Lean inventory strategy and fashion retailers’ profitability: exploring the moderating effects of information technology intensity Open Access

The fashion industry is one of the major industries that have a negative impact on the environment. This negative impact occurs at every stage, from the production of raw materials to the sale of final products (Muposhi and Chuchu, 2022). The large amount of unsold inventory generated at the retail stage is a particularly serious challenge. In general, leftover inventory is common in the fashion industry (Wen et al., 2019) owing to the industry’s characteristics of short product life cycles, volatile and unpredictable demand, tremendous product variety, long and inflexible supply processes, and a complex supply chain (Şen, 2008). If the inventory is not sold by the end of the selling season, these garments are generally disposed of by incineration or sent to the landfill as waste. Thus, these discarded clothing items—whose production already has a negative impact—end up being disposed of unconsumed, further damaging the environment (Niinimäki et al., 2020). This is a significant waste of resources and a considerable threat to the environment.

To mitigate this threat, the lean inventory strategy can be employed as a sustainable practice with great potential to reduce unconsumed garment waste (Islam et al., 2021; Medcalfe and Miralles Miro, 2022). The lean inventory strategy is defined as inventory replenishment strategy aimed at maintaining inventory levels that are less than those of similar-sized peer firms (Barker et al., 2022; Elking et al., 2017; Eroglu and Hofer, 2011a). By managing the reorder period and ordering quantity, fashion retail companies keep their inventory levels low. Thus, this strategy can reduce unsold garment waste generated at the retail stage and mitigate the negative environmental impacts. As a large amount of unconsumed garments continue to be generated every year, more than 1 billion unconsumed garments are discarded (Nakamura and Fujita, 2018). A report released by government agencies also noted that “the first priority for the fashion business must be converting itself into a sustainable system” (Study Group on Fashion Futures, 2022, p. 8). Thus, the potential benefit of the lean inventory strategy in transforming fashion from an unsustainable industry to a sustainable one seems to be increasing in Japan and garnering increased interest.

Most previous studies reveal a positive relationship between the lean inventory strategy and firm financial performance. However, they do not provide clear and convincing insight into whether the lean inventory strategy always has a positive impact on firm financial performance. Thus, managers may hesitate to invest in implementing this strategy despite its potential for improving environmental and financial performance. First, there have been mixed findings regarding the relationship between inventory leanness and firm financial performance. Some scholars have found an inverted U-shaped relationship (Barker et al., 2022; Chuang et al., 2019; Eroglu and Hofer, 2011a, b, 2014; Hofer et al., 2012; Isaksson and Seifert, 2014; Liu et al., 2024; Modi and Mishra, 2011), whereas others have demonstrated a moderated relationship (Deb et al., 2023; Elking et al., 2017; Kovach et al., 2015; Kroes et al., 2018; Manikas et al., 2021). Notably, the evidence supporting these claims is somewhat unconvincing. An inverted U-shaped effect can masquerade as an interaction effect (or vice versa) due to their mathematical similarity (Belzak and Bauer, 2019). However, except for the study by Kroes et al. (2018), no study explicitly denies the competing explanation. Therefore, if only one of the two effects—inverted U-shaped or interaction—is included in a hypothesized model, it is possible to incorrectly conclude that one of this effects exists even when it does not (Belzak and Bauer, 2019). Thus, whether the lean inventory strategy has a positive impact on firm financial performance remains unconfirmed. That is, the boundary conditions under which inventory leanness positively impacts firm financial performance remain unclear.

To fill this research gap, this study explores the boundary conditions of the lean inventory strategy using a two-step approach (Daryanto, 2019). First, it proposes a theory-driven moderation hypothesis and tests this hypothesis using a moderated regression analysis with large panel data. Specifically, the author expects a firm’s information technology (IT) intensity to moderate the relationship between the lean inventory strategy and firm profitability. This moderation hypothesis is formulated based on the organizational information processing theory (OIPT) (Galbraith, 1974) and organizational economics (Milgrom and Roberts, 1988). Next, the author examines the robustness of the moderation hypothesis against nonlinearity using information-theoretic techniques (Daryanto and Lukas, 2022). By doing so, the author attempts to verify that the findings of this study are not a spurious moderation effect.

The rest of this paper is organized as follows. Section 2 presents the conceptual background and develops hypotheses. Section 3 provides an overview of the data and methodology, and Section 4 presents the results of the empirical analysis. Section 5 provides a discussion of the present study’s implications for theory and practice. Finally, Section 6 provides the conclusions and limitations.

The relationship between inventory leanness (i.e. lean inventory strategy) and firm performance has long been investigated by various scholars (Cannon, 2008; Capkun et al., 2009; Chuang et al., 2019; Elking et al., 2017; Eroglu and Hofer, 2011a, 2014; Isaksson and Seifert, 2014; Koumanakos, 2008; Kroes et al., 2018; Liu et al., 2024; Shockley and Turner, 2015). Most studies have demonstrated a positive relationship between inventory leanness and firm performance. However, model specifications have varied. Some studies fit a quadratic model for inventory leanness and assess the significance of the quadratic term. Upon finding a negative and significant coefficient, they reveal an inverted U-shaped relationship between inventory leanness and firm performance (Barker et al., 2022; Chuang et al., 2019; Eroglu and Hofer, 2011a, b, 2014; Hofer et al., 2012; Isaksson and Seifert, 2014; Liu et al., 2024; Modi and Mishra, 2011). In contrast, other studies fit an interaction model and assess the significance of the interaction term. After finding a positive or negative significant coefficient, they show a moderated relationship between inventory leanness and firm performance (Deb et al., 2023; Elking et al., 2017; Kovach et al., 2015; Kroes et al., 2018; Manikas et al., 2021).

These different views reveal different implications regarding the boundary conditions under which inventory leanness positively impacts firm performance. Specifically, the former type of studies demonstrated an inverted U-shaped relationship, indicating that firms can increase their performance to the point where the quadratic curve reaches its maximum. Hence, their findings reveal that the level of inventory leanness itself can be regarded as a boundary condition of the strategy because the maximum point of an inverted U curve (i.e. turning point) is determined by the level of the inventory leanness itself. The latter type of studies found a moderated relationship, denoting that the effect of inventory leanness was dependent on another variable as the boundary condition. In summary, previous studies do not provide clear insights into the boundary conditions under which inventory leanness positively impacts firm performance.

Furthermore, previous studies do not provide convincing evidence to identify the boundary conditions of the strategy. In particular, most previous studies use multiple regression, and scholars assess the statistical significance of regression coefficients on squared or interaction terms. If the coefficient of the squared term is negative and significant, an inverted U-shaped relationship is indicated, and if the coefficient of the interaction term is significant, a moderated relationship is implied. However, the regression models used in studies can produce false-positive conclusions. The regression models of these two methods are similar in mathematical structure; thus, an inverted U-shaped effect can masquerade as a moderation effect (or vice versa) (Belzak and Bauer, 2019). For example, assuming that variable X is the focal predictor and variable M is the moderator, if the correlation between X and M increases, the correlation between their squared and interaction terms will also increase (Belzak and Bauer, 2019; Daryanto, 2019; Ganzach, 1998). Nonetheless, except for the study by Kroes et al. (2018), no study presents an alternative explanation for the empirical analysis. The existing literature has not provided convincing evidence that the boundary condition depends on the levels of inventory leanness itself or on another variable.

Therefore, confirming the performance effect of the lean inventory strategy and its boundary conditions may provide valuable insights for fashion retail managers who must handle both financial and environmental performance. This is because if the boundary conditions remain unclear, they will hinder managers from investing in the lean inventory strategy. If the relationship between this strategy and firm performance is inverted U-shaped, very high levels of inventory leanness can have a negative effect on firm performance. Therefore, managers who are primarily concerned with the financial performance of lean inventory strategy may hesitate to invest in pursuing it. This will waste an opportunity to solve issues in the fashion industry from an economic as well as environmental perspective. Therefore, it is imperative to understand the boundary conditions of the strategy.

In this study, the author extends the literature discussed above by examining the boundary conditions of the lean inventory strategy. Specifically, this study first formulates theory-driven moderation hypotheses in the next section. Thereafter, the study tests these hypotheses and determines the robustness of the results using information-theoretic techniques (Daryanto and Lukas, 2022).

This study expects that a lean inventory strategy will have a positive effect on firm profitability because of the following reasons. First, most studies reported a substantially positive relationship between inventory leanness and firm performance. Even if the inverted-U hypothesis is supported empirically, several studies demonstrated that the maximum point of the inverted-U relationship lies at the extreme end of the sample (Isaksson and Seifert, 2014; Modi and Mishra, 2011). Therefore, most firms have the potential to increase firm performance by becoming leaner.

Second, considering the characteristics of the fashion retail industry, the positive effect of inventory leanness may be more pronounced. From a general perspective, excessive inventory (not lean) causes high inventory holding costs, including storage, insurance, taxes, obsolescence, and interest (Modi and Mishra, 2011). Moreover, in the fashion industry, leftover inventory is common, caused by its highly volatile consumer demand and short selling season (Şen, 2008; Wen et al., 2019). This leftover inventory negatively impacts firm profitability due to end-of-season clearance sales (Şen, 2008) and inventory write-off expenses (Kesavan and Mani, 2013).

Third, the dependent variable in this study is firm profitability. Compared with the dependent variables used in the existing literature such as financial market performance, this variable is directly affected by a firm’s operating expenses. It is expected that inventory-related expenses would severely affect fashion retail companies’ profitability. Therefore, a lean inventory strategy has a positive impact on a fashion retailer’s profitability, and this study proposes the following:

Similar to previous studies that employed the inverted-U and moderation approaches, this study expects the benefit of a lean inventory strategy to be context specific. It predicts that the performance effect of a lean inventory strategy is contingent more on variables rather than inventory leanness itself. IT intensity moderates the effect of a lean inventory strategy. These arguments are based on OIPT and organizational economics and proceed as follows.

First, from an OIPT perspective, the fit between information processing needs and capabilities can cause high task effectiveness or optimal performance (Galbraith, 1974). To contend with their uncertain environment, firms can either reduce their information processing needs (e.g. carrying slack resources) or increase their information processing capacity (Galbraith, 1974; Srinivasan and Swink, 2015, 2018). Following this framework, pursuing a lean inventory strategy is associated with a firm’s information processing needs. Particularly, a lean inventory strategy signals an increase in a firm’s information processing needs. Because a lean inventory strategy is a strategy streamlining inventory but not carrying slack resources, a firm pursuing it must increase its information processing capacity to meet its increased information processing needs. Here, one way to increase a firm’s information processing capacity is investing in vertical information systems (Galbraith, 1974; Srinivasan and Swink, 2015). Therefore, the effect of a lean inventory strategy on firm performance is dependent on a firm’s IT intensity.

This argument is also supported by Milgrom and Roberts (1988), who stated that inventories and communications of information about demand are substitutes. Thus, to adapt to its uncertain environment, a firm can use either inventories or learning demand (communications of information). From an organizational economics perspective, this choice—inventory and communication—depends on the relative cost of communications about inventories. Therefore, adopting a lean inventory strategy requires replacing inventory with the communication of information. To justify adopting a lean inventory strategy in terms of profit maximization, the relative cost of communications must be lower than that of inventories. Here, assume that investing in vertical information systems leads to a substantial decrease in the relative cost of communications within an organization or between organizations. Accepting this assumption, a firm with vertical information systems can maintain a lower cost of communications than inventories. Hence, this firm can justify adopting a lean inventory strategy and realize higher performance than firms with no vertical information systems.

Therefore, this study expects that a firm’s IT intensity will positively moderate the lean inventory strategy—performance relationship by increasing its information processing capacity or reducing the relative costs of communications within or between organizations. Therefore, the following hypothesis is proposed:

This study empirically investigated the proposed hypotheses using a secondary dataset—the Basic Survey of Japanese Business Structure and Activities (BSJBSA). It is an official statistics and an annual panel survey conducted by the Ministry of Economy, Trade and Industry (METI) of Japan. The BSJBSA data were obtained from METI under confidentiality agreements and were handled in accordance with the provisions of Articles 42 and 43 of the Statistics Act in Japan.

The BSJBSA scope covers enterprises with 50 or more employees and whose paid-up capital or investment fund is over 30 million yen. In the BSJBSA survey, firms are classified according to the Japan Standard Industrial Classification (JSIC). JSIC was established in 1949, with the current 13th revision of 2013 (JSIC Rev.13; Ministry of Internal Affairs and Communications, 2023). This study used this coding system to identify firms of interest and selected firms classified under the two-digit JSIC code 57 (woven fabrics, apparel, apparel accessories, and notions) ^[1].

The research period spanned from 2005 to 2019, and a longitudinal data set was constructed ^[2]. However, to calculate some variables, such as the sales growth value of firm i at year t, data from 2004 were also used. Hence, the initial data contained 5,724 firm-year observations of 895 firms from 2004 to 2019. The sample was selected through the following process. First, 1,742 observations were removed because of missing sales or missing or zero inventories. Firms with zero stores were also removed. Second, after introducing the control variables, 3,175 observations were removed because of missing data. Thus, this study employed a sample of 807 firm-year observations of 57 firms. Finally, to avoid presenting results influenced by outliers, we winsorized the data at the 1% and 99% levels (Dbouk et al., 2020; Kroes et al., 2018). Moreover, all monetary variables were deflated for 2015 using the producer price index calculated by the Bank of Japan (Bank of Japan, 2022).

Herein, firm profitability was measured by a firm’s return on assets (ROA). ROA has been frequently used in previous research (e.g. Cannon, 2008; Elking et al., 2017; Eroglu and Hofer, 2011a, 2011b; Isaksson and Seifert, 2014; Koumanakos, 2008; Shockley and Turner, 2015). Although some studies used Tobin’s q and stock market returns (Kroes et al., 2018; Modi and Mishra, 2011), these measures were not employed because the sample of this study was not restricted to listed firms.

ROA was defined as follows:

where $EBITDAit$ equals earnings before interest, tax, depreciation, and amortization of firm i in year t, and $TAit$ equals total assets of firm i in year t.

Lean inventory strategy was measured using the empirical leanness indicator (ELI). ELI evaluates a firm’s inventory leanness relative to firms of comparable size within a narrowly defined industry (Eroglu and Hofer, 2011a). In previous research, ELI was commonly used to measure inventory leanness (Chuang et al., 2019; Eroglu and Hofer, 2011a, b; Hofer et al., 2012; Isaksson and Seifert, 2014). As firms with a lean inventory strategy operate with fewer inventories than those not pursuing a lean inventory strategy, ELI was used to measure lean inventory strategy.

Following Eroglu and Hofer (2011b), ELI was calculated as follows. First, the inventory was calculated by averaging the prior- and current-year period-ending inventories. Second, we regressed the natural logarithm of sales on the natural logarithm of inventory for each year. Finally, we studentized the residuals and multiply them by −1 for positive ELI values to correspond to lean firms.

In line with previous IT research, the ratio of IT spending per employee (in year t) was used to measure IT intensity (Bardhan et al., 2013; Mithas et al., 2012). IT intensity was defined as follows:

where $ITspendingit$ equals IT expense of firm i in year t, and $EMPit$ is the average of the prior- and current-year period-ending total employments of firm i in year t. To capture a firm’s IT spending, “Information processing and communication expenses” ^[3] recorded in the BSJBSA survey were used. This item reflects the amount of money spent on a firm’s IT infrastructure per year. The ratio of IT spending per employee can be interpreted as the strength of a firm’s IT infrastructure supporting an employee’s information processing activities. Therefore, the higher a firm’s IT intensity, the greater is employees’ information processing capability. Hence, a firm with a higher IT intensity can absorb the increased information processing load associated with the lean inventory strategy. Thus, the ratio of IT spending per employee was used to measure IT intensity.

This study used a broad set of control variables, many of which are common in previous research: (1) size, measured as logarithm of the annual sales revenue of firm i in year t (Eroglu and Hofer, 2014; Isaksson and Seifert, 2014); (2) sales growth (Eroglu and Hofer, 2011a; Isaksson and Seifert, 2014; Shockley and Turner, 2015), measured as the year-over-year percentage change in sales revenue of firm i in year t; (3) number of stores, measured as the average of the prior and current-year period-ending total number of stores of firm i in year t (Kesavan and Mani, 2013) as a retail firm with more stores may be more likely to have massive inventory; (4) store growth, measured as the year-over-year percentage change in the number of stores of firm i in year t (Chuang et al., 2019); (5) PPE, measured as the ratio of annual sales to net plant, property, and equipment of firm i in year t (Kroes et al., 2018); (6) ADVT, measured as the ratio of advertising expense to sales of firm i in year t.

Herein, the observation unit was a fashion retail company. There may be idiosyncratic and unobservable factors, such as business models or the main product category they sell, affecting a firm’s profitability and inventory leanness. In this situation, ordinary least squares estimators would be biased, causing omitted variable bias (Wooldridge, 2002).

Given the longitudinal nature of the data, this study used a fixed-effect method. This method can account for time-invariant idiosyncratic and unobservable factors and avoid omitted variable bias (Wooldridge, 2002). Hence, this study selected a fixed-effect method to test the hypotheses. It used the R statistical software package for data analysis (R Core Team, 2021).

To avoid spurious moderation effects, it used the information-theoretic approach in moderation analysis proposed by Daryanto (2019) and Daryanto and Lukas (2022). According to Daryanto (2019, p. 110), spurious moderation can occur when a predictor and moderator variable correlate and the true nature of the relationships between predictors and a dependent variable are nonlinear. Although this study hypothesized and tested the theory-driven moderation hypothesis, several studies have demonstrated the inverted-U relationships between inventory leanness and firm performance (Chuang et al., 2019; Eroglu and Hofer, 2011a, b, 2014; Isaksson and Seifert, 2014; Modi and Mishra, 2011). Therefore, to verify the moderation hypothesis, this study followed the recommendations by Daryanto (2019) and Daryanto and Lukas (2022).

Table 1 presents the descriptive statistics and correlations of the variables included in the analysis.

To ensure that potential sources of collinearity do not affect the results, variance inflation factors (VIFs) for variables are presented. None of the variables had a VIF over 1.375. As these values were well below the threshold set by Chatterjee and Hadi (2012) (a cutoff value of 10), collinearity was not a significant issue in the analysis.

Table 2 presents the results of the fixed-effect estimations. Model 1 is a baseline model that contains only control variables.

H1 predicts that a lean inventory strategy positively affects firm profitability. Column 2 of Table 2 tests this hypothesis by adding ELI to the regression in Column 1. The ELI coefficient is positive and significant (estimate = 3.066; SE = 1.224; p < 0.05). Thus, H1 is supported.

H2 predicts that IT intensity positively moderates the relationship between a lean inventory strategy (measured by ELI) and IT intensity on firm profitability. To test this, Model 3 adds the interaction term between ELI and IT intensity to the regression model (Model 2). In Model 3, the total effects of ELI on firm profitability can be demonstrated by the following partial derivatives:

Based on this equation, this study’s results can be interpreted as follows: The ELI coefficient is positive but not statistically different from zero (estimate = 2.359, SE = 1.244, p > 0.1), whereas the coefficient of the interaction effect is significantly positive (estimate = 0.089, SE = 0.04, p < 0.05). These results imply that the ELI does not affect firm profitability when IT intensity is zero (Echambadi and Hess, 2007; Goldsby et al., 2013). If the ELI coefficient is significantly positive, similar to Model 2, the total effects of the ELI are not zero even if IT intensity is zero. However, after the interaction term is added to Model 2, the ELI’s significantly positive effect becomes nonsignificant in Model 3. Thus, when IT intensity is zero, the total effect of the ELI is not statistically different from zero. This is because although the coefficient of the interaction term is significantly positive in Model 3, the product term of the ELI and IT intensity is zero because IT intensity is zero.

Although IT intensity is conceptually treated as a moderating variable, this study also empirically obtains the same insights regarding the effects of IT intensity on firm profitability. In Model 3, the coefficient of IT intensity captures the conditional effect of IT intensity on ROA when the ELI is zero (Echambadi and Hess, 2007; Goldsby et al., 2013). In Model 3, this coefficient is significantly negative (estimate = −0.057, SE = 0.026, p < 0.05; see Model 3). The mean value of the ELI is 0.009, with a standard deviation of 0.857 (Table 1). For firms with an average ELI, the effect of IT intensity is negatively related to ROA. However, the total effects of IT intensity on firm profitability can be calculated using partial derivatives as follows:

This equation implies that if a firm has some degree of inventory leanness, IT intensity becomes positively related to ROA due to the positive interactive effect. When a firm has an ELI index above 0.64 (0.057/0.089), IT intensity is positively related to ROA. To generate a positive effect of IT intensity, a firm must be leaner than average, that is, ELI +0.75 SD (0.64/0.857). The above analysis indicates that IT intensity’s effect on firm profitability is contingent on a firm’s inventory leanness.

This study conducted several tests to check whether the results are robust. First, the present study used alternative measures for firm profitability. The analyses were re-run using return on sales as an alternative measure of firm profitability. The results are presented in Table 2 (M4–M6) and are very similar to those in Table 2 (M1–M3). All regression coefficients were consistent with those in the original models in terms of signs, magnitudes, and significance levels. Thus, the results are robust despite using different performance measures.

Second, this study used an information-theoretic approach (Burnham and Anderson, 2004; Daryanto, 2019; Daryanto and Lukas, 2022) to check the robustness of the moderation model (Model 3 in Table 2) against nonlinearity. In particular, to determine whether the interaction effect between ELI and IT intensity is confounded by their quadratic effects on firm profitability, the quadratic terms of the independent variables were added to the model. According to Daryanto (2019), seven plausible models can be derived from a linear additive model. In addition to these models, Daryanto (2019) includes another plausible model (PM 6 in Table 3) recommended by past studies (Ganzach, 1997, 1998). Therefore, following Daryanto (2019), eight plausible models were estimated and the best model was selected based on the amount of information contained in the models.

Table 3 provides information regarding the quality of these eight models such as Akaike information criterion (AIC) values and distance (D). D is the rescaled value of the AIC and is the difference in AIC of the model relative to that of the best-performing model (Daryanto, 2019). It can be interpreted as the likelihood of a model, given the available data. A model is highly plausible if D ≤ 2; however, if D > 10, the model should not be considered (Burnham and Anderson, 2004; Daryanto, 2019; Daryanto and Lukas, 2022).

In this robustness check procedure, some important results are obtained. First, using the D < 2 criterion, PM 4, PM 6, and PM 8 have substantial support. Among them, PM 4, the theory-driven moderation model developed in this study (Model 3 in Table 2), is selected as the best model. Moreover, PM 6—which contains the ELI squared term and completes the D < 2 criterion—does not provide any evidence to support the inverted-U effect of inventory leanness. Furthermore, although PM 8—which contains the ELI squared term and completes the D < 2 criterion—does not provide evidence to support the moderation hypothesis, the VIF is above Chatterjee and Hadi’s (2012) suggested cutoff value of 10. Hence, the positive moderation effect found is unlikely to be misleading or spurious due to nonlinearity effects.

This study aims to explore the boundary conditions of the lean inventory strategy. Specifically, the conditions under which the lean inventory strategy improves firm performance were investigated. There have been two competing views regarding these boundary conditions. Some scholars have considered the level of inventory leanness itself as the boundary condition (Barker et al., 2022; Chuang et al., 2019; Eroglu and Hofer, 2011a, b, 2014; Hofer et al., 2012; Isaksson and Seifert, 2014; Liu et al., 2024; Modi and Mishra, 2011). Others have posited that other factors form the boundary conditions (Deb et al., 2023; Elking et al., 2017; Kroes et al., 2018). In the present study, empirical results supported the latter view. Specifically, the results reveal that the lean inventory strategy has a positive effect on fashion retail companies’ profitability only when IT intensity is not zero, and this positive effect is strengthened by IT intensity. Moreover, the study confirms that the moderation effect is not spurious due to nonlinearity effects. Thus, this study concludes that IT intensity is the boundary condition of the strategy.

This research adds to the extant literature by providing evidence that supports the moderation hypothesis and rejects the inverted U-shaped hypothesis, which is a key theoretical contribution. An inverted U-shaped effect can masquerade as a moderation effect (or vice versa) due to their mathematical similarity (Belzak and Bauer, 2019). Nonetheless, except for the study by Kroes et al. (2018), no study has explicitly eliminated their rival explanations. This makes previous arguments weak and unconvincing. Further, we cannot exclude the possibility of a false-positive conclusion. Hence, to provide a more convincing argument, it is necessary to explicitly eliminate an alternative explanation.

In addition, this study elucidates the boundary conditions of the strategy. Specifically, it confirms that the lean inventory strategy has positive effects on fashion retailers’ profitability only when IT intensity is not zero. Furthermore, this positive impact of the strategy is augmented by firm IT intensity. These findings support previous results that proved that the effect of inventory leanness is moderated by variables other than inventory leanness (Deb et al., 2023; Elking et al., 2017; Kroes et al., 2018). In particular, the findings support those of Deb et al. (2023), who noted that the negative effect of days inventory outstanding on firm performance is positively moderated by IT infrastructure intensity. However, the findings of this study differ from those that indicate that inventory leanness moderated its own relationship with firm financial performance (Eroglu and Hofer, 2011a, b, 2014; Isaksson and Seifert, 2014; Liu et al., 2024; Modi and Mishra, 2011).

A possible theoretical explanation for the results is as follows. Pursuing the lean inventory strategy (i.e. streamlining inventory slack) has both positive and negative impacts on firm performance. The relative strength of these opposing effects determines the net effect of the lean inventory strategy. The results of the present study indicate that the performance effect of the strategy is not pronounced when IT intensity is zero. As companies cannot cope with the increase in information processing needs associated with inventory slack reduction without IT infrastructure, the performance effect of the strategy is not pronounced because the opposing effects are balanced. However, companies with higher IT intensity can address the increase in information processing needs. Thus, they are likely to receive more benefits from pursuing the lean inventory strategy.

This study reveals that the lean inventory strategy has positive effects on fashion retailers’ profitability when IT intensity is not zero. These findings offer important implications for managers who are responsible for both financial and environmental performance. First, the findings of the present study reveal that pursuing the lean inventory strategy to a substantial extent does not have a negative impact on firm profitability. The findings also reject the inverted U-shaped relationship between inventory leanness and firm profitability. Second, they indicate that IT infrastructure is a necessary condition for improving profitability using the lean inventory strategy. These findings imply that to generate business value from pursuing the lean inventory strategy, managers should also invest in IT infrastructure. Third, this study indicates that IT intensity strengthens the positive effect of the lean inventory strategy on firm profitability. Therefore, to gain greater benefit from pursuing the strategy, managers should increase their investment in IT infrastructure.

Nowadays, the fashion industry is under increasing criticism for its destructive effects on the environment (Li and Leonas, 2022; Muposhi and Chuchu, 2022). Stakeholders such as consumers and investors are demanding increasing environmental responsibility from fashion retail companies. These companies should cope with such stakeholders’ demands to ensure favorable consumer purchase intentions (Lundblad and Davies, 2016; Pérez et al., 2022; Stringer et al., 2020) and investor investment decisions (Solomon et al., 2004; Wen, 2009). Thus, managers should not hesitate to adopt the lean inventory strategy to improve financial as well as environmental performance.

This study hypothesized and tested the effect of a lean inventory strategy on fashion retail companies’ profitability. The results demonstrated the positive effect of the lean inventory strategy and the moderation effect of IT intensity on firm profitability. However, this study has three limitations that should be addressed in future research. First, IT types that affected the relationship between lean inventory strategy and firm performance are not discussed. Nowadays, numerous technologies are available in the fashion industry (Dal Forno et al., 2023). However, in this study, IT was analyzed at an aggregated level. Hence, a natural extension of this study would be to measure and examine the IT types that affect the relationship between lean inventory strategy and firm performance. An empirical study examining the moderation effect of specific IT systems would shed further light on the complexities of a lean inventory strategy.

Second, this study used a fixed-effect method to consider time-invariant idiosyncratic and unobservable factors that may cause omitted variable bias. However, multilevel linear modeling can examine the lean inventory strategy effect to capture the possible clustering structures within the dataset. There can be different firms operating in various segments of the fashion retail industry, so clustering structures can be considered. Moreover, multilevel linear modeling enables researchers to examine cross-level interactions between firm- and industry-level variables, thereby providing deeper insight into the complex effects of a lean inventory strategy.

Third, this study hypothesized the moderation effect based on OIPT, and the level of analysis was firm wide. However, the lean inventory strategy impact—inventory slack reduction—on increasing information processing needs can be more pronounced at the interfunctional level, including in the buying and sales departments. Therefore, a future study can set the level of analysis at the interfunctional level and examine the moderation effect of IT intensity at this level.

Funding: This work was supported by the Japan Society for the Promotion of Science under the grant JP20K13620.