The determinants and value-relevance of voluntary disclosure of supply chain information

Abstract

We use the voluntary nature of supply chain information disclosure in China’s stock market, including both major customers and suppliers information, to study the determinants and value relevance of proprietary information voluntary disclosure. Consistent with information asymmetry concern, disclosure is more likely when firms are seeking external finance or operating with a more concentrated supply chain where the needs of reducing information asymmetry are higher. Supply chain disclosure is found to be associated with a lower firm valuation on average. Good corporate governance reduces such voluntary disclosure, further confirming protecting proprietary information is one of the key considerations of non-disclosure. The disclosure of supplier identity is less value relevant than customer identity.

Keywords:

• voluntary disclosure
• supply chain
• customer
• supplier
• information asymmetry
• proprietary costs
• corporate governance

Acknowledgments

The authors appreciate the helpful suggestions from Timothy King, Isabel Wang and Tong Yu. We thank seminar participants at the 11th International Accounting & Finance Doctoral Symposium, the 2018 NWSSDTP Job Market and Employability Skills Workshop for Accounting and Finance, the 2019 CAPANA annual conference and the 2019 CFRI conference for helpful comments. All authors have contribution equally to this work.

Data availability

Data used in this research are from a commercial database CSMAR and supplement with manual collection. Due to the licence restriction, public sharing of the data is prohibited.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 See, for example, the effects of customer concentration on performance (Patatoukas 2012), cost of capital (Dhaliwal et al. 2016, Cen et al. 2016), capital structure (Banerjee et al. 2008), dividend policy (Wang 2012), cash holdings (Itzkowitz 2013), tax planning (Cen et al. 2017) and earnings management (Raman and Shahrur 2008).

2 The US Securities and Exchange Commission (SEC) Regulation S-K and the Statement of Financial Accounting Standards (SFAS) No.131 both state that companies listed in the US capital market should disclose the name (or identity) and trade volume of any customer if aggregate sales to them are equal to 10% or more of total revenues.

3 We survey the regulation and show examples of supply chain information disclosure in China in Section 2.1 and Online Appendix A.

4 Ellis et al. (2012) also study some evidence of voluntary disclosure in the context of disclosing none major customer information. They find some opportunistic behaviour in disclosing none major customers’ identity confirming the existence of a good news bias.

5 See for example, Chilton and Sarfty’s (2017) discussion on the limitations of supply chain disclosure regimes and Jarman et al. (2016) on encouraging private sector to disclosure their supply chain information for public benefit.

6 The relatively lower disclosure of the supplier information from companies is partly due to encouragement of voluntary disclosure for supplier information is three years later than that of the costumer information. See section 2.1 for more details about the disclosure regimes.

7 Specifically, the Inverse Mills ratios (IMR) from the analysis of the likelihood of disclosure are included in synchronicity and firm value regressions.

8 Particularly, firms in the durable goods and/or highly competitive industries would be more concerned about proprietary information because of their specific investment and unique products (Banerjee et al. 2008) or their current competitive position with respect to rivals (Li 2010).

9 Although our evidence cannot speak directly about the effect of supply chain disclosure on issues such as modern slavery, it contributes to this debate with new evidence (see, an example of the policy debate in the UK regarding its Modern Slavery Act 2015 (Home Office 2019)). One of the concerns about compulsory disclosure of supply chain information is its negative impact on firm value due to the revealing of proprietary business information. Our empirical evidence on its value relevant shows that there is a circumstance that disclosing such information is not value-destroying. Especially when the demand for information is high from the investors given the risk embedded in supply chain measured by concentration in our context. With the increasing awareness of the importance of supply chain sustainability, more and more investors will concern about this aspect of the risk in the supply chain and demand for more disclosure of such information. Our findings suggest that disclosing such information given the demand is potentially value enhancing.

10 Madsen (2017)findsthat investors search for customer information in orderto conjecture a company’s earnings,and this cannot be achieved without customeridentity information. Investorscould also make use of information oncustomer type to assess a company’s product quality (Fynes et al. 2005) and supplier type to assess company performance (Steven et al. 2014). In particular, large customers and suppliers can act as certifying entities, given that they have higher incentives and lower costs when monitoringthe company (Lanier et al. 2010, Steven et al. 2014).

11 In addition to shareholder protection, the government also uses firms’ supply chain information for other purposes. Cen et al. (2017) show significant correlation between a firm’s customer relationship and its tax avoidance behaviour. Thus, the supply chain relationships could be the target of government investigation. In the practice of China, the government also uses firms’ supply chain information to inspect commercial corruption as business tends to be involved in corruption (Cai et al. 2011).

12 There is an alternative view on the effect of corporate governance from the stakeholder’s point of view. Disclosing supply chain information (especially suppliers) may also have its social benefit given the rising concern regarding the sustainability of the supply chain and modern slavery. A good governance company may have a more sustainable supply chain and more likely to disclose them. Given this possible alternative, the combined effect of corporate governance will be unclear. However, in the context of our research, the influence of stakeholder is weaker in emerging market and therefore the main shareholder alignment effect of governance is more likely to be observed.

13 In our main analysis, both $R\mathrm{\&}DExpenses$ and $Paten{t}_{Ap{p}_{Avg}}$ are included in the regressions. We also only use either $R\mathrm{\&}DExpenses$ or $Paten{t}_{Ap{p}_{Avg}}$ in our regressions and the main conclusions remain. Results are available with request on authors.

14 The odds ratios are retrieved by taking exponentials of the coefficients.

15 The odds ratios are Exp(0.1954) = 1.22 and Exp(0.7332) = 2.08 for columns 2 and 4, respectively.

16 Proprietary cost is amore significant determinant of the disclosure of identity than of the tradeproportion of individual suppliers/customers. This is because identity disclosure is more relevant to revealing proprietary information than isthe disclosure of anonymous trading volumes.

17 There is a potential endogeneity concern regarding companies with high R&D and patent activities and their disclosure choice. To this end we conduct an IV approach as a robustness test. We use the regional R&D expenditure input per GDP ($R\mathrm{\&}{D}_R$), students in Regular Institution of Higher Education per one million population ($Studen{t}_R$), the industry mean of R&D intensive ($R\mathrm{\&}{D}_I$) and the industry mean of patent applicants ($Paten{t}_I$) as the IVs for firm’s R&D intensity and patent applications. We then employ the IV-probit model. The results are consistent with those present here and can be found in the online Appendix C5. We thank an anonymous referee for highlighting this potential concern. 

18 The odd ratio of equity incentive for column 4 is exp(−0.3308) = 0.72; and this suggests when equity incentive taken value of 1 there is 1−0.72 = 0.28 or 28% lower in the odds of disclosing supplier’s identity.

19 In addition to the variables in our disclosure determinant model in Table 3, in the valuation determinant model in equation 2 we also add control variables for profitability (Chi 2005), institutional ownership (McConnell and Servaes 1995) and liquidity (Fang et al. 2009) measured by stock turnover ratio (Datar et al. 1998).

20 The standard deviation of Tobin’s Q for the sample is 2.42.

21 Furthermore, stronger results documented in the identity (3 and 4) than in volume disclosure regressions (1 and 2) are partly expected. This is because disclosing the identity of the customer/supplier, in addition to the volume, is one step further in terms of disclosing firm specific information. We further confirm this difference between customer and supplier information disclosure in an online appendix. We classify company according to their combined disclosure behaviour for the customer and supplier information into four groups including companies disclosing: (1) customer only, (2) supplier only, (3) both and (4) none. We find that for firm-year observations only disclosing supplier’s information, there is no significant negative value effect while the negative valuation effect still observed for those disclosing either customer only or both.

22 http://www.csrc.gov.cn/pub/csrc_en/newsfacts/release/200708/t20070816_69104.html

Additional information

Funding

This work was supported by the National Natural Science Foundation of China: Research Project (grant number 72102244): ‘A study on the effectiveness of CSRC's random inspection’ and Research Project (grant number 71772188): ‘Fraud regulation and bank loan contracts’, and the National Social Science Fund of China: Research Project (grant number 20BJL016): ‘Macro-economic resilience in production and credit networks’.