Long-term orientation and corporate social responsibility in small and medium-sized enterprises: the role of strategy formation mode

Purpose – This study aims to explore the effects of long-term orientation (LTO) and strategy formation mode on corporate social responsibility. While many researchers have investigated how large businesses address corporate social responsibility (CSR), there is little empirical evidence on how small-andmedium-sizedbusinessesimplement CSR orwhatindividualdriversshapethis process. Design/methodology/approach – The paper surveyed 282 small and medium-sized managers from the United Kingdom.Therespondents wererecruited usingplatform ProlificAcademic. Findings – The findings reveal that LTO is a prerequisite for developing CSR and shapes strategy formation mode. The findings also suggested that deliberate strategies are positively related to CSR. The results are consistent across different components of LTO (futurity, continuity and perseverance) and CSR types (internal and external). Originality/value – The results show that all aspects of LTO are relevant for CSR in SMEs. Besides LTO, deliberatestrategy formation model isanimportantfactorcontributing toCSR.Thepaperpresents as first an empirical contribution to the strategy literature by examining positive relationship between LTO and deliberatestrategy formationmode.


Introduction
Small and medium-sized enterprises (SMEs) represent the backbone of most economies (Santos, 2011).In today's dynamic and complex business environment, where society demands solutions to complex phenomena such as climate change, unemployment, migration and demographic changes (Dartey-Baah and Amoako, 2021), SMEs face the imperatives of sustainability and social responsibility, which have emerged as significant trends in business (Oduro et al., 2021).Thus, many companies have to change their systems and processes to reduce environmental impact, produce environmentally friendly products and find ways of improving the well-being of communities within which they operate (Aguilera et al., 2007).Corporate social responsibility (CSR) is an established part of management research whose origins can be traced back to the beginning of this field.The initial definition by Bowen (1953) defines CSR as the obligations of businessmen to implement such guidelines, which are desirable in terms of achieving firm goals and society's values.In the last century, many firms did not adopt CSR policies because middle management saw them as expensive and difficult to measure (Lee, 2008).However, at the turn of the 20th and 21st centuries, there was a radical change in the perception of social direct as well as indirect effects on CSR via the strategy formation mode.An empirical verification of these effects would contribute to a better understanding of time horizon as a mechanism shaping CSR activities in SMEs (Choi et al., 2023).
To address these research gaps, we formulate and test our hypothesis that strategy formation mode partially mediates the relationship between LTO and the adoption of CSR.In a survey among 282 managers of SMEs from the United Kingdom, we found support for this hypothesis, presenting several contributions to the management literature.First, we complement and extend recent research on the link between future orientation and CSR (Choi et al., 2023) by considering undisclosed CSR activities as well as the multifaceted natures of both LTO and CSR.Second, by examining the interplay between LTO and strategy formation mode, we provide insights on the mechanism through which these drivers collectively influence CSR, addressing recent calls published in systematic literature reviews focused on CSR in SMEs (Dartey-Baah and Amoako, 2021).Third, we contribute to the ongoing debate on CSR adoption by SMEs (Stoian and Gilman, 2017) by examining whether they display similarities to larger corporations concerning the interaction between strategy and CSR.For example, while existing strategic literature highlights that wellstructured plans in larger corporations can effectively tackle social and environmental concerns (Luederitz et al., 2021), the understanding of a similar scenario in SMEs remains limited.Overall, our findings shed light on the formation of CSR strategies within SMEs, contributing to an area of research that still remains open (Luederitz et al., 2021).

Long-term orientation and corporate social responsibility
Existing for over six decades and spanning various disciplines, the concept of CSR has been subject to an abundant number of definitions (Barauskaite and Streimikiene, 2021).Song and Dong (2022) posit that the most widely embraced definitions include Carroll's (1991) four-part definition of CSR, stating that CSR "encompasses the economic, legal, ethical, and discretionary expectations that society has of organizations at a given point in time" (Carroll, 1991, p. 283).A key feature of Caroll's understanding of CSR is that these responsibilities transcend the present, extending as enduring commitments that reach into future generations of stakeholders (Carroll, 2016, p. 6).Thus, CSR inherently encompasses a forward-looking perspective, with its advantages likely to manifest when a company embraces a long-term approach to profit maximization (B enabou and Tirole, 2010).This suggests a plausible connection between LTO and CSR.In the following, we explain the reasons for this connection, building upon the notion that LTO is a multifaceted construct.In particular, we build on Lumpkin and Brigham's (2011) work that has identified three dimensions of LTO: futurity, continuity and perseverance.Futurity is a company's belief that planning for long-term goals and assessing their success or failure are important; continuity connects the past, present and future, attributing value to preservation and constancy as well as longevity; and perseverance refers to the willingness to work hard and display patience to create future value (Brigham et al., 2014).
We first consider the role of futurity, which reflects "how far ahead the firm looks into the future in planning its strategies and operations" (Miller and Friesen, 1978, p. 923).Firms with a high future orientation believe that "forecasting, planning, and evaluating the longrange consequences of current actions have utility" (Lumpkin andBrigham, 2011, p. 1152).Consequently, these firms prioritize the adoption of long-term goals (Miska et al., 2018).The effects of CSR are not immediately felt; instead, they manifest over time (Graafland and Noorderhaven, 2020).From a psychological viewpoint, entrepreneurs with a strong sense of futurity feel comfortable imagining what will happen in the future.As a result, they perceive the benefits of CSR as "psychologically possible and real" (Das and Teng, 1997, p. 78) to a greater extent than short-term-oriented entrepreneurs.Since future-oriented entrepreneurs derive utility from the long-term consequences (Lumpkin and Brigham, 2011) that CSR j SOCIAL RESPONSIBILITY JOURNAL j brings about, they are more inclined to pursue socially responsible behaviors.Evidence suggests that companies in cultures characterized by a greater emphasis on future orientation display a higher degree of economic, social and environmental sustainability practices (Miska et al., 2018).This provides empirical support for the idea that a firm's futurity is positively related to its socially responsible actions.
Continuity is another aspect of LTO that is likely to be related to a firm's CSR.Companies prioritizing continuity emphasize the preservation of long-standing traditions (Lumpkin and Brigham, 2011).Surveys conducted among Swiss SMEs (Looser and Wehrmeyer, 2015) and Italian SMEs (Coppa and Sriramesh, 2013) underscore the importance of traditions in shaping responsible corporate values and promoting CSR.Historical legacies and traditions provide a foundation for the values and expectations of the involved actors (Antal and Sobczak, 2007) and influence the decisions about a company's future.Routines and habits stemming from a company's long-lasting traditions shape its sense of accountability and ethical conduct (Rendtorff, 2011).Aligning with traditions can foster a sense of purpose and responsibility, encouraging the company to make positive contributions to society, uphold ethical standards and adopt environmentally friendly practices.Furthermore, longstanding traditions create a sense of trust and credibility among stakeholders (Conz et al., 2023), enhancing the company's reputation and encouraging support for its CSR activities.However, continuity can be threatened by social and reputational sanctions, making firms that prioritize continuity even more responsive to the needs of external stakeholders (G omez-Mejı ´a et al., 2011).As such, a link emerges between continuity and a firm's socially responsible activities.
Finally, CSR's connection to LTO can be examined through its aspect of perseverance, i.e. the conscientiousness that is required to persist over time (Lumpkin and Brigham, 2011).Long-term-oriented companies recognize the value and potential benefits of CSR, which is reflected in their strategic actions (Wang and Bansal, 2012).Such firms are able to develop strategic resources even if they imply higher costs.This is manifested, for example, in the patient innovation strategies of long-term-oriented SMEs (Wang and Bansal, 2012).Perseverant businesses are likely to use patient capital, i.e. the capital invested without the threat of liquidation for long periods (Dobrzynski, 1993).This patient capital, in turn, facilitates the execution of long-term strategies (Sirmon and Hitt, 2003).Given that CSR is a long-term strategy initiative (B enabou and Tirole, 2010), it can be expected that perseverance, as manifested by the patience needed for cultivating future value, will be positively related to CSR.
The above discussion leads us to the expectation that CSR is a long-term initiative (B enabou and Tirole, 2010) and that long-term-oriented companies will adopt CSR to a greater extent, thus aligning their goals and strategic actions with their temporal preferences.In other words, it can be expected that: H1.There is a positive relationship between LTO and CSR.

Long-term orientation and strategy formation mode
Strategy represents the "pattern in a stream of significant decisions" (Mintzberg, 1972) that helps set the direction of the business.Mintzberg and Waters (1985) distinguish two main modes of strategy formationdeliberate (planned) strategy and emergent strategy.Deliberate strategies require the organization's full control over the strategy process, a predictable environment and a detailed and precise specification of the desired state.On the other hand, emergent strategies are realized despite or without intentions (Mintzberg and Waters, 1985) and fit better into dynamic and competitive environments.The consensus within the present literature is that a firm's strategy formation resides on a continuum from planned to emergent strategies (Neugebauer et al., 2016).j SOCIAL RESPONSIBILITY JOURNAL j Strategy formation requires the statement and clarification of long-term goals (Schendel and Hofer, 1979).Hence, strategy formation necessarily considers the factor of time (Covin et al., 2006).The extent to which the factor of time is considered, or alternatively, the temporal depth of a firm's strategy formation, is influenced by the temporal orientation of the business (Kallmuenzer et al., 2018;Zheng et al., 2020).Firms with a long-term perspective tend to make decisions that will prepare them for the future (Harris and Ogbonna, 2006).Moreover, SMEs that focus on long-term survival instead of financial performance will maintain tight control over the strategy processes (Kreiser et al., 2006), which presents a suitable condition for the adoption of deliberate strategies.
Empirical evidence suggests that long-term thinking and time horizons are positively associated with strategic thinking and a structured approach to strategic planning (Miller, 1983;Stonehouse and Pemberton, 2002).Similar observations can be made at the level of individuals.The personality and time orientation of executives significantly influence the strategy formation of the business (Harris and Ogbonna, 2006).A company in which the board of directors is long-term oriented will more likely implement a deliberate strategy (Kopmann et al., 2017).Thus, the company's futurity can be expected to result in a deliberate, rather than emergent, approach to strategy formation.
A deliberate strategy can enhance the long-term sustainability of businesses (Herrera-Echeverry et al., 2020).Initially associated with larger corporations, strategic planning has also proven advantageous for the growth of small businesses (Miller and Cardinal, 1994).Within this context, a deliberate strategy assumes a pivotal role in ensuring the smoothness, efficiency, stability and regularity of SME functioning, akin to the traits exhibited by "planners" as outlined by Miller (1983).Planned strategies in SMEs facilitate harmonious alignment with the business environment, foster stakeholder trust, and facilitate the integration of organizational goals (Mazzola et al., 2008).For SMEs undergoing a generational leadership transition, strategic planning emerges as a crucial tool in facilitating the development and succession of the succeeding generation (Mazzola et al., 2008).Thus, it can be assumed that SMEs prioritizing continuity are more inclined to embrace a deliberate strategy.
Based on the above arguments, we expect that long-term-oriented businesses will prefer adopting a deliberate, rather than emergent, strategy.In other words, our expectation is that: H2.There is a positive relationship between LTO and deliberate vs emergent strategy.

Strategy formation mode and corporate social responsibility
Both deliberate and emergent strategy formation modes are undoubtedly related to a firm's CSR, albeit they influence it via different mechanisms.While carefully planned strategies help align the environmental, social and economic concerns, emergent strategies may help integrate the organisation's purpose with stakeholders and individual business divisions (Luederitz et al., 2021).Also, emergent strategies are better for dealing with "wicked" problems, while deliberate strategies are more suitable for non-wicked ("tame") issues (Neugebauer et al., 2016).As CSR is associated with much uncertainty regarding the behavior of the actors involved (Lepoutre et al., 2007), many problems associated with CSR are "wicked" problems.Nevertheless, in smaller firms, CSR strategy making is neither purely planned nor purely emergent (Luederitz et al., 2021).Furthermore, with respect to the topic of this study, it is necessary to differentiate between the ability to solve CSR problems and the propensity to engage in CSR.Put differently, an interesting question is whether the extent of a SME's involvement in CSR might be contingent upon its chosen strategy formation mode.
Smaller firms often lack experience and financial and human resources needed to engage in CSR.For such firms, the first step toward a successful implementation of CSR should be the identification of key stakeholders and understanding their interests (Falck and Heblich, 2007).
While both deliberate and emergent strategies can contribute to the development of CSR (Neugebauer et al., 2016), it appears that firms using deliberate strategies are able to better understand individual stakeholders' needs, interests and demands, enabling them to easier demonstrate CSR (Galbreath, 2010).In addition, businesses that systematically monitor and evaluate their internal and external environment while using a formal planning process can easily respond to market and nonmarket concerns, making it easier for them to develop CSR (Galbreath, 2010).Moreover, CSR activities are often capital-intensive and expensive (Tang et al., 2012).This becomes especially important in SMEs, which face significant resource constraints, and thus, it can be expected that those SMEs that emphasize planning and rationality will be better positioned to engage in CSR.Finally, the success of CSR depends on "whether a firm engages in CSR in a predictable or erratic pattern, whether it concentrates its CSR efforts or disperses its CSR engagement in many directions" (Tang et al., 2012(Tang et al., , p. 1295)).Since an emergent strategy evolves through ongoing modifications, and "emerges out of practice in a bottom-up or undirected way" (Neugebauer et al., 2016, p. 325), it may lack coherence and build on "external events that overshadow long-term trends and structured changes" (Idenburg, 1993, p. 136).However, anticipating and adapting to long-term trends is essential for meeting the interests of both current and future stakeholders, and thus, integrating CSR into a SME's culture demands a sense of constancy, logical coherence and rationality that is typically fostered by a deliberate strategy formation.Consequently, although there is limited literature clarifying the relationship between strategy formation mode and CSR, we maintain the perspective that firms prioritizing deliberate strategies over emergent ones are more likely to be favorably positioned for increased CSR involvement.Alternatively, we expect that: H3.There is a positive relationship between deliberate vs emergent strategy and CSR.

Mediating role of strategy formation mode
The previous discussion leads to the existence of a mediation model whose graphical representation is displayed in Figure 1.First, we posit a positive relationship between LTO and CSR (H1), representing the direct effect of LTO on CSR.Second, we anticipate a positive relationship between LTO and deliberate strategy formation mode (H2).This deliberate strategy formation mode, in turn, is expected to foster CSR (H3).The combined effect of these relationships represents the indirect effect of LTO on CSR via deliberate strategy formation mode.The total effect of LTO on CSR is expected to be positive and composed of the direct and indirect effects.Formally, we anticipate that: H4.The relationship between LTO and CSR is partially mediated by deliberate vs emergent strategy.

Data collection and sample size
To collect the data set, an online self-administered questionnaire was prepared in Qualtrics and distributed through the platform Prolific Academic (Derfler-Rozin et al., 2021).The platform is often described as an alternative to Amazon MTurk, which is explicitly targeted at academic research (Palan and Schitter, 2018) and whose pool appears to provide more attentive, naı ¨ve and honest respondents as compared to MTurk or conventional nonprobability research pools (Adams et al., 2020).Among other applications, Prolific Academic has recently been used to recruit and study managers of all ranks (Anisman-Razin et al., 2023;Babalola et al., 2023;Roseck a and Machek, 2022;Sherf et al., 2019;Sherf and Morrison, 2020).
The data collection took place in the first quarter of 2022 and consisted of two steps.The purpose of the first step ("prescreening") was to reach managers who work in companies that operate in the United Kingdom and have more than 10 employees.Microenterprises with less than 10 employees were not considered for inclusion in the sample because, unlike SMEs and large firms, they often lack explicit management systems and CSR strategies and display an exceptionally low degree of formalization (Russo and Tencati, 2009).Furthermore, to ensure that the data is reliable, we imposed a minimum threshold of 97% approval rate in Prolific.We addressed a total of 3,000 respondents, out of which 598 respondents met the desired criteria.In the second step, we distributed the questionnaire and obtained 429 complete responses.Subsequently, we eliminated 49 cases with inadequately low completion times (less than 5 min) and 15 cases that failed to pass our attention checks (e.g."If you pay attention to this survey, please select 'I agree'").A total of 15 cases were not considered because of inconsistency in answers (e.g. the number of managers was reportedly greater than the number of employees).As a strict measure to prevent potentially unreliable answers, we also did not admit 68 cases because of possible "straightlining," i.e. providing identical or almost identical answers in batteries of questions (Johnson, 2016).The final sample of respondents consists of 282 managers from the United Kingdom.
In terms of demographic profiles of respondents, 52% were women and 48% were men.As to the age of respondents, 20% were younger than 30 years, 35% were 31-40 years old, 20% were 41-50 years old, 17% were 51-60 years old and 7% were older than 60 years.When considering the sample's representativeness, it is important to acknowledge its connection to the data collection method.For instance, the government's Longitudinal Small Business Survey among SME Employers reported that 53% of SMEs in the UK had women in managerial roles in 2021.While our sample had 52% women, other surveys, such as Meta's Quarterly Small Business Survey in 2022, indicated a slightly lower proportion of UK SMEs, with predominantly female management at 45%.Similarly, while our sample showed 55% of respondents under 40 years old, Meta's Quarterly Small Business Survey suggests a somewhat lower figure of 40%.Thus, our sample possibly has a younger structure with a slightly higher female proportion, which is a common case in organizational research when online panel data is used (Keith et al., 2017).Although this study does not depend on individual respondent traits, their characteristics could impact the types of businesses they are associated with.As to firm-level demographics, we consider the Longitudinal Small Business Survey among SME Employers from 2022 as a reference survey.First, 10.29% of SMEs in our sample were engaged in manufacturing, closely aligned with the UK's 9%.Likewise, 6.76% of our sample's firms were in retail, similar to the UK's 7.5%.Lastly, 18.5% of firms in the sample were in the service sector.While a direct comparison to UK figures is challenging due to the sector's diverse nature, it roughly corresponds to the aggregate percentages of firms operating in Administrative and Support Service Activities (8.60%), Accommodation and Food Service Activities (4%), and Arts, Entertainment and Recreation (5%).To summarize, although we cannot guarantee j SOCIAL RESPONSIBILITY JOURNAL j complete representation of all UK SMEs in our sample, its characteristics do not exhibit substantial deviations from the overall figures.Moreover, in light of meta-analytic evidence indicating a close convergence in quality between online panel data and conventionally sourced data (Walter et al., 2019), we maintain the belief that the chosen method of data collection does not compromise the integrity of the results.

Measures
Participants were requested to express their level of agreement with statements using a five-point Likert scale, with responses ranging from 1 (strongly disagree) to 5 (strongly agree).The dependent variable (CSR) was measured using the CSR scale developed and validated by Turker (2009).The factorial structure is four-dimensional, encompassing CSR toward both social and nonsocial stakeholders (e.g."Our company makes investment to create a better life for the future generations"), employees (e.g."Our company supports employees who want to acquire additional education"), customers (e.g."Our company protects consumer rights beyond the legal requirement") and government (e.g."Our company always pays its taxes on a regular and continuing basis").Like in Turker's (2009) study, the scale, consisting of 17 items, displays a good internal consistency (a ¼ 0.906).The independent variable (LTO) was measured by the scale developed and validated by Brigham et al. (2014).The scale consists of 12 items that consider three dimensions of LTO: futurity (sample item is "The management in our firm focuses in particular on long-term profitability"), continuity (sample item is "Preserving reputations for the longevity of the business is important to our management") and perseverance (sample item is "The management in our firm demonstrates patience for future rewards").The construct also shows a good internal consistency (a ¼ 0.897).The mediator (deliberate vs emergent strategy) was measured through the five-item emergent-to-planned strategy scale theoretically developed and validated by Slevin and Covin (1997).Following the notion that a firm's strategy formation lies on a continuum from emergent to planned strategies (Neugebauer et al., 2016), the emergent-to-planned strategy scale measures the pattern of strategy formation along this continuum.Low scores indicate an emergent strategy formation mode, while higher scores indicate a planned or intended strategy formation mode.The scale also demonstrates good internal consistency (a ¼ 0.818), with sample items such as "Formal strategic plans serve as the basis for our competitive actions," "My business unit's strategy is carefully planned and well understood before any significant competitive actions are taken" or "We typically don't know what the content of our business strategy should be until we engage in some trial and error actions" (reverse scored).
The analysis includes control variables for other factors that might influence strategy formation mode or CSR.To address maturation effects, firm age (measured as years since incorporation) is controlled for, which has been recognized as relevant in previous research on both strategy formation mode (Covin et al., 2006) and CSR (Galbreath, 2010;Wang and Bansal, 2012).Similarly, firm size is also controlled for, as it has been shown to be relevant in the study of strategy formation mode (Kopmann et al., 2017) and CSR (Wang and Bansal, 2012).Additionally, following previous authors (Graafland and Noorderhaven, 2020;Herrera-Echeverry et al., 2020;Miska et al., 2018), industry-level effects are controlled by using four dummy variables representing industry affiliation (manufacturing, wholesale and retail, services and other industries) (Miller and Cardinal, 1994).

Reliability, validity and common method bias
Confirmatory factor analysis in IBM SPSS Amos was conducted to test the quality of the measurement model.Following the literature review, LTO and CSR were treated as secondorder reflective constructs.The measurement model displays an acceptable overall fit (x 2 /df ¼ 1.790, RMSEA ¼ 0.059, CFI ¼ 0.912, TLI ¼ 0.902, PNFI ¼ 0.737), satisfying standard thresholds for goodness-of-fit indices (Hair et al., 2010).Table 1 presents the j SOCIAL RESPONSIBILITY JOURNAL j measures of construct reliability and validity.The composite reliabilities (CRs) of all constructs are higher than 0.6 and, except for deliberate strategy, the average variances extracted (AVEs) are greater than 0.5.As to deliberate strategy, AVE greater than 0.4 can be accepted if CR is greater than 0.6 (Fornell and Larcker, 1981).Thus, the latent variables display acceptable reliability and convergent validity (Hair et al., 2010).LTO and CSR also display a good discriminant validity, as the AVEs are greater than their MSVs (Hair et al., 2010).For deliberate strategy, AVE is less than MSV, suggesting a questionable discriminant validity due to a high correlation between deliberate strategy and the other constructs.Thus, we evaluate the intercorrelations and their statistical significance (Table 2).Since no correlation is higher than 0.85 (Kline, 2005), we consider that discriminant validity is not a major concern in our analysis (Chowdhury, 2018) and proceed with the use of previously validated scales (Sivertstøl, 2018).Finally, to evaluate the presence of common method bias, we conducted the Harman single-factor test.The first factor only explains 11.59% of the variance, suggesting that common method bias is not an issue in the analysis.

Data analysis methods
To test the proposed path relationships, we used the PROCESS macro for SPSS.PROCESS was developed by Hayes (2018) and allows for testing complex models with both parallel and serial mediators.In our analysis, the direct effects are estimated using OLS with heteroscedasticity-robust standard errors, and the indirect effects are estimated using the bootstrapping method (Preacher and Hayes, 2004).The procedure uses 5,000 bootstrap samples and constructs 95% confidence intervals (CIs).A direct or indirect effect is considered significant if the corresponding CI does not contain the zero value (Hayes, 2018), i.e. its lower or upper bounds are both positive or negative.Furthermore, in order to address potential self-selection bias (see Section 4.3), we used the heckit procedure from the sample selection package in R. j SOCIAL RESPONSIBILITY JOURNAL j intercorrelated, which raises concerns about multicollinearity issues.Therefore, we observed the variance inflation factors (VIFs) and tolerances.The maximum VIF was 1.808 and the minimum tolerance was 0.553, suggesting that multicollinearity is not an issue in the analysis as all VIFs are less than 10 and all tolerances are greater than 0.1 (Hair et al., 2010).

Regression results and the direct and indirect effects
Table 3 displays the results of model testing.The upper part of the table includes the regression results.Consistent with Hypothesis 2, we find that there is a positive relationship between LTO and deliberate vs emergent strategy (b ¼ 0.62, p < 0.001).The strategy formation mode does not seem to be dependent upon firm size or age.Hypothesis 3, which predicted a positive relationship between deliberate vs emergent strategy and CSR, is also supported (b ¼ 0.26, p < 0.001).CSR is also determined by LTO, thus supporting Hypothesis 1 (b ¼ 0.37, p < 0.001).Surprisingly, there is a weak and marginally significant negative relationship between firm size/age and CSR.The bottom part of Table 3 displays the total, direct and indirect effects.All the effects are positive and statistically significant, supporting Hypothesis 4, which anticipated that the relationship between LTO and CSR is partially mediated by deliberate strategy.

Post hoc tests
Our initial analysis revealed that the items forming the individual components of LTO were intercorrelated to such an extent that the three components of LTO (continuity, futurity and perseverance) could not be treated as first-order constructs and that LTO is a multidimensional construct.Nevertheless, to verify whether the results are consistent across various model specifications, we assessed a series of models in which LTO was measured by its individual components.We also considered the internal and external dimensions of CSR, as some firms (those with family involvement in management or ownership) seem to display asymmetric attitudes toward internal and external social dimensions (Cruz et al., 2014).Table 4 displays the direct and indirect effects.The results remain consistent irrespective of the dimension of LTO, and they do not depend on whether we consider the internal or external social dimensions, which supports the robustness of our results.
Additionally, given our use of online panel data, it becomes important to address potential self-selection bias.Online panel data displays two types of self-selection: topical self-selection, where the announced subject of the study influences the profile of respondents, and economic self-selection, whereby participant involvement is primarily driven by compensation considerations (Lehdonvirta et al., 2021).The former is not a concern in our analysis as the study's topic was not disclosed to participants.However, the possibility of economic self-selection remains.Consequently, we conducted a robustness test using the Heckman (1979) two-stage regression model aimed at reflecting the probability of a person answering the survey (Good et al., 2022).In our case, we build on research on online panel data, which shows that non-naı ¨ve respondents are more likely to engage in economic self-selection (Haug, 2018;Lehdonvirta et al., 2021).To capture this phenomenon, we introduced a dummy variable indicating whether a respondent's total number of approvals surpasses the sample average (Dong and Yang, 2020) 3.This indicates that self-selection bias is not an issue in our analysis (Issah et al., 2023).

Discussion and conclusion
While deliberate strategies in large corporations can appropriately address social and environmental issues, the question of how smaller firms pursue sustainability is not clearly answered (Luederitz et al., 2021).We hypothesized and found support for a partial mediation model in which an SME's CSR is affected by its LTO, both directly and indirectly, through deliberate strategy formation mode.These findings are consistent with the commonly held belief that CSR represents obligations that extend into the future and whose effects materialize in the long term (B enabou and Tirole, 2010;Carroll, 2016;Graafland and Noorderhaven, 2020), leading to a positive link between LTO and CSR.This result aligns with recent findings that future orientation is positively related to CSR (Choi et al., 2023), supporting our assumption that future-focused SMEs will be more likely to engage in CSR due to its long-term benefits (Lumpkin and Brigham, 2011).And although SMEs generally tend to have shorter planning horizons than large firms (Street et al., 2017), we also argue that when SMEs put emphasis on continuity and perseverance, two additional crucial elements of LTO (Chrisman and Patel, 2012), they will become more responsive to the needs of stakeholders and more willing to make investments that take longer to show financial returns.This finding complements and extends recent research on the relationship between future orientation and CSR in companies that publicly disclose their CSR (Choi et al., 2023).We provide a more finegrained understanding of how LTO contributes to CSR in SMEs, demonstrating that not only futurity but also the emphasis on long-lasting traditions (continuity) and the wilingness to invest effort and patience are pertinent factors in how LTO promotes CSR in SMEs.This presents a noteworthy contribution to the management literature.Our second finding is that LTO is positively related to deliberate strategy formation mode.
To the best of our knowledge, we are the first authors to examine this relationship, presenting an empirical contribution to the literature.We complement the existing studies that found relationships between long-term thinking and strategic planning (Stonehouse and Pemberton, 2002) or between "future preparedness" and deliberate strategy implementation (Kopmann et al., 2017) and explain how the individual facets of LTO can foster deliberate strategy in SMEs.
Our results also suggest that, besides LTO, an important factor that contributes to the CSR of a SME is its deliberate strategy formation mode.The finding, presenting the third contribution to the literature, is perhaps the most controversial one, as many authors consider that CSR is a "wicked" problem (Lepoutre et al., 2007;Mees-Buss and Welch, 2014), for which emergent and flexible strategies are more suitable (Neugebauer et al., 2016).However, it should be noted that we do not suggest that CSR should be tackled through deliberate strategies.Instead, we suggest that SMEs, characterized by constrained human and financial resources alongside a lack of experience, would be more inclined to actively pursue CSR when they opt for a deliberate strategy.This choice, of course, depends on various other considerations, including the prevailing environmental dynamics and the level of hostility within the firm's operational context (Neugebauer et al., 2016).The deliberate strategy will, however, make it easier to accommodate long-term trends, understand the needs of current and future stakeholders, conduct systematic environmental monitoring (Galbreath, 2010) and assess the costs and benefits of CSR, which is a costly and capital-intensive endeavor (Tang et al., 2012).Finally, our results show that there is a specific relationship between two important drivers of CSR -LTO and strategy formation modethereby addressing recent calls for an investigation into mediated relationships between drivers of CSR (Dartey-Baah and Amoako, 2021).
The study also brings about several practical implications.LTO appears to be an essential prerequisite for CSR.In other words, SME owners focusing on short-term results or displaying insufficient patience should know that their CSR may not materialize into the expected benefits.More broadly, our findings provide additional evidence that excessive short-termism, frequently observed among SMEs (Street et al., 2017), can harm the success of the firm (Kraus et al., 2006).On the other hand, we concur with Graafland and Noorderhaven (2020) that, in contrast to national cultures, the organizational LTO can be influenced by the firm, at least in part.For example, owners can encourage managers to follow long-term goals, and managers can set up corporate policies that reflect an extended time horizon.It is also possible to emphasize the business's continuity (including traditions and key values) at the organizational level.Such activities are likely to foster the CSR of the business.Also, CSR can be supported by adopting a more deliberate strategy, which requires committed leadership, engaging employees, clear vision, comprehensive planning, adaptive execution and a supportive culture that aligns with long-term goals.
Seeking external expertise and incentivizing performance can further enhance the process.
For sure, this study is not without limitations.Our data is cross-sectional, collected at one instance at a time, making it difficult to make causal inferences.Also, the data was collected through an online panel.While there are challenges associated with this type of data collection (Johnson, 2016), recent studies suggest that online panel data are increasingly used in management research (Porter et al., 2019), and their quality converges with the quality of data collected through "conventional" methods (Walter et al., 2019).We also took multiple measures to prevent and recognize participant speeding and straightlining (Johnson, 2016) and ensure that the data is reliable.
Our inquiry also suggests that there are several research gaps that future research could address.The major research gap we see is the question of how CSR relates to organizational culture.LTO can be seen as one facet of organizational culture, i.e. "the pattern of basic assumptions that a given group has invented, discovered, or developed in j SOCIAL RESPONSIBILITY JOURNAL j learning to cope with its problems of external adaptation and internal integration" (Schein, 1995, p. 222).The management literature offers multiple dimensions of organizational culture, including LTO, individual vs group orientation, internal vs external orientation and centralization vs decentralization, among others (Zahra et al., 2004).Future studies could investigate the effects of organizational culture on corporate CSR, as the results could provide important implications for management theory and practice.Additionally, future studies could consider not only the role of strategy formation mode (deliberate vs emergent strategy) but also the role of other possible typologies of corporate strategy.While it is believed that CSR is a popular tool of differentiation strategy (Chun and Cho, 2017), recent research highlights that the exclusive focus on traditional "pure" strategies (Miles et al., 1978;Porter, 1985) or "hybrid" strategies could provide misleading results.Instead, there appears to be the need for understanding the role of configurations of generic strategies and their interdependencies and contingencies with strategic attributes, such as firm size or slack resources (Greckhamer and Gur, 2021).Thus, future studies could investigate the role of strategy configurations in fostering the CSR of SMEs.

Figure
Figure 1 Theoretical framework

Table 2
Table 2 displays the basic descriptive statistics and pairwise Pearson correlation between the model variables.As expected, firm age is positively correlated with firm size.Also, surprisingly, smaller firms tend to display lower levels of CSR.The three constructs being investigated in this study -LTO, deliberate vs emergent strategy, and CSRare highly Descriptive statistics and Pearson correlations Construct reliability and validity Ã ÃNotes: ÃÃ p < 0.01, Ã p < 0.05Table 1

Table 4
(Zhang et al., 2020)2015)nt's non-naı ¨ve status.In the initial stage, a probit model was applied to test the selection equation, forecasting this dummy variable.Since the first stage necessitates the inclusion of at least one variable that serves as a regressor in the selection equation but not in the regression equation, besides firm-level control variables, we consider respondents' gender (1 ¼ female, 0 ¼ male)(Matthijsse et al., 2015)and ethnicity (1 ¼ white, 0 ¼ other)(Zhang et al., 2020), as previous research shows that both factors are predictors of survey respondents' non-naı ¨vety.Moreover, both variables are unrelated to the outcome variables in this study, making them suitable instrumental variables.For the second stage, we computed the inverse Mill's ratio (IMR) and included it as an additional predictor in two outcome equations, predicting strategy formation mode and CSR.The results, displayed in Table5, show that the IMR lacks statistical significance, and Direct and indirect effects considering alternative model specifications ÃÃ p < 0.01, Ã p < 0.05; LLCI ¼ lower bound of 95% CI, ULCI ¼ upper bound of 95% CI j SOCIAL RESPONSIBILITY JOURNAL j the overall results, including the direct and indirect effects in the mediation model, are consistent with those presented in Table