Advertising 𝛽2𝑠𝑡 _0.0099 _15.1557 _0.0000*** ST Display Advertising 𝛽3𝑠𝑡 0.0072 6.9830 0.0000*** ST Price Increase 𝛽4𝑠𝑡 -1.1585 -15.5869 0.0000*** ST Price Decrease 𝛽5𝑠𝑡 -2.3657 -33.2643 0.0000*** LT Advertising 𝛽6𝑙𝑡 0.0028 4.1625 0.0000*** LT Feature Advertising 𝛽7𝑙𝑡 0.0126 7.1651 0.0000*** LT Display Advertising 𝛽8𝑙𝑡 0.0095 3.9195 0.0000*** LT Price Increase 𝛽9𝑙𝑡 -0.6452 -7.0422 0.0000*** LT Price Decrease 𝛽10𝑙𝑡 -0.6214 -6.7016 0.0000*** Q1 𝛽11 -0.0198 -3.6569 0.0001*** Q2 𝛽12 -0.0236 -4.2340 0.0000*** Q3 𝛽13 -0.0099 -1.7752 0.0379** Adjustment π -0.4351 -35.1038 0.0000*** * p < 0.10. ** p < 0.05. *** p < 0.01.

Regarding the advertising effects for follower firms we find significant short-term and long-term effects for mass advertising expenditures, feature advertising and display advertising.

The mass advertising expenditures (𝛽1𝑠𝑡_{) have a positive short-term elasticity 0.0007} (p<0.01) on the volume sales of followers, for the long-term (𝛽6𝑙𝑡) we found a positive and significant elasticity of 0.0028 (p<0.01). We expected to have larger mass advertising effects for the long-term, and smaller effects for the short-term. The expectation turned out to be true since the weighted β for the long-term is indicating a larger effect.

Short-term feature adverting (𝛽2𝑠𝑡_{) has a positive significant elasticity of 0.0099 (p<0.01),} the long-term feature advertising (𝛽7𝑙𝑡) has a positive and significant elasticity of 0.0126 (p<0.01). Short-term display advertising (𝛽3𝑠𝑡_{) has a positive and significant elasticity of} 0.0072 (p<0.01) on the volume sales of followers, the long-term display advertising elasticity (𝛽8𝑙𝑡) is positive and significant 0.0095 (p<0.01). For both variables we expected stronger short-term effects compared to the long-term effects. The results of the follower analysis indicate the opposite is true, for both feature advertising and display advertising we found stronger long-term effects.

(28)

(29)

6. DISCUSSION

6.1 Summary

The differences in the effectiveness of marketing mix tools between market leaders and follower firms have not been addressed by prior research. Four years of weekly data from the three largest supermarket chains in the Dutch FMCG-sector gave the opportunity to research the potential differences between market leaders and followers for several product categories. These product categories fall within three product classes: beverages, food and personal care. The research investigates the short-term and long-term effects of the marketing mix tools mass advertising expenditures, feature advertising, display advertising, price increases and price decreases on the volume sales of brands, for both market leaders and followers.

Comparing the advertising frequencies (% of weeks with advertising) for mass advertising expenditures, feature advertising and display advertising between market leaders and followers, we found higher frequencies for the market leaders for all three advertising variables in both the beverages and food classes. In the personal care class we found a counterintuitive result for mass advertising and feature advertising. Followers have a higher frequency of advertising on both variables. For display advertising frequency it is again the market leader that advertises more often and so is in line with the ratio of all the other numbers.

The ECM model (3.1 Error correction model) has been estimated twice, once for the market leader brands and once for the follower brands. For the market leader model we find 8 significant marketing mix parameters. The analysis confirms that for market leaders significant short-term advertising elasticities exist for feature advertising and display advertising (e.g. Srinivasan & Grover, 1992; Papatla, 1996; Mehta et al., 2003). Besides significant long-term advertising elasticities are found for mass advertising expenditures (Sethuraman et al., 2011) and feature advertising (e.g. Mehta et al., 2003). Regarding the pricing effects for market leaders the research proved significant short-term and long-term elasticities for both price increases and price decreases (Greenleaf, 1995; Pauwels, 2007).

(30)

significant short-term and long-term elasticities for both price increases and price decreases (Greenleaf, 1995; Pauwels, 2007).

6.2 Managerial implications

The complex quest for managers to explain the performance differences for the marketing mix tools is an fundamental issue for companies. By describing and elaborating on the effects of the marketing mix tools, this section will guide managers to gain understanding, depending on being the market leader or a follower firm. While some results are consistent with the expected results from the literature study, some results indicated a surprising counterintuitive result.

Mass advertising expenditures

Table 6

Elasticities mass advertising expenditures

Marketing mix tool Market leader Follower Mass advertising Short-term 𝛽1𝑠𝑡 Non-significant 0.0007

Long-term 𝛽6𝑙𝑡 0.0024 0.0028

(31)

therefore feasible for managers of follower firms. Followers still benefit from building brand image, which makes consumers loyal to the brand and will eventually lead to consumers making a purchase decision (Mabkhot et al., 2017; Yoo et al., 2000). Regarding the long-term a slightly larger elasticity for followers than for market leaders is found. The difference is too small to consider as a real advantage and therefore we conclude that the main advantage for followers is gained on the short-term mass advertising effect. In addition, the measured mass advertising elasticities are the smallest among the measured marketing mix tools effects in this research. This is in line with prior research from Allenby and Hanssens (2004) who stated that lower advertising elasticities are common in CPG categories.

Feature advertising

Table 7

Elasticities feature advertising

Marketing mix tool Market leader Follower Feature advertising Short-term 𝛽2𝑠𝑡 0.0087 0.0099

Long-term 𝛽7𝑙𝑡 0.0125 0.0126

(32)

that, although about online ads, even short-term oriented tools like search engine marketing and banner ads have long-term effects that differ per target group. When looking at the differences between market leaders and followers, we found a really small short-term advantage for the followers which is probably not economically significant. Also the long-term cumulative effect of feature advertising is equal for market leaders and followers. We therefore conclude that there is not really an advantage for market leaders or followers and so the feature advertising effectiveness is equal.

Display advertising

Table 8

Elasticities display advertising

Marketing mix tool Market leader Follower Display advertising Short-term 𝛽3𝑠𝑡 0.0123 0.0072

Long-term 𝛽8𝑙𝑡 Non-significant 0.0095

(33)

other. Concluding, when we compare the display advertising elasticity between market leaders and followers we found the market leaders to have a short-term advantage compared to the followers. Regarding the long-term followers experience some small delayed effects, for market leaders those effects do not exist because of the post promotion dip (e.g. Leone, 1987; van Heerde et al., 2000).

Pricing

Table 9

Elasticities pricing3

Marketing mix tool Market leader Follower Price increase Short-term 𝛽4𝑠𝑡 -0.7900 -1.1585

Long-term 𝛽9𝑙𝑡 -0.3695 -0.6452

Price decrease Short-term 𝛽5𝑠𝑡 -1.9837 -2.3657

Long-term 𝛽10𝑙𝑡 -0.3896 -0.6214

Finally, we found some really interesting results regarding the pricing effects between market leaders and followers (table 9). For market leaders and followers, both short-term and long-term price increase and price decrease elasticities have significant effects on the volume sales from the brand. When looking at the short-term elasticities for market leaders and followers, we found that price decreases have a larger positive effect on volume sales than price increases will lead to a loss in sales. If we compare the long-term price increase elasticity with the long-term decrease elasticity, they are (almost) symmetrical for both market leaders and followers, so the cumulative long-term effect for price increase and decrease elasticities cancels out. For both market leaders and followers the price decrease elasticity is stronger on the short-term than the long-term, this means that the increase in the short-term volume sales is followed by a post promotion dip on the long-term. Van Heerde et al. (2000) and Leone (1987) already indicated that consumers tend to accelerate their purchases in response to a promotion, so they buy earlier and/or purchase larger quantities than they would in the absence of the promotion.

If we compare the short-term results of market leaders and followers we see that market leaders suffer less from a price increase, but also experience a smaller price decrease effect.

(34)

So, price reductions are more effective for followers, and market leaders are better able to handle a price increase. For managers of market leader brands this means that they are able to increase their price with loosing less sales compared to the followers who would follow the price increase. In many cases market leaders have already spend a heavy amount on building a positive image of the brand in the consumer’s mind. Brand image creates brand awareness and high awareness prompts the consumer towards making a purchase decision (Yoo et al., 2000). Also, a stronger brand image has a significant and positive effect on the brand loyalty (Mabkhot et al., 2017). So for market leader brands with a higher brand loyalty the consumer keeps loyal regardless of a price increase (to a certain extent). For managers of followers this implies they should not blindly follow a price increase of the market leader brand, since their loss on volume sales is bigger. On the other hand, concerning price decreases follower brands have higher elasticities compared to the market leader. When the prices in a certain market are decreased, this finding is very useful for companies. Managers of follower firms that decrease their price, will gain more sales compared to market leaders who would also decrease their price with the same amount. In this case, managers of market leader brands should think twice about a price decrease since the effect for market leaders is small and follower firms will outperform the market leader brands on the effectiveness of the price decrease.

(35)

6.3 Limitations and future research

This study assessed the differences in the short- and long-term effectiveness of marketing mix tools on the volume sales of brands, between market leaders and follower firms. Although we presented interesting results, this study has a number of limitations which open up future research directions.

First, this research indicated that the short-term mass advertising elasticity for market leader brands is no longer adding any value because the mass advertising effect is already beyond the optimum. In many cases market leaders have already spend a heavy amount on building a positive image of the brand in the consumer’s mind, which creates brand awareness and brand loyalty, that will eventually lead to consumers making a purchase decision (Mabkhot et al., 2017; Yoo et al., 2000). For followers this short-term mass advertising elasticity still is adding value. This result leads to an interesting question for managers and therefore a clear future research direction. Managers of market leaders brands are interested in the amount of money they can save by cutting mass marketing expenditures to the exact optimum. The matching question therefore is: to what extent are market leaders able to drop the mass advertising expenditures, so it is at the exact optimum and no money is wasted? For managers of follower brands the question on how much to increase the mass advertising expenditures to get to the optimum, is an interesting question to address.

Secondly, this research only included advertising and pricing as marketing mix tools. This research therefore adds to a long list of research that investigated the effects of both tools (e.g. Chintagunta et al., 2006; Dorfman & Steiner, 1954; M. C. Gupta & Di Benedetto, 2007; Kadiyali, 1997; Karray & Amin, 2015; Lu et al., 2016). Recent research from Ataman et al. (2010) found higher short-term and long-term sales elasticities for product and distribution, and smaller elasticities for advertising and price decreases. To get a more complete view on the differences in the effectiveness of marketing mix tools between market leaders and followers, this research setting can be replicated with adding product and distribution as important marketing mix tools.

(36)

expect them to have smaller short-term effects, and so perhaps stronger long-term effects. The post promotion dips as described by van Heerde et al. (2000) and Leone (1987) might not hold for these kind of products since it makes no sense to stock them. As a future research direction, this research setting can be replicated with less storable product categories.

(37)

REFERENCES

Ailawadi, K. L., Farris, P. W., & Parry, M. E. (1999). Market share and ROI: Observing the effect of unobserved variables. International Journal of Research in Marketing, 16(1), 17–33. https://doi.org/10.1016/S0167-8116(98)00012-3

Akerlof, G. A. (1970). The Market for “Lemons”: Quality Uncertainty and the Market Mechanism. The Quarterly Journal of Economics, 84(3), 488. https://doi.org/10.2307/1879431

Allenby. G. M.. & Hanssens. D.M. (2004). Advertising response. MSI special report 05-200.

Cambridge. MA: Marketing Science Institute.

Amountzias, C., Dagdeviren, H., & Patokos, T. (2017). Pricing decisions and market power in the uk electricity market: A VECM approach. Energy Policy, 108(June), 467–473. https://doi.org/10.1016/j.enpol.2017.06.016

Andrews, R. L., & Srinivasan, T. C. (1995). Studying Consideration Effects in Empirical Choice Models Using Scanner Panel Data. Journal of Marketing Research, 32(1), 30. https://doi.org/10.2307/3152108

Apelbaum, E., Gerstner, E., & Naik, P. A. (2003). The effects of expert quality evaluations versus brand name on price premiums. Journal of Product & Brand Management, 12(3), 154–165. https://doi.org/10.1108/10610420310476915

Arndt, J., & Olsen, L. (1975). A research note on economies of scale in retailing. The Swedish Journal of Economics, 77(2), 207–221. https://doi.org/10.2307/3438917

Assmus, G., Farley, J. U., & Lehmann, D. R. (1984). How Advertising Affects Sales: Meta-Analysis of Econometric Results. Journal of Marketing Research, 21(1), 65. https://doi.org/10.2307/3151793

Ataman, M. B., Van Heerde, H. J., & Mela, C. F. (2010). The Long-Term Effect of Marketing Strategy on Brand Sales. Journal of Marketing Research, 47(5), 866–882. https://doi.org/10.1509/jmkr.47.5.866

Bijmolt, T. H. A., van Heerde, H. J. van, & Pieters, R. G. M. (2005). New Empirical Generalizations on the Determinants of Price Elasticity. Journal of Marketing Research, 42(2), 141–156. https://doi.org/10.1509/jmkr.42.2.141.62296

(38)

Differences between New and Existing Customers. Journal of Interactive Marketing, 26(3), 155–166. https://doi.org/10.1016/j.intmar.2011.12.001

Buzzell, R. D., Gale, B. T., & Sultan, R. G. M. (1975). Market share: Key to profitability? Harvard Business Review, January(April), 97–106. https://doi.org/10.1108/eb053944 Caves, R. E., & Porter, M. E. (1977). From Entry Barriers To Mobility Barriers: Conjectural

Decisions and Contrived Deterrence To New Competition. Quarterly Journal of Economics, 91(2), 241–261. https://doi.org/10.2307/1885416

Chintagunta, P. K., Kadiyali, V., & Vilcassim, N. J. (2006). Endogeneity and Simultaneity in Competitive Pricing and Advertising: A Logit Demand Analysis. Journal of Business, 79(6), 2761–2787. https://doi.org/10.1086/507998

Coolsbee, A., Levitt, S., & Syverson, C. (2012). Microeconomics, University Of Chicago

publications

Corstjens, M., & Steele, R. (2008). An international empirical analysis of the performance of manufacturers and retailers. Journal of Retailing and Consumer Services, 15(3), 224– 236. https://doi.org/10.1016/j.jretconser.2007.06.004

Debruyne, M., & Reibstein, D. J. (2005). Competitor See, Competitor Do: Incumbent Entry in

New Market Niches. Marketing Science, 24(1), 55–66.

https://doi.org/10.1287/mksc.1040.0064

Derfus, P. J., Maggitti, P. G., Grimm, C. M., Smith, K. G., Pamela, J., Maggitti, P. G., … Queen, R. (2016). THE RED QUEEN EFFECT : COMPETITIVE ACTIONS AND FIRM PERFORMANCE quest Evolutionary, 51(1), 61–80.

Dorfman, R., & Steiner, P. O. (1954). Optimal Advertising and Optimal Quality. The American Economic Review, 44(5), 826–836. https://doi.org/10.2307/1807704

Ehrenberg. Andrew S.C. (1974). "Repetitive Advertising and the Consumer." Journal of

Advertising Research. 14 (2. April). 25-34.

Fader, P. S., & McAlister, L. (1990). An Elimination by Aspects Model of Consumer Response to Promotion Calibrated on UPC Scanner Data. Journal of Marketing Research, 27(3), 322–332. https://doi.org/10.2307/3172589

(39)

Industrial Organization, 46, 63–76. https://doi.org/10.1016/j.ijindorg.2016.01.004

Ferrier, W. J., Smith, K. G., & Grimm, C. M. (1999). The role of competitive action in market share erosion and industry dethronement: A study of industry leaders and challengers. Academy of Management Journal, 42(4), 372–388. https://doi.org/10.2307/257009 Geyskens, I., Gielens, K., & Gijsbrechts, E. (2010). Proliferating Private-Label Portfolios:

How Introducing Economy and Premium Private Labels Influences Brand Choice. Journal of Marketing Research, 47(5), 791–807. https://doi.org/10.1509/jmkr.47.5.791 Gijsbrechts, E. (1993). Prices and pricing research in consumer marketing: Some recent

developments. International Journal of Research in Marketing, 10(2), 115–151. https://doi.org/10.1016/0167-8116(93)90001-F

Gijsenberg, M. J. (2014). Going for gold: Investigating the (non)sense of increased advertising around major sports events. International Journal of Research in Marketing, 31(1), 2–15. https://doi.org/10.1016/j.ijresmar.2013.09.004

Gijsenberg, M. J. (2017). Riding the Waves: Revealing the Impact of Intrayear Category Demand Cycles on Advertising and Pricing Effectiveness. Journal of Marketing Research, 54(2), 171–186. https://doi.org/10.1509/jmr.14.0576

Givon, M., & Horsky, D. (1990). Untangling the Effects of Purchase Reinforcement and Advertising Carryover. Marketing Science, 9(2), 171–187. https://doi.org/10.1287/mksc.9.2.171

Golder, P. N. (2000). Historical Method in Marketing Research with New Evidence on Long-Term Market Share Stability. Journal of Marketing Research, 37(2), 156–172. https://doi.org/10.1509/jmkr.37.2.156.18732

Gotlieb, M. R., Scholl, R. M., Ridout, T. N., Goldstein, K. M., & Shah, D. V. (2017). Cumulative and long-term campaign advertising effects on trust and talk. International Journal of Public Opinion Research, 29(1), 1–22. https://doi.org/10.1093/ijpor/edv047 Greene, W. H. (2003). Econometric analysis (5th ed.)Englewood Cliffs, NJ: Prentice Hall.

Greenleaf, E. A. (1995). The Impact of Reference Price Effects on the Profitability of Price Promotions. Marketing Science. https://doi.org/10.1287/mksc.14.1.82

(40)

Marketing Management, 36(4), 540–548. https://doi.org/10.1016/j.indmarman.2006.02.004

Gupta, S. (1988). Impact of Sales Promotions on When, What, and How Much to Buy. Journal of Marketing Research, 25(4), 342. https://doi.org/10.2307/3172945

Haviv, A. (2015). Does Purchase Without Search Explain Counter-Cyclic Pricing ? Working Paper, Rotman School of Management, University of Toronto, 1–49.

He, X., Prasad, A., & Sethi, S. P. (2009). Cooperative advertising and pricing in a dynamic stochastic supply chain: Feedback stackelberg strategies. Production and Operations Management, 18(1), 78–94. https://doi.org/10.3401/poms.1080.01006

Im, K. S., Pesaran, M. H., & Shin, Y. (2003). Testing for unit roots in heterogeneous panels. Journal of Econometrics, 115(1), 53–74. https://doi.org/10.1016/S0304-4076(03)00092-7

Ingene, C. A. (1984). Scale Economies in American Retailing: A Cross-Industry Comparison. Journal of Macromarketing. https://doi.org/10.1177/027614678400400205

Inman, J. J., Mcalister, L., Hoyer, W. D., Inman, J. J., Mcalister, L., & Hoyer, W. D. (1991). Promotion Signal : Proxy for a Price Cut ? Linked references are available on JSTOR for this article : Promotion Signal : Proxy for a Price Cut ? Journal of Consumer Research, 17(1), 74–81. Retrieved from http://www.jstor.org/stable/2626826

Jackson, S. B., & Wilcox, W. E. (2000). Do Managers Grant Sales Price Reductions to Avoid Losses and Declines in Earnings and Sales? Quarterly Journal of Business and Economics. Retrieved from http://www.jstor.org/stable/40473306

Jacobson, R. (1988). Distinguishing Among Competing Theories of the Market Share Effect.

Journal of Marketing, 52(4), 68–80. Retrieved from

http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=6352828&site=ehost-live&scope=site

Kadirov, D. (2015). Private labels ain’t bona fide! Perceived authenticity and willingness to pay a price premium for national brands over private labels. Journal of Marketing Management, 31(17–18), 1773–1798. https://doi.org/10.1080/0267257X.2015.1031265 Kadiyali, V. (1997). Exchange rate pass-through for strategic pricing and advertising : An

(41)

Kahneman, D., & Tversky, A. (1979). Prospect Theory : An Analysis of Decision under Risk. Econometrica, 47(2), 263–292. https://doi.org/https://doi.org/10.2307/1914185

Kalyanaram, G., & Winer, R. S. (1995). Empirical Generalizations from Reference Price Research. Marketing Science. https://doi.org/10.1287/mksc.14.3.G161

Karray, S., & Amin, S. H. (2015). Cooperative advertising in a supply chain with retail competition. International Journal of Production Research, 53(1), 88–105. https://doi.org/10.1080/00207543.2014.925602

Katila, R., Chen, E. L., & Piezunka, H. (2012). All the right moves: How entrepreneurial firms compete effectively. Strategic Entrepreneurship Journal, 6(2), 116–132. https://doi.org/10.1002/sej.1130

Kotler, P. and Keller, K. L. (2006) Marketing Management, 12th edition. New Jersey:

Prentice Hall

Lamey, L., Deleersnyder, B., Steenkamp, J.-B. E. ., & Dekimpe, M. G. (2012). The Effect of Business-Cycle Fluctuations on Private-Label Share: What Has Marketing Conduct Got to Do with It? Journal of Marketing, 76(1), 1–19. https://doi.org/10.1509/jm.09.0320

Leone, R. P. (1987). Forecasting the Effect of an Environmental Change on Market Performance. International Journal of Forecasting, 3, 463–478.

Levin, A., Lin, C. F., & Chu, C. S. J. (2002). Unit root tests in panel data: Asymptotic and finite-sample properties. Journal of Econometrics, 108(1), 1–24. https://doi.org/10.1016/S0304-4076(01)00098-7

Lu, L., Gou, Q., Tang, W., & Zhang, J. (2016). Joint pricing and advertising strategy with reference price effect. International Journal of Production Research, 54(17), 5250–5270. https://doi.org/10.1080/00207543.2016.1165878

Mabkhot, H. A., Shaari, H., & Salleh, S. M. (2017). The Influence of Brand Image and Brand Personality on Brand Loyalty, Mediating by Brand Trust: An Empirical Study. Jurnal Pengurusan, 50, 18.

Mehta, N., Rajiv, S., & Srinivasan, K. (2003). Price Uncertainty and Consumer Search: A Structural Model of Consideration Set Formation. Marketing Science, 22(1), 58–84. https://doi.org/10.1287/mksc.22.1.58.12849

(42)

Advertising on Consumer Brand Choice. Source Journal of Marketing Research, 34(2), 248–261. https://doi.org/10.2307/3151862

Merton, R. K. (1968). The Matthew Effect in Science: The reward and communication systems of science are considered. Science, 159(3810), 56–63. https://doi.org/10.1126/science.159.3810.56

Mukhopadhyaya, J. N., Roy, M., & Raychudhuri, A. (2012). Determinants of Market Share, Profitability and Market Power at the Firm Level in the Cement Industry of India. Vilakshan: The XIMB Journal of Management, 9(1), 95–114. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&db=bah&AN=74258378&site=ehost -live

Palda, K. S. (1965). The measurement of cumulative advertising effects. Journal of Business.

Papatla, P. (1996). A Multiplicative Fixed-Effects Model of Consumer Choice. Marketing Science, 15(3), 243–261. https://doi.org/10.1287/mksc.15.3.243

Pauwels, K. (2007). How retailer and competitor decisions drive the long-term effectiveness of manufacturer promotions for fast moving consumer goods. Journal of Retailing, 83(3), 297–308. https://doi.org/10.1016/j.jretai.2006.03.001

Posen, H. E., Lee, J., & Yi, S. (2013). THE POWER OF IMPERFECT IMITATION. Strategic Management Journal, 34, 149–164. https://doi.org/10.1002/smj

Rao. Ambar G. and Peter B. Miller (1975). "Advertising/Sales Response Functions." Journal

of Advertising Research. 15 (No. 2. April). 7-15.

Rhoades, S. A. (1985). Market share as a source of market power: Implications and some evidence. Journal of Economics and Business, 37(4), 343–363. https://doi.org/10.1016/0148-6195(85)90027-X

Roehm, M. L., Pullins, E. B., & Roehm, H. a. (2002). Designing Loyalty-Building Programs for Packaged Goods Brands. Journal of Marketing Research, 39(May), 202–213. https://doi.org/10.1509/jmkr.39.2.202.19085

Rosenthal, R. (1991). Meta-analytic procedures for social research. Newbury Park, CA: Sage

Publications

(43)

8(4), 340–351. https://doi.org/10.1108/10610429910284319

Sethuraman, R., Tellis, G. J., & Briesch, R. A. (2011). How Well Does Advertising Work? Generalizations from Meta-Analysis of Brand Advertising Elasticities. Journal of Marketing Research, 48(3), 457–471. https://doi.org/10.1509/jmkr.48.3.457

Sharp, B., Riebe, E., Dawes, J., & Danenberg, N. (2002). A marketing economy of scale-big brands lose less of their customer base than small brands. Marketing Bulletin, (1992), 1– 8.

Smith, K. G., Ferrier, W. J., & Grimm, C. M. (2001). King of the hill: Dethroning the industry

leader. Academy of Management Executive, 15(2), 59–70.

https://doi.org/10.5465/AME.2001.4614896

Srinivasan, V., & Grover, R. (1992). Evaluating the Multiple Effects of Retail Promotions on Brand Loyal and Brand Switching Segments. Marketing Research, XXIX(February), 76– 89. https://doi.org/10.2307/3172494

Szymanski, D. M., Bharadwaj, S. G., & Varadarajan, P. R. (1993). An analysis of the market share-profitability relationship. Journal of Marketing, 57(3), 1. https://doi.org/10.2307/1251851

van Heerde, H. J., Gijsenberg, M. J., Dekimpe, M. G., & Steenkamp, J.-B. E. . (2013). Price and Advertising Effectiveness over the Business Cycle. Journal of Marketing Research, 50(2), 177–193. https://doi.org/10.1509/jmr.10.0414

van Heerde, H. J., Leeflang, P. S. H., & Wittink, D. R. (2000). The Estimation of Pre- and Postpromotion Dips with Store-Level Scanner Data. Journal of Marketing Research, 37(3), 383–395. https://doi.org/10.1509/jmkr.37.3.383.18782

William G. Shepherd (1970). Market Power and Economic Welfare: An Introduction..

Random House.

Wood, L., & Poltrack, D. (2015). Measuring the long-term effects of television advertising: Nielsen-CBS study uses single-source data to reassess the “two-times” multiplier. Journal of Advertising Research, 55(2), 123–131. https://doi.org/10.2501/JAR-55-2-123-131

(44)

https://doi.org/10.1177/0092070300282002

Yuen, E. F. T., & Chan, S. S. L. (2010). The effect of retail service quality and product quality on customer loyalty. Journal of Database Marketing & Customer Strategy Management, 17(3–4), 222–240. https://doi.org/10.1057/dbm.2010.13