Hunted hunter: the role of competitive comparison in product survival

Open Access
07-03-2025
Original Paper

Published in:

Authors: Fernando Campayo-Sanchez; Francisco Mas-Ruiz; Juan Luis Nicolau
Published in: Review of Managerial Science | Issue 12/2025

Activate our intelligent search to find suitable subject content or patents.

search-config

AI-assisted search

Patentsearch

Off

Abstract

The article 'Hunted hunter: the role of competitive comparison in product survival' delves into how competitive comparisons influence product survival in the automotive industry. It introduces the concept of competitive comparison, defined as a focal firm comparing its product with a rival's product, and argues that this comparison can increase the survival of the focal firm's product. The study uses a longitudinal empirical model to test this hypothesis, finding that competitive comparisons lead to subsequent actions by the focal firm, such as price changes and increased advertising investment. These actions, in turn, enhance the product's survival. The research challenges the prevailing view that competition decreases product survival and introduces the Awareness-Motivation-Capability (AMC) framework to explain the counterintuitive findings. The study highlights the importance of understanding competitive dynamics at the product level rather than assuming all products in a segment are de facto competitors. The findings have significant implications for marketing and product management strategies, emphasizing the need to consider competitive comparisons as a strategic tool rather than a threat.

AI Generated

This summary of the content was generated with the help of AI.

Extras
Notes

Supplementary file1 (DOCX 274 KB)

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s11846-025-00870-5.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

1 Introduction

The commercial longevity of an organization’s product reflects the evolutionary aptitude of a product in the market and is a standard variable to measure its performance (Wang and Chen 2018). Although product survival has been studied from multiple theoretical perspectives, none have captured the attention of academics as much as competition (Carroll et al. 2010).

To date, the literature that focuses on the relationship between a firm’s product commercial longevity and its rivalry with the products of other companies predominantly analyzes the degree of competition in terms of segment crowdedness (see Greenstein and Wade 1998; Ruebeck 2002, 2005; Requena-Silvente and Walker 2005, 2009; de Figueiredo and Kyle 2006; Barnett et al. 2022). Researchers argue that competition decreases the life of the product because it reduces sales and profits over time, thereby anticipating its exit (e.g., Barroso et al. 2016; de Figueiredo and Kyle 2006; Wang 2019). However, by considering all products traded in the same segment as direct competitors, researchers cannot address the possible asymmetry of competitive perceptions and behaviors at the product level for each dyad of firms (Chen et al. 2007). This asymmetry assumes situations where Company “X” may consider that its product “A” competes with Company “Y’s” product “B;” however, the latter does not consider its product “B” to compete with Company “X’s” product “A.”

To overcome this problem that derives from considering all products traded in the same segment as direct competitors, we focus on the concept of competitive comparison, which is defined by Kim and Tsai as “the comparison between a focal firm and a target company in a given product market” (2012, p. 115). This tool is ideal for analyzing the competition for two reasons: (1) the comparison can occur in any of the multiple dimensions that define the product space, and (2) it is in tune with the competitive dynamics literature because it admits the asymmetric nature of the subjective evaluation of competitors. To date, Kim and Tsai (2012) have argued that company “X” publicly compares its product with the product of firm “Y” because it seeks to improve its image in the eyes of buyers. Building on this premise, we argue that this circumstance will encourage “X” to take action against the said target to confirm its comparison, which should reduce the commercial longevity of “Y’s” product in the market. The work of Kim and Tsai (2012), which employs a cross-sectional perspective over a year, finds that the competitive comparison of a focal firm “X” with a target firm “Y” increases the former firm’s sales relative to the latter firm’s sales. However, two limitations require emphasis. First, it does not use a longitudinal database that accounts for changes in competitive comparisons between firms resulting from the exchange of actions and responses. Second, it does not consider the potential reaction from the company whose product has been identified as a comparison target.

Our article seeks to address these deficiencies by developing a longitudinal integrative model that answers the following research questions: How will a rival’s competitive comparison with a focal firm’s product influence the subsequent actions carried out by the latter company? How will this situation affect the survival of its product? Kim and Tsai (2012) postulated that engaging in comparative attacks against competitors’ products leads to increased sales for the firm initiating these competitive moves, which should in turn result in extended commercial longevity for the products marketed by that firm. Paradoxically, the thesis we defend suggests the opposite: a competitive comparison of a rival’s product with a focal firm’s product will enhance the survival of the focal firm’s product. Based on the awareness–motivation–capability (AMC) framework (Chen 1996), we argue that knowledge of this comparative information disseminated by the rival will provide the focal firm with awareness and motivation to act, prolonging the permanence of its product in the market.

In short, this work adopts a new approach to understand the effect of competition on the commercial longevity of the product, which represents an advance for the research on competitive dynamics. Following these ideas, we carried out a longitudinal empirical study (which applies a binomial logit model with random coefficients, a linear regression model with random intercepts, and a Cox proportional-hazards model) on practically all the vehicles commercialized in the Spanish car market between 2008 and 2017.

2 Theory and research hypotheses

2.1 Competitive comparison with the focal firm’s product

One of the most extensively studied issues in the product survival literature is competition (see Table 1). To date, the dominant research approach argues, according to the population ecology of organizations (Hannan and Freeman 1977) the following: when the density of a segment increases, the products marketed by companies face greater competitive pressure. Two consequences arise from this situation: (1) the lack of attention on the part of consumers toward less competitive products, which inevitably leads to their exit (e.g., Barroso et al. 2016; Khessina and Carroll 2008) and (2) the fall in prices and profit margins per unit sold, which encourages companies to withdraw their products due to the loss of profitability (de Figueiredo and Kyle 2006; Wang 2019; Zuckerman 2000). Based on this argumentation, a negative relationship between market density and product survival is found empirically by Greenstein and Wade (1998), Ruebeck (2002, 2005), Requena-Silvente and Walker (2005, 2009), de Figueiredo and Kyle (2006), and Barnett et al. (2022).

Table 1

Overview of research on the relationship between competition and product survival

Authors (Year)	Method	Industry	Competition measure	Findings	Competition-product survival relationship
Greenstein and Wade (1998)	Empirical (secondary)	Commercial mainframe computer	Number of competitors (systems) in a niche	Density in a niche increases the product’s failure rate	Negative
			Dummy variable that captures if the firm introduces another newer product in an existing system’s niche	Cannibalization of a system increases the product’s failure rate	Negative
Ruebeck (2002; 2005)	Empirical (secondary)	Hard drives	Number of rivals’ competing drives	Inter-firm competition is insignificant	–
			Number of the firm’s own drives that compete	Intra-firm competition influences positively product exit	Negative
Requena-Silvente and Walker (2005; 2009)	Empirical (secondary)	Automobile	Number of rival models in the segment	Inter-firm competition increases the product’s failure rate in all segments by predominating in luxury/sport cars	Negative
			Number of own models in the same segment	Intra-firm competition increases the product’s failure rate in all segments, by predominating in small and large family cars	Negative
de Figueiredo and Kyle (2006)	Empirical (secondary)	Laser printer	Number of models, and models in the same niche	Competition increases a product’s failure rate	Negative
Talay and Townsend (2015)	Empirical (secondary)	Automobile	Product’s recent exposure time to all rivals in the segment	A greater duration of recent competitive experience improves the product’s survival	Positive
			Product’s exposure time to all rivals in the distant past in the segment	A greater duration of competitive experience in distant past decreases the product’s survival	Negative
Barnett et al. (2022)	Empirical (secondary)	Popular music	Number of popularity-ranked rival K-pop songs	Competition from rival songs intensify the brevity of success of the focal song	Negative

An alternative study by Talay and Townsend (2015) adopts an evolutionary perspective (Barnett and Hansen 1996) to propose that a longer duration of recent competitive experience of a product with all segment rivals (measured by the time a product was recently exposed to this competition) induces greater survival of that product, while a longer duration of competitive experience with rivals in the distant past reduces its survival. The underlying logic is based on the reciprocal sequence of competitive action and reaction, known as the “Red Queen” effect in evolutionary biology, which can lead to improved product performance and consequently, enhanced survival chances. By contrast, Talay and Townsend (2015) argue that recent competitive experiences bring information to the market that allows the company to update its practices and increase the chances of product survival; however, distant (historical) experiences become dispersed and less informative from a competitive standpoint, leading the company to retain outdated practices and thus reducing the chances of product survival.

Specifically, most of this research (see Table 1) has focused on the automobile industry, highlighting the work of Requena-Silvente and Walker (2005, 2009), whose findings show a negative relationship between car model density in a segment and model survival; and that of Talay and Townsend (2015), whose results show a positive (negative) relationship between a model’s recent (distant) competitive exposure to its rivals in the segment and the survival of the model. Finally, we have not identified any research on the competition-product survival relationship in the Spanish auto industry.

A sharp analysis of these studies would help us easily deduce the following: they all start from the assumption that the products of a segment are de facto competitors. However, given that rivalry is a relational and subjective phenomenon (Kilduff et al. 2010), this approach suffers from a serious background imprecision that could give rise to false reasonings by not admitting the existence of competitive asymmetries (Chen 1996). To correct this deficiency, we will start from Kim and Tsai’s notion of competitive comparison (see above).Their empirical work focuses on the firm-dyad level; however, to date, we have not identified any study that has concentrated on competitive comparison at the product level, despite of being an interesting theoretical question in the product demography literature (see Carroll et al. 2010); this theoretical framework suggests the influence of competition at the product level on product survival. Alternatively, in this study, we use the product as a unit of analysis.

Notably, remarkable achievements in the field of strategy have occurred at an excessively high level of aggregation (as the authors focus primarily on firms), which masks the inherent structure of competition at the product level (Bromiley et al. 2002; Gelei and Dobos 2024). Those arbitrary boundaries often have tenuous links to actual competitive interactions (which most often occur at the product level). By moving away from the mainstream, we attempt to dodge these simplifications rooted in the competitive dynamics literature. Thus, we bet on a more realistic perspective on how to identify rival products in the marketplace.

Car companies often post their own competitive comparisons on their websites. This information, which is public and official, is provided on its own initiative by the brands in such a way that the listed vehicles are identified on the basis of their own subjective evaluations. We keep in mind that car manufacturers’ websites are essentially marketing tools. Therefore, when a brand identifies its product’s rivals, it will only point to those products against which it wants potential consumers to compare its product.

Logically, an organization will limit itself to listing as a competitive priority of those products from other companies with which it considers it can compete from its own subjective perspective—either in terms of price or visual level, mechanical or equipment. Otherwise, its product would lose in the comparison, which would imply an implicit recognition of the superiority of rivals’ products over potential consumers.

We continue to ask the following question: does an incentive exist that induces firms to make competitive comparisons? Several authors have confirmed that the dissemination of information at the product level by organizations will allow them to influence the attitude that consumers have toward the brand (e.g., Aaker 1996; Keller 1998). On the basis of this approach, Kim and Tsai (2012) postulated that an organization’s sales will more largely increase than its competitor’s sales when they compare their products with those of more valued companies (self-affirmed comparisons) or when they avoid comparing them with those of others that are worse considered (self-discarded comparisons). In the first case, the firm seeks to delight in reflected glory to improve its image (Elsbach and Kramer 1996); in the second, the company attempts to cut off the reflected failure by dissociating itself from an organization of lesser status to safeguard its reputation (Snyder et al. 1986).

A firm targets another company’s product because it considers it an important rival that poses a competitive threat to the extent that it has invested time and resources to carry out this comparison. Thus, Kim and Tsai (2012) argued that the organization will not only limit itself to identifying another company’s product as a competitive priority; most likely, this circumstance will instead motivate it to develop another type of action against the said stated target to confirm its comparison. For example, Lexus compared the BMW i3 with its CT 200 h (hybrid) model in an advertisement in 2014; it indicated that the autonomy of the German vehicle was insufficient to reach Las Vegas because it had to recharge every 200 km. In short, the competitive comparison is not only a movement aimed at strengthening the image of the firms’ products in the eyes of buyers but could also become a powerful spur that impels organizations to take action against some of their main competitors’ products.

However, the work of Kim and Tsai (2012) contains two important limitations. First, it shows the impossibility of undertaking a dynamic analysis of competition that considers the change in the competitive comparisons of firms due to the exchange of actions and responses (Chen et al. 2007). The comparison targets of the companies can vary over time depending on the competitive movements and counterattacks of the organizations, potentially directly affecting the survival of the product. This gap is precisely what triggers the second limitation: that the authors ignore the reaction from the company whose product has been considered a competitive priority from the subjective perspective of another organization.

Hence, although firms can use their rivals to try to be successful in the market, in this research, we used longitudinal data on competitive comparisons to study the influence of this type of actions on the commercial longevity of products that have been identified as competitors. Consequently, when analyzing the competition, we will not focus on which are the rivals of a firm’s product but will explore whether that product is seen by another organization as a competitor to any of the products that are part of its portfolio. This reference point is extremely important in our study because it will allow us to capture competitive asymmetries. The underlying logic is that all companies are going to compare each of their products with other products. However, some products might not be perceived as rivals (our approach) by the rest of the players in the industry.

We assume that when a focal firm’s product is in crosshairs of another company, it will stay in the market longer. However, what is the argument that serves as the basis to explain this relationship? We maintain that when a product is identified as a comparison target, the firm that markets it responds by instinct of conservation; that is, it reacts automatically with a defensive aim against the escalation of movements that other organizations direct against its product. In the next section, we will turn to the AMC framework (Chen 1996; Chen et al. 2007) to study this question in detail.

2.2 AMC framework of competitive dynamics research

The competitive dynamics literature conceptualizes rivalry between firms as the successive exchange of actions and responses (Chen et al. 2007; Kang et al. 2021; Nicolau-Gonzálbez and Ruiz-Moreno 2014). In dyadic competitive relationships, companies are highly interdependent in the sense that the failure of one serves as a bridge to the success of the other and vice versa (Chen et al. 2017). In this way, any movement directed against the product of another organization will be compromising its position in the market (Rindova et al. 2004).

From this perspective, researchers have identified the existence of three antecedents of response to an external competitive threat: awareness, motivation, and capability (Chen 1996). A company will not be able to react against the escalating movements of its rivals, unless it is aware of the attacks and motivated and sufficiently capable of acting on it (Chen and Miller 2015; Downing et al. 2019; Yu and Cannella 2007).

Investigating competitive dynamics, especially the AMC framework, offers a particularly useful perspective for examining how competitive comparison with a focal firm’s product can influence the actions that this company will take—whether it be restylings, price changes, or aggressive advertising campaigns. We advocate that competitive comparison will provide the focal firm with awareness and motivation to act, ultimately helping the company’s product achieve greater commercial longevity.

2.2.1 Awareness

This behavioral driver not only precedes motivation and the capability to act but also directs firms’ attention and intelligence-gathering efforts before rivalry emerges (Downing et al. 2019; Luoma et al. 2022). In the absence of awareness, even sufficiently motivated and capable organizations will not be able to spot competitive signals (Withers et al. 2018) and will have trouble defending themselves against attacks from other companies (Guo et al. 2017). A meager or erroneous conscience makes friends of enemies indiscernible (Sytch and Tatarynowicz 2014; Zakrzewska-Bielawska et al. 2022), precludes identifying and prioritizing threats, prevents firms from channeling their motivation and ability to counterattack, and hinders competing against rivals (Chen et al. 2007).

We advocate that when company “X” identifies firm Y’s product as a comparison target, the latter will develop an acute awareness of the former’s perception. Competitive comparisons are movements that organizations use to strengthen and improve the image of their products in the eyes of buyers (Kim and Tsai 2012). Given that these actions can only be effective if they are known to consumers, “X” will publicly disseminate its comparisons through its website. This circumstance is precisely what allows “Y” to be aware that “X” perceives its product as a rival. Furthermore, the fact that the latter organization has used time and resources to carry out this comparison clearly shows that it considers that the product of the former company poses a competitive threat (Kim and Tsai 2012). Therefore, “X” has a high probability of being motivated to pursue other measures with the goal of strengthening the position that its product holds in the market. Ultimately, the public nature of the competitive comparison would enable “Y” to focus on “X” as an important competitor worth monitoring to avoid being surprised by its actions.

2.2.2 Motivation

The idea that any advantage achieved by a firm is ephemeral because it can be pulverized by attacks from other companies has been a tacitly assumed premise in the competitive dynamics literature (Chen et al. 2010; Chen and McMillan 1992). Organizations that navigate in environments characterized by uncertainty and volatility must distinguish the competitive signals emitted by their rivals and take advantage of this information to design actions that help them defend and improve their position in the market (Calderon-Monge and Ribeiro-Soriano 2024; Chen et al. 2017; Rojas-Córdova et al. 2023). In this regard, researchers have been theorizing that the most aggressive firms may end up being more profitable because they have a greater probability of hindering the movements of their rivals (Andrevski and Ferrier 2019; Chen and Hambrick 1995). This perspective is consistent with the argument made by D’Aveni (1994) that in a hypercompetitive environment, organizations are forced to adopt a belligerent attitude in the hope of creating a series of temporary advantages.

Evolutionary and ecological theories that delve into the Red Queen hypothesis also portray how entities dynamically interact and coevolve with each other (Barnett and McKendrick 2004; Talay and Townsend 2015; Talay et al. 2014). This system can be seen as a contest in which the performance and viability of an organization depend on whether it matches or exceeds the actions of its rivals in the market (Derfus et al. 2008). Consequently, when firm “A” takes action, these movements can improve its competitive position at the expense of other companies. However, these actions could also spur these rivals to fight back, negatively affecting the performance of firm “A.” This theoretical approach is in line with the argument adduced by Schumpeter (1976) that a process of “creative destruction” occurs when companies take actions to obtain an advantage in the market that can be eroded by competitive movements from their adversaries. Therefore, each organization is constrained by others to participate in this dynamic in such a way that they all end up running as fast as they can to maintain their relative position with respect to their competitors (Wiggins and Ruefli 2005).

In sum, an organization cannot remain lethargic when other companies attempt to knock it out. When an organization compares its product with the product of a focal firm, it not only seeks to improve the image of its product in the market; most likely, it will be encouraged to develop other types of actions against the said target (Kim and Tsai 2012). Therefore, knowing this comparison is the factor that motivates the focal firm to develop its own actions to win the competitive race and not succumb to the attacks of more aggressive rivals who seek to undermine its status quo (Ferrier et al. 1999). Therefore, the following hypothesis is proposed:

Hypothesis 1

A focal firm is more likely to make a subsequent competitive action when it is compared competitively with the product of another company.

2.3 Influence of the competitive comparison on the survival of the focal firm’s product

When the product of a firm is compared by another organization with any of the products that are part of its portfolio, the first will be forced to fight back. We maintain that the successive development of actions by the focal firm is the mechanism that explains that its product has greater commercial longevity.

In markets where products are complex (e.g., automobiles), buyers cannot know in detail all of the organizations’ products. Consequently, when a brand makes a competitive comparison with another company’s product, it will not only limit itself to identifying it but will also thoroughly specify the specific dimensions―be it the guarantee, the price, or whatever other mechanical, technical, or equipment-related characteristics—which should tip the consumer’s balance in favor of its model (Kim and Tsai 2012). However, this circumstance renders competitive comparison a double-edged sword. To improve its reputation, a firm needs to disseminate valuable information that can be used by the organization whose product has been identified as a comparison target. In our opinion, this knowledge will allow the latter company to respond to the movements of the former with more effective attacks, which would, as it were, cast an image similar to that of the hunted hunter. Therefore, the following hypothesis is proposed:

Hypothesis 2

The survival of the product of a focal firm increases when it is compared competitively with the product of another company.

3 Methodology

To test Hypothesis 1, we analyze the effect of competitive comparison on two strategic actions: a price change and a change in advertising investment.

Considering that price change is measured through a dichotomous variable, we use a binomial logit model with random coefficients for panel data. This approach is intended to control for unobserved heterogeneity that arises from the presence of unmeasured covariates; i.e., by using this methodology, we assume the possible existence of unobserved factors at the product level that cause the estimated parameters to have different impacts on each of the cars in our sample (a very realistic assumption in practice). The empirical model is as follows:

$${Price}_{nt}^{*}={\beta }_{1n}{RIVDENS}_{nt-1}+{\beta }_{2n}{{Controls}_{Product}}_{nt-1}+{\beta }_{3n}{{Controls}_{Firm}}_{nt-1}+{\varepsilon }_{nt},$$

where ${Price}_{nt}^{*}$ is a latent process representing the probability that a firm will change the price of product $n$ in period $t$; ${RIVDENS}_{nt-1}$ is the independent variable that has been calculated from the competitive comparisons made by the companies at the product level in period $t-1$; and ${{Controls}_{Product}}_{nt-1}$ and ${{Controls}_{Firm}}_{nt-1}$ represent the control variables at the product and firm (brand) levels, respectively.

Meanwhile, the change in advertising investment made by the firm is measured using a continuous variable. Thus, we employ a regression model with random intercepts to avoid potential biases that can arise when observations are not independent of each other. Given that the dependent variable is constructed at the firm level, we had to aggregate all product-level variables to provide firm-level information (for other details, see Sect. 4.4). Accordingly, the estimated model can be represented by the following expression:

$$\text{log}({\left|Dif\_AI\right|)}_{ft}={\beta }_{1}{RIVDENS}_{ft-1}+{\beta }_{2}{{Controls}_{Product}}_{ft-1}+{\beta }_{3}{{Controls}_{Firm}}_{ft-1}+{u}_{f}+{\varepsilon }_{ft},$$

where $\text{log}({\left|Dif\_AI\right|)}_{ft}$ is the logarithm of the absolute value of the difference between the advertising investment made by the focal firm at time $t$ and at time $t-1$; ${RIVDENS}_{ft-1}$ is the number of competitive comparisons made by other firms against all products marketed by the focal firm $f$ in period $t-1$; ${u}_{f}\sim N(0,{\delta }_{f}^{2})$ is the random intercept associated with automakers; and ${\varepsilon }_{ft}\sim N(0,{\delta }_{\varepsilon }^{2})$ is the error term included in the regression.

To contrast Hypothesis 2, we use a survival model, specifically the proportional hazards model of Cox (1972) owing to its flexibility because it does not need to define how the hazard function depends on time, being general in nature. In turn, this regression allows us to accommodate different characteristics of the sample, such as covariates that are dependent on time and censorship (Therneau and Grambsch 2000).

Therneau’s (2018) approach is used to cope with the presence of unobserved heterogeneity due to the underlying nesting in the data at the product and firm level. In particular, two nested random effects are established, namely, $f=1,\dots, F$ automobile brands, with each $f$ brand confirmed by $n=1,\dots, {n}_{f}$ vehicles (each of which has different observations in each monthly period that remains in the Spanish market). The empirical model is as follows:

$$h\left(t\right)={h}_{0}\left(t\right)\text{exp}\left[{\beta }_{1}RIVDENS+{\beta }_{2}{Controls}_{Product}+{\beta }_{3}{Controls}_{Firm}+\left({u}_{fn}+{u}_{f}\right)\right],$$

where $h\left(t\right)$ is the dependent variable (or probability that a product exits the market at time $t$ given that it had not exited until then); ${h}_{0}\left(t\right)$ is the baseline hazard function (or hazard that the event of interest occurs simply with the passage of time); ${u}_{f}\sim N(0,{\delta }_{f}^{2})$ is the random intercept associated with automakers; and ${u}_{fn}\sim N(0,{\delta }_{fn}^{2})$ is the fragility term that captures the nested effect at the vehicle and organization level. Similar to Asplund and Sandin (1999), given that our regressions are based on monthly statistics, the decision to withdraw a product is considered supported by information from the previous month. Thus, we also use lagged independent variables in the research design.

Finally, given that several studies argue that when we analyze the entry and exit of products in the market, unmeasured confounding factors complicate the identification of causal effects (e.g., Berry and Reiss 2007; Jia 2008; Mazzeo 2002; Seim 2006); accordingly, we have implemented an alternative method to address the possible endogeneity required by the use of instrumental variables. These predictors must be correlated with the potential endogenous regressors (in our case $RIVDENS$) and must be independent of the error term included in the model that assesses the commercial longevity of the product. Following Martinez-Camblor et al. (2019), we will use the two-stage residual inclusion algorithm with frailty (2SRI-F) (for other details, see Web Appendix W1). To obtain valid instruments to predict the variables of interest, we will rely on Barroso et al. (2016), who used “the differences, lagged by one or more periods, in the core variables of interest with respect to their individual means:”

$${RIVDENS}_{nt}^{**}={RIVDENS}_{nt}-\frac{1}{{T}_{n}}\sum_{\tau }{RIVDENS}_{nt}.$$

This dynamic panel estimator (lagged 6 months) should predict our covariate of interest while anticipating uncorrelatedness with unmeasured confounders. Additionally, the estimated coefficient of this variable in the first-stage Poisson regression is positive and statistically significant [β = 0.308 (0.003), p-value = 0.0000] (see Web Appendix W2). Furthermore, by applying a chi-square test to evaluate the improvement in fit as a consequence of including ${RIVDENS}_{nt}^{**}$, we can verify that its value is above the recommended thresholds. Therefore, this variable meets all the requirements to be considered a statistically meaningful instrument.

4 Sample, data, and variables

The Spanish car market is an interesting case study owing to different aspects. First, Spain is the eighth largest automobile producer worldwide (after China, USA, Japan, Germany, India, South Corea, and Mexico) and ranked twelfth in the ranking of countries with the most units registered in 2017 (OICA 2018). In addition, Spain is the second-largest producer in Europe (accounting for 15% of production, following Germany with 29.7%) and the fifth-largest European automobile market by sales (8.34% of registrations) in 2017, following Germany, the United Kingdom, France, and Italy. That year, Spain had manufacturing facilities for vehicle manufacturers such as Citroën, Peugeot, Opel, Volkswagen, Ford, Audi, Seat, Renault, Mercedes, Nissan, and Iveco. Many vehicles exclusively produced in these factories were destined for sale across Europe (e.g., the Nissan Navara, Nissan Pulsar, and Ford Kuga) and globally (e.g., Opel Meriva, Ford S-Max, Audi Q3, Seat Ibiza, and Renault Captur, etc.). Second, after the dismantling of tariffs in 1992, several studies (e.g., Barroso and Giarratana 2013; Miravete et al. 2018) detected strong competition at the product level between firms. This rivalry, which is reflected in the competitive comparisons of the carmakers, might affect the survival of their products.

4.1 Sample

An analysis is conducted on practically all products of firms that operated in the Spanish automobile market between 2008 and 2017. The car model is defined by the organizations by name (Carroll et al. 2010) and by the absence of a substantial change in their physical characteristics (Berry et al. 1995). In this sense, each technological generation released by a firm with respect to a vehicle will be a different model for us, although it has the same name. According to Rhee and Haunschild (2006), we consider each car manufacturer (represented by a brand, e.g., BMW) as a single entity that competes with others in the market.

The sample consists of 41,135 observations of 824 car models produced by 61 firms and marketed in Spain between January 2008 and December 2017 (one monthly observation per product). The registrations of these vehicles represent practically 100% of the car market in Spain, with an average of 342 automobiles commercialized each month. Those vehicles not constantly produced or marketed by the brands over time are excluded (their 0.00002% market share is marginal).

4.2 Dependent variable

To contrast Hypothesis 1, the dependent variables are the focal firm’s competitive actions/inactions of price and advertising investment. The first variable is dichotomous, taking the value 1 if the focal company altered the product price at time $t$; otherwise, 0. The second variable is continuous and measures the natural logarithm of the absolute value of the difference between the advertising investment made by the focal firm at time $t$ and the advertising investment made by the focal firm at time $t-1$.

To contrast Hypothesis 2, the dependent variable “duration” is measured by the number of months a car model survives in the market (for other details, see Web Appendix W3). Following de Figueiredo and Kyle (2006), we define the exit of the product as the first month in which a vehicle is no longer present in our database. This finding implies that the manufacturers have ceased the production and marketing of the product. In the study period, we observed the launching and withdrawal of most of the products that are part of our sample. However, we also include censored cases. Consequently, the models that were present on the market before January 2008 may be subject to analysis to the extent that we have information regarding the specific moment at which they were introduced in the market. In turn, vehicles whose exits have not been observed (because they occurred after December 2017) will also be included given the suitability of our methodology to address this problem.

4.3 Independent variable

Regarding the testing of Hypotheses 1 and 2, the independent variable used to explain the competitive actions of the focal firm and the survival of the focal firm’s product is determined by the competitive comparison made by the firms at the product level with the vehicle of the focal company. Specifically, we use a density measure of this dyadic rivalry at the product level (RIVDENS) through the number of vehicles that have been compared by any car brand that markets them with the model $n$ of the focal firm $f$ in the period $t-1$. However, when analyzing the change in advertising investment as the dependent variable, we had to transform this regressor to capture the number of attacks in the form of competitive comparisons received by all products marketed by the focal firm $f$ in period $t-1$. In this way, RIVDENS provides information at the firm level.

Automakers’ competitive comparisons are published on their web pages, allowing a focal company to know if its product is the rival of another organization’s product. These comparisons for the models commercialized in Spain were collected monthly from 2008 to 2017 by a magazine specializing in automobiles so that the secondary information available constitutes the entire period of the sample. We have a total of 95,514 competitive dyadic comparisons of carmakers in that time frame for the products in the sample. In turn, among the 824 existing models, 604 have been considered rivals of at least one vehicle from another firm.

4.4 Control variables

The control variables at the product level are as follows:

(i)

Rivalry between products of the same firm (RIV_INTFIRM) based on competitive comparisons made by companies. RIV_INTFIRM is a dummy variable, where 1 indicates that the product $n$ of focal firm $f$ is compared by this same organization with another of its products in moment $t-1$; otherwise, 0. When examining the impact of this variable on the alteration of advertising investment by the focal firm, we measure the number of competitive comparisons that the focal firm makes between its own products in period $t-1$.

(ii)

Market segment. To capture the influence of the market segment, several dummies are included. Each car can be classified into one of the eight segments (small, compact, intermediate, mid-size luxury, luxury, sport, 4 × 4, and minivan) as defined by the National Association of Automobile Manufacturers. The luxury car is used as the baseline. In the model analyzing the change in advertising investment, these variables indicate whether the focal firm $f$ markets one or more vehicles in each of the identified segments.

(iii)

Competition between products of firms in terms of segment crowding in moment $t-1$. To control the crowdedness of the segment (de Figueiredo and Kyle 2006), we consider the number of products in the same segment as the vehicle of the focal company. We use the number of products marketed in the segment where the focal firm $f$ has proliferated the most during period $t-1$ as the variable when analyzing advertising investment actions.

(iv)

Product quality in the period $t-1$. This variable is obtained by a specialized automotive magazine and registers monthly the average rating (between 0 and 10) of each car model in the aspects of equipment, safety, and road behavior. With this variable, we intend to measure the perceived quality of the product, which has an impact on customer loyalty and the perceived value of the product (Sen et al. 2023) and consequently, on product survival. When analyzing advertising investment, we measure the average quality of all the products in the focal firm $f$’s portfolio during period $t-1$.

(v)

Replacement. We use a dummy variable that takes the value 1 when the focal firm withdraws a product while introducing another in the same segment at time t; otherwise, 0.

(vi)

Product’s order of entry. An ordinal variable captures the age of the population as a function of the time in which the product was marketed for the first time (Khessina and Carroll 2008). In the regression related to advertising investment, this variable measures the average of the values assigned to each of the focal firm’s products.

The control variables at the firm (brand) level are as follows:

(i)

Competition between the products of the same organization in terms of segment crowding. This variable reflects the focal brand’s portfolio breadth in the same segment as the product under consideration in the period $t-1$ (de Figueiredo and Kyle 2006).

(ii)

Multimarket contact. This variable is measured by using the index proposed by Baum and Korn (1999) in moment $t-1$.

(iii)

Firm size. This variable is measured by the monthly number of new vehicle registrations of the focal company at time $t-1$.

(iv)

Firm market share. This variable attempts to capture if the withdrawal of a product is supported by the market performance of the company in period $t-1$ (Asplund and Sandin 1999). It is measured as the percentage of the total sales in the market that a company holds over a given period. We will use the sum of the registrations of new vehicles obtained by the brands.

(v)

European firms. A dummy variable takes the value 1 if car manufacturers are headquartered in countries belonging to the European Union; otherwise, 0. It attempts to control whether the commercial longevity of an organization’s product may be due to the advantages derived from its geographical location.

(vi)

Generalist firms. Following resource partitioning theory (e.g., Dobrev and Kim 2006; Verhaal et al. 2017), we have calculated a dichotomous variable derived from a longitudinal cluster analysis (see Web Appendix W4), which takes the value 1 if the organization has been classified as a generalist and 0 if it is a specialist.

Table 2 shows the descriptive statistics of the variables used.

Table 2

Descriptive statistics and correlation matrix

Variable	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)	(14)	(15)	(16)	(17)	(18)	(19)	(20)	(21)	(22)
1. Product withdrawal	1
2. RIVDENS	− 0.05	1
3. RIV_INTFIRM	0.01	− 0.05	1
4. Small	0.00	0.01	− 0.05	1
5. Compact	− 0.00	− 0.05	0.07	− 0.19	1
6. Intermediate	0.00	0.01	− 0.05	− 0.15	− 0.14	1
7. Mid-size Luxury	0.01	0.03	− 0.03	− 0.10	− 0.10	− 0.08	1
8. Sport	0.01	− 0.01	0.03	− 0.17	− 0.16	− 0.13	− 0.09	1
9. 4 × 4	− 0.01	0.02	0.01	− 0.25	− 0.23	− 0.19	− 0.13	− 0.21	1
10. Minivan	0.01	0.00	0.01	− 0.16	− 0.15	− 0.12	− 0.08	− 0.14	− 0.20	1
11. Segment crowding	− 0.00	0.02	0.02	− 0.08	− 0.21	− 0.13	− 0.13	0.31	0.63	− 0.32	1
12.Competition intra firm	0.01	0.09	0.05	− 0.05	− 0.11	− 0.03	0.02	0.12	0.18	− 0.07	0.33	1
13. Multimarket contact	0.00	− 0.14	− 0.02	0.02	− 0.12	− 0.11	− 0.08	0.16	0.16	− 0.08	0.14	− 0.01	1
14. Quality	− 0.01	0.15	0.00	− 0.31	0.10	0.16	0.08	− 0.02	0.07	− 0.01	0.00	0.08	− 0.26	1
15. Replacement	0.01	0.00	0.00	0.01	0.00	0.00	0.00	0.01	− 0.01	0.00	− 0.01	− 0.02	0.00	0.00	1
16. Size	− 0.01	0.14	0.01	0.12	0.15	0.00	− 0.05	− 0.21	− 0.11	0.16	− 0.12	0.11	− 0.43	0.17	0.01	1
17. Generalist	0.00	0.10	− 0.02	0.01	0.16	0.15	0.10	− 0.29	− 0.10	0.12	− 0.17	− 0.03	− 0.65	0.28	0.00	0.42	1
18. European firm	− 0.02	0.09	− 0.04	− 0.01	0.05	− 0.01	0.04	0.12	− 0.17	− 0.04	− 0.06	0.23	− 0.08	− 0.01	− 0.01	0.24	− 0.09	1
19. Product’s order of entry	− 0.07	− 0.13	− 0.05	0.02	0.06	0.02	− 0.03	− 0.01	0.03	− 0.09	− 0.03	0.00	0.08	0.15	− 0.08	0.04	0.01	− 0.03	1
20. Firm market share	− 0.01	0.15	0.01	0.13	0.17	0.00	− 0.05	− 0.22	− 0.12	0.18	− 0.13	0.11	− 0.48	0.18	0.00	0.93	0.45	0.26	0.00	1
21. Price action		0.04	− 0.01	0.03	0.04	0.01	0.00	− 0.04	− 0.02	− 0.01		0.02	− 0.04	0.04		0.06	0.05	0.04	− 0.05	0.09	1
22. Advertising investment action		0.44	0.07	0.35	0.48	0.31	0.17	− 0.08	0.32	0.37		0.29	− 0.43	0.30		0.45	0.48	0.02	0.02	0.48		1
Mean	0.01	2.33	0.02	0.17	0.15	0.10	0.05	0.13	0.24	0.11	58.81	2.65	0.59	7.10	0.01	2,595.3	0.83	0.64	107.58	0.29	0.10	13.0
SD	0.11	2.67	0.15	0.37	0.36	0.31	0.22	0.33	0.42	0.32	20.96	1.51	0.15	0.71	0.10	2,607.6	0.38	0.48	43.19	0.27	0.31	2.33
Min	0	0	0	0	0	0	0	0	0	0	13	1	0.38	2	0	0	0	0	0	0	0	1.39
Max	1	14	1	1	1	1	1	1	1	1	103	9	1	8.50	1	12,536	1	1	210	1.12	1	16.6

5 Results

Table 3 presents the results of a binomial logit model with random coefficients using panel data and a linear regression with random intercepts to test Hypothesis 1. To examine the presence of multicollinearity, we conducted a variance inflation factor (VIF) test for each of the regressions presented. The highest VIF value is 9.05 in Model 1 and 7.77 in Model 2, with the average VIF values for all variables being 3.04 and 2.85, respectively, in these two regressions. Given that these values are below 10 (Hair et al. 2010), multicollinearity is not a concern affecting our results. Additionally, the likelihood ratio test shows that both regressions fit the data significantly better than the null model (the one without covariates explaining the behavior of the dependent variables analyzed).

Table 3

Results of a binomial logit model with random coefficients and a linear regression model with random intercepts to test Hypothesis 1 (standard errors in parenthesis and p-values in brackets)

	Model 1 (price)		Model 2 (Advertising investment)
	b	SD of β	β
Constant	− 2.341*** (0.249) [0.0000]		8.215*** (0.797) [0.000]
RIVDENS	0.039*** (0.007) [0.0000]	0.001 (0.392) [0.9974]	0.009*** (0.002) [0.000]
Control variables
RIV_INTFIRM	− 0.464** (0.200) [0.0201]	0.714*** (0.227) [0.0016]	− 0.171*** (0.057) [0.003]
Competition intra firm	0.127 (0.116) [0.2726]	0.017 (3.130) [0.9957]	0.262*** (0.032) [0.000]
Multimarket contact	− 0.680*** (0.218) [0.0018]	0.321 (0.274) [0.2408]	− 1.270*** (0.344) [0.000]
Quality	0.296 (0.208) [0.1548]	0.382** (0.156) [0.0144]	0.302*** (0.099) [0.002]
Size	0.239** (0.112) [0.0337]	0.172 (0.841) [0.8383]	0.000 (0.000) [0.497]
Product’s order of entry	− 0.308*** (0.116) [0.0078]	0.211 (0.317) [0.5070]	0.002*** (0.001) [0.007]
Firm market share	0.723*** (0.105) [0.0000]	0.002 (2.984) [0.9995]	0.597*** (0.215) [0.006]
Generalist	− 0.050 (0.087) [0.5678]		0.955*** (0.117) [0.000]
European firm	0.137*** (0.048) [0.0040]		0.085 (0.064) [0.182]
Small	0.214*** (0.063) [0.0007]		0.530*** (0.089) [0.000]
Compact	0.364*** (0.066) [0.0000]		0.882*** (0.095) [0.000]
Intermediate	0.198*** (0.075) [0.0080]		− 0.139* (0.073) [0.057]
Mid-size Luxury	0.130 (0.106) [0.2205]		− 0.083 (0.079) [0.293]
Sport	− 0.203** (0.083) [0.0146]		− 0.483*** (0.071) [0.000]
4 × 4	0.053 (0.050) [0.2905]		0.873*** (0.118) [0.000]
Minivan	− 0.075 (0.077) [0.3297]		− 0.039 (0.076) [0.607]
${\sigma }_{f}^{2}({\sigma }_{f})$			3.101 (0.080)
Observations	26,707		4,721
Log-likelihood	− 8,583.68		− 9,370.625
Likelihood ratio test	702.14*** [0.0000]		2,661.313*** [0.000]

The price action regression model has been estimated through a binomial logit model with random coefficients and panel data. This methodology allows for the possible disparity of the estimated parameters depending on the vehicle considered. Thus, we can control for unobserved heterogeneity (Sarrias 2016). This model was estimated from January 2011 to December 2017 because we only had price data at that time interval. The advertising investment action regression model was estimated using a linear regression model with random intercepts to avoid biases derived from the differences that may exist between the companies that are part of the sample (explained by the existence of unmeasured covariates). All the variables that provide information at the product level have been aggregated to provide information at the company level (given that the dependent variable measures the natural logarithm of the change in advertising investment of the focal company between period t and period t-1). See Sect. 4 for further details

*p <.10, **p <.05, ***p <.01

If we focus on Model 1 (which analyzes price changes), we observe that the estimated mean of the variable RIVDENS is positive and significant at the 1% level [b = 0.039 (0.007), p = 0.000]. This finding provides evidence that when a rival competitively compares any of its products to the focal firm’s vehicle, the latter is more likely to retaliate in the market segment where the rival’s product is commercialized through a competitive action, such as price change. We find identical conclusions when analyzing changes in advertising investment in Model 2 [β = 0.009 (0.002), p-value = 0.000]. Specifically, we observe that the greater the number of competitive comparisons made by other companies against the focal firm’s products, the higher the probability that the latter will respond by altering its advertising investment to avoid reputational damage (which occurs when consumers perceive its products as having lower quality than those of the rivals making these comparative attacks). These results support Hypothesis 1, as they indicate that companies cannot remain passive in the face of aggressive moves by other organizations if they do not want to succumb to their rivals in the market. As a visual aid, Fig. 1a graphically illustrates these results.

Fig. 1

Significant relationships between variables

Full size image

Regarding the control variables, we have two interesting findings. First, when companies make comparisons between their own products, they are less likely to attack other rivals by carrying out price changes or altering the advertising investment. When companies engage in such comparative actions, it often happens with the deliberate intention of having their products compete against each other (which, in the words of Chandy and Tellis (1998), would be considered a strategy of intentional cannibalization). In this scenario, companies seem to be less concerned with attacking external rivals and instead prefer to focus on building competitive relationships among their own products. By contrast, the greater the multimarket contact, the lower the probability that companies will engage in competitive moves against their rivals through price changes or advertising investment. This result supports the mutual forbearance hypothesis, as it shows that multimarket contact mitigates rivalry among firms competing in the market (Baum and Korn 1999).

Table 4 shows the results of eight Cox regressions to test Hypothesis 2. We have also used the VIF test to examine the existence of multicollinearity. Its highest value is 8.78, the mean being 2.99 for all the variables in Model 5. Its value is between 2.45 and 2.70 for the remaining estimated regressions. Given that these magnitudes do not exceed the threshold of 10, our estimates should not be biased due to multicollinearity. Regarding the goodness of fit, the likelihood ratio test, which is statistically significant at the 1% level, indicates a preference for the estimated regressions over the null model. By contrast, the information criteria of Akaike (AIC) and Schwartz (BIC), although they are not conclusive due to their disagreement, opt for models that incorporate two nested random effects, reinforcing the convenience of the chosen design.

Table 4

Results of the Cox regression models to test Hypothesis 2 (standard errors in parenthesis and p-values in brackets)

	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7		Model 8
	β	Exp(β)	Β	Exp(β)	β	Exp(β)	β	Exp(β)	β	Exp(β)	β	Exp(β)	β	Exp(β)	β	Exp(β)
RIVDENS	− 0.331 (0.030)	0.718*** [0.0000]	− 0.373 (0.032)	0.688*** [0.0000]	− 0.373 (0.032)	0.689*** [0.0000]	− 0.373 (0.032)	0.688*** [0.0000]	− 0.370 (0.033)	0.691*** [0.0000]	− 1.000 (0.096)	0.368*** [0.0000]	− 1.009 (0.101)	0.365*** [0.0000]	− 1.005 (0.102)	0.366*** [0.0000]
Control variables
RIV_INTFIRM	0.182 (0.240)	1.199 [0.4478]	0.170 (0.286)	1.185 [0.5526]	0.179 (0.291)	1.196 [0.5371]	0.168 (0.288)	1.183 [0.5596]	0.205 (0.290)	1.227 [0.4802]	− 0.047 (0.252)	0.955 [0.8535]	− 0.044 (0.267)	0.957 [0.8683]	− 0.037 (0.275)	0.964 [0.8931]
Small									0.924 (0.364)	2.519** [0.0111]	0.852 (0.308)	2.343*** [0.0056]	0.891 (0.323)	2.437*** [0.0058]	0.748 (0.336)	2.113** [0.0256]
Compact									0.764 (0.358)	2.146** [0.0328]	0.561 (0.306)	1.752* [0.0672]	0.576 (0.322)	1.778* [0.0737]	0.435 (0.331)	1.545 [0.1889]
Intermediate									0.742 (0.367)	2.101** [0.0433]	0.514 (0.317)	1.672 [0.1052]	0.525 (0.334)	1.690 [0.1155]	0.428 (0.341)	1.535 [0.2094]
Mid-size Luxury									0.670 (0.401)	1.954* [0.0949]	0.503 (0.345)	1.653 [0.1444]	0.538 (0.364)	1.713 [0.1389]	0.467 (0.370)	1.595 [0.2078]
Sport									0.901 (0.349)	2.463*** [0.0099]	0.738 (0.297)	2.091** [0.0130]	0.801 (0.313)	2.227** [0.0106]	0.670 (0.322)	1.955** [0.0376]
4 × 4									0.386 (0.359)	1.472 [0.2812]	0.334 (0.307)	1.397 [0.2765]	0.331 (0.322)	1.392 [0.3049]	0.207 (0.332)	1.230 [0.5324]
Minivan									0.539 (0.366)	1.714 [0.1411]	0.462 (0.311)	1.587 [0.1374]	0.479 (0.327)	1.615 [0.1426]	0.326 (0.337)	1.386 [0.3332]
Segment crowding							0.004 (0.003)	1.004 [0.2013]
Competition intra firm	0.105 (0.034)	1.110*** [0.0018]	0.127 (0.039)	1.136*** [0.0010]	0.180 (0.044)	1.198*** [0.0000]	0.157 (0.046)	1.170*** [0.0007]	0.156 (0.045)	1.169*** [0.0006]	0.118 (0.038)	1.125*** [0.0020]	0.123 (0.040)	1.131*** [0.0019]	0.147 (0.043)	1.158*** [0.0007]
Multimarket contact	− 0.672 (0.451)	0.511 [0.1363]	− 0.822 (0.523)	0.440 [0.1163]	− 0.608 (0.632)	0.545 [0.3363]	− 0.664 (0.625)	0.515 [0.2879]	− 0.816 (0.613)	0.442 [0.1828]	− 1.097 (0.484)	0.334** [0.0234]	− 1.121 (0.509)	0.326** [0.0277]	− 0.955 (0.581)	0.385* [0.0999]
Quality	− 0.031 (0.072)	0.969 [0.6647]	− 0.045 (0.086)	0.956 [0.6036]	− 0.017 (0.089)	0.984 [0.8522]	− 0.028 (0.088)	0.973 [0.7533]	0.036 (0.094)	1.037 [0.6996]	0.231 (0.082)	1.260*** [0.0048]	0.237 (0.087)	1.267*** [0.0068]	0.242 (0.091)	1.274*** [0.0076]
Replacement							0.801 (0.406)	2.228** [0.0484]	0.778 (0.405)	2.177* [0.0547]
Size	0.000 (0.000)	1.000** [0.0142]	0.000 (0.000)	1.000** [0.0190]	0.000 (0.000)	1.000** [0.0269]	0.000 (0.000)	1.000** [0.0303]	0.000 (0.000)	1.000** [0.0435]	0.000 (0.000)	1.000*** [0.0028]	0.000 (0.000)	1.000*** [0.0033]	0.000 (0.000)	1.000*** [0.0037]
Generalist	0.237 (0.170)	1.267 [0.1624]	0.180 (0.206)	1.197* [0.3820]	0.236 (0.264)	1.266 [0.3717]	0.259 (0.261)	1.296 [0.3209]	0.141 (0.259)	1.151 [0.5872]	0.141 (0.191)	1.151 [0.4593]	0.125 (0.203)	1.133 [0.5391]	0.160 (0.240)	1.174 [0.5053]
European Firm	− 0.393 (0.101)	0.675*** [0.0001]	− 0.523 (0.123)	0.593*** [0.0000]	− 0.539 (0.180)	0.583*** [0.0028]	− 0.502 (0.178)	0.605*** [0.0049]	− 0.580 (0.171)	0.560*** [0.0007]	− 0.344 (0.111)	0.709*** [0.0020]	− 0.392 (0.119)	0.676*** [0.0010]	− 0.457 (0.157)	0.633*** [0.0036]
Product’s order of entry	− 0.003 (0.002)	0.997** [0.0480]	− 0.001 (0.002)	0.999 [0.5137]	− 0.001 (0.002)	0.999 [0.6092]	− 0.000 (0.002)	1.000 [0.8485]	− 0.000 (0.002)	1.000 [0.9767]	− 0.005 (0.002)	0.995*** [0.0053]	− 0.004 (0.002)	0.996** [0.0234]	− 0.004 (0.002)	0.996** [0.0357]
Firm market share	− 1.417 (0.438)	0.243*** [0.0012]	− 1.462 (0.459)	0.232*** [0.0015]	− 1.664 (0.506)	0.189*** [0.0010]	− 1.620 (0.504)	0.198*** [0.0013]	− 1.635 (0.500)	0.195*** [0.0011]	− 1.535 (0.454)	0.215*** [0.0007]	− 1.527 (0.462)	0.217*** [0.0009]	− 1.577 (0.492)	0.207*** [0.0013]
Residual first stage											0.730 (0.099)	2.075*** [0.0000]	0.725 (0.105)	2.064*** [0.0000]	0.721 (0.106)	2.057*** [0.0000]
Observations		41,135		41,135		41,135		41,135		41,135		36,313		36,313		36,313
Events		492		492		492		492		492		467		467		467
Log likelihood		− 2,502.91		− 2,498.91		− 2,492.64		− 2,490.36		− 2,482.91		− 2,323.84		− 2,323.48		− 2,320.54
Null		− 2,617.97		− 2,617.97		− 2,617.97		− 2,617.97		− 2,617.97		− 2,459.17		− 2,459.17		− 2,459.17
Likelihood ratio test		230.1*** [0.0000]		238.1*** [0.0000]		250.7*** [0.0000]		255.2*** [0.0000]		270.1*** [0.0000]		270.7*** [0.0000]		271.4*** [0.0000]		277.3*** [0.0000]
AIC		5,025.81		5,019.82		5,009.28		5,008.73		5,005.82		4,683.69		4,684.97		4,681.09
BIC		5,067.80		5,066.01		5,059.66		5,067.51		5,089.79		4,758.32		4,763.75		4,764.01
AIC^a		210.12		216.11		226.65		227.20		230.11		234.66		233.38		237.26
BIC^a		168.14		169.93		176.27		168.43		146.14		160.02		154.60		154.33
ANOVA				7.99*** [0.0047]		12.55*** [0.0004]		4.55 [0.1028]		14.91** [0.0210]				0.72 [0.3957]		5.88** [0.0153]
ANOVA^b										19.46** [0.0126]						6.60** [0.0368]
Fixed effects		Yes		Yes		Yes		Yes		Yes		Yes		Yes		Yes
Random effects		No		Yes		Yes		Yes		Yes		No		Yes		Yes
${\sigma }_{n}^{2}({\sigma }_{n})/{\sigma }_{fn}^{2}({\sigma }_{fn})$				0.403 (0.635)		0.345 (0.588)		0.324 (0.569)		0.345 (0.587)				0.111 (0.333)		0.095 (0.308)
${\sigma }_{f}^{2}({\sigma }_{f})$						0.174 (0.417)		0.164 (0.405)		0.114 (0.338)						0.094 (0.306)

Model 6 has been calculated with the 2SRI method and Models 7 and 8 with the 2SRI-F algorithm to control for endogeneity (for other details, see the methodology section and Web Appendix W1)

^*p < 0.10, **p < 0.05, ***p < 0.01

a The improvement in the tested model regarding AIC and BIC compared with the null model

b Chi-square tests where the improvement in the fit of Model 5 vs. 3 and Model 8 vs. 6 is evaluated

Following Therneau (2018), we use chi-square tests (presented under the label ANOVA in Table 4) to sequentially evaluate which of the estimated regressions has the greatest explanatory power. In this sense, we can see an improvement in the fit as a consequence of the gradual increase in the maximum likelihood function; that is, Model 2 is preferable to Model 1 [χ²₍₁₎ = 7.99, p value = 0.0047], Model 3 to 2 [χ²₍₁₎ = 12.55, p value = 0.0004), and Model 5 to 4 [χ²₍₆₎ = 14.91, p value = 0.0210] and 3 [χ2₍₈₎ = 19.46, p value = 0.0126]. Therefore, we will comment on the results regarding Model 5, whose estimated parameters are similar in terms of significance and magnitude with those of the remaining regressions with two nested random effects.

Regarding Hypothesis 2, we analyze how the variable RIVDENS affects the survival of the product. This variable presents a negative and significant estimated coefficient at 1% [β = − 0.370 (0.033), p-value = 0.0000]. However, we focus on the exponential of the coefficient, which reflects the multiplicative change that an independent variable produces in the risk that the exit of the product occurs. Thus, we can verify that each time the product of the focal firm is identified as a comparison target by another organization, it decreases the risk of exit by 30.9% (1 − 0.691 = 0.309). This result seems to support Hypothesis 2 to the extent that an increase in survival is detected. Figure 1b provides a graphical representation of this statistically significant result.

Regarding the control variables at the product level, the coefficient for intra-firm competition in Model 5 is significant at a level below 1%. This finding suggests that when a focal firm compares one of its products with another of its products, it reduces the survival of the compared product. This outcome is explained by the cannibalization effect, which has been documented in previous studies (see Greenstein and Wade 1998; Ruebeck 2002; Requena-Silvente and Walker 2005). Additionally, significant coefficients exist for the small, compact, intermediate, mid-size luxury, and sport market segments in Model 5. The results indicate that the estimated market exit risk for models in these segments is higher than the estimated exit risk for models in the baseline segment (luxury segment). Finally, the exponential of the replacement coefficient is greater than one and is significant at a level below 10%; as noted by de Figueiredo and Kyle (2006), this finding suggests that market exit is more likely when a company introduces an additional product in the same segment at the time it withdraws another.

As for the firm-level control variables, firm market share has a significant coefficient; as expected (see Ruebeck 2005), it suggests that a product manufactured by a firm with market power is less likely to exit the market. Following Khessina and Carroll (2008), we also considered the possible influence of the firm’s geographic region and found that if the firm is European, the survival of its products in the market increases significantly. This phenomenon is explained by the fact that a European firm tends to align more closely with the tastes and preferences of Spanish consumers.

Finally, although the previous models eliminate a wide range of confounders, we sought to test the robustness of our findings by performing other estimations to control for the presence of endogeneity. Specifically, we applied—as has been exposed in the methodology section and in Web Appendix W1—the 2SRI and 2SRI-F algorithms (Martínez-Camblor et al. 2019). These procedures lead to the last three regressions presented in Table 4. We can observe, through chi-square tests, that Model 8 (which incorporates two nested random intercepts) is superior to Model 7 (which adds a product-level frailty term) and 6 [χ²₍₁₎ = 5.88, p-value = 0.0153; χ²₍₂₎ = 6.60, p-value = 0.0368]. Therefore, we will comment on the results relative to Model 8, whose estimated parameters are similar in terms of significance and magnitude with those of the remaining regressions that control for endogeneity. Notably, the RIVDENS coefficient remains negative and statistically significant [β = − 1.005 (0.102), p-value = 0.0000]. However, given that this dynamic panel instrument only satisfies the exclusion restriction when the sources of endogeneity do not vary over time, we consider these results to be supportive rather than apodictic evidence.

6 Discussion and conclusions

As rooted in the literature on the survival of the product, we have demystified the trend of thinking that all the products in the market are de facto competitors (see Greenstein and Wade 1998; Ruebeck 2002, 2005; Requena-Silvente and Walker 2005, 2009; de Figueiredo and Kyle 2006; Talay and Townsend 2015; Barnett et al. 2022). From this premise, our analyses have revealed several key findings. First, segment crowding, used by most researchers to analyze the relationship between competition and product survival (see Greenstein and Wade 1998; Ruebeck 2002, 2005; Requena-Silvente and Walker 2005, 2009; de Figueiredo and Kyle 2006; Barnett et al. 2022), can inadequately capture rivalry between firms’ products. This variable does not even influence the commercial longevity of the product in our study (see Tables 4 and Web Appendix W5). This non-significant result, combined with the significant influence of competitive comparison on product survival, underscores the importance of asymmetric rivalry between product dyads of different firms.

Second, our study finds that a rival’s competitive comparison with a focal firm’s product increases the focal firm’s subsequent actions related to pricing and advertising investment. These results can be explained through the AMC framework because the focal firm becomes aware of the attack and is motivated to retaliate in terms of pricing and advertising investment. To date, the only identified study, conducted by Kim and Tsai (2012), found that a focal firm’s competitive comparison with a target firm is positively associated with the focal firm’s sales growth, attributing this phenomenon to basking in the reflected glory of the reputable target firm. However, that study does not examine competitive comparison at the product level, nor does it explore the influence of a focal firm’s comparative attack on the rival’s retaliation.

Third, we found empirical evidence that a product from one focal company will stay in the market longer when it is identified as a comparison target by another organization. This result does not support the predominant view in the literature, which advocates a negative relationship between competition and product survival (see Table 1). Instead, it supports a positive relationship theoretically underpinned by the AMC framework. To date, we have identified only one study, by Talay and Townsend (2015), which assumes a positive relationship between a product’s recent competitive experience duration—measured by the time a product is exposed to all the rival products in the market, considering that all products in the market are de facto competitors—and product survival, theoretically based on the Red Queen effect. However, that study does not analyze the reciprocal sequence of action and competitive response implicit in the Red Queen competition. Following the AMC framework, our work alternatively proposes that asymmetric rivalry between competing products, in terms of a rival firm’s competitive comparison with a focal firm’s product, positively affects the focal firm’s product survival.

6.1 Managerial implications

The main managerial implications derived from our results are as follows. First, the findings facilitate the analysis of the impact of a decision, such as a firm’s competitive comparison against a rival’s product, thereby enabling the prediction of the rival’s most likely response. Managers should assess potential competitive actions from the perspective of the target firm’s awareness of the attack and its response motivation. This perspective offers a valuable framework for the target firm whose product is subject to the comparative attack. Consequently, the firm initiating the competitive comparison should anticipate strong retaliation from the target firm, which is likely to act as a competitive agent responding to the attack through price adjustments and increased advertising investments. For instance, in the automobile industry, we can highlight the 2009 billboard battle in Santa Monica between Audi and BMW. Audi launched its A4 model with the slogan “Your move, BMW,” to which BMW responded by promoting its M3 model with the retort “Checkmate.”

Second, with respect to the strategic maneuvers, a firm must consider that a comparative attack directed against the product of a rival is rarely carried out with impunity; and that the effectiveness of this action, in terms of the survival of the rival product largely depends on the response used in retaliation by the rival. A company that takes its adversary lightly can end up prolonging the commercial longevity of its rival’s product in the market. Therefore, before tackling a rival, an organization must be careful when devising its strategies (Chen et al. 2021; Ketchen et al. 2004), especially—as with competitive comparisons—when it has to disseminate valuable information to improve the image of its product. Otherwise, it could end up being the victim of its own attack, facilitating the survival of the rival’s product.

This study focuses on the Spanish automobile market, which serves as a representative case for the European automobile market (see Sect. 4). Unlike other regions, such as the United States (see Miravete et al. 2018), the European automobile market has historically been characterized by a high preference for diesel vehicles. This preference emerged due to government interventions, including diesel fuel excise tax policies introduced in the 1970s and European Union emission regulations implemented in the 1990s, which favored diesel vehicles. By contrast, emission standards in the United States created a regulatory environment that effectively led to the disappearance of diesel vehicles from the market. As government interventions, consumer demand, technological advancements, and competitive forces can potentially create dynamic markets, lagging behind the competition can prove highly detrimental. The importance of understanding these factors and integrating the competitive environment into a firm’s strategic planning becomes evident when considering the state of ongoing changes in the global automobile industry. As companies strive to develop global products, the idiosyncrasies of competition across different markets play a crucial role. Our work highlights the crucial task of companies to examine specific regional market conditions and industry dynamics when evaluating the potential impact of competitive strategies on the survival of their marketed products. This emphasis ensures that firms align their strategic initiatives with the complexities and nuances of evolving competitive environments.

6.2 Theoretical contributions

Two key theoretical contributions emerge from our empirical model. First, our findings challenge the basic assumptions in studies that analyze the impact of rivalry on product survival. Many papers have argued that competition directly affects the “mortality” of the products that are traded in the market (e.g., Barroso et al. 2016). Specifically, they advocate the following: faced with an increase in rivalry understood in terms of density, consumers lose their interest in less competitive products, inevitably leading to their exit. By contrast, other studies have focused on competition as a factor that firms consider when making decisions regarding the products that are part of their portfolio (e.g., de Figueiredo and Kyle 2006; Wang 2019). In this sense, they maintain that a greater density of products in the market leads to a fall in prices and profit margins, which encourages companies to withdraw their products due to the loss of profitability. Both approaches conceptualize competition by considering that all the products commercialized in the market are de facto competitors. In other words, they consider that in crowded market segments, a kind of “war” is anticipated to happen between all the products commercialized that will reduce their survival.

The main problem from which these works suffer is that they do not assume the existence of competitive asymmetries. By abstracting from the context in which products fight their battles and the extent to which two products compete with each other, scholars have been unable to delve into the nature of each competitive interaction (Chen 1996). Only through a peer-to-peer analysis that focuses on these subjective relationships (Kilduff et al. 2010) can the peculiarities of competition and rivalry be unraveled. In this study, we start from the investigation of competitive dynamics to offer a particularly novel vision to explore the factors that drive the development of actions and responses against the products of certain firms.

Second, our work contributes to the competitive dynamics literature. Although previous research has shown how competitive signals from a rival influence the product-level strategy developed by the focal firm (e.g., Chen et al. 2017), we are unaware of the existence of any empirical study that has examined how competitive comparisons can influence rivalry between companies’ products. In the competitive dynamics literature, the components of awareness and motivation have traditionally been measured through the size of the firm and the volume of the rival’s attack (e.g., Chen et al. 2007; Gao et al. 2017; Shi et al. 2020). However, by using this type of structural variable (unable to identify whether the action is performed with a competitor in mind), nearly all studies cannot address the thorny question of intentionality. In this research, we use competitive comparison because it allows us to overcome this deficiency.

On the basis of the AMC framework (e.g., Chen 1996; Chen et al. 2007), we argue that when a firm X compares its product with the product of another company (here called Y), the latter will have awareness and motivation to develop actions against the product of the first. Specifically, the fact that X disseminates this information through its website is precisely what enables Y to recognize this competitive signal. Furthermore, the previous literature has found that the first organization is highly likely to launch a barrage of attacks against the second company’s product to confirm its comparison (Kim and Tsai 2012). We understand that this threatened existence is precisely the condition for Y to counterattack in defense of its competitive status quo (Ferrier et al. 1999).

6.3 Limitations and future research

The limitations of our empirical work also suggest interesting avenues for future studies. First, while we examine whether a competitor compares its product with a focal company’s product, we do not measure the degree to which such comparison between two products is emphasized relative to other pairs. Future research could use surveys to explore why two products tend to be compared more frequently and how firms prioritize each of the identified targets.

Second, our work theorizes that when a rival competitively compares its product to the product of a focal firm, the latter company is induced to fight back, which will end up prolonging the survival of its product. Specifically, we have carried out an empirical contrast of these responses in the form of price and advertising investment changes. However, anecdotal evidence suggests that the focal firm may also resort to other types of actions, such as restylings or promotional campaigns, that we have not been able to operationalize due to a lack of data. While we expect competitive comparisons to result in all types of moves and counterattacks at the product level, analyzing their influence on the development of other types of actions (aside from a price and advertising investment change) by the focal firm could help corroborate our findings.

Third, future research could develop longitudinal analyses that reveal the impact of the opinions of other stakeholders on the competitiveness activity of organizations and therefore, in the commercial longevity of their products in the market. Shedding additional light on these views would provide an excellent opportunity to gain a better understanding of competition analysis and the consequences of rivalry between firms’ products.

Lastly, although product proliferation decisions seem to be made by large multinational corporations at the global level, many cases exist where specific differences are observed in the life histories of products that vary depending on the countries to which the companies are directed. The Ford Edge model, for example, has been withdrawn from different European markets due to its meager sales (including Spain); however, it continues to be offered in some countries in the Schengen area such as Germany, Italy, Switzerland, Austria, Poland, and Belgium. Ultimately, factors such as the idiosyncratic characteristics of the audience and their specific tastes and preferences diverge in different national markets, which are considered by brands when making decisions at the local level on the commercial longevity of their products in each territory. Although our work is adequate to analyze how the competitive comparison with the product of a focal company affects its survival, our findings are specific to the Spanish automobile market. Care should be taken not to overgeneralize our findings. Given that markets vary in the degree to which organizations compete at the product level, specific contingencies can arise that affect the relationships between these factors. Future research could examine the extent to which our framework holds up in other markets, helping corroborate and generalize the theory and hypotheses presented in this paper.

Acknowledgements

Grant PID2021-123453NB-I00 funded by MCIN/AEI/https://doi.org/10.13039/501100011033 (Spain) and by ERDF A way of making Europe.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

previous article Values and subjective well-being of European entrepreneurs: a configurational analysis across technological levels

next article Expanding overseas: the contribution of board role performance from an emerging market perspective

Metadata
Appendix
Literature

Title: Hunted hunter: the role of competitive comparison in product survival
Authors: Fernando Campayo-Sanchez
Francisco Mas-Ruiz
Juan Luis Nicolau
Publication date: 07-03-2025
Publisher: Springer Berlin Heidelberg
Published in: Review of Managerial Science / Issue 12/2025
Print ISSN: 1863-6683
Electronic ISSN: 1863-6691
DOI: https://doi.org/10.1007/s11846-025-00870-5

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 274 KB)

Aaker DA (1996) Building strong brands. Free Press, New York

Andrevski G, Ferrier WJ (2019) Does it pay to compete aggressively? Contingent roles of internal and external resources. J Manage 45(2):620–644. https://doi.org/10.1177/0149206316673718CrossRef

Asplund M, Sandin R (1999) The survival of new products. Rev Ind Organ 15(3):219–237. https://doi.org/10.1023/A:1007708612713CrossRef

Barnett WP, Hansen MT (1996) The Red Queen in organizational evolution. Strat Man J 17:139–157CrossRef

Barnett WP, McKendrick DG (2004) Why are some organizations more competitive than others? Evidence from a changing global market. Adm Sci Q 49(4):535–571. https://doi.org/10.2307/4131490CrossRef

Barnett WP, Rhee M, Tak E (2022) Manufacturing popularity: an ecological model of time-based competition. Strat Sci 7(4):267–283CrossRef

Barroso A, Giarratana MS (2013) Product proliferation strategies and firm performance: The moderating role of product space complexity. Strateg Manage J 34(12):1435–1452. https://doi.org/10.1002/smj.2079CrossRef

Barroso A, Giarratana MS, Reis S, Sorenson O (2016) Crowding, satiation, and saturation: the days of television series’ lives. Strateg Manage J 37(3):565–585. https://doi.org/10.1002/smj.2345CrossRef

Baum JAC, Korn HJ (1999) Dynamics of dyadic competitive interaction. Strateg Manage J 20(3):251–278. https://doi.org/10.1002/(SICI)1097-0266(199903)20:3%3c251::AID-SMJ23%3e3.0.CO;2-HCrossRef

Berry S, Reiss P (2007) Empirical models of entry and market structure. Handb Ind Organ 3:1845–1886. https://doi.org/10.1016/S1573-448X(06)03029-9CrossRef

Berry S, Levinsohn J, Pakes A (1995) Automobiles prices in market equilibrium. Econometrica 63(4):841–890. https://doi.org/10.2307/2171802CrossRef

Bromiley P, Papenhausen C, Borchert P (2002) Why do gas prices vary, or towards understanding the micro-structure of competition. Manag Decis Econ 23(4–5):171–186. https://doi.org/10.1002/mde.1060CrossRef

Calderon-Monge E, Ribeiro-Soriano D (2024) The role of digitalization in business and management: a systematic literature review. Rev Manage Sci 18(2):449–491

Carroll GR, Khessina OM, McKendrick DG (2010) The social lives of products: analyzing product demography for management theory and practice. Acad Manage Ann 4(1):157–203. https://doi.org/10.5465/19416521003732362CrossRef

Chandy RK, Tellis GJ (1998) Organizing for radical product innovation: the overlooked role of willingness to cannibalize. J Mark Res 35(4):474–487. https://doi.org/10.1177/002224379803500406CrossRef

Chen MJ (1996) Competitor analysis and interfirm rivalry: towards a theoretical integration. Acad Manage Rev 21(1):100–134. https://doi.org/10.5465/amr.1996.9602161567CrossRef

Chen MJ, Hambrick DC (1995) Speed, stealth, and selective attack: How small firms differ from large firms in competitive behavior. Acad Manage J 38(2):453–482. https://doi.org/10.5465/256688CrossRef

Chen MJ, MacMillan IC (1992) Nonresponse and delayed response to competitive moves: the roles of competitor dependence and action irreversibility. Acad Manage J 35(3):539–570. https://doi.org/10.5465/256486CrossRef

Chen MJ, Miller D (2015) Reconceptualizing competitive dynamics: a multidimensional framework. Strateg Manage J 36(5):758–775. https://doi.org/10.1002/smj.2245CrossRef

Chen MJ, Su KH, Tsai W (2007) Competitive tension: the awareness-motivation-capability perspective. Acad Manage J 50(1):101–118. https://doi.org/10.5465/amj.2007.24162081CrossRef

Chen MJ, Lin HC, Michel JG (2010) Navigating in a hypercompetitive environment: the roles of action aggressiveness and TMT integration. Strateg Manage J 31(13):1410–1430. https://doi.org/10.1002/smj.891CrossRef

Chen T, Tribbitt MA, Yang Y, Li X (2017) Does rivals’ innovation matter? A competitive dynamics perspective on firms product strategy. J Bus Res 76:1–7. https://doi.org/10.1016/j.jbusres.2017.02.016CrossRef

Chen MJ, Michel JG, Lin W (2021) Worlds apart? Connecting competitive dynamics and the resource-based view of the firm. J Manage 47(7):1820–1840. https://doi.org/10.1177/01492063211000422CrossRef

Cox DR (1972) Regression models and life-tables. J R Stat Soc 34(2):187–220. https://doi.org/10.1111/j.2517-6161.1972.tb00899.xCrossRef

D’Aveni RA (1994) Hypercompetition: managing the dynamics of strategic maneuvering. Free Press, New York

de Figueiredo JM, Kyle MK (2006) Surviving the gales of creative destruction: the determinants of product turnover. Strateg Manage J 27(3):241–264. https://doi.org/10.1002/smj.512CrossRef

Derfus PJ, Maggitti PG, Grimm CM, Smith KG (2008) The red queen effect: competitive actions and firm performance. Acad Manage J 51(1):61–80. https://doi.org/10.5465/amj.2008.30708624CrossRef

Dobrev SD, Kim TY (2006) Positioning among organizations in a population: moves between market segments and the evolution of industry structure. Adm Sci Q 51(2):230–261. https://doi.org/10.2189/asqu.51.2.230CrossRef

Downing ST, Kang JS, Markman GD (2019) What you don’t see can hurt you: awareness cues to profile indirect competitors. Acad Manage J 62(6):1872–1900. https://doi.org/10.5465/amj.2018.0048CrossRef

Elsbach KD, Kramer RM (1996) Members’ responses to organizational identify threats: encountering and countering the business week rankings. Adm Sci Q 41(3):442–476. https://doi.org/10.2307/2393938CrossRef

Ferrier WJ, Smith KG, Grimm CM (1999) The role of competitive action in market share erosion and industry dethronement: a study of industry leaders and challengers. Acad Manage J 42(4):372–388. https://doi.org/10.5465/257009CrossRef

Gao H, Yu T, Cannella AA Jr (2017) Understanding word responses in competitive dynamics. Acad Manage Rev 42:129–144. https://doi.org/10.5465/amr.2014.0211CrossRef

Gelei, A, Dobos, I (2024) Micro-coopetition: Conceptualizing and operationalizing coopetitive managerial decision-making over time—a game theoretic approach. Rev Manage Sci: 1–25. https://doi.org/10.1007/s11846-023-00676-3

Greenstein S, Wade J (1998) The product life cycle in the commercial mainframe computer market, 1968–1982. RAND J Econ 29(4):772–789CrossRef

Guo W, Yu T, Gimeno J (2017) Language and competition: communication vagueness, interpretation difficulties, and market entry. Acad Manage J 60(6):2073–2098. https://doi.org/10.5465/amj.2014.1150CrossRef

Hair JF, Black WC, Babin BJ, Anderson RE (2010) Multivariate data analysis. Cengage, Boston, MA

Hannan MT, Freeman JH (1977) The population ecology of organizations. Amer J Sociol 82(5):929–964CrossRef

Jia P (2008) What happens when Wal-Mart comes to town: an empirical analysis of the discount retailing industry. Econometrica 76(6):1263–1316. https://doi.org/10.3982/ECTA6649CrossRef

Kang E, Thosuwanchot N, Gomulya D (2021) Mitigating industry contagion effects from financial reporting fraud: a competitive dynamics perspective of non-errant rival firms exploiting product-market opportunities. Strateg Organ. https://doi.org/10.1177/14761270211025947CrossRef

Keller KL (1998) Strategic brand management. McGraw-Hill, New York

Ketchen DJ Jr, Snow CC, Hoover VL (2004) Research on competitive dynamics: recent accomplishments and future challenges. J Manage 30(6):779–804. https://doi.org/10.1016/j.jm.2004.06.002CrossRef

Khessina OM, Carroll GR (2008) Product demography of De Novo and De Alio firms in the optical disk drive industry, 1983–1999. Organ Sci 19(1):25–138. https://doi.org/10.1287/orsc.1070.0301CrossRef

Kilduff GJ, Elfenbein HA, Staw BM (2010) The psychology of rivalry: a relationally dependent analysis of competition. Acad Manage J 53(5):943–969. https://doi.org/10.5465/amj.2010.54533171CrossRef

Kim KH, Tsai W (2012) Social comparison among competing firms. Strateg Manage J 33(2):115–136. https://doi.org/10.1002/smj.945CrossRef

Luoma J, Laamanen T, Lamberg JA (2022) Toward a routine-based view of interfirm rivalry. Strateg Organ 20(2):433–446. https://doi.org/10.1177/147612702093135CrossRef

Martínez-Camblor P, Mackenzie T, Staiger DO, Goodney PP, O’Malley AJ (2019) Adjusting for bias introduced by instrumental variable estimation in the Cox proportional hazards model. Biostatistics 20(1):80–96. https://doi.org/10.1093/biostatistics/kxx062CrossRef

Mazzeo MJ (2002) Product choice and oligopoly market structure. RAND J Econ 33(2):221–242. https://doi.org/10.2307/3087431CrossRef

Miravete J, Moral MJ, Thurk J (2018) Fuel taxation, emissions policy, and competitive advantage in the diffusion of European diesel automobiles. Rand J Econ 49(3):504–540. https://doi.org/10.1111/1756-2171.12243CrossRef

Nicolau-Gonzálbez JL, Ruiz-Moreno F (2014) Who performs a stronger response to whom? Detecting individual competitive actions and reactions. Rev Manage Sci 8:385–403. https://doi.org/10.1007/s11846-013-0109-1CrossRef

OICA (2018) Production and sales statistics. International Organization of Motor Vehicle Manufacturers, OICA, Paris

Requena-Silvente F, Walker J (2005) Competition and product survival in the UK car market. App Econ 37:2289–2295CrossRef

Requena-Silvente F, Walker J (2009) The survival of differentiated products: an application to the UK Automobile market, 1971–2002. Manch Scho 77(3):288–316CrossRef

Rhee M, Haunschild PR (2006) The liability of good reputation: a study of product recalls in the US automobile industry. Organ Sci 17(1):101–117. https://doi.org/10.1287/orsc.1050.0175CrossRef

Rindova VP, Becerra M, Contardo I (2004) Enacting competitive wars: competitive activity, language games, and market consequences. Acad Manage Rev 29(4):670–686. https://doi.org/10.5465/amr.2004.14497655CrossRef

Rojas-Córdova C, Williamson AJ, Pertuze JA, Calvo G (2023) Why one strategy does not fit all: a systematic review on exploration–exploitation in different organizational archetypes. Rev Manage Sci 17(7):2251–2295. https://doi.org/10.1007/s11846-022-00577-xCrossRef

Ruebeck CS (2002) Interfirm competition, intrafirm cannibalisation and product exit in the market for computer hard disk drives. Econ Soc Rev 33(1):119–131

Ruebeck CS (2005) Model exit in a vertically differentiated market: interfirm competition versus intrafirm cannibalization in the computer hard disk drive industry. Rev Ind Organ 26:27–59CrossRef

Sarrias M (2016) Discrete choice models with random parameters in R: The Rchoice Package. J Stat Softw 74(10):1–31. https://doi.org/10.18637/jss.v074.i10CrossRef

Schumpeter JA (1976) Capitalism, socialism, and democracy. George Allen & Unwin, London

Seim K (2006) An empirical model of firm entry with endogenous product-type choices. RAND J Econ 37(3):619–640. https://doi.org/10.1111/j.1756-2171.2006.tb00034.xCrossRef

Sen SS, Alexandrov A, Jha S, McDowell WC, Babakus E (2023) Convenient= competitive? How Brick-And-Mortar Retailers can cope with Online Competition. Rev Manage Sci 17(5):1615–1643. https://doi.org/10.1007/s11846-022-00566-0CrossRef

Shi W, Connelly B, Hoskisson R, Ketchen D (2020) Portfolio spillover of institutional investor activism: an awareness–motivation–capability perspective. Acad Manage J 63(6):1865–1892. https://doi.org/10.5465/amj.2018.0074CrossRef

Snyder CR, Lassegard M, Ford CE (1986) Distancing after group success and failure: basking in reflected glory and cutting off reflected failure. J Pers Soc Psychol 51(2):382–388. https://doi.org/10.1037/0022-3514.51.2.382CrossRef

Sytch M, Tatarynowicz A (2014) Friends and foes: the dynamics of dual social structures. Acad Manage J 57(2):585–613. https://doi.org/10.5465/amj.2011.0979CrossRef

Talay MB, Townsend JD (2015) Do or die: competitive effects and Red Queen dynamics in the product survival race. Ind Corp Chan 24(3):721–738CrossRef

Talay MB, Calantone RJ, Voorhees CM (2014) Coevolutionary dynamics of automotive competition: Product innovation, change, and marketplace survival. J Prod Inn Man 31(1):61–78CrossRef

Therneau TM, Grambsch PM (2000) Modeling survival data: extending the Cox model. Springer, BerlinCrossRef

Therneau, TM (2018) Mixed effects Cox models. Mayo Clinic.

Verhaal JC, Hoskins JD, Lundmark LW (2017) Little fish in a big pond: legitimacy transfer, authenticity, and factors of peripheral firm entry and growth in the market center. Strateg Manage J 38(12):2532–2552. https://doi.org/10.1002/smj.2681CrossRef

Wang T, Chen Y (2018) Capability stretching in product innovation. J Manage 44(2):784–810. https://doi.org/10.1177/0149206315594847CrossRef

Wang P (2019) Price space and product demography: evidence from the workstation industry, 1980–1996. Res Policy. https://doi.org/10.1016/j.respol.2019.05.007CrossRef

Wiggins RR, Ruefli TW (2005) Schumpeter’s ghost: Is hypercompetition making the best of times shorter? Strateg Manage J 26(10):887–911. https://doi.org/10.1002/smj.492CrossRef

Withers MC, Ireland RD, Miller D, Harrison JS, Boss DS (2018) Competitive landscape shifts: the influence of strategic entrepreneurship on shifts in market commonality. Acad Manage Rev 43(3):349–370. https://doi.org/10.5465/amr.2016.0157CrossRef

Yu T, Cannella AA (2007) Rivalry between multinational enterprises: An event history approach. Acad Manage J 50(3):663–684. https://doi.org/10.5465/amj.2007.25527425CrossRef

Zakrzewska-Bielawska A, Czakon W, Gantert TM (2022) Old guards or new friends? Relational awareness and motivation in opportunities seizing. Eur Manage J. https://doi.org/10.1016/j.emj.2022.06.003CrossRef

Zuckerman EW (2000) Focusing the corporate product: securities analysts and de-diversification. Adm Sci Q 45(3):591–619. https://doi.org/10.2307/2667110CrossRef

Advanced Search

Search Operators:

Springer Professional

Hunted hunter: the role of competitive comparison in product survival

Abstract

Abstract

Supplementary Information

Publisher's Note

1 Introduction

2 Theory and research hypotheses

2.1 Competitive comparison with the focal firm’s product

2.2 AMC framework of competitive dynamics research

2.2.1 Awareness

2.2.2 Motivation

2.3 Influence of the competitive comparison on the survival of the focal firm’s product

3 Methodology

4 Sample, data, and variables

4.1 Sample

4.2 Dependent variable

4.3 Independent variable

4.4 Control variables

5 Results

6 Discussion and conclusions

6.1 Managerial implications

6.2 Theoretical contributions

6.3 Limitations and future research

Acknowledgements

Publisher's Note

Supplementary Information

Other articles of this Issue 12/2025

Expanding overseas: the contribution of board role performance from an emerging market perspective

When do entrepreneurs form intermediary relationships? A study of start-ups that use venture capital law firms in fundraising

The coopetition management process: a science mapping analysis

Explaining hotel breakfast pricing under spatial heterogeneity and competitive environments

‘Burned out’ millennials! Can job crafting promote employee well-being in the SMEs?

Values and subjective well-being of European entrepreneurs: a configurational analysis across technological levels

Premium Partner