Dodo6/Topic_Modelling_using_LDA

1Article2 3Financial Spillover Effects in Supply Chains:4Do Customers and Suppliers Really Benefit?5Erik Hofmann * and Yannick Sertori6Institute of Supply Chain Management, University of St. Gallen, 9000 St. Gallen, Switzerland7* Correspondence: erik.hofmann@unisg.ch8Received: 2 February 2020; Accepted: 4 March 2020; Published: 10 March 20209 10Abstract: Studies have shown that leading supply chain companies are associated with significantly11higher company financial ratios than competitors. In contrast, little research has focused on the12financial performance of the affiliated suppliers and customers of such supply chain leader (SCL)13companies. Thus, the central purpose of this paper is to determine, from a financial perspective,14whether suppliers and customers benefit or lose by participating in a SCL network (so called15“financial spillover effects”). Companies that were ranked in the Gartner Supply Chain Top 25 were16selected as SCLs. For each selected firm, the five largest suppliers and customers were identified17and compared with a control sample from the same industry. In order to elaborate on existing18insights into the (financial) outcome of supply chain relationships, we applied an explorative19approach with abductive reasoning, while comparing the secondary data for 224 SCL supplier (5620firms) and 168 SCL customer (42 firms) firm-years with 1940 (485 firms) and 1544 (386 firms) control21firm-years, respectively. The following insights are made: First, the superior financial performance22of SCLs was confirmed. Second, the financial performance of suppliers and customers showed23superior liquidity and activity ratios but inferior profitability ratios. Third, suppliers showed much24more significant results than customers.25Keywords: financial performance; supply chain excellence; supply chain relationship; supply chain26finance; liquidity redistribution; profitability waiver; Wilcoxon signed rank test27 281. Introduction29“Firms do not survive and prosper, solely through their own individual efforts, as each firm’s30performance depends in important ways on the activities and performance of others and hence on31the nature and quality of the direct and indirect relations a firm develops with these counterparts”32[1] (p. 123). This dependency between companies implies that relationships among companies are of33increasing importance today. As various research papers indicate, company relationships based on34collaboration are linked to higher performance [2–6]. In recent years, consensus has emerged that35competition in the market has shifted from firm against firm to supply chain against supply chain [7–3610]. Therefore, supply chain management (SCM) has greatly increased in importance and with supply37chains, today is seen as a potential source of competitive advantage [11–13]. Many studies have38shown that excellent SCM practices generate an increase in financial performance [11,13–17].39However, although many researchers have come to comparable conclusions, each study was carried40out in different ways and analyzed financial performance with different measures [10]. Although one41group of researchers used accounting-based measures [11,13,16], another group used market-based42measures [18], or a combination of both [14,17]. While most of these studies tried to differentiate43between more and less successful firms, some investigations used established rankings. For instance,44Greer and Theuri [15] applied Gartner’s (formerly AMR) Supply Chain Top 25 ranking, in order to45link SCM superiority to multifaceted firm financial performance. The authors showed the superior46Logistics 2020, 4, 6; doi:10.3390/logistics401000647 48www.mdpi.com/journal/logistics49 50Logistics 2020, 4, 651 522 of 2753 54financial performance of these supply chain leaders (SCLs). However, within their analysis, Greer55and Theuri [15] neglected to examine the impacts on the affiliated supply chain business partners of56these leading companies. Do the affiliated suppliers and customers of these SCLs benefit to the same57extent, or is the success of the SCLs, in contrast, based on the exploitation of their market position?58In a more generalized way, Crook and Combs [19] pointed out that today there is consensus on the59positive influence of effective SCM on firms’ performance, but less attention is given to the60distribution of the gains through a supply chain. It is interesting to understand whether other61members in an “outperforming” supply chain derive benefit from this kind of leadership, too, and, if62so, how. Thus, we try to close this gap by focusing on suppliers and customers of SCLs and analyzing63their (accounting-based) financial performance. We call such a financial reaction in the supply chain64a “financial spillover effect”.65The question of whether and how affiliated suppliers and customers benefit from leading supply66chain companies is mainly discussed with the topic of supply chain relationships. In that sense, Kim67and Henderson [20] classified supply chain relationships as of either a competitive (associated with68power) or cooperative (associated with embeddedness) nature. Commonly, studies in the context of69power relations have analyzed the dominant position of buying companies (customers) and70highlighted a negative impact on suppliers’ financial performance [21–23], especially a negative effect71on profitability [24]. Even confronted with powerful customers, in some cases, suppliers have also72been associated with positive returns in terms of liquidity ratios, such as shorter cash conversion73cycles [25]. Contrarily, studies analyzing the cooperative nature of supply chain relationships use74different—often non-financial—performance data [26,27], making a comparison difficult.75Cooperative-dominant studies even focus on the affiliated suppliers’ or buyers’ operational76performance improvements or efficiency effects [27–30]. Thus, there has only been limited77comprehensive analysis in the literature of the financial performance implications for the affiliated78suppliers and customers of leading supply chain firms [21,22,31]. Herein, “comprehensively” means79an impact analysis in terms of liquidity, financial activity and profitability measures.80Based on these gaps, the present study tries to answer the following research question: “Do81suppliers and customers of SCLs demonstrate superior financial ratios compared to those of other82companies from the same industry?” The purpose of this study is to investigate the financial spillover83effect in supply chains and whether suppliers and customers may derive beneficial advantages from84participating in an SCL’s network. The study is carried out by selecting firms from the three central85samples (SCLs, SCL suppliers and SCL customers). We collect financial statement information about86SCLs’ key customers and suppliers. The control samples are established by taking firms from the87same industry. The analysis is based on secondary data (financial ratios). The usage of secondary88data to build or elaborate theory is common in finance research. This is also becoming increasingly89common in supply chain and industrial organization research, as suggested by Rabinovich and90Cheon [32] and Busse [33]. We assume that publicly available practices—affiliation with a certain91supply chain network—can explain variations in firm-level (financial) performance. Based on92Gartner’s Supply Chain Top 25 ranking, we identified a set of SCLs. A Wilcoxon signed rank test was93conducted. Following an explorative study approach with abductive reasoning, we identify some94interesting correlations and state propositions at the end.95This study shows that suppliers and customers are associated, on one hand, with superior96liquidity and activity and, on the other, with inferior profitability ratios compared with peers from97the same industry. These findings confirm the existence of financial spillover effects in supply chains,98highlighting correlations between supplier and customer ratios and demonstrating that SCLs’99business partners can derive financial performance benefits (and losses) from their participation in100an SCL network. In accordance with Kim and Henderson [20], it is suggested that in such supplier–101customer relations, competitive and cooperative relationships co-exist. The present findings also102contribute to the work of Gulati and Sytch [28] who, in relation to interdependency, distinguished103between asymmetry dependency (in terms of the logic of power) and joint dependency (in terms of104the logic of embeddedness). The contribution for practitioners is a profound understanding of the105 106Logistics 2020, 4, 6107 1083 of 27109 110potential consequences of participating in an SCL’s network. Nevertheless, this topic warrants further111exploration in future studies.112This paper is organized as follows: In the second section, the theory and a literature review about113the central topics are provided. In the third section, the methodology is presented, explaining the114explorative study approach with abductive reasoning, data sources, sample selection and data115analysis methods. In the fourth section, the results for each of the three samples are highlighted116separately. In the fifth section, the findings are discussed, and propositions are put forward. Finally,117in the sixth section, a conclusion is given, as well as implications for theory, practice and future118research.1192. Research Context1202.1. Supply Chain Management and Firm Financial Performance121Johnson and Templar [34] state that SCM “directly impacts organizational profitability, liquidity122and productivity and the effects can be measured and reported via the financial statements and ratios123used to externally monitor organizational performance” (p. 91). In practice, the Pricewaterhouse124Coopers Global Supply Chain Survey 2013 came up with six key findings about how leader firms125manage supply chains. One of these findings is that improving a company’s supply chain leads to126around 70% higher financial and operational performance [35]. In addition, Deloitte Consulting LLP127[36] (p. 2), in their 2014 Global Supply Chain Survey, which selected leader and follower companies,128showed that the majority of SCLs demonstrate well above average revenue growth and earnings129before interest and taxes (EBIT) margins. In this study, leaders and followers were selected by130questioning a large number of executives (more than 400) in manufacturing and retail firms about131distinctive supply chain approaches and by using two performance metrics (inventory turnover and132percentage of deliveries that are on time and in full) [36] (p. 1).133The link between the efficiency of supply chain processes and financial performance, which134finally improves value for shareholders, has been established in the literature [37–39]. Christopher135[40] depicted the three main financial dimensions of a firm that are influenced by SCM: cash flow,136profit and resource utilization. A firm’s profitability depends to a large extent on how costs are137managed in a supply chain [41]. An increase in supply chain efficiency, which reduces costs, increases138a company’s short-term profit [42]. To calculate cost efficiency, the cost to sales (cost of goods sold139(COGS) to sales) ratio is a commonly used metric [43]. In terms of liquidity, the cash conversion cycle140is a metric for measuring the net cash flow generated from a firm’s assets [44]. In literature, it is also141named cash-to-cash [45], cash cycle [24] or net operating cycle [46]. For the purpose of uniformity142during the study, the name cash conversion cycle (CCC) will be used. A reduction in inventory and143accounts receivable and an extension of accounts payable are the drivers for improving the CCC144financial ratio.145The literature on the financial impacts of SCM performance can roughly be clustered into two146groups [10]: supply chain excellence and supply chain disruptions. While the first group of literature147analyzes the contribution of exemplary SCM practices [11,14–17,47], the second group focuses on148supply chain disruptions and their negative financial impact [48–50]. For the present analysis, the149first group is of greater interest and is further examined.150D’Avanzo et al. [14] showed that leader firms, compared with the industry average growth rate,151are associated with significantly higher market capitalization growth rates. In this study, leaders were152selected through “a statistical correlation between companies’ financial success and the depth and153sophistication of their supply chains” (p. 40). The leading companies were found through different154sources, including, among others, the Gartner Supply Chain Top 25 ranking [51]. Whereas Ellinger155et al. [11] showed that leading companies have a higher Altman Z-score compared to their peers,156Greer and Theuri [15] found that SCLs outperform non-SCL competitors in terms of accountingbased costs, activity and liquidity ratios, proving that SCLs are financially healthier. Swink et al. [17]157showed that leading companies in SCM, compared to competitors, are associated with significantly158better financial and market metrics.159 160Logistics 2020, 4, 6161 1624 of 27163 164Regarding the financial strength of the SCLs, two other issues are combined. First, the question165arises from where the better financial and market metrics come. Second, it is of interest what happens166with these financial resources (retention vs. disbursement). The SCM literature provides various167theoretical explanations for both points.168Regarding the first point (the source of superior financial metrics), two different perspectives169are to be distinguished. A positive “success” was realized based on their own strength, including the170unilateral exploitation of an asymmetrical power situation [28], which amounts to the temptation to171pursue opportunistic behavior [52]. If this is the dominant point of view, it is not expected that related172suppliers and customers will profit to the same extent [53]. In contrast, an argument can also be made173for the joint activities in the supply chain as the reason for the positive success. A possible explanation174can be found in the relational exchange and the related joint learning and knowledge transfer175approach [54,55]. The joint dependency approach also contributes to this [28]. Gulati and Sytch [28]176found that while joint dependency improves the performance of manufacturers in procurement177relationships, this effect is partially moderated “by the level of joint action and the quality of178information exchange between the partners” (p. 32). Especially in long-term relationships, the supply179chain efficiency improves [56], and the performance of key suppliers increases [57].180Regarding the use of the financial resources, there are again two possible approaches. Initially,181the resources could stay in the company as profit retention, or they could be distributed to own182shareholders (both leading to a negative spillover effect). Furthermore, an additional possible183explanation could be found in the supply chain-oriented interpretation of the resources dependence184theory, in which the supply chain members recognize that dependence can create forbearance and185trust [58]. In such a model, SCL-related suppliers and customers would positively profit though an186inter-organizational redistribution of financial resources in the supply chain.187Regarding these issues, research gaps arise as the perspectives predominantly applies to SCLs188has not yet been completely clarified. Furthermore, it is unclear to what extent the considerations189between the categories of financial figures differ. It is quite possible that the attitude of SCLs differs190regarding the provision of liquidity compared to profit sharing.1912.2. Supplier and Customer Performance Implications from Supply Chain Relationships192As Tan et al. [59] (p. 1047) point out, companies today cannot exist in isolation, and for this193reason, the integration of suppliers with their customers, as well as customers with their suppliers, is194central to achieve, among other indicators, financial goals. This fact leads to the question of what195exactly the benefits are for suppliers and customers in supply chains.196To date, different studies have explored the benefits for different actors operating in supply197chains. It is striking how many ways the researchers who have investigated this topic differ, in areas198such as industry focus and/or the element of SCM that is analyzed, such as total quality management199(TQM) [59,60], efficient consumer response [61] or information technology [31]. Moreover, central to200this academic field is the analysis of the benefits that can be drawn from different relationships201between actors in supply chains. Various studies, among others, have analyzed the performance202effects of supply chain relationships. Whereas research concerned with supply chain relationships203differs in the analysis of the members in a relationship (e.g., supplier–manufacturer [21] or supplier–204retailer [28]), the literature on the benefits of suppliers and customers in supply chains has come to205different results and conclusions. Kim and Henderson [20] categorize the existing literature on supply206chain relationships as being of either a competitive or cooperative nature.207Although competitive relationships are often analyzed in power situations, the cooperative208viewpoint focuses on embeddedness. Deepening the study’s findings related to power contexts, Kim209and Wemmerlöv [23] found that customer power, created by the supplier’s dependency, is positively210associated with increasing cooperation, which has a negative impact on the supplier’s financial211performance. By analyzing the relationship between suppliers and main customers, Patatoukas [25]212has shown that the buyer’s power has a negative effect on the supplier’s profitability by reducing the213gross margin and has a positive influence by reducing selling, general and administrative expenses214(SG&A), enhancing asset utilization and shortening the CCC. Through an examination of supplier–215 216Logistics 2020, 4, 6217 2185 of 27219 220seller–buyer relationships, Lanier et al. [24] suggest that, whereas profitability in supply chain221relationships is more likely to be sourced from downstream supply chain partners, CCC benefits are222gained by every member in the chain. Furthermore, Gosman and Kohlbeck [21] analyzed the223consolidation in the retail market. They concluded that the increase in bargaining power for retailers224decreased suppliers’ gross margin. Confirming these results, other studies proved that customer225concentration and power increase the dependency of suppliers [62] and decrease suppliers’226profitability [22]. Furthermore, in line with Duffy and Fearne [63], the literature shows that power227asymmetry can have a negative impact on the distribution of benefits in supply chain partnerships.228Although powerful buyers have a negative impact on a supplier’s performance, and especially229on profitability [21], in some cases, suppliers are also associated with positive returns in liquidity230ratios, such as a shorter CCC [24,25]. A question that arises in connection with the power asymmetry231and the supposed dependency of suppliers from their customers and of customers from their232suppliers is about the reasons that lead to this situation, particularly as it involves a more or less233exposed SCL. According to the strategic choice approach, from a supplier’s perspective, it could have234been a conscious decision based on the attractiveness of winning the SCL as a customer. The235attractiveness of the SCL, from the supplier’s perspective, can be based upon its brand strength or its236reputation or due to its size and the related market potential. Within an existing supplier–buyer237relationship, the degree of dependence varies depending on how much a creditor can afford to lose238[64]. In order to acquire the SCL as a customer, the suppliers can make a specific investment (e.g.,239implementation of a specific IT system or adaptation of processes). Such investments or the240acceptance of a discount on their sales lead to a reduction in the supplier’s profitability.241A similar argumentation can be made from the perspective of a SCL customer. Studies suggest242that a retailer’s (customer’s) ability to partner with focal suppliers (SCLs) is key to category243management effort and performance [53]. Due to the importance or criticality of the products of the244SCL, the customer even feels obliged (as a strategic choice) to accept the products of the SCL against245a substantial price premium. In doing so, the customer’s COGS rise, and his or her profitability sinks.246Thus, we call this reasoning “profitability waiver thesis”. If possible (products or materials of the SCL247supplier are substitutable), the customer can reduce the dependency by the development of dual or248even multiple sourcing.249Summing up, although the link between SCM and financial performance and the link between250supply chain excellence and financial performance superiority has been studied in depth, only a few251studies have analyzed suppliers or customers of SCL firms or the (financial) performance distribution252through supply chains. No study has comprehensively explored the financial implications253(spillovers) of being member of an SCL’s network for suppliers and customers. This fact is254astonishing, as the rationales behind supply chain relationships—a competitive viewpoint based on255power regimes vs. cooperative viewpoints based on embeddedness—are for their part again well256examined. Thus, the central purpose of this paper is to identify the potential implications for257suppliers and customers who participate in an SCL’s network.2583. Methodology2593.1. Approach and Data Source260We follow an explorative research approach with abductive reasoning [65]. Meaning, our261reasoning starts with a deviating empirical observation, followed by an iterative reflection262(discussion) with existing theory. According to the consistence (“degree of matching”) between the263empirical results and existing theory, new insights in the form of propositions are suggested.264The present study is based on secondary financial data. The usage of secondary data is becoming265increasingly common in logistics and supply chain research, as suggested by Rabinovich and Cheon266[32] and Busse [33]. A key advantage of using secondary financial data is that the sample firms must267not be kept anonymous. All of the examined companies of the study at hand are listed in appendix268A. The two databases used to export the data for this study were (i) the Bloomberg terminal and (ii)269the Compustat database.270 271Logistics 2020, 4, 6272 2736 of 27274 275“Bloomberg is the popular subscription-based data service that provides a vast array of financial,276economic and general information to users” [66] (p. 49). It provides all kinds of information on277equities, bonds, etc. [67,68]. For the present study, the function “SPLC <GO>” is important. This action278provides information about the supply chains of companies and allows users to explore and deepen279the topic of SCM. This function shows the key suppliers, customers and competitors of a focal firm280and displays money flows between these firms.281In addition to Bloomberg, the Compustat database was used. This database covers282approximately 98% of the world’s total market capitalization, more than 65,000 firms globally, of283which more than 45,500 are non-North American securities [69]. In similar studies, this database was284used, among others, to find control firms for the statistical test and important information about285companies [4,15,70,71]. For the present study, the Compustat database was used to export and select286the control sample firms.2873.2. Sample Selection288The investigation is carried out from a simple supply chain perspective. We analyze the first-tier289suppliers (upstream) and the first-tier customers (downstream) of long-term SCLs. Thus, the sample290selection is based on real supplier–SCL or SCL–customer relationships. In order to diminish one-time291effects, we looked for firms with complete data sets from a four-year time period (2011–2014). The292overall sample selection process encompasses a five-step procedure (Figure 1).293 294.295Figure 1. Procedure for selecting the samples.296 297Step 1 encompasses the identification and selection of SCL firms (central sample 1). For this298purpose, the Gartner Supply Chain Top 25 was used [72]. The Supply Chain Top 25 ranking was299created from AMR Research in 2004. In 2010, AMR Research was acquired by the Gartner Group,300which then took over the execution of this study [73]. The main goal behind the Gartner classification301and analysis is “identifying global supply chain leaders and highlighting their best practices for heads302 303Logistics 2020, 4, 6304 3057 of 27306 307of supply chain and strategy organizations” [74] (p. 15). Gartner [75] uses a composite score to308determine its top 25 list. The composite score is divided into two components, a financial component309and an opinion component, which are weighted 50% of the total score. The financial component310comprises three financial metrics: return on assets (ROA; 25%) = net income / total assets; inventory311turns (15%) = COGS / inventory; and revenue growth (10%) = change in revenue from the prior year.312The opinion component is composed of two different panels, which are equally weighted: Gartner313analyst experts (25%) and peer opinionists (25%). Finally, all the information is normalized on a 10point scale and combined with the weightings named above in a total composite score. To select314“real” and long-term SCL firms, we decided to choose companies that were leaders for several years.315For this reason, central sample 1 was restricted to companies that were listed as leaders in the Gartner316Supply Chain Top 25 ranking (in every one of the years 2011 through 2015). The information about317the companies was retrieved from Aronow et al. [76], Aronow et al. [74], Hofman [77], Hofman and318Aronow [78] and Hofman et al. [79]. After eliminating one firm without a substantial group of control319firms, data of 17 unique SCL firms (68 firm-years) was gathered.320In Step 2, the biggest suppliers (central sample 2) and the biggest customers (central sample 3)321of each SCL were selected. The “<SPLC> GO” function in the Bloomberg terminal was used to detect322the firms of the SCL networks. Looking upstream the supply chain, it is possible to break down four323groups of suppliers, depending on the different types of costs of the SCL they supply to: COGS324suppliers, SG&A suppliers, CAPEX suppliers and R&D suppliers. As these different suppliers tie325different costs of focal firms (e.g., while from an accounting view COGS and SG&A expenses are326stated in the profit and loss statement, CAPEX and R&D expenses have to be capitalized and are327listed on the balance sheet), they must be analyzed separately. For a practical purpose, this study328restricts the analysis to the five biggest COGS suppliers (measured as % of COGS of SCLs) only,329because the COGS is important for SCM, given that it “reflects the direct costs and overhead330associated with the physical production of products for sale” [80] (p. 276). After adjustments (e.g.331elimination of duplicates), the central sample 2 “biggest suppliers per each SCL” covered 56 unique332firms, making—according the chosen four years period to elaborate long term relationships with the333SCL (see above)—224 firm-years. Looking downstream the supply chain, the five biggest customers334per SCL (measured as % of revenue of SCLs) were selected. After adjustments, the central sample 3335“biggest customers per each SCL” included 42 unique firms and 168 firm-years. Table 1 summarizes336the two steps and sub-steps for identifying and selecting the three central samples.337Table 1. Procedure for identifying the firms of the three central samples.3381.13391.23401.33411.43422.13432.23442.33452.43462.5347 348Step 1: SCLs sample selection349SCLs350Firms that were ranked at least one year between 2011 and 2015 in the Top 2535136352Firms that were listed every of the last five years (2011–2015) in the Top 2535318354Firms after eliminating companies due to a lack of control firms in the same35517356industry357Firm-years (2011–2014)35868359Step 2: Suppliers and customers sample selection360Suppliers Customers361Firms*3629036390364Firms after eliminating duplicates**3657136649367Firms after eliminating companies due to a lack of control firms3685636943370Firms after eliminating companies due to a lack of available data3715637242373Firm-years (2011–2014)374224375168376* The five biggest suppliers (measured as % of COGS of SCLs) and customers (measured as % of377revenue of SCLs) of each of the identified SCL firms identified in sub-step 1.2.378** Firms that are suppliers or customers for two or more SCL firms are eliminated.379 380In Steps 3 to 5, the control samples are selected. The present study compares SCLs, SCL-suppliers381and SCL-customers with their peer groups in the same industry. This means that for every company382in the central samples, all appropriate firms in the same industry (the same four-digit Standard383 384Logistics 2020, 4, 6385 3868 of 27387 388Industrial Classification (SIC) code) were taken as the control sample [71]. To determine the389benchmark of the industry, the median of the ratios was then calculated. Medians of at least three390ratios (three available sets of company data) were used. To refine the control firms’ selection and the391benchmarking, the following procedure was applied. First, every firm still active in the Compustat392database was downloaded with the four-digit SIC code (Step 3). Second, only the firms with an exact393match based on the four-digit SIC codes of the central sample firms were considered (Step 4). Third,394this sample was further restricted by including only firms that, in 2014, were at least mid-caps,395meaning a market capitalization > USD 2 billion (Step 5). Appendix B shows the distribution of the396three central samples’ companies in market capitalization. This last step is relevant for selecting more397robust control samples, considering that the central samples also contain bigger firms. Therefore, it398makes sense to take large companies, which thus serve as stronger control firms, and the results can399then be more significant. We consciously decided to apply the following design in order to have a400larger database. On one hand, if we had decided on control sample firms with a market capitalization401> USD 5 billion, we would have had to remove 7 of the 17 SCL companies from the analysis, due to402missing of control sample data, thus drastically reducing the number of analyzed firms. On the other403hand, we decided to use control sample firms > USD 2 billion as this is the entry level barrier for midcap companies. Figure 2 displays the number of firm-years (N) in the samples and summarizes the404sources or databases that were used to retrieve the information.405 406Figure 2. Number of firm-years in the samples and sources of data.407* Sample selection source: Gartner Supply Chain Top 25.408** Sample selection source: Bloomberg.409*** Sample selection source: Compustat. All company ratios are downloaded from Bloomberg.410— Supply chain relationship.411- - Wilcoxon signed rank test.412 4133.3. Data Analysis414The present study is based on Greer and Theuri’s [15] work of analyzing the cost, liquidity and415activity ratios of SCL firms. In order to take the study one step further, some ratios were added. The416liquidity ratios are the operating cycle, CCC, current ratio and cash turnover. In contrast to Greer and417Theuri’s [15] study, the operating cycle was classified as the liquidity ratio, considering that it418measures the time from acquisition of inventory and the realization of cash from sales. At the same419time, the liquidity ratio is also very similar to the CCC, also named the net operating cycle, which is420definitely classified as a liquidity ratio [46]. The CCC extends the operating cycle analysis by421introducing accounts payable. Stewart [81] describes the CCC as “a composite metric describing the422average days required to turn a dollar invested in raw material into a dollar collected from a423customer” (p. 43). A positive CCC means that the focal company has increasing costs because of424additional borrowed money [45]. If the CCC is short or even negative, the firm is receiving its425accounts receivable from customers before paying its accounts payable to suppliers. In such a426constellation, the company increases its liquidity and is financed by suppliers [44].427 428Logistics 2020, 4, 6429 4309 of 27431 432As activity ratios for this study, the following are selected: receivables turnover, asset turnover433and days inventory outstanding (DIO). This is similar to the activity ratio analyzed by Greer and434Theuri [15]. As mentioned, the operating cycle was displaced in the liquidity category.435Profitability is a new category, in comparison on Greer and Theuri’s [15] work. For the purpose436of this study, the two cost ratios were added to the profitability category, because of their direct437impact on a firm’s profit. The two additional profitability ratios, EBIT margin and return on capital438employed (ROCE), look at the profit earned by the firm. While the EBIT margin concentrates on the439income statement, the ROCE also incorporates the balance sheet and looks at the earning power of440the total capital. Table 2 lists all the chosen ratios.441Table 2. Selected financial ratios.442 443Profitability444 445Activity446 447Liquidity448 449Ratios450 451Formula452 453Bloomberg description*454Measures the time455Accounts receivable turnover in between the acquisition of456inventory and the457Operating cycle**458days +459realization of cash from460inventory turnover in days461sales of inventory.462Metric which expresses463Accounts receivable turnover in464the length of time, in days,465days +466that it takes for a company467inventory turnover in days –468Cash conversion cycle469to convert resource inputs470accounts payables turnover in471into cash flows.472days473Ratio to indicate the474company's ability to pay475Current ratio476Current assets / Current liabilities477back its short-term478liabilities with its shortterm assets.479Cash and near cash480turnover ratio measures481Cash turnover482Sales / Cash483how effective a company484is utilizing its cash.485Ratio that measures how486many times a business can487Receivable turnover488Sales / Average total receivables489collect its average accounts490receivable during the year.491The ratio is an indicator of492the efficiency with which a493Assets turnover494Sales / Average total assets495company is deploying its496assets.497Average number of days498that goods remain in499Days inventory outstanding500(Inventory / COGS) x 360501inventory before being502sold.503Earnings before interest504EBIT margin505(EBIT / Sales) x 100506and taxes (EBIT) as a507percentage of net sales.508Ratio which indicates the509company's ability to use510Return on capital employed (EBIT / Capital employed) x 100511its capital investment512efficiently.513 514Logistics 2020, 4, 6515 51610 of 27517 518COGS to sales519 520(COGS / Sales) x 100521 522SG&A to sales523 524(SG&A / Sales) x 100525 526Percentage of revenue (net527sales) used to pay costs of528goods sold.529Selling, general, and530administrative (SG&A)531expenses as a percentage532of total sales.533 534*Considering that the data for the analysis are downloaded from the Bloomberg database, if not535otherwise specified, the ratio definitions are retrieved from the find fields in the Bloomberg tab. The536names and formulas of the ratios were simplified and unified to promote understanding.537**Bloomberg does not have a direct formula for the operating cycle, so it must be calculated with the538two indicated components.539 540The data was gathered from 2011 until 2014 for each company from the Bloomberg database.541The ratios of the firms in the control samples were used to calculate the industry benchmark. The542median of all ratios of the same year of each firm in the same four-digit SIC code was calculated. In543this study, the median is preferred to the arithmetic mean because “medians are considered to be544better indicators. As accounting data are not normally distributed, the medians are extremely robust545to outliers and other deviations from normality” [70] (pp. 181–182).546In order to identify significant differences between the ratios of suppliers and customers of SCLs547and the ratios of control firms in the same industry, we used the Wilcoxon signed rank test (for a548theoretical deepening of the approach, see Weiers [82] (pp. 505–517); for additional examples of the549application of the approach in literature, see Greer and Theuri [15], Santhanam and Hartono [71],550Bharadwaj [70] and Kalwani and Narayandas [4]). This approach is a nonparametric test, which can551be used for comparing paired samples. As the name reveals, the test compares a sample with another552(paired) sample [82] (p. 513). According to Weiers [82], “[a] nonparametric test is one that makes no553assumptions about the specific shape of the population from which the sample is drawn” (p. 506).554Thus, the Wilcoxon signed rank test is the nonparametric counterpart of the paired-samples t-test555and compares two dependent samples. Compared to the paired t-test, the Wilcoxon signed rank test556is seen as more robust against outliers [83]. Another advantage is the small sizes samples can have.557The logic and procedure for detecting the test statistic W is listed (the procedure is from Weiers558[82] (pp. 508–517)):5591.5602.5613.562 5634.5645.565 566Create a column with the xi and yi samples and structure them in a way that the567paired samples are on the same line.568Calculate the difference between the paired observations, di = xi – yi. This is also the569measurement of interest for the purpose of the test.570Calculate the absolute value of di = IdiI, by ignoring di = 0 values and calculating for571tied rows the average of their position. The IdiI has to be ranked in descending order,572so that the smallest IdiI has rank 1.573List the rank of observations where xi > yi in the R+ column.574Finally, calculate the test statistic by summing these positive differences (W = ∑R+).575 576If the number of the nonzero differences (di ≠ 0) exceeds 20 (n ≥ 20), it is possible to use a normal577approximation of the Wilcoxon signed rank test. For the purpose of the study, this assumption is578satisfactory. For the normal approximation, we need the z-formula, which looks like the following579[82] (p. 510):580(581 582𝑧=583 584(585 586)(587 588)589)590 591,592 593where W is the sum of the R+ ranks, and n is the number of observations for which di ≠ 0.5944. Results595 596(1)597 598Logistics 2020, 4, 6599 60011 of 27601 6024.1. Relationship Distribution603Before the results are described, the relationship between the supply chain members and the604central samples’ distribution in industries is briefly shown. This provides the basis for the discussion605and helps to interpret the findings. The relationship between suppliers and SCLs and SCLs and606customers is displayed in Figure 3. It shows the companies’ percentage of revenue or COGS they607impact. For the data information, the mean, median and standard deviation (SD) of the analyzed608companies are provided. Generally, it appears that SCLs have a lower percentage (in the mean and609the median) in costs and revenues than the suppliers and customers. This shows that SCLs are less610dependent on their relationship with suppliers and customers than vice versa. Looking at the611suppliers and customers, it becomes clear that suppliers (SCLs % of revenues) are the most dependent612companies. They exceed the customers (SCLs % of COGS) by 1.82 percentage points in the mean and613by 1.09 percentage points in the median.614 615Figure 3. Relationship strength within the SCL networks.616Table 3 classifies the central samples in the SIC divisions, in order to get an initial impression of617the position of the sample companies in the supply chains. Although most of the SCLs (70.6%) and618suppliers (85.7%) are active in the manufacturing division, 40.5% of the customer firms work in the619retail trade business. Many customer companies are also active in manufacturing (38.1%). See more620information about the classification of the samples in SIC code in Appendix A and about the samples621descriptive data in Appendix B.622Table 3. Distribution of central sample firms in SIC divisions.623SIC division624Manufacturing625Transportation & Public Utilities626Wholesale Trade627Retail Trade628Finance, Insurance, Real Estate629Services630 631N6321263316344635-636 637SCLs638%63970.6%6405.9%64123.5%642-643 644Suppliers645N646%64748 85.7%64836495.4%65036515.4%65216531.8%65416551.8%656 657Customers658N659%6601666138.1%662666314.3%66436657.1%6661766740.5%668-669 670The descriptive results illustrate that whereas SCLs and suppliers are one step before the671consumption point, most of the customers have direct contact with end consumers.6724.2. Supply Chain Leaders (SCLs)673 674Logistics 2020, 4, 6675 67612 of 27677 678In line with Greer and Theuri [15], we conducted an analysis to confirm their results, that SCL679firms have significantly better financial ratios than control firms in the same industries. Table 4680presents the results for SCLs. Although 76.5% of the analyzed companies are from the US, all firms681are at least large caps (70.6%) or even bigger, mega-caps (29.4%). Far more than half of the companies682(70.6%) have more than 100,000 employees.683Table 4. Comparison of ratios for SCLs and control firms.684Ratios685 686Liquidity687 688Operating cycle689Cash conversion cycle690Current ratio691Cash turnover692 693Activity694 695Receivable turnover696Asset turnover697Days inventory outstanding698 699Profitability700 701EBIT margin702Return on capital employed703COGS to sales704SG&A to sales705 706Sample707SCLs708Control709SCLs710Control711SCLs712Control713SCLs714Control715SCLs716Control717SCLs718Control719SCLs720Control721SCLs722Control723SCLs724Control725SCLs726Control727SCLs728Control729 730N731687321043733647349407356873611147376873811117396874010817416874211047436874410577456874611157476774885574968750107775160752882753 754Mean75589.569756120.90575729.06275862.5847591.6127601.82276114.26576210.63276318.72476413.2137651.0597661.10976759.31676875.60876918.40877010.22977142.76177221.15377351.81277459.33177523.59877622.649777 778Median77991.383780112.07278134.48378246.5347831.3517841.51278511.1307868.26078710.6927889.1057890.9147901.02379160.54179263.50579318.33579410.14979532.30679620.23479751.92479855.58479925.30780020.320801 802Z8035.505***8044.641***8052.542***806−3.391***807−3.660***8080.9848094.729***810−6.251***811−5.703***8123.837***813−1.229814 815Significance: * 10% level, ** 5% level, *** 1% level.816 817In the following, each ratio result is briefly described (mainly the medians).818Liquidity Ratios. The operating cycle (91.383 for SCLs vs. 112.072 for control firms) and the CCC819(34.483 for SCLs vs. 46.534 for control firms) of SCL firms are statistically significantly lower (at the8201% level). The current ratio (1.351 for SCLs vs. 1.512 for control firms) is also shown to be statistically821significantly lower (at the 1% level). A general rule for the current ratio is that it should total 200%822(this means looking at the results 2) and be at least 100%. In this case, the control firms seem to have823safer current ratios, which contradicts Greer and Theuri’s [15] (pp. 102–103) findings. Even if the824control firms seem to have a better ratio, the SCLs’ current ratio nevertheless exceeds the minimum825limit of 1. The cash turnover (11.130 for SCLs vs. 8.260 for control firms) of SCLs, however, is826statistically significantly higher (at the 1% level) compared with that of the control firms. The lower827operating cycles and CCCs together with the higher cash turnover ratios imply that SCL firms are828associated with significantly better liquidity ratios than control firms in the same industries.829Activity Ratios. The statistically significantly higher (at the 1% level) receivables turnover (10.692830for SCLs vs. 9.105 for control firms), together with the statistically significantly lower (at the 1% level)831DIO (60.541 for SCLs vs. 63.505 for control firms) show that SCL firms have better activity ratios832compared to firms in the same industries.833Profitability Ratios. The EBIT margin (18.335 for SCLs vs. 10.149 for control firms) and the ROCE834(32.306 for SCLs vs. 20.234 for control firms) of SCL firms are statistically significantly higher (at the8351% level) than their respective industry benchmarks. The COGS to sales ratios (51.924 for SCLs vs.836 837Logistics 2020, 4, 6838 83913 of 27840 84155.584 for control firms) are shown to be statistically significantly lower (at the 1% level) than that of842peers in the same industry. The SG&A to sales ratios are higher (25.307 for SCLs vs. 20.320 for control843firms) for SCLs, but the results do not reach statistical significance. This seems to be in line with the844results obtained by Bharadwaj [70] (p. 182) for IT leaders. In sum, the higher EBIT margin and ROCE845and lower COGS to sales show that SCL firms are associated with better profitability ratios.846In summary, SCLs are associated with significantly superior liquidity, activity and profitability847ratios. These results are in harmony with that of Greer and Theuri [15]. The following analysis extends848their study by analyzing the suppliers and customers of SCL firms.8494.3. Suppliers850The suppliers of SCLs are, similar to the SCLs for their part, mostly US companies (51.8%).851Around one fifth (21.4%) of the firms in this sample are from Asian countries: Japan (8.9%), South852Korea (7.1%) and Taiwan (5.4%). The majority of the suppliers are large caps (42.9%) and mid-caps853(39.3%). Most (67.8%) have more than 10,000 employees, 21.4% of which have more than 100,000854workers. Compared to the SCLs sample, the suppliers’ sample contains smaller companies: 85.7% of855these firms are active in the manufacturing SIC division. The results for the suppliers of SCL firms856are displayed in Table 5.857Table 5. Comparison of ratios for suppliers of SCLs and control firms.858Ratios859 860Liquidity861 862Operating cycle863Cash conversion cycle864Current ratio865Cash turnover866 867Activity868 869Receivable turnover870Asset turnover871Days inventory outstanding872 873Profitability874 875EBIT margin876Return on capital employed877COGS to sales878SG&A to sales879 880Sample881Suppliers882Control883Suppliers884Control885Suppliers886Control887Suppliers888Control889Suppliers890Control891Suppliers892Control893Suppliers894Control895Suppliers896Control897Suppliers898Control899Suppliers900Control901Suppliers902Control903 904N9051989061407907191908127990921691017109112209121915913213914167191521791619079171999181419919217920171292116792298292320792415849251569261233927 928Mean929104.820930116.17293153.79993264.1289331.8339341.82393518.72493613.1149379.3689387.7179391.1739400.98694158.35494266.4429436.49894411.15394537.90394619.02994773.54494868.45794914.81595014.334951 952Median95396.995954113.55895544.47395659.8939571.4959581.73295911.1649608.1509618.4209627.4779630.9419640.77196551.29096663.8369676.58296810.69796915.78497017.77597178.89097266.89597310.12997411.887975 976Z9773.840***9784.258***9793.008***980−4.416***981−3.874***982−3.558***9834.021***9845.023***985−0.686986−3.958***9870.177988 989Significance: * 10% level, ** 5% level, *** 1% level.990 991Liquidity Ratios. The operating cycle (96.995 for suppliers vs. 113.558 for control firms) and the992CCC (44.473 for suppliers vs. 59.893 for control firms) of the SCL suppliers are statistically993significantly lower (at the 1% level) compared to that of control firms in the same industry. The994current ratio (1.495 for suppliers vs. 1.732 for control firms) is also statistically significantly lower (at995the 1% level). Similar to the SCLs’ current ratio, in this case, the control firms’ current ratio is closer996to 2. Consequently, the same as for the SCLs is true here. The cash turnover (11.164 for suppliers vs.9978.150 for control firms) of suppliers, however, is statistically significantly higher (at the 1% level)998 999Logistics 2020, 4, 61000 100114 of 271002 1003compared with that of the control firms. The lower operating cycle and the CCC together with the1004higher cash turnover ratios imply that SCL suppliers are associated with significantly better liquidity1005ratios than control firms in the same industries.1006Activity Ratios. The Wilcoxon signed rank test indicates that the receivables turnover (8.420 for1007SCL suppliers vs. 7.477 for control firms) and the asset turnover (0.941 for suppliers vs. 0.771 for1008control firms) are statistically significantly higher (at the 1% level) than those of the control firms in1009the same industries. In contrast, the DIO (51.290 for suppliers vs. 63.836 for control firms) are1010statistically significantly lower (at the 1% level). This result shows that SCL supplier firms also1011possess significantly better activity ratios.1012Profitability Ratios. There is evidence (at the 1% level) that SCL suppliers’ EBIT margins (6.582 for1013SCL suppliers vs. 10.697 for control firms) are statistically significantly lower than the industry1014benchmarks. The COGS to sales (78.890 for suppliers vs. 66.895 for control firms) ratios are1015statistically significantly higher (at the 1% level) than that of the control firms in the same industry.1016As the ROCE and the SG&A to sales ratios do not present statistically significant results, SCL1017suppliers possess smaller profitability ratios.1018In summary, suppliers of SCLs are associated with significantly superior liquidity and activity1019but inferior profitability ratios.10204.4. Customers1021The majority of the SCL customer firms are US companies (47.6%). A smaller portion of1022companies (26.1%) are from Japan (9.5%), Great Britain (9.5%) and France (7.1%). The majority of the1023firms are at least large caps (76.1%), of which 7.1% are mega-caps. Most of the customers (61.9%) have1024more than 100,000 employees. This shows that this sample, compared to the SCL suppliers’ sample,1025is made up of bigger companies. The majority (40.5%) of these firms are active in the retail trade SIC1026division: 38.1% of the firms are in the manufacturing sector and 14.3% in the transportation and1027public utilities SIC division. The results for the customers of SCL firms are displayed in Table 6.1028Table 6. Comparison of ratios for customers of SCLs and control firms.1029Ratios1030 1031Liquidity1032 1033Operating cycle1034Cash conversion cycle1035Current ratio1036Cash turnover1037 1038Activity1039 1040Receivable turnover1041Asset turnover1042Days inventory outstanding1043 1044Profitability1045 1046EBIT margin1047Return on capital employed1048COGS to sales1049SG&A to sales1050 1051Sample1052Customers1053Control1054Customers1055Control1056Customers1057Control1058Customers1059Control1060Customers1061Control1062Customers1063Control1064Customers1065Control1066Customers1067Control1068Customers1069Control1070Customers1071Control1072Customers1073Control1074 1075N1076149107712951078147107912621080168108115211082167108315141084159108514551086167108714781088158108913361090168109115161092145109310401094164109513781096138109711071098 1099Mean110079.549110188.494110229.408110327.09611041.27711051.263110635.554110716.106110826.753110923.09011101.43311111.329111247.368111351.90811147.59711158.007111628.598111720.267111870.504111968.925112018.020112117.7511122 1123Median112466.288112588.494112625.360112727.06211281.09111291.210113014.830113111.538113211.56711339.48711341.18911351.392113638.229113748.95611385.93411395.574114018.441114120.118114273.134114374.717114419.010114514.6431146 1147Z11482.954***1149−1.18711501.2101151−4.068***1152−2.411***11530.85911542.931***11552.176**1156−0.3421157−2.102**11580.2311159 1160Logistics 2020, 4, 61161 116215 of 271163 1164Significance: * 10% level, ** 5% level, *** 1% level.1165 1166Liquidity Ratios. The operating cycle of SCL customers (66.288 for SCL customers vs. 88.494 for1167control firms) is statistically significantly lower (at the 1% level) compared to that of control firms in1168the same industry. In addition, the CCC as the current ratio presents misleading results if the mean1169and the median are compared. Further, neither ratio has statistically significant results. The cash1170turnover (14.830 for customers vs. 11.538 for control firms) of customers is statistically significantly1171higher (at the 1% level) compared to that of control firms in the same industry. The lower operating1172cycles, together with the higher cash turnover, imply that SCL customer firms are associated with1173significantly better liquidity ratios than control firms in the same industry.1174Activity Ratios. Although the receivables turnover for SCL customers (11.567 for SCL customers1175vs. 9.487 for control firms) is statistically significantly higher (at the 1% level), the DIO (38.229 for SCL1176customers vs. 48.956 for control firms) are statistically significantly lower (at the 1% level) than that1177of control firms in the same industry. These results show that SCL customer firms are associated with1178significantly better activity ratios.1179Profitability Ratios. The test statistic shows that SCL customers’ EBIT margins are statistically1180significantly lower (at the 5% level) than the industry benchmarks. In addition, although the SCL1181customers’ mean is smaller than that of the control samples’, the median is higher. This says1182something about the distribution of the data. This effect could be because the average distance of the1183firms’ ratios above the median is smaller, and/or the average distance of the customers’ ratios below1184the median is larger. Further, the COGS to sales ratio is statistically significantly higher (at the 5%1185level) than that of control firms in the same industry. For the COGS to sales ratio, the same1186phenomenon as for the EBIT margin appears, only inverted. Although the ROCE as the SG&A to1187sales ratio does not present significant results, SCL customers have smaller profitability ratios than1188their peers.1189Summing up, customers of SCL firms are associated with significantly superior liquidity and1190activity but inferior profitability ratios.11915. Discussion11925.1. SCL Networks as a Source of Liquidity1193By analyzing supplier–buyer relationships, Patatoukas [25] showed that suppliers, among1194others, can reduce their CCC through relationships with major customers. Lanier et al. [24], by1195investigating a concentrated three-firm SC, demonstrated that positive CCC results are distributed1196across the supply chain, with some indication that upstream members benefit more. Furthermore,1197Gosman and Kohlbeck [21] showed that suppliers of powerful retailers are associated with a lower1198CCC. These results are, to some extent, in line with our results in which suppliers and SCLs have1199statistically significantly superior liquidity ratios. Although SCL suppliers display a 15.4 day shorter1200CCC against peers, SCL firms for their part show a 12.1 day shorter CCC compared to their control

Showing the first 1,200 of 2307 lines. Download the file for the rest.