08/02/2014
Corporate strategy
and competitive
advantage
Bachelor Thesis Business studies Student: Astrid Mangnus Student nu mber: 10000392 Supervisor: Hung-Yao Liu2 Abstract
The literature on corporate strategy is not unanimous on its effect on profitability. Influential papers (Schmalensee, 1985, Rumelt, 1991, McGahan & Porter, 1997) have argued that any corporate effect on profitability is negligible. However, in the last decade the importance of corporate strategy is perceived valid again (Bowman & Helfat, 2001). As more results (Chang & Singh, 2000; Ruefli & Wiggins, 2003; Misangyi et al., 2006) support the idea that corporate efforts have an impact on performance, it is a logical next step in this line of research to study the types of corporate strategy that are of the greatest influence. The present study links the return on assets to the corporate strategies of a sample of U.S.-based multi-business firms. The goal of this research is to provide clarity to multi-business firms about which strategies to pursue and to contribute to the academic debate about the importance and impact of corporate strategy. The results indicate that a cost leadership strategy leads to the greatest monetary performance amongst its competitors.
Introduction
The dynamics of business are in a constant forward motion. In most industries, this leads to fierce competition among firms, and to a battle in which short-lived success is by no means a guarantee for a persistent outperforming of competitors. In this study, the performance of multi-business firms is the focal point. The main goal is to gain further insights into the effect that different types of corporate strategy have on the profitability of multi-business firms. The management layer to which the business level reports formulates a firm’s corporate strategy. This corporate strategy deals with the ways in which a corporation manages a set of businesses together (Grant, 1995), as opposed to business strategy, which deals with the ways in which a single-business firm or an individual business unit of a larger firm competes within a particular industry or market. Therefore, corporate strategy is especially important in multi-business firms, which consist of smaller business units that report to the top management team.
3 Although a corporate effect on performance has been established, a brief overview of the existing literature will show that there is still a lot of debate about the levels of influence and importance of corporate strategy. This discussion originated in the 1980’s, when a debate rose about the influence that the corporate level and its strategy has on a multi-business firm’s performance. Authors such as Schmalensee (1985) and Rumelt (1991) presented results that attributed the level of performance to the business and firm levels. After Bowman & Helfat (2001), a number of other authors (Chang & Singh, 2000; Ruefli & Wiggins, 2003; Misangyi et al., 2006) have further established that there is in fact an effect of corporate strategy on performance. This effect is therefore well researched, although researchers’ findings still differ in their conclusions on its magnitude. Furthermore, little is known about the exact corporate strategies that lead to the highest performance. It is a logical next step in the literature to further study the nature of these corporate effects on performance and which actions yield which results.
The goal of this study is thus to bridge the gap in the current literature by gaining further insights into the effects that different types of corporate strategy have on a multi-business firm’s performance.Which are the most prominent corporate strategies at the moment and what kind of effect do they have on the performance and possible competitive advantage of the firms that implement them? The present study investigates this research question by focusing on the performance of multi-business firms. Firstly, a literature review shapes an insight into different views of corporate strategy, leading to present-day ideas on the topic that can be grouped into types. From this analysis, Porter’s (1980) generic strategies emerge as the most solid strategic types that are fit for an empirical model. The two strategies of Cost Leadership and Differentiation are selected to be included in a multiple linear regression model. Secondly, a group of multiple business firms is identified, whose performance and level of possible competitive advantage are measured. This sample of multiple business companies and the data on their profitability are retrieved from Compustat Historical Segments database and span a period starting in 1997 and ending in 2012. Thirdly, the other factors that might influence performance are identified and implemented in the
4 model as well. After completing these three steps a regression is made from which the relationship of the strategy types to firm performance can be established.
The regression analysis does not yield the exact expected outcome. The differentiation efforts measured in the model have in fact an adverse effect on performance. Of the cost leadership efforts, only one out of the two has a significant positive effect on performance. This could imply a number of things. The first one is that cost leadership has a stronger effect on performance. This would mean that cost leadership is the best strategy to follow when a firm wants to enhance its performance. It could thus be concluded that differentiation is the lesser strategy, since all firms can naturally be expected to seek profit maximization. However, quality, a broad range of products and customer satisfaction are also examples of business elements that are vital to this strategic type and are not being measured in the current operationalization of performance. However, the model and its output are valid and therefore indicate a significant phenomenon. Furthermore, the literature does support the outcome of the model and supports the greater effect on return on assets when a cost leadership strategy is followed. This implies that, although model, data and strategic types might all be limited in a way, it is justified to conclude from the outcomes that cost leadership is the corporate strategy that leads to the greatest firm performance.
The insights of this study can firstly advance the search for an answer to the question in which way corporate strategy matters. By showing how two different kinds of corporate practice influence firm performance not only the effect is shown, but a basic foundation is made for the distinguishing of best practices for corporate strategy in an empirical way. Secondly the knowledge of the concept of corporate strategy itself is deepened by the results of this study. From the literature review it appeared that few studies researched the effects of corporate strategy in an empirical model using a database. Finally, in light of the current literature confirming the significant effect of corporate strategy, it is highly relevant for firms in practice to establish which are strategy types are the best ones to choose when aiming for a deliberate outperforming of competitors. The results of this study might provide a direction for these firms to pursue.
5 Theoretical Background
Outperforming the competition through strategy
Convergence theory (Arrow & Debreu, 1954; Makowski & Ostroy, 2001) argues that all exceptional performers in a sample of firms ultimately converge towards the mean profit, making a pattern of true superiority impossible. The outperforming of competitors in a free market is therefore a phenomenon that is at the center of discussion, with critics arguing that it is an illusion. However, obstacles such as competition, imitation and resource redeployment get in the way of convergence and lead to chances of superior performance (Chakar &Vissa, 2005). Since authors such as Lippman & Rumelt (1982), Jacobsen (1988) and Mueller (1977) have studied the possibility of and causes for superiority of certain firms over others, even in market equilibrium, this phenomenon has become the subject of many more studies and gained its place in present-day research.
Mueller (1977) hypothesizes that the converging of all profits towards one mean will not always occur in a sample of firms. Indeed, according to his results, above or below average performance is not always a temporary disequilibrium. Lippman & Rumelt (1982) study entry into a market where profitability is high. According to convergence theory, the abnormally high profitability should attract new entrants who all want their share, bringing costs and profits back into equilibrium again. However, Lippman & Rumelt find that even under conditions of free entry, the abnormal performance of firms in this market could suggest an inimitability that actually discourages possible entrants. Jacobsen (1988) finds that there are three factors that management can influence when aiming for a true outperforming of competitors: vertical integration, market intensity and market share. Consequently, a firm is able to influence its profitability through its strategic choices. This makes strategy an important factor to consider, since the corporate strategy of choice has an impact on how the corporate elements mentioned by Jacobsen are organized. The following paragraph will therefore look into detail at the ways in which corporate strategy has been researched and how it has been found to influence performance. Subsequently, the last paragraph will elaborate on the
6 ways in which corporate strategy has been theorized and which types and frameworks are developed.
The influence of corporate strategy on performance
A vast body of literature has been written on corporate strategy following its academic ascent in the 1980’s. However, the results of studies seeking to verify the relevance of efforts made at a firm’s corporate level have not always supported their actual importance for a firm’s performance. Schmalensee (1985) hypothesizes three points of view on the most important influence on the profitability of a firm. Firstly, in a classical economist point of view, profitability is possible in market equilibrium by collusion and entry barriers put up by the established firms. Secondly, the revisionist viewpoint argues that rather than through the exclusion of competition, a firm becomes profitable because of its rate of efficiency. Lastly, the author introduces the managerial viewpoint, where profitability is explained by managerial practices. After testing the importance of market share, firm effect and industry effect with a descriptive model in an analysis of variance framework, Schmalensee (1985) found that the industry determines 75% of the profits of a firm, with market share determining most of the other 25% and a negligible role for firm efforts. He expresses surprise about this result in his conclusion section. He adds that although the results are valid, the absence of firm effects might not be a generally applicable result.
To better distinguish stable from fluctuating effects, Rumelt (1991) extends Schmalensee’s study over a longer period of time. This is done by measuring the profits of firms in Schmalensee’s 242 4-digit FTC-industries over four years instead of one. Rumelt’s approach yielded results that emphasized the importance of the business unit level for profitability over that of the industry level. In this study, the effects of corporate efforts appeared to be negligible as well. Research in the subsequent decade elaborated on the findings of Schmalensee (1985) and Rumelt (1991). Examples of these are McGahan & Porter (1997), who support Schmalensee’s findings, and Roquebert, Philips
7 & Westfall (1996), who support Rumelt’s findings while also detecting a small corporate effect on profitability.
In their 2001 article, Bowman & Helfat address the virtual absence or minor presence of a corporate effect. They argue that the absence of corporate effects on profitability in the aforementioned studies is merely due to methodological choices. They find a significant result with respect to corporate strategy by analyzing a considerable number of empirical studies that use variance decomposition techniques. In the studies they analyze, corporate effects at times explain 18% or more of profitability, which is the case in Roquebert, Philips & Westfall (1996), one of the studies that are included. The methods used by Schmalensee (1985), Rumelt (1991) and other authors are listed and subsequently discussed for underestimating the corporate effect. This possible underestimation is likely caused by a number of factors. First of all, the samples used in research on corporate effect often include single-business firms, which by definition means that there is only a firm effect and no possible corporate effect. Secondly, the broad definition of industry and business level effects cannibalizes the corporate effect. Thirdly, variance decomposition method of analysis is a hierarchical regression and the order of the variables can therefore have a crucial impact. Lastly, in not all of the studies covariance was measured, thereby possibly reducing the measured corporate effects.
These results mark the beginning of a renewed emphasis on corporate effects and corporate strategy, as in the wake of Bowman & Helfat’s 2001 article, a number of other studies have emerged that support their results and elaborate on them. A first example of this comprises further study in specific sectors, as is done for manufacturing by Chang & Singh, 2000. By using market share as an indicator of competitive position instead of Rumelt’s (1991) variance decomposition technique, Chang & Singh find that corporate effects on market share are significantly larger than zero. The influence of the corporate level is especially present in small but developing business units and less in established ones. A second focus of study is employed by Misangyi et al. (2006) as they use a multilevel approach to investigate the relative importance of firm, industry and corporate efforts.
8 Their results show that while business segments, the management level that is closest to actual operations, play the most important role in influencing a firm’s performance, industry and corporate strategy are also of vital importance. Industry concentration and prosperity make the industry level relevant for a firm’s position and its performance. At the same time, the resources that the corporate level allocates determine performance in a crucial way as well.
Corporate strategy in theory
The previous paragraph shows that the current state of the literature indicates a significant role for the corporate level in influencing the profitability of a multi-business firm. Strategic decisions at this level are therefore of vital importance. Corporate strategy is the pattern of decisions in a firm that determines its objectives, purposes or goals. It also produces the principal policies and plans for achieving those goals. Furthermore, it defines the range of businesses the company is to pursue, the kind of economic and human organization it intends to be and the nature of the economic and noneconomic contributions it intends to make to its shareholders, employees, customers and communities. It consists of two parts: formulation and implementation (Andrews, 1987). The present study will focus on the completed formulation and the effect that the direction chosen by a firm has on its profitability. Since the 1980s, the managerial decisions that lead firms into a certain direction or position have been subject to extensive research. Several studies, which will be discussed below, have tried to capture generic strategies or to identify strategic types: established directions that are highly probable to lead to a superior competitive position.
In 1978, Miles et al. proposed a theoretical framework that described the ways in which firms formulate their corporate strategy and their way of reaching their strategic goals. This led to the formulation of four types. Firstly, the defender’s strategy is to capture and defend a stable territory within the marketplace. Secondly, the prospector pursues an opposite strategy: it is constantly shifting territories in an ongoing stream of innovations. Thirdly, the analyzer adopts a corporate strategy that combines both of the previous two dependent on the situation the firm is in, making it
9 the best way to reach a firm’s strategic goals according to the authors. Finally, the reactor is considered a failure: its corporate strategy is both inconsistent and unstable. On a comparable note to Miles et al. (1978), the work of Michael Porter rose to prominence in the 1980’s. Porter (1980) formulates corporate strategy as the answer of a firm to the question how they should react to five key external forces: the threat of new entrants, the bargaining of suppliers and buyers, the threat of substitution and the competition within the market. In 1980, he published a book in which he defines three generic strategies, of which firms must choose one. Many authors have elaborated on his framework, some of which are included here for clarification.
The differentiation strategy seeks to create value through customer loyalty, price inelasticity, and higher margins, which they achieve via brand image, advertising intensity, premium pricing, novelty, fashion and exclusive distribution networks (Chaganti et al., 1989, Mason et al., 1991). Associated with differentiation are innovative and growth strategies, which according to Galbraith & Merrill (1991) are more likely to yield results in the long run than in the short run. Also, there is a greater chance of failure compared to less innovative strategies. Innovative and growth strategies also adversely affect short-term earning and performance measures (Rappaport, 1978). Research finds a greater use of non-financial measures in companies pursuing strategies relating to innovation and/or quality (i.e. differentiation strategy). It can be expected that lower emphasis is placed on contemporaneous accounting measures in firms pursuing differentiation strategy, as the performance of such firms will be evaluated based on a broader set of measures, including non-financial ones (Ittner et al., 1997). The antithesis of the differentiation strategy is the cost leadership strategy. This strategy yields low cost producers for a given level of quality (Porter, 1980). Consumers buy these products because they are priced below competitors' equivalent products (Hambrick, 1983). A firm that pursues cost leadership will normally sacrifice high profit margins and instead generate profits through volume, placing a high volume-low margin approach at the center of their businesses (Rea and Kerzner, 1997, Daly, 2001, Campbell et al., 2002, Peng, 2008). Rea and Kerzner
10 (1997) argue that cost leadership requires high market share, while Daly (2001) states that one way of pursuing a cost leadership strategy is to be the high volume producer.
The strategy designs of Miles et al. (1978) and Porter (1980) are arguably quite similar in the categories they establish for firms to fit into in order to outperform its competitors: it is either about cost or about quality. What all authors seemingly agree on is the importance of pursuing some strategy. When companies fail to adopt one of the proposed strategies, the assumption is that they will fall prey to competitors who did deliberately adopt one. Porter (1980) calls that situation being stuck in the middle; other authors have designed strategic types for it such as the Miles et al.’s (1978) reactor. As it becomes clear from the described development of corporate strategy, the mentioned authors tend to focus on the placement of a company within an industry. This view is a defining characteristic of the positioning school, which reached its peak in popularity in the 1980’s. In 1999, Mintzberg developed a framework of ten schools, into which he grouped different organization types and the strategies they yield. Mintzberg’s final conclusion is that the schools are in fact a constraint and that research should strive towards one answer to strategic management and implementation, a blend of all the schools.
When looking at more recent literature on corporate strategy, others share Mintzberg’s opinion. Ruefli & Wiggins (2002) studied the drivers for competitive advantage, and found first of all that it is extremely hard to achieve. Furthermore, their findings show that instead of positioning the highest performing firms achieved this by using inimitable resources. In addition to rejecting the rigid positioning school, a second important focus is on the changing nature of business and therefore dynamic corporate strategies. Adner & Helfat (2003) found that corporations change their structures, control systems and other attributes over time. Therefore, the corporate effect generated by the corporate strategy changes inevitably. Another pair of authors that discusses the need for dynamics in strategy is Doz & Kosonen (2007, 2010). The results from their study of three years of in-depth research in a dozen large global companies show a new leadership style of very strategically agile individuals with a great need for power and independence. What the papers by Adner & Helfat and
11 Doz & Kosonen illustrate is a rejection of rigid and generic corporate strategies, in favor of a dynamic approach led by a highly skilled top management. However, in contrast with the seemingly new view on strategy by the mentioned authors that published more recently, generic strategy types are all but forgotten. This stems mostly from their ease of use in empirical research. Authors such as Balsam, Fernando & Tripathy (2011) still use cost leadership and differentiation as the two strategy types in their study on how strategy influences executive compensation.
Methodology
Defining Types of Corporate Strategy
The goal of this study is to bridge the gap in the current literature by gaining further insights into the types of corporate strategy on performance. The previous paragraph has explored the existing corporate strategies. When defining the appropriate way to incorporate these strategies into an empirical method for testing the effects on profitability, a couple of options have to be considered. The first thing that can be concluded from the literature review is that authors have either attempted to measure Porter’s generic strategies empirically or have tried to venture into new strategies (Parnell & Wright (1993), Kotha & Vadlamani (1995)). Campbell-Hunt (2000) signaled it as a common problem that although there is an extensive body of research employing generic strategies, there is very little consensus in the empirical methods that are used, resulting in studies that are difficult to compare. After employing meta-analysis to aggregate 17 empirical papers researching generic strategies, the author finds that Porter’s (1980) generic strategies of cost leadership, diversification and focus are highly important, but he suggests that there are more interesting strategic directions to follow, encouraging future researchers to refine Porters strategies. Many authors have thought about classifications of corporate strategy types and the majority of them are derived from theory. An article that could be said to provide a refinement to Porter’s strategies while simultaneously shaping more types of strategies into an empirical model is a 1983 study by Galbraith
12 & Schendel, who distinguish between consumer goods and industrial products: six strategy types were identified for consumer products namely harvest, builder, cashout, niche or specialization, climber, and continuity, while four were identified for industrial products: low commitment, growth, maintenance and niche or specialization. These types are designed for empirical use by using data from the SPIYR database, in which a cluster analysis is performed. A second way in which researchers have tried to formulate their own strategic types is the conduction of a survey. By asking managers directly about the goals of and direction within the company, authors have an opportunity to employ a bottom-up approach to discover strategy types. However, most researchers connect their results to generic strategies such as Porter’s strategies from 1980 (Kumar & Subramanian (1998)) or Mintzberg’s from 1988 (Kotha & Vadlamani (1995)). They do not formulate new types as Galbraith & Schendel (1983) did, but confirm the effectiveness of the three generic strategies and therewith the disadvantage of being “stuck in the middle”: not adopting a strategy at all. Since the surveys are mostly carried out on a smaller scale than is achievable with the use of a database, and often in a confined firm or industry, the results are not directly applicable in this study.
Since all of the aforementioned papers acknowledge the importance of Porter’s (1980) generic strategies, the method that this study employs is the one formulated by Balsam, Fernando & Tripathy (2011). In their article, the strategic position of firms is operationalized through a number of variables, which the authors have derived from the Compustat, CRSP and Execucomp databases. The variables can be divided in two groups, which respectively stand for a cost leadership strategy or a differentiation strategy. Porter’s third generic strategy, focus, is not included in the method, since Balsam, Fernando & Tripathy (2011) argue that it is an extension of the differentiation strategy as it focuses on high quality output as well. The only difference is that in the case of a niche strategy, only one specific high-quality product is offered instead of many. This, they say, makes it hard to distinguish between the two in a model and overlap may occur. This assumption can be disputed, since in his book Porter explicitly states that the focus strategy is “quite different” from the other two. However, he does split the focus strategy into two forms: cost focus, where the firm tries to
13 reach the lowest cost for its target segment, and differentiation focus, which occurs when a firm tries to differentiate as much as possible in its niche. This definition does imply that the focus strategy will always have a certain overlap with either the differentiation or the cost leadership strategy. Because of this and since the present study adopts Balsam, Fernando & Tripathy’s method, their reasoning with regard to the focus strategy is accepted in this case as well. The focus strategy is therefore not included in the model.
Based on the strategy profiles of the differentiation strategy and the cost leadership strategy as they are also described in the literature review, Balsam, Fernando & Tripathy (2011) have designed six variables, three for each strategic type. In their study, consequently, the variables are used to connect strategy with methods of executive compensation. In the present case, it will be connected to corporate performance. Although this study does not employ the more elaborate method of Galbraith & Schendel (1983), it will provide a more general outcome than a study using surveys and will connect to the existing literature using Porter’s generic strategies. After developing a model and running a regression, this method makes it is possible to determine whether there is a significant difference in profitability for any of the mentioned variables that indicate a strategic type.
Sample
Multi-business firms exist in several forms, from being very diversified to consisting of a specialized set of business units. In this study, the performance of multi-business firms will be obtained through Compustat. Rocquebert, Philips & Westfall (1996) operationalized multi-business firms as companies with two or more businesses or business units. By explicitly studying multi-business firms, the authors generated results that support the findings of Rumelt (1991) but also suggested for the first time the existence of a corporate effect. The sample used in this study is derived from Standard & Poor’s Compustat Historical Segment database. The derived Historical Business Segment data ranges from 1997 until 2012. From this segment data it is possible to derive the number of segments in which a corporation is active, so single-business firms can be filtered out according to the suggestions
14 of Bowman & Helfat (2001). The level of diversification can be determined as well by way of the NAICS-codes. The chosen interval that the Compustat-data spans is adequate for measuring firm performance whilst controlling for an exceptional year. It has as an additional advantage because it is a time span for it is possible to use solely NAICS-codes. This prevents them being mixed with SIC-codes. Since NAICS-codes replaced SIC-codes in 1997, the two different codes could prove a complication for the final model when used together.
The dataset consists of 804,593 observations, each of a specific business unit in a certain year.Following McGahan & Porter (1999), 97091 observations that do not contain a primary NAICS code and thus do not have a core business are eliminated. Furthermore, 146,435 corporations with sales less than 10 million are removed. In addition, 97,070 observations of corporations with total assets less than 10 million are removed as well. According to McGahan & Porter, these data could be particularly affected by anomalies rather than operating activities. After the filtering actions, the original dataset provides a sample of 561067 business segment observations over sixteen years. After preparing the sample in this manner, it is ready for the creation of the variables that are needed to identify Porter’s strategies of cost leadership and differentiation. Subsequently, a benchmark for superior performance can be formulated in order to estimate the relationship between the elements of the model.
Dependent variable
The variable Performance, or Competitive Advantage, can be operationalized as the prolonged benefit of implementing some unique value-creating strategy that is not simultaneously being implemented by any current or potential competitors. Additionally, it possesses an inability to duplicate the benefits of this strategy (Hoffman, 2000). The performance of the firms in the sample is measured in return on assets (ROA), an accounting construct, since it is among the most widely used measures for performance in studies with a related subject (Schmalensee, 1985, Rumelt, 1991, Bowman & Helfat, 2001, 2003, Ruefli & Wiggins 2003). After constructing the variable, ROA is
15 connected to each of the strategic types using a regression analysis in Stata. Return on assets is an accounting tool that is made out of the ratio of operating profit to total assets (Ruefli & Wiggins 2003). For the used sample these necessary data items were derived from Compustat Business Segment data and created into the variable ROA. After the creation of the independent variables, elaborated in the following paragraph, the missing observations of ROA are dropped from the sample in order to create a dataset with equal observations for each variable. 10 observations were deleted.
Independent variables
Balsam, Fernando & Tripathy (2011) have listed six variables that signal the type of strategy that firms have adopted, all of those ratios are derived from Compustat Historical Segments data. The first three variables indicate a differentiation strategy if their numbers are high. The first variable, which is labeled Differentiation 1, is the ratio of selling, general and administrative expenses to net sales. According to the authors, firms pursuing a differentiation strategy will invest in a variety of marketing activities (which will materialize in higher SG&A) to differentiate themselves from competitors. The second variable, Differentiation 2, is the ratio of research and development to net sales: a key success factor for a firm pursuing a differentiation strategy is the offering of high quality and innovative products and services. It is likely that firms pursuing a differentiation strategy spend bigger budgets on research and product design. Thirdly, the variable called Differentiation 3 is the ratio of net sales to cost of goods sold. The success of a firm adopting a differentiation strategy is measured by its ability to command higher prices. Consequently, a higher level of SALES/COGS can reasonably be expected to be associated with a differentiation strategy.
The last three variables indicate a cost leadership strategy. The first of those and consequently the fourth variable is the ratio of net sales to capital expenditures on property, plant and equipment. In the analysis, it is referred to as Cost Leadership 1. A higher SALES/CAPEX indicates a more efficient use of assets, signaling an attempt to lower costs and to be subsequently able to lower prices. The fifth variable, Cost Leadership 2, also a signal of higher efficiency, is the ratio of net
16 sales to net book value of plant and equipment. In this case, it indicates a low-budget policy in the acquisition of plant and equipment. Finally, the ratio of employees to total assets is used to measure labor efficiency and is labeled Cost Leadership 3. Since human capital is a costly asset for any firm, a lower relative number of employees can for example indicate a sparse hiring policy, but it can also mean that the standardized products do not need much manual labor to manufacture them. In the Compustat sample, combining the aforementioned data elements into ratios creates the six variables. To create an equal set that contains equal numbers of observations for each variable, missing values are dropped from the sample. This results in 367232 eliminations for Differentiation 1, 44632 for Differentiation 2, 625 for Differentiation 3, 1822 for Cost Leadership 1, 1608 for Cost Leadership 2 and 3387 for Cost Leadership 3. After dropping these entries, the final dataset consists of 44681 observations.
Control variables
In addition to the main variables, a number of control variables are included in the model in order to make it more thorough. The first control variable is Timetrend, which controls for anomalies n the model that become apparent over the sample’s time span (Collins, Maydew & Weiss, 1997). The second control variable is Firmsize. It is operationalized as the natural log of net sales (Hovakimian, Hovakimian & Tehranian, 2004). This controls for the different sizes of the firms in the sample and the effect that this might have on their performance as measured by the model. Lastly, since the sample spans all industries in the Compustat Historical Segments database, it is necessary to control for the effect that these industries could have on the outcome of the model. Based on the firms’ four-digit NAICS-code, which indicates their industries, 683 dummies are created. The dummies are jointly integrated in the model to form the controlling Industry Dummies (Blomstrom & Sjoholm, 1999).
17 The first step in converting the established variables into an empirical model is the assessment of their validity. To ensure that the variables that were created based on previous research measure the two types of strategy, an exploratory factor analysis is conducted.
Table 1a. Exploratory factor analysis of Differentiation 1-3 and Cost Leadership 1-3
Variable Differentiation Cost Leadership Uniqueness
factor loading factor loading
Differentiation 1 0.9106 -0.0172 0.1705 Differentiation 2 0.9113 -0.0186 0.1691 Differentiation 3 0.1452 0.4923 0.7366 Cost Leadership 1 -0.0532 0.5852 0.6547 Cost Leadership 2 -0.0640 0.7638 0.4125 Cost Leadership 3 -0.3510 -0.1586 0.8516
As can be seen in table 1a, not all of the variables adopted from the study by Balsam, Fernando & Tripathy (2011) contribute to the accuracy of their respective constructs in this case. The variables Differentiation 3 and Cost Leadership 3, marked in italics, do not measure a common factor and measure even an opposite or the other factor. Based on these findings it can be concluded that the theory does not match the reality of this particular sample and situation. It would therefore make the model more accurate if Differentiation 3 and Cost Leadership 3 are no longer included here. A new factor analysis, of which the results are displayed in table 2b, confirms the benefit of this decision for the two constructs. Two clear factors can be distinguished from this second test that are also both convincingly measured by the variables that were constructed for the Differentiation and Cost Leadership strategies.
Table 1b. Exploratory factor analysis of Differentiation 1-2 and Cost Leadership 1-2
Variable Differentiation Cost Leadership Uniqueness
factor loading factor loading
Differentiation 1 0.9318 -0.0134 0.1315
18
Cost Leadership 1 0.0024 0.7581 0.4253
Cost Leadership 2 -0.0333 0.7537 0.4308
Empirical model
When incorporating the verified variables from the exploratory factor analysis it is possible to construct the following model, designed for a multiple linear regression analysis:
ROAi,t =α0 + β1Firmsizei,t+ β2Timetrendi,t+ β3Industry,i,t+ β4Diffi,t + β5CLi,t + Industry Dummies + εi,t
where for firm i in year t
ROAi, t Return on assets. Ratio of operating profit to identifiable total assets. Firmsize Natural log of net sales. Controls for the size of firms.
Timetrend Construct to measure yearly trends
Industry Set of industry dummies based on NAICS-code
Diff Construct to capture differentiation strategy. Continuous variable based on the factor analysis of the t-1 to t-16 ratios of SG&A Expense to Sales and R&D expenses to Sales
CL Construct to capture cost leadership strategy. Continuous variable based on the factor analysis of the t-1 to t-16 ratios of sales to capital expenditures and sales to net book value of plant and equipment
Results
Descriptive statistics
19 units, measured over a sixteen-year time span, each firm has approximately 2 business units on average. In Stata, the six independent variables that were formulated by Fernando, Balsam & Tripathy (2011) were reconstructed creating ratios using the Compustat Historical Segment variables
Selling, General & Administrative, Net Sales, Research and Development, Cost of Goods Sold, Capital Expenditures, Property, Plant & Equipment, Employees, Identifiable Total Asset and Operating Profit.
This has resulted in the variables Differentiation 1 and 2 and Cost Leadership 1 and 2 as well as the dependent variable Return on Assets. Table 2 lists the descriptive statistics of the created variables, as observed after the clearing of the sample.
Table 2. Descriptive Statistics
Variable Obs Mean Std. Dev. Min Max
Differentiation 1 44681 .4849099 .5525495 .0013118 20.71154 Differentiation 2 44681 .1342592 .2564815 -.0023181 9.503352 Cost Leadership 1 44681 80.5486 1198.408 -483.3975 136250 Cost Leadership 2 44681 15.62816 58.40436 .0136783 5664.957 ROA 44681 .0030879 .2347603 -6.71857 1.503757 Regression results
The first step in the regression analysis is the computation of Pearson’s correlation coefficients, listed in table 3a. Pearson’s coefficients measure the relationships of the different variables in the model to one another.
Table 3a. Correlation coefficients
Pearson ROA Diff 1 Diff 2 Cost 1 Cost 2 Timetrend Firmsize ROA 1.0000
Diff 1 -0.5498 1.0000
Diff 2 -0.4357 0.7374 1.0000
20 Cost 2 0.0315 -0.0315 -0.0329 0.1432 1.0000
Timetrend 0.1146 -0.0761 -0.0392 0.0217 0.0437 1.0000
Firmsize 0.2912 -0.1691 -0.0903 -0.0251 -0.0995 0.1491 1.0000
After establishing Pearson’s correlation coefficients, the multiple linear regression analysis is conducted. The outcomes of the regression analysis that was based on the model formulated in the previous paragraph are displayed in table 3b. As can be seen, all of the outcomes are significant, except for cost leadership 1. This is the variable constructed out of the ratio of Net Sales to Capital Expenditures. The higher Cost Leadership 1, the more it is supposed to indicate an efficient use of assets, signaling an attempt to lower costs and to be subsequently able to lower prices. In addition to this one insignificant result, an element that stands out is that the coefficients of the other independent variables are all negative, signaling a negative relationship to Return On Assets. This would first of all mean that Differentiation 1, or the ratio of selling, general and administrative expenses to net sales, indicating the intensity of attempting to differentiate from competitors, leads to a lower ROA. The second independent variable, Differentiation 2, is the ratio of research and development to net sales: offering high quality and innovative products and services. Innovation and quality also influence ROA negatively, according to the regression results. After the third variable, which has proven not to be significant follows the fourth variable, Cost Leadership 2. This is a signal of higher efficiency of assets like its predecessor. It is measured in the ratio of net sales to net book value of plant and equipment. This is the only variable that is significantly positively related to Return On Assets. A third observation is that the three control variables are all three significantly influencing ROA (Selling & Stickney, 1989).
Table 3b. Effects of differentiation and cost leadership strategies on corporate performance ROA Coef. Std. Err. t P>|t| Differentiation 1 -.1848596 .0025622 -72.15 0.000* Differentiation 2 -.0818632 .0054335 -15.07 0.000*
21 Cost Leadership 1 -5.47e-07 7.78e-07 -0.70 0.482
Cost Leadership 2 .0002322 .0000186 12.50 0.000* Timetrend .0017385 .000244 7.13 0.000* Firmsize .0331648 .0006062 54.71 0.000* Industry Dummies -.0904018 .0039704 -22.77 0.000* *p>.001 (n=44681) Discussion
Some elements of the regression output are unexpected. Based on the literature reviewed before conducting the empirical research through the model it could reasonably be expected that both variables indicating Differentiation and Cost Leadership would lead to a higher ROA. This would follow from both Porter (1980), who theorized that firms should never be stuck in the middle, and from Bowman & Helfat (2001), who argued that the decisions made at a firms’ corporate level are of significant influence on the firms’ profitability, next to the industry and business levels. Following the existing literature in their search for the best strategies to pursue, the present study concluded that most researchers connect their results to generic strategies such as Porter’s strategies from 1980 (Kumar & Subramanian (1998)) or Mintzberg’s from 1978 (Kotha & Vadlamani (1995)). Even if they formulate new types, like Galbraith & Schendel (1983) did, either efficiency or high quality is at the core of these. Almost all of the existing empirical papers emphasize the importance of Porter (1980), including the study that designed the method that this paper employs: Balsam, Fernando & Tripathy (2011). What appears to be accepted as a given in many of the mentioned studies is the notion that without a deliberate adoption of a certain strategy companies find themselves “stuck in the middle”: not adopting a strategy at all. It is therefore logical to expect for both of the strategic types that are formulated and measured using an empirical model to have a positive, significant effect on performance. However, this is not the case.
This outcome could either indicate a methodological error or new information regarding corporate strategy that differs from the common point of view in research. Balsam, Fernando &
22 Tripathy (2011) study the linkage of the strategy to executive compensation, while Porter and the present study link strategy to performance. The authors find that while firms pursuing a cost leadership strategy place a significantly higher weight on sales, firms that pursue a differentiation strategy place significantly lower weight on ROA. In addition to this, firms adjust their reward systems to provide management with incentives to pursue the corporate strategy of choice. For example: differentiators may place greater emphasis on non-financial measures of performance, while cost leaders value a high volume of sales since that is where they gain their profits. As such, a circle is formed in which certain objectives are constantly pursued in a certain strategy. The fact that these objectives differ so fundamentally amongst firms that pursue different strategies is a possible hazard for the model used in this study. The fact that firms pursuing a differentiation strategy might value non-financial measures of performance, such as quality or long-term customer commitment, could implicate that return on assets is an operationalization of performance that favors the cost leadership strategy. Since the cost leader strives to achieve higher return on assets, the model could falsely attribute to them a higher performance. However, return on assets is, as described before, a widely used and generally applicable measure of performance in studies with a related subject (Schmalensee, 1985, Rumelt, 1991, Bowman & Helfat, 2001, 2003, Wiggins & Ruefli, 2003). Considering the fact that this study focuses on for-profit firms, they can be expected to search profit maximization (Wiggins & Ruefli, 2002). If this profit maximization is assumed as a given, return on assets would, in light of previous studies, be the optimal measure for performance, verifying the outcome of the present study and proving it to be relevant.
The fact that the literature at this moment already suggests a far more elaborate picture of corporate strategy than the empirical possibilities is also a point of discussion for the model used in the present study. Although Porter’s generic strategies are still widely used in empirical papers, an extensive body of literature written by different authors has followed it up, arguing different views and measurements of strategy. An important focus lies on the changing nature of business and therefore dynamic corporate strategies. Adner & Helfat (2003) found that corporations change their
23 structures, control systems and other attributes over time. Therefore, the corporate effect generated by the corporate strategy changes inevitably. In their study, they develop a descriptive model and measure the corporate effects on profitability in U.S. multiple business firms over the period 1977-1999. They find that managerial human capital, social capital and cognition determine strategic decisions and outcomes. These dynamic managerial capabilities are crucial in a firm’s strategy and in whether it will be successful or not. Another pair of authors that discusses the need for dynamics in strategy is Doz & Kosonen (2007, 2010). In the first of the two studies, they conducted three years of in-depth research in a dozen large global companies. The results showed a new leadership style at the top employing very strategically agile individuals with a great need for power and independence as managers. They formulated a strategic agility framework, consisting of the three dimensions strategic sensitivity, leadership unity and resource fluidity, to frame this “new deal”. In their 2010 article, these three dimensions were complemented by a set of suggested actions for a firm’s top management. These examples suggest a more holistic view of firms and their strategies, connecting elements such as efficiency and performance to not only the behavior of a firm, which can be measured externally, but also a firms architecture and internal organization. The incorporation of the more recent theoretical findings into an empirical model could provide a more complete view and could possibly explain the unexpected outcome of the model.
At the moment the model is still valid and reliable, meaning that the results can impossibly be merely a methodological error. It appears therefore that, within the boundaries of the present study, cost leadership is the more effective strategy for a firm that seeks to maximize its monetary performance. This is a significant result because first of all it shows that differentiation is a strategy that could not only be less effective but even counterproductive to a firm’s monetary performance. Secondly it clearly shows the more effective strategy, which is one of volume, efficiency and the offering of goods for the lowest possible price: cost leadership. These findings are significant for theory as well as practice. For the existing literature, the results of this study challenge the general idea that the two generic strategies of cost leadership and differentiation achieve the same goal,
24 namely an optimal performance in terms of revenue, employing different ways. According to the present study, the results are not the same. This is also a useful indicator for managerial practice. It signals the direction that the corporate management of a firm should go when seeking maximization of return on assets or performance in the broader sense. It also shows that although a differentiation strategy is proven to be effective on many fronts such emphasizing and improving quality and customer relations (Balsam, Fernando & Tripathy (2011), monetary performance is not necessarily one of them. Consequently, when considering pursuing a strategy that focuses on quality, customer satisfaction and innovative product strategies, among others, the results of this study show the downside of that choice. It provides an insight into the implications regarding profit maximization, commonly accepted as a goal that all firms seek to achieve, when a multi-business firm is not pursuing a cost leadership strategy.
Further research would possibly lead to a clarification of some of the points of doubt mentioned above. The regression could be conducted again, incorporating more indicators for strategy rather than limiting the model to two strategic types. Whereas the present model provides an indication of the effectiveness of the differentiation and cost leadership strategies, the literature suggests that there are more types that are interesting directions for further study. Often, these strategies are derived from theory or small-scale surveys. By extracting these already partly verified strategic types and implementing them in a large-scale model using many observations from a database, the discussion of corporate strategy could be advanced and deepened along the lines that the present study has set up. Another adjustment that could be made if the model was tested again is the measure of performance. In the present study, performance was measured as return on assets. This way, elements such as for example relative price that a firm can ask for its goods or return on assets relative to price level were not taken into account. While it is a common way to measure performance that is also used in the majority of other studies on this subject, a more elaborate operationalization of corporate performance might provide a useful addition. The understanding of
25 performance, especially when it is related to strategic types that each have different objectives, could be further deepened by doing so.
Conclusion
This paper studies what the most prominent corporate strategies are at the moment and what effect they have on the performance and possible competitive advantage of the firms that implement them. This study does not yield the outcome that was expected beforehand. The differentiation efforts measured in this analysis have an adverse effect on performance. Of the cost leadership efforts, only one has a significant positive effect on performance. This could imply a number of conclusions. The first one is that cost leadership has a stronger effect on performance, at least when performance is operationalized as return on assets. This would mean that cost leadership is the best strategy to follow when a firm wants to enhance its performance. It could thus be concluded that differentiation is the lesser strategy, since all firms naturally seek profit maximization. However, the goals of the differentiation strategy could be different and harder to measure when utilizing return on assets. Quality, a broad range of products and customer satisfaction are vital to this strategic type as well and they are elements that are not being measured in the current operationalization of performance. In addition to this, it could be argued that Porter’s two generic strategies are underestimating the complex nature of organizations. However, the model and its output are valid and therefore indicate a significant phenomenon. Furthermore, the literature does support the outcome of the model and supports the greater effect on return on assets when a cost leadership strategy is followed. This implies that, although the model, data and strategic types might be somewhat limited, it is justified to draw the conclusion from the outcomes that cost leadership is the corporate strategy that leads to the greatest performance.
26 References
Adner, R., & Helfat, C. E. (2003). Corporate effects and dynamic managerial capabilities. Strategic Management Journal, 24(10), 1011–1025.
Andrews, K. R. (1984). The concept of corporate strategy. In N. J. Foss (Ed.), Resources, firms and strategy (pp. 52–60). Oxford Management Readers.
Arrow, K. J., & Debreu, G. (1954). Existence of an Equilibrium for a Competitive Economy. Econometrica, 22(3), 265–290.
Balsam, S., Fernando, G. D., & Tripathy, A. (2011). The impact of firm strategy on performance measures used in executive compensation. Journal of Business Research, 64(2), 187–193. Blomstrom, M., & Sjoholm, F. (1999). Technology transfer and spillovers: Does local participation
with multinationals matter? European economic review, 43, 915–923.
Bowman, E. H., & Helfat, C. E. (2001). Does corporate strategy matter? Strategic Management Journal, 22(1), 1–23.
Campbell-Hunt, C. (2000). What have we learned about generic corporate strategy? A meta-analysis. Strategic Management Journal, 21(2), 127–154.
Campbell, D., Stonehouse, G., Houston, W. (2002). Business strategy. 1st edition: Butterworth-Heinemann.
Chaganti, R.S., Chaganti, R., Mahajan, V. (1989). Profitable strategies under different types of competition. Entrepeuneurial Theory Practice, 13(3), 21–35.
Chacar, A., & Vissa, B. (2005). Are emerging economies less efficient? Performance persistence and the impact of business group affiliation. Strategic Management Journal, 26(10), 933–946. Chang, S.J., & Singh, H. (2000). Corporate and industry effects on business unit competitive position.
Strategic Management Journal, 21(7), 739–752.
Collins, D. W., Maydew, E. L., & Weiss, I. S. (1997). Changes in the value-relevance of earnings and book values over the past forty years. Journal of accounting and economics, 24, 39–67.
27 Daly, J.L. (2001). Pricing for profitability: activity-based pricing for competitive advantage. 1st edition:
Wiley.
Doz, Y. L., & Kosonen, M. (2007). The new deal at the top. Harvard Business Review, 85(6), 98–104. Doz, Y. L., & Kosonen, M. (2010). Embedding strategic agility. Long Range Planning, 43(2), 370–382. Galbraith, C.S., Merrill, G.B. (1991). The effect of compensation program and structure on SBU
competitive strategy: a study of technology intensive firms. Strategic Management Journal, 12, 353–70.
Galbraith, C., & Schendel, D. (1983). An Empirical Analysis of Strategy Types. Strategic Management Journal, 4(2), 153–173.
Hambrick, D.C., MacMillan, I.C. (1984). Asset parsimony — managing assets to manage profits. Sloan Management Review, 25(2), 67–74.
Hoffman, N. P. (2000). An Examination of the “Sustainable Competitive Advantage” Concept: Past, Present, and Future. Academy of Marketing science review, (4), 1–16.
Hovakimian, A., Hovakimian, G., & Tehranian, H. (2004). Determinants of target capital structure: The case of dual debt and equity issues. Journal of Financial Economics, 71(3), 517–540.
Ittner, C.D., Larcker, D.F., Rajan, M.V. (1997). The choices of performance measures in annual bonus contracts. Accounting Review, 72(2), 231–55.
Jacobsen, R. (1988). The persistence of abnormal returns. Strategic Management Journal, 9(5), 415– 430.
Kotha, S., & Vadlamani, B. L. (1995). Assessing generic strategies: an empirical investigation of two competing typologies in discrete manufacturing industries. Strategic Management Journal, 16(1), 75–83.
Kumar, K., & Subramanian, R. (1998). Porter’s strategic types: difference in internal processes and their impact on performance. Journal of Applied Business Research, 14(1), 107–124.
Lippman, S. A., & Rumelt, R. P. (1982). Uncertain imitability: an analysis of interfirm differences in efficiency under competition. The Bell Journal of Economics, 13(2), 418–438.
28 Makowski, L., & Ostroy, J. M. (2001). Perfect Competition and the Creativity of the Market. Journal of
Economic Literature, 39(June), 479–535.
Mason, J.B., Mayer, M.L., Ezell, H.F. (1991). Retailing. 2nd edition: Irwin.
McGahan, A. M., & Porter, M. E. (1997). How much does industry matter, really? Strategic Management Journal, 18(Summer Special Issue), 15–30.
McGahan, A. M., & Porter, M. E. (1999). The persistence of shocks to profitability. The review of economics and statistics, 81(1), 143–153.
Misangyi, V. F., Elms, H., Greckhamer, T., & Lepine, J. a. (2006). A new perspective on a fundamental debate: a multilevel approach to industry, corporate, and business unit effects. Strategic Management Journal, 27(6), 571–590.
Miles, R. E., Snow, C. C., Meyer, A. D., Coleman, H. J. (1978). Organizational strategy, structure, and process. The Academy of Management review, 3(3), 546–562.
Mintzberg, H. (1988). Generic strategies: toward a comprehensive framework. Advances in strategic
management, 5(1), 1-67.
Mintzberg, H., & Lampel, J. (1999). Reflecting on the strategy process. Sloan Management Review, 40(3), 21–30.
Mueller, D. C. (1977). The Persistence of profits above the norm. Econonomica, 44(176), 369–380. Parnell, J. A., & Wright, P. (1993). Generic strategy and performance: an empirical test of the Miles
and Snow typology. British Journal of Management, 4(1), 29–37. Peng, M.W. (2008). Global strategy. 2nd edition: South-Western. Porter, M. E. (1980). Competitive Strategy. Free Press, New York.
Rappaport A. (1978). Executive incentives vs. corporate growth. Harvard Business Review, 56(4), 81– 88.
Rea, P. J., & Kerzner, H. R. (1997). Strategic planning: a practical guide. John Wiley & Sons, Hoboken.
Ruefli, T. W., & Wiggins, R. R. (2003). Industry, corporate, and segment effects and business performance: a non-parametric approach. Strategic Management Journal, 24(9), 861–879.
29 Ruefli, T. W. & Wiggins, R. R. (2002). Sustained Competitive Advantage: temporal dynamics and the incidence and persistence of superior economic performance. Organization science, 13(1), 82– 105.
Rumelt, R. P. (1991). How much does industry matter? Strategic Management Journal, 12(3), 167– 185.
Roquebert, J. A., Phillips, R. L., & Westfall, P. A. (1996). Market vs. management: what drives profitability? Strategic Management Journal, 17(8), 653–664.
