Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/ssrn/the-check-is-in-the-mail-can-disclosure-reduce-late-payments-to-8pkTYkcmUd
References
References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.
- Copyright
- Copyright © 2021 Elsevier Inc.
- eISSN
- ARCH-4059
- DOI
- 10.2139/ssrn.3775656
- Publisher site
- See Article on Publisher Site
Abstract
The check is in the mail: Can disclosure reduce late payments to suppliers? Elizabeth Chuk Associate Professor UC Irvine elizchuk@uci.edu Ben Lourie Assistant Professor UC Irvine blourie@uci.edu Il Sun Yoo PhD Student UC Irvine yoois@uci.edu January 18, 2021 A common corporate cash management strategy is to delay payments owed to suppliers. We examine whether buyers pay suppliers faster in response to a recent regulatory change in the United Kingdom that mandates the public disclosure of buyers’ payment practices. We find that after the regulatory change, UK firms subject to this regulation shortened their payment periods relative to several control samples of firms that were not subject to the regulation. In cross- sectional tests, we predict and find that this shortening of the payment period is attenuated for firms that are less able to bear the costs of paying suppliers faster. Specifically, we find that the reduction in the payment period is smaller for buyers that (i) have longer operating cycles, (ii) depend more heavily on trade credit as a source of external financing, and (iii) pay dividends. In supplemental tests using proprietary data, we find that buyers subject to this regulation reduced their trade credit that is overdue by 30 days or more. However, this reduction is partially offset by an increase in the trade credit that is overdue by less than 30 days. Our findings are important in light of the ongoing debate in other regimes (e.g., the United States) on whether to require additional disclosures of trade credit. Electronic copy available at: https://ssrn.com/abstract=3775656 1. INTRODUCTION Trade credit, which is credit extended by suppliers who allow delayed payment from buyers, is an important source of external financing for firms. For example, in the United States, about one quarter of corporate debt is trade credit, and the aggregate amount of trade credit is roughly three times as large as the aggregate amount of bank loans (Rajan and Zingales 1995; Barrot 2016; Ivashina, Iverson, and Smith 2016). Trade credit is especially crucial for firms that cannot raise capital through more traditional channels such as bank credit (Petersen and Rajan 1997; Giannetti, Burkart, and Ellingsen 2008). Thus, firms dedicate significant time and resources to manage working capital and trade credit (Long, Malitz, and Ravid 1993). It is a common cash management policy for buyers to strategically delay payments to suppliers in order to boost short-term capital, which can be used to earn returns from other investments before eventually paying off their obligations to suppliers (Monga and Trentmann 2020; Emery 1984). As a result, it is not uncommon for trade credit to become past due, thereby potentially leaving suppliers in a cash-strapped position. We examine whether firms in the United Kingdom alter their behavior by paying suppliers more quickly in response to a recent regulatory change that requires firms to publicly disclose their payment practices and timeliness in paying invoices. Section 3 of the Small Business, Enterprise and Employment Act 2015 (hereafter referred to as Section 3 for brevity) became effective for fiscal years beginning on or after April 6, 2017, requiring large UK firms to self-report their payment practices, policies, and performance. Importantly, Section 3 mandates disclosure of the average time taken to pay invoices and the percentage breakdown of invoices based on the number of days taken to pay (e.g., percentage Section 3 defines large firms as those that exceed at least two of the following three size thresholds on both of their last two balance sheet dates: (i) £36 million in annual sales, (ii) £18 in total assets, and (iii) 250 employees. Electronic copy available at: https://ssrn.com/abstract=3775656 of invoices paid within 30 days, percentage of invoices paid between 31 and 60 days, percentage of invoices not paid within agreed terms, and the longest standard payment period). Thus, stakeholders—such as suppliers—are now armed with additional relevant information to aid in decision-making, such as whether to trade with a particular buyer or how to structure contracts to protect themselves against late-paying buyers. Prior to Section 3, buyers typically did not voluntarily disclose their payment practices and performance. As such, it was not possible for a supplier to evaluate one buyer relative to another in terms of trade credit behavior. Some information about a particular buyer would be available if a supplier has conducted trade with the buyer in the past. However, even if a supplier had previously conducted trade with a particular buyer, the supplier only has knowledge about its own trading terms with the buyer but does not have access to information about the payment practices between the buyer and its other suppliers. Thus, Section 3 provides incrementally useful information to suppliers about buyers, regardless of the existence of—or lack thereof—a past trading relationship. In recent years, the evaluation of a buyer’s payment practices and performance extends beyond the interests of suppliers. In a recent survey, institutional investors identify supply chain issues such as trade credit behavior as one of their top concerns in consideration of a firm’s environmental, social, and governance (ESG) performance (Human 2021). A variety of stakeholders—including suppliers, investors, customers, and employees—are increasingly interested in firms’ approach to ESG issues. As a result, many firms are actively attempting to improve their engagement with the community over ESG issues (Human 2021, Kishan and Marsh 2021). The Section 3 disclosures can allow on-time paying buyers to cultivate a favorable Electronic copy available at: https://ssrn.com/abstract=3775656 reputation among suppliers and other stakeholders, whereas late-paying buyers develop poor reputations when their payment practices are publicly exposed under Section 3. Late payments to suppliers has increasingly become common practice. For example, the largest 1,000 public firms in the US increased their payment delays from 40 days in 2008 to 56.7 days in 2017 (Shumsky and Trentmann 2018). In Australia, a survey found that 14% of businesses are owed more than AUD 100,000 in late payments, and about half of small businesses reported receiving late payments for 40% of their invoices (Waters 2017). In the UK, the construction giant Carillion typically paid its subcontractors with a 120-day delay. At the time of its bankruptcy filing in 2018, it was revealed that Carillion owed GBP 2 billion to 30,000 suppliers (Torrance 2019). Hence, in recent years, the pattern of delaying payment to suppliers has become a trend with global impact. Late payments impose financial burdens on suppliers by restricting cash flows to the suppliers, thereby hampering investment and economic growth (Department of Business, Energy & Industrial Strategy 2016; Barrot and Nanda 2020). Firms with cash constraints cannot invest in positive-NPV projects. This leads to impaired growth and inefficient resource allocation in an economy. The topic of late payments to suppliers has received increased attention during the COVID-19 pandemic, with many firms unable to pay suppliers on time due to business closures and reductions in revenues. Late payments to suppliers have exacerbated supply chain disruptions and working capital shortages during the pandemic, threatening to “trigger a tidal wave of failures” (Caniato, Moretto, and Rice 2020). Yet, in another pandemic-inspired twist, some large firms have voluntarily adopted payment policies that defy conventional cash management strategies: Some firms are actually Electronic copy available at: https://ssrn.com/abstract=3775656 paying their suppliers early (Monga and Trentmann 2020). For example, Micron Technology Inc. has been paying some suppliers within 10 days instead of 50 days, which was the firm’s average delay before the COVID-19 pandemic. In addition, Lockheed Martin Corporation has been paying some invoices in half the time it previously took (Monga and Trentmann 2020). Both companies depend on complex global networks for parts and services, and they are injecting money to their suppliers due to concerns about disruptions in their supply chain if their suppliers fail (Monga and Trentmann 2020). Regardless of the incentives that might motivate these changes, it is clear that the length of the payment period is an impactful firm policy. Thus, firms devote substantial resources to manage these policies effectively. Our paper examines whether firms reduce the length of their payment period to pay suppliers in response to a regulatory change that mandates public disclosure of firms’ payment practices and policies. Using a difference-in-differences design, we find that UK firms subject to the disclosure regulation reduced the length of their payment period after the regulation 8.8% more than a control sample of UK firms not subject to the regulation. Furthermore, we document that this reduction in the length of the payment period is larger for firms that are financially capable of bearing the costs of shortening their payment periods, but this reduction is smaller for firms that are less able to bear such costs. Specifically, in cross-sectional tests, we find that the reduction in the length of the payment period is smaller for (i) firms with longer operating cycles, (ii) firms that rely more heavily on trade credit as a source of financing, and (iii) firms that pay dividends. In supplemental tests, we choose an alternative control group of firms in Germany and France. These firms are not subject to the UK disclosure regulation, but they are large enough to satisfy the firm size thresholds specified by the UK regulation. Therefore, they have similar firm Electronic copy available at: https://ssrn.com/abstract=3775656 size characteristics as the treatment sample. We find that large UK firms reduce the length of their payment periods after the regulation compared to a control group of large firms from Germany and France. These results mitigate concerns that the results of our main test (which uses small UK firms as the control group) are driven by differences in firm size characteristics across the treatment and control groups. In additional tests using proprietary data, we document that after the disclosure regulation, UK firms subject to the regulation reduced the amount of payables past due beyond 30 days and increased the amount of payables past due within 30 days. Together, these results suggest that the regulation was effective in reducing payables overdue by a longer time period (i.e., past due by more than 30 days), but this reduction is partially offset by an increase in payables overdue by a shorter time period (i.e., past due by 1–30 days). The economic magnitude of the regulation is nontrivial. Compared to a control sample of UK firms not affected by the regulation, the firms subject to Section 3 reduced their percentage of trade credit past due by more than 30 days by 9.7%, but they increased the percentage of trade credit past due within 30 days by 9.1%. It appears that after the disclosure regulation, UK firms shortened the time period for which payables are past due, but the regulation does not completely curb tardiness in payment. Our study makes several contributions. First, our study offers insights based on a regulatory change in the UK, but these findings can be generalized to other countries because the pattern of delayed payments to suppliers is an increasing global trend (Waters 2017; Shumsky and Trentmann 2018; Torrance 2019). The regulatory change we examine represents a unique setting in which only disclosure was mandated without any changes in measurement or recognition rules in the financial reports. Nor were there any mandated requirements for firms to Electronic copy available at: https://ssrn.com/abstract=3775656 alter their payment practices and policies. To our knowledge, no other country has implemented a similar regulation that requires the disclosure of payment practices and policies. To the extent that regulatory agencies in other countries are considering similar regulatory changes to address the problem of increasingly tardy payments to suppliers, our study shows that mandating the disclosure of payment practices and policies has created incentives for firms to reduce the length of their payment periods to pay suppliers. We remain agnostic about whether a reduction in the payment period is a desirable outcome. Nonetheless, we emphasize that such an outcome can have impactful ripple effects throughout an economy, since late payments to suppliers can impair growth and lead to inefficient resource allocation. Our second contribution relates to the recent increase in the use of reverse factoring. In a reverse factoring transaction, a buyer reaches out to a financial institution to become involved as an intermediary to pay its suppliers faster in exchange for a discount. This reduces the receivables for the supplier, who pays a fee to the financial institution. This fee is typically based on the creditworthiness of the buyer. At a later date, the buyer then pays the invoices to the financial institution. As a result, the supplier receives cash payment earlier, while the buyer enjoys more time to pay the invoices. The rising popularity of reverse factoring can lead to a reduction in the informativeness of accounts payable because the payment terms agreed upon with the financial institution are typically different from the terms of traditional accounts payable owed directly to suppliers. According to a recent Moody’s report in 2019, almost half of companies on the payment side of supply chain financing transactions use some form of reverse factoring, and another 37% report that they are considering it (Moody’s, September 19, 2019). Currently, the SEC and FASB are considering whether to require increased disclosures related to reverse factoring and other Electronic copy available at: https://ssrn.com/abstract=3775656 aspects of trade credit relationships (Steinberg 2020). The results of our study, based on the effects of a disclosure requirement regarding payment practices in the UK, can inform the SEC and FASB about how to design the potential regulation that may be enacted in the US. Third, we contribute to the literature on the real effects of accounting regulations, which has documented that firms are willing to undertake actions with cash flow consequences in order to report favorable financial numbers (Imhoff and Thomas 1988; Graham, Hanlon, and Shevlin 2011; Chuk 2013). Our research setting offers a distinctive feature in contrast to prior studies, which tend to assume that investors are the key stakeholder group that incentivizes managers to alter firm transactions in order to report favorable financial numbers. However, in our setting, suppliers comprise one key stakeholder group that incentivizes managers to report favorable financial numbers. Financial regulators have long expressed the view that investors are not the only users of financial reporting (e.g., the FASB mission statement). Our study helps us understand suppliers as a group of financial reporting users and understand the incentives this group creates for managers. In our setting, investors as a stakeholder group would likely encourage managers to pay suppliers more slowly, given that conventional cash management strategies suggest that buyers should pay later. Meanwhile, managers exploit investment opportunities regarding the cash on hand, as long as late payments do not jeopardize their business relationships with suppliers. In contrast, suppliers as a stakeholder group likely encourage managers to pay suppliers faster in order to maintain relationships with suppliers. The real effect that we document is a shortening of the payment period, so our results suggest that managers’ response to Section 3 was, on average, driven by the desire to report favorably to suppliers as a stakeholder group rather than to their investors. Therefore, we contribute to the real effects literature by documenting that managers Electronic copy available at: https://ssrn.com/abstract=3775656 are willing to undertake actions that have cash flow consequences in order to report favorable financial numbers to suppliers as a stakeholder group, even when the preferences of suppliers contradict those of the stakeholder group of investors. Last, we contribute to the literature on trade credit by documenting that the mandated disclosure of payment practices and policies can compel firms to alter those practices and policies. Prior studies have documented various determinants of the amount and extent of the trade credit used by firms (Petersen and Rajan 1997; Fisman and Love 2003; Giannetti, Burkart, and Ellingsen 2008). In the context of these documented determinants, firms weigh the costs against the benefits of delaying payment when deciding when and how much to pay their suppliers. Our results suggest that mandated public disclosure imposes an additional cost for delaying payment to suppliers, and this additional cost leads managers to pay suppliers more quickly after the regulatory change. The rest of the paper is organized as follows. In Section 2, we discuss the institutional background of the regulation and develop our hypotheses. Section 3 describes our data and empirical design. Section 4 reports our results, and we provide additional analysis in Section 5. Section 6 summarizes and concludes. 2. INSTITUTIONAL BACKGROUND AND HYPOTHESIS DEVELOPMENT Regulation in the UK Section 3 was enacted in the UK to (i) increase the transparency and public scrutiny of large businesses’ payment practices and performance and (ii) give small business suppliers better information so they can make more informed decisions about who they trade with, negotiate fairer terms, and challenge late payments (Department of Business, Energy & Industrial Strategy 2016). Electronic copy available at: https://ssrn.com/abstract=3775656 Section 3 became effective for fiscal years beginning on or after April 6, 2017, and it applies to large companies and large limited liability partnerships (LLPs) in the UK that exceed at least two of the following three size thresholds on both of their last two balance sheet dates: (i) GBP 36 million in annual sales, (ii) GBP 18 million in total assets, and (iii) 250 employees. Under Section 3, qualifying UK firms must disclose information about their payment practices, policies, and performance twice per fiscal year by publishing their disclosures on a government website, which is open to the public. Of particular importance, firms must disclose (i) the average time they take to pay invoices (from the date of receipt of the invoice) and (ii) the percentage breakdown of invoices paid during the reporting period that were paid within 30 days, between 31 and 60 days, and over 60 days. Appendix A describes the required disclosures mandated by the regulatory change. Before Section 3, thousands of businesses experienced severe administrative and financial burdens every year because they did not receive payments on time (Department of Business, Energy & Industrial Strategy 2019). Despite initial proposals to require these disclosures as part of the annual financial reporting process, Section 3 ultimately requires its disclosures to be reported separately from the annual financial reports due to concerns that annual reports are not prepared on a timely basis. Section 3 requires firms to report their disclosures every six months on a public government website within 30 days of the end of the reporting period. Failure to report, or reporting false or misleading information, is a criminal offense. A company director is required to sign off on this report to ensure the accuracy of the information. However, these disclosures are not subject to audit. The website is https://check-payment-practices.service.gov.uk/search. Electronic copy available at: https://ssrn.com/abstract=3775656 Potential managerial responses to the regulatory change One of the main determinants of the level of accounts payable reported on a firm’s balance sheet is the amount of credit the firm’s suppliers offer (Petersen and Rajan 1997). Suppliers have advantages in lending compared to traditional financial institutions (Biais and Gollier 1997) because suppliers have access to private information that banks do not have (Mian and Smith 1992). In particular, suppliers have an information advantage over bank lenders when investigating the creditworthiness of their clients (Petersen and Rajan 1997). Furthermore, suppliers can threaten to cut off future supplies in the event of borrower actions that reduce the likelihood of repayment (Petersen and Rajan 1997). For these reasons, suppliers can offer trade credit or withhold trade credit from a buyer based on the buyer’s creditworthiness. In a similar spirit, one of the stated objectives of Section 3 is to give small business suppliers better information in order to make informed decisions about who to trade with, negotiate fairer terms, and challenge late payments. Before the regulation, buyers chose whether and when to pay their invoices by weighing the benefits of delaying payment (in the form of higher short-term capital that could be used for other investments) against the costs of delaying payment (in the form of potentially damaging relationships with suppliers). Section 3 imposes a new cost for delaying payment to suppliers. Under the new regulatory regime, suppliers can make more informed decisions about whether and how to work with buyers that tend to pay late. In response to this regulatory change, we expect that firms pay their suppliers faster in order to avoid unfavorable reporting of their payment practices, policies, and performance: H1: After the regulatory change, firms subject to the Section 3 disclosure requirements pay their suppliers faster than a control group of firms not subject to the regulatory change. Electronic copy available at: https://ssrn.com/abstract=3775656 Cross-Sectional Predictions We expect this change in behavior to be stronger for firms that are more able to bear the costs of paying faster (i.e, the cost of reducing short-term capital), and we expect the change to be smaller for firms that are less able to bear such costs. In other words, we expect cross- sectional variation in firms’ responses to the regulatory change. We identify several firm characteristics that impact a firm’s ability to bear the costs of shortening the payment period. The first firm characteristic we identify is the length of a firm’s operating cycle, which likely impacts the firm’s ability to pay its payables faster. The length of a firm’s operating cycle is defined as the period of time from the moment the firm purchases raw materials to the moment cash is received from customers for a sale. Firms with shorter operating cycles can more quickly convert raw materials to cash from sales; therefore, they are more likely able to bear the costs of paying their payables faster after Section 3. In contrast, firms with long operating cycles are likely less able to reduce short-term capital by paying suppliers more quickly. H2A: After Section 3, the increase in firms’ speed of payment to their suppliers is less pronounced for firms that have longer operating cycles. A firm’s dependence on trade credit is likely another cross-sectional determinant of the strength of the impact of the regulation on changes in managerial behavior. Firms that are less dependent on trade credit have access to additional alternative sources of financing, such as borrowing from financial institutions (Fisman and Love 2003). Compared to firms that heavily depend on trade credit as a source of financing, firms that do not depend on trade credit can more easily find substitute sources of capital to make up for reductions in short-term capital resulting Electronic copy available at: https://ssrn.com/abstract=3775656 from the decision to pay suppliers more quickly. That is, that costs of shortening the payment period to pay suppliers are higher for firms that depend more heavily on trade credit. H2B: The increased speed of payment to suppliers resulting from Section 3 is less pronounced for firms that depend more heavily on trade credit. Finally, a firm’s dividend policy is likely to impact the firm’s response to Section 3. Dividend policy tends to be sticky over time because it is difficult for firms to adjust their dividend payouts from year to year (e.g., Lintner 1956). Given that dividend payouts require cash distributions that reduce internally generated funds, firms that pay dividends have less flexibility to shorten their payment periods to pay suppliers. In contrast, after Section 3, firms that do not pay dividends have greater flexibility to bear the costs of paying suppliers faster. H2C: The increased speed of payment to suppliers resulting from Section 3 is less pronounced for firms that pay dividends. There are reasons to believe that we might not observe the predicted changes in firm behavior in response to Section 3. First, it may be too burdensome on operations for firms to pay suppliers more quickly, so it may not be feasible for firms to shorten their payment period in response to Section 3, even if they desire to do so. Second, suppliers have access to private information about buyers (Petersen and Rajan 1997), and the disclosures required under Section 3 might not provide incrementally useful information to suppliers. Under this scenario, the buyers believe that before Section 3, sellers already had access to the information that buyers are required to disclose under Section 3. If buyers believe that these disclosures do not add to the information available to suppliers, then buyers would have no reason to alter their behavior after Section 3. Electronic copy available at: https://ssrn.com/abstract=3775656 Third, these disclosures might not provide strong enough incentives for firms to alter their behavior, because the regulatory change mandates disclosure but does not explicitly mandate any actual reduction in payment periods. Nor does the regulation impose any sanctions for late payments to suppliers; sanctions are imposed only for failure to disclose. Last, firms could anticipate Section 3 because it was initially intended to be implemented in April 2016, but it was delayed until April 2017. Therefore, firms could have altered their behavior before the effective date that we observe, thereby reducing the power of our empirical tests. For these reasons, it is possible that we will not detect the predicted managerial responses. Thus, whether firms respond to Section 3 by paying suppliers more quickly remains an empirical question. 3. DATA AND EMPIRICAL DESIGN Constructing a sample of UK firms Section 3 became effective for fiscal years beginning on or after April 6, 2017. We define the first year that Section 3 is effective as year t = 0. Our pre-period includes three years before Section 3 (years t = −3, −2, and −1). Our post-period includes the first two years for which Section 3 is effective (years t = 0 and 1). For a December 31 firm, the pre-period includes years 2015, 2016, and 2017; the post-period includes years 2018 and 2019. We obtain a sample of UK firms using Compustat Global. Table 1 outlines our sample selection process. We begin with the 7,418 firm–years observed for UK firms on Compustat Global during the event years t = −3, −2, −1, 0, and 1. Then, we delete 3,711 firm–years that have missing data for our variables of interest, and we delete 219 firm–years for firms that do not have at least one year in the pre-period and at least one year in the post-period. We further Electronic copy available at: https://ssrn.com/abstract=3775656 remove 311 firm–years for which the value of the cost of goods sold was below GBP 1 million in order to avoid the small denominator problem when computing our dependent variable. Our final sample consists of 3,177 firm–year observations, of which 1,983 (1,194) observations are above (below) the firm size thresholds as defined by the Section 3 regulation. Measuring managerial behavior in response to regulatory change To measure whether UK firms reduce their payment periods in response to Section 3, we measure our dependent variable as the natural log of the days payable outstanding. Days payable outstanding is computed as the average accounts payable times 365, divided by the cost of goods sold. For several reasons, we use days payable outstanding computed based on financial statement data from Compustat as opposed to the payment period disclosures self-reported by firms under Section 3. First, a long line of literature uses days payable outstanding based on financial statement data to measure the length of the payment period in transactions involving trade credit (e.g., Long et al. 1993, Dechow 1994, CFA Institute 2011, Wahlen et al. 2014). Thus, using the same proxy as the one used in prior literature facilitates comparison between our inferences and the findings established in prior studies. Second, as a practical matter, the Section 3 payment period disclosures are not available for the pre-period, so it would not be possible to conduct pre-post tests on managerial behavior. Also, the Section 3 payment period disclosures are not available for firms that are not subject to Section 3 (i.e., the disclosures are not available for any control firms), thus precluding any comparisons between the treatment and control groups. Thus, we use the financial statement–based measure of days payable outstanding as our proxy for firm policies regarding their payments to suppliers. Electronic copy available at: https://ssrn.com/abstract=3775656 Nonetheless, we compare our financial statement–based measure of days payable outstanding to the self-reported disclosures under Section 3. Given that the Section 3 self- reported disclosures are available only for large UK firms in the post-period, we can perform the comparison only for this subsample. For all large UK firms in our post-period, we hand-collect the payment period disclosures required under Section 3 directly from the firms’ self-reported filings. We then compute the difference between the Section 3 disclosure of the payment period and the financial statement–based measure of days payable outstanding. In untabulated analyses, we find that for 64% of the observations, the days payable outstanding is larger than the payment periods self-reported under Section 3. On average, the median (mean) difference is 12.2 (20.6) days. In the subsample for which the days payable outstanding is larger (smaller) than the payment period self-reported under Section 3, the median difference between the two measures is 30 (9.3) days. By construction, there are several reasons to expect that these two measures will not be the same. First, the self-reported disclosures under Section 3 are based on nonconsolidated numbers, while the Compustat data are based on consolidated numbers. Second, the self-reported disclosures under Section 3 are required every six months, while the Compustat-based days payable outstanding variable is computed annually. Third, for the disclosures under Section 3, if a transaction is not sufficiently linked to the UK, firms are not required to include that transaction in its disclosure, while Compustat does not make this distinction. Fourth, the Section 3 disclosures allow firms to selectively eliminate intercompany invoices from their reported statistics, while Compustat does not make this distinction. In summary, these differences in measurement methodologies create differences between the payment periods disclosed under Section 3 and the financial statement–based days payable Electronic copy available at: https://ssrn.com/abstract=3775656 outstanding. Although these two measures do not have exactly the same value, we find that the Pearson (Spearman) correlation coefficient between them is 37.6% (31.6%), which is significantly different from zero at p-value < .01. Empirical model To test our H1, we estimate the following model: 𝐷𝑎𝑦𝑠𝑃𝑎𝑦𝑎𝑏𝑙𝐿𝑛𝑒 = 𝛼 + 𝛼 𝑠𝑡𝑃𝑜 ∗ 𝐿𝑎𝑟𝑔𝑒 + 𝛼 𝑟𝑙𝑜𝑠𝑡𝑛𝐶𝑜 + 𝐹𝑖𝑟𝑚 + 𝑖𝑡 0 1 𝑡 𝑖 𝑘 𝑖𝑡 𝑌𝑒𝑎𝑟 + 𝜀 , (Equation 1) 𝑖𝑡 where Large is a dummy variable that equals 1 for firms that exceed the firm size thresholds defined by Section 3 and equals 0 for small firms that are exempt from Section 3. The dummy variable Post is equal to 1 for years in which Section 3 is effective and equal to 0 for years before Section 3. Our main variable of interest is Post × Large, which we predict to be negative under H1. We cluster standard errors by firm, and we include firm fixed effects and year fixed effects. Accordingly, we do not report results for the main effects for Large and Post. In Equation 1, we control for other determinants of days payable outstanding. These determinants are largely related to either the buyer’s need for a longer payment period or the buyer’s ability to negotiate a longer payment period. We control for the length of the operating cycle (LnOperatingCycle) because firms with longer operating cycles are likely to need more time to settle their liabilities. We control for leverage (Leverage) because firms with higher leverage are likely to have more limited access to debt financing and are therefore more likely to rely on the flexibility of trade credit. We control for firm size, measured as total assets (LnAT), sales (LnSale), and the number of employees (LnEmployee), because these are the three size measures used by Section 3 to define a large firm. Electronic copy available at: https://ssrn.com/abstract=3775656 𝐹𝐸 𝐹𝐸 In general, we expect that large buyers possess more purchasing power to negotiate with suppliers for longer payment periods. Thus, we expect a positive relation between firm size and days payable outstanding. However, we note that for one of these size measures, LnSale, sales can also be negatively related to the dependent variable because firms that have higher sales tend to also have a higher cost of goods sold. A higher cost of goods sold would mechanically translate to a lower days payable outstanding, since the cost of goods sold appears in the denominator of days payable outstanding. Next, we control for firm performance (ROA), since better performing buyers are more likely to negotiate longer payment periods with suppliers. We control for firm growth (Growth) because growth firms tend to be less established, so they are likely less able to negotiate longer payment periods with suppliers. Finally, we control for the amount of cash holdings (Cash), as buyers with larger cash reserves are less likely to require a delay in payment to suppliers. All variables are described in detail in Appendix B. To test H2A, H2B, and H2C, we estimate the following model, using a triple interaction to measure the cross-sectional variation in the impact of Section 3: 𝐷𝑎𝑦𝑠𝑃𝑎𝑦𝑎𝑏𝑙𝐿𝑛𝑒 = 𝛼 + 𝛼 𝑠𝑡𝑃𝑜 ∗ 𝐿𝑎𝑟𝑔𝑒 + 𝛼 𝑎𝑙𝑛𝑉𝑖𝑆𝑒𝑜𝐶𝑎𝑟𝑠𝑠𝑐𝑡𝑟𝑜 𝑖𝑡 0 1 𝑡 𝑖 2 𝑖𝑡 +𝛼 𝑠𝑡𝑃𝑜 ∗ 𝑎𝑙𝑛𝑖𝑉𝑆𝑒𝑜𝐶𝑎𝑟𝑠𝑠𝑐𝑡𝑟𝑜 + 𝛼 𝐿𝑎𝑟𝑔𝑒 ∗ 𝑎𝑙𝑛𝑉𝑖𝑆𝑒𝑜𝐶𝑎𝑟𝑠𝑠𝑐𝑡𝑟𝑜 3 𝑡 𝑖𝑡 4 𝑖 𝑖𝑡 +𝛼 𝑠𝑡𝑃𝑜 ∗ 𝐿𝑎𝑟𝑔𝑒 ∗ 𝑎𝑙𝑛𝑖𝑉𝑆𝑒𝑜𝐶𝑎𝑟𝑠𝑠𝑐𝑡𝑟𝑜 + ∑ 𝛼 𝑟𝑙𝑜𝑠𝑡𝑛𝐶𝑜 + 𝐹𝑖𝑟𝑚 + 5 𝑡 𝑖 𝑖𝑡 𝑘 𝑖𝑡 𝑌𝑒𝑎𝑟 + 𝜀 , (Equation 2) 𝑖𝑡 All variables are measured the same as in Equation 1. We estimate Equation 2 three times separately, with each estimation substituting one of the three cross-sectional partitioning variables for CrossSectionalVar. To test H2A, we use LnOperatingCycle as CrossSectionalVar. To test H2B, we use TradeCreditDependence as CrossSectionalVar. To test H2C, we use DividendPayer as Electronic copy available at: https://ssrn.com/abstract=3775656 𝐹𝐸 𝐹𝐸 CrossSectionalVar. The variable of interest is the triple interaction term Post × Large × CrossSectionalVar, which we predict to be positive in H2A, H2B, and H2C because we expect these cross-sectional determinants to mitigate the impact of Section 3 on the reduction of payment periods. 4. EMPIRICAL RESULTS Table 2 Panel A provides descriptive statistics for UK firms below the size thresholds (i.e., control group), and UK firms above the size thresholds (i.e., the treatment group) are shown in Panel B. As expected, the UK firms in the treatment group are larger than the UK firms in the control group along all three size dimensions: total assets (GBP 2,258.6 million for treatment firms versus GBP 49.82 million for control firms), sales (GBP 1,803.1 million for treatment firms versus GBP 27.97 million for control firms), and the number of employees (13.71 for treatment firms versus 0.18 for control firms). The UK treatment firms, on average, have a shorter days payable outstanding than the UK control firms (75.10 days for treatment firms versus 89.28 days for control firms). Table 3 presents the difference-in-differences results for Equation 1. Consistent with H1, the interaction Post × Large is significantly negative with a p-value < .01. We follow the procedure described in Ford (2018) to interpret the coefficient when using a log-transformed dependent variable. Thus, we exponentiate the coefficient, subtract 1 from this number, and then multiply the result by 100: (exp(0.084) – 1) × 100 = 8.8%. Our results suggest that after Section 3, treatment firms reduced the length of their payment period 8.8% more than the control firms not subject to Section 3. For the control variables, we find that the length of the payment period is positively related to the length of the operating cycle, such that firms with longer operating cycles are less Electronic copy available at: https://ssrn.com/abstract=3775656 able to settle payables quickly. Of the three size thresholds defined by Section 3, we find that the length of the payment period is positively associated with total assets and the number of employees, and it is negatively associated with sales. This is likely due to the fact that sales are positively related to the cost of goods sold, which appears in the denominator of the dependent variable. This means that a higher value of the cost of goods sold mechanically reduces the value of the dependent variable. Table 4 presents the results of the cross-sectional tests shown in Equation 2. In Table 4 Panel A, Post × Large is significantly negative, with a p-value < .01, and the triple interaction Post × Large × LnOperatingCycle is significantly positive, with a p-value < .01. Together, these results suggest that UK treatment firms reduce their payment periods after Section 3 compared to UK control firms, but this effect is mitigated for firms that have longer operating cycles, since firms with longer operating cycles are likely to have less flexibility to settle payables faster. In Table 4 Panel B, Post × Large is significantly negative, with a p-value < .01, and the triple interaction Post × Large × TradeCreditDependence is significantly positive, with a p- value < .01. Together, these results suggest that UK treatment firms reduced their payment periods after Section 3 compared to UK control firms, but this effect is mitigated for firms that depend more heavily on trade credit as a source of financing, since these firms have fewer financing options for settling payables faster. In Table 4 Panel C, Post × Large is significantly negative, with a p-value < .01, and the triple interaction Post × Large × DividendPayer is significantly positive, with a p-value < .01. Together, these results suggest that UK treatment firms reduced their payment periods after Section 3 compared to UK control firms, but this effect is mitigated for firms that pay dividends, Electronic copy available at: https://ssrn.com/abstract=3775656 since these firms have more cash committed to paying dividends and thus have less flexibility to settle payables faster. Together, our results in Tables 3 and 4 support H1 and H2, respectively. Robustness tests In our main difference-in-differences test shown in Table 3, we assign large UK firms to the treatment group and small UK firms to the control group, using the size thresholds defined by the Section 3 regulation. To mitigate potential concerns that the observed results are driven by characteristics related to differences in firm size (rather than being driven by the impact of the Section 3 regulation), we perform robustness tests using an alternative control sample of firms that have the both of the following features: (i) buyers that satisfy the definition of large firms under the size thresholds specified by Section 3 and (ii) buyers that are not subject to the Section 3 regulation. We construct such a sample using large firms from Germany and France. We choose these two countries due to the institutional similarities among these countries and the UK (Antoniou, Guney, and Paudyal 2006). The sample of large firms from Germany and France satisfies the Section 3 size thresholds, but these firms are not subject to this regulation because they are outside the UK. Table 2 Panel C (D) reports descriptive statistics for the firms in Germany and France that are above (below) the size thresholds specified in Section 3. After comparing the means in Panel D (i.e., large firms in Germany and France) to the means in Panel B (i.e., large firms in the UK), we note that days payable outstanding is longer for large firms from Germany and France than for large UK firms (82.85 days versus 75.10 days, respectively). Additionally, large firms in Germany and France tend to have longer operating cycles than large UK firms (181.18 days versus 137.80 days). Along all three dimensions of firm size specified by Section 3, large firms from Germany and France are larger than the large UK firms (GBP 5,800.6 million versus GBP Electronic copy available at: https://ssrn.com/abstract=3775656 2,258.6 million for total assets, GBP 4,240.4 million versus GBP 1,803.1 for sales, and 19,110 versus 10,480 for the number of employees). Table 5 reports the results of estimating Equation 1 using large firms from Germany and France as an alternative control sample. In Column 1 of Table 5, we retain the same treatment group as in the original estimation of Equation 1 (namely, large UK firms), but we remove small UK firms from the control sample and substitute large firms from Germany and France as the control group. We define a dummy variable UK, which equals 1 for UK firms and equals 0 for firms in Germany and France. The interaction term Post × UK is significantly negative, with a p- value <.05. This is consistent with the idea that large UK firms subject to Section 3 reduce their payment periods after Section 3 by 3.98% more than a control sample of large firms not subject to Section 3. In Column 2 of Table 5, we expand the control sample by including (i) large firms from Germany and France, (ii) small firms from Germany and France, and (iii) small UK firms. Recall that our dummy variable Large equals 1 for large firms in all countries (i.e., UK, Germany, and France) and equals 0 for small firms in all countries. The interaction term Post × Large × UK, which measures the incremental effect of firms that are large and located in the UK in the post- period, is significantly negative, with a p-value < .01. This suggests that large UK firms reduced their payment periods after Section 3 compared to a control group composed of large firms in Germany and France and small firms from Germany, France, and the UK. Together, the results shown in Table 5 mitigate concerns that our main results in Tables 3 and 4 are driven by differences in firm characteristics related purely to firm size; rather, they are driven by a managerial response to Section 3. Following Ford (2018), we compute the difference-in-differences effect as (exp(0.039) – 1) × 100 = 3.98%, where −0.039 is the regression coefficient on POST × UK. Electronic copy available at: https://ssrn.com/abstract=3775656 5. ADDITIONAL ANALYSIS To supplement our tests using Compustat data, we also obtain data from a proprietary dataset compiled by Credit Risk Monitor, which is a financial risk analysis and news service for credit, supply chain, and financial professionals. Credit Risk Monitor covers 56,000 global public companies, with a total of USD 63.8 trillion in corporate revenue. We obtain data to compute the (i) percentage of trade credit owed by the buyer that is past due between 1 and 30 days and (ii) the percentage of trade credit owed by the buyer that is past due by 31 days or more. We tabulate these percentages at the bottom of Table 2 for our various subsamples. Table 2 Panels A and B show that large UK firms, compared to small UK firms, have a lower percentage of trade credit past due between 1 and 30 days (16% for large UK firms versus 24% for small UK firms). However, the large firms also have a higher percentage of trade credit past due by 31 days or more (10% for the large firms versus 5% for the small firms). Comparing Panels C and D, we note that the percentage of trade credit past due between 1 and 30 days is similar across both large and small firms in Germany and France (16% for both groups), while large firms from Germany and France have a higher percentage of trade credit past due by 31 days or more compared to small firms from Germany and France (7% for the large firms versus 4% for the small firms). In Table 6, we report the results of tests using these two measures of trade credit as dependent variables. In Panel A, we use a specification similar to our main tests reported in Table 3. We assign large UK firms to the treatment group and small UK firms to the control group. We find that Post × Large is marginally negative, with a p-value <.10 at explaining the percentage of trade credit past due by 31 days or more; the coefficient implies that after Section Electronic copy available at: https://ssrn.com/abstract=3775656 3, the percentage of trade credit past due by 31 days or more decreased by 9.7% more for large UK firms relative to small UK firms. However, we also find that Post × Large is significantly positive, with a p-value < .05. This explains the percentage of trade credit past due between 1 and 30 days. The magnitude of the coefficient implies that after Section 3, the percentage of trade credit past due between 1 and 30 days increased by 9.1% more for large UK firms relative to small UK firms. Thus, the impact of Section 3 is economically significant, and it triggered changes in the percentage of trade credit past due by almost 10% more for buyers affected by the regulation relative to buyers not subject to the regulatory change. Table 6 Panel B shows results similar to Panel A after redefining the control group by removing small UK firms from the analysis and substituting large firms from Germany and France as the control group. The interaction term Post × UK is marginally negative, with a p- value < .10. This explains the percentage of trade credit past due by 31 days or more. The coefficient implies that after Section 3, the percentage of trade credit past due by 31 days or more decreased by 3.8% more for large UK firms relative to large firms from Germany and France. However, we also find that Post × Large is marginally positive, with a p-value < .10 at explaining the percentage of trade credit past due between 1 and 30 days. The magnitude of the coefficient implies that after Section 3, the percentage of trade credit past due between 1 and 30 days increased by 2.7% more for large firms in the UK relative to large firms in Germany and France. In Table 6 Panel C, we expand the control sample by including (i) large firms from Germany and France, (ii) small firms from Germany and France, and (iii) small UK firms. The triple interaction term Post × Large × UK is significantly negative, with a p-value < .05 when Electronic copy available at: https://ssrn.com/abstract=3775656 explaining the percentage of trade credit past due by 31 days or more. The coefficient implies that after Section 3, the percentage of trade credit past due by 31 days or more decreased by 14.8% more for large UK firms relative to the control sample consisting of large firms from Germany and France as well as small firms from the UK, Germany, and France. However, we also find that Post × Large is marginally positive, with a p-value < .10 when explaining the percentage of trade credit past due between 1 and 30 days. The magnitude of the coefficient implies that after Section 3, the percentage of trade credit past due between 1 and 30 days increased by 7.6% more for large UK firms relative to the control sample consisting of large firms from Germany and France as well as small firms from the UK, Germany, and France. Again, the magnitude of the economic impact of Section 3 is nontrivial, with incremental changes in past due trade credit of 7.6% to 14.8% for buyers in the treatment group relative to buyers in the control group. Together, these results suggest that the regulation was effective at reducing payables overdue by a longer time period (i.e., overdue by more than 30 days), but this reduction was partially offset by an increase in payables overdue by a shorter time period (i.e., overdue between 1 and 30 days). It appears that after the disclosure regulation, UK firms shortened the time period in which payables are past due, but the regulation does not completely curb tardiness in payment. 6. CONCLUSION The buyer’s decision on when to pay suppliers bears economically important consequences for both the buyer and the supplier. For the buyer, a large portion of corporate debt consists of trade credit, so buyers spend significant firm resources to devise cash management strategies to delay payments to suppliers. For the seller, delays in receiving payment can impose Electronic copy available at: https://ssrn.com/abstract=3775656 a sustained strain on cash flows, thus potentially threatening the financial health and viability of suppliers. Moreover, a persistent cash strain imposed on suppliers can hamper economic growth by generating ripple effects across interconnected firms, potentially leading to a stagnant economy. When viewed through the lens of broad economic growth, on-time payments to suppliers represent a public good because on-time payments create positive externalities for the economy. However, a buyer does not reap all the benefits from paying on time; therefore, in equilibrium, the buyer typically “underinvests” in paying on time. Although government regulators may desire that buyers pay their invoices on time, we remain agnostic about whether reducing the payment period is desirable. In fact, it could be argued that is actually desirable for buyers to delay payment. For example, buyers that delay payment can invest the cash they would have used to pay suppliers. By doing so, buyers can maximize returns from those other investments before paying suppliers, and such returns can spur financial growth for buyers. This in turn can facilitate general economic growth. Therefore, this reduction in the payment period does not necessary improve large-scale economic growth. We do not wish to engage in a normative debate about trading off suppliers’ welfare versus buyers’ welfare; nonetheless, we acknowledge that altering the length of the payment period can substantially affect the economy. Our study documents that buyers pay their suppliers more timely in response to a recent regulatory change that mandates the public disclosure of buyers’ payment practices and buyers’ timeliness in paying invoices. Our findings, which are based on a regulatory change in the UK, can inform regulators who are considering similar requirements in other regimes such as the US. Electronic copy available at: https://ssrn.com/abstract=3775656 The FASB and the SEC are currently considering whether and how to require increased disclosures about reverse factoring and other issues related to trade credit. It is the official position of the FASB and the SEC that their regulatory goal is to promote high-quality financial reporting to facilitate efficient resource allocation in the economy, as opposed to encouraging any type of “real effects.” Because we show that required disclosures can lead to unintended consequences, our findings can help guide the FASB and SEC as they design potential regulations related to trade credit. Electronic copy available at: https://ssrn.com/abstract=3775656 References Antoniou, A., Y. Guney, and K. Paudyal. 2006. The determinants of corporate debt maturity structure: Evidence from France, Germany and the UK. European Financial Management 12, 161–94. Barrot, J.-N. 2016. Trade credit and industry dynamics: evidence from trucking firms. Journal of Finance 71, 1975–2016. Barrot, J.-N, and R. Nanda. 2020. The employment effects of faster payment: Evidence from the federal Quickpay reform. Journal of Finance 75, 3139-3173. Biais, B, and C. Gollier. 1997. Trade credit and credit rationing. Review of Financial Studies 10, 903-937. Caniato, F, A. Moretto, and J. Rice. 2020. A financial crisis is looming for small suppliers. Harvard Business Review. Available at https://hbr.org/2020/08/a-financial-crisis-is-looming-for- smaller-suppliers. Chuk, E. 2013. Economic consequences of mandated accounting disclosures: Evidence from pension accounting standards. The Accounting Review 88(2), 395-427. Dechow, P. 1994. Accounting earnings and cash flows as measures of firm performance: The role of accounting accruals. Journal of Accounting and Economics 18, 3-42. Department of Business, Energy & Industrial Strategy 2016. Available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file /574312/duty-to-report-on-payment-practices-and-performance-government-response.pdf. Department of Business, Energy & Industrial Strategy. 2019. Available at https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file /831507/payment-practices-performance-reporting-requirements.pdf. Emery, G. 1984. A pure financial explanation for trade credit. Journal of Financial and Quantitative Analysis 19, 271-285. Fisman, R, and I. Love, 2003. Trade credit, financial intermediary development, and industry growth. Journal of Finance 58, 353–374. Ford, C. 2018, Interpreting log transformations in a linear model. University of Virginia Library. Available at https://data.library.virginia.edu/interpreting-log-transformations-in-a-linear-model/ Electronic copy available at: https://ssrn.com/abstract=3775656 Giannetti, M., M. Burkart, and T. Ellingsen. 2011. What you sell is what you lend? Explaining trade credit contracts. Review of Financial Studies 24(4), 1261–1298. Graham, J., M. Hanlon, and T. Shevlin. 2011. Real effects of accounting rules: Evidence from multinational firms’ investment locations and profit repatriation decisions. Journal of Accounting Research 49(1), 137-185. Henry, E., T. R. Robinson, and J. H. van Greuning. 2011. Financial analysis techniques. CFA Institute. Available at https://www.cfainstitute.org/-/media/documents/support/programs/cfa/2019-L1V3R26- footnotes.pdf. Human, T. 2021. Investors focus on climate while social issues are lower concerns, study finds. IR Magazine. Available at https://www.irmagazine.com/buy-side/investors-focus-climate-while- social-issues-are-lower-concern-finds-survey Imhoff, E. and J. Thomas. 1988. Economic consequences of accounting standards: The lease disclosure rule change. Journal of Accounting and Economics 10(4), 277-310. Ivashina V., B.C. Iverson, and D.C. Smith. 2016. The ownership and trading of debt claims in Chapter 11 restructurings. Journal of Financial Economics 119, 316-335 Kishan, S. and A. Marsh. 2021. Biodiversity, supply chain rank among biggest ESG themes in 2021. Bloomberg. Available at https://www.bloomberg.com/news/articles/2021-01- 08/biodiversity-supply-chain-rank-among-biggest-esg-themes-in-2021 Lintner, J. 1956. Distribution of incomes of corporations among dividends, retained earnings, and taxes. American Economic Review 46 (2): 97–113. Long, M.S., I.B. Malitz and S.A. Ravid 1993. Trade credit, quality guarantees, and product marketability’, Financial Management 22, 117–27. Mian, S. L. and C. Smith, 1992. Accounts receivable management policy: Theory and evidence. Journal of Finance 47, 169-200. Monga, V and N. Trentmann. 2020. Lockheed Martin, Micron, and others pay bills early to protect suppliers. Wall Street Journal. Available at https://www.wsj.com/articles/lockheed- micron-and-others-pay-bills-early-to-protect-suppliers-11599742800. Moody’s. 2019. Reverse factoring’s rising popularity comes with high, but hidden, risks. Moody's Investors Service Ltd. Available at https://www.moodys.com/research/Moodys- Reverse-factorings-rising-popularity-comes-with-high-but-hidden--PBC_1195322 Petersen, M.A. and R.G. Rajan, 1997. Trade credit: Theories and evidence, Review of Financial Studies 10, 661-669. Electronic copy available at: https://ssrn.com/abstract=3775656 Rajan, R. and L. Zingales. 1995. What do we know from capital structure: Some evidence from the international data, Journal of Finance 50, 1421-1460. Shumsky, T. and N. Trentmann 2018. Delaying payments to suppliers helps companies unlock cash. Wall Street Journal. Available at https://www.wsj.com/articles/delaying-payments-to- suppliers-help-companies-unlock-cash-1530178201. Steinberg, J. 2020. SEC asks Boeing, Coca-Cola to disclose more about popular financing tool, Wall Street Journal. Available at https://www.wsj.com/articles/sec-asks-boeing-coca-cola-to- disclose-more-about-popular-financing-tool-11598526651 Torrance, J. 2019. Calls for action on late payments a year after Carillion collapse. The Telegraph. Available at https://www.telegraph.co.uk/business/2019/01/15/calls-step-late- payments-crackdown-year-carillion-collapse/ Wahlen, J. M, S.P. Baginski, and M. T. Bradshaw. 2014. Financial reporting, financial statement analysis, and valuation: A strategic perspective. Boston: Cengage. Waters, C. 2017. Call for payment legislation as small business used as ‘cheap form of finance’. The Sydney Morning Herald. Available at https://www.smh.com.au/business/small-business/call- for-payment-legislation-as-small-businesses-used-as-cheap-form-of-finance-20170412- gvj8e7.html Electronic copy available at: https://ssrn.com/abstract=3775656 Appendix A Required Disclosures Mandated by Section 3 of the Small Business, Enterprise and Employment Act 2015 This appendix describes the required disclosures mandated by Section 3 of the Small Business, Enterprise and Employment Act 2015. Section 3 requires that large companies and large limited liability partnerships in the UK report the following every six months: Narrative descriptions of: - The organization’s payment terms, including the standard contractual length of time taken to pay invoices, the maximum contractual payment period, any changes to standard payment terms, and whether suppliers have been notified or consulted about these changes. - The organization’s process for the resolution of disputes related to payment. Statistics on: - The average time taken to pay invoices from the date the invoice is received. - The percentage of invoices paid within the reporting period that were paid in 30 days or less, between 31 and 60 days, and over 60 days. - The proportion of invoices that were due within the reporting period but were not paid within the agreed terms. Statements (i.e., a tick box) about: - Whether an organization offers e-invoicing. - Whether an organization offers supply chain finance. - Whether the organization’s practices and policies cover deducting sums from payments as a charge for remaining on a supplier’s list, and whether they have done this during the reporting period. - Whether the organization is a member of a payment code. If so, the name of the code. Section 3 applies to large companies and limited liability partnerships. To qualify as large, an organization must exceed at least two of the following three size thresholds on both of the firm’s last two balance sheet dates: (i) GBP 36 million in annual sales, (ii) GBP 18 million in total assets, and (iii) 250 employees. Electronic copy available at: https://ssrn.com/abstract=3775656 Appendix B Variable Definitions Variable Description An indicator variable that equals 1 for firm–year observations Post whose fiscal years begin on or after 6 April, 2017, and 0 otherwise, according to the disclosure requirement starting date. An indicator variable that equals 1 for firm-year observations that meet the size criteria for Section 3 regulation, and 0 otherwise. An indicator variable that equals 1 for firm–year observations whose assets, sales, and employees exceed 18 million, 36 million, and Large 250, respectively, in year t−1 and year t−2 (i.e., the two years before the regulation) and 0 otherwise, according to the three official thresholds that determine which firms qualify for the disclosure requirement. An indicator variable that equals 1 for UK firms and 0 for firms in UK Germany or France. The natural log of days payable outstanding (i.e, the average LnDaysPayble accounts payable ÷ COGS × 365). The natural log of days sales outstanding (i.e., the average LnDaysSales accounts receivable ÷ sales × 365). The natural log of days inventory outstanding (i.e., the average LnDaysInventory inventory ÷ COGS × 365). The natural log of days sales outstanding plus days inventory LnOperatingCycle outstanding. LnAT The natural log of assets. LnSale The natural log of sales. LnEmployee The natural log of employees. Leverage The ratio of total debt to lagged total assets. Growth The growth in sales. The ratio of income before extraordinary items to lagged total ROA assets. Cash The ratio of cash to lagged total assets. TradeCreditDependence The ratio of lagged accounts payable to lagged total liabilities. An indicator variable that equals 1 for firm–year observations DividendPayer paying dividend in the year, and 0 otherwise. Electronic copy available at: https://ssrn.com/abstract=3775656 Past due 31 days or The amount of trade credit owed by the firm that is more than 30 more days past due, scaled by total trade credit. Past due between 1 and The amount of trade credit owed by the firm that is between 1 and 30 30 days past due, scaled by total trade credit. Electronic copy available at: https://ssrn.com/abstract=3775656 Table 1. Sample Selection All UK firms available in Compustat (Global) 7,418 Less: UK firms that have missing values for necessary variables (3,711) Less: UK firms available only in the pre-period or the post-period (219) Less: UK firms with COGS less than GBP 1 million (311) Final Sample 3,177 firm–year obs Treated Group: UK firms above thresholds for regulation 1,983 firm–year obs Control Group: UK firms below thresholds for regulation 1,194 firm–year obs Electronic copy available at: https://ssrn.com/abstract=3775656 Table 2. Descriptive statistics Panel A Panel B Panel C Panel D UK UK France/Germany France/Germany Below thresholds Above thresholds Below thresholds Above thresholds VARIABLES mean sd mean sd mean sd mean sd Compustat Data Days Payable 89.28 99.45 75.10 97.38 107.21 120.19 82.85 81.62 LnDaysPayable 4.13 0.87 3.97 0.82 4.26 0.94 4.15 0.73 Operating Cycle 178.92 197.46 137.80 120.37 241.90 236.96 181.18 141.35 LnOperatingCycle 4.83 0.84 4.65 0.77 5.15 0.83 5.00 0.62 Assets (GBP in millions) 49.82 158.46 2258.6 5524.6 52.53 169.75 5800.6 14612.2 LnAT 3.17 1.06 6.21 1.67 3.19 1.04 6.63 2.01 Sales (GBP in millions) 27.97 47.13 1803.1 3824.0 30.18 51.84 4240.4 10112.8 LnSale 2.87 0.97 6.18 1.58 2.84 1.04 6.52 1.88 Employees (in thousands) 0.18 0.24 10.48 22.00 0.12 0.13 19.11 45.01 LnEmployees 0.15 0.15 1.54 1.17 0.11 0.10 1.73 1.40 Leverage 0.18 0.28 0.23 0.22 0.22 0.22 0.25 0.19 ROA −0.06 0.22 0.05 0.09 −0.07 0.21 0.04 0.08 Growth 0.22 0.54 0.09 0.21 0.11 0.41 0.07 0.17 Cash 0.24 0.26 0.12 0.12 0.24 0.25 0.16 0.15 DividendPayer 0.21 0.41 0.66 0.47 0.11 0.31 0.38 0.49 TradeCreditDependence 0.23 0.18 0.21 0.17 0.23 0.17 0.20 0.13 Proprietary Trade Credit Data Past due 31 days or more 0.05 0.40 0.10 0.30 0.04 0.37 0.07 0.24 Past due between 1 and 0.24 0.23 0.16 0.20 0.16 0.22 0.16 0.19 The table presents descriptive statistics. All variables are defined in Appendix B. Electronic copy available at: https://ssrn.com/abstract=3775656 Table 3. The effect of the UK payment disclosure requirement on payment periods 𝐷𝑎𝑦𝑠𝑃𝑎𝑦𝑎𝑏𝑙𝐿𝑛𝑒 = 𝛼 + 𝛼 𝑠𝑡𝑃𝑜 ∗ 𝐿𝑎𝑟𝑔𝑒 + ∑ 𝛼 𝑟𝑙𝑜𝑠𝑡𝑛𝐶𝑜 + 𝐹𝑖𝑟𝑚 𝑖𝑡 0 1 𝑡 𝑖 𝑘 𝑖𝑡 +𝑌𝑒𝑎𝑟 + 𝜀 𝑖𝑡 (1) VARIABLES Pred. LnDaysPayable Post × Large − −0.084*** (−2.81) LnOperatingCycle + 0.319*** (4.73) Leverage + 0.065 (1.36) LnAT + 0.197*** (2.86) ROA + −0.003 (−0.04) LnSale +/− −0.355*** (−3.81) Growth − −0.106*** (−3.32) LnEmployee + −0.047 (−0.41) Cash − −0.093 (−1.35) Constant 3.367*** (7.22) Observations 3,177 Fixed effects Firm, Year SE clustered by Firm Adjusted R-squared 0.886 The table presents the result from the model that tests the effect of the UK payment disclosure requirement on payment periods by using a difference-in-differences approach. The sample includes only UK firms. Large is an indicator variable that equals 1 for firm-year observations that meet the size criteria for Section 3 regulation, and 0 otherwise. Post is an indicator variable that equals 1 for firm–year observations whose fiscal years begin on or after 6 April, 2017, and 0 otherwise. The dependent variable is the natural log of days payable outstanding. All variables are defined in Appendix B. *, **, and *** denote two-tailed p-values significant at 10%, 5%, and 1%, respectively. Electronic copy available at: https://ssrn.com/abstract=3775656 𝐹𝐸 𝐹𝐸 Table 4. Cross-sectional tests for the change in payment periods 𝐿 𝑛 𝐷𝑎𝑦𝑠𝑃𝑎𝑦𝑎𝑏𝑙𝑒 = 𝛼 + 𝛼 𝑠𝑡𝑃𝑜 ∗ 𝐿𝑎𝑟𝑔𝑒 + 𝛼 𝑎𝑙𝑛𝑉𝑖𝑆𝑒𝑜𝐶𝑎𝑟𝑠𝑠𝑐𝑡𝑟𝑜 𝑖𝑡 0 1 𝑡 𝑖 2 +𝛼 𝑠𝑡𝑃𝑜 ∗ 𝑎𝑙𝑛𝑖𝑉𝑆𝑒𝑜𝐶𝑎𝑟𝑠𝑠𝑐𝑡𝑟𝑜 + 𝛼 𝐿𝑎𝑟𝑔𝑒 ∗ 𝑎𝑙𝑛𝑉𝑖𝑆𝑒𝑜𝐶𝑎𝑟𝑠𝑠𝑐𝑡𝑟𝑜 3 𝑡 𝑖𝑡 4 𝑖 𝑖𝑡 +𝛼 𝑠𝑡𝑃𝑜 ∗ 𝐿𝑎𝑟𝑔𝑒 ∗ 𝑆𝑒𝐶𝑟𝑠𝑠𝑐𝑡𝑜 𝑎𝑙𝑛𝑖𝑉𝑜𝑎𝑟 + ∑ 𝛼 𝑟𝑙𝑜𝑠𝑡𝑛𝐶𝑜 + 𝐹𝑖𝑟𝑚 + 𝑌𝑒𝑎𝑟 + 𝜀 5 𝑡 𝑖 𝑖𝑡 𝑘 𝑖𝑡 𝑖𝑡 Panel A. The length of operating cycles (1) VARIABLES LnDaysPayable Post × Large −0.439** (−2.55) LnOperatingCycle 0.243*** (3.27) Post × LnOperatingCycle −0.040 (−1.42) Large × LnOperatingCycle 0.156 (1.58) Post × Large × LnOperatingCycle 0.073** (2.00) Observations 3,177 Controls Yes Fixed effects Firm, Year SE clustered by Firm Adjusted R-squared 0.887 Electronic copy available at: https://ssrn.com/abstract=3775656 𝐹𝐸 𝐹𝐸 𝑖𝑡 Panel B. Trade credit dependence (1) VARIABLES LnDaysPayable Post × Large −0.178*** (−3.65) TradeCreditDependence 1.254*** (8.92) Post × TradeCreditDependence −0.363*** (−3.01) Large × TradeCreditDependence 0.277 (1.12) Post × Large × TradeCreditDependence 0.451*** (3.04) Observations 3,177 Controls Yes Fixed effects Firm, Year SE clustered by Firm Adjusted R-squared 0.898 Panel C. Dividend payments (1) VARIABLES LnDaysPayable Post × Large −0.138*** (−3.28) DividendPayer −0.034 (−0.54) Post × DividendPayer −0.036 (−0.79) Large × DividendPayer −0.037 (−0.50) Post × Large × DividendPayer 0.106* (1.86) Observations 3,177 Controls Yes Fixed effects Firm, Year SE clustered by Firm Adjusted R-squared 0.886 The table presents results from cross-sectional tests for the effect of the UK payment disclosure requirement on payment periods. The sample includes only UK firms. Large is an indicator variable that Electronic copy available at: https://ssrn.com/abstract=3775656 equals 1 for firm-year observations that meet the size criteria for Section 3 regulation, and 0 otherwise. Post is an indicator variable that equals 1 for firm–year observations whose fiscal years begin on or after 6 April, 2017, and 0 otherwise. The dependent variable is the natural log of days payable outstanding. All variables are defined in Appendix B. *, **, and *** denote two-tailed p-values significant at 10%, 5%, and 1%, respectively. Electronic copy available at: https://ssrn.com/abstract=3775656 Table 5. Robustness check: Using German and French firms as the control group 𝐷𝑎𝑦𝑠𝑃𝑎𝑦𝑎𝑏𝑙𝐿𝑛𝑒 = 𝛼 + 𝛼 𝑠𝑡𝑃𝑜 ∗ 𝑈𝐾 + ∑ 𝛼 𝑟𝑙𝑜𝑠𝑡𝑛𝐶𝑜 + 𝐹𝑖𝑟𝑚 + 𝑌𝑒𝑎𝑟 + 𝜀 𝑖𝑡 0 1 𝑡 𝑖 𝑘 𝑖𝑡 𝑖𝑡 𝐷𝑎𝑦𝑠𝑃𝑎𝑦𝑎𝑏𝑙𝐿𝑛𝑒 = 𝛼 + 𝛼 𝑠𝑡𝑃𝑜 ∗ 𝐿𝑎𝑟𝑔𝑒 + 𝛼 𝑠𝑡𝑃𝑜 ∗ 𝑈𝐾 𝑖𝑡 0 1 𝑡 𝑖 2 𝑡 𝑖 +𝛼 𝑠𝑡𝑃𝑜 ∗ 𝐿𝑎𝑟𝑔𝑒 ∗ 𝑈𝐾 + ∑ 𝛼 𝑟𝑙𝑜𝑠𝑡𝑛𝐶𝑜 + 𝐹𝑖𝑟𝑚 + 𝑌𝑒𝑎𝑟 + 𝜀 3 𝑡 𝑖 𝑖 𝑘 𝑖𝑡 𝑖𝑡 (1) (2) German/French firms above Triple differences thresholds as the control group VARIABLES LnDaysPayable LnDaysPayable Post × Large 0.046* (1.75) Post × UK −0.039** 0.086** (−2.12) (2.51) Post × Large × UK −0.125*** (−3.21) LnOperatingCycle 0.343*** 0.316*** (4.55) (5.67) Leverage 0.015 0.082** (0.28) (2.17) LnAT 0.252*** 0.173*** (3.13) (3.26) ROA 0.084 0.051 (0.84) (0.77) LnSale −0.421*** −0.319*** (−4.34) (−5.29) Growth −0.149*** −0.086*** (−2.62) (−3.20) LnEmployee 0.077 −0.030 (0.74) (−0.35) Cash −0.180** −0.118** (−2.30) (−2.40) Constant 3.382*** 3.323*** (5.55) (9.27) Observations 4,942 7,535 Fixed effects Firm, Year Firm, Year SE clustered by Firm Firm Adjusted R-squared 0.921 0.898 The table presents results from models that test the effect of the UK payment disclosure requirement on payment periods by using German and French firms as the control group. In Column 1, the sample includes only large firms in the UK, Germany, and France. In Column 2, the sample includes all firms in the UK, Germany, and France. UK is an indicator variable that equals 1 for UK firms and 0 for firms in Germany or France. Large is an indicator variable that equals 1 for firm-year observations that meet the Electronic copy available at: https://ssrn.com/abstract=3775656 𝐹𝐸 𝐹𝐸 𝐹𝐸 𝐹𝐸 size criteria for Section 3 regulation, and 0 otherwise. Post is an indicator variable that equals 1 for firm– year observations whose fiscal years begin on or after 6 April, 2017, and 0 otherwise. The dependent variable is the natural log of days payable outstanding. All variables are defined in Appendix B. *, **, and *** denote two-tailed p-values significant at 10%, 5%, and 1%, respectively. Electronic copy available at: https://ssrn.com/abstract=3775656 Table 6. Additional test: Examining past due trade credit Panel A Panel B Panel C UK firms below thresholds as the German and French firms above Triple differences control group thresholds as the control group VARIABLES Past due between Past due 31 days Past due between Past due 31 days Past due between Past due 31 days 1 and 30 or more 1 and 30 or more 1 and 30 or more 0.024 0.050 Post × Large 0.091** −0.097* (2.50) (−1.79) (0.86) (1.08) Post × UK 0.027* −0.038* −0.050 0.111 (1.79) (−1.87) (−1.18) (1.56) Post × Large × UK 0.076* −0.148** (1.69) (−2.00) LnOperatingCycle 0.010 0.001 0.021 −0.028 0.005 0.006 (0.50) (0.03) (0.60) (−0.48) (0.23) (0.19) Leverage −0.084 0.071 −0.085 −0.060 −0.086* 0.041 (−0.86) (0.84) (−1.25) (−0.73) (−1.69) (0.68) LnAT 0.043 −0.023 0.019 0.189** 0.001 0.035 (0.94) (−0.20) (0.53) (2.40) (0.02) (0.53) ROA −0.048 0.083 0.157* −0.130 −0.021 −0.031 (−0.56) (0.57) (1.67) (−0.79) (−0.34) (−0.31) LnSale −0.052 0.064 −0.020 −0.003 −0.022 0.084 (−0.93) (0.56) (−0.41) (−0.03) (−0.62) (1.36) Growth −0.002** −0.003* −0.061* 0.083 −0.003*** −0.002** (−2.52) (−1.76) (−1.90) (1.51) (−5.42) (−2.42) LnEmployee 0.069 0.067 −0.016 −0.088 −0.009 −0.032 (1.04) (0.65) (−0.37) (−1.31) (−0.25) (−0.53) Cash 0.029 0.019 −0.162* −0.064 −0.015 −0.007 (0.36) (0.18) (−1.68) (−0.76) (−0.28) (−0.12) Constant 0.041 −0.257 0.131 −0.903* 0.308 −0.681** (0.13) (−0.56) (0.39) (−1.83) (1.34) (−2.16) Observations 1,115 1,115 2,502 2,502 2,989 2,989 Fixed effects Firm, Year Firm, Year Firm, Year Firm, Year Firm, Year Firm, Year SE clustered by Firm Firm Firm Firm Firm Firm Adjusted R−squared 0.341 0.276 0.404 0.340 0.401 0.344 Electronic copy available at: https://ssrn.com/abstract=3775656 The table presents results from models that test the effect of the UK payment disclosure requirement on past due trade credit. In Column 1, the sample includes only UK firms. In Column 2, the sample includes only large firms in the UK, Germany, and France. In Column 3, the sample includes all firms in the UK, Germany, and France. UK is an indicator variable that equals 1 for UK firms and 0 for firms in Germany or France. Large is an indicator variable that equals 1 for firm-year observations that meet the size criteria for Section 3 regulation, and 0 otherwise. Post is an indicator variable that equals 1 for firm–year observations whose fiscal years begin on or after 6 April, 2017, and 0 otherwise. The dependent variable is the natural log of days payable outstanding. All variables are defined in Appendix B. *, **, and *** denote two-tailed p-values significant at 10%, 5%, and 1%, respectively. Electronic copy available at: https://ssrn.com/abstract=3775656
Journal
ARN Conferences & Meetings – SSRN
Published: Jan 18, 2021
Keywords: regulation, disclosure, supplier payment period, trade credit, customer–supplier relationships, supply chain