Should Law Interfere with Failed Negotiations? Notes on Economic Analysis of Pre-contractual Reliance.

Tereza Nováková

1. Introduction

The modern law of contract is based on the freedom to choose whether or not to enter into a contract as well as the freedom to choose the contracting partner and freely negotiate the terms of such a contract.

However, before a binding contract is reached, there is often a long period when parties negotiate over the terms and make reliance investments that would increase the value of the project if the contract were made. If the contract is agreed between the parties, it will stipulate how to divide the surplus from the transaction, taking into account the reliance investments made by one or both parties. If negotiations break down before a binding contract is made, however, the question arises if one party should be held liable for the loss the other party suffered in reasonable reliance that the contract will be entered into.

Since the legal rules of liability for failed negotiations affect the willingness of parties to make reliance investments, I first analyse the incentives of the parties to invest in the future transaction under the different liability regime. In this respect, I consider how the legal rules should be designed to produce a socially optimal outcome. As a normative criterion of social optimality, I use efficiency.[1]

Secondly, using the conclusions from the economic analysis of the different liability regimes and their incentives for the parties’ behaviour during negotiations, I compare the Czech liability regime for failed negotiations with the Dutch one and discuss which approach is more socially desirable. I chose these two jurisdictions since both seem to protect good faith during negotiations, but the interpretation of such duty differs. I will suggest that such difference may affect the incentives of the parties to make reliance expenditures during negotiations.

2. Efficient reliance decisions

Complex, commercial contracts often require long negotiations and costly preliminary actions that increase the surplus from the future contract. Such actions are thus considered investments in the future transaction should the contract be concluded. The potential surplus from the contract is then divided between the parties according to their bargaining powers.[2] If contract is not reached, the reliance investment is fully wasted or it has only some scrap value for the investing party.[3]

When deciding whether to make a reliance investment, a party takes into account the amount of such investment, the probability that agreement on the contract will be reached, the increase of the surplus resulting from such investment, and the scrap value of the investment (if any). For the reliance investment to be efficient, it must be less than its expected benefit. Mathematically, it can be expressed as follows: C < * p + V * (1-p), where C is the cost of reliance investments, S is the surplus from the transaction, p is the probability that the contract will be reached, and V is the scrap value of the investment.[4] We see that the more likely the contract is concluded, the more likely the reliance costs will still be efficient, all else being equal.

Let’s assume that only one party makes the pre-contractual investment and that the bargaining position of both parties is equal.[5]

  • No liability regime

The first possible legal regime is no liability for the failed negotiations. In other words, if agreement on a contract is not reached, the investing party bears its own costs incurred during the negotiations regardless of the reason for such failure. According to Bebchuk and Ben-Shahar[6], under such a regime, the parties will not engage in effective reliance. They will be naturally incentivised to underinvest in reliance, or overinvest in the case of information asymmetry, as will be explained further.

The economic explanation is as follows: The expected return of the investing party from the transaction is half of the surplus (given our assumption of equal bargaining powers) less its investment costs, and such return is discounted for one period[7], because there is one period between its investment and the future returns from such investment.[8]

Since we assume that one party makes reliance investments while the other does not, it is possible that an expected return of the investing party (half of the surplus less its investment costs) is negative even if the expected value of the project as a whole is positive.[9]

In such case, the investing party would not pursue such a transaction despite the fact that it would be socially desirable to do so, since each party only takes into account its own return from the transaction. Given the assumption of equal bargaining positions, the party investing nothing (e.g. the buyer) does not take into consideration the sunken investment costs of the seller. This creates a ‘hold up’ problem. It means that the seller invests less than what is optimal because it expects to capture only a part of the benefit gained (mainly) from its investments, in other words, it expects to be held up.[10]

Under this regime, the transaction that has a positive value might not be pursued and the natural incentive to make a reliance investment is suboptimal.[11]

Symmetric information

Let’s now assume that the information of both parties is symmetric, i.e. both have the same information about the parameters influencing the reliance decision.[12] In such case, it is very likely that the investing party will be incentivised by its contracting partner to engage in efficient reliance, because the transaction increases their joint welfare.[13]

The non-investing party will be aware of the lack of natural incentive to engage in efficient reliance (the problem of underinvestment) and thus will be willing to pay part of the efficient costs to the other party as long as its return from the transaction stays positive.

Eventually, under the assumption of information symmetry, the reliance decisions of the investing party will be efficient, and there is no need to incentivise the parties by any special liability rules.

Asymmetric information – the investing party has inferior information

If the investing party has less information about some parameter influencing its reliance decision, its decision will not be efficient. Let’s assume that it reasonably believes, due to the behaviour of the non-investing party[14], that the probability of reaching the contract is higher than it really is.[15] Let’s say that the actual probability is 40% but the non-investing party behaves in a way that makes the other party believe that agreement on the contract will be reached with a probability of 80%. The investing party is thus willing to invest more than it would have (to overinvest) if it had known the real probability. Such decision is again socially suboptimal.

  • Strict liability regime

Under a strict liability[16] regime, one party must reimburse the other for the costs incurred during negotiations even if it was not the one breaking off the negotiations. Under such a regime, the parties will overinvest.[17] It is because the investing party does not take into consideration the probability that the contract will be formed when making a reliance decision since it is able to reimburse all costs if no agreement is reached. Thus, the absence of risk[18] that agreement on the contract will not be reached incentivises the investing party to invest above the optimal level.[19]

  • Optimal liability regime

It has been argued that strict liability for failed negotiations is socially undesirable since it only leads to overinvestments in the pre-contractual stage, because the business risk of negotiations failing is fully transferred to the non-investing partner.

A complete absence of liability for failed negotiations is also undesirable since it leads to inefficient reliance decisions. In the case of symmetric information, there is, however, no need for the law to incentivise a party to invest efficiently, because the non-investing party will do it itself by voluntary participation in such investment costs.

On the other hand, in the case of asymmetric information enabling the non-investing party to induce the investing party to make inefficient reliance investments, it is desirable to set a liability rule that deters such inefficiency.

If the non-investing party faces liability for damage[20] suffered by the investing party who relied on misleading information from the non-investing party on the probability of reaching agreement on the contract, and if such liability has a probability of 60%[21], in most cases, the expected loss may exceed the expected gain from the transaction and thus deter the non-investing party from misleading the other.

To express this mathematically, let’s say that the reliance costs C=10 increase the surplus from the transaction by 20 (S=20), and the probability that the transaction will be pursued is 40%. However, the relying party reasonably believes (being misled by the other party) that such probability is 80%. The scrap value of the investment for the investing party is 4. It is clear that the relying party would not engage in the reliance investment if it had known the real probability of reaching the agreement, because the costs would exceed its private benefit from the transaction (10 > 0.4*10 + 4).

However, such a party will be willing to invest if the probability is 80% since the benefit of such reliance will exceed its costs (10 < 0.8*10 + 4). Where the liability rule exists, the non-investing party will have to compensate the other party’s costs with a probability of 60%, i.e. the expected loss of such misleading is 6 (0.6*10) whereas its expected private benefit is 4 (0.4*10).

Thus, the liability imposed on the party who misrepresented the probability of reaching agreement on the contract and thus induced the other party to make an inefficient reliance decision, deters such inefficiency and results in a socially optimal outcome.

3. Liability for failed negotiations under Czech law

Act No. 89/2012 Sb., the Civil Code, is based on the freedom of contract and autonomy of will.[22] As a first code, however, it defined criteria under which a negotiating party may be liable for failed negotiations. For liability to be found under Section 1729 of the Civil Code, the parties must have reached an advanced stage of negotiations so that reaching agreement on the contract seems very likely and the parties were acting in a way that one could reasonably expect agreement to be reached. The party breaking off the negotiations without having a legitimate reason is considered dishonest and will be liable for damage suffered by the other party.

The basis of liability for failed negotiations under the Czech Civil Code lies in creating the expectation that the contract will very likely be formed and that the other party reasonably relied on that.         The liability for failed negotiations was in the same way interpreted by the Czech Supreme Court before the Civil Code became effective, and it imposed liability for failed negotiations based on the general prevention duty.[23] The Czech Supreme Court repeatedly stated[24] that the liability for failed negotiations may be imposed where one party, due to the behaviour of its partner, believes in good faith that agreement on the contract will be reached and the first party breaks off the negotiations without having a legitimate reason.

I assume that, under the Civil Code, the courts will focus on whether one party misled the other about the likelihood of reaching agreement on the contract which the other party could have reasonably believed (i.e. whether the investing party reasonably believed that the contract will be formed due to the behaviour of the other partner). The stage of negotiations should not, therefore, be relevant as such but may help in considering if it was reasonable for the relying party to assume that agreement on the contract will likely be reached.

We see that the liability rule for failed negotiations basically corresponds to the optimal liability regime as suggested above. Liability is limited to cases in which the investing party was given a reasonable expectation by the other party that agreement on the contract would very likely be reached (the case of information asymmetry) but the negotiations failed eventually for no satisfactory reason.

For the rule to be efficiently applied, the courts should look at the behaviour of both parties during negotiations. It is not sufficient for one party to believe that the contract will be formed because of the advanced state of negotiations, which could produce inefficiencies consisting of overinvestment.

4. Liability for failed negotiations under Dutch law

The Dutch Supreme Court defined in 1982[25] a three-stage liability test for failed negotiations. In the first stage (when the negotiations have begun), either party may break off the negotiations without any liability. Once the negotiations proceed further (the second stage), the party who breaks off the negotiations is liable to compensate the other party for the costs incurred in connection with such negotiations. In the third stage, neither party can break off the negotiations without facing a sanction corresponding to a breach of contract.[26]

The Dutch Supreme Court held that there is a possibility that a negotiation concerning a contract has reached a certain stage which will render breaking off of the negotiation under the circumstances as an act contrary to good faith as parties might rely on each other that any concluded contract will, in any case, emanate from the negotiation.[27]

The Supreme Court thus indicated that protection is provided to the party who believes that agreement on a contract will be reached and such belief results from the advanced stage of negotiations. It is thus not relevant if there is any information asymmetry dishonestly used by the one who broke off negotiations.

This rule is, however, socially undesirable since it may lead to the overinvestment problem as suggested in the section describing the “strict liability regime” above. This is because the risk of the failure of negotiation is fully transferred to the non-investing party even if there is no dishonesty on its part.

There were attempts to codify the conclusions from the Plas v. Valburg decisions in the Civil Code (Burgerlijk Wetboek), but those attempts failed because the concept of liability for the breakdown of negotiations was considered not sufficiently developed yet.[28] More case law was needed to clearly establish the limits of duty for failed negotiations.

The subsequent case law seems to bring the liability for the breakdown of negotiations closer to the interpretation of the Czech Supreme Court and thus to a more efficient outcome. In Combinatie v. de Staat[29], it was held that the circumstances of the breakdown of negotiations have to be assessed, so the mere advanced stage of negotiations is not sufficient for the court to find liability. In De Ruiterij v. MBO[30], the Dutch Supreme Court held that the court must take into account the extent to which one party caused the other party’s reliance, and importantly, it stated that ‘unforeseen circumstances’ leading to termination of negotiations may justify such a break down.[31]

5. Conclusion

In this paper, I considered whether it is socially desirable to impose liability for failed negotiations using the normative criterion of efficiency. In an analysis of the two polar regimes of no-liability and strict liability, I proved that both regimes lead to inefficiencies consisting mainly of overinvestment in reliance by the investing party. Thus, it is necessary to have a legal rule that incentivises the investing party to only engage in efficient reliance.

Such liability should be imposed in the cases of asymmetric information between parties when the investing party has inferior information and the non-investing party uses its superior information about some relevant parameters to make a reliance investment that it would otherwise not have made. The liability should not, however, be broadened to include any breach of negotiations (even if those negotiations proceeded to an advanced stage) since such stricter liability tends to lead to the problem of overinvestment.

Comparing the results from the analysis of different legal regimes with the wording of Section 1729 of the Czech Civil Code and the case law of the Czech Supreme Court, I discovered that the liability for failed negotiations in Czech law basically corresponds to the optimal legal regime. On the other hand, the position taken by the Dutch Supreme Court in Plas v. Valburg could lead to overinvestment in reliance since it is close to what we called “strict liability”. The more recent case law of the Dutch Supreme Court, however, “weakens” the liability for the breakdown of negotiations and brings the rule rather closer to that established by Section 1729 of the Czech Civil Code.

Tereza Nováková graduated in 2015 from the Faculty of Law of Charles University and from the Faculty of International Relations, University of Economics in Prague. She spent one year at the Cardiff University Law School as an exchange student. Currently, she works as a law clerk in the District court for Prague 6. At the same time, she is a Ph.D. candidate at the Faculty of International Relations, University of Economics in Prague.


[1] As efficient, we understand an action which increases the joint welfare of both parties.

[2] Bebchuk, L. A. and Ben-Shahar, O. (2001) Precontractual Reliance. Journal of Legal Studies: Vol. 30: No. 2, Article 6., [online]. Available at: http://chicagounbound.uchicago.edu/jls/vol30/iss2/6 [Accessed 4 September 2016], p. 9.

[3]Wils, W., (1993) Who Should Bear The Costs Of Failed Negotiations? A Functional Inquiry Into Precontractual Liability. Discussion Paper No. 111, 6/92, Harvard Law School, Cambridge, MA 02138, ISSN 1045-6333, [online]. Available at: http://www.law.harvard.edu/programs/olin_center/papers/pdf/Wils_111.pdf [Accessed 5 November 2016], p. 21.

3 Wils, op. cit., note 3, p. 9.

[5] i.e. the surplus from the transaction will be divided equally between both parties.

[6] Bebchuk and Ben-Shahar, op. cit., note 2, p. 10.

[7] Schwartz, A. and Scott, R. E. (2007) Precontractual Liability and Preliminary Agreements. Faculty Scholarship Series. Paper 301, p. 24 [online]. Available at http://digitalcommons.law.yale.edu/fss_papers/301 [Accessed on 3 September 2016].

[8] Economically, each party’s expected return looks as follows: π[1/2(δS)]-C where π is the probability that the project will be successful, δS is the discounted surplus from the project by one period and C is the investment costs incurred by the respective party.

[9] Schwartz and Scott, op. cit., note 7, p. 24.

[10] Bebchuk and Ben-Shahar, op. cit., note 2, p. 432.

[11] Wils, op. cit., note 3, p. 10.

[12] Wils, op. cit., note 3, p. 9.

[13] Ibid., pp. 11–12.

[14] The non-investing party takes only the benefit from the transaction without participating in the precontractual costs. This is why they tend to induce the investing party to have reliance even if it leads to overinvestment of the investing party (under the condition of equal bargaining powers and a no liability regime).

[15] Wils, op. cit., note 3, p. 21.

[16] The strict liability regime will be the only option if the courts can determine only the investment costs of the parties and not their bargaining powers and behaviour during negotiations.

[17] Bebchuk and Ben-Shahar, op. cit., note 2, pp. 433–434.

[18]The risk is shifted to the non-investing party which has to reimburse all the costs of the investing party if agreement on the contract is not reached for any reason.

[19] The investing party will be willing to invest up to its expected share of the surplus plus the scrap value of the investment (if any).

[20] The loss consists of the costs actually incurred in reliance that the contract will be concluded and potentially the lost profit from the second best opportunity which must be real and not merely hypothetical.

[21] which is the real probability of failure of negotiations assumed in the example on page 4 above.

[22] Explanatory Memorandum to Act No. 89/2012 Sb., Civil Code, [online]. Available at: http://obcanskyzakonik.justice.cz/images/pdf/Duvodova-zprava-NOZ-konsolidovana-verze.pdf [Accessed on 5 November 2016], p. 432.

[23] Section 415 of Act No. 40/1964 Sb., the Civil Code.

[24] E.g. Czech Supreme Court, dated 11 October 2006, Ref. No. 29 Odo 1166/2004.

[25] Dutch Supreme Court, dated 18 June 1982, N.J. 1983, 723 (Plas v. Valburg).

[26] In the case of the Dutch Supreme Court, dated 11 March 1983, NJ 1983, 585, specific performance was granted.

[27] Rslasut’s Blog. Pre-contractual Phase: A comparison between Dutch law and English law as an

indication of an alternative approach for the Indonesian practice. [online] Available at <http://rslasut.wordpress.com/pre-contractual-phase-a-comparison-between-dutch-law-and-english-law-

as-an-indication-of-an-alternative-approach-for-the-indonesian-practice/> [Accessed on 6 November 2016].

[28] Zimmermann R., Whittaker S., Good Faith in European Contract Law, Cambridge University Press 2000, p. 247.

[29] Hoge Raad 4 October 1996, NJ 1997 (Combinatie v. de Staat).

[30] Hoge Raad 14 June 1996, NJ 1997, 481 (De Ruiterij v. MBO).

[31] Hondius E.H., Towards a European Civil Code, Kluwer Law International, 2004, 847 pages, p. 374.