A Ten Country-Company Study of Sustainability and Product-Market Performance

Institutional Theory

As applied in this study, institutional theory holds that to attain legitimacy, a company tends to be isomorphic to other companies in its country and, as a result, over time companies within the same country will resemble each other and be an extension of their environment (Dacin 1997). DiMaggio and Powell (1983) identify three mechanisms – coercive, mimetic, and normative pressures – through which such resemblance of companies occurs.

Coercive isomorphism is driven by two different forces – political influence and the problem of legitimacy (DiMaggio and Powell 1983). Within the country context, isomorphism is an idea of what contemporary societies should be that is addressed by the institutionalization of certain models that are constructed and promulgated through cultural and associational processes, such as the United Nations Sustainable Development Goals (e.g., Griggs et al. 2013; Nilsson, Griggs, and Visbeck 2016; United Nations 2015). Importantly, as specified by the United Nations, the expectation is not that all countries will adhere to all 17 Sustainable Development Goals equally from 2015 and on. Instead, countries have signed on to use the United Nations framework and work toward becoming more sustainable and, ultimately, achieve certain acceptable thresholds by 2030 (United Nations 2015). Within this frame-of-reference, to gain legitimacy for sustainability initiatives, companies imitate other companies that operate within their own country and, overall, these companies are, in essence, coerced to adhere to the sustainability infrastructure in the country (cf. McFarland, Bloodgood, and Payan 2008). Intra-country company-dependence, politically and from the standpoint of legitimacy, are key factors in coercive isomorphic change (Pfeffer and Salancik 1978). Since a company depends on its home country’s sustainability infrastructure to achieve its own sustainability goals, the company will imitate the structure, climate, and behavioral focus of other companies in the country as well as adhere to country attributes, institutions, and national sustainability policy (cf. Clark 1990).

Mimetic isomorphism arises under conditions of environmental uncertainty (DiMaggio and Powell 1983; McFarland, Bloodgood, and Payan 2008). When faced with ambiguous situations, as applied to our theorizing about country-company sustainability initiatives, companies model themselves after other companies in their home country, especially after those they regard as more legitimate or successful (Lieberman and Asaba 2006). These successful companies collectively have been able to work to achieve performance goals within the constraints resulting from country attributes, institutions, and national policy. Additionally, companies are also likely to mimic other companies in their country-environment with whom they have boundary-spanning ties, which favors intra-country connections over inter-country connections (cf. Galaskiewicz and Wasserman 1989). Managers who have intra-country ties to peers in other companies are able to see how such companies deal with country constraints and “learn what is and what is not acceptable” (Galaskiewicz and Wasserman 1989, p. 456). This learning influences the way they behave themselves in implementing sustainability initiatives.

Normative isomorphism results from the diffusion of ideas and normative rules via social networks (Mizruchi and Fein 1999). Normative rules dictate the organizational and professional behaviors that are considered appropriate within the environment (Guler, Guillen, and Macpherson 2002). In particular, professional and trade associations often serve as vehicles for the establishment and dissemination of such normative rules (DiMaggio and Powell 1983). Through these professional networks, members of one company influence those of another (Mizruchi and Fein 1999). For example, Beliveau, Cottrill, and O’Neill (1994) examined factors that predict sustainability and found that a company is more likely to engage in sustainability initiatives if other companies in its industry have done so.

Resource-Based Theory

A number of theoretical perspectives focus on company-level effects to explain variation in performance outcomes across companies (e.g., Hult 2011). Resource-based theory (Kozlenkova, Samaha, and Palmatier 2014; cf. Hunt and Morgan 1996), originating as the resource-based view (Barney 1991; Wernerfelt 1984), is very applicable for the country-company dynamics of sustainability initiatives since such initiatives build on resource use and constraints. Consequently, resource-based theory identifies the company’s idiosyncratic, strategic resources as the primary determinant of competitive advantage and company performance (Barney 1991). The theory portrays resources as those tangible and intangible strategic assets and capabilities possessed by a company that enable it to implement valuable strategies that improve its efficiency and effectiveness (Wernerfelt 1984). As examples, company resources include brand names, patents, corporate culture, trade contacts, knowledge, management skills, and efficient procedures (Barney 1991; Grant 1991; Wernerfelt 1984). In order for resources to be a source of sustained competitive advantage, they must be valuable, rare, imperfectly imitable, and non-substitutable (Barney 1991).

An important aspect of resource-based theory is that it explains long-lived variation in the profitability of companies within the same industry and country (Bamiatzi, Bozos, Cavusgil, and Hult 2016; Makino, Isobe, and Chan 2004; cf. Hunt and Morgan 1996). Due to the accumulation of unique resources, companies evolve differently, thus exhibiting distinct company structures and implementing different marketing strategies (Day 1994). By having unique strategic marketing resources, companies erect barriers to imitation with the objective of generating above-normal returns and protecting their competitive advantage in the marketplace (Wernerfelt 2005). Rumelt (1984, 1987) argues that through isolating mechanisms – such as response lags, producer learning, and reputation – companies prevent imitative competition.

The company’s portfolio of unique strategic resources also influences the strategies that it adopts (Peteraf 1993; Wernerfelt 2005). According to resource-based theory, these resources enable the company to either achieve superior performance through differentiation of its products or to have the ability to manufacture more efficiently to achieve a low-cost position (Conner 1991; Hunt and Morgan 1996; Porter 1980). At a given point in time, customers have divergent preferences, but due to constraints in inputs, the company is unable to satisfy all of the customers’ wants and needs (cf. Peteraf 1993). Instead, the company will seek a balance between the external market demands and the internal resources and capabilities to satisfy a segment of the market (Grant 1991). Different companies possess different resources and hence will target different segments, giving rise to within-country intra-industry heterogeneity and profit level variation (cf. Hunt and Morgan 1996). As such, idiosyncratic company characteristics, such as strategic resources allocated to sustainability initiatives, explain differences in product-market performance across companies (cf. Barney 1991).

Hypotheses

H₁ serves as a benchmark examination of the theoretical tension between institutional theory (DiMaggio and Powell 1983) and resource-based theory, as it applies in marketing (e.g., Kozlenkova, Samaha, and Palmatier 2014). The basic notion for H₁ is to provide guidance on what is possible in implementing sustainability initiatives based on country-company competitiveness dynamics (Buckley and Casson 2011). On one hand, institutional theory suggests that for a company to attain legitimacy regarding its sustainability efforts, the company tends to be isomorphic to the sustainability infrastructure in its home country and to other intra-country companies’ sustainability efforts (Dacin 1997). On the other hand, resource-based theory explains why there can be variation in the product-market performance of companies within the same country and industry (e.g., Wernerfelt 1984). The idea is that, by combining unique strategic resources idiosyncratically, companies erect barriers to imitation of their sustainability initiatives, with the objective of generating above-normal returns (Wernerfelt 2005).

While institutional theory is a robust theory which has been used in a variety of macro and micro settings, the theory has been used particularly effectively in country-level international research to explain macro phenomena (e.g., Udry 2003). Consequently, institutional theory provides certain “boundaries” for our expectation of what can be achieved by companies in the 10 countries studied (Hult 2011, p. 509). Beyond these boundaries, however, we suggest that resource-based theory coupled with the marketing literature on sustainability (e.g., Huang and Rust 2011; Hunt 2011; White and Simpson 2013), corporate social responsibility (e.g., Iyer and Soberman 2016; Luo and Bhattacharya 2009), and green marketing (e.g., Cronin et al. 2011; Trudel, Argo, and Meng 2016) provide a plausible justification for why companies’ sustainability efforts should reap performance advantages above the effects of the country’s sustainability infrastructure.

Perceptibly, Crittenden et al. (2011, p. 71) recognize that “sustainability is a major concern for marketers in the 21st century since marketing strategies and activities are inextricably linked to the future of the natural environment that sustains all life” (cf. Kotler 2011). However, more than just a recognition, which sometimes can be dismissed as a viewpoint or desired state, Luo and Bhattacharya (2009, p. 210) showed that CSR-driven marketing strategies can meaningfully meet Wall Street’s demands, such as product-market performance, and “create moral capital” as well as provide an “insurance-like protection” for companies’ shareholder wealth. This narrative is echoed by Cronin et al. (2011, p. 158) who state that “as companies note the positive gains that can accrue through environmentally friendly marketing strategies and the potential pitfalls associated with non-environmentally friendly strategies, going green is beginning to take center stage in boardrooms around the world.”

Consequently, our initial hypothesis (H₁), which is tested with data from 10 countries, builds on the theoretical tension between institutional theory and resource-based theory. The theoretical weight, however, is allocated to companies’ potential to leverage their idiosyncratically organized strategic resources to implement sustainability initiatives that build on but also surpass the country’s sustainability effects. Thus, as a benchmark hypothesis (for H₂ through H₅ to come), we formally state H₁ as:

H₁: A company’s sustainability activities have a positive effect on the company’s product-market performance above the sustainability efforts provided by the company’s home country.

Some companies are “doing good” (e.g., Kotler and Lee 2005) and some customers acquire a “warm glow” (Habel, Schons, Alavi, and Wieseke 2016) regardless of costs involved, product prices, or benefits received in the exchange. These companies and customers have adopted a “self-accountability” (Peloza, White, and Shang 2013, p. 104) that creates a unique value proposition for them. But, most companies and most customers, we argue, are not value-seeking without some logical cost-and-benefit reasoning. Value is the function of benefits “received and what is given” (Zeithaml 1988, p. 14). This rationale applies to both companies and customers; both receive benefits and both give up something in the exchange.

At the customer level, “value is a customer’s perceived preference for and evaluation of those product attributes, attribute performances, and consequences arising from use that facilitate (or block) achieving the customer’s goals and purposes in use situations” (Woodruff 1997, p. 142). Customer value is tightly connected to “price fairness” (Habel, Schons, Alavi, and Wieseke 2016), which is defined as “a consumer’s subjective sense of a price as right, just, or legitimate versus wrong, unjust, or illegitimate” (Campbell 2007, p. 261). Customer value is also connected to the notion of a “warm glow” (i.e., positive feelings people experience when they support products by a company implementing sustainability initiatives) in that increased warm glow also increases the perceived value received (Giebelhausen, Chun, Cronin, and Hult 2016).

At the company level, there are a number of value propositions and definitions, but the marketing focus on customer lifetime value (CLV) applies fittingly to sustainability initiatives since such initiatives are targeting, at least partially, the fundamental idea of gaining customers’ repeat business because of “doing good,” creating a “warm glow,” and/or offering sustainable products at fair prices (cf. Holm, Kumar, and Rohde 2012). In this case, “the economic value of repeat business is determined by the discounted net present value of future cash flows from current customers” (Fornell, Morgeson, and Hult 2016, p. 94). CLV leads to an increase in stock prices (Kumar and Shah 2009) and to increased company profits (Venkatesan and Kumar 2004).

With the assumption that customers are likely to purchase more quality products from a company implementing sustainability initiatives, if those initiatives do not affect costs or prices, a greater effect of sustainability on product-market performance than expected in the baseline hypothesis (H₁) should also follow. The reasoning is rooted in sentiments stemming from companies “doing good” (Luo and Bhattacharya 2009), customers experiencing a “warm glow” by the value received or by adjusting their “warm glow thermostat” (Giebelhausen, Chun, Cronin, and Hult 2016, p. 56), and/or by “perceived price fairness” (Habel, Schons, Alavi, and Wieseke 2016, p. 84). Formally stated as H₂:

As we stated in the theoretical development of H₂, some companies are “doing good” (e.g., Kotler and Lee 2005) and some customers acquire a “warm glow” (Habel, Schons, Alavi, and Wieseke 2016). In fact, some companies and customers cultivate the “doing good” and “warm glow” phenomena even if costs and prices are very high or benefits are minimal or non-existent. From the standpoint of marketing strategy, consumer behavior, or ultimately product-market performance, theoretically there is no consistent reasoning that can be applied to prescriptively explain such phenomena for a complete customer base of a company or even a specific profitable customer segment (cf. Luchs, Naylor, Irwin, and Raghunathan 2010). Likewise, normatively it is also difficult to describe why this behavior ought to happen or ought to be rational. The logic of “self-accountability” (Peloza, White, and Shang 2013, p. 104) for the good of humanity is as close as we will get to encapsulate the unique value proposition for these companies and customers. However, building on the “value” rationale in H₂, we argue that very high company costs, very high customer costs, and/or very high product prices are not viable. Therefore, H₃ is a complementary hypothesis to H₂ and states that:

H₃: A company’s sustainability activities have no effect on the company’s product-market performance when (a) company costs are high, (b) customer costs are high, and (c) product prices are expensive.

The logic, but also harshness, of H₃ suggests that there is a theoretical and practical middle ground that should be explored for products delivered by companies advocating and implementing sustainability initiatives. The literatures on “doing good” (Kotler and Lee 2005; Luo and Bhattacharya 2009), “warm glow” (Giebelhausen, Chun, Cronin, and Hult 2016), and “price fairness” (Habel, Schons, Alavi, and Wieseke 2016) theoretically demand such sensitivity analyses. That is, while very high company costs, very high customer costs, and/or very unfair prices seemingly – for most people – will be off putting and result in no purchase, some modest increases in costs and/or prices while maintaining a positive effect on performance seem plausible. In support, marketing research provides an indication that would lead us to believe that costs and prices being modestly higher would still entice customers to buy as a part of their “mindful consumption” (Sheth, Sethia, and Srinivas 2011, p. 21), socially responsible buying habits (Iyer and Soberman 2016), sustainable behaviors (Huang and Rust 2011), and “greening” deeds (cf. Leonidou, Katsikeas, and Morgan 2013, p. 151).

In parallel to H₂ being a sensitivity analysis of H₁, H₄ is structured as a sensitivity analysis of H₃. The rationale is that a company ought to, in a normative mode, be able to build on the positive embodiments of “doing good” (Luo and Bhattacharya 2009) and the customers’ sentiments toward wanting to experience a “warm glow” (even if it means adjusting their “warm glow thermostat,” Giebelhausen, Chun, Cronin, and Hult 2016) to implement sustainability activities that result in some increased company costs, increased customer costs, or increased product prices and still have a positive effect on the company’s product-market performance. Though indications exist that a 30 percent increase in product prices have been received favorably by customers (Enax, Krapp, Piehl, and Weber 2015), theoretically there is limited rationale to specify a specific breaking point for acceptable increases of company costs, customer costs, and product prices but the effects should be above market-competitive costs and prices. Thus, without specifying a cut-off point, theory (and practice) submit that:

H₄: A company can implement sustainability activities that result in some (a) increased company costs, (b) increased customer costs, and (c) increased product prices and still have a positive effect on the company’s product-market performance.

The rationale for H₅ comes full circle from where we started the hypothesis development in H₁. That is, the tension between institutional theory (Dacin 1997; DiMaggio and Powell 1983) and resource-based theory (Barney 1991; Kozlenkova, Samaha, and Palmatier 2014; Wernerfelt 1984) mandate a deeper theoretical and practical examination of the give-and-take between the two theories, as they apply to country-company dynamics, competitiveness, and sustainability (cf. Buckley and Casson 2011; Buckley, Pass, and Prescott 1998).

Again, institutional theory holds that to attain legitimacy, a company tends to be isomorphic to actors in its home country and develop similar sustainability efforts as other companies in the country due to the country’s sustainability-infrastructure support (Dacin 1997; DiMaggio and Powell 1983). For example, the globally adopted United Nations Sustainable Development Goals are a form of coercive isomorphism to adhere to certain goals; modeling a company after other companies in the country or country initiatives regarding sustainability is a form of mimetic isomorphism; and following normative rules unique to a country is a form of normative isomorphism. The coercive, mimetic, and normative pressures of institutional theory suggest that companies cannot waver too far up or down on sustainability initiatives.

At the same time, resource-based theory suggests that even in the restrictive scenario presented in the country-company dynamics of sustainability initiatives, companies can leverage unique, idiosyncratic strategic resources above that of the country in which they reside (cf. Barney 1991; Wernerfelt 1984) to achieve superior product-market performance. At minimum, the amalgamation of strategic resources a company uses to address sustainability initiatives has a strong possibility to be valuable, rare, imperfectly imitable, and non-substitutable (Barney 1991) due to the mixture of country, industry, competitors, and customers involved.

Integrating the logic from institutional theory and resource-based theory, we apply the rationale that there is “tension” between the two theories, and that institutional theory presents boundary conditions for the unique leverage that can be realized by resource-based theory. We do, however, expect companies to be able to develop and implement sustainability initiatives that are to some extent above that of their home country and nonetheless at least maintain a positive product-market performance effect. To theoretically guide us in this endeavor, we build on configuration theory and the notion of “ideal profiles” of companies (e.g., Doty, Glick, and Huber 1993). Profile deviation studies normally select the highest performing 10% of companies to calibrate ideal profiles (e.g., Van de Ven and Drazin 1985, Venkatraman and Prescott 1990). In parallel, as a theoretical starting point and practical guide for the analyses, we expect that companies can implement sustainability initiatives that are 10 percent greater than the sustainability level achieved in a country’s infrastructure (cf. Van de Ven and Drazin 1985). As such, integrating institutional theory and resource-based theory and using guidance from configuration theory, H₅ states:

H₅: A company can implement sustainability activities at 10 percent above that of the sustainability infrastructure provided by the company’s home country and have a positive effect on the company’s product-market performance.

Data Collection

We tested the five hypotheses using country, company, and customer-focused data gathered in 10 countries, including the nine Spanish-speaking sovereign countries in South America (Argentina, Bolivia, Chile, Colombia, Ecuador, Paraguay, Peru, Uruguay, and Venezuela) and the United States. South America occupies the entire southern part of the supercontinent of the Americas, and there are 12 independent countries and three territories on this continent. For data equivalence purposes (Hult et al. 2008), we excluded the three independent South American countries where the primary language is not Spanish (Brazil – Portuguese, Guyana – English, and Suriname – Dutch) as well as the three South American territories which are not considered sovereign states (Falkland Islands, French Guiana, and South Georgia and the South Sandwich Islands). The United States was included as a benchmark and independent sample due to its efforts on international competitiveness.

Professional marketing research companies in the 10 countries were used to identify the sampling frame, correlate the sampling frame across countries, and collect the data online using Qualtrics. For the survey instrument, we used a “parallel translation approach” and multiple translators to ensure that the questionnaire had semantic equivalence across languages (English and Spanish), conceptual equivalence across the countries, and normative equivalence to the source survey (which was created in English). Specifically, we used two translators to create the Spanish version of the questionnaire from English. Then we had two Spanish-English fluent translators attend to semantic, conceptual, and normative equivalences. Any differences of opinion between the translators was solved in conference. As such, we took precautions to ensure that the words and sentence structure in the translated items expressed the same meaning as in English (semantic equivalence), that the constructs used were the same across groups (conceptual equivalence), and, if needed, there was an ability of the translated items to address social norms that differ across the countries (normative equivalence).

In collecting the data, our focus was on (1) statistical power (Cohen 1988) and (2) data collection equivalence, the latter referred to as “whether the sources of data, the methods of eliciting data and the resulting samples are comparable across cultures, and can be viewed in relation to three elements: sampling frame comparability, data collection procedure, and sample comparability” (Hult et al. 2008, p. 1037). To eliminate the alternative explanation that differences in the sampling frame, data collection methods, and final samples account for differences across the 10 countries, we took steps to safeguard that (1) the sampling frame was the same across countries (senior manager working full-time for a company headquartered in the country), (2) the data collection method was the same for each country (online via Qualtrics), and (3) a random selection of companies was used from a country-specific representative sample to alleviate any uncontrolled, systematic errors biasing the estimators.

The survey instrument, statistical power, and data collection criteria led us to a sample of n = 4,051 for the 10 countries. The country samples are: Argentina, n = 437; Bolivia, n = 355; Chile, n = 286; Colombia, n = 480; Ecuador, n = 548; Paraguay, n = 247; Peru, n = 297; Uruguay, n = 484; Venezuela, n = 388; and the USA, n = 529. On average, the companies in the database are 44 years old (range across countries: 28 to 69 years), employ 2,377 people (range across countries: 938 to 3,680 people), and are dispersed across all ten economic sectors of the globally used Harmonized Commodity Description and Coding System (agriculture, forestry, and fishing, n = 110; mining, n = 60; construction, n = 249; manufacturing, n = 344; transportation, communications, electric, gas, and sanitary services, n = 352; wholesale trade, n = 270; retail trade, n = 394; finance, insurance, and real estate, n = 259; services, n = 1,433; and public administration, n = 580). The company age range (28 to 45 years) and company size range (769 to 1,486 people) among the South American countries are relatively consistent while the US companies are, on average, older (69 years) and larger (3,680 people employed). Descriptive statistics (means, standard deviations) and bivariate correlations are included in Table 1 for the overall sample (n = 4,051).

OPEN IN VIEWER

Table 1.

Descriptive Statistics and Correlations for the Overall Sample.

	Mean	Standard Deviation	1	2	3	4	5	6	7	8	9
1. Country Sustainability	6.15	2.67	1.00
2. Company Sustainability	6.25	2.56	.60	1.00
3. Company Costs	5.83	2.23	.24	.24	1.00
4. Customer Costs	5.96	2.23	.19	.23	.68	1.00
5. Product Price	6.73	2.31	.31	.47	.29	.28	1.00
6. Product Competitiveness	6.91	2.29	.30	.46	.29	.26	.82	1.00
7. Company Age	43.89	45.38	.08	.02	-.05	.00	-.06	-.05	1.00
8. Company Size	2377.46	3410.04	.03	-.05	-.06	-.03	.01	.01	.54	1.00
9. Industry	---	---	.02	.01	-.02	-.01	-.03	-.03	.17	.08	1.00

Correlations > .07 are significant at p < .05.

Dependent Variable

To align with the study’s focus on the effects of a company’s sustainability activities – rooted in the United Nations’ Sustainable Development Goals (United Nations 2015) – on product-market performance (e.g., Morgan 2012; O’Sullivan and Abela 2007), the dependent variable was anchored by growth of the market competitiveness of the company’s portfolio of products (e.g., Buckley, Pass, and Prescott 1998). Performance in t₀ was the temporal reference point to t₁ (compared with last year, our company’s products are much more competitive in the marketplace; en comparación con el año anterior, los productos de nuestra empresa son mucho más competitivos en el mercado). To assess the validity of the dependent variable (DV) prior to the hypothesis testing, we tested the bivariate correlations between product-market performance (“product competitiveness”) and return-on-investment (.27, p<.01), sales growth (.25, p<.01), Tobin’s Q (.56, p<.01), and Altman’s Z (.38, p<.01). In each case, the DV of product competitiveness is significantly related to traditional firm-level accounting and financial-market performance variables, as would be expected if the variable has criterion and concurrent validity (Standards for Educational and Psychological Testing 2014) and is temporally embedded in the marketing-performance outcome chain (Katsikeas, Morgan, Leonidou, and Hult 2016).

Independent Variables

Company costs, customer costs, and product price

The two measurement scales for Company Costs and Customer Costs were adapted from work by Kirmani (1990). Each scale has three reflective items anchored by “a very high cost” (un costo muy alto) = 0 to “no cost at all” (sin costo) = 10. For the scale measuring Company Costs, the items are: (1) our company bears considerable costs for its sustainable development activities (COcost1; nuestra empresa tiene costos considerables por sus actividades de desarrollo sostenible); (2) the money our company invests on sustainable development activities is very large (COcost2; el dinero que nuestra empresa invierte en actividades de desarrollo sostenible es muy alto); and (3) our company incurs very high costs for its sustainable development engagement (COcost3; nuestra empresa incurre en altos costos por su compromiso al desarrollo sostenible). For Customer Costs, the parallel items are: (1) our customers bear considerable costs for our company’s sustainable development activities (CUcost1; nuestros clientes/consumidores tienen costos considerables por las actividades de desarrollo sostenible de nuestra empresa); (2) the money our customers invest in our company’s sustainable development activities is very large (CUcost2; el dinero que nuestros clientes/consumidores invierten en actividades de desarrollo sostenible de nuestra empresa es muy alto); and (3) our customers incur very high costs for our company’s sustainable development engagement (CUcost3; nuestros clientes/consumidores incurren en altos costos por el compromiso al desarrollo sostenible de nuestra empresa). The price competitiveness item was based on Buckley, Pass, and Prescott (1998), with a scaling from “strongly disagree (totalmente en desacuerdo)” = 0 to “strongly agree (totalmente de acuerdo)” = 10: Compared with last year, our company’s prices are much more competitive in the marketplace (en comparación con el año anterior, los precios de nuestra empresa son mucho más competitivos en el mercado).

Table 2 summarizes the results of the measurement analyses for the reflective scales of Company Costs and Customer Costs by country, region, and overall samples. The psychometric properties were evaluated via confirmatory factor analyses (CFA) using LISREL 9.20 (Jöreskog et al. 2000). The model fits were evaluated using the Comparative Fit Index (CFI) and the Root Mean Square Error of Approximation (RMSEA). Composite reliability was calculated using the procedures outlined by Fornell and Larcker (1981) based on Werts, Lin, and Jöreskog (1974). In each of the 10 countries, region of South America, and overall sample, the parameter estimates (factor loadings) ranged from .67 to .99, the composite reliabilities ranged from .86 to .96, and the average variances extracted ranged from .67 to .89. The CFI ranged from .88 to .99 and the RMSEA ranged from .02 to .09. Each measurement statistic indicates valid and reliable scales.

OPEN IN VIEWER

Table 2.

Measurement Analysis of Reflective Scales.

	Argentina	Bolivia	Chile	Colombia	Ecuador	Paraguay	Peru	Uruguay	Venezuela	South America	USA	Overall
COcost1	.72	.77	.78	.78	.77	.69	.82	.75	.84	.77	.79	.78
COcost2	.95	.95	.98	.94	.83	.86	.88	.97	.96	.93	.89	.92
COcost3	.93	.95	.86	.81	.86	.96	.97	.95	.92	.91	.90	.91
CR	.91	.92	.91	.88	.86	.88	.92	.92	.93	.91	.90	.90
AVE	.76	.80	.77	.72	.67	.71	.80	.80	.82	.76	.74	.76
CUcost1	.94	.92	.89	.88	.90	.86	.89	.89	.85	.89	.83	.88
CUcost2	.96	.92	.99	.92	.92	.92	.75	.94	.93	.94	.90	.92
CUcost3	.93	.93	.94	.87	.96	.91	.81	.88	.92	.91	.81	.88
CR	.96	.95	.96	.92	.95	.92	.86	.93	.93	.94	.88	.92
AVE	.89	.85	.89	.79	.86	.80	.67	.82	.81	.83	.72	.80
CFI	.99	.98	.98	.98	.97	.99	.88	.96	.97	.99	.99	.99
RMSR	.04	.03	.05	.05	.03	.02	.09	.04	.06	.03	.03	.03
n	437	355	286	480	548	247	297	484	388	3,522	529	4,051

Note: COcost = parameter estimates (loadings) for Company Costs, CUcost = parameter estimates (loadings) for Customer Costs, CR = composite reliability, AVE = average variance extracted, CFI = comparative fit index, and RMSR = root mean square residual.

However, since our research incorporates data collected in 10 countries, a need exists to establish measurement equivalence across countries prior to the hypothesis testing (Steenkamp and Baumgartner 1998). The assumption is that factor loadings, factor correlations, and error variances are invariant across countries. To test these premises, four different CFA models were analyzed and a chi-square (χ²) difference test was performed on the nested models. First, we ran a model where the factor loadings, factor correlations, and error variances are all invariant across the 10 countries (χ² = 1,668.75, df = 197). Second, we analyzed a model where the factor loadings are allowed to differ across each of the 10 countries (χ² = 1,632.17, df = 143). The Δχ² = 36.58 and Δdf = 54, which is not significant. Adding two more models (freely estimated error variances and factor loadings; freely estimated error variances only), as expected, resulted in no improvements. Thus, the measurement scales are robust across countries.

Discriminant validity between Company Costs and Customer Costs was assessed using the procedures recommended by Anderson (1987), Bagozzi and Phillips (1982), and Fornell and Larcker (1981). First, this entails analyzing the pair of constructs in a series of two-factor CFA models for the country, region, and overall samples. Each model was run twice – once constraining the phi coefficient (φ) to unity and once freeing this parameter. A chi-square (χ²) difference test was performed on the nested models to assess if the χ² values were significantly lower for the unconstrained models. The critical value (▵χ² ₍₁₎>3.84) was exceeded in all cases. Second, we examined the confidence interval (+/− two standard errors) built around the correlation between pairs of constructs to see whether it includes 1.0 (Anderson 1987). None of the intervals includes 1.0, indicating discriminant validity between constructs. Third, we examined the difference between the Average Variances Extracted (AVE) for each construct and compared it with the Shared Variances (SV) across constructs. In all cases, the AVE was higher than the corresponding SVs. The SVs were also .46 (lower than the preferred .50 value) while the AVEs were .76 to .89 for Company Costs and Customer Costs, respectively (higher than the preferred 50 percent). This indicates that the measures possessed robust discriminant validity.

In addition, we used the CFA structure to assess common method variance. We employed a CFA approach to Harman’s one-factor test (McFarlin and Sweeney 1992; Sanchez and Brock 1996). The rationale for this test is that if common method bias poses a serious threat to the analysis and interpretation of the data, a single latent factor would account for all manifest variables (Jarvis, MacKenzie, and Podsakoff 2003). A worse fit for the one-factor model would suggest that CMV does not pose a serious threat. In all cases, the fits are worse for the unidimensional models, suggesting that CMV is not a threat to the analyses.

Country-Company Sustainability

This study’s results were predicated on levels and changes in country and company sustainability (cf. Fornell, Morgeson, and Hult 2016). In the theory testing of the five research hypotheses we did not address levels, per se, even though they were incorporated into the analyses. The logic for focusing on changes in levels of sustainability is theoretically clear; it is of little consequence to companies in Argentina what the sustainability level is in Bolivia, for example, except as a competitive benchmarking comparison. The reason is that the tension between institutional theory and resource-based theory would only allow for a certain degree of company variance in sustainability within-country regardless of between country comparisons (as indicated previously, the range is -30.0% to +24.0%, see Table 8). However, levels and changes in sustainability need to be incorporated to understand what companies are able to do relative to countries’ sustainability efforts. As such, for a better understanding of the sustainability benchmarking-levels used in studying changes in those levels, Figure 2 reports on the sustainability scores of the countries and companies within countries. Venezuela, or officially the Bolivarian Republic of Venezuela, has the lowest level of country sustainability (x¯ = 4.68), as perhaps expected in the relatively tightly centralized and controlled federal republic, while Ecuador, officially the Republic of Ecuador, has the highest (x¯ = 7.26).

OPEN IN VIEWER

Figure 2.

Sustainability Levels of Countries and Companies.

The bar charts in Figure 2 also indicate that in four of the countries (Argentina, Chile, Uruguay, and the USA), the countries themselves have achieved higher levels of sustainability infrastructure than the efforts demonstrated by the countries’ companies. This presents a great marketing opportunity for companies since in each of those four countries, a company can gain marketplace advantages by implementing sustainability efforts above that of the country (see Table 8). For example, in Argentina, where companies can gain positive product-market performance by implementing sustainability initiatives at 124 percent of the country’s efforts, companies right now underperform the country efforts by 3.17 percent (Δ 27.17%). Similarly, companies in Chile (Δ 15.44%), Uruguay (Δ 12.56%), and the USA (Δ 17.51%) can gain positive effects on product-market performance by increasing their sustainability efforts.

Companies in Colombia and Ecuador should, at this time, rely on their countries’ sustainability infrastructure (see Table 8). Interestingly, Colombian and Ecuadorian companies in actuality implement initiatives that are above their respective countries. This means that companies in Colombia (Δ -22.24%) and companies in Ecuador (Δ -32.55%) are spending resources on sustainability initiatives that do not directly have as much effect on product-market performance, a noble albeit performance-detrimental action. That said, despite costs, these Colombian and Ecuadorian companies are admirably building a movement toward sustainability in their countries to reach both higher country and company levels of sustainability over time. The remaining four countries have mixed results on country-company alignment (Bolivia, 2.96%; Paraguay, 12.20%; Peru, 4.80%; and Venezuela, 22.90%). Companies in Bolivia (<14.0%) and Peru (<10%) are well within the sustainability efforts supported by the sensitivity analyses (Table 8), while companies in Paraguay (>9.0%) and Venezuela (>8.0%) are not.

Customer Costs, Company Costs, and Price Sensitivities

The analyses involving customer costs, company costs, and product prices were foundationally rooted in institutional theory (DiMaggio and Powell 1983) and resource-based theory (Barney 1991; Wernerfelt 1984) like the country-company analyses. These cost/price factors have been associated with “warm glow” (customer costs), “doing good” (company costs), and price fairness (product prices) in the literature. With respect to the costs/prices, institutional theory guides what cost and price levels are appropriate in a country while resource-based theory guides companies on how to use resources effectively to develop the value proposition for customers. However, the substantive literatures of “warm glow” (Giebelhausen, Chun, Cronin, and Hult 2016), “doing good” (Lou and Bhattacharya 2009), “price fairness” (Habel, Schons, Alavi, and Wieseke 2016), and “customer value” (e.g., Woodruff 1997; Zeithaml 1988) drive much of the theorizing and the implications that can be drawn for customer costs, company costs, and product prices, respectively.

Table 4 through Table 7 have numerous results relating to containing and increasing individual and multiple costs/prices. As expected, when costs and prices are contained (i.e., neither the company nor the customer has increased costs associated with purchasing a product and the prices of that product are market-competitive), company sustainability has a positive and significant effect on product-market performance (see Table 4 and Table 6). The implication is rather straightforward, company stakeholders (Hillebrand, Driessen, and Koll 2015) like to buy products from companies practicing and incorporating sustainability initiatives in their operations if the costs and prices are market competitive. Containing company costs and/or containing customer costs, when the sensitivity analyses were involved, resulted in product-market performance improvements vis-à-vis the benchmark model in H₁. Logically, cost effectiveness is preferred by companies and customers. However, the effects of all of the models involving containing product prices (as a form of price competitiveness or “price fairness”) were reduced. This indicates that companies prefer a “doing good” frame-of-reference and customers prefer a “warm glow” feeling over price competitiveness. In the opposite, when costs and prices are perceived to be increased significantly, the effects of company sustainability on product-market performance are inconsistent and in the vast majority of scenarios not significant in Table 5 and Table 7 (insignificant beta coefficients, ΔR², and/or equations). Too high of a cost and too high of a price result in less interest and an inconsistent sustainability-performance relationship. Again, this is not controversial, per se, although some heavily oriented “warm glow” customers may prefer that sustainability products were always preferred by everyone (cf. Fenton 2014).

Finally, where marketing strategy-making is important relative to our study’s scope is in the large “band” of options between the two extremes of containing costs/prices (company costs ≥ 6; customer costs ≥ 6; and price competitiveness ≥ 6) and drastically increasing these costs/prices (company costs < 2; customer costs < 2; and price competitiveness < 2). This middle-band-range is basically represented by the results that are in between what is reported in Table 4 and Table 5 on one hand and Table 6 and Table 7 on the other (with the exception of the last row of Table 5). The basic notion of cost flexibilities and price sensitivities build on “warm glow” feelings (customers), “doing good” (companies), and “price fairness” (customers’ perception of product prices). For example, as a comparison, fifty-five percent of customers across 60 countries were found in a study by The Nielsen Company to be willing to pay more for products from companies that are committed to sustainability initiatives (Fenton 2014). In our case, we found that customer and company costs as well as product prices can increase by 27 to 72 percent without being detrimental to a company’s product-market performance. Of the cost/price alternatives in Table 5, product prices are the most inflexible in the U.S., where only a 27 percent price premium can be added without loss of the effects of sustainability initiatives on a company’s product-market performance. The company costs in South America has the greatest flexibility, with a 69 percent above-normal cost possible to be added without significant detrimental effects on a company’s product-market performance.