Pankey v. Petco Animal Supplies, Inc.

Filed 6/24/20
CERTIFIED FOR PUBLICATION

COURT OF APPEAL, FOURTH APPELLATE DISTRICT

DIVISION ONE

STATE OF CALIFORNIA

ANDREW J. PANKEY, D072779

Plaintiff and Appellant,

v. (Super. Ct. No. 37-2014-00004156-
CU-PO-CTL)
PETCO ANIMAL SUPPLIES, INC.,

Defendant and Respondent.

APPEAL from a judgment of the Superior Court of San Diego County,

Eddie C. Sturgeon, Judge. Affirmed.

Gomez Trial Attorneys, John H. Gomez, Bibianne U. Fell; Higgs Fletcher &

Mack, John Morris, and Rachel E. Moffitt for Plaintiff and Appellant.

Horton, Oberrecht, Kirkpatrick & Martha, Kimberly S. Oberrecht; Greines,

Martin, Stein & Richland, Robert A. Olson, Cynthia E. Tobisman, and Eleanor A. Ruth

for Defendant and Respondent.

INTRODUCTION

Plaintiff Andrew J. Pankey (Andrew) filed a products liability claim against Petco

Animal Supplies, Inc., after his son Aidan contracted a rare bacterial infection from a rat
purchased at Petco. Aidan later died as a result of complications related to his infection.

Andrew alleged, among other things, that Petco was strictly liable for injuries resulting

from the sale of the pet rat, which he argued was a product for purposes of strict products

liability. The trial court instructed the jury on negligence under ordinary negligence and

negligent failure-to-warn theories, as well as three theories of strict products liability:

(1) failure to warn, (2) manufacturing defect, and (3) design defect under a risk-benefit

test. The jury returned verdicts in favor of Petco.

On appeal, Andrew contends the trial court erred by refusing to instruct the jury on

an alternative strict liability design defect theory, the consumer expectations test. He

argues there was sufficient evidence from which the jury could have concluded the pet rat

purchased from Petco failed to perform as safely as an ordinary consumer would expect

when used in an intended or reasonably foreseeable manner.

We conclude a live pet animal sold in its unaltered state is not a product subject to

the design defect consumer expectations theory of strict products liability. Accordingly,

we affirm. Because we affirm on this basis, we need not reach a conclusion regarding the

applicability of the consumer expectations test or the prejudicial effect of its exclusion.

Were we to do so, however, we would affirm.

FACTUAL AND PROCEDURAL BACKGROUND

A. Aidan Pankey

In May 2013, 10-year-old Aidan Pankey and his grandmother visited a Petco store

to purchase a companion for Aidan's female pet rat. They selected and purchased a male

2
rat. Aidan kept his male rat in a vivarium at his grandmother's home, but he occasionally

handled it outside the vivarium.

Two weeks after purchasing the male rat, Aidan developed a fever, grew

increasingly lethargic, and at one point collapsed on the floor. Aidan's grandmother

called an ambulance; Aidan passed away at the hospital shortly after he arrived.

The autopsy did not disclose a cause of death or any sign of a bite or scratch.

However, analysis from Aidan's tissue samples revealed the presence of streptobacillus

moniliformis, a bacteria carried by some rats. Analysis of blood and tissue samples taken

from Aidan's rats revealed the bacteria was carried by the male rat, but not the female rat.

The parties to this suit stipulated that Aidan's death resulted from rat bite fever (RBF), an

infection caused by streptobacillus moniliformis.

B. Pretrial Proceedings

Andrew and Aidan's mother, individually and as administrators of Aidan's estate,

sued Petco and Barney's Pets, the exclusive supplier of pet rats for Petco's California

stores during the relevant time frame, alleging negligence and strict products liability.

Petco moved for summary adjudication of the strict products liability cause of

action on the basis that a pet rat is not a product. The court denied the motion,

concluding a pet rat is a product for purposes of California's strict products liability

doctrine.

Aidan's mother settled and dismissed her claims against the defendants.

3
C. Trial

Andrew's case proceeded to a jury trial on four theories: (1) negligence, (2) strict

liability based on a warning defect, (3) strict liability based on a manufacturing defect,

and (4) strict liability based on design defect.

1. Streptobacillus Moniliformis

Plaintiff's infectious disease expert, Dr. Joseph Vinetz, testified that 10 to 100

percent of wild rats carry the bacteria streptobacillus moniliformis. The bacteria is a

commensal organism that has adapted to live in rats, but it "doesn't have to be" present in

rats. Rats that carry streptobacillus moniliformis appear healthy and for the most part are,

but they can transmit the bacteria to one another through direct contact or their

environments, and they can transmit the bacteria to humans through bites and scratches,

or through their saliva, feces, and urine.

In humans, streptobacillus moniliformis can cause RBF, which typically manifests

itself in flu-like symptoms, a fever, a rash, and/or joint pain. Possible complications from

RBF include endocarditis (a heart valve infection), meningitis (an infection of the lining

of the brain), septic arthritis, and for 7 to 13 percent of untreated patients, death.

Sometimes, RBF resolves on its own.

Dr. Vinetz testified it is "fairly simple" to breed pathogen-free rats that do not

carry streptobacillus moniliforms, and he testified at least two laboratories had done so.1

1 The chief operations officer from Barney's testified that he purchased pathogen-
free rats from one of the laboratories Dr. Vinetz cited, Harlan Laboratories, but rats from
that lab tested positive for the bacteria.
4
According to Dr. Vinetz, pathogen-free rats can be bred by (1) performing molecular

testing to screen out carrier rats, (2) segregating noncarrier rats in a clean facility free

from the bacteria where they are allowed to breed, (3) delivering the first-generation pups

via caesarian section (to avoid a minimal risk of transmission through the placenta),

(4) using noncarrier rats to feed new pups, and (5) permitting future generations of the

pathogen-free pups to interbreed and give birth naturally.

Randall Buck, the vice president of operations for Barney's (exclusive provider of

rats to Petco), testified that he attempted to breed pathogen-free rats. He set up a clean

facility free from the bacteria, and he purchased rats from labs that provided

documentation the rats had tested negative for the streptobacillus moniliformis. He

testified that once the rats left their original lab environments, some of them seemed to

have compromised immune systems and became ill. Although the labs provided

documentation that the rats tested negative for the streptobacillus moniliformis, he was

unsuccessful at purchasing rats that were 100 percent free from the bacteria. He testified

that the only way to be certain a rat did not carry the bacteria was to euthanize the animal

and test brain samples. He also testified that even if rats were bred in a lab to be free

from the bacteria, they might not fare well in a pet store environment because their

immune systems would be compromised. He based this belief on his own experience in

trying to do it.

2. Petco's Policies and Practices

Petco sells about 400,000 rats nationally every year. It makes periodic,

unannounced visits to the facilities of its suppliers, including Barney's, to ensure that the

5
animals are cared for, clean, and well-fed. Rats are not tested for streptobacillus

moniliformis as part of these audits. However, Petco requests that its suppliers provide

rats that do not carry the bacteria.

Rats can be tested for streptobacillus moniliformis through molecular testing,

called a PCR (polymerase chain reaction). Petco does not tell customers the test is

available, but if a customer asks, Petco will segregate the rat, swab its mouth, and send

the swab for testing at Petco's expense. It is not clear from the record how many

customers have requested such testing, but Petco's vice president of veterinary medicine

testified there were two instances he knew of when a customer had made such a request.

Petco will also test a rat if it bites someone. Petco tested a total of 381 rats in 2013, the

year Aidan died, and 196 of them, or 51.4 percent, tested positive for the bacteria.

If a rat bites someone and tests positive for streptobacillus moniliformis, Petco

will euthanize the rat, inform the supplier about the test result, halt further sales of rats

from the same batch (i.e., rats of the same size and gender as the carrier rat), and inform

customers who have already purchased rats from the same batch about RBF and the

positive test result.

Petco warns customers about streptobacillus moniliformis and the dangers of

bacterial transmission in three ways. First, a sign near the rat habitats entitled "Safe

Small Animal Handling" states: "All small animals can carry germs, which may infect

humans." Second, Petco posts "Care Sheets" near its rat habitats, advising customers that

"all rats are potential carriers of infectious diseases, such as Rat Bite Fever . . . ." Finally,

at the point of sale it requires that customers who intend to buy animals sign a

6
Companion Animal Purchase Card, which attests that the customer has reviewed caution

statements on the backside of the card. One such caution statement advises customers of

RBF, the means by which it can be transmitted to humans, and its flu-like symptoms.2

3. Consumer Expectations

Andrew elicited testimony from Aidan's grandmother and several parents of

children who contracted RBF from pet rats purchased from Petco over a period of more

than a decade. Aidan's grandmother testified she expected the pet she purchased to be

healthy and safe for her grandson. She was aware wild rats could carry disease, but she

did not expect a pet she purchased from Petco to be carrying a deadly disease. Other

parents testified they believed rats purchased from Petco would be "disease free" and that

Petco would sell rats that were "safe for children and families." Parents also testified that

they did not have any expectations about the biological makeup of the rat, that

independent research about rats explained there was a chance of getting a disease or

germs from rats, and that a few looked over the warnings, but "didn't exactly read all of

them."

2 It reads: "Rat Bite Fever ('RBF') is a bacterial infection transmissible to people
from rats. Rats are carriers of this bacterium and show no signs of illness, however, they
can pass the bacteria on to people via bites, scratches, or accidental ingestion of
contaminated rat feces. This ingestion can take place by failing to wash your hands after
having contact with the pet or its bedding. Symptoms of infection will usually occur 2-10
days after exposure to an infected rodent and include abrupt onset of chills and fever,
vomiting, pain in the back and joints, headaches and muscle pain. A qualified physician
can make the diagnosis of RBF based on symptoms and testing for the specific strain of
bacteria causing RBF. If a rat bite occurs it would be advisable to consult a physician."
7
 4. Jury Instructions and Special Verdict

Over Andrew's objection, the trial court declined to instruct the jury on the

consumer expectations test, a test by which a trier of fact may find a design defect if "the

circumstances of the product's failure permit an inference that the product's design

performed below the legitimate, commonly accepted minimum safety assumptions of its

ordinary consumers." (Soule v. General Motors (1994) 8 Cal.4th 548, 568-569 (Soule).)

The court opined it would be "too much" to expect a consumer to form minimum safety

assumptions about a pet rat.

The trial court instructed the jury on the risk-benefit test for purposes of design

defect theory. The risk-benefit test asks the trier of fact to determine whether the plaintiff

has proved the product's design proximately caused his injury and, if so, whether the

defendant has established that the benefits of the product's design outweigh the risk of

injury inherent in such a design. (Kim v. Toyota Motor Corp. (2018) 6 Cal.5th 21, 30

(Kim).) The court also instructed the jury on other causes of action, including negligence,

negligent failure-to-warn, strict liability failure-to-warn, and strict liability manufacturing

defect.

The jury found in favor of defendants on all causes of action. On the negligent

failure-to-warn claim, the jury determined that Petco knew or reasonably should have

known that the pet rat was dangerous or was likely to be dangerous when used or misused

in a reasonably foreseeable manner (11-1 vote), and that Petco knew or reasonably should

have known users would not realize the danger (12-0 vote), but found no liability because

Petco did not fail to adequately warn of the danger or instruct on the safe use of the pet

8
rat (10-2 vote). The jury also returned a defense verdict on the ordinary negligence cause

of action (12-0 vote).

The jury found in favor of defendants on each of the strict liability causes of action

as well. For the manufacturing defect claim, the jury decided the pet rat did not contain a

manufacturing defect when it left defendants' possession (11-1 vote). It also concluded

there was no strict liability for a warning defect because, although the pet rat had

potential risks that were known or knowable (11-1 vote), they did not present a

substantial danger to persons using or misusing the pet rat in an intended or reasonably

foreseeable way (10-2 vote). Finally, the jury found Petco not liable for a strict liability

design defect under the risk-benefit test because the risk of the pet rat's design did not

outweigh its design benefits (9-3 vote).

Following entry of judgment, Andrew and Barney's settled. As a result, this

appeal only concerns the judgment in favor of Petco.

DISCUSSION

Andrew challenges the judgment contending the trial court erred by refusing to

instruct the consumer expectations test as a theory on which the jury could conclude

Aidan's pet rat had a design defect. Petco offers three responses. First, it argues the

appeal is moot because Andrew did not appeal the judgment in favor of Barney's, and it

has not been vacated. Petco contends issue preclusion forecloses Andrew from

relitigating some or all issues relating to the design defect claim. Second, Petco argues a

pet rat is not a product under the strict products liability doctrine. Finally, Petco argues

9
the alleged design defect involves "novel and complex scientific matters" incapable of

resolution under the consumer expectations test.

We conclude Petco has not demonstrated mootness on issue preclusion grounds.

We further conclude a live pet sold in its natural form without modification does not

constitute a product under California's strict products liability design defect doctrine.

Because we conclude a pet rat is not a product for purposes of the strict products liability

design defect, we need not reach the remaining issue of the propriety of the consumer

expectations test, but were we to do so, we would conclude it was not improper for the

court to refuse that jury instruction.

I

MOOTNESS

Petco contends the appeal is moot because the judgment entered in Barney's favor,

which Andrew did not appeal, apparently due to the parties' postjudgment settlement,

precludes Andrew from "relitigating" the issues against Petco. It maintains that under the

10
doctrine of issue preclusion, the final judgment precludes Andrew "from pursuing or

continuing to pursue" a design defect theory of liability against Petco.3

We have some difficulty understanding the precise contours of Petco's issue

preclusion argument. Petco does not articulate any particular issue it believes is subject

to preclusion, instead vaguely contending the judgment in Barney's favor resolved

undefined "strict liability issues" that, according to Petco, cannot be now relitigated.

Because Petco does not identify the issue on which it bases this argument, we are inclined

to deem Petco's argument waived. (People ex rel. City of Santa Monica v. Gabriel

(2010) 186 Cal.App.4th 882, 887 ["Absent a more thorough explanation of this issue and

its purported effect, we deem the argument waived."].)

Petco may be contending that the unappealed judgment in Barney's favor

conclusively determined that Aidan's pet rat did not have a design defect; that was the

position Petco took during oral argument. But even if that were the intent, we would

reject the argument on its merits.

3 Petco sometimes indicates the judgment in Barney's favor precludes Andrew from
relitigating a strict products liability claim against Petco. However, Petco relies on the
doctrine of issue preclusion, not claim preclusion, as the basis for its mootness argument.
"It is important to distinguish these two types of preclusion because they have different
requirements." (DKN Holdings LLC v Faerber (2015) 61 Cal.4th 813, 824 (DKN).)
Because Petco provides no analysis or legal authority to support a claim preclusion
argument, we do not consider it. (Road Sprinkler Fitters Local Union No. 669 v. G&G
Fire Sprinklers, Inc. (2002) 102 Cal.App.4th 765, 772, fn. 6 [claim preclusion argument
raised for first time in appellate oral argument waived]; see City of Santa Maria v. Adam
(2012) 211 Cal.App.4th 266, 286-287 [appellant must supply cogent argument supported
by legal analysis; court may disregard arguments that fail to disclose appellant's
reasoning].)
11
 Issue preclusion "prohibits the relitigation of issues argued and decided in a

previous case, even if the second suit raises different causes of action." (DKN, supra, 61

Cal.4th at p. 824.) It applies "(1) after final adjudication (2) of an identical issue

(3) actually litigated and necessarily decided in the first suit and (4) asserted against one

who was a party in the first suit or one in privity with that party." (Id. at p. 825.)

Petco does not satisfy these requirements because Andrew did not litigate to

decision a design defect claim based on the consumer expectations test. The court never

offered the consumer expectations instruction; thus, the design defect issue that was

actually and necessarily decided, the risk-benefits theory, is not identical to the design

defect issue that Andrew asks to be addressed on remand. (Curtis v. State of California

ex rel. Dept. of Transportation (1982) 128 Cal.App.3d 668, 685 [contributory negligence

not actually litigated and decided in prior action because it was "specifically and

deliberately taken away from the jury"].) Accordingly, the mootness argument lacks

merit.

II

UNMODIFIED PET RAT IS NOT A PRODUCT

Petco next argues that the judgment should be affirmed because a live animal

cannot be a "product" for purposes of strict liability.

A. Additional Facts

Prior to trial, the defendants moved for summary adjudication on the ground that a

living animal is not a product for the purposes of strict products liability. The parties

briefed the issue, and the court denied the defendants' motion. The trial court explained:

12
"California courts have not addressed whether an animal is a 'product,' nor have they

adopted the Restatement (Third) for products liability, notwithstanding it has been in

existence since 1989."4

B. Products Liability: General Principles

Summary adjudication is properly granted when the moving party establishes the

right to entry of judgment as a matter of law. (Code Civ. Proc., § 437c, subd. (f).)

Whether a rat is a product for purposes of a strict products liability design defect cause of

action is a legal question, which we review de novo. (Jacks v. City of Santa Barbara

(2017) 3 Cal.5th 248, 273; Buss v. Superior Court (1997) 16 Cal.4th 35, 60; Bialo v.

Western Mutual Ins. Co. (2002) 95 Cal.App.4th 68, 76-77.) We review the decision, not

the court's rationale. (Bialo, at p. 76.)

" 'Products liability is the name currently given to the area of the law involving the

liability of those who supply goods or products for the use of others to purchasers, users,

and bystanders for losses of various kinds resulting from so-called defects in those

products.' " (Merrill v. Navegar, Inc. (2001) 26 Cal.4th 465, 478.) Liability attaches not

only to manufacturers, but also to others in the vertical chain of distribution, including

retailers, licensors, and wholesale and retail distributors. (Bay Summit Community

Assn. v. Shell Oil Co. (1996) 51 Cal.App.4th 762, 773.)

4 California courts have adopted various portions of the Restatement Third of Torts.
(See Johnson v. United States Steel Corp. (2015) 240 Cal.App.4th 22, 31-33 (Johnson)
[quoting the definition of "product"].)
13
 "Strict liability has been invoked for three types of defects—manufacturing

defects, design defects, and 'warning defects,' i.e., inadequate warnings or failures to

warn." (Anderson v. Owens-Corning Fiberglas Corp. (1991) 53 Cal.3d 987, 995.)

"[T]he term defect as utilized in the strict liability context is neither self-defining

nor susceptible to a single definition applicable in all contexts." (Barker v. Lull

Engineering Co. (1978) 20 Cal.3d 413, 427 (Barker).) A product has a manufacturing

defect if it "differs from the manufacturer's intended result or from other ostensibly

identical units of the same product line." (Id. at p. 429.) A warning defect exists if the

manufacturer does not adequately warn the customer "of a particular risk that was known

or knowable in light of the generally recognized and prevailing scientific and medical

knowledge available at the time of the manufacture and distribution." (Anderson v.

Owens-Corning Fiberglas Corp., supra, 53 Cal.3d at p. 1002.) A product has a design

defect if it "fails to meet ordinary consumer expectations as to safety" or "the design is

not as safe as it should be." (Barker, at p. 432.) Regardless of the theory, a plaintiff must

establish "that the object or instrumentality claimed to be defective was in fact a

'product.' " (Brooks v. Eugene Burger Management Corp. (1989) 215 Cal.App.3d 1611,

14
1626.) We focus our analysis here on whether the rats Petco sells are products for

purposes of design defect theory.5

California law has not addressed whether an animal is a product for purposes of

strict liability. However, both the Second and Third Restatements of Torts provide some

guidance regarding what constitutes a product for strict liability claims, and whether we

apply section 402A of Restatement Second of Torts (hereafter Second Restatement) or

section 19 of the Restatement Third of Torts (hereafter Third Restatement), we reach the

same conclusion: a pet rat carrying the streptobacillus moniliformis bacteria is not

subject to a products liability design defect claim.

" ' "The Restatement . . . does not constitute a binding authority, but, considering

the circumstances under which it has been drafted, and its purposes, in the absence of a

contrary statute or decision in this state, it is entitled to great consideration as an

argumentative authority." ' " (Brady v. Calsol, Inc. (2015) 241 Cal.App.4th 1212, 1225.)

Although California has not adopted the Third Restatement, California courts have in

some circumstances cited approvingly to portions, including the definition of "product"

5 We requested supplemental briefing regarding whether the existence of
streptobacillus moniliformis in rats, to the extent it could be considered a defect, is a
manufacturing defect rather than one of design. "In general, a manufacturing or
production defect is readily identifiable because a defective product is one that differs
[(1)] from the manufacturer's intended result or [(2)] from other ostensibly identical units
of the same product . . . ." (Barker, supra, 20 Cal.3d at p. 429.) Both parties responded
that the alleged failure of the product, if it is a defect, is a design defect and not a
manufacturing defect. Without offering any opinion on the question, we accept the
parties' agreement for purposes of this appeal.
15
found in section 19. (Johnson, supra, 240 Cal.App.4th at p. 31; Jenkins v. T&N PLC

(1996) 45 Cal.App.4th 1224, 1229 [citing definition of "product" in draft Restatement].)

Section 19 of the Third Restatement defines a product for purposes of strict

liability as "tangible personal property distributed commercially for use or consumption,"

excluding human blood and tissue. (Rest.3d Torts, Products Liability, § 19, subds. (a),

(c).) Comment b to this section recognizes the division among courts regarding whether

living animals should be considered products for determining a commercial seller's

liability in tort and states that "when a living animal is sold commercially in a diseased

condition and causes harm to other property or to persons, the animal constitutes a

product for purposes of this Restatement." (Id., § 19, com. b.)

The definition offered in section 19 of the Third Restatement applies to all product

liability claims, but we focus on the narrower question of whether an item is a product for

the purpose of design defect claims. Comment b to the Third Restatement centers its

discussion on diseased animals, noting that diseased animals that are themselves harmed

are not eligible for recovery under the Third Restatement, while animals in a "diseased

condition" that cause harm to other property or to persons are products for which

damages are available under the Third Restatement. We focus on the issue of disease

because it is the disease from which the claim of design defect arises; absent a designed

product, there can be no design defect claim.

The dissent accuses us of failing to "address the reality" that a pet sold at a pet

store is—diseased or not—" 'tangible personal property distributed commercially for use

or consumption,' " thereby falling within the definition of "product" taken from section

16
19 of the Third Restatement. (Dis. opn., at pp. 5-6.) This misunderstands our point,

which is that the comment recognizes nuances involved with living animals. Comment b

to the Third Restatement does not expressly distinguish "healthy" living animals from

ones in a "diseased condition"; however, it does make clear that an animal in a "diseased

condition" is a defective product. Moreover, although it acknowledges the division

among courts regarding how to characterize living animals, comment b to the

Restatement chooses only to address animals "in a diseased condition" as products, even

though some of courts have held living animals are not defective products when the

"defect" derives from the animal's personality or dangerous propensity. (See, e.g.,

Blaha v. Stuard (S.D. S.Ct. 2002) 640 N.W.2d 85, 88-89 (Blaha) [pet dog bit child];

Kaplan v. C Lazy U Ranch (D.Colo. 1985) 615 F.Supp. 234, 237 (Kaplan) [biting and

kicking horse].) The animal's "diseased condition" is the evidence of a defective product

that brings it under Third Restatement. (See also Rest.3d Torts, supra, § 21, cmts. d-f

[discussing harms caused by defect in products as establishing eligibility for recovery

under Third Restatement].) We likewise concentrate on the role of "disease" in

determining whether a pet rat living in its natural condition, without modification, can be

a defective product.

Andrew suggests that the word "diseased" should be considered from the

perspective of the human who ultimately contracts RBF. However, "diseased" is an

adjective modifying the condition of the living animal. Neither party defines "diseased,"

but Webster's Third New International Dictionary defines it as "affected with or as if with

a disease[;] lacking health or soundness[;] sickly . . . ." (Webster's 3d New Internat. Dict.

17
(2002) p. 648.) Thus, under the Third Restatement, in assessing whether the rat here is a

product, we ask whether a rat carrying streptobacillus moniliformis is in a diseased

condition, i.e., affected with a disease or lacking in health or soundness. It is not, and the

parties do not disagree on this point.

Andrew's expert Dr. Vinetz explained that RBF, developed by some humans after

exposure to the streptobacillus moniliformis bacteria, is a disease of humans, not of rats.

He testified streptobacillus moniliformis bacteria is a nonessential bacteria not part of a

rat's natural flora, but instead operating as a commensal organism that resides in the

upper respiratory tract section, or the upper oropharyngeal area.6 As a commensal

organism, the bacteria adapts to the rat and the rat adapts to it. The bacteria does not

cause a disease in the rat that makes the rat sick. Thus, an animal carrying the bacteria

also does not appear sick; it has glossy fur, bright eyes, and looks alert. This testimony

makes clear that the streptobacillus moniliformis bacteria can be disease-causing in

humans, but the carrier rat is not diseased. Thus, where an animal (i.e., the rat) coexists

with an organism (streptobacillus moniliformis bacteria) and that organism causes no

harm to the animal, the animal itself is not diseased and is therefore not a product.

California has adopted outlines of the Restatement Second of Torts, section 402A.

(Cronin v. J.B.E. Olson Corp. (1972) 8 Cal.3d 121, 131.) Section 402A provides that a

6 Defense expert Dr. Sean Elliott, a pediatric infectious disease specialist, testified
that streptobacillus moniliformis is "normal flora," or part of a rat's normal intestinal
bacteria. We are not influenced by this apparently-contradictory testimony because even
considering only Dr. Vinetz's testimony, we reach the same conclusion about how to
characterize a rat carrying streptobacillus moniliformis.
18
party that sells a product in a defective condition is liable for the harm thereby caused to

the consumer or end user of the property as long as the seller is engaged in the business

of selling the product and the product is "without substantial change in the condition in

which it is sold." (Rest.2d Torts, supra, § 402A, subd. (1)(b).) However, its text and

comments are silent regarding the application of strict liability to live animals. (See

Rest.2d Torts, supra, § 402A.)

Courts considering strict liability claims under the Second Restatement, section

402A have reached conflicting conclusions about whether diseased animals and animals

with dangerous personalities are products. One line of cases follows a 1977 New York

opinion, Beyer v. Aquarium Supply Co. (Div. of Hartz Mountain Corp.) (N.Y. 1977) 404

N.Y.S.2d 778 (Beyer), which concluded that public policy supports holding accountable a

breeder, distributor, or vendor who places a diseased animal in the stream of commerce.

(Id. at p. 779 [diseased hamster]; see also Sease v. Taylor's Pets, Inc. (Or. App. 1985) 700

P.2d 1054 (Sease) [rabid skunk]7; Worrell v. Sachs (Conn. 1989) 563 A.2d 1387, 1387

(Worrell) [parasite-carrying puppy].)8 In Beyer, the court compared a diseased hamster

to a complex manufactured product and concluded the risk to human health from a

7 In Sease, the court reasoned that a living creature could be a product because there
was no requirement that a product be manufactured or processed, and Oregon statutes
"cover[ed] products subject both to natural change and intentional alteration," providing
an explicit defense when the seller can show the product is not in substantially the same
condition as it was when it was sold. (Sease, supra, 700 P.2d at p. 1058.)

8 The Worrell court concluded the animal's status as a product should be considered
separately from whether it had a defect and viewed a disease as a defect. (Worrell, supra,
563 A.2d at p. 1388.)
19
diseased animal is at least as great as that posed by a manufactured product.9 (Beyer, at

p. 779.) These cases largely concluded that a seller could control an animal's diseased

condition up to the point of sale or to address public health concerns, but they are

distinguishable from the present case because unlike the diseased animals in those cases,

rats carrying streptobacillus moniliformis are not diseased. Additionally, although

streptobacillus moniliformis may pose a risk to human health, as we explain ante, that

alone does not justify application of the design defect theory of strict liability, particularly

where—as here—other available theories of liability achieve similar goals.

Another line of cases follows Illinois decisions which have held that living

animals are not products because they are participants in "a constant interaction with the

environment around them as part of their development" and live "in a constant process of

internal development and growth." (Anderson v. Farmers Hybrid Cos. (Ill. App. 1980)

408 N.E.2d 1194, 1199 (Anderson); see also Blaha, supra, 640 N.W.2d at pp. 88-89 [pet

dog bit child]; Kaplan, supra, 615 F.Supp. at p. 237 [biting and kicking horse];

Zendejas v. Redman (S.D.Fla. June 13, 2017, No. 15-81229-CIV) 2017 U.S.Dist.Lexis

90503, *4, *19-*20 (Zendejas) [show horse that refused to jump].) These courts have

noted the animals are not "designed, assembled, fabricated . . . , produced, [or]

constructed" (Kaplan, at p. 238); they are "living, breathing, sentient creature[s] that

constantly grow[], learn[], experience[], and change[] based on both external and internal

factors that often cannot be controlled by the seller . . . ." (Zendejas, at p *19.) Under

9 The Beyer court did not address the mutable nature of animals. (Beyer, supra, 404
N.Y.S.2d at p. 779.)
20
this line of reasoning, live animals are not products for purposes of strict liability because

there is no way to attribute their behavior solely to their status at the point of sale, as their

behavior is influenced by the surrounding environment. At the same time, these cases

recognize that the animals sold in their natural state are not "designed." (Kaplan, at

p. 238.)

Courts have also used mutability to explain why diseased animals are not products

for purposes of strict liability. For example, in Anderson plaintiffs purchased female

breeding pigs that had a contagious and infectious disease which infected the buyers'

swine herd, requiring expensive treatment and leading to the death of several members of

the herd. (Anderson, supra, 408 N.E.2d at pp. 1195-1196.) The appellate court

concluded that to be a product, the good's nature must be fixed at the time it leaves the

manufacturer's or seller's control, and the good must reach the user without substantial

change. (Id. at p. 1199.) The court recognized that imposing strict liability helps insure

costs of injury are borne by those who market the product rather than an injured person

powerless to protect himself or herself. (Ibid.) However, it concluded that purpose is

defeated if liability is imposed on those goods whose character is "easily susceptible" to

changes beyond the control of the seller. (Ibid.) In those situations, neither party has the

power to protect against the good's changing nature; thus, it would be unfair to force the

burden on the seller over the user.

Ohio and Connecticut courts have similarly held that sick birds are not products.

In Malicki v. Koci (Ohio App. 1997) 700 N.E.2d 913 (Malicki), the court considered

21
whether a parakeet that transmitted psittacosis, or "parrot fever"10 to its new owners was

a product. The court held it was not, and it explained: "Imposing strict liability upon the

defendants in this case would yield the harsh result of holding them responsible as

absolute insurers of the health of a living organism whose health can be affected by many

factors totally outside the defendant's control." (Id. at p. 915.) Similarly, in Latham v.

Wal-Mart Stores, Inc. (Mo. App. 1991) 818 S.W.2d 673, a Missouri appellate court

considered whether a parrot that transmitted psittacosis was a product. The court decided

it was not: "It seems unreasonable for us to hold a seller liable for changes potentially

wrought upon a 'product' by the purchaser, while the item was completely outside the

seller's control." (Id. at p. 676.) The mutability in these cases addressed the influence of

environmental factors on the physical health of the animals rather than their personalities

or nature, but the theory remains the same: the seller cannot control outside influences or

the internal biological forces that impact the health or dangerousness of the animal.

The dissent argues that none of these authorities adopts our reasoning that only

diseased animals displaying symptoms of illness are a product. (Dis. opn., at p. 9.) This

mischaracterizes our reasoning and our conclusion. Pet rats in their natural state, even

carrying the bacteria, are neither diseased nor designed. Rats are susceptible to their

10 "Parrot fever" is caused by chlamydia psittaci. The bird in Malicki appeared
healthy upon visual inspection, and the seller's expert veterinarian testified that bird's
tissue samples showed no signs of psittacosis. (Malicki, supra, 700 N.E.2d at p. 914.)
The virologist from the Animal Disease Diagnostic Lab at Ohio State University testified
the bird was in a subclinical phase of psittacosis and had acted as a carrier of the disease.
(Ibid.) Unlike streptobacillus moniliformis, which is carried by rats as a normal
gastrointestinal, respiratory, and oral secretions and rarely affects the rat, chlamydia
psittaci can cause symptoms in the bird.
22
environment because nothing prevents a rat from contracting the bacteria from other rats

or the environment where the bacteria is found. To hold the seller responsible under the

design defect theory would make the seller an absolute insurer of the rat's health,

biological condition, and even behavior, even though those things are affected by factors

beyond the scope of the seller's control. (See Malicki, supra, 700 N.E.2d at p. 915.)

Additionally, even accepting that retailers and manufacturers are in the best position to

bear the cost of injuries resulting from the goods they sell (see Jimenez v. Superior Court

(2002) 29 Cal.4th 473, 477 (Jimenez)), we do not foreclose equitable cost spreading or

the apportionment of losses with our conclusion here that rats are not products for

purposes of design defect theory. Those goals can be achieved through warning and

manufacturing claims.

A principal purpose behind strict products liability is to relieve injured parties of

onerous evidentiary burdens required in negligence causes of action. (Barker, supra, 20

Cal.3d at p. 431.) In design defect cases, once the plaintiff shows injury caused by a

product's design, the burden shifts to a defendant to prove the product is not defective.

(Ibid.) But this presumes there is a design to the product in the state in which it is sold,

and as we have explained pet rats living in their natural state are not designed.

We also recognize that manufacturers and retailers are generally in a strategic

position to promote the safety of the products they design. (Jimenez, supra, 29 Cal.4th at

pp. 477-478; Union Pac. Co. v. S. Cal. Edison Co. (1985) 163 Cal.App.3d 700, 709.) In

general, strict product liability promotes safer products by placing the responsibility on

the party in the best position to prevent the harm caused by their danger or defect, thereby

23
incentivizing proper design and quality control. (Rest.3d Torts, supra, § 2, com. a.)

However, here the offending item is a creature living in its natural state.

Finally, Andrew separately argues an animal's health is not relevant to its status as

a product for purposes of strict liability. (See Worrell, supra, 563 A.2d at p. 1387

[concluding an animal's mutability is an issue of proximate causation to be considered

separately from the animal's status as a product]; Harvey, The Applicability of Strict

Products Liability to Sales of Live Animals (1982) 67 Iowa L. Rev. 803, 815 [arguing it is

an analytical flaw to base product status on product mutability because they concern

separate elements of strict liability] (hereafter Harvey).) Under this theory, a good's

status as a product for purposes of strict liability is evaluated separately from the

requirement that it undergoes a substantial change after the seller surrenders its control.11

(Harvey, supra, 67 Iowa L. Rev. at pp. 815, 819-820.)

The difficulty with setting apart the defect from the definition of the product here

is that the policy rationale underpinning strict liability for design defect relies on the good

being defective to justify its application. For example, the cost allocation policy asks

how to appropriately allocate the cost of injury resulting from the defect. (See Barker,

supra, 20 Cal.3d at p. 431.) Here, we never reach the question of whether the bacteria is

11 One law review article promoting this approach recommends treating live animals
as products, attributing to them a presumption of mutability that results in the relevant
defect, then requiring a plaintiff to prove the animal did not experience any substantial
change after leaving the seller, i.e., the defect was present at the time of sale. (Harvey,
supra, 67 Iowa L. Rev. at p. 821.) However, this approach would be problematic in the
case before us, even if we were to consider the rat carrying the streptobacillus
moniliformis bacteria to be diseased because there is an incubation period for animals'
contraction of the bacteria; the PCR test can be negative one day and positive the next.
24
a defect because we have concluded it is not part of a design and does not make the rat

diseased.

Still, we understand that a consumer may be unable to recognize and take

additional precautions against disease-causing bacteria an animal may carry, particularly

here, where 10 to 100 percent of pet rats carry the bacteria. The real risk is contracting

RBF without treating it. RBF is rare; there are approximately 200 documented cases in

United States history, and Petco alone sells about 400,000 rats per year. Dr. Vinetz

estimated that between zero and 10 people are admitted annually to the hospital for RBF,

but that it can sometimes resolve on its own, too.12 Those risks are better addressed by

negligence, negligent warning, and warning and manufacturing strict liability causes of

action, all of which were available to the plaintiffs and considered by the jury here.

III

APPLICABILITY OF CONSUMER EXPECTATIONS TEST

Andrew requested that the trial court instruct the jury on two alternative tests to

determine whether Aidan's pet rat suffered from a design defect—the risk-benefit test and

the consumer expectations test. The court provided instructions on the risk-benefit test

12 According to the Center for Disease Control, the mortality rate for untreated RBF
is 7 to 13 percent. In closing arguments, Petco's counsel argued: "Five million rats, 45
alleged claims of rat-bite fever. And the likelihood of harm occurring, again, .0009
percent . . . . [¶] . . . [¶] Are there risks of rat-bite fever? Very, very, very minimal risks.
Very, very, very minimal. The percentages are infinitesimal. They're tiny. [¶] . . . [¶]
Did the potential risk present a substantial danger to persons using or misusing the pet? I
would suggest to you that the plaintiffs have not proven there is a substantial danger that
you're going to get rat-bite fever from a rat . . . ." Even if Petco sales were attributed for
the entire universe of documented RBF cases, our calculations indicate the mortality rate
would be less than .0065 percent.
25
but not the consumer expectations test. Andrew contends this was error. Petco defends

the court's decision on grounds that a pet rat is inherently complex and evaluating the

circumstances of the alleged failure in this case requires scientific and medical

knowledge beyond the understanding of ordinary consumers.

Having concluded a pet rat carrying streptobacillus moniliformis is not a product

for purposes of strict liability, design defect, there is no need to consider whether the

court properly instructed the jury regarding the available tests. However, were we to

draw a conclusion on this issue, we would affirm the court's approach.

A. Consumer Expectations Test: General Principles

We review the propriety of jury instructions de novo. (Cristler v. Express

Messenger Systems, Inc. (2009) 171 Cal.App.4th 72, 82.) A plaintiff may prove the

existence of a design defect under either of two alternative tests, the consumer

expectations test (if applicable) or the risk-benefit test.13 (Barker, supra, 20 Cal.3d at

p. 432.) Under the consumer expectations test, the plaintiff may prove a design defect by

showing that the product failed to perform as safely as an ordinary consumer would

expect when used in an intended or reasonably foreseeable manner. (Ibid.) "The

rationale of the consumer expectations test is that '[t]he purposes, behaviors, and dangers

of certain products are commonly understood by those who ordinarily use them.'

[Citation.] Therefore, in some cases, ordinary knowledge of the product's characteristics

13 The Third Restatement rejects the consumer expectations test in favor of the risk-
benefit test and whether there was a better alternative design available. (Webb v. Special
Electric Co., Inc. (2016) 63 Cal.4th 167, 184, fn. 8, citing Rest.3d Torts, supra, § 2,
com. g.)
26
may permit an inference that the product did not perform as safely as it should."

(Saller v. Crown Cork & Seal Co., Inc. (2010) 187 Cal.App.4th 1220, 1232 (Saller).)

The consumer expectations test considers minimum safety expectations of the product's

users, not the expectations of the general consuming public. (Demara v. The Raymond

Corp. (2017) 13 Cal.App.5th 545, 559.)

Alternatively, under the risk-benefit test, plaintiffs can prove a design defect if

they show the product's design proximately caused an injury and the defendant fails to

show the design's benefits outweigh the risk of danger inherent in the design. (Barker,

supra, 20 Cal.3d at p. 432.) When weighing the benefits and risks, a jury may consider a

nonexhaustive list of factors, including the gravity of the potential harm resulting from

the use of the product, the likelihood the harm would occur, the feasibility of an

alternative safer design at the time of manufacture, and the financial cost and

disadvantages of an alternative design. (Kim, supra, 6 Cal.5th at p. 30.)

"The tests are not mutually exclusive, and a plaintiff may proceed under either or

both." (McCabe v. American Honda Motor Co. (2002) 100 Cal.App.4th 1111, 1126

(McCabe).) However, the consumer expectations test is reserved for cases in which "the

everyday experience of the product's user permits a conclusion that the product's design

violated minimum safety assumptions," regardless of the merits of the design. (Soule,

supra, 8 Cal.4th at p. 567.) The inherent complexity of a product is not controlling on

whether the consumer expectations test applies (Saller, supra, 187 Cal.App.4th at

p. 1232), but a complex product can "cause injury in a way that does not engage its

ordinary consumers' reasonable minimum assumptions about safe performance." (Soule,

27
at pp. 566-567.) In such cases, the consumer expectations test does not apply. (Id. at

p. 568; see also Morson v. Superior Court (2001) 90 Cal.App.4th 775, 793-795

(Morson).)

B. Inapplicability of Consumer Expectations Test

Petco contends the evidence adduced at trial was insufficient to warrant a jury

instruction on the consumer expectations test because the alleged failure of the product at

issue was "esoteric and inherently complex," and it could be properly understood only

with knowledge of "the biological mechanisms of pathogens, infectious diseases, and

technicalities of rat breeding for commercial and laboratory purposes." The trial court

accepted this argument, finding that RBF and the circumstances of the pet rat's alleged

failure were "too much" for an ordinary consumer to understand.

We assume for purposes of discussion that the biology of rats and the pathology of

RBF are complex topics. We recognize that the consumer expectations test can apply to

complex products, depending on the circumstances of their alleged failure. (Saller,

supra, 187 Cal.App.4th at p. 1232.) But when the ultimate issue of design defect calls for

a "careful assessment of feasibility, practicality, risk, and benefit, the case should not be

resolved simply on the basis of ordinary consumer expectations." (Soule, supra, 8

Cal.4th at p. 562.) "The critical question is whether the 'circumstances of the product's

failure permit an inference that the product's design performed below the legitimate,

commonly accepted minimum safety assumptions of its ordinary consumers.' " (McCabe,

supra, 100 Cal.App.4th at p. 1122, quoting Soule, at pp. 568-569.) However, "[s]ome

products cause injury in a way that does not engage its ordinary consumers' reasonable

28
minimum assumptions about safe performance." (Id. at p. 1122.) For example, when a

consumer has no idea how safe a product should be made against all foreseeable hazards,

that is, in those cases where the plaintiff's theory of defect seeks to examine obscure

behavior that falls outside the ordinary experience of the consumer, the consumer

expectation test is inapplicable, and the defect can only be proved by the risk-benefit

analysis. (Ibid.)

Because the design defect is the existence of the bacteria causing RBF, central to

the question here is whether the bacteria causing RBF can be designed out of rats, and a

consumer's ordinary experiences do not introduce it to the potential complexities of

designing a rat to be free from streptobacillus moniliformis, a bacteria which is not

readily visible in the pet rat.

Andrew's infectious disease expert Dr. Vinetz told the court RBF could not be

bred out of the rats because it was not a genetic condition. As an alternative, he testified

it would be "simple" to design a rat free from the bacteria that causes RBF, and he

described the process: The breeder would conduct a PCR to test the rats before their

breeding, only allow rats that tested negative for the bacteria to breed, deliver the pups by

cesarean section, then use rats that had tested negative to feed the new pups in a clean

facility. Dr. Vinetz testified there could be a zero percent incident rate of RBF in a

"highly controlled setting." He pointed to a research lab which provided data indicating

that over an 18-month period, it had not detected streptobacillus moniliformis in the rats

housed there. That lab bred and housed rats in its research facility; those rats never left

the facility and lived in other environments.

29
 Defendants offered testimony regarding their unsuccessful attempts to "design"

rats without streptobacillus moniliformis through the breeding process, by purchasing rats

with documentation that they were negative for the bacteria, placing them in a separate,

clean facility, and breeding them. These efforts resulted in some offspring that tested

positive for streptobacillus moniliformis. Defendants also offered testimony that lab rats

were not always healthy and did not always fare well in the pet store environment.

Although the parties disagreed about the viability of designing rats without

streptobacillus moniliformis, both parties agreed that PCR testing could be used to assess

whether a rat carried the bacteria with 99 percent sensitivity, but the only way to be

certain a rat was free from streptobacillus moniliformis was through testing of tissue

samples after euthanizing the animal.14 Additionally, Dr. Vinetz testified that healthy

domesticated or lab rats demonstrate streptobacillus moniliformis 10 to 100 percent of the

time.

We draw two conclusions from this evidence. First, there is no way to be certain

that any given, living rat is free from the bacteria that causes RBF. Second, it is unclear

14 Alex, the rat, was euthanized before it was tested for streptobacillus moniliformis,
and the record indicates that the initial swab was negative for the bacteria. Dr. Vinetz
testified that Alex's swab was taken from the nasopharynx instead of the oropharynx, and
he suggested the reason it came back negative was that the swab reflected the nasal
turbinates, when the bacteria is transmitted through "spit," not "snot." Subsequent tissue
testing found the bacteria in the rat's tongue, adrenal gland, several lymph nodes, and the
sciatic nerve. However, the salivary gland, soft palate, and trachea, and gastrointestinal
system were negative for streptobacillus moniliformis. Dr. Vinetz believed an oral swab
would have been positive if it had been performed.
One other witness testified the rat presumed to have caused him RBF tested
negative for RBF.
30
whether it is within Petco's or any supplier's control to design a rat free from

streptobacillus moniliformis. Because there is competing evidence about the

designability of a rat free from the bacteria, this is more properly a question for the jury.

However, even if the jury accepts Dr. Vinetz's testimony, this merely demonstrates there

may be a better, alternative design than the one in which rats naturally exist. This type of

alternative design is one for a jury to consider when weighing a product's benefits against

its risks to determine whether there is a design defect at all. (See Barker, supra, 20

Cal.3d at pp. 431-432.)

Ultimately, we would conclude that ordinary experience and understanding absent

a warning would not inform a consumer regarding how safely a rat should perform under

any particular set of circumstances because rats contract streptobacillus moniliformis

without notice or symptom. As a practical matter, here it is "impossible to eliminate the

balancing or weighing of competing considerations in determining whether a product is

defectively designed or not. . . ." (Barker, supra, 20 Cal.3d at p. 433.)

Additionally, as we explain post, "the issue of design defect cannot be resolved by

standardless reference to the 'expectations' of an 'ordinary consumer' " (Soule, supra, 8

Cal.4th at p. 567, citing Barker, supra, 20 Cal.3d at pp. 433-434) because the warning

here regarding the potential occurrence of RBF contributes to the " 'legitimate, commonly

accepted minimum safety assumptions' of . . . ordinary consumers." (Morson, 90

Cal.App.4th at p. 792, quoting Soule, at p. 569.)

31
C. Plaintiff Suffered No Prejudice

Petco maintains that even if instructions on the consumer expectations test should

have been given, the failure to do so was not prejudicial in light of the jury's conclusions

that it was not liable on either a negligence or strict liability failure-to-warn theory. We

would be inclined to agree.

1. Additional Facts

The jury was instructed on two failure-to-warn theories, strict liability and

negligence. With respect to the strict liability failure-to-warn claim, the jury answered

"yes" to the first question: "Did the pet rat have potential risks that were known or

knowable in light of the scientific and medical knowledge that was generally accepted in

the scientific community at the time of the sale?" It answered "no" to the next question:

"Did the potential risks present a substantial danger to persons using or misusing the pet

rat in an intended or reasonably foreseeable way?" With respect to strict liability, the

jury never reached the questions of whether an ordinary consumer would have recognized

potential risks and whether Petco failed to adequately warn or instruct of potential risks.

With respect to negligent failure to warn, the jury answered "yes" to the first

question: "Did Petco Animal Supplies Inc. know or should it reasonably have known that

the pet rat was dangerous or was likely to be dangerous when used or misused in a

reasonably foreseeable manner?" It answered "yes" to the second question: "Did Petco

Animal Supplies know or should it reasonably have known that users would not realize

the danger?" But it answered "no" to the third question: "Did Petco Animal Supplies

Inc. fail to adequately warn of the danger or instruct on the safe use of the pet rat?"

32
 2. Instructional Error: General Principles

The failure to give a requested instruction in a civil suit will not result in a reversal

unless the error was prejudicial, and the appealing party must show a "reasonable

probability" that the error affected the result. (See Soule, supra, 8 Cal.4th at pp. 580-582;

People v. Watson (1956) 46 Cal.2d 818, 836.) In this context, "probability" means "a

reasonable chance, more than an abstract possibility." (Kinsman v. Unocal Corp. (2005)

37 Cal.4th 659, 682; College Hospital Inc. v. Superior Court (1994) 8 Cal.4th 704, 715.)

In assessing an instructional error's prejudicial impact, we consider the " 'natural and

probable effect on a party's ability to place his full case before the jury,' " and the

likelihood of actual prejudice, considering the state of the evidence, effect of other

instructions, effect of counsel's arguments, and any indications by that it was misled.

(Veronese v. Lucasfilm Ltd. (2012) 212 Cal.App.4th 1, 31; see Soule, at pp. 580-581.)

3. Harmless Error

The parties agree that all evidence related to the consumer expectation test was

presented to the jury. Andrew presented testimony from multiple witnesses about their

expectations when purchasing pet rats from Petco, including testimony from Aidan's

grandmother, who purchased Alex. Plaintiff's counsel asked jurors to consider this

testimony in evaluating a consumer's expectations in the context of the failure-to-warn

claim.

The dispute turns on whether it is reasonably probable, in light of the jury's

verdicts on the other causes of action, that the jury would have returned a verdict for

Andrew under the consumer expectations test. Taking into consideration the state of the

33
evidence, the effect of other instructions, the effect of counsel's arguments, and the

special verdicts, were we considering prejudicial effect of the exclusion of the consumer

expectations test jury instruction, we would conclude its exclusion was harmless. The

jury returned verdicts in Petco's favor on both the strict liability failure-to-warn and the

negligent failure-to-warn causes of action. Andrew contends he suffered prejudice

because the consumer expectations test is a distinct cause of action, unaffected by other

claims and their corresponding instructions. While we recognize these other causes of

action are distinguishable from the consumer expectations test, their interplay with the

ordinary consumer's expectation informs our conclusion.

The jury received instructions for the negligent and strict liability failure-to-warn

causes of action, as well as for the risk-benefit design defect claim. In general, a retailer

is strictly liable for failure to warn if the warning was feasible and its absence caused the

injury; reasonableness of the failure to warn is immaterial. (Webb v. Special Electric Co.,

Inc. (2016) 63 Cal.4th 167, 180.) A negligent failure to warn cause of action differs

because the plaintiff must prove the defendant's conduct fell below the standard of care;

thus, if a seller could reasonably decline to offer the warning, the failure to do so is not

negligent. (Id. at p. 181.)

In reaching the strict liability failure-to-warn verdict, the jury concluded the pet rat

had potential risks known and accepted in the scientific community at the time of the

sale, but those risks did not present a substantial danger to persons using the rat in an

intended or reasonably foreseeable way. The jury did not reach the question of whether

an ordinary consumer would have recognized potential risks or whether Petco failed to

34
adequately warn or instruct regarding those risks because it concluded they did not pose a

substantial danger.

In reaching the negligence verdict, however, the jury did answer those questions.

It concluded that Petco knew a pet rat was likely to be dangerous, and that it should have

known its users would not realize the danger. In so holding, the jury implicitly

recognized that absent a warning, the ordinary consumer would not recognize the

potential danger wrought by a pet rat. However, the jury nonetheless concluded that

Petco did not fail to adequately warn consumers about the danger. In other words, taking

these verdicts together, the jury concluded pet rats posed a danger that the ordinary

consumer would not realize, but the danger was not substantial, and Petco supplied an

adequate warning to alert consumers to the danger. Given this conclusion, it is not

probable that the addition of the consumer expectations instruction would have resulted

in a different outcome; to hold otherwise would be to conclude it is probable the jury

would have rendered an inconsistent verdict.

Andrew explains negligent and strict liability failure-to-warn causes of action are

separate and distinct claims that do not collapse into each other merely because a plaintiff

alleges both. He then argues that just as negligent and strict liability causes of action are

different, so too are failure-to-warn and design defect causes of action. And further, the

conclusion regarding one cause of action cannot predict the jury's conclusion regarding

the other. In essence, Andrew argues that one cannot speculate as to whether adequate

warnings help frame a consumer's expectations.

35
 Andrew cites to Saller to support this argument. There, the plaintiff filed design

defect and negligent and strict liability failure-to-warn claims against the manufacturer of

insulation containing asbestos. The trial court instructed the jury on the risk-benefit test

and a negligent failure-to-warn theory but refused the instructions for the consumer

expectations test and strict liability failure-to-warn. (Saller, supra, 187 Cal.App.4th at

p. 1230.) The jury returned defense verdicts for the claims on which it was instructed.

(Id. at pp. 1230-1231.) On appeal, the plaintiff argued the court should have instructed

on the consumer expectations test and strict liability failure-to-warn claim. (Id. at

p. 1231.) In response, the manufacturer asserted that the plaintiff did not present

evidence of what an ordinary consumer would expect, and any alleged error was harmless

due to the defense verdicts on the risk-benefit design defect and negligent failure-to-warn

claims. (Ibid.) The Court of Appeal rejected these arguments and reversed. (Id. at

pp. 1236, 1239.)

Saller is distinguishable because it answered a different question than the one

before us now. The court did not consider whether a jury's conclusion about the

adequacy of a warning would impact the verdict on a consumer expectations test in light

of the given warning; no warning was offered to consumers in Saller at all. (Saller,

supra, 187 Cal.App.4th at p. 1240.) The court concluded that the findings on negligent

failure-to-warn did not preclude a different outcome under the strict liability failure-to-

warn cause of action. (Ibid.)

36
 Moreover, the issue here is not whether warning and design defect causes of action

are distinct theories;15 it is whether the inconsistent outcome that plaintiffs propose

makes it less than reasonably probable the outcome here would be different taking into

consideration the existence and adequacy of the warning that contributes to a consumer's

expectations. The parties agree that if verdicts that simultaneously conclude consumers

were adequately warned of known risks and that consumers would reasonably expect the

product to perform more safely than the warnings indicate are contradictory, they would

be fatally contradictory.16

Andrew maintains that were the instruction offered, the jury would probably have

found in his favor because the consumer expectations test does not consider the

complexity of the product, the magnitude of the danger, the feasibility of a warning, or

the reasonability of the seller. It merely considers the failure of the product without other

considerations. But this argument ignores the influence of the warning on a consumer's

15 Failure-to-warn and designed defect based on the consumer expectations test are
distinct causes of action, and plaintiffs may allege both. (Arnold v. Dow Chemical Co.
(2001) 91 Cal.App.4th 698, 717; Soule, supra, 8 Cal.4th at p. 567.)

16 Andrew explains that such an inconsistent verdict is reversible error, necessitating
a new trial. (See, e.g., City of San Diego v. D.R. Horton San Diego Holding Co., Inc.
(2005) 126 Cal.App.4th 668, 682 [inconsistent verdict rule requires a jury to resolve all
controverted issues].)
37
expectation.17 (See, e.g., Hansen v. Sunnyside Products, Inc. (1997) 55 Cal.App.4th

1497, 1501 [recognizing the relevance of warnings under the risk/benefit test]; see also

Rest.3d Torts, supra, § 2, com. g [explaining a consumer's expectations are influenced by

how products are portrayed and marketed].)

Andrew nonetheless contends that customers expect the rats to be healthy and

points to testimony by consumers and Petco employees about those expectations. This

testimony alone does not lead to the conclusion that it is reasonably probable the jury

would have reached a different verdict under the consumer expectations test. There was

also testimony indicating consumers had no expectations about rat biology or that they

weighed risks because they had general knowledge about the potential for disease but

purchased pet rats anyway.18

The dissent also points to testimony of consumer expectations about the safety or

health of pet rats based on factors such as Petco's slogan and pet-related custom and

17 Although no published California cases have expressly held that warning
information helps form the basis of the ordinary consumer's expectations, other
jurisdictions have drawn that conclusion. (Ford Motor Co. v. Trejo (Nev. 2017) 402 P.3d
649, 656; Knitz v. Minster Mach. Co. (Ohio Ct.App. 1987) 1987 Ohio App. Lexis 5828,
at *125; Hoffman v. Hercules Chem. Co. (N.D.Ill. 2004) 2004 U.S. Dist. Lexis 22505, at
*24).) Where there are federal, statutorily-mandated warnings, such warnings have
precluded liability under the consumer expectations test because the ordinary consumer's
expectation has been preempted by the statutory warning. (See Papike v. Tambrands Inc.
(9th Cir. 1997) 107 F.3d 737, 743 [explaining consumer not entitled to expect product to
perform more safely than government-mandated warnings and concluding warnings met
federal requirements, so design defect claim failed consumer expectations tests].)
However, we are not suggesting that a product labeled with an adequate warning will
automatically fail the consumer expectations test under California tort law, only that the
warning will inform a consumer's expectations.
18 One witness testified that he returned to Petco to get a replacement rat after the
first rat he purchased bit his son, causing RBF.
38
social practices as evidence that a jury may have returned a different verdict under the

consumer expectations test than it did on the risk-benefit test. (Dis. opn., at p. 15.) This

evidence points to consumer expectations without consideration of the risks arising out of

complex rat biology and the potential interaction with human beings. The dissent's

recognition that the risk of serious injury or death from a pet rat may not be consistent

with the expectations of ordinary consumers even in light of a warning that tells

consumers there is such a risk (dis. opn., at p. 17) further bolsters our conclusion that use

of standardless consumer expectations test in this case could lead a jury to completely

ignore known risks, contradicting its conclusion that Petco offered an adequate warning.

If jurors applying the consumer expectations test were to consider all the factors, as the

dissent explains they must, jurors could not simply ignore the warning, which informed

consumers that they could contract RBF from handling the pet rat. Here Andrew does

not challenge any of the jury's verdicts—not the conclusion that Petco did not act

negligently in selling the rat, nor the verdicts that Petco did not fail to adequately warn

consumers of the risk of RBF. Given consumers' knowledge of that risk, it is less than

reasonably probable the outcome here would be different under the consumer

expectations test.

Finally, Andrew argues that the jury's unanimous finding that consumers would

not realize the dangers inherent in Petco's product means it is reasonably probable that

the jury would have returned a verdict in Andrew's favor under a consumer expectations

theory. Courts consider the closeness of a verdict as a factor indicating prejudice.

(Griesel v. Dart Industries, Inc. (1979) 23 Cal.3d 578, 584-585; Norman v. Life Care

39
Centers of America (2003) 107 Cal.App.4th 1233, 1252-1253.) But here the verdict

related to the adequacy of the warning was split at 10-2, not 9-3. The verdict that was

split at 9-3 concluded the benefits of the rat's design outweighed its risks. But that does

not offer any meaningful interpretation of what the ordinary consumer's expectation was,

only that the balancing of risks and benefits was a close call. Moreover, this conclusion

is undercut by the jury's other findings, which, as we have explained, necessarily impact

what the ordinary consumer in these situations expects.

We are sympathetic to Andrew and his family for the tragic loss of Aidan.

However, because we conclude an unmodified rat is not a designed product for purposes

of design defect strict products liability, we are compelled to affirm the judgment.

DISPOSITION

The judgment is affirmed. Parties to bear their own costs on appeal.

HUFFMAN, Acting P. J.

I CONCUR:

HALLER, J.

40
Dato, J., dissenting.

The beginning of this story is familiar to many parents and grandparents

throughout California; the end is not. A child wants a pet, and the parent or grandparent

sees it as an opportunity to provide companionship and teach the child responsibility.

Yet the story in this case took a tragic turn when the pet rat purchased by Aidan Pankey's

grandmother from national retailer Petco Animal Supplies, Inc. carried bacteria that was

transmitted to 10-year-old Aidan and caused his death. Aidan's family is left to wonder

how this could have happened.

This lawsuit by Aidan's father claims that Aidan's rat was a defective product

placed in the stream of commerce by Petco. To recover damages on a strict products

liability theory, a plaintiff must first show that injury or death was caused by a "product."

He or she must then demonstrate that the product was defective.

Addressing an issue of first impression under California law, the majority opinion

purports to answer the first question by conflating it with the second: it holds that a live

animal is not a "product" because the rat in this case wasn't really "diseased." But "[a]

'product' is broadly defined to include any 'tangible personal property distributed

commercially for use or consumption.' " (Johnson v. United States Steel Corp. (2015)

240 Cal.App.4th 22, 31 (Johnson), quoting Rest.3d Torts, Products Liability, § 19,

subd. (a).) A mass-marketed pet rat—Petco sells 400,000 every year—plainly satisfies

this expansive definition. Whether Aidan's rat was "diseased" is relevant to neither the

rat's status as a product, nor the question of whether it was defective. A product is
defective if it creates an unacceptable risk of harm to human beings. That the rat did not

look or feel sick has no bearing on this analysis.

The more difficult question in this case is the second one—whether the rat

carrying streptobacillus moniliformis was defective, and that would generally be an issue

for the jury to decide. The problem here is that the trial court refused to instruct the jury

on one of the two available tests to show a design defect—the consumer expectations

test—concluding it would be "too much" to expect ordinary consumers to have safety

assumptions about pet rats. To the contrary, however, ordinary retail pet store customers

can form reasonable safety expectations about their purchases, including that a pet rat

will not kill their child. I believe the trial court misapplied established California law in

reaching its conclusion, and that the majority opinion errs in approving that decision.

Accordingly, I respectfully dissent.

A. A Pet Rat as a Product

Petco argued in the trial court that a live animal cannot be a "product" for purposes

of strict products liability, and it reprises that argument here. The trial court rejected

Petco's contention, relying in part on the broad definition of "product" found in section 19

of the Restatement Third of Torts, Products Liability (hereafter Restatement), and cited

approvingly by California courts. (See, e.g., Johnson, supra, 240 Cal.App.4th at p. 31.)

The majority opinion agrees with Petco, disagrees with the trial court, and holds as a

matter of first impression under California law that an "unaltered" or "unmodified" pet rat

is not a product for purposes of asserting a design defect theory. (Maj. opn., at pp. 2, 41.)

2
 The area of law known as "products liability" refers to a set of legal principles and

doctrines that impose liability on economic entities that produce, distribute and sell

defective products which cause damage or injury. (See Merrill v. Navegar, Inc. (2001)

26 Cal.4th 465, 478.) Strict liability is imposed for defective products, and has been

applied to three different categories or types of defects—manufacturing defects, design

defects, and warning defects. (Anderson v. Owens-Corning Fiberglas Corp. (1991) 53

Cal.3d 987, 995.) But regardless which type of defect the plaintiff alleges, he or she must

first establish, as a threshold matter, "that the object or instrumentality claimed to be

defective was in fact a 'product.' " (Brooks v. Eugene Burger Management Corp. (1985)

215 Cal.App.3d 1611, 1626 (Brooks).) This inquiry depends, in large part, on the

policies underlying the strict products liability doctrine. (Fluor Corp. v. Jeppesen & Co.

(1985) 170 Cal.App.3d 468, 475; Pierce v. Pacific Gas & Elec. Co. (1985) 166

Cal.App.3d 68, 83 (Pierce).)

As Justice Traynor explained in his seminal opinion for the California Supreme

Court in Greenman v. Yuba Power Products, Inc. (1963) 59 Cal.2d 57, a principal

purpose of strict products liability is "to insure that the costs of injuries resulting from

defective products are borne by the manufacturers that put such products on the market

rather than by the injured persons who are powerless to protect themselves." (Id. at

p. 63.) The doctrine eases the plaintiff's burden where negligence is present but difficult

to prove, incentivizes improved product safety, induces allocation of resources towards

safer products, and spreads the risk of loss among all product users. (Pierce, supra,

3
166 Cal.App.3d at p. 83.) Each of these policies supports the determination that the pet

rat sold by Petco in this case was a product.

Most notably, recognizing a pet rat as a product appropriately allocates the cost of

a defective animal to the breeders, suppliers, and retailers who placed the animal on the

market and are in the unique position to detect and prevent the spread of streptobacillus

moniliformis before sale to the public. This policy is especially significant for pet

animals because consumers are in many instances unable to recognize—much less take

precautions against—harmful conditions in the animals they purchase. Significantly

here, witnesses from both sides agreed a consumer cannot determine whether a pet rat

carries streptobacillus moniliformis based on its physical appearance.1 Considering the

inconspicuous nature of these kinds of alleged animal defects, public policy strongly

supports imposing strict liability on animal breeders, suppliers, and retailers for injuries

proximately caused by the defective animals they place on the market. Moreover,

treating a pet rat as a product would incentivize Petco and other members of the

distribution chain to reduce the risk that rats with streptobacillus moniliformis will reach

the consuming public. (Union Pac. Ins. Co. v. S. Cal. Edison Co. (1985) 163

Cal.App.3d 700, 709 ["The manufacturer is held strictly liable because it is in a peculiarly

strategic position to promote the safety of its products."].)

1 Plaintiff's expert testified that rats carrying streptobacillus moniliformis have no
"externally visible signs of illness" and "don't look sick." Likewise, Petco's vice
president of veterinary medicine agreed that the bacteria represents a "hidden danger" and
testified that streptobacillus moniliformis cannot be detected by visual inspection.
4
 Further, strict products liability ensures that costs attributable to defective animals,

like other defective goods, can be equitably apportioned among members of the

distribution chain and internalized in the price of the product. (Jimenez v. Superior Court

(2002) 29 Cal.4th 473, 485 (Jimenez) [strict liability "promote[s] equitable spreading and

apportionment of the losses resulting from physical injuries as a cost of doing

business"].) As the tragic circumstances of this case make clear, a defective pet rat can

present a substantial threat to the health and safety of consumers and cause irreplaceable

loss. A recognition that pet animals sold by retail merchants are products would ensure,

at a minimum, that the economic value of these losses would be equitably apportioned

among retailers, suppliers, and other members of the vertical distribution chain, who

could in turn spread those costs to all of their animal-purchasing customers. A contrary

conclusion would force these sometimes-catastrophic losses upon the handful of

customers unlucky enough to have received a bacteria-carrying animal.

The majority opinion does not dispute that these policy considerations support a

conclusion that a commercially distributed pet animal is a product for purposes of strict

products liability. (Maj. opn., at pp. 23–25.) Nor does it address the reality that a pet

animal sold at a retail establishment is "tangible personal property distributed

commercially for use or consumption," and therefore surely satisfies the Restatement

5
definition of a product.2 (Rest.3d Torts, supra, § 19, subd. (a).) Instead, it focuses

entirely on the statement in comment b to section 19 that "when a living animal is sold

commercially in a diseased condition and causes harm to other property or to persons, the

animal constitutes a product for purposes of this Restatement." (Maj. opn., at p. 16,

quoting Rest.3d Torts, supra, § 19, com. b.) Without additional analysis, the majority

opinion assumes that a "diseased condition" is a necessary requirement for treating a live

animal as a "product," and ultimately concludes that Aidan's pet rat was not "diseased"

because streptobacillus moniliformis did not "make[] the rat sick." (Maj. opn., at p. 18.)

Accordingly, a rat carrying the streptobacillus moniliformis bacteria "is not diseased and

is therefore not a product." (Maj. opn., at p. 18.)

But the Restatement comment is intended to be illustrative, not definitional. This

is clear when the reference to "diseased condition" is read in context:

Courts are divided regarding whether living animals, such as pets or
livestock, should be considered to be products for the purpose of
determining a commercial seller's liability in tort. Frequently, as
when diseased livestock are sold and subsequently must be
destroyed, the claim to recover for their value involves a claim for
harm to the product itself and thus represents a claim for pure
economic loss not permitted by this Restatement. See [Restatement
Third of Torts, Products Liability,] § 21. But when a living animal
is sold commercially in a diseased condition and causes harm to
other property or to persons, the animal constitutes a product for
purposes of this Restatement. (Rest.3d Torts, supra, § 19, com. b.)

2 As one commentator has correctly summarized, "pets comfortably fit within any
definition of the term ['product']." (Parent, Every Dog Can Have Its Day: Extending
Liability Beyond the Seller by Defining Pets As "Products" Under Products Liability
Theory (2006) 12 Animal L. 241, 243.)

6
The first sentence indicates that the purpose of the comment is to address a conflict in the

case law as to whether live animals, such as pets and livestock, are products for purposes

of tort liability. (Rest.3d Torts, supra, § 19, com. b.) The Restatement answers that

question directly—yes, they are because they are "tangible personal property distributed

commercially for use or consumption."3 (Rest.3d Torts, supra, § 19, subd. (a).)

As an example, comment b mentions "diseased" livestock without defining the

term. If diseased livestock is sold—presumably for human consumption—and then has

to be destroyed because it can't be used for its intended purpose, the economic loss to the

buyer is not compensable in tort because it is harm to the product itself. (See Jimenez,

supra, 29 Cal.4th at p. 482 [discussing the economic loss doctrine].) But the comment

goes on to explain that if the defect in the product "causes harm to other property or to

persons," it is the proper subject of a product liability action. The critical question is

whether the product creates a risk of harm to persons or other property. (See Lanetti,

Toward a Revised Definition of Product under the Restatement (Third) of Torts:

Products Liability (2000) 35 Tort & Ins. L.J. 845, 873 ["The risk presented to the

consumer when purchasing the diseased animal is as great as, and arguably greater than,

when purchasing a traditional defective product." (Italics added.)].) It makes no

difference whether the animal (i.e., the product itself) is symptomatic.

3 "[P]ets are considered property of their owners." (Kimes v. Grossner (2011) 195
Cal.App.4th 1556, 1559; see Civ. Code, § 655.) Further, the parties agree that a
consumer "uses" a pet for companionship. (People v. Chung (2010) 195 Cal.App.4th
721, 728, fn. 4 [discussing the companionship function of pets].)
7
 As the Restatement commentary reflects, there is limited case law from other

jurisdictions that reaches a contrary conclusion—that live animals are not products for

purposes of strict products liability. These cases rely on "mutability" concerns,

suggesting that the purposes of strict products liability "would be defeated" if liability

were imposed for products with a "changeable nature" subject to "events and conditions

outside the control of the seller." (E.g., Anderson v. Farmers Hybrid Cos., Inc. (Ill. App.

1980) 408 N.E.2d 1194, 1199; Latham v. Wal-Mart Stores, Inc. (Mo. App. 1991) 818

S.W.2d 673, 676 ["We tend to agree with the Illinois view, that due to their mutability

and their tendency to be affected by the purchaser, animals should not be products"];

Malicki v. Koci (Ohio App. 1997) 700 N.E.2d 913, 915 [parrot is not a product because

its "health can be affected by many factors totally outside the defendant's control"].)

The reasoning of these cases is both unpersuasive and inconsistent with broader

principles of California law.4 Mutability may affect whether a defendant is ultimately

4 Better reasoned case law and scholarly commentary concludes that live animals
can be products. For example, in Beyer v. Aquarium Supply Co. (Div. of Hartz Mountain
Corp.) (N.Y. 1977) 404 N.Y.S.2d 778 (Beyer), a New York court followed the California
approach of analyzing the policies underlying strict products liability. It concluded "there
is no reason why a breeder, distributor or vendor who places a diseased animal in the
stream of commerce should be less accountable for his actions than one who markets a
defective manufactured product." (Beyer, supra, 404 N.Y.S.2d at p. 779; see also
Lannetti, supra, 35 Tort & Ins. L.J. at pp. 864, 873 [strict liability for sellers of diseased
animals "allows an equitable distribution of losses and promotes the marketing of safe
products"].) Agreeing with Beyer, the Oregon Court of Appeals and a Connecticut trial
court specifically rejected the argument that the fixed or unfixed nature of property
defines whether it is a product. (Worrell v. Sachs (Conn. 1989) 563 A.2d 1387, 1388;
Sease v. Taylor's Pets, Inc. (Or. App. 1985) 700 P.2d 1054, 1058.) As the Connecticut
court explained, mutability is "highly significant on the issue of liability" in a given case
because liability depends, in part, on whether a product reaches its consumer without
8
liable for injuries caused by a product. For instance, a substantial and unforeseeable

change to an item may make it difficult or impossible for a plaintiff to prove a defect or

causation. (Jimenez, supra, 29 Cal.4th at p. 480 ["What matters is whether the [products]

were defective when they left the factory, and whether these defects caused the alleged

injuries."]; see also Vermeulen v. Superior Court (1988) 204 Cal.App.3d 1192, 1202

[a change to the quality or character of something after it leaves a defendant's control

may constitute an intervening and superseding cause that precludes liability].) But

mutability is not relevant to the foundational question whether property is a product.

(See Pierce v. Pacific Gas & Elec. Co., supra, 166 Cal.App.3d at p. 83 [differentiating

between the predicate issue of product status and the issue of liability]; Harvey, The

Applicability of Strict Products Liability to Sales of Live Animals (1982) 67 Iowa L.Rev.

803, 817 [although a change to a product after its leaves defendant's control is relevant to

liability, it is "analytically flawed" to base product status on product mutability].) More

importantly, perhaps, none of the out-of-state authorities cited by Petco adopt the

reasoning of the majority opinion here—that only diseased animals displaying symptoms

of an illness are products.

Implicitly acknowledging the novelty of its doctrinal rationale, the majority justify

the approach based on the "difficulty with setting apart the defect from the definition of

substantial change. (Worrell, supra, 563 A.2d at p. 1387, italics added.) However, "it
does not necessarily follow logically that inability to prove a case because of mutability
means that an animal is not a product at all." (Ibid., italics added; Sease, supra, 700 P.2d
at p. 1058 [Oregon strict products liability doctrine "cover[s] products that are subject to
both natural change and intentional alteration"].) In other words, the mutability concern
"confuses proof of liability with [product] status." (Worrell, supra, 563 A.2d at p. 1387;
Sease, supra, 700 P.2d at p. 1058.)
9
the product" because "strict liability for design defect relies on the good being defective

to justify its application." (Maj. opn., at p. 25.) But the policy justification for strict

liability always relies on the product being "defective" in some way. This is true

regardless of whether the theory of defect is based on the product's design, its production,

or the failure to provide adequate warnings. The presence of a defect is always a second

and distinct step of the analysis. The majority's reasoning is circular because whether a

live animal is a product depends on whether it has a defect, so that only defective live

animals are products. And if an animal is "diseased" so that it is a product, when would it

not also be defective? Two distinct analytic questions are collapsed into one.

The majority opinion's intermittent focus on whether Petco's rats are products for

purposes of a design defect theory raises additional concerns about how the holding will

be applied, and ultimately renders the proffered rationale unworkable. Consistent with

both the Restatement discussion5 and precedent, the parties briefed the central question

whether a rat or a live animal is a product "for the purposes of strict products liability."

And the majority opinion often uses similar phraseology. (Maj. opn., at pp. 2, 12, 15,

21.) Indeed, this is consistent with the majority's recognition that it is necessary to

determine whether something is a "product" regardless of what type of defect the plaintiff

is alleging. (Maj. opn., at pp. 14–15, citing Brooks, supra, 215 Cal.App.3d at p. 1626.)

5 Section 19 appears in Restatement Chapter 4, "Provisions of General
Applicability," under Topic 3, "Definitions." Comment a to section 19 explains that the
question is "whether something distributed in commerce is a product for purposes of tort
liability." (Rest.3d Torts, supra, § 19, com. a, italics added.)
10
 Yet on other occasions, especially when stating its conclusion, the majority

opinion appears to limit its holding to what is a product "for purposes of design defect

theory" (maj. opn., at pp. 15, 23; see also maj. opn., at pp. 2, 16, 41, italics added),

suggesting that live animals in general (or pet rats in particular) might be products for

purposes of a different type of product defect theory. The majority go so far as to posit

that we need not be concerned about denying Aidan's father a design defect claim (on

grounds that a pet rat is not a product) because liability for Aidan's death is better

addressed as a strict liability claim for manufacturing defect or a failure to warn. (Maj.

opn., at p. 26.) The necessary implication is that Aidan's pet rat is a product for purposes

of defect theories other than design, but the majority offer no authority to support such an

approach or mechanism for deciding when something is a product for one purpose but not

another.

In my view, there is no principled basis to distinguish pet rats from millions of

other items of personal property placed into the stream of commerce by retailers like

Petco. They are products, subject to being deemed "defective" if the plaintiff can meet

any of the accepted tests for establishing a product defect. The remaining question in this

case is whether the trial court erred in precluding the jury from considering one of these

recognized tests—the consumer expectations test—in deciding whether the pet rat

suffered from a design defect.

11
B. The Applicability of the Consumer Expectations Test

Having decided that a pet rat is not a "product," the majority opinion explains that

"there is no need to consider whether the court properly instructed the jury regarding the

available tests." (Maj. opn., at pp. 26–27.) Were it to do so, however, it would agree

with the trial court that the circumstances of this case did not allow the jury to consider a

design defect theory based on the consumer expectation test. (Maj. opn., at p. 27.)

Unlike the question of first impression relied on to decide this appeal, this second

issue raised by the parties is one discussed in a number of California cases, including

relatively recent authority from this court. Under California law, there are two alternative

tests to prove the existence of a design defect—the consumer expectations test or the risk-

benefit test. (See Barker v. Lull Engineering Co. (1978) 20 Cal.3d 413, 432.) Generally

speaking, "a plaintiff may proceed under either or both." (McCabe v. American Honda

Motor Co. (2002) 100 Cal.App.4th 1111, 1126 (McCabe); see also Bresnahan v. Chrysler

Corp. (1995) 32 Cal.App.4th 1559, 1569.) As the majority opinion explains, the

consumer expectations test focuses on whether the product failed to perform as safely as

an ordinary consumer would expect. (Maj. opn., at p. 27.)

In this case the plaintiff requested instructions on both tests. Past cases have

recognized, however, that the consumer expectations test is not appropriate in every case.

Rather, it applies where "the circumstances of the product's failure permit an inference

that the product's design performed below the legitimate, commonly accepted minimum

safety assumptions of its ordinary consumers." (Soule v. General Motors (1994)

8 Cal.4th 548, 568–569.) Here, Petco told the trial court that no instructions on consumer

12
expectations were warranted because ordinary consumers don't understand the biology of

rats as carriers of streptobacillus moniliformis.6 Ultimately the court refused to instruct

on the consumer expectations test, concluding it would be "too much" to expect that an

ordinary layperson-consumer would have safety assumptions about a pet rat.

The majority assume for purposes of discussion that the biology of rats and the

pathology of rat bite fever (RBF) are complex topics. I don't disagree. However,

characterizing a product as "complex" is the beginning, not the end, of the analysis

because the consumer expectations test can apply to complex products, depending on the

circumstances of the alleged failure. (Saller v. Crown Cork & Seal Co., Inc. (2010) 187

Cal.App.4th 1220, 1232.) This court emphasized this very point only a few years ago in

Demara v. The Raymond Corp. (2017) 13 Cal.App.5th 545, 559 (Demara). The plaintiff

in Demara, injured when his foot was crushed by a forklift, brought a strict products

liability claim alleging that the forklift was defectively designed. The defendant argued

that the consumer expectations test did not apply because "the ordinary consumer has no

experience, and thus no expectation, as to the design and safety of the [forklift]—a

complex and technical product." (Id. at p. 558.) We concluded otherwise, observing that

"where a technically complex product performs 'so unsafely that the defect is apparent to

the common reason, experience, and understanding of its ordinary consumers,' a lay jury

is competent to determine whether the product's design is unsafe." (Id. at pp. 558–559.)

6 As Petco attempts to characterize it on appeal, the alleged failure of the product
was "esoteric and inherently complex," and could be properly understood only with
knowledge of "the biological mechanisms of pathogens, infectious diseases, and
technicalities of rat breeding for commercial and laboratory purposes."
13
We further explained: " 'If the facts permit an inference that the product at issue is one

about which consumers may form minimum safety assumptions in the context of a

particular accident, then it is enough for a plaintiff, proceeding under the consumer

expectation test, to show the circumstances of the accident and "the objective features of

the product which are relevant to an evaluation of its safety" [citation], leaving it to the

fact finder to "employ '[its] own sense of whether the product meets ordinary

expectations as to its safety under the circumstances presented by the evidence.' " ' "7

(Id. at p. 559; see Jones v. John Crane, Inc. (2005) 132 Cal.App.4th 990, 1003 ["[t]he

fact that expert testimony [is] required to establish legal causation for plaintiffs' injuries

does not mean that an ordinary user of the product would be unable to form assumptions

about the safety of the products"].)

Consistent with this court's Demara reasoning and the evidence presented in this

case, there was nothing overly complex about a pet rat or the circumstances of its alleged

defect as a carrier of infectious bacteria that would preclude a jury from applying the

consumer expectations test. Most people are familiar with, and have formed safety

7 The majority opinion correctly recognizes that "[t]he critical question is whether
the 'circumstances of the product's failure permit an inference that the product's design
performed below the legitimate, commonly accepted minimum safety assumptions of its
ordinary consumers.' " (Maj. opn., at p. 29, quoting McCabe, supra, 100 Cal.App.4th at
p. 1122, italics added; see Demara, supra, 13 Cal.App.5th at p. 561.) But this is not, as
the majority suggest, the same thing as asking "whether the bacteria causing RBF can be
designed out of rats" (maj. opn., at p. 30), a question that might be relevant, if at all, in
applying the risk-benefit test. The consumer expectations test would allow the jury to
find that even if it were not feasible to design a bacteria-free rat, selling pet rats that can
infect and sometimes kill pet owners and their family members is inconsistent with
consumers' reasonable safety expectations.
14
expectations about, pet animals sold for consumer use by a pet retailer. The jury could

reasonably find that Petco itself shaped expectations about the health and safety of its

pets—including the 400,000 pet rats it sells annually—in its corporate slogan, which

reads, "Petco. Healthier Pets. Happier People. Better World." Indeed, Aidan's

grandmother and others testified that, based on Petco's slogan and other factors, they

expected Petco would sell "healthy" and "disease free" pet rats. This evidence does not,

as Petco claims, merely reflect the general truism that consumers expect their products to

be safe. Instead, it supports a finding that the specific product characteristic challenged

here—the presence of hazardous and potentially deadly bacteria that can be transmitted to

pet owners and their families—was something that ordinary consumers considered and

formed expectations about.8

To be sure, the record includes other evidence that might tend to show the pet rat

in this case did not fall below ordinary consumer expectations because Petco informs

customers of bacterial transmission and RBF in several ways. And this evidence could

ultimately be helpful to Petco in trying to convince the jury that the pet rat it sold to

Aidan's grandmother satisfied the minimum safety expectations of ordinary consumers

8 Additionally, the circumstances of the product's alleged failure were not esoteric.
While transmission of streptobacillus moniliformis to humans may be rare, and a fatal
outcome even rarer, nothing in the record indicates Aidan's response to the bacterial
infection was idiosyncratic. On the contrary, it appears that many individuals with RBF
suffer the same symptoms that Aidan experienced. (Morton v. Owens-Corning Fiberglas
Corp. (1995) 33 Cal.App.4th 1529, 1535 [consumer expectations test applicable where
"[t]he injury [plaintiff] incurred was not the result of esoteric circumstances"]; cf. Trejo v.
Johnson & Johnson (2017) 13 CalApp.5th 110, 160 [consumer expectations test
inapplicable where plaintiff experienced "idiosyncratic reaction[]" to drug].)
15
because such consumers understand the bacterial transmission risks associated with pet

rats. But for purposes of this issue of instructional error—where the standard of review

requires us to consider the evidence in the light most favorable to the party requesting the

jury instruction—the fact that Petco provided information to customers about bacteria

and RBF actually supports plaintiff's position that an ordinary consumer can and does

form expectations about this kind of alleged defect.

Finally, the majority opinion suggests that even if the trial court should have

instructed the jury on the consumer expectations test, the failure to do so was harmless

error. (Maj. opn., at pp. 32, 34.) This conclusion is based primarily on answers the jury

provided to special verdict questions on other causes of action—specifically negligence

and strict liability failure-to-warn theories. (Maj. opn., at pp. 33, 35–36.) According to

the majority, these answers make it "less than reasonably probable the outcome here

would be different taking into consideration the existence and adequacy of the warning

that contributes to a consumer's expectations." (Maj. opn., at pp. 37–38.)

The majority opinion recognizes at several points that a cause of action for design

defect based on a failure to meet consumer expectations is independent of and

analytically distinct from a failure-to-warn claim. (Maj. opn., at pp. 35, 37–38.) It

likewise appears to acknowledge that a defense verdict on one claim does not preclude a

plaintiff's verdict on another theory. (Maj. opn., at p. 38, fn. 17.) But California law

makes clear that a "consumer's reasonable expectations are not confined to those arising

from information disseminated by the manufacturer or distributor of the product."

(Hufft v. Horowitz (1992) 4 Cal.App.4th 8, 18, fn. 8.) Rather, they are informed by a host

16
of other factors that include custom and social practice. This is particularly true with

regard to a topic as universal as human interaction with pets. (See generally Bradshaw,

The Animals Among Us: How Pets Make Us Human (2017).)

So for purposes of an alleged design defect, the question is not whether "warning

information helps form the basis of the ordinary consumer's expectations." (Maj. opn., at

p. 38, fn. 17.) It certainly can. The real question is whether, considering all the factors

that inform the expectations of ordinary consumers, the product performed as safely as

consumers would expect. Here we know the jury found that Petco was not liable in

negligence or strict liability for failure to warn. But this does not mean that the risk of

serious injury or death from a pet rat was consistent with the expectations of ordinary

consumers. The jury's special verdict answers make it quite clear the jury believed

otherwise, because it specifically found that Petco sold a dangerous pet rat and knew or

should have known that ordinary consumers would not realize the danger.

The reason the jury ultimately determined that Petco was not liable on either a

negligence or strict liability failure-to-warn theory had nothing to do with consumer

expectations being met. Indeed, Petco's counsel told the jury that the reason it should

return a defense verdict on the failure-to-warn claims was because the likelihood of

contracting RBF was infinitesimal. And it appears the jury followed this advice, finding

that although the rat was dangerous, the risk of death or serious injury was insufficiently

"substantial."

Given the dangers associated with overwarning (see generally, e.g., Dowhal v.

SmithKline Beecham Consumer Healthcare (2004) 32 Cal.4th 910, 932), the

17
substantiality of a risk is certainly a relevant consideration for a jury to take into account

in deciding whether a product seller should have included a different or additional

warning. But even a very small likelihood of a very bad result may be inconsistent with

the expectations of ordinary consumers and therefore unacceptable. Here, Petco provided

general warnings about the existence of RBF and the fact that small animals like rats can

carry bacteria or other germs which may cause flu-like symptoms. But it never advised

purchasers there was any risk of death or serious injury. That it chose not to do so is

hardly surprising—the thought of bacteria-laden rodents transmitting a deadly illness

conjures medieval images that could hardly have been good for pet rat sales. And

perhaps a jury could rationally conclude that the incidence of serious consequences from

RBF is sufficiently rare that no more specific warning was required. But it does not

mean that ordinary consumers would expect to die—or, worse yet, watch their child or

grandchild die—after playing with a furry little creature purchased for benefits of

companionship. In those instances, liability based on the product's design could be

properly imposed on the product seller so that cost of a rare and unexpected—but

nonetheless devastating—injury can be internalized in the price of the product and spread

among all users. In that way, the price of the product could reflect its true cost.

Consistent with the well-established consumer expectations test, the jury here should

have made that call.

I would reverse and remand for a new trial at which the court would instruct the

jury that it may return a plaintiff's verdict if it finds, along with other elements in CACI

18
No. 1203, that the pet rat which caused Aidan Pankey's death was not as safe as the

ordinary consumer would have expected.

DATO, J.

19