Teaching Assistants: The Battle of the Contracts Law Titans over the "Eisenberg Uncertainty Principle"

This is the eighth in a series of posts on Victor Goldberg's work.  Today's post is about Chapter 11 of his book, Rethinking Contract Law and Contract Design (RCL).  Links to related posts follow this one.

In 2009, Mel Eisenberg published Impossibility, Impracticability, and Frustration, in which he made a sophisticated, nuanced, and carefully calibrated argument about when performance should be excused and what damages should be available in "unexpected circumstances cases."  Professor Goldberg thought Professor Eisenberg got it wrong, and Chapter 11 of RCL seems to be a reprint of his critique with a few additions.  You can read the original version here.  In this post,  I will first lay out Professor Goldberg's criticisms and then summarize Professor Eisenberg's response to those criticisms.  That seems prudent, as Professor Eisenberg thought that Professor Goldberg had misstated his argument.

According to Professor Goldberg, Professor Eisenberg proposes a "beefed-up" excuse doctrine through two tests relating to remote risks: the shared-assumption test, which permits excuse when the parties shared a tacit assumption  about some low-probability event, and the bounded-risk test, which allows for excuse when a change in prices leaves a promisor with a loss significantly greater than could have ben expected (RCL, 137).  Professor Goldberg's assessment is that the shared-assumption test would be hard to apply and that the bounded-risk test is fundamentally wrong (RCL, 138).

The analysis of the shared-assumption test begins with an illuminating discussion of the insurance market relevant to the classic frustration-of-purpose case, Krell v. Henry.  Krell sought to recover £50 owed under a contract to rent rooms with a view of the parade commemorating the coronation of Edward VII.  Due to the King's illness, the coronation did not take place.  The court found that the non-occurrence of the coronation parade frustrated Henry's purpose in leasing the flat and thus excused him from his contractual obligation to pay.

In Professor Eisenberg's language, the coronation parade was a shared tacit assumption, the non-occurrence of which excused performance.  But Professor Goldberg points out (quite wittily) that "[t]he likelihood that a 60-year-old, grossly overweight, heavy smoker, who had been the target of at least one assassination attempt might be unavailable was not trivial."  In fact, there was an active insurance market related to the coronation, and Lloyds was quoting odds of 300-1 against cancellation.  That cancellation was not an event that could not have been foreseen and against which the parties could not have protected themselves (RCL, 139-43).

Professor Goldberg next discusses Taylor v. Caldwell, the first English case to recognize the doctrine of impossibility, a doctrine expanded in Krell v. Henry.  There is not much to be said here.  Taylor establishes a default rule for when a venue is destroyed and performance is therefore impossible.  Default rules are not that important, because parties routinely contract around them.  Today, parties allocate risks with express contractual language.  We learn, for example, that Paula Abdul's contract provides that she gets paid whether or not she performs, so long as she is not at fault.  Rod Stewart's contract was a bit more involved, leading to a decade of litigation.   The point is not that unexpected events should never excuse performance; rather, Professor Goldberg maintains that the cases on which Professor Eisenberg relies would fail under his test, because the events that occurred were not unforeseen, and thus Professor Eisenberg's theory cannot provide the basis for the default rule (RCL, 144-45).

Professor Goldberg's attack on the bounded-risk test is multi-pronged, but in the end, he just doesn't think Professor Eisenberg's approach actually provides much guidance that would help determine actual cases.  According to Professor Goldberg, Professor Eisenberg thinks that the bounded-risk test would apply in cases of market-wide price increases (but see Professor Eisenberg's response, summarized below), but Professor Goldberg points out that excuses tend to work best in litigated cases involving fact-specific events, like fires, burst pipes, etc.  The only case that fits the market-wide paradigm is Alcoa v. Essex, in which the court imposed a floor under the seller's profits, but Professor Eisenberg does not rely on that case perhaps, Professor Goldberg suggests, "because he feared that his analysis would be tainted if it were associated with such a bad decision."  Here, as in most cases, Professor Goldberg prefers negotiated solutions to Professor Eisenberg's proposal, which he views as setting a cap on prices established by courts in response to sudden and dramatic price increases.  The reality is that parties do not choose gross inequity clauses that excuse performance in case of dramatic price changes.  If the parties disfavor such a solution, courts should not impose it on them (RCL, 146-53).

In his response, Professor Eisenberg clarifies that the shared assumption test is not about the parties' inability to foresee remote risks.  The could see the risks, but in actuality they don't -- or at least, they do not address those risks in the contract.  His test does not turn on whether the parties should have foreseen the risk; it turns on their tacit shared assumptions, regardless of foreseeability: "But reasonable foreseeability normally should be only an index, not the test. The test should be what the parties tacitly assumed (Eisenberg 2009, 216).  From this perspective, Krell v. Henry was not wrongly decided, as Professor Goldberg argues.  Yes, the cancellation of the coronation parade was foreseeable, but Krell and Henry did not foresee it.  

Professor Eisenberg proposes a default rule; Professor Goldberg does not.  He thinks it matters little, given that parties can contract around default rules.  Professor Eisenberg proclaims, "you can't fight something with nothing" (Eisenberg 2010, 390).  That said, Professor Goldberg also notes the general tendency of default rules to be sticky.  Courts are wary of parties' attempts to evade them, and so a bad default rule is to be avoided.  Professor Goldberg may be justified in fighting something with nothing if he prefers a return to more traditional approaches.

As to the bounded-risk test, Professor Eisenberg admits to one misstatement of his test.  Professor Goldberg seized upon that misstatement to highlight the theory's erroneousness.  But the bounded-risk test is not limited to instances of market-wide cost changes.  It applies to both seller-specific and market-wide cost increases, so long as they are dramatic (Eisenberg 2010, 390-91).  Professor Eisenberg then responds to Professor Goldberg's argument that the bounded-risk test should not apply to market-wide increases in cost.  The hypo author is the master of the hypo, so long as his hypothetical actors do not act unreasonably (Eisenberg 2010, 391-93).

While Professor Eisenberg's response is mostly dedicated to demonstrating the extent to which Professor Goldberg has misstated his thesis, in his conclusion, Professor Eisenberg clarifies the two authors' substantive differences and the stakes.  Professor Eisenberg's purpose was to provide a theoretical grounding for our current approach to excuse doctrine.  In resisting that approach, Professor Eisenberg argues, Professor Goldberg seeks to return contracts doctrine to its pre-Restatement classical period (Eisenberg 2010, 395).

A post on Chapters 8-10 (consequential damages) is here.

A post on Chapter 7 (liquidated damages) is here.

A post on Chapters 5 & 6 (speculative damages) is here.

A post on Chapter 4 (lost-volume damages) is here.

A post on Chapter 3 (timing for assessing damages) is here.

A post on Chapter 2 (the flexibility/reliance trade-off) is here.

The introductory post is here.