Interfaces with Other Disciplines
Risk analysis with contractual default. Does covenant breach matter?

https://doi.org/10.1016/j.ejor.2013.04.047Get rights and content

Highlights

  • Covenant violations are an underhand and often overlooked risk.

  • We introduce a methodology for modeling the consequences of covenant breach.

  • Results show statistically significant difference in risk profiles.

  • Risk measures need to be readjusted to account for default risk.

Abstract

Mergers and acquisitions (M&A), private equity and leveraged buyouts, securitization and project finance are characterized by the presence of contractual clauses (covenants). These covenants trigger the technical default of the borrower even in the absence of insolvency. Therefore, borrowers may default on loans even when they have sufficient available cash to repay outstanding debt. This condition is not captured by the net present value (NPV) distribution obtained through a standard Monte Carlo simulation. In this paper, we present a methodology for including the consequences of covenant breach in a Monte Carlo simulation, extending traditional risk analysis in investment planning. We introduce a conceptual framework for modeling technical and material breaches from the standpoint of both lenders and shareholders. We apply this framework to a real case study concerning the project financing of a 64-million euro biomass power plant. The simulation is carried out on the actual model developed by the financial advisor of the project and made available to the authors. Results show that both technical and material breaches have a statistically significant impact on the net present value distribution, and this impact is more relevant when leverage and cost of debt increase.

Introduction

Risk analysis is a central tool for supporting investments decision-making. The main outcome of a risk analysis is the distribution of the valuation criterion (henceforth, risk profile). From the risk profile, analysts estimate all risk measures. Thus, it is crucial that the risk profile reflects as accurately as possible the reality of the investment at hand.

Covenants are contractual clauses that impose on a borrower either obligations to do something (positive covenants) or to refrain from doing something (negative covenants) Smith (1993). Covenants are set tightly ex-ante (i.e., before the closing of the loan agreement), after extensive negotiation between lenders and shareholders (see the bondholder-stakeholder conflict in Smith and Warner, 1979). Because covenants are defined to provide lenders with early warning signals and safety buffers, their breach does not necessarily occur in states of actual financial distress of the borrower Chen and Wei (1993). This fact prompts the question of whether covenant breach affects an investment’s risk profile. The answer is still debated, yet more importantly, it has not been the subject of a direct quantitative investigation.

Covenant breach challenges standard risk analysis practices, raising three main questions: (1) How likely is covenant breach? (2) Given a breach (either a technical default or a material breach as defined below), what consequences might sponsors and lenders face? (3) How significant is the impact on risk profiles and, consequently, on risk measures?

We design a quantitative model of the consequences of covenant breach and apply it to the special case of project finance where the borrower is represented by a Special Purpose Vehicle (SPV) Esty (2003). The model has two main features. It takes into account the important distinction between technical and material breach, and it differentiates the perspectives of sponsors and lenders. A breach is material when it corresponds to a “material” event SEC, 1999, Novo, 2007, that is, lenders consider the violation capable of disrupting the orderly continuation of the deal and force the borrower to default. Conversely, if the event is not deemed material, lenders “waive” the contractual breach and we are in the presence of a technical default. The determination of whether a default is material or not is achieved by carrying out a “materiality test”. The materiality test requires the ex-ante definition of a range of values of the cover ratios discussed in Section 3. At the upper end of this range there is a preset value that triggers technical default but allows creditors and the SPV shareholders to continue managing the transaction with enforced monitoring or renegotiated loan terms. At the lower end there is a final value below which the breach is considered material. At this point, creditors consider the loan subject to resolution; they accelerate the loan and force the deal into bankruptcy without the possibility of renegotiation. In this respect, covenant violations expose sponsors to a risk, which, if overlooked, can lead to grave consequences for both shareholders and lenders. The survey by Davison et al. (2010) reveals material violations in around 8% of the project finance deals analyzed in their paper. In these circumstances, lenders are in a position to take over control of the project, and can continue the initiative with the objective of recovering the amount lent to the SPV Davison et al. (2010, p. 44). In the case of a technical breach, our model accounts for the impact on the investment cash flow determined by the change in the debt schedule repayment. In fact, lenders impose a dividend lock-up on sponsors and require them to use all the available cash in the year of technical breach to accelerate debt repayment (cash sweep). In the case of a material breach, the model accounts for the shareholders’ loss of equity cash flows subsequent to the lenders’ enforcement of the security package. For lenders, the model accounts for the probability of workout and recovery of the outstanding loan amount.

We translate the model into a simulation methodology that nests into the traditional Monte Carlo simulation.

We apply the simulation procedure to a real project finance tra nsaction involving a recent biomass power plant located in Italy worth around 64 million euro which underwent financial closing in 2009 with General Electric-Interbanca as lead arranger. For this project, the authors had access to the financial model used by sponsors and the mandated lead arranging bank in their evaluation of the deal. The availability of the financial model allows us to examine quantitative results relying on real data and not on fictitious or simulated deals. Results show that the effect of covenant breach on the cash flow distributions to the SPV’s sponsors is negligible if the violations are not material. Conversely, in the case of material breach, the cumulative distribution function of the equity net present value (NPV) exhibits a fat left tail, with a considerable change in the value of risk measures. Indeed, our findings confirm that overlooking the consequences of a material breach can lead both shareholders and lenders to underestimate the risks involved in the transaction. The impact on the risk profile is statistically significant and the values of the risk measure (Conditional Value at Risk or CVaR) can be remarkably different from those obtained in an analysis that fails to take into account the consequence of a covenant breach.

We think our paper can contribute to the existing literature on investment risk analysis from different standpoints.

First, available papers on covenants have analyzed contractual clauses almost exclusively in the context of corporate finance settings, i.e. in case of bond or loan contracts between lenders and already existing firms Aghion and Bolton, 1992, Beneish and Press, 1995, Chava and Roberts, 2008. However, much less is known about the importance and the role of covenants included in lending agreements designed for project finance, and more generally for structured deals. The typical project finance transaction is a fully self-contained, one-time financing event with a definite economic life-cycle Gatti et al. (2013). With such a deal, the previous lending relations between lenders and the SPV shareholders are much less important than the soundness of the stand-alone deal to be financed. These characteristics are true for no other corporate financing sample, making project finance deals ideal to study the disciplining role of covenants in isolation from other borrower-specific factors.

Second, we discuss the use of covenants for a significant segment of international capital markets which is not extensively covered in previous theoretical and empirical papers. Thomson Reuters reports that syndicated loans amounted to US$3.9 trillion at the end of 2011, of which US$214.5 bn were PF loans. These figures compare with US$180 bn of private equity invested globally in 2010 and with global securitization issuance of mortgage and asset backed securities of US$799 bn at the end of 2010.

Third, we design a methodology that is useful to model the effect of technical and material breach of covenants for sponsors and lenders when deciding on individual investment projects. Knowledge of the expected frequency of covenant violations provides analysts with a degree of confidence as to the project’s ability to sustain stresses. This information can then be used by lenders to fine-tune the loan tenor, the spread and the debt/equity ratio. Our methodology can also help creditors assess unexpected loss and allocate equity capital in accordance with Basel II and Basel III rules (Basel Committee on Banking Supervision, 2009).

The remainder of the paper is organized as follows. Section 2 provides the motivation for the paper, discussing covenants included in credit agreements typically used by lenders and sponsors in project finance. Section 3 details the most often used financial covenant for structured deals, the Debt Service Cover Ratio (DSCR). Section 4 proposes a cursory review on risk measures and risk analysis. Section 5 presents our model. Section 6 illustrates the simulation procedure. Section 7 introduces the case study analysis and show the results. Section 8 concludes.

Section snippets

Covenants: an overview

Covenants are defined as “... supplemental obligations of the borrower in addition to the basic obligation to repay the lenders the amount due on the scheduled maturity dates [...] These supplemental obligations may be either correlated to loan repayment – as in the case the borrower does’t take certain actions that will hamper debt repayment at the scheduled dates – or required by lenders in order to monitor their credit investment and verify that it is being managed properly” Novo (2007, p.

The debt service cover ratio

A standard categorization of covenants is not available in the existing literature. However, most authors agree on the broader categories of positive, negative, and financial covenants Rosenbaum and Pearl, 2009, Smith, 1993. In the financial covenants category, Demerjian, 2007, Nini et al., 2009 distinguish between cover ratios, current ratios/liquidity ratios, leverage ratios, gearing ratios, and net worth ratios. Demerjian (2007) indicates that in highly leveraged transactions (like project

Risk analysis and risk measures: a cursory review

The term “risk analysis” originates in the seminal works of Hillier, 1963, Hertz, 1964, Hillier, 1965, Van Horne, 1966, Wagle, 1967. These authors propose the use of the Monte Carlo simulation to obtain the distribution of an NPV or IRR. This has become best practice in the analysis of industrial investments and business planning see Carmichael and Balatbat (2008). More recently, the term “probabilistic sensitivity” has also come into use as a synonym for risk analysis Hazen and Huang (2006).

A model of covenant breach

Before introducing the model, let us state some notation conventions. We consider an investment project that develops over T periods. For simplicity of notation, the construction phase is assumed to be in period t = 0, while operation takes place from periods 1 to T. We let A0 denote the investment cost and Idc represent the interest capitalized during construction. The investment is financed through equity and debt, with ℓ denoting the debt proportion to A0. The symbols FCFt, It, Pt and TL

Simulation procedure and risk analysis insights

Finally we come to how to nest the modeling of the consequences of default described above in the context of the Monte Carlo simulation. In this section, we outline how our conceptual model can be turned into a simulation procedure. The first step consists of running a probabilistic simulation of the financial model registering the values of both lenders and shareholders’ criteria at the end of each simulation (Fig. 1). Then, the examination of covenant breach at scenarios n = 1, 2, … , N starts.

Project background and economic rationale

In mid 2007, Bonollo Distilleries,one of the most renowned high-end liquor producers in Italy, started studying the project of a biomass plant to be annexed to its already existing distillery. According to Bonollo Distilleries’ management, the plant was needed to reduce both energy costs and polluting emissions and to diversify the energy sources currently employed by the factory.

At that time, Bonollo Distilleries already used some biomass energy to cover approximately 30% of its heat energy

Conclusions

We investigated the consequences of including covenant breach in the risk analysis of large industrial projects, and provided an objective view on their effect and significance. We have analyzed both theoretical and managerial implications of the violation of financial covenants included in debt contracts. To achieve our objective, we developed both a conceptual model and a simulation procedure. The model accounts for the different perspectives of sponsors and lenders and distinguishes the

Acknowledgements

We wish to thank Roman Slowinski (the Editor-in-Chief) for the editorial attention and the anonymous referees for the careful and very perceptive comments on the original version of the manuscript. We would also like to thank Andrea Resti, Andrea Sironi, Carlo Chiarella, Giovanni Puopolo and Alessandro Sbuelz for useful comments on earlier drafts of the paper. We are also grateful to Carefin Bocconi Research Center for generous financial support. We alone are responsible for any remaining

References (74)

  • R.T. Rockafellar et al.

    Conditional value-at-risk for general loss distributions

    Journal of Banking and Finance

    (2002)
  • C.J. Smith et al.

    On financial contracting: an analysis of bond covenants

    Journal of Financial Economics

    (1979)
  • A. Sweeney

    Debt covenant violations and managers’ accounting responses

    Journal of Accounting and Economics

    (1994)
  • Achleitner, A.K., Braun, R., Hinterramskogler, Tappeiner, F., 2010. Structure and Determinants of Financial Covenants...
  • P. Aghion et al.

    An incomplete contracts approach to financial contracting

    Review of Economic Studies

    (1992)
  • G.J. Alexander et al.

    A comparison of VaR and CVaR constraints on portfolio selection with the mean-variance model

    Management Science

    (2004)
  • T.W. Anderson et al.

    Asymptotic theory of certain goodness-of-fit criteria based on stochastic processes

    Annals of Mathematical Statistics

    (1952)
  • G.E. Apostolakis

    The concept of probability in safety assessment of technological systems

    Science

    (1990)
  • P. Artzner et al.

    Coherent measures of risk

    Mathematical Finance

    (1999)
  • Basel Committee on Banking Supervision, 2009. Strengthening the Resilience of the Banking Sector. Bank for...
  • M. Beneish et al.

    The resolution of technical default

    The Accounting Review

    (1995)
  • J. Berkowitz et al.

    Evaluating value-at-risk models with desk-level data

    Management Science

    (2011)
  • Blanc-Brude, F., Goldsmith, H., Valila, T., 2006. Ex ante Construction Costs in the European Road Sector: A Comparison...
  • V. Bonetti et al.

    Offtaking agreements and how they impact the cost of funding for project finance deals: a clinical case study of the Quezon Power Ltd., Co.

    Review of Financial Economics

    (2009)
  • D.G. Carmichael et al.

    Probabilistic DCF analysis and capital budgeting and investment – a survey

    Engineering Economist

    (2008)
  • S. Chava et al.

    How does financing impact investment? The role of debt covenants

    Journal of Finance

    (2008)
  • K. Chen et al.

    Creditors’ decisions to waive violations of accounting-based debt covenants

    The Accounting Review

    (1993)
  • F. Corielli et al.

    Risk shifting through nonfinancial contracts: effects on loan spreads and capital structure of project finance deals

    Journal of Money, Credit and Banking

    (2010)
  • C. Crnkovic et al.

    Quality control

    RISK

    (1996)
  • D. Cumming et al.

    Advice and monitoring in venture finance

    Financial Markets and Portfolio Management

    (2007)
  • D. Cumming

    Contracts and exits in venture capital finance

    Review of Financial Studies

    (2008)
  • Dailami, M., Hauswald, R., 2003. The Emerging Project Bond Market: Covenant Provisions and Credit Spreads. Working...
  • Davison, A., Keisman, D., Kelhofer, K., 2010. Default and Recovery Rates for Project Finance Bank Loans, 1983–2008,...
  • Demerjian, P., 2007. Financial Ratios and Credit Risk: The Selection of Financial Ratio Covenants in Debt Contract....
  • C. Demiroglu et al.

    The information content of bank loan covenants

    Review of Financial Studies

    (2009)
  • H. Elsinger et al.

    Risk assessment for banking systems

    Management Science

    (2006)
  • Benjamin. Esty

    Structuring loan syndicates: a case study of the Hong Kong disneyland project loan

    Journal of Applied Corporate Finance

    (2001)
  • Cited by (19)

    • Risk-based optimization of the debt service schedule in renewable energy project finance

      2021, Utilities Policy
      Citation Excerpt :

      Lenders and financial institutions may set these ratings as a target in negotiations with sponsors. Studies primarily have focused on the DSCR and loan life coverage ratio to control creditor financial concerns (e.g. (Xueqing, 2005; Yun et al., 2009; Borgonovo and Gatti, 2013; EsteveBorràs et al., 2019)). Gatti (Gatti et al., 2007) has stated that the consideration of credit risk in project finance transactions is of particular importance for lenders.

    • Bayesian updating of solar resource data for risk mitigation in project finance

      2020, Solar Energy
      Citation Excerpt :

      Furthermore, by estimating the probability distribution of the debt service coverage ratio (DSCR) generated via Monte Carlo simulations (MCS), we introduce the conditional value-at-risk (CVaR) as a controlling risk measure to differentiate between technical and material defaults. Technical default can frequently be happening during the loan life of the project when an agreed loan covenant between the SPV and lenders is violated, as long as the SPV is not in severe financial distress (Borgonovo and Gatti, 2013). In this situation, the project could be continued only if the lenders consider the covenant breach is not a material event.

    • Portfolio optimization under loss aversion

      2016, European Journal of Operational Research
      Citation Excerpt :

      They show that investors are loss averse: they consider the deviations of their terminal wealth as gains and losses starting from a reference level θ , and they react differently to gains than to losses. There is a variety of reasons triggered by real-life constraints explaining how one ends up having a critical level of potential losses, see for example Borgonovo and Gatti (2013) who investigated the consequences of including covenant breach in the risk analysis of large industrial projects, and provided an objective view on their effect and significance. There are many papers dealing with loss aversion: Benartzi and Thaler (1995) and Barberis, Huang, and Santos (2001) use numerical techniques to solve the portfolio problem of loss averse investors, Gomes (2005) provides an exact solution in a model with two states of the world, Berkelaar, Kouwenberg, and Post (2004) extend these results toward closed form solutions assuming a complete markets setting, Lien (2001) studies the implications of loss aversion for futures hedging, Zakamouline and Koekebakker (2009) show that a decision maker exhibits more than loss aversion, Fortin and Hlouskova (2011) and Best, Grauer, Hlouskova, and Zhang (2014) study the asset allocation for the particular case of linear loss-averse investor.

    • Investment timing, debt structure, and financing constraints

      2015, European Journal of Operational Research
      Citation Excerpt :

      Thus, the more severe the debt issuance bounds are, the lower are the credit spreads and default probabilities. These links between risk and return are the same as those traditionally suggested in the literature and also match the results by Gomes and Schmid (2010) and Borgonovo and Gatti (2013). The next characterizes the properties of financing constraints.

    • Risk transfer in project finance loans for toll road using credit default swaps

      2023, Journal of Financial Management of Property and Construction
    View all citing articles on Scopus
    View full text