Long-Term Impact of Marketing

A Compendium

Dominique M Hanssens

• Publisher: WSPC
• Year: 2018
• ISBN:
  9789813229815
• Topic: Commerce
• Subtopic: Études marketing

-->

Long-Term Impact of Marketing: A Compendium summarizes four decades of marketing science research by Professor Hanssens and coauthors. The book focuses on the topic of inferring long-term marketing impact on business performance from real-world data. It presents time-series analytic methods to measure the short- and long-term effects of marketing on business performance. As marketing data increase in quantity and quality, the application of the principles in the book are becoming more relevant and important.

--> Contents:

• Dedication
• About the Author
• Introduction (Dominique M Hanssens)
• Market Response, Competitive Behavior, and Time-Series Analysis (Dominique M Hanssens)
• Modeling Asymmetric Competition (Gregory S Carpenter, Lee G Cooper, Dominique M Hanssens and David F Midgley)
• Measuring the Long-Term Effects of Public Policy: The Case of Narcotics Use and Property Crime (Keiko Powers, Dominique M Hanssens, Yih-Ing Hser and M Douglas Anglin)
• The Persistence of Marketing Effects on Sales (Marnik G Dekimpe and Dominique M Hanssens)
• Sustained Spending and Persistent Response: A New Look at Long-Term Marketing Profitability (Marnik G Dekimpe and Dominique M Hanssens)
• The Category-Demand Effects of Price Promotions (Vincent R Nijs, Marnik G Dekimpe, Jan-Benedict E M Steenkamps and Dominique M Hanssens)
• The Long-Term Effects of Price Promotions on Category Incidence, Brand Choice, and Purchase Quantity (Koen Pauwels, Dominique M Hanssens and S Siddarth)
• New Products, Sales Promotions, and Firm Value: The Case of the Automobile Industry (Koen Pauwels, Jorge Silva-Risso, Shuba Srinivasan and Dominique M Hanssens)
• Competitive Reactions to Advertising and Promotion Attacks (Jan-Benedict E M Steenkamp, Vincent R Nijs, Dominique M Hanssens and Marnik G Dekimpe
• Performance Regimes and Marketing Policy Shifts (Koen Pauwels and Dominique M Hanssens)
• The Impact of Marketing-Induced Versus Word-of-Mouth Customer Acquisition on Customer Equity (Julian Villanueva, Shijin Yoo, and Dominique M Hanssens)
• Marketing and Firm Value: Metrics, Methods, Findings, and Future Directions (Shuba Srinivasan and Dominique M Hanssens)
• The Direct and Indirect Effects of Advertising Spending on Firm Value (Amit Joshi and Dominique M Hanssens)
• Consumer Attitude Metrics for Guiding Marketing Mix Decisions (Dominique M Hanssens, Koen H Pauwels, Shuba Srinivasan, Marc Vanhuele, and Gokhan Yildirim)
• Performance Growth and Opportunistic Marketing Spending (Dominique M Hanssens, Fang Wang and Xiao-Ping Zhang)

-->
--> Readership: Graduate students and researchers in the fields of marketing and econometrics. -->
Keywords:Marketing Impact;Long Term Marketing Effects;Marketing Science;Time-series ModelsReview:

"Rarely does a scholar consistently produce impactful research for over four decades. Dominique Hanssens is one of those rare scholars whose legendary work on the long-term impact of marketing and public policy is beautifully captured in this book."

Sunil Gupta
Edward W Carter Professor of Business Administration
Harvard Business School

"This compendium affords a long-term perspective on long-term marketing effects. Collectively, the papers in this volume span four decades of research and convincingly demonstrate that marketing is an investment, not an expense. Dominique Hanssens's pioneering contributions are a must-read for those interested in marketing dynamics."

Carl Mela
T. Austin Finch Foundation Professor
Duke University Fuqua School of Business

"Marketing strategy and accompanying actions result in long-term, forward-looking, multi-period benefits. The brand and market power of incumbents is felt across industries ranging from consumer goods to industrial products to technology-based markets. Unfortunately, we tend to measure the impact of marketing investments in retrospective, short-term frameworks that grossly underestimate their contributions to the firm in terms of profitability and growth. Dominique Hanssens is one of the few who gets it right. He and his coauthors document long-term effects of market-based assets (brand, customers, distribution partners), capabilities and actions (advertising, promotions) on category growth, brand share and protection from price competition. For great insights, look no further!"

Rajendra Srivastava
Dean and Novartis Professor of Marketing Strategy and Innovation
Indian School of Business
Key Features:

• Summarizes general topics of inferring long-term marketing impact on business performance from real-world data
• Presents time-series analytic methods to measure the short- and long-term effects of marketing on business performance

Chapter 1

Market Response, Competitive Behavior, and Time-Series Analysis^∗

Dominique M. Hanssens

University of California, Los Angeles

The author’s principal objective is to present a framework for market analysis which specifically models primary demand, competitive reaction, and feedback effects of the market variables. The approach is an extension of earlier work by Clarke and by Lambin, Naert, and Bultez on the relationship among the elasticities of the marketing variables. The author develops this framework and formulates an approach for empirical applications based on principles of time-series analysis. In particular, Granger’s well-known causality definition is used in conjunction with Box–Jenkins) analysis to find the nonzero elements in the marketing model. These principles are applied empirically to the case of a city pair of the U.S. domestic air travel market, where three major airlines compete on the basis of flight scheduling and advertising. The analysis reveals that flight scheduling has a market-expensive or a competitive effect, depending on the competitor, and that advertising does not have a significant impact on performance. In addition, several patterns of competitive reactions are found. The author offers observations on the theoretical and empirical aspects of this approach to marketing model building.

In an increasingly complex and risky business environment, the development of quantitative models of markets is a difficult but rewarding task. In the past decade, such models have been of strong interest among academicians, as shown by the large number of articles published in this field. Also, models of markets are gaining popularity in industry, where they are used for forecasting as well as evaluation of market plans (e.g., Stryker, 1978; The Wall Street Journal, 1977).

The structural relationships that are part of a market mechanism can he categorized into sales response effects, competitive reactions, and feedback effects. Though the importance of these types of market relationships has certainly been recognized, very few empirical studies have included all of them simultaneously. For example, numerous studies have analyzed the marketing mix effects on sales of single products, isolated from the market in which the products operate. The main reason for such an approach is probably the lack of good data, because most market research is done in the profit sector where data are typically scarce and/or proprietary. Unfortunately, failure to include the relationships among certain variables may result in severe model misspecification and, ultimately, unreliable research findings.

The main purpose of this chapter is to propose a systematic modeling of the various relationships that characterize markets. To recognize the dynamic nature of market variables, interest is focused on longitudinal data sets, which are the most common and the most useful, because marketing planning is dynamic. The first part of the chapter is marketing theoretical. After a brief review of the literature, the theoretical models proposed by Clarke (1973), Lambin, Naert, and Bultez (1975), and Schultz and Wittink (1976) are examined. The findings of those researchers can be integrated and extended into a full-scale dynamic model of a market, i.e., a model which incorporates all the potential structural relationships previously described. The complexity or such a model necessitates the description of an empirical approach for using the model. This is done by application of recent developments in univariate and multiple time-series analysis in conjunction with econometrics. Specifically, the issues of discovering relationships and ruling out spurious associations are discussed.

The second part of the chapter is an empirical illustration of the theoretical ideas, based on a submarket of the U.S. domestic air travel industry. The model considers each of the three competitors in this market separately and includes the two most important marketing instruments for the period studied, number of flights and advertising expenditures.

1.Prior research

Some important contributions in the evolution of modeling sales response, feedback, and competitive behavior are reviewed hereafter. Parsons and Schultz (1976) and Naert and Leeflang (1978) provide a more thorough discussion of the literature.

1.1.Sales response elements

Marketing model builders have devoted most of their efforts toward developing functions of sales response to the marketing mix. The earlier applications typically considered only one product and one marketing variable at a time, for example, Palda’s (1964) regression models on Lydia Pinkham sales and advertising, which stimulated a subsequent research stream on the dynamic or carryover effects of advertising (e.g., Bass and Clarke, 1972; Houston and Weiss, 1975), At about the same time, more studies appeared which considered two or more marketing mix variables, for example, Lambin’s (1970) study on small electric appliances and Little’s (1975) work on a packaged food product. Also, some researchers added competitive marketing efforts as explanatory variables in response functions, for example Sexton (1970) and Urban (1969) on frequently purchased branded goods. Finally, there was a definite trend toward analyzing markets rather than single product sales, as exemplified by the more complex models by Beckwith (1972) and McCann (1974), also on frequently purchased branded goods.

One issue in sales response modeling which has not received sufficient attention is the distinction between market-expansive (i.e., primary demand) effects and competitive (i.e., secondary demand) effects of the marketing variables. As Parsons and Schultz (1976) point out for the case of advertising, most studies really have not been designed to test for the presence of primary demand effects. Yet the question is very important because in the absence of market expansion the marketing efforts of the competitors may cancel each other out. For example, a study by Metwally (1978) indicates “. . . that advertising in a number of Australian industries is self-cancelling and escalating.” Fortunately, two theoretical developments are very useful for the study of expansive versus competitive effects: (1) a simple mathematical equality, due to Clarke (1973), which states that the elasticity of a marketing instrument on sales (n_S) equals the primary demand elasticity (n_PD) plus the market share elasticity (n_m), and (2) a set of theoretical conditions for the existence of primary demand, primary sales, competitive, and mixed effects of advertising derived by Schultz and Wittink’ (1976). These contributions are used in the model development hereafter.

1.2.Feedback elements

The possible presence of a feedback relationship between sales and the marketing mix variables has not been investigated thoroughly, in spite of an early warning by Quandt (1964), Bass and Parsons (1969) used predictive testing and simultaneous equations to include the effects of past sales on future advertising budgets in a model of the cigarette industry. Similar efforts were made by Schultz (1971) and Wildt (1974), but are encompassed under the more general case of endogenous marketing decision variables. In terms of model estimation, the modeling of feedback is necessary only if there is a true simultaneous relationship. Because most marketing models are built on data for relatively short time intervals such as months or quarters they are more likely to be recursive in sales and, say, advertising. In such cases, failure to model feedback should not affect the reliability of the market response coefficients.

1.3.Competitive behavior elements

The explicit modeling of competitive behavior is also fairly rare in market model building, primarily because data on competitive marketing expenditures are very difficult to obtain. Some empirical examples are Lambin’s (1970) study of a consumer durable, which includes reaction functions for advertising, and Schultz’ (1971) model for air travel which considers flight share and advertising share equations. Perhaps the most complete empirical study is Wildt’s (1974) which models competition on the basis of advertising, price, promotion, and new products. These and other studies confirm that competitive reactions are usually very strong. A theoretical contribution by Lambin, Naert, and Bultez (1975) also deserves close attention. The objective of these authors was to generalize the Dorfman–Steiner theorem to cases of oligopoly with competitive reactions and market expansion, for which they derived the following fundamental relationship:

where:

Eq,u

=

vector of total sale...