skip to main content


Sweeping Connecticut Supreme Court Decision Refines Existing Tort Law

Connecticut Supreme Court Declines to Adopt the Restatement (Third) of Torts, Products Liability

January 10, 2017

In arguably the most important Connecticut tort-law decision in decades, the Connecticut Supreme Court in Bifolck v. Philip Morris, Inc., --- A.3d ---, 2016 WL 7509118 (Conn. Dec. 29, 2016), declined to adopt the approach of the Restatement (Third) to product liability design-defect claims and  “reaffirm[ed] its allegiance” to a “true strict liability” standard under § 402A of the Restatement (Second).  The Court also made a number of “modest refinements” to the Court’s existing interpretation of § 402A.  Most importantly, the Court held that every product liability design-defect claim must allege that the product was “unreasonably dangerous,” but declined to box plaintiffs into one definition of that term for purposes of stating a claim.  The Court also refused to limit punitive damages under the Connecticut Product Liability Act (“CPLA”) to the “litigation expenses less costs” limit under the common-law rule set forth in Waterbury Petroleum Products, Inc. v. Canaan Oil & Fuel Co., 193 Conn. 208, 477 A.2d 988 (1984).  Given the Court’s cautious approach to remaking the state’s tort law, Bifolck is in practice a reaffirmation of the status quo in Connecticut—at least for now.  The Court did leave open the possibility that it might adopt the Restatement (Third) at some point in the future should its standards under § 402A prove “unworkable.”

The Bifolck decision was the second Connecticut Supreme Court decision in 2016 to consider the operation of the  CPLA in the design-defect context.  In the first, Izzarelli v. R.J. Reynolds Tobacco Co., 321 Conn. 172, 136 A.2d 1232 (2016), the Court held that the “modified consumer expectation test” would be the primary test in a strict liability action based on defective design, while the “ordinary consumer expectation test” would be reserved for res ipsa-type cases in which the allegedly defective product failed to meet the ordinary consumer’s minimum safety expectations.  The Bifolck decision, fortunately, does away with the confusing names of these “tests” without changing the holding of Izzarelli.[1]

Rejecting (or Postponing) Adoption of the Restatement (Third)

The Connecticut Supreme Court is the latest state Supreme Court to decline to adopt the approach of Restatement (Third) to product liability design-defect claims.  The Court pointed out that § 2(b) of the Restatement (Third), which provides that a product “is defective in design when the foreseeable risks of harm posed by the product could have been reduced or avoided by the adoption of a reasonable alternative design by the seller or other distributor . . . and the omission of the alternative design renders the product not reasonably safe . . . ,” imposes two requirements not mandated under the Court’s § 402A tests:  first, proof that the harm was foreseeable; second, proof that a reasonable alternative design existed that would have reduced or avoided the danger.  These requirements, the Court asserted, “would appear to make consequential changes to our product liability law.”  The first requirement deviates from § 402A’s “true strict liability standard” and “effectively require[s] proof of negligence.”  The second requires proof of availability of an alternative design; under the Court’s § 402A tests, such evidence is relevant but not required.

The Court declined to adopt the Restatement (Third) largely in light of these additional requirements.  The Court explained that requiring plaintiffs to prove that the risk of harm was foreseeable was “manifestly inconsistent with the court’s concern in Potter [v. Chicago Pneumatic Tool Co., 241 Conn. 199, 694 A.2d 1319 (1997),] about the burdens of expert testimony . . . and its unequivocal determination that policy considerations favored adherence to strict liability.”  And requiring plaintiffs to prove availability of an alternative design “would preclude valid claims for products for which there is no alternative design”—even where the plaintiff alleges that “the product is so dangerous that it should not have been marketed at all.” 

In arriving at its holding, the Court rejected various of the defendant’s arguments in favor of adoption, including that the § 402A tests are “unworkable” (“In the almost two decades since this court adopted our modified consumer expectation test . . ., there has been no evidence that our § 402A strict liability tests have proved to be unworkable.”); that other jurisdictions have adopted the Restatement (Third) (“It suffices for our purposes that several other jurisdictions apply similar standards to ours, some for many years.”); and that evidence of a reasonable alternative design is routinely presented (“[S]imply because in cases of factually-marginal applications courts have found evidence relating to alternative designs to be particularly probative and persuasive . . . does not necessarily support a thesis that adducing such evidence is dispositive of whether a plaintiff has carried his/her burden of proof.”).

The court emphasized that it had not concluded it could not adopt the Restatement (Third)—just that it should not at the present juncture: “if we defer further consideration of the Restatement (Third) until such time as we have a case in which our current standards have demonstrated themselves to be unworkable or result in a manifest injustice, not only might we make a better informed decision, but the legislature might, in the interim, make its own reforms.” This is an interesting invitation, but it seems doubtful the Court will rush to reconsider Bifolck and Izzarelli in short order barring truly exceptional facts.

Pushing back against the proponents of Restatement (Third) on policy grounds, the Court “reaffirm[ed] its allegiance” to a “true strict liability” standard under § 402A of the Restatement (Second).

Refining Existing Standards

Having “reaffirmed [its] allegiance to a strict liability standard under § 402A,” the Court turned to whether § 402A and comment (i) apply to a product liability claim for negligence.  The Court agreed with the defendant that “no product liability action can succeed without proof of a defective condition unreasonably dangerous to the consumer or user”—even an action based on negligence.  (“[A]ny product liability claim, no matter the type or theory, is governed by the same essential elements.”)  The Court disagreed, however, that there was “a single definition of unreasonably dangerous[] as provided in comment (i)”—that is, that “a product is unreasonably dangerous only when it is dangerous to an extent beyond that contemplated by an ordinary consumer.”  Izzarelli made clear that a product may be unreasonably dangerous if it fails to meet consumers’ minimum safety expectations or if its risks exceed its utility.  And because the consumer may know of the risk of danger but fail to fully appreciate that danger or know how safe the product could be made,” a plaintiff may still recover for damages caused by a product where the product’s danger was open and obvious. 

In so holding, the Court sought to “clarify the plaintiff’s burden of proof in strict liability cases.”

  1. The Court dropped the confusing nomenclature it recently discussed in Izzarelli. Going forward, Connecticut law will recognize the “modified consumer expectation test” as a “risk utility test” and the “ordinary consumer expectation test” as a “consumer expectation test.” 
  2. The Court held that a plaintiff must allege, and thereby put the defendant on notice, whether the allegedly defective product is claimed to be “unreasonably dangerous” because (a) a reasonable alternative design could have reduced or avoided the danger, (b) the design of the product marketed is manifestly unreasonable in that the risk of harm from the product so clearly exceeds its utility that a reasonable, informed consumer would not purchase the product, or (c) both.
  3. Where a plaintiff proceeds on the theory that a product is unreasonably dangerous because it lacked some feature that would have reduced or avoided the injury, the plaintiff must “simply” prove that the alternative design was feasible (technically and economically) and that the alternative design would have reduced or avoided the harm.  Other factors may be relevant, but the failure to present proof on these other factors will not keep the case from the jury. 
  4. A defect may be established under the consumer expectation test by proof of the allegedly defective product’s noncompliance with safety statutes or regulations or a product seller’s express representations. 

Statutory Punitive Damages Not Limited by Common-Law Punitive Damages Rule

The Court further addressed whether the common-law rule of punitive damages articulated in Waterbury (limiting punitive damages to litigation expenses less costs) applies to limit an award of statutory punitive damages under the CPLA.  The CPLA provides that “[i]f the trier of fact determines that punitive damages should be awarded, the court shall determine the amount of such damages not to exceed an amount equal to twice the damages awarded to the plaintiff.”  Conn. Gen. Stat. § 52-240b. 

Applying the general rules of statutory construction but taking pains not to “extend[], modif[y], repeal[] or enlarge[]” the scope of § 52-240b, the Court concluded that the common-law rule does not act as a limit on the statutory rule—in other words, the Court would not construe the act to equate statutory punitive damages with litigation expenses.  First, the Court explained that the rules are inconsistent in many respects—for example, statutory damages are measured in relation to a multiple of compensatory damages, not litigation expenses.  Second, the Court indicated that if it were to construe the act to equate statutory punitive damages with litigation expenses, the statute would, in some cases, have no effect or frustrate the purpose of the common-law rule—as when litigation expenses are less than two times the damages or when a plaintiff’s compensatory damages are low in comparison to his or her litigation costs.  Third, if punitive damages in § 52-240b were interpreted to mean common-law punitive damages, then both §§ 52-240a (which provides for attorney’s fees under the CPLA) and 52-240b would provide for attorney’s fees, but under different conditions—for example, attorney’s fees under § 52-240a are not capped as are punitive damages in § 52-240b. 

Finally, the court underscored that it had developed the common-law punitive damages rule as a check on the “caprice and prejudice of [juries],” which can assess damages “in amounts which are unpredictable and bear no relation to the harmful act.”  “By vesting the court with the authority to determine the amount of punitive damages and by limiting the amount of those damages in the act, the legislature provided an alternative method of reining in excessive punitive damages . . . .”

Bifolck reaffirms Connecticut’s allegiance to strict liability and the status quo, leaving open only a remote possibility that the Supreme Court might adopt the Restatement (Third) where its § 402A tests prove “unworkable” or result in “manifest injustice.” 

[1] See Adam M. Masin, The Connecticut Supreme Court Modifies the State’s Consumer Expectation Test by Adopting the Modified Test, Drug & Device L. Blog (May 20, 2016),

© Shipman & Goodwin LLP 2019. All Rights Reserved.