A. Context: The “Advance” Purchase of Vaccines and the AstraZeneca Litigation

The Covid 19-pandemic has had a revealing effect on numerous legal relationships. Central themes have been “rights” issues, the extent of the regulatory power of the State, and the precarious balance between risks of life and compensation schemes for keeping things going. Turning to contract law, we observe a multitude of adjustment needs caused by the changes and interruptions of all production cycles, including education and culture. My focus here is on structural problems of the contracts concerning the “fight against the virus,” the EU vaccine purchase contracts. In this context we find contract clauses that have received a recent prominence and wide use in the world of international business contracting, “best efforts” clauses. They are employed for “softening” obligations. Their functions and problems, and their place in a “taxonomy of obligations,” will be developed in this Article from the perspective of a draftsperson.

Contracts are individualizing legal relationships. No important contract can be understood without its “personal” narrative and its context. Consequently, I will first shortly turn to the narrative and context of the EU vaccine deals, a multi-billion-dollar game of highest social and political priority.

The news about the first outbreaks and the deadly potential of the infectious disease in late December 2019 prompted immediate research on and development of curing and immunization strategies. In my hometown of Marburg in Germany, a traditional center for virus research and vaccine production, I witnessed as early as February 2020 the rapid start of related academic projects, including immediate public research subsidies, for addressing the development of an effective viral vector vaccine that is presently in the last trial phase.Footnote ¹ Concurrently, 100 kilometers to the southwest, in Mainz, a small high tech drug developer, BioNTech rushed to modifying an ongoing cancer vaccine project on an innovative “mRNA” technology. Driven and assisted by the global outreach of US pharma giant Pfizer, this led to a highly successful Covid 19 vaccination formula in the unprecedented time of less than a year. Similar projects were conducted around the globe. In the US in early 2020, the Trump administration started “Operation Warp Speed.” It included the provision of subsidies that finally amounted to an authorization of US $20 billion for research, development, the building of new production facilities, and prioritized national procurement of vaccines.Footnote ² Within the European Union, subsidies were initially offered by some individual Member States. A report of the German government in a “Small Inquiry” procedure of the Bundestag of February 1, 2021, lists authorized subsidies, totaling 740 million Euro to three firms operating in the Federal Republic.Footnote ³

To avoid an unequal treatment within the EU, the Member States decided for a common purchase and distribution of the vaccines. They authorized the EU Commission as a common agent for negotiating and concluding the relevant agreements with the most promising production firms. The mentioned report to Members of the Bundestag lists six firms with which the Commission, represented by the Commissioner for Health and Safety, signed so-called “advance purchase agreements” or “buying options” between the end of August 2020 and January 2021.Footnote ⁴

All agreements were classified “confidential.” In view of a protracted start of the vaccination programs and an alarming scarcity of vaccine deliveries, the public, and finally the European Parliament, required the disclosure of the relevant contract documentation. Thereafter, the sizable agreements were made public. All documents that are now available by special request, contain partially blacked-out “sensitive” clauses. The “editing” has been done—as far as can be seen—largely by discretion of the firms, with uneven results concerning the intensity.Footnote ⁵ For purposes of analysis, I will use the six published contracts,Footnote ⁶ but I will mainly refer to the most readable sets, those between the EU and AstraZeneca, the existing “Oxford” vector vaccine, and CureVac, the “Tübingen” mRNA vaccine, that is still under review for approval, as of June 2021.

The AstraZeneca contract is of special interest because the EU Commission blames the firm for breach of contract. From their legal point of view, the firm has failed to deliver the specified number of vaccine doses. The firm contends that it did not violate the agreement because the delivery targets mentioned in the schedule are limited by a best efforts clause. Both parties did not reach a settlement in the conciliation period of 20 days, stated in the agreement.Footnote ⁷ After this period, the Commission filed a lawsuit before the court appointed in the contract, the Tribunal de première instance francophone de Bruxelles, Section Civile, Belgian law being applicable. They asked for the delivery of 300 million doses and for the assessment of penal damages (“astreinte”) corresponding to this amount. According to a press release they also cancelled the contract.Footnote ⁸

On June 18, 2021, the single judge handed down a preliminary decision with a Solomonic result.Footnote ⁹ The 67-page decision reports the complete negotiating history with all relevant clauses translated into French. The judge is particularly concerned with a priority delivery clause of 100 million doses in a contract between the University of Oxford and the UK subsidiary AstraZeneca UK Ltd. dated May 17, 2020.Footnote ¹⁰ In an agreement on August 24 and 28, 2020, this subsidiary signed a respective priority agreement with the UK Government.Footnote ¹¹ As noted in the reference to the contract text,Footnote ¹² the EU signed their agreement with the mother company, AstraZeneca AB, Sweden, on August 27, containing Article 13, stipulating that the firm:

In the judgment which would need a special extensive treatment, the judge “pierces the corporate veil” by looking at AstraZeneca as one unit. She partly discards the priority agreement by putting all delivered doses together. But she clearly underlines that all deliveries are “best efforts” deliveries.Footnote ¹³ Applying the French/Belgian doctrine of an “obligation de moyens,” that will be explained below,Footnote ¹⁴ she reaches the conclusion that currently AstraZeneca is only obliged to deliver, at specified dates in July, August, and September 2021, a total of 50 million doses, instead of the 300 million doses requested. But she also assesses penal damages for the case of non-delivery. Commenting on the judgment, AstraZeneca already quipped that they would follow the judgment with pleasure.Footnote ¹⁵ The President of the EU, Ursula von der Leyen also welcomed the decision, stating: “It … demonstrates, that it [the vaccination campaign] was founded on a sound legal basis.”Footnote ¹⁶

In the following, I will leave the “interpretive” view of a court concerned with contract litigation. I will rather concentrate on the relevant clauses from a drafting perspective. At first sight, a cautious approach of the firms, by promising their “best efforts,” seems more than warranted in our situation. At the time of negotiating the deals, the products to be delivered were, at most, in a trial stage.

But contracts are concerned with future events. And events in the future can be made present by using conditions. If the developers will be ultimately successful in their efforts, there will be a point in time when the developed and admitted vaccines will become marketable goods. No doubt, the production, supply, and delivery will be complex. How is this process reflected in the contracts? What are the projected contract risks, the relevant contract clauses, their treatment of performance issues, and the underlying theoretical aspects?

B. The Use of “Best Efforts” Clauses in Two Vaccine Contracts

Despite the complexities associated with the development of Covid 19, vaccines the drafting of a purchase contract is not rocket science, even if billions of Euros are concerned. States and private health organizations routinely procure newly developed vaccines. We have been witnessing the annual need for fighting influenza with adapted new vaccines and the outbreak of no less than three pandemics since the turn of the century, not accounting for the local or regional events with Covid, ZIKA, and the MERS outbreaks. In an earlier short paper on the structural defects of the vaccine contracts, I tried to show that a drafting exercise would have to contemplate at least three different factual stages for designing an appropriate contractual coverage:Footnote ¹⁷

Let us look how these stages are technically covered in the contracts, using the examples of CureVac and AstraZeneca.

C. “Best Efforts” as a Mode of Performance – the Obligation and Performance Riddle

Is the concept of “best efforts” a standard, or rather a mode of performance? The easiest way of settling this divergence in the functions of the clause might be the verification that it is both. We might be confronted with specific traits of the different legal orders, and we would have to live with the differences. The iconic comparative law text of Zweigert & Kötz seems to point in this direction. In the context of “claims to performance,” the authors discuss the different approaches.Footnote ³⁵ They start with German law dealing with the theoretical priority of a claim for specific performance, which is “not of much use.”Footnote ³⁶ For qualified personal services, they point to a rule of civil procedure—Section 888 of the statute on civil procedure—that limits enforcement.Footnote ³⁷ Here, obviously the “character of the obligation” matters. For French and Belgian law, a seemingly related problem is being discussed in the section on breach of contract and the claim for damages, particularly concerning the issue whether a seller who fails to deliver may excuse himself or herself by claiming that he or she was not at fault. The old rule of the French Code civil, Article 1147 —until 2017—would deny this. Only force majeure or cas fortuit would be a reason for excuse. But in certain situations, fault is a viable excuse. Tony Weir’s English translation of Zweigert & Kötz nicely restates the original French doctrine:

It is not by accident that Tony Weir, a most distinguished English comparative lawyer known for the subtlety of his translations, uses the term “best efforts” obligation, for giving a proper meaning to the idea of an “obligation de moyens” in this context.

Obviously, the French doctrine since the analysis of René Demogue,Footnote ³⁹ followed also by Belgian law,Footnote ⁴⁰ looks at the contracts from an objective point of view. As a matter of fact, the treatment of doctors and the efforts provided by the service of lawyers are typically treated as these supposedly softer “obligations de moyens.”Footnote ⁴¹ This is the judicial perspective. The crucial question comes up under a drafting perspective. Can—if one uses the French distinctionFootnote ⁴²—a “statutory” obligation “de résultat” be converted into an obligation “de moyens” in the contract between the parties?

The answer is “yes, within limits,”Footnote ⁴³ not only confined to French or Belgian law,Footnote ⁴⁴ but in all major transaction orders on the globe that accept the premise of a general freedom of contract. The standard sales example is simple enough. Of course, it would be classed as an “obligation de résultat” under French law. However, in a commercial context, I could reasonably contract for a delivery of last resort, probably of an uncertain quantity and/or an uncertain time, with unspecified quality changes, likely at a lower price. It may be in the interest of both parties to share upstream risks in the production, storing, and shipping phases. Moreover, I could deal with an even softer scheme “to be considered” in a seller market. Clearing sales in food or fruit markets are daily illustrations in which quality and quantity risks are shifted to the customer, and even these deals, if happening regularly, may be covered by umbrella agreement with the purchasers. The commitment “I’ll try to get that widget for you” is standard under many circumstances and mostly meant as a binding commitment, weak as it may be. At the other end of the scale are fixed, guaranteed, and dedicated deliveries or services “just-in-time,” with iron clad penalties/liquidated damages clauses for non-performance.

The underlying theoretical question that has already been implicitly raised by the examples just given is whether the world of drafted obligations follows a clear-cut division between “result obligations” and “best efforts obligations.” Here, the answer is no. But the drafting practitioners should be happy that the French doctrine, circling around the alignment of the seemingly contradicting old Articles 1137 and 1147 of the venerable Civil code of 1804, has been producing a simple and plausible clear cut “modular” structureFootnote ⁴⁵ to their efforts of entertaining a practical distinction for aligning the interests of the parties for “softer” forms of bindingness in individual cases. In that sense, the original French concept, which has been implicitly around in all legal orders in some way, is a good starting point for detailing the individual obligations in the negotiating process.Footnote ⁴⁶

As in many other cases of the institutional development of private law, the result/effort distinction is rooted in the changes of contract law in the 18th and 19th centuries. In this period, in both the “romanistic” and the “common law” legal orders, a limited, action-based “law of contracts remedies” has been transformed into a free-wheeling substantive law of contract, of course leaving some traces of the former approach.Footnote ⁴⁷ An example is the mentioned contradiction of articles 1137 and 1147 Code civil.

The debates about the transformation of the original Roman contract actions into a “modern” substantive law of obligations in the middle of the 19th century are telling. The most erudite legal scholar of his time, Friedrich Carl von Savigny, takes pains to specify and clarify the point. This has been largely forgotten in the remainder of his century. In his final work, the unfinished two volume treatise on the Law of Obligations as a Part of the Present Roman Law, Savigny stresses that there is a pervasive complexity of obligations that should not be put in the straitjacket of the simple divisions of the former Roman law actions.Footnote ⁴⁸ The Roman sources are, indeed, ambiguous at this point. The important Roman jurist Gaius, looking at the actions in personam, Footnote ⁴⁹ suggests dividing the obligations between “dare”—a “giving,” technically a “handing out,” probably lurking around in the French “résultat;” “facere”—a “doing, acting,” later including “omissions,” vaguely related to the French “moyens;” and a fairly dubious “praestare,” possibly derived from the formulae in tort actions.Footnote ⁵⁰ Savigny insists that the admissible variety of obligations, to be shaped by the will of the parties, forbids a neat subdivision of this kind.Footnote ⁵¹ He stresses that the terms used by Gaius are rather “flexible.”Footnote ⁵² Indeed, in a footnote, he can show that in one reported case, the “facere” obligation is not less stringent than the “dare” obligation but even embraces the “dare.”Footnote ⁵³

The technically least sophisticated contract system, that of the various common law jurisdictions, has had the easiest time for adaptation to the openness of negotiated obligations. It never faced the constraints of codification and the related doctrinal fights in the brave new world of freedom of contract. It is questionable, however, whether Farnsworth’s analysisFootnote ⁵⁴ for the jurisdictions in the US, to understand “best efforts” as an overarching “bilateral standard,” is correct in general. Here, as in many other cases, we should not entertain the comparative law cuckoo that, despite similar results in practice, the different spirits of the common and continental laws emanating from the different legendary deep forests of Nottingham, or those of the Druids in the middle of France, require a fundamentally distinct conception of legal problems.

“Best efforts” may become a “bilateral standard” if both parties are interested in an individual obligational commitment of the same kind, namely, not to perform in a straightforward manner but to do their best that the specified individual targets in the contract are being met. In other words, “best efforts” is a matter of the individually specified obligation and its performance. The more frequent generic cases for best efforts that are mentioned in the French hornbooks, like the doctor-patient relationship, concern cases in which one party makes a serious, well understood effort, and the other is willing to react straightforwardly by paying the contract bill.Footnote ⁵⁵ However, in the sphere of negotiated commercial contracts, there are many nuances of obligation. A simple division between “strict” and “best efforts” obligations would hardly be adequate for understanding these arrangements. Here, the resolution of disputes about the commitments of the parties is a matter of looking at the contractual text in its context and of abstaining from simple black and white classifications.

D. A Primer on the Taxonomy of Obligation

In the negotiation and drafting of commercial contracts, “bindingness” is a central aspect.Footnote ⁵⁶ Parties try to hedge their commitments to the very last minute of signing. If costly commitments beyond the normal outlays are concerned, we witness a sort of “Coasean bargaining in contrahendo”Footnote ⁵⁷ concerning the coverage of costs, neatly separated from the issue of whether a final agreement has been reached. Substantial agreements contain clauses specifying when and under which conditions the agreement will become binding.Footnote ⁵⁸ In the course of negotiating a deal, the parties will try to fix individual chapters and leave others open, adding a note at the margin “to be discussed,” abbreviated typically with “TBD.” This indicates that the finished chapters contain a sort of “social” bindingness. Although it is clear to the parties that the text is not binding legally, reneging comes, at least, at a reputational cost. At this point, a lack of cooperative behavior will be sanctioned by blaming and shaming. “You might not wish to make a deal with this counterpart,” sometimes leading to strategies of the other parties to add a matching risk component in their first offer. Tougher reactions may be the factual exclusion from new business, or even exclusion from “clubs” like business meetings, commercial platforms, or information circles. Numerous important documented business arrangements,Footnote ⁵⁹ in some cases named “letter of intent,” never reach a legally binding stage despite being honored meticulously by the parties.

The world of binding commercial contracts starts with arrangements in which parties are hedging their commitments with “softener” clauses, especially using the term “best efforts,” mostly mitigated by explanations of what they consider as reasonable under the circumstances. A good example is the CureVac Contract, treated above, including the potentially tricky effort to include commitments to third parties, as an excuse for non-performance.

Although “best efforts” may be used for a varying intensity of obligation, the arrangements of this kind may be grouped under this catchword. As mentioned before, there are instances of generic best effort-contracts, although in the canonical French examples the requisite “efforts” of lawyers and doctors could also be framed by detailing their obligations, namely their professional duties under the given circumstances. In a litigation about professional malpractice, the statement “But I exercised my best efforts” alone would hardly excuse the defendant.Footnote ⁶⁰ This indicates that “best efforts” is mostly used if the parties are not prepared to detail their duties, particularly if the expected benefits from specification are lower than the drafting costs, including the frustrations of commercial parties when lawyers elaborate on the obvious.Footnote ⁶¹

The “stronger” commitments may be classed under a rubric of “performance contracts” and “guarantee contracts,” with two immediate caveats. First, all three groupings of legally binding agreements are on a sliding scale; they are not new ”contract types,” as treated in the Continental contract law codifications. Second, the term “performance contacts” should not be understood as a verbal collocation of the obvious. All contracts are technically concerned with a future, named performance. Here, the term “performance” is used for distinguishing them from the arrangements in which parties try to hedge their obligations by reference to their serious efforts to perform.

In all groups of arrangement, the characteristic “risk reduction technology” differs. In best efforts contracts we find typically a “reasonableness” limit that is at the forefront of constraining the obligation to use the best efforts to perform. In “performance contracts,” this role is assumed by force majeure clauses. These do also show up in best efforts contracts, but they do not technically matter there, because the incidence of normal “force majeure” events would make the best efforts performance “unreasonable” anyhow. If I use “strict force majeure” as the “risk reduction technology” for the strictest of all arrangements, the guarantee contracts, I am commenting on instances in the drafting and negotiating exercise in which “force majeure” is extended to, for example, labor disputes or administrative consent, events that are not completely out of the control of the debtor.Footnote ⁶² This leads to the following scheme of a tentative taxonomy of bindingness and enforcement of commercial arrangements (Table 1):

E. Searching for a Fitting Contract Scheme

Whereas the classical codification movement of the last two centuries has been a high-flying effort of capturing emanations of a “jurisprudentia perennis,” the prosaic exercise of drafting commercial agreements follows the needs and variety of the day. Reading large contracts for some decades, I could witness all sorts of “contract fashions.” A prominent example for the 60’s to the 80’s has been the “joint venture” movement in all possible and impossible situations. The financial sector supplied the idea of covenants, which rapidly spread out to other fields with a limited adaptation potential. This also happened with MAC-clauses, including their entropic growth in size in the contract documentation. “Best efforts” clauses have been around for a long time because they were fitting to certain types of arrangement in which—demonstrated by the examples of the French doctrine—an obligation could not be framed simply in result terms. But they are currently en vogue in the industry, with dubious consequences.

Discussing the vaccine development above, I pointed to the initial Stages 1 and 2 concerning the open-end research and the uncertain development of a functioning drug, including the complexity of testing and organizing a new production and distribution scheme. These stages are captured adequately by the concept of “best efforts” from a drafting perspective. There are many commitments in practice in which experienced draftspersons will be happy to start working with the best efforts-idea, detailing it in the contract text. But it is both fashion and/or strategy using the best efforts-approach in a clear-cut production and delivery context. That is not to say that using the idea is illegal. But it may not work.

At first blush, it may sound smart to keep the arrangement as open as possible. And the “office memo” to the principals may state: “We have succeeded in negotiating a deal that keeps all options open for the firm.” Scott and Triantis, in their analysis of best efforts clauses, point to the downside of this strategy.Footnote ⁶³ The “cheap” openness ex ante may become costly ex post. It is trivial that, fortunately, only very few commercial transactions dealing with definable performances, using best efforts clauses for hedging the commitment, go to court. However, as Scott and Triantis explain, in this unlikely situation, the court will have to deal with a “vague term” and be required to “complete the contract.Footnote ⁶⁴”

It should have been easy to formulate in the vaccine contracts that, after reaching the point of admission of the vaccine and the demonstration of the viability of a large-scale production, the parties would enter into a standard production and delivery relation, a “performance contract” in the meaning developed above, including a highly detailed force majeure clause expressing all potential constraints in the production and delivery as excuses. Whenever an “openness” of a “contract future” is at stake, conflict avoidance demands to think twice about the selection of the starting point for a fitting draft.

F. An Afterthought

The EU vaccine contracts are a story of failed ambitions and misleading contract choice. The lawyers of AstraZeneca may be disappointed because they introduced a common law style “industry contract” which might work between commercial parties, or vis-a-vis the UK National Health Service, if it would have stayed in a common law setting, including, preferably, commercial arbitration. They accepted a reading of this contract under Belgian law by a Belgian civil court that will be influenced by the seemingly fundamental distinction between “obligations de résultat” and “obligations de moyens.” They were fortunate finding a judge who expressed an understanding for the freedom of the parties to hedge their risks in these terms. If the judge would have taken the objective route for the present arrangement, she might have arrived at a textbook classification:the delivery of vaccines concerns a “résultat.” They took an unnecessary risk and got away with a fuzzy contract that could have been re-written by a court.

Turning to CureVac, it is likely that the introduction of an escape clause in the surprising context of the definition of “reasonableness” will not matter in practice. The query of third party commitments will probably be irrelevant in view of a generous supply of other mRNA vaccines in the second half of 2021. Why they styled a future delivery contract as a best efforts arrangement will remain their secret.

The critique of the drafting efforts of AstraZeneca and CureVac does not exonerate the other producers and their contract management. To the contrary, their extensive blackening of clauses does not show strength. A closer analysis would probably also show a similar fuzziness in these “copy & paste” contracts.

The visible flaws and associated queries of the vaccine contracts contain a general message for the international drafting profession. Best efforts clauses are no panacea for negotiating and drafting efficient major business deals. A stronger effort for specification of performances makes good sense under many circumstances. Best efforts clauses will work for the parties if they fit the transaction.

Table 1. Taxonomy of Bindingness and Enforcement

Note, that these four stages of bindingness in contracting are on a sliding scale. They are NOT “new” contract types.