Skip to main content
SpringerLink
Log in

State contract law and the use of accounting information in debt contracts

  • Published:
Review of Accounting Studies Aims and scope Submit manuscript

Abstract

This paper examines the relation between state contract law and the use of accounting information in debt contracts. Contract theory suggests that balance sheet based covenants resolve debtholder-shareholder conflicts ex ante, whereas income statement based covenants serve as trip- wires that trigger the switch of control rights ex post. It is more difficult for lenders to exert their control rights ex post if the contract law is more favorable to debtors (i.e., the law is pro-debtor), suggesting that balance sheet based covenants are more efficient in these jurisdictions. We therefore test and find evidence that lenders using pro-debtor (pro-lender) law are more (less) likely to rely on balance sheet based covenants. We measure reliance using both the weight of balance sheet covenants relative to income statement covenants and the covenant strictness. Our analysis further shows that contracts with performance pricing grids are less likely to include interest increasing grids when the law is more favorable to debtors. The results provide initial evidence that contract law is an important determinant for the design of debt contracts.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

Notes

  1. Honigsberg et al. (2014) document that the higher interest rates are limited to out-of-state borrowers, who opt into favorable (more pro-debtor) state laws.

  2. Christensen and Nikolaev (2012) classify covenants as capital covenants and performance covenants. Although there is a difference in terminology, these classifications measure the same construct as balance sheet-based covenants and income-statement covenants, respectively.

  3. We perform additional robustness tests and analyses that we do not tabulate for brevity (e.g., we re-estimate our performance pricing model at the facility rather than deal level and obtain weaker but consistent results).

  4. Within country designs are not limited to US settings. Recent work analyzing changes in bankruptcy laws in Italy has shown that bankruptcy laws affect loan interest rates and firm investment decisions (Bonetti 2017; Rodano et al. 2016).

  5. In the context of commercial contracts, courts will enforce the contracting parties’ choice of law only if that state has a “reasonable relationship” to the contract. However, the “reasonable relationship” term has been interpreted very broadly in recent decades (Eisenberg and Miller 2010). Additionally, in an effort to create work for state-licensed attorneys, many states have enacted statutes that are meant to ensure choice of law clauses in commercial contracts are enforced. Combined, these factors provide parties to commercial contracts with substantial flexibility in selecting the law that will govern their agreement (Eisenberg and Miller 2010).

  6. Using a sample of over 3000 debt contracts, Honigsberg et al. (2014) show that the borrower’s state of incorporation differs from the state law governing the debt contract more than 90% of the time.

  7. Bankruptcy proceedings do not affect our analysis. When firms file for bankruptcy, the bankruptcy courts will apply the law that was selected in the contract. Further, although there are some differences in procedure that are controlled by the local jurisdiction, bankruptcy law itself is federal.

  8. In their review of the literature on debt covenants, Armstrong et al. (2010) note that debtholders value some accounting attributes but not others, and they urge researchers to analyze factors that lead debtholders to favor certain types of accounting information.

  9. Aghion and Bolton (1992) show two types of efficient control allocations: unilateral and contingent control allocations. The unilateral control allocation regime assigns control to the borrower (and in some cases to the lender) by aligning the interests of the lender and borrower at the initiation of the contract, while the contingent control allocation regime reallocates control between the lender and borrower based on signals during the term of the contract.

  10. Covenants that include a mix of both balance sheet and income statement numbers (e.g., debt to EBITDA ratio) are classified as income statement-based covenants, because income statement numbers are flow numbers that use the stock number as a scalar. Balance sheet-based covenants include balance sheet numbers only. See Appendix 2 for covenant mix classifications.

  11. See, e.g., Layne v. Ft. Carson Nat’l Bank, 655 P.2d 856 (Colo. 1982) (borrower sued bank after a 5% increase in the interest rate, alleging that the bank acted in bad faith); Homelife Props. Ltd. v. Fahey Banking Co., 2010 Ohio Misc. LEXIS 522 (borrower sued bank, alleging that it improperly tried to increase the interest rate on a commercial loan); First Nat’l Mont. Bank v. McGuinness, 705 P2d 579 (borrowers sued bank after the bank informed the borrowers that the interest rate would be raised on the final year of a contract extension).

  12. This matching link data is available until 2017. http://finance.wharton.upenn.edu/~mrrobert/styled-9/styled-12/index.html

  13. We identified the number of lender liability lawsuits by relying on Cappello (2009), the leading treatise on lender liability. This treatise is focused on the branch of law that seeks to protect borrowers—not equity holders or other stakeholders—from unfair lending practices. To eliminate cases unrelated to state law, we eliminated all cases brought exclusively under federal law (e.g., the bankruptcy code). The hundreds of remaining cases in the treatise largely consist of claims arising under contract law, deceptive practices statutes, environmental law, fiduciary relationships, and sales of collateral.

  14. The intuition is that parties are more likely to settle rather than litigate when the legal rules are certain, because they can anticipate how the courts will rule and will not waste time and money in litigation. Hence it is only worthwhile to litigate when the case outcome is uncertain.

  15. Common law was traditionally very friendly to lenders, but over time a series of borrower friendly cases have muddled the law in certain jurisdictions and created uncertainty. As an example, consider the most famous lender liability case: K.M.C. Co. v. Irving Trust Co., 757 F.2d 752 (6th Cir. 1985). In this case, the lender refused to provide additional funds that were available under the borrower’s line of credit, because the lender believed that the borrower posed a credit risk. When the borrower went out of business and sued the lender for violating the duty of good faith, the jury awarded the borrower $7,500,000. The case was highly unusual at the time, because the lender was found liable for taking an action that was expressly permitted by the contract terms. As many would expect, the frequency of similar claims spiked after this case. What had previously seemed clear in the law—that the lender could withhold funds if the contract permitted him to do so—was now uncertain. As such, plaintiffs had incentives to bring litigation in instances where there was previously no reason for them to waste their time. Hence this case illustrates the idea that uncertainty in this particular area of law is bad for lenders.

  16. Demerjian (2011) uses a subset of the covenants used by Christensen and Nikolaev (2012). For consistency, we follow the classification of Christensen and Nikolaev (2012) as described in Appendix 2.

  17. http://faculty.washington.edu/pdemerj/data.html

  18. In addition to controlling for year fixed effects, we conduct three pseudo-falsification tests with the New York subsample (untabulated): (1) we estimate the same regression over the pre period from 1995 to 2003, assuming a pseudo-change in year 2000; (2) we estimate the same regression over the post period from 2004 to 2017, assuming a pseudo-change in year 2011; and (3) we estimate the model over the sample of contracts governed by states other than New York during the full sample period, assuming a pseudo-change in year 2004. These pseudo change indicators do not show significant coefficients.

  19. In untabulated tests, we experiment with less restrictive variations of this approach, identifying subsamples where (i) the home state of the lead lender is the same as the contract law, (ii) the home state of the borrower is the same as the contract law, and (iii) the lead lender and borrower are from the same state. In all such cross-sectional tests, the interaction term is positive and statistically significant.

  20. Our descriptive analyses show that deals with bank dependent borrowers are more frequently governed by pro lender state contract laws.

  21. We are grateful to an anonymous reviewer for suggesting this proxy for bargaining power. Although untabulated, we also use borrowers with non-investment grade debt (S&P long-term bond rating lower than BBB) as a proxy for bank dependence and find similar results.

  22. In an untabulated analysis, we tested whether our results are stronger for the sample of firms with highly volatile income by interacting net income volatility with the Pro-Debtor Index. Although our main result continued to hold, the interaction term was not significant, suggesting that our results are not driven by firms with high income volatility. Similarly, our results are robust to controlling for income and cash volatility in the model.

  23. In untabulated tests, we find no evidence of association between the state contract law and the inclusion of interest decreasing performance pricing grids.

  24. We use the presence of security as a benchmark because prior literature has shown that the security feature of the loan is an important determinant of performance pricing grid design (Christensen and Nikolaev 2012; Hollander and Verriest 2016).

  25. We perform performance pricing tests across two alternative sample specifications (untabulated). First we estimate model (3) over the sample of loans that include performance pricing grids. Results are robust, albeit weaker. Second, since the performance pricing grids are designed at the facility level, we estimate the model (3) at facility level, recalculating deal-level control variables at the facility level. The results resemble those reported in Table 7.

  26. We estimated our Performance Pricing model over within sample settings reported in Tables 3 and 4 for robustness. Due to the many fixed effects and small sample sizes, the logit model cannot be estimated in many cases and yields insignificant results when converged.

  27. For example, in Diversified Foods, Inc. v. First National Bank of Boston (1991), Me. Super. LEXIS 84, the borrower sued the lender, alleging that the lender violated the duties of good faith and fair dealing by taking actions such as raising the APR. In evaluating the borrower’s claims, the court looked at the borrower’s actions preceding the lender’s alleged breaches and stressed repeatedly that the borrower had voluntarily breached the contract by, for example, violating the covenant related to intercompany advances. The lender, the court stated, simply exercised its rights under the loan agreement in response to the borrower’s breaches. In effect, the borrower’s voluntary breach created a defense for the lender against lender liability claims.

  28. We developed this algorithm after reading numerous debt contracts to identify their governing law.

  29. We use the firms’ primary state of operations as their home states, not the state where the firms are incorporated.

  30. We check several diagnostics (untabulated) to assess the validity of our instrumental variable approach, following Larcker and Rusticus (2010). We find a partial R-squared of 37% with a significant first-stage partial F-statistic, which supports the choice of instruments. However, the Durbin-Wu-Hausman tests yield insignificant chi-square statistics for endogeneity, which can be seen as evidence of lack of endogeneity.

  31. Dealscan typically treats these agreements as new agreements and assigns a new unique deal identifier.

  32. We do not include renegotiation events that are coded as “amended” in Roberts’ (2015) sample, as these events do not lead to a new contract.

References

  • Aghion, P., & Bolton, P. (1992). An incomplete contracts approach to financial contracting. Review of Economic Studies, 59, 473–494.

    Google Scholar 

  • Aier, J. K., Chen, L., & Pevzner, M. (2014). Debtholders’ demand for conservatism: Evidence from changes in directors’ fiduciary duties. Journal of Accounting Research, 52, 993–1027.

    Google Scholar 

  • Armstrong, C., Guay, W., & Weber, J. (2010). Role of information and financial reporting in corporate governance and debt contracting. Journal of Accounting and Economics, 50, 179–234.

    Google Scholar 

  • Asquith, P., Beatty, A., & Weber, J. (2005). Performance pricing in debt contracts. Journal of Accounting and Economics, 40, 101–128.

    Google Scholar 

  • Ball, R., Robin, A., & Sadka, G. (2008). Is financial reporting shaped by equity markets or by debt markets? An international study of timeliness and conservatism. Review of Accounting Studies, 13, 168–205.

    Google Scholar 

  • Beatty, A., Weber, J., & Yu, J. J. (2008). Conservatism and debt. Journal of Accounting and Economics, 45, 154–174.

    Google Scholar 

  • Bonetti, P. (2017). Debt renegotiation and the properties of accounting information: Evidence from a bankruptcy reform. Working paper. Available at SSRN: https://ssrn.com/abstract=2887944 or https://doi.org/10.2139/ssrn.2887944. Accessed 21 July 2020.

  • Bozanic, Z., Cheng, L., & Zach, T. (2018). Soft information in loan agreements. Journal of Accounting, Auditing and Finance, 33(1), 40–71.

    Google Scholar 

  • Brown, D., Ciochetti, B., & Riddiough, T. (2006). Theory and evidence on the resolution of financial distress. Review of Financial Studies, 19, 1357–1397.

    Google Scholar 

  • Burton, S. J. (1994). Good faith in articles 1 and 2 of the U.C.C.: The practice view. William and Mary Law Review, 35, 1533–1564.

    Google Scholar 

  • Cappello, A. B. (2009). Lender liability – 4th edition. Juris Publishing, Inc.

  • Chava, S., & Roberts, M. R. (2008). How does financing impact investment? The role of debt covenants. The Journal of Finance, 63, 2085–2121.

    Google Scholar 

  • Christensen, H. B., & Nikolaev, V. V. (2012). Capital versus performance covenants in debt contracts. Journal of Accounting Research, 50, 75–116.

    Google Scholar 

  • Christensen, H. B., Nikolaev, V. V., & Wittenberg-Moerman, R. (2016). Accounting information in financial contracting: The incomplete contract theory perspective. Journal of Accounting Research, 54, 397–435.

    Google Scholar 

  • Coase, R. H. (1937). The nature of the firm. Economica, 4, 386–405.

    Google Scholar 

  • Costello, A. M., & Wittenberg-Moerman, R. (2011). The impact of financial reporting quality on debt contracting: Evidence from internal control weakness reports. Journal of Accounting Research, 49, 97–136.

    Google Scholar 

  • De Franco, G., Vasvari, F., Vyas, D., & Wittenberg-Moerman, R. (2019). Similarity in the restrictiveness of bond covenants. European Accounting Review. https://doi.org/10.1080/09638180.2019.1664311.

  • Demerjian, P. R. (2011). Accounting standards and debt covenants: Has the “balance sheet approach” led to a decline in the use of balance sheet covenants? Journal of Accounting and Economics, 52, 178–202.

    Google Scholar 

  • Demerjian, P. R., & Owens, E. L. (2016). Measuring the probability of financial covenant violation in private debt contracts. Journal of Accounting and Economics, 61, 433–447.

    Google Scholar 

  • Dichev, I., & Skinner, D. (2002). Large-sample evidence on the debt covenant hypothesis. Journal of Accounting Research, 40, 1091–1123.

    Google Scholar 

  • Eisenberg, T., & Miller, G. P. (2007). Do juries add value? Evidence from an empirical study of jury trial waiver clauses in large corporate contracts. Journal of Empirical Legal Studies, 4, 539–588.

    Google Scholar 

  • Eisenberg, T., & Miller, G. P. (2010). The market for contracts. Cardozo Law Review, 30, 2073–2098.

    Google Scholar 

  • Francis, J., Philbrick, D., & Schipper, K. (1994). Shareholder litigation and corporate disclosures. Journal of Accounting Research, 32, 137–164.

    Google Scholar 

  • Gergen, M. (1993). A cautionary tale about contractual good faith in Texas. Texas Law Review, 72, 1235–1276.

    Google Scholar 

  • Givoly, D., Hayn, C., & Katz, S. (2016). The changing relevance of accounting information to debt holders over time. Review of Accounting Studies, 22(1), 64–108.

    Google Scholar 

  • Grossman, S. J., & Hart, O. D. (1986). The costs and benefits of ownership: A theory of vertical and lateral integration. Journal of Political Economy, 94, 691–719.

    Google Scholar 

  • Gulati, M., & Scott, R. E. (2012). The three and a half minute transaction: Boilerplate and the limits of contract design. The University of Chicago Press.

  • Hart, O., & Moore, J. (1990). Property rights and the nature of the firm. Journal of Political Economy, 98, 1119–1158.

    Google Scholar 

  • Heninger, W. G. (2001). The association between auditor litigation and abnormal accruals. The Accounting Review, 76, 111–126.

    Google Scholar 

  • Hollander, S., & Verriest, A. (2016). Bridging the gap: The design of bank loan contracts and distance. Journal of Financial Economics, 119, 399–419.

    Google Scholar 

  • Hong, H. A., Hung, M., & Zhang, J. (2016). The use of debt covenants worldwide: Institutional determinants and implications on financial reporting. Contemporary Accounting Research, 33(2), 644–681.

    Google Scholar 

  • Honigsberg, C., Katz, S., & Sadka, G. (2014). State contract law and debt contracts. Journal of Law and Economics, 57(3), 1031–1061.

    Google Scholar 

  • Klein, B., Crawford, R. G., & Alchian, A. A. (1978). Vertical integration, appropriable rents, and the competitive contracting process. Journal of Law and Economics, 21, 297–326.

    Google Scholar 

  • La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. (1997). Legal determinants of external finance. Journal of Finance, 52, 1131–1150.

    Google Scholar 

  • La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. (1998). Law and finance. Journal of Political Economy, 106(6), 1113–1155.

    Google Scholar 

  • Larcker, D., & Rusticus, T. (2010). On the use of instrumental variables in accounting research. Journal of Accounting and Economics, 49, 186–205.

    Google Scholar 

  • Li, N. (2010). Negotiated measurement rules in debt contracts. Journal of Accounting Research., 48, 1103–1144.

    Google Scholar 

  • Li, J., Elayan, F. A., Meyer, T. O., & Pacharn, P. (2012). The outcome of backdating investigations: Economic consequences, market overreaction and management motives. Investment Management and Financial Innovations, 9, 70–87.

    Google Scholar 

  • Mansi, S. A., Maxwell, W. F., & Wald, J. K. (2009). Creditor protection laws and the cost of debt. Journal of Law and Economics, 52, 701–726.

    Google Scholar 

  • Mian, A., Sufi, A., & Trebbi, F. (2015). Foreclosures, house prices, and the real economy. Journal of Finance, 70, 2587–2634.

    Google Scholar 

  • Miller, D. P., & Reisel, N. (2012). Do country-level investor protections impact securities-level contract design? Evidence from foreign bond covenants. Review of Financial Studies, 25, 408–438.

    Google Scholar 

  • Minnis, M., & Sutherland, A. (2017). Financial statements as monitoring mechanisms: Evidence from small commercial loans. Journal of Accounting Research, 55, 197–233.

    Google Scholar 

  • Murfin, J. (2012). The supply-side determinants of loan contract strictness. Journal of Finance, 67, 1565–1601.

    Google Scholar 

  • Nini, G., Smith, D. C., & Sufi, A. (2009). Creditor control rights and firm investment policy. Journal of Financial Economics, 92, 400–420.

    Google Scholar 

  • Nyarko, J. (2020). Stickiness and incomplete contracts. The University of Chicago Law Review, forthcoming.

  • Priest, G. L., & Klein, B. H. (1984). The selection of disputes for litigation. The Journal of Legal Studies, 13(1), 1–55.

    Google Scholar 

  • Qi, Y., & Wald, J. K. (2008). State laws and debt covenants. Journal of Law and Economics, 51(2), 179–207.

    Google Scholar 

  • Qian, J., & Strahan, P. E. (2007). How laws and institutions shape financial contracts: The case of bank loans. The Journal of Finance, 62, 2803–2834.

    Google Scholar 

  • Roberts, M. R. (2015). The role of dynamic renegotiation and asymmetric information in financial contracting. Journal of Financial Economics, 116, 61–81.

    Google Scholar 

  • Roberts, M., & Sufi, A. (2009). Control rights and capital structure: an empirical investigation. Journal of Finance, 64, 1657–1695.

  • Rodano, G., Serrano-Velarde, N., & Tarantino, E. (2016). Bankruptcy law and bank financing. Journal of Financial Economics, 120(2), 363–382.

    Google Scholar 

  • Saavedra, D. (2018). Syndicate size and the choice of covenants in debt contracts. The Accounting Review, 93(6), 301–329.

    Google Scholar 

  • Sanga, S. (2018). Incomplete contracts: An empirical approach. The Journal of Law, Economics, and Organization, 34(4), 650–679.

    Google Scholar 

  • Santos, J., & Winton, A. (2008). Bank loans, bonds, and information monopolies across the business cycle. Journal of Finance, 63, 1315–1359.

  • Skinner, D. J. (2011). Discussion of “accounting standards and debt covenants: Has the “balance sheet approach” led to a decline in the use of balance sheet covenants?”. Journal of Accounting and Economics, 52, 203–208.

    Google Scholar 

  • WRDS Research Macros. (2010). Wharton Research data services, The Wharton School, University of Pennsylvania, wrds.wharton.upenn.edu. Accessed 12 July 2019.

  • Zhang, J. (2008). The contracting benefits of accounting conservatism to lenders and borrowers. Journal of Accounting and Economics, 46, 27–54.

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Stanford Law School, Stanford, CA, USA

    Colleen Honigsberg

  2. INSEAD, Fontainebleau, France

    Sharon P. Katz

  3. Kennesaw State University, Kennesaw, GA, USA

    Sunay Mutlu

  4. University of Texas at Dallas, Dallas, TX, USA

    Gil Sadka

Authors
  1. Colleen Honigsberg
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Sharon P. Katz
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Sunay Mutlu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Gil Sadka
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Colleen Honigsberg.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix 1A: Provisions to Construct the Pro-Debtor Index

The Pro-Debtor Index consists of six legal practices and standards that reflect the lender’s ability to enforce a debt contract. The six features included in the index are not the only differences between states’ laws but are meant to reflect the differences considered most relevant in the legal literature on lender liability (Honigsberg et al. 2014). The features are described in detail by Honigsberg et al. (2014), but we describe them briefly here. Each state receives a ranking of 1–3 for each feature, where 1 is considered pro-lender and 3 is considered pro-debtor.

  1. (1)

    Does the state enforce predispute jury trial waivers?

It is common for commercial contracts to include a provision stating that, if litigation arises from the contract, the parties will not have the option to litigate in front of a jury (Eisenberg and Miller 2007). Instead, the case can only be heard by a judge. Some states allow these provisions to be enforced, but others do not. The Pro-Debtor Index codes states as 1 if such waivers are enforceable, 2 if the law is uncertain, and 3 if the waivers are unenforceable.

  1. (2)

    Does the state have specialized business courts?

Many states have established specialized business courts to hear commercial disputes. These courts tend to benefit businesses by, for example, increasing the pace of litigation and ensuring that the judge has relevant expertise. From a lender’s perspective, such courts can allow them to enforce a judgment more rapidly—hopefully before the debtor has spent all the money in question. The Pro-Debtor Index codes states as 1 if the state had business courts, 2 if the state had an alternative to the standard business court, and 3 if the state had no relevant business court or similar structure.

  1. (3)

    Does the state recognize the tort of deepening insolvency?

A limited number of states recognize the tort (a wrongful civil act leading to legal liability) of deepening insolvency. This tort allows harmed parties to bring a claim against a party, such as a secured lender, who has wrongfully prolonged the corporation’s life (e.g., a secured lender who has continued to provide funds rather than cause the corporation to file for bankruptcy). Lenders in states that recognize the tort face the prospect of liability under this additional cause of action. The Pro-Debtor Index codes the states that declined to recognize the tort as 1, the states that had not ruled on the tort (and where lenders would thus be concerned that the state would recognize the tort) as 2, and the states that explicitly recognized the tort as 3.

  1. (4)

    Does the state enforce waivers of lender liability?

To avoid claims of lender liability, many lenders require borrowers to sign an agreement waiving all claims the borrower could theoretically bring against the lender. Some states are more comfortable enforcing these waivers than others. For example, some states hold that such waivers are unenforceable if the lender required the borrower to sign the waiver as a condition to a loan renegotiation, whereas other states may enforce the waiver under the same circumstances. The Pro-Debtor Index ranks the states as 1 if the case law surrounding waivers is the most beneficial to lenders, 2 if the case law is moderate, and 3 if the case law poses the highest risk to lenders.

  1. (5)

    Does the state have statutes allowing for its law and forum to be used for commercial contracts of a minimum dollar value?

To provide work for state-licensed attorneys, states are often thought to compete to provide contract law for large commercial contracts (Eisenberg and Miller 2010). One of the first steps for the states that desire to compete is for the state legislature to pass statutes providing that parties to any commercial contract over a minimum dollar value (e.g., $1 million) can use the state’s laws and courts. These statutes can benefit lenders by providing them with greater certainty that the lawsuit will be resolved without unnecessary, time-consuming dispute over the choice of law, forum, or both. Further, these statutes provide lenders with greater certainty that their choice of law will be enforced, allowing them to act in accordance with the relevant law. The Pro-Debtor Index codes states as 1 if the state has both choice of law and choice of forum statutes, 2 if the state has only one statute, and 3 if the state has neither statute.

  1. (6)

    To what extent must a lender act in good faith when taking discretionary action?

Courts will generally allow lenders to take discretionary actions permitted by the contract—such as calling a loan after the borrower has defaulted—even if the action harms the debtor. However, some states are more likely to impose restrictions on the lenders’ discretionary actions if the action will harm the borrower—even if the contract states that the lender can take the action in question (Burton 1994; Gergen 1993). The Pro-Debtor Index ranks the states as 1 if the relevant case law is the most beneficial to lenders, 2 if the case law is moderate, and 3 if the case law poses the highest risk to lenders.

Appendix 1B

Table 9 Pro-Debtor Index
Full size table

Appendix 1C

Table 10 Perceived Litigation Risk
Full size table

Appendix 2

Table 11 Covenant Mix
Full size table

Appendix 3: Variable Definitions

ADV: Advertising expenditures scaled by total revenues. Compustat items: XAD/REVT. Missing values of XAD are set to zero

AGE: Natural logarithm of the number of years a firm exists in Compustat

ALTZ: Altman’s (1968) Z score. Compustat items: 1.2*((ACT-LCT)/AT) + 1.4*(RE/AT) + 3.3*(PI/AT) + 0.6*((PRCC_F*CSHO)/LT) + 0.999*(REVT/AT)

Balance sheet ratio: The ratio of the number of balance sheet covenants over the sum of balance sheet-based covenants and income statement-based covenants.

Balance Sheet Covenant Strictness: The ex ante probability of violation of balance sheet-based covenants in the loan agreement. This measure is calculated and provided by Demerjian and Owens (2016).

BTM: Book-to-market ratio of equity. Compustat items: SEQ/(PRCC_F*CSHO).

DEALSIZE: Natural logarithm of total loan deal size specified in Dealscan.

DecGrid: An indicator variable that takes the value of 1 if the performance pricing grid includes interest decreasing grids, zero otherwise.

DIVYIELD: Dividend yield, calculated as the cash dividends paid divided by the market capitalization. Compustat items: DVC/(PRCC_F*CSHO).

Event Duration: The number of months between the origination of the loan path and the renegotiation event.

Income Statement Covenant Strictness: The ex ante probability of violation of income statement-based covenants in the loan agreement. This measure is calculated and provided by Demerjian and Owens (2016).

LEADSIZE: Natural logarithm of the number of lead creditors that Dealscan indicates participated in the loan deal.

LENDFREQ: Lending frequency, as calculated by the number of loans a borrower received over the last five years according to Dealscan.

LEV: Leverage, calculated as long-term liabilities divided by market value of total assets. Compustat items: DLTT/(AT-SEQ+(PRCC_F*CSHO)).

Loan Duration: The number of months between the origination and the termination of the loan path.

LOSS: Loss indicator, taking the value of 1 if income before extraordinary items is negative, zero otherwise. Compustat item: IB.

MATURITY: Natural logarithm of the maturity of loan deal size specified in Dealscan.

NUMCOV: Natural logarithm of the number of financial covenants reported in Dealscan.

PP: Indicator variable taking the value of one if Dealscan indicates that the loan deal includes performance pricing indicator, zero otherwise.

RATING: S&P domestic long-term issuer credit rating, recoded numerically from 1 to 22, with 1 being “AAA” and 22 being “D.” For firms not rated by S&P, we estimate the ratings following Beatty et al. (2008). Debt rating is first regressed on a set of financial variables, including log of assets, ROA, leverage, dividend indicator, subordinated debt indicator and a loss indicator, with industry and year fixed effects for rated firms. The firm’s financial information is then used to compute a credit rating for each firm in each year. The computed rating values are winsorized at 1 and 22.

RD: Research and development expenditures scaled by total revenues. Compustat items: XRD/REVT. Missing values of XRD are set to zero.

Renegotiation Round: The sequence number of the renegotiation over the loan path. The first renegotiation takes the value of one, and so on.

RETVOL: Natural logarithm of return volatility, calculated over the last 24 months using the CRSP monthly file.

REVOLVER: Indicator variable taking the value of one if Dealscan indicates that the loan deal includes a revolving facility, zero otherwise.

ROA: Return on assets. Compustat items: IB/AT.

SECURED: Indicator variable taking the value of one if Dealscan indicates that the loan deal is secured, zero otherwise.

SIZE: Natural logarithm of the market value of assets. Compustat items: AT-SEQ + (PRCC_F*CSHO).

SPREAD: Natural logarithm of the all in-drawn spread over LIBOR as reported in Dealscan.

TANG: Asset tangibility. Compustat items: PPENT/AT.

Appendix 4: Extending the Sample using WRDS SEC Analytics Suite

First, using the text-parsing macro described in WRDS Research Macros (2010), we search and identify all 10-K, 10-Q, and 8-K filings in the SEC Edgar system with the following 10 terms in capital letters: “credit agreement”, “loan agreement,” “credit facility,” “loan and security agreement,” “loan & security agreement,” “revolving credit,” “financing and security agreement,” “financing & security agreement,” “credit and guarantee agreement,” “credit & guarantee agreement.” We also add two additional terms to this search, following Bozanic et al. (2018): “credit and security agreement” and “credit & security agreement.”

Second, we require the filings above to include the term “table of contents” within 6000 characters after the initial search terms. We allow the term “table of contents” to be case insensitive. WRDS SEC Analytics Suite allows us to associate each filing with the firm identifier (GVKEY) and filing date. Using these two parameters, we link contracts with contract details in Dealscan database, using the Compustat-Dealscan matching table described by Chava and Roberts (2008). We identify 3891 loan deals with matching SEC filing information during the period from 2006 to 2017. As in the work of Bozanic et al. (2018), the frequency of new loan deals significantly decreases around the financial crisis in 2008. These factors explain the relatively low density of observations in the additional sample of 2006–2017 period, as compared to Nini et al.’s (2009) initial sample of 1996–2005 period.

Finally, we also use the WRDS paragraph-parsing macro to identify the jurisdiction of the contract. Specifically, we search for the term “governed by” in the contract and identify cases where it is followed by terms “law,” “laws,” “state,” and “commonwealth” after two lines and parse out these instances.Footnote 28 Then we search for the occurrences of names of the 50 US states in these parsed out texts. If the search algorithm returns nonmatching results or multiple state matches for a given contract, we drop these observations from the sample. We collect 2455 contracts with matching jurisdiction information for the period of 2006–2017. However, as explained in the Sample Selection section above, the sample is reduced because we lack test variables for some contracts.

Appendix 5: Instrumental Variable Analysis

Our instrumental variable analysis uses indicators for the home states of both the lender and borrower, and their interactions, as instruments.Footnote 29 The economic intuition behind the instruments is as follows. The home states of the contracting parties are primary candidates for the state law that will govern the debt contract. This leads to a strong correlation between home states (i.e., instruments) and contract law jurisdictions (i.e., regressors), and we believe this is the primary way through which the home states (i.e., instruments) affect the mix of covenants (i.e., dependent variable). The home state may relate to some cross-sectional characteristics such as industry, but we do not expect that the home state will be associated with other temporal variations in firm-specific characteristics. Moreover, although the home states of the contracting parties are likely to determine the state law governing the contract, this association is not overwhelming. In our sample, only 10% of the contracts are governed by the home states of borrowers, while 30% are governed by the home states of lenders. We believe this eliminates the concern of a spurious strong correlation between the instruments and the endogenous regressor.

Table 12 reports the second stage results of our instrumental variable analysis. We do not report the first stage estimation as we have many instruments as dummy variables.Footnote 30 Overall, when we control for potential endogeneity with instruments, we find similar results to those reported in Table 2, which provides support for our first hypothesis that borrower-lender pairs are more likely to use balance sheet-based covenants when the state contract law is more favorable to debtors.

Table 12 Instrumental Variable Analysis
Full size table

Appendix 6: Changes in State Contract Law for the Same Loan Path

To address loan specific correlated omitted variable concerns, we identify instances in which the same borrower-lender pair changes the jurisdiction that governs the loan contract. The loan renegotiation sample introduced by Roberts (2015) provides us with a suitable setting for this identification. Roberts (2015) selects a random sample of 501 loan tranches and, manually investigating the SEC filings, analyzes the history of these loans to identify major renegotiations to provide descriptive evidence that renegotiation is an important mechanism to dynamically allocate control rights.

Roberts (2015) defines loan paths for each tranche that begin with an origination, include renegotiations, and end with a termination. We focus on the renegotiations where the lender-borrower pair changes some contractual features. During some of these renegotiations, an amended and restated credit agreement is written, which replaces the original contract.Footnote 31 Although rare, we identify instances in which the governing law for the amended and restated credit agreement differs from that in the original contract. This variation provides us with a setting to estimate a change-in-change specification.

After merging our sample with the 501 loan paths provided by Roberts (2015), we have a sample of 63 loan paths with 95 amendment or restatement events that replace the original agreement.Footnote 32 There is a change in the Pro-Debtor Index of the governing state contract law in six of these loan paths. We regress the change in the Pro-Debtor Index on the change in balance sheet ratio. The results are tabulated in Table 13. We include controls for borrower characteristics and event variables (e.g., event duration, loan duration at the time of the event, and renegotiation round), following Roberts (2015). In column 1, the change in the Pro-Debtor Index is positively associated with the change in the balance sheet ratio (t-statistic is 2.65). In columns 2 and 3, we break down the change in the Pro-Debtor Index into indicators of increase and decrease. The results show that a decrease in the Pro-Debtor Index is negatively associated with a change in the balance sheet ratio (t-statistic is −1.86). Similarly, an increase in the Pro-Debtor Index is positively associated with a change in the balance sheet ratio, but the significance level is marginally under the accepted levels (t-statistic is 1.65). These results support our main finding that the pro-debtor state contract law is associated with a higher use of balance sheet-based covenants in the loan contract. However, we caution the reader about the generalizability of the loan subsample test, as the sample size is very limited. (There are only six cases where the contracting parties change the state law governing the contract.)

Table 13 Changes in State Contract Law for the Same Loan Path
Full size table

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Honigsberg, C., Katz, S.P., Mutlu, S. et al. State contract law and the use of accounting information in debt contracts. Rev Account Stud 26, 124–171 (2021). https://doi.org/10.1007/s11142-020-09559-4

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11142-020-09559-4

Keywords

JEL Classification

Search

Navigation