Houston Area Safety Council, Inc. and Psychemedics Corporation, Petitioners, v. Guillermo M. Mendez, Respondent (2023)

This case turns on the threshold inquiry for any common-law negligence claim: whether a duty exists. To make that determination, the Court today adheres to our precedents by applying the Phillips factors. But at least for a case like this, we should not need those factors. Drug testing is a familiar area of life and law that, in many respects, is already highly regulated. Judicial meddling with the careful balances that the other branches can strike and have struck is unnecessary and improper.

I nonetheless join the Court's thorough, well-reasoned, and well-written opinion that declines to recognize a duty in this context. No party has asked us to revisit those precedents. I address this case in a separate writing, however, for three reasons. First, I explain why we can reach today's result wholly aside from the Phillips factors. Second, I want to clarify that while the Phillips factors were a salutary and important improvement in the law when they were adopted, they do not have the ability to predictably constrain judicial adventurism. A judge who wants to impose a duty or create a new cause of action can readily do so and claim fidelity to Phillips—and, because the factors are amorphous, that claim cannot readily be rebutted. And third, I emphasize that I am not at all antagonistic to the common law. To the contrary, my approach would ensure that the common law plays its important but properly limited role in our large legal superstructure. In other words, we honor the common law by applying it where it exists, reading other laws (including the Constitution) in light of established common-law principles, and—importantly—declining to expand it where there is no vacuum for it to fill.

In short, I again suggest that our jurisprudence may benefit from a different approach when we respond to requests to enlarge the common law of torts.¹

I

One feature of today's case is, in my view, dispositive: that the alleged duty would arise in a highly familiar and significantly regulated context. Testing for drugs—within the broader context of drug abuse and detection more generally—is not some remote area of life that federal and state law have never contemplated. Over fifty years ago, the President declared that drug abuse was “public enemy number one.”² It is no secret that the improper use of drugs remains a widespread danger that has generated a host of public-policy responses. Nor is it a secret that drug use in some contexts—especially by those in positions capable of affecting public safety—is an even higher priority. Quite relevant here, both Congress and this State's legislature have passed laws that encourage drug-free workplaces. See Drug-Free Workplace Act of 1988, Pub. L. No. 100-690, 102 Stat. 4304 (codified as amended at 41 U.S.C. §§ 8101–8106) (instituting requirements for federal contractors and grant recipients); Tex. Lab. Code § 21.120 (stating that declining to employ individuals who illegally use or possess controlled substances does not constitute unlawful employment discrimination).

One specific response to drug use has been an emphasis on workplace drug testing, especially in fields with heightened safety concerns. To address drug testing in such employment contexts, a broad range of laws and regulations have been enacted by both the federal ³ and state ⁴ governments. These enactments usually provide protection for employees, whether through an appeal process, by establishing required testing procedures, or both. See, e.g., Mission Petroleum Carriers, Inc. v. Solomon, 106 S.W.3d 705, 713–15 (Tex. 2003) (discussing protections for transportation workers under specified federal drug-testing regulations); Tex. Lab. Code § 207.026(d) (requiring appeal process in unemployment-benefits context); Tex. Health & Safety Code § 555.022(b) (requiring appeal process for employees of state-supported living centers). In this case, Psychemedics points to regulatory requirements for laboratories addressing licensing, safety, and testing quality. The record indicates that Psychemedics is licensed or certified both by Texas (as well as other states) and the U.S. Department of Health and Human Services. The degree to which policymaking efforts have already been directed toward workplace drug testing is unsurprising. The briefing and the Court's opinion, see ante at 12–13, merely confirm what is widely known.

Nothing suggests that legislation cannot add (or subtract) more or different requirements as needed, including private causes of action against drug-testing laboratories. Likewise, as noted above, the reason that drug testing is so prevalent and pervasively regulated in some contexts is because public safety requires it. Consider, for example, an airline pilot who should have been flagged as under the influence, but who instead received a false-negative result. This one mistake could prove fatal to dozens or hundreds of lives.

Importantly, we do not live in an era in which courts alone are attuned to and responsible for such risks. Identifying and weighing them implicates the institutional competence and authority of the political branches of government. Those branches can consider the full landscape of costs and benefits, both to the public and to individuals; conduct detailed studies; gather and assess comprehensive technical data; account for the input of all affected groups; prioritize competing goals; and determine the proper extent of regulation. Unlike judges, they are not confined to the single record of a case and the parties before them. As one prominent scholar said plainly more than a century ago, “[i]t is a sound instinct in the community that objects to the settlement of questions of the highest social import in private litigations between John Doe and Richard Roe.” Roscoe Pound, Common Law and Legislation, 21 Harv. L. Rev. 383, 404 (1908). In short, judges should not be lawmakers when actual lawmakers are on the job.

Historically, of course, “judges were lawmakers by necessity as the common law took form.” Elephant Ins., 644 S.W.3d at 155 (Young, J., concurring). For so many centuries of the common law's development, the rest of the government made no effort to prescribe essential rules of conduct; the common law properly occupied the vacuum. The circumstances before us today provide no such vacuum.⁵ Insisting on judicially regulating in this context would amount instead to elbowing aside the institutions that exist for the purpose of, and have already undertaken the duty of, legislating and regulating. When the executive and legislative branches step up to the plate, the judicial creation of new common-law duties or actions is akin to treating a regulatory scheme like a Google Doc where we judges are invited to add a paragraph here or there.

Such judicial tinkering is hardly costless. It instead poses serious risks of destabilizing difficult balances that the other branches have struck—including when a deliberative scheme of regulation does not provide for liability in a given context.⁶ We cannot disregard “the presumption that the Legislature acts with full knowledge of ․ extant law.” Am. Nat'l Ins. Co. v. Arce, ___ S.W.3d ___, 2023 WL 3134718, at *4 n.14 (Tex. Apr. 28, 2023). Particularly when the legislature has focused deeply on an area of highly regulated law, it knows if the common law does not already impose a duty. When the legislature regulates many aspects of such a legal terrain (especially one as prominent as drugs in the workplace) but refuses to impose new legally enforceable obligations (such as a laboratory's duty as requested in this case), we should assume this result to be purposeful. We should credit the legislature with having considered taking such a step and, for any number of reasons, declining to do so. Maybe it prefers to await further data and experience. Maybe it has concluded that overregulation carries more risks than benefits.⁷ But the one thing that is improper to assume, at least in the context of detailed regulation or a highly familiar public-policy problem, is that the legislature just wants to see what this Court will do.⁸ Regardless, the presence of one or both indicates that perceived gaps or shortfalls should be remedied legislatively, not by the courts.⁹

* * *

In short, the existence of a regulatory scheme governing the type of conduct alleged to constitute negligence—at least when there is no impediment to the political branches' continued regulation—generally should foreclose the imposition of a common law negligence duty. Thus, even if the Phillips factors remain intact for other situations, we should take the opportunity—at least in a future case in which the parties so urge—to carve out circumstances like the one before us now and hold that such factors will play no further analytical role.

II

My desire to move away from the Phillips factors—at least in contexts like today's case, and perhaps more generally—does not connote disrespect for them. When they were announced in Greater Houston Transportation Co. v. Phillips, 801 S.W.2d 523 (Tex. 1990), the enumerated factors facilitated judicial restraint. It was a necessary and welcome speedbump following an extravagant era of largely uninhibited judicial creativity. Given Phillips's important place in our legal history, it is impossible to regard its now-eponymous factors without a healthy measure of gratitude. They made it easier for courts to pause and reflect on whether another branch of government would be better equipped to determine how best to order societal responsibilities and resources.

My dissenting friends and colleagues, however, invoke the very principles that led Phillips to attempt a recalibration of judicial excess. The common law, they say, “is not frozen or stagnant, but evolv[es]” when “the need [arises] in a changing society.” Post at 2, 16 (Boyd, J., dissenting) (quoting El Chico Corp. v. Poole, 732 S.W.2d 306, 310, 311 (Tex. 1987)). As those societal changes occur, the dissent adds, “it is the duty of this court,” id. at 2 (quoting Poole, 732 S.W.2d at 310), to evaluate whether the “changing social standards and increasing complexities of human relationships in today's society” “justify imposing a duty” we have not previously recognized, id. at 1–2 (quoting Otis Eng'g Corp. v. Clark, 668 S.W.2d 307, 310 (Tex. 1983)).

These statements were anachronistic when they were made in 1983 and 1987.¹⁰ Those cases' conception of the judicial role betrays an undue confidence in the judiciary's ability, much less authority, to oversee society. Judges are not at all equipped to measure the “complexities of human relationships in today's society,” and we lack the resources to survey (much less understand) our “changing society['s]” various “need[s].” Judges are elected for legal analysis, such as interpreting the meaning of documents and clearly articulating the meaning of existing law. These are very important and powerful roles that have little, if anything, to do with telling a society how to run its affairs. Fortunately, other branches of government do have that authority: those who are elected to devise and implement policy choices.

The irony is that the dissent would harness the Phillips factors in service of the repudiated vision of an invasive judiciary, when the point of Phillips was to do the opposite. But I cannot fault the dissent for this approach—Phillips itself makes it possible because its factors remain inherently malleable. Social questions, political questions, economic questions, social utility, burdens and benefits, and all the rest—who can say how such a “balance” turns out, or that it is balanced wrongly?

The Phillips factors, in other words, provided (and perhaps still provide) a useful yield sign. But after that pause, they do not in and of themselves prevent the determined judicial driver from zooming forth again. Fortunately, our Court has rarely put the foot back on the gas. The Phillips factors, however, are not themselves why we have discontinued broadly recognizing a host of new duties or causes of action. We do not need any factors to know that it is “highly unlikely that we could properly ‘discover’ a new duty lurking in the shadows after all these generations.” Elephant Ins., 644 S.W.3d at 159 (Young, J., concurring).

One unfortunate aspect of the concept of “balancing” tests—where facts and factors are weighed—is that it plays on the common judicial imagery of “the scales of justice.” As attractive as such an image is for some purposes, it is inapposite here. Balancing implies some calculable and reliable result. But the Phillips factors cannot provide for any such thing. The weighing of these factors manufactures an ability for judges to place a thumb on the scale—consciously or not—and then depict, to ourselves and others, the result as following from a neutral weighing. The idea is that the judge is no more responsible for the outcome of that weighing than the doctor is when we step onto his scales.

That is not usually how “balancing” tests actually work in the law. As one of my colleagues has previously noted, “[a] judge applying such a [balancing] test can become indistinguishable from a policymaker, for nothing in the test ‘stop[s] [him] from arriving at any conclusion he sets out to reach.’ ” Abbott v. Anti-Defamation League Austin, Sw., & Texoma Regions, 610 S.W.3d 911, 928 (Tex. 2020) (Blacklock, J., concurring) (quoting McDonald v. City of Chicago, 561 U.S. 742, 799 (2010) (Scalia, J., concurring)). And as another eminent jurist has commented in a different context, “what happens when the factors point in different directions[?] ․ No one knows. You get to guess.” Axon Enter., Inc. v. FTC, 598 U.S. 175, 207 (2023) (Gorsuch, J., concurring in judgment) (preliminary print). I must agree with all of this. A test that can be reverse-engineered seems less like an exercise of judicial discretion than a disguised exercise of the legislative function.

The problem is not that the tests do not generate unanimity—the justices of this Court routinely disagree about all sorts of questions. But they are usually legal questions. Multifactor balancing tests based on considerations more conducive to legislative than judicial determinations—like the balancing of values and risks at the forefront of the drug-testing context—are especially prone to unpredictable answers. Predictability, like objectivity, is essential to the rule of law. But both suffer because “[s]uch ‘balancing tests’ often invite courts to engage in open-ended cost-benefit analysis with no objective guideposts”—“a quintessentially legislative way of making decisions.” Abbott, 610 S.W.3d at 928 (Blacklock, J., concurring).

This all sounds very critical, but I am not critical of the judges who use the Phillips factors. For one thing, our current precedents require that effort; the Court today, and all its members (including me), are right to adhere to them. And in addition to my recognition that the Phillips factors have a positive origin, I also am certain that no justice of this Court uses them cynically to simply reach a desired result. Rather, I only note that such an especially indeterminate and subjective test has the potential to be used in a less than unbiased manner, even if done so unwittingly.

* * *

Instead of the current factor-ridden balancing test on which we must rely, I await a day in which the Court is presented with a new exercise—one that does not depend on subjective interpretations of policy considerations flexibly wielded to reach a particular result. In the meantime, while I acknowledge our broad authority to recognize new duties or causes of action, I am grateful that the Court continues to exercise extreme caution when asked to do so—especially in a highly regulated context like this.

III

Finally, and most briefly, I believe that the principles that I have articulated are among the best ways for us to honor the common law and our role as its custodians.

Adopting clear rules about when we will not accept invitations to expand the common law is a way to preserve the common law's integrity. Jurisdictional rigor involves an analogous point. “[J]udicial duty is not less fitly performed by declining ungranted jurisdiction than in exercising firmly that which the Constitution and the laws confer.” Ex parte McCardle, 74 U.S. (7 Wall.) 506, 515 (1869).¹¹ Our job is to decide cases—when we have jurisdiction. And our job is to expand and contract common-law duties and actions—when the conditions for doing so exist.

Recognizing the proper boundaries of the common law, therefore, would not, as our dissenting colleagues contend, forfeit our “role as guardian of the common law.” Post at 3. Recognizing these limits instead ensures that, like all judicial decisionmaking, common-law emanations arise by necessity and not through will or mere preference. There is no necessity to regulate an area that the other branches are already regulating.

Importantly, I do not contend that the Court lacks the power to take the steps that the dissent would take. In this sense the analogy to subject-matter jurisdiction is imperfect. The question instead is one of sound judicial administration, respect for the separation of powers, and a desire to ensure that the common law is playing only roles of undoubted propriety.

Many such roles remain, of course. The common law is not going anywhere and I am not advocating otherwise. Indeed, the common law has a privileged jurisprudential position. One of the “[s]tabilizing [c]anons” of construction is that “statutes will not be interpreted as changing the common law unless they effect the change with clarity.” Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 318 (2012). This Court accordingly “follow[s] an ‘opt-out’ approach that incorporates common-law principles absent the Legislature's clear repudiation.” Taylor v. Tolbert, 644 S.W.3d 637, 650 (Tex. 2022). Unsurprisingly, this Court still routinely decides a host of common-law cases in many legal areas—including torts. Of course, “as other sources of law proliferate, our common-law garden will require more pruning than fertilizing.” Elephant Ins., 644 S.W.3d at 158 (Young, J., concurring). But any pruning must satisfy the requirements of stare decisis. Until this Court says otherwise, or the legislature clearly supersedes it by statute, the rather substantial body of common law remains wholly in effect.

The common law, moreover, plays another vital role in the life of the law: informing and infusing our understandings of myriad legal texts, both old and new. To determine what later-enacted laws mean, as I recently discussed in a concurring opinion, “the common law is especially valuable. ‘[W]e construe statutory language against the backdrop of common law, assuming the Legislature is familiar with common-law traditions and principles.’ ” Arce, 2023 WL 3134718, at *15 (Young, J., concurring) (quoting Marino v. Lenoir, 526 S.W.3d 403, 409 (Tex. 2017)).

These important roles make it all the more important that we exercise caution in changing—and especially in expanding—the common law's reach. Because the Court exercised such caution in this case, I am pleased to concur.

FOOTNOTES

1.   See also, e.g., Elephant Ins. Co. v. Kenyon, 644 S.W.3d 137, 155–60 (Tex. 2022) (Young, J., concurring); Am. Nat'l Ins. Co. v. Arce, ___ S.W.3d ___, 2023 WL 3134718, at *17 (Tex. Apr. 28, 2023) (Young, J., concurring). I expand upon those observations here because this case provides a prime example of when the Phillips factors can be less than helpful and may in fact cause more harm than good.

2.   Richard Nixon, President of the U.S., Remarks About an Intensified Program for Drug Abuse Prevention and Control (June 17, 1971), https://www. presidency.ucsb.edu/documents/remarks-about-intensified-program-for-drug-abuse-prevention-and-control.

3.   See, e.g., 49 U.S.C. § 5331 (public transportation); id. § 20140 (railroads); id. § 31306 (commercial motor vehicles); id. § 45102 (aviation); 51 U.S.C. § 31102 (NASA employees and contractors); 14 C.F.R. pt. 120 (aviation); 49 C.F.R. pt. 40 (DOT drug-testing procedures); id. pt. 199 (pipeline facilities); id. pt. 219 (railroads); id. pt. 382 (commercial motor vehicles).

4.   See, e.g., Tex. Health & Safety Code § 242.052 (convalescent and nursing facilities); id. § 555.022 (state-supported living centers for those with intellectual disabilities); Tex. Occ. Code § 2303.161 (vehicle-storage facilities); id. § 2308.158 (towing operators); Tex. Hum. Res. Code § 42.057 (DFPS-licensed or certified residential childcare facilities); Tex. Lab. Code § 207.026 (unemployment benefits for certain categories of workers); 26 Tex. Admin. Code § 554.423 (providing model drug-testing policy, including appeal process, for nursing facilities that choose to drug-test employees); id. § 745.4151 (providing model drug-testing policy for residential childcare facilities, including an appeal process and minimum standards for testing methods); 16 Tex. Admin. Code § 85.725 (establishing minimum standards for drug testing for employees of vehicle-storage facilities and detailing reporting and reviewing procedures); id. § 86.710 (same for towing operators).

5.   “Unlike the judges who developed the law of negligence, today's judges routinely see comprehensive statutory and regulatory schemes,” and “it is increasingly less likely than ever before that there are gaps that judges alone can (much less should) fill.” Elephant Ins., 644 S.W.3d at 157 (Young, J., concurring).

6.   By proposing a bright-line rule of restraint when the area of law at issue is highly regulated, I do not suggest that a lack of regulation indicates that courts should feel liberated or obligated to recognize new tort duties or actions. While legislative exertion more obviously demonstrates an attempt to finely tune a policy balance, inaction can also signal a policy choice, particularly in contexts like this one that are not “genuinely new.” Id. at 159 n.7. My point is that even if we keep the Phillips factors generally, we at least need not use them in contexts like this one.

7.   Avoiding overregulation in this context, for example, would be consistent with our State's approach to private-employer drug testing more generally. See Aaron S. Demerson, Tex. Workforce Sols. & Tex. Workforce Comm'n, Texas Guidebook for Employers 216 (2022), https://efte.twc.texas.gov/ texas-guidebook-for-employers-2022.pdf (noting that, “[u]nder Texas and federal laws, there is almost no limitation at all on the right of private employers to adopt drug and alcohol testing policies for their workers”).

8.   This case again provides a ready illustration. As the Court observes, “the Fifth Circuit made an Erie guess ․ that, under Texas law, an independent laboratory does not owe a legal duty ‘to persons tested to perform its services with reasonable care.’ ” Ante at 10 (quoting Willis v. Roche Biomed. Lab'ys, Inc., 61 F.3d 313, 316 (5th Cir. 1995)). Until the judgment below, we are aware of no Texas courts that deviated from that understanding. For decades, therefore, the legislature has known that such claims would not be heard in any forum. It is implausible that the legislature, aware of this legal regime and fully attuned to the complex issues surrounding drug testing, would not have stepped in if it believed that the absence of new tort liability was undermining its regulatory program or if such new liability would be advantageous to it.

9.   The courts obviously retain important roles in highly regulated legal areas. For one thing, courts remain the chief fora in which those laws and regulations may be interpreted and enforced. And not just enforced, but also tested—in proper adversary contexts, it remains a core judicial function to determine whether the legislature (and, by extension, agencies) have complied with all relevant principles of constitutional (and administrative) law.My point, therefore, is not that courts must supinely yield to anything an agency (whether expert or otherwise) might say. Quite to the contrary, courts should and must ensure that the political branches and the agencies they empower do not overstep. But when it comes to the lawmaking function, the very fact that the legislature has focused on a problem (including when it delegates to agencies, whether properly or not) may mean that we have much less need to act as common-law judges in that area, and perhaps no need at all.

10.   Notably, in one of the areas of common law left to the U.S. Supreme Court—implying causes of action—that Court went through a similar process of exuberance and retrenchment, including with respect to the indisputably important goal of protecting constitutional rights.In his concurring opinion in Bivens, for example, Justice Harlan forthrightly put it this way: “[T]he range of policy considerations we may take into account is at least as broad as the range ․ a legislature would consider with respect to an express statutory authorization of a traditional remedy.” Bivens v. Six Unknown Named Agents of Fed. Bureau of Narcotics, 403 U.S. 388, 407 (1971) (Harlan, J., concurring in judgment) (emphasis added).Things quickly changed. The U.S. Supreme Court stopped implying other Bivens actions around the time that this Court made the statements about the evolution of the common law in the two cases cited above. And fast-forward to last year, in which the U.S. Supreme Court put it this way: “A court faces only one question: whether there is any rational reason (even one) to think that Congress is better suited to weigh the costs and benefits of allowing a damages action to proceed.” Egbert v. Boule, 142 S. Ct. 1793, 1805 (2022) (internal quotations omitted). “[E]ven a single sound reason to defer to Congress is enough to require a court to refrain from creating such a remedy.” Id. at 1803 (emphasis added) (internal quotations omitted).We need not go so far with respect to all common law; after all, state courts do possess a broader repository of common-law authority than federal courts. But I do not see why it should be terribly controversial, at least with respect to highly regulated areas of the law, to agree with Egbert's logic. For those areas, unless there is no good reason to think that the legislature “is better suited to weigh” all the various factors involved in creating a new duty to govern civil and commercial relationships (and that would be, to put it mildly, rare), we should stay our hand.

11.   Many citations of McCardle use 1868 for its year of decision. But while it was heard in December Term 1868, the Court decided it in 1869. See William H. Rehnquist, Judicial Independence, 38 U. Rich. L. Rev. 579, 590 n.41 (2004) (explaining both the confusion regarding McCardle's citation and the correct date to use). Since Chief Justice Rehnquist's article, the U.S. Supreme Court has not used the erroneous 1868 date.

Evan A. Young Justice