Credit: Maxim_Kazmin/Adobe Stock

Wala Al-Daraji
PhD Candidate in Law, University of Reading
w.al-daraji@pgr.reading.ac.uk

Introduction

Performance bonds are contracts of guarantee used in many industries including construction. In 1977, a performance bond was considered a ‘new business transaction’,¹ however, it was still perceived to be similar to a letter of credit. Lord Denning defined it as ‘a guarantee by a bank that suppliers will perform their obligations under the contract’.² He further described them as ‘promissory notes payable on demand’.³

There are different forms for performance bonds provided by different professional bodies and contract types. One example is the International Chamber of Commerce (ICC) Uniform Rules for Demand Guarantees (URDG).⁴ The use of the URDG in construction is worth considering as it reinforces the autonomy principle, which is fundamental in drafting, execution and judicial interpretation of bonds in many jurisdictions. The URDG is also considered to ‘reflect international standard practice’ and ‘balance the legitimate interests of all parties’.⁵

While on-demand bonds are expected to be paid out without challenge, case law shows exceptions. Contractors are normally unsuccessful in preventing a demand for payment except for fraud. However, the decision in Simon Carves Ltd v Ensus UK⁶ shows that courts may restrain payment without finding fraud. In contrast, the Supreme Court of South Africa, in Aveng-Strabag JV v Sanral [2020], maintained the traditional approach of refusing contractor’s action to restrain payment.

This article will highlight the main features of the URDG that may make it attractive in construction projects and presents a comparison between the South African and the English approach in relation to performance bonds. It will also address when a call upon a bond can be restrained and answer the question whether performance bonds are efficient risk allocation tools taking into account the case law presented.

ICC URDG

The URDG provides for the autonomy principle under Article 5 Independence of Guarantee and Counter-Guarantee.⁷ ‘A guarantee is by its nature independent of the underlying relationship […] A reference in the guarantee to the underlying relationship for the purpose of identifying it does not change the independent nature of the guarantee.’⁸ In the construction industry, the underlying relationship is the main contract between the contractor and the employer. In addition, Article 5 provides that the guarantor is only bound by the relationship between itself and the beneficiary and is not concerned with claims arising from other contracts. Thus, the autonomy principle is catered for under the URDG. Furthermore, payment to the beneficiary is protected from claims or counterclaims made under the underlying contract.

Another important provision under the URDG is that demand guarantees are documents only, thus guarantors are not concerned with performance under the main contract or goods provided.⁹ Article 15 deals with Requirements for Demand and it requires that, when making a demand, a beneficiary shall provide documents as specified in the guarantee. There is no requirement to check whether contractual performance took place or not. Arguably, this makes it easy to cash a bond based only on documents without getting into the nature of the dispute between the parties. The URDG also requires a statement by the beneficiary explaining breach in the underlying contract; however, this requirement can be waived if it is excluded from the guarantee. Thus, while the autonomy principle is valued, there is a requirement to inform the guarantor of what went wrong in the underlying relationship. Article 27 provides indemnification to the guarantor from liability for the form of document, the description of performance, goods or services, and good faith of any person issuing or referred to in any document presented to it.¹⁰ However, it is worth highlighting that fraud is not mentioned under the rules and the guarantor is not exempt for failing to act in good faith.¹¹

Performance bonds in South Africa

In JV Aveng Strabag v SANRAL,¹² the contractor, a joint venture (JV), claimed force majeure due to civil unrest that stopped work on site for 84 days. The employer, SANRAL, denied force majeure and instructed the JV to resume works. The JV did not return to site and the dispute was referred to arbitration. The FIDIC Contract (Red Book) was used by the parties. It defined four events under sub-clause 4.2 which allowed SANRAL to make a claim under the performance bond.¹³ The events included failure by the JV to extend the validity of the bond, to pay the employer an amount due, to remedy a default and circumstances that entitle the employer to terminate. The last event, or rather, category, arguably provides the employer with room to use the bond more than the others.¹⁴ Circumstances entitling termination would include breach of the contract’s express and implied terms. For example, late delivery of specialist equipment may allow termination.

When the JV asked SANRAL for assurances that the second will not call upon the bond until arbitration proceedings are complete, SANRAL confirmed that it intends to call upon the guarantee. Accordingly, the JV applied for an interlocutory interdict which was dismissed but then allowed to appeal. In the appeal, Judge Makgoka confirmed that the autonomy principle of the guarantee from the main contract is recognised in South African law. J Makgoka referred to Edward Owen where Lord Denning emphasised the principle by explaining that a bank is not concerned with the relationship between the supplier and customer, whether there is default or not. The appeal was dismissed as the JV failed to show that the parties have intended that the employer is entitled to payment prior to determination of any dispute between them. The existence of an ongoing arbitration did not affect the judge’s decision as a dispute in the underlying contract does not automatically translate to stopping a beneficiary from calling upon the performance bond. The autonomy principle was also recognised in Loomcraft.¹⁵ Scott AJA referred to it as ‘documentary credit’¹⁶ and emphasised that fraud is the only exception that would allow a bank not to honour a performance guarantee.

Kwikspace¹⁷ is another South African case. In this case, Australian law was the law of contract. Cloete JA confirmed that under Australian common law, a contractor can restrain an employer from making a call on a bond if it proves that there would be a breach of the main contract. This arguably means that the autonomy principle is not absolute in Australia as a performance guarantee may be subject to a contractual qualification in the main contract.¹⁸ Thus, prior to cashing a bond, the main contract conditions must be fulfilled first. This undermines the liquidity of the bond as the beneficiary may be denied sums based on the main contract conditions. It also makes it a less effective guarantee in some jurisdictions than others. The URDG does not permit non-documentary conditions; the only condition it allows is either a date or lapse of a period of time.¹⁹

Performance bonds in England

Turning to England, the case of Edward Owen Engineering v Barclays Bank is considered an important authority to review.²⁰ The claimant brought action to restrain the bank from paying its performance bond, which stipulated ‘payable on demand without proof or conditions’.²¹ The claimant argued that as there was no default, the bond cannot be cashed. They further argued that bad faith exists where a bondholder cashes a bond knowing there is no default. The appeal failed as sufficiency of performance was not required. Browne LJ quoted Lord Denning: ‘the bank here is simply concerned to see whether the event has happened upon which its obligations to pay has arisen’.²² The decision concurs with other decisions where injunctions to stop payment are refused except for fraud.

In contrast to Edward Owen, a recent case shows that a contractor can succeed in restraining an employer from making a call on a performance bond. In Simon Carves v Ensus,²³ an injunction was granted on the basis that there was a reasonably good arguable case that the bond had become null and void.²⁴ The case involved a process plant construction project under a contract incorporating the Institution of Chemical Engineers General Conditions of Contract. Issues arose in relation to responsibility for rectifying defects raised in enforcement notices issued by the Environment Agency.

The wording of the bond was of an unconditional bond, stating ‘The Bank hereby irrevocably and unconditionally undertakes to pay to the Purchaser.’²⁵ J Akenhead highlighted the problem of seeking to restrict a call on a bond where there is no fraud, stating: ‘There has been little jurisprudence on the circumstances which arise in which there are contractual provisions between contractor and purchaser/employer which impose restrictions or which prevent calls being made on bonds or letters of credit.’²⁶ The special conditions of the contract provided that under sub-clause 3.7 the bond shall become null and void upon the issue of the Acceptance Certificate.²⁷

J Akenhead concluded that if the underlying contract prevents the beneficiary from making a demand under the bond, the court will restrain such demand. He then went on to say that the court can decide to restrain the beneficiary either as it is in breach of the underlying contract or because it is fraud ‘in that the beneficiary is seeking to call on the bond when it knows or can be taken to know that the underlying contract forbids it from doing so’.²⁸ Thus, the illegitimacy of making a call upon a bond can qualify as fraud. The action succeeded because upon issuing the Acceptance Certificate by the employer, the bond has become null and void in respect of ‘any pending or previously notified claims’.²⁹

When can a call upon a performance bond be restrained?

As mentioned above, the main exception where a performance bond may not be called is fraud. The strict rule was established in the United States case of Sztejn v J Henry Schroder Banking Corporations.³⁰ The case emphasised separability of the letter of credit from the main contract between the seller and the buyer; that is, the autonomy principle. The bank is concerned with documents presented rather than the actual goods or services. Performance or lack thereof under the underlying contract was not the test. The case further stressed that a letter of credit is important for trade and it is important to preserve its efficiency. The argument for honouring letters of credit as important instruments in international commerce was also highlighted in other cases such as R D Harbottle (Mercantile) v National Westminster Bank.³¹ J Kerr argued: ‘The machinery and commitments of banks are on a different level. They must be allowed to be honoured, free from interference by the courts. Otherwise, trust in international commerce could be irreparably damaged.’³²

Different authorities show that a contractor seeking an injunction to stop a call upon payment of a bond has to overcome three barriers: knowledge of the beneficiary’s call, fraud and bank knowledge of fraud.³³ The contractor needs to be aware of a potential call on the bond, which may not be easy to know. There is a game theory interpretation here as there is information asymmetries unless the employer makes an explicit ‘threat’ to call upon the bond. Otherwise, the contractor arguably has no way of knowing if a call will be made or not.

Eveleigh LJ said: ‘in principle I do not think it is possible to say that in no circumstances whatsoever, apart from fraud, will the court restrain the buyer. The facts of each case must be considered’.³⁴ Thus, fraud is not the only exception but also ‘if the contract is avoided or if there is a failure of consideration between buyer and seller’. The underlying contract can be relied upon to restrain a buyer from calling upon the bond as witnessed in Simon Carves and Australian case law. Another area where the courts may grant a beneficiary making a call on a bond is where there is a risk of assets being moved to another jurisdiction to the detriment of the other party. In Mavera Compania Naviera SA v International Bulkcarriers SA,³⁵ the Mavera principle was established, which provides for securing assets of the defendant so that the claimant is not left with nothing.

Are performance bonds efficient risk allocation tools?

Having reviewed case law from England and South Africa, we now turn to the efficiency of bonds. Are bonds efficient tools to transfer risk between contractors and employers in construction? The risk in question is not only that of a contractor defaulting but also the risk of bringing a contractor to ruin. Lundberg’s collective theory of risk can be used to determine a contractor’s probability of ‘ruin’³⁶ if all or some of its performance bonds are called upon within a period of time. However, given the different parameters involved in each construction contract, such probability may not be easy to quantify. Pareto’s efficiency has one criterion, which is ‘any change that puts one member of society in a better position without making somebody else worse off is a Pareto improvement’.³⁷

A performance bond is arguably beneficial for the employer as it provides a security of usually 10–20 per cent of the contract value, which in theory is cashable on demand regardless of the existence of a dispute. However, it is arguably inefficient because contractors have to provide an equivalent security to the bank for the duration of the project. Accordingly, the contractor has to find 10–20 per cent of its contract value and place it as a security with the guarantor. This places a contractor working on multiple projects under financial pressure as it has to provide 10–20 per cent for each contract value it is working on as a security for the bank. There is an opportunity cost for contractors to use these sums more efficiently and innovatively instead of having them tied down to be released and then blocked for the next project. Performance bonds exist in a string of guarantees, between banks of suppliers and buyers, suppliers and their banker, and buyers and their banker. This increases transaction costs for both suppliers and buyers as they negotiate the terms of such guarantees among themselves and with their respective bankers.

One may ask why we are still using bonds despite their impact on transaction costs. Bonds are there for two reasons: security and risk allocation as to ‘who shall be out of pocket pending resolution of a dispute’.³⁸
A performance bond could be seen as a ‘liquidated damages’ where there is a substantial breach of contract. It provides the buyer with a quick remedy as opposed to suing the supplier as long as the value of the bond is sufficient to cover the damage. Lord Denning warned that it could even be used when the breach is ‘insubstantial or trivial’,³⁹ in which case it can be classified as a penalty. Thus, the possibility of using performance bonds as liquidated damages is ‘so real’.⁴⁰ Accordingly, construction contractors need to allow for such possibility in their price. Naturally, given the low profit margins of construction contractors, there is pressure on contractors to submit a low bid to win the works. However, such low price may prove to be insufficient to deal with such high possibility. Moreover, calling a bond carries ‘a very real risk of damage to the commercial reputation, standing and creditworthiness’ of a contractor.⁴¹ This may affect their ability to pre-qualify for other tenders as well as being able to obtain finance.

Conclusion

While performance bonds are here to stay, their efficiency is questionable and their transaction costs are arguably high. Comparative case law from South Africa and England shows similarities and differences in judicial approach when dealing with actions to restrain a call upon a bond by a beneficiary. Despite the similarities, there is still a grey area as to when an action to restrain a call in the absence of fraud or Mavera can be successful. The decision in Simon Carves may give contractors a higher chance of success in restraining action subject to the particulars of each case. The costs involved in restraining calls upon bonds cannot be ignored given the contracting industry’s typically low profit margins. The autonomy principle is arguably undermined when the main contract decides whether a bond can be cashed or not. Judicial clarity on what constitutes grounds to restrain calling upon a bond is needed. Moreover, the use of documents-only bonds such as the URDG may limit litigation between contractors and employers as disputes between the parties within the underlying contract have no say in restraining or granting a call upon a performance bond. A balance is needed between allowing actions to restrain a call upon a bond on the one hand and preserving the important role of bonds in international trade on the other.

Notes

1 Edward Owen Engineering v Barclays Bank [1978] QB 159, 164

2 Ibid, 166.

3 Ibid, 170.

4 ICC Academy, ICC Uniform Rules for Demand Guarantees (URDG 758) eBook Summary at https://icc.academy/urdg-758 accessed 9 October 2022. The URDG was officially endorsed by the United Nations Commission on International Trade Law (UNCITRAL) in 2011 and is also recognised by FIDIC.

5 The World Bank, Public Private Partnership Legal Resource Center at https://ppp.worldbank.org/public-private-partnership/library/icc-uniform-rules-demand-guarantees accessed 9 October 2022.

6 Simon Carves Ltd v Ensus UK Ltd [2011] EWHC 657 (TCC), 2011 WL 1060082.

7 ICC Uniform Rules for Demand Guarantees (URDG 758) www.cipcic-bragadin.com/wp-content/uploads/2015/09/ICC-URDG-758.pdf accessed 9 October 2022.

8 Ibid, Art 5(a).

9 Ibid, Art 6.

10 Ibid, Art 27(a), (c)–(d).

11 Ibid, Art 30 Limits on Exemption from Liability.

12 Joint Venture between Aveng (Africa) (Pty) Ltd and Strabag International GmbH v South African National Road Agency Soc Ltd and Another (Case no 577/2019) [2020] ZASCA 146 (13 Nov 2020).

13 Ibid, 19.

14 Ibid.

15 Loomcraft Fabrics CC v Nedbank Ltd and Another (70/94) [1995] ZASCA 127.

16 Ibid, 5.

17 Kwikspace Modular Buildings Ltd v Sabodala Mining Co Sarl and Another [2010] ZASCA 15; 2010 (6) SA 477 (SCA).

18 Fletcher Construction Australia Ltd v Varnsdorf Pty Ltd [1998] 3 VR 812.

19 See n 7 above, Art 7.

20 See n 1 above.

21 Ibid, 159.

22 Howe Richardson Scale Co Ltd v Polimex-Cekop [1977] EWCA Civ J0623-1

23 See n 6 above.

24 Ibid.

25 Ibid.

26 Ibid, 29.

27 Ibid.

28 Ibid, 34.

29 Ibid, 37(a).

30 (1941) 31 NYS 2d 631.

31 [1978] QB 146.

32 RD Harbottle (Mercantile) Ltd v National Westminster Bank Ltd [1978] QB 146, 156.

33 Issaka Ndekugri, ‘Performance Bonds and Guarantees: Construction Owners and Professionals Beware’ [1999] 125 Journal of Construction Engineering and Management 6, 428–36.

34 Potton Homes Ltd v Coleman Contractors Ltd [1984] 28 BLR 19 CA, 28.

35 [1975] 2 Lloyd’s Rep 509.

36 Karl Borch, ‘The Theory of Risk’ [1967] 29 Journal of the Royal Statistical Society 3, 432.

37 Klaus Mathis Efficiency Instead of Justice? Searching for the Philosophical Foundations of the Economic Analysis of Law translated by Deborah Shannon (Springer 2009), 33.

38 Fletcher Construction Australia Ltd v Varnsdorf Pty Ltd [1998] 3 VR 812, 826.

39 See n 1, 170.

40 Ibid.

41 See n 6, 41.