All the World’s a Platform? “Marketplace Platform” Employment Laws

Establishment of the Independent Contractor Classification

Legal academics note that there are essentially two dominant currents of overarching theoretical perspective on the gig economy: it is either an utter canard explained by the desire to evade existing law or an evolution of the very nature of work.⁴ It has been suggested that one way to break through this conceptual bifurcation may be to “effectively distinguish app employees from bona fide independent contractors . . . [by] ask[ing] the same question that initially sparked the independent contractor designation: Do today’s ‘independent contractors’ possess sufficient skill and economic autonomy to stand alone in the marketplace?”⁵

The problem with this perspective is that it is not at all clear that the original establishment of the independent contractor classification had anything much to do with imposing liability on those persons able to stand alone in the marketplace. Rather, creation of the independent contractor category may simply have been a strategy allowing masters to escape tort liability created by putative servants. In other words, creating a category of “not servants” may have had more to do with a “push” policy of tort liability avoidance than a “pull” policy of consciously allocative tort compensation (leaving entirely to one side questions of employment law—including workers’ compensation—coverage).

The distinction between a servant or employee and an independent contractor has been said to lie in the fact that the latter does the work on his own time, in his own way, and under no one’s directions but his own, so that the one who selected the contractor has no control or right of control over the manner in which it is done. It is therefore to be regarded as the contractor’s own enterprise, and he, rather than the one who retained his services, is the proper party to be charged with the risk. Also, no doubt, there is in the picture, although not often expressed, the fact that, taken as a class, independent contractors are in the main financially responsible parties, at least to a far greater extent than employees.⁶

Thus, the foundation for the original independent contractor category was probably that the “master” ought not to be held accountable for circumstances over which the master had no control, for reasons of fairness; and it was, occasionally, a happy fortuity that some entities deemed to be independent contractors turned out to be “financially responsible parties” (at least in comparison to penniless “employees”), though one may very much doubt that this was the primary reason behind the employee–independent contractor bifurcation.⁷

Furthermore, the whole idea of vicarious liability (respondeat superior) was “bitterly” contested by common-law judges throughout the 19th century.⁸ The earliest English precedents fail to cite any authority for the proposition that vicarious liability existed; and no less a luminary than Oliver Wendell Holmes thought the common-law version of respondeat superior was a fiction derived from Roman law: If my servant (a chattel) harmed you, you had the right to kill or maim my servant.⁹ To avoid this result, I could pay you compensation. Over time, some links in the chain were lost, and it became the rule that the one who was harmed could simply approach the master for compensation for the wrongs done by the servant, though the master was in no sense at fault.¹⁰ There seems no getting around the fact that respondeat superior was (and is) a form of strict liability,¹¹ and that without the respondeat superior doctrine there would be no disputes over “independent contractors” in tort cases because masters would have no need to escape liability for the torts of their servants.

So much for a clean tort theory on the origins of the independent contractor doctrine being grounded somehow in the liability-worthiness of contractors. And if respondeat superior was a mere public policy expedient—the brute ability to recover from deep pockets—small wonder that disputes over whether a worker is truly a “servant” have been ongoing since the mid-19th century: “If you are going to arbitrarily fleece me under respondeat superior, regardless of your good social intentions, I am going to resist.” Nineteenth-century judges seemed extremely reluctant to hold the users of “workers” liable in all cases, and the instinct to prevent the perceived injustice of such strict liability seemed strong.¹²

Still, respondeat superior “won out,” and there is probably more than one economic explanation for why it did. Perhaps an explanation hitting closer to the mark than the placement of liability on parties who could pay simply because they could pay (i.e., employees, no liability because unable to pay; independent contractors, liability because able to pay) is that of the “cheapest cost avoider.” Liability might simply be placed on the party having control of the risks inherent in an activity because that party could take precautions against those risks. This idea provides a plausible explanation for why workers’ compensation (and other employment) law has remained firmly connected to common-law notions of control—the entity that controls workplace conditions may take precautions against various forms of “antisocial” conduct. It is also a public policy seeming to marry ideas of justice, general deterrence, and instrumental realism.¹³

“Classical” Workers’ Compensation

Whatever the economic justifications for creation of the independent contractor doctrine in the 19th century, and despite 19th-century judicial resistance to concepts of respondeat superior in tort contexts, the first English workers’ compensation statute (of 1897)—which was the prototype for early American workers’ compensation statutes—made no bones about significantly interfering with attempts by employers to unilaterally designate workers as independent contractors:

Where, in an employment to which this Act applies, the undertakers, as hereinafter defined, contract with any person for the execution by or under such contractor of any work, and the undertakers would, if such work were executed by workmen immediately employed by them, be liable to pay compensation under this Act to those workmen in respect of any accident arising out of and in the course of their employment, the undertakers shall be liable to pay to any workman employed in the execution of the work any compensation which is payable to the workman (whether under this Act or in respect of personal negligence or wilful act independently of this Act) by such contractor, or would be so payable if such contractor were an employer to whom this Act applies.

Provided that the undertakers shall be entitled to be indemnified by any other person who would have been liable independently of this section.

This section shall not apply to any contract with any person for the execution by or under such contractor of any work which is merely ancillary or incidental to, and is no part of, or process in, the trade or business carried on by such undertakers respectively.¹⁴

Readers may recognize this language as a foundational form of “statutory employment.”¹⁵ If an “undertaker” employed contractors to do work that was not merely “ancillary or incidental” to and was a part of, or process in, the trade or business carried on by the undertaker, the contractors were covered “workmen” under the act. It is on this “part of, or process in” question that market contractor laws seem to depart significantly from workers’ compensation theory.

Platform Laws Meet the 1897 Test

In 2018, a number of states passed “market contractor” laws under the apparent lobbying impetus of the company Handy Inc.¹⁶ As will be shown, these laws unquestionably make it significantly easier for companies to classify their workers as independent contractors rather than employees, thereby lowering their operating costs but also creating a class of individuals with fewer legal protections when suffering on-the-job injuries.¹⁷

Although styled as laws applicable only to retired handymen and similar part-time workers, the text of the laws is much broader than this innocuous characterization suggests.¹⁸ Simply put, they do not, by their terms, apply only to handymen (or similar categories of workers), nor do they apply only to part-time workers. But even the innocuous characterization of the laws might give one pause. In what sense are laws applicable to handymen touching on a revolutionary new economy (the apparent premise for why gig laws are justifiable in the first place)? These are very broad laws with no obvious, narrower connection to the “mysterious” world of high-tech coders and the like, whose work, we are told, defies analysis under traditional restatement factors. What makes the employment seem “gig-like” to legislators appears to be that workers are dispatched to their work with an app, or even just through a computer.

Almost identical laws were passed in Arizona,¹⁹ Florida,²⁰ Indiana,²¹ Iowa,²² Kentucky,²³ Tennessee,²⁴ and Utah²⁵—not all of these are locations that might evoke images of a booming, high-tech economy.²⁶ Handy laws were defeated in Alabama, California, Colorado, and Georgia.²⁷ This article aims to show briefly how expansive the marketplace platform laws are by focusing closely on statutory language. The laws impact workers’ compensation significantly because most of them carry the talismanic phrase that the law applies “for all purposes under state and local laws, rules, ordinances, and resolutions.”²⁸

The advocacy group National Domestic Workers Alliance argued, in connection with enactment of a Handy law in Tennessee:

The [Chattanooga] Times Free Press reported on the legislation in the Senate, reporting on sponsor Senator Watson likening “the marketplace platforms to business advertising in newspapers.” We invite Senator Watson to talk to a Handy worker, who will explain that marketplace platforms set the rates of workers, charge workers fines for changing their work schedule, facilitate payments and change rates of workers according to reviews. Not much like a newspaper when you think about it, is it?²⁹

Thus, the argument centers on a debate as to whether a marketplace platform is no more than a passive information clearinghouse merely offering “disinterested” space for contractors and third parties to enter into contractual relationships without the platform having any input into what happens thereafter.³⁰ In other words, are marketplace platforms “undertakers” (to use the language from classical workers’ compensation) contracting “with any person for the execution by or under such contractor of any work,” or is the work of contractors “merely ancillary or incidental to, and is no part of, or process in, the trade or business carried on by such undertakers”?

Analysis of a Handy/Contractor Law—Tennessee

Because the platform laws under discussion are nearly identical, analysis of just one of them, Tennessee’s, will suit the purposes of this article.

Under Tennessee’s marketplace contractor law, a “marketplace contractor” working for a “third party” at the direction of a “marketplace platform” is an independent contractor as a matter of law if the following conditions are specified in a written agreement:

(1) The marketplace platform and marketplace contractor agree in writing that the contractor is an independent contractor with respect to the marketplace platform;

(2) The marketplace platform does not unilaterally prescribe specific hours [of work]. If a marketplace contractor posts the contractor’s voluntary availability to provide services, the posting does not constitute a prescription of hours . . . ;

(3) The marketplace platform does not prohibit the marketplace contractor from using any online-enabled application, software, website, or system offered by other marketplace platforms;

(4) The marketplace contractor may, at its discretion, enlist the help of an assistant to complete the services, and the marketplace platform may require the assistant to complete the marketplace platform’s standard registration and vetting process. If the marketplace contractor enlists the help of an assistant, the marketplace contractor, not the marketplace platform, is responsible for paying the assistant;

(5) The marketplace platform does not restrict the marketplace contractor from engaging in any other occupation or business;

(6) The marketplace platform does not require marketplace contractors to use specific supplies or equipment;

(7) The marketplace platform does not control the means and methods for the services performed by a marketplace contractor by requiring the marketplace contractor to follow specified instructions governing how to perform the services. However, the marketplace platform may require that the quality of the services provided by the marketplace contractor meets specific standards and requirements;

(8) The agreement or contract between the marketplace contractor and the marketplace platform may be terminated by either the marketplace contractor or the marketplace platform with or without cause;

(9) The marketplace platform provides no medical or other insurance benefits to the marketplace contractor, and the marketplace contractor is responsible for paying taxes on all income derived as a result of services performed to third parties from the assignments or connections received from the marketplace platform; and

(10) All, or substantially all, payment to the marketplace contractor is based on performance of services to third parties who have engaged the services of the marketplace contractor through the marketplace platform.³¹

Before setting out more of the law, it may be worth pausing to consider that factor (7) (the statute, like the Restatement (Second) of Agency, utilizes 10 factors)—control—would be important in any employee versus independent contractor analysis. Indeed, under the ABC test that is at the heart of the AB 5/Dynamex rule in California,³² the burden would be on the putative employer to establish its absence of control. Under the Tennessee law, however, the control factor is merely one of 10, and the other nine seem designed to make it very easy for a marketplace platform to establish that a marketplace contractor is an independent contractor. For example, rather than determining whether a putative independent contractor in fact engages in another business or calling—as would be the question in a restatement analysis³³—it is enough for the platform “not to have restricted” the worker from doing so. Similarly, the question under the Handy laws is whether the platform “requires” the contractor to use specific supplies or equipment or “prevents” the contractor from seeking other employment, not whether the contractor has actually done these things.³⁴ Thus, in addition to providing a simplified factor test weighted in favor of finding independent contractor status, the law also seems designed to disregard consideration of economic realities surrounding the relationship between the contractors and the platform.

Continuing with exposition of the law, a “marketplace contractor” is any individual, corporation, partnership, sole proprietorship, or other business entity that:

(A) Enters into an agreement with a marketplace platform to use the platform’s online-enabled application, software, website, or system to [be given an assignment, or otherwise] receive connections to third-party individuals or entities seeking its services in this state; and

(B) In return for compensation from the third-party or marketplace platform, offers or provides services to third-party individuals or entities upon being given an assignment or connection through the marketplace platform’s online-enabled application, software, website, or system.³⁵

A “marketplace platform,” in turn, is a corporation, partnership, sole proprietorship, or other business entity operating in the state that “[o]ffers an online-enabled application, software, website, or system that enables the provision of services by marketplace contractors to third-party individuals or entities seeking services.”³⁶ Under this definition, it is difficult to conceive of any company that could not characterize itself as a marketplace platform. Companies use “online-enabled” websites, after all. Customers may be recast as “third parties.” When the customer contacts the company, the customer is “seeking services” from service-provider employees, now recast as “marketplace contractors” who, after all, “agreed” in advance and in writing that they were contractors (something that students of the history of workers’ compensation may recognize as a thinly disguised preinjury waiver of liability).³⁷

Imagine a situation in which a contractor is subject to discipline if the contractor does not comply with a work schedule “voluntarily” posted by the platform; imagine a contractor who does not in fact use other “platforms” but has used the same “platform” exclusively for a period of years; imagine a contractor who does not in fact engage in any occupation or business other than work through the platform (and never has); imagine a contractor who in fact uses platform-provided supplies or equipment (and always has); and imagine a platform that in fact supplies supervision to the contractor. None of this may matter for purposes of Handy-law independent contractor analysis if the worker has “agreed in writing” that the worker is not an employee, and if a company offers an online-enabled application.

Thus, if one calls a plumbing company this year (just as one called the plumbing company last year), the company might announce that it is not a plumbing company, after all, and that you are not a “customer” but rather a “third party.” The plumbing company has reimagined itself as a technology company because it possesses a website and dispatches to jobs its now-contractor plumbers through use of a Google app and a couple of handfuls of TracFones. If the benefit gained through such a metamorphosis is a complete release of the company from the obligation to comply with its state’s labor and employment laws, none of this is difficult to imagine.

Conclusion

It is a simple axiom of workers’ compensation policy that an industry must pay for the costs of its injuries. One can pretend that Handy is not in the industry of handymen or Uber is not in the industry of transportation. In a sense, this is true in the way it is also true that no business is permanently tethered to its business. Apple (an actual technology company) happens to be in the business of providing technology because it is profitable. If it were not profitable, Apple would exit the business and invest its capital elsewhere. But while it is in the technology business, it is in the technology business. Handy may only temporarily happen to be in the industry of providing handymen, but while it is providing handymen, it is engaged in the industry of handymen. It seems very bad policy to exempt companies within an industry from sharing the burden of the costs of injuries to its “workers” without fully exploring the relationship between the platform and those workers.

It has always been fundamental to workers’ compensation that workers engaged in the process of a venturer’s business may be treated as employees if the business would otherwise have used “employees” to carry on its core activities (the “B” factor in the ABC test). In the language of the “old” workers’ compensation law, workers should be covered if they are part of the trade or business carried on by the “undertakers” of productive activity.

A society might come to a different conclusion about how to allocate the costs of injury (or how to disseminate other social policies embedded in employment statutes) than to cover “employees” by industry sectors. But changing the rules for coverage—or effectively discontinuing coverage altogether—should be carefully discussed and deliberated upon by the broader society. Regrettably, platform workers, including handymen, will be injured in the course of work. Who will pay for the costs of injury, and how much will be paid? These questions have historically been decided transparently by the body politic, and experience counsels the robust use of the democratic process before embarking on a public policy adventure.

Marketplace platform laws also aggravate the growing nonuniformity in “employment status” determinations throughout the United States. To the ABC and restatement tests is now added the seven-state marketplace platform test. One may recall that a central justification for enactment of the Employee Retirement Income Security Act of 1974 (ERISA)—the federal employee benefit statute—was to relieve employers from the burdens of having to comply with different benefit structures throughout the country. “Requiring ERISA administrators to master the relevant laws of 50 States and to contend with litigation would undermine the congressional goal of ‘minimiz[ing] the administrative and financial burden[s]’ on plan administrators—burdens ultimately borne by the beneficiaries.”³⁸ Even within states, employee definitions can differ under state employment statutes.³⁹ Now it appears that we may effectively have created national “zones” in which employee determinations may significantly (or even completely) differ: ABC, restatement/common law, and marketplace platforms.⁴⁰ In the din arising from this nonuniformity, original employment law policies seem all but lost.

Notes

1. “Gig economy” refers to “economic activity that involves the use of temporary or freelance workers to perform jobs typically in the service sector.” Gig Economy, Merriam-Webster, https://www.merriam-webster.com/dictionary/gig%20economy (last visited Jan. 19, 2021).

2. See, e.g., Tenn. Code Ann. §§ 50-8-101 to -103 (2019). Most legal analyses emphasize whether the employer or the worker has “control” in the workplace, though other factors are considered as well. See Restatement (Second) of Agency § 220(2) (Am. Law Inst. 1958); Restatement (Third) of Agency § 7.07(f) (Am. Law Inst. 2006). The “ABC” test, in use in several states, also analyzes control, along with other factors, but places the burden on the putative employer, rather than the putative employee, to show that it lacks control of the details of the employee’s work. The recently codified California version of the ABC test (not applicable to workers’ compensation) states:

[A] person providing labor or services for remuneration shall be considered an employee rather than an independent contractor unless the hiring entity demonstrates that . . .

(A) The person is free from the control and direction of the hiring entity in connection with the performance of the work . . . .

(B) The person performs work that is outside the usual course of the hiring entity’s business.

(C) The person is customarily engaged in an independently established trade, occupation, or business . . . .

Cal. Lab. Code § 2750.3 (emphasis added).

3. See infra notes 19–25 and accompanying text.

4. Keith Cunningham-Parmeter, Gig-Dependence: Finding the Real Independent Contractors of Platform Work, 39 N. Ill. U. L. Rev. 379, 381–82 (2019).

5. Id. at 385.

6. Victor E. Schwartz et al., Prosser, Wade and Schwartz’s Torts: Cases and Materials 697 n.1 (12th ed. 2010) (emphasis added).

7. Charles W. Pierson, A Recent Attempt to Limit the Independent Contractor Doctrine, 8 Yale L.J. 63, 64 (1898) (noting that the theoretical justification for respondeat superior was weak, and “[a] sense of its harshness and a desire to restrict its operation are no doubt partly responsible for the evolution of the Independent Contractor Doctrine”).

8. David G. Hanes, The First British Workmen’s Compensation Act, 1897 (1968).

9. Id. at 9–10. As in the case of a “deodand,” “a thing that by English law before 1846 was forfeited to the crown and thence to pious uses because it had been the immediate cause of the death of a person.” Deodand, Merriam-Webster, https://www.merriam-webster.com/dictionary/deodand (last visited Jan. 19, 2021).

10. Id.

11. John C. P. Goldberg et al., Tort Law: Responsibilities and Redress 561 (2016) (explaining that some commentators view vicarious liability as “strict”).

12. See, e.g., Blake v. Ferris, 5 N.Y. 48 (1851) (upholding a challenge to the trial court’s refusal to instruct the jury that the “employer” could not be held liable for the acts of its independent contractor).

13. But see Guido Calabresi, Does the Fault System Optimally Control Primary Accident Costs?, 33 L. & Contemp. Probs. 429, 441 (1968) (explaining asymmetrical operation of the cheapest cost avoider principle in workers’ compensation contexts).

14. English Workers’ Compensation Act of 1897, 60 & 61 Vict. c. 37, § 4.

15. For a modern discussion of these issues, see 6 Lex K. Larson & Thomas A. Robinson, Larson’s Workers’ Compensation Law ch. 70 (2020).

16. Sarah Kessler, Handy Is Quietly Lobbying State Lawmakers to Declare Its Workers Aren’t Employees, Quartz at Work (Mar. 30, 2018), https://qz.com/work/1240997/handy-is-trying-to-change-labor-law-in-eight-states. Texas enacted by administrative rule a similar market contractor test in April 2019 that applies solely to employment status for unemployment benefit qualification. See Erin Douglas, Texas Adopts Rule to Classify Digital Gig Economy Workers as Independent Contractors, Hous. Chron. (Apr. 9, 2019), https://www.houstonchronicle.com/business/article/Texas-adopts-rule-to-classify-digital-gig-economy-13754532.php.

17. Rebecca Smith, “Marketplace Platforms” and “Employers” under State Law—Why We Should Reject Corporate Solutions and Support Worker-Led Innovation, Nat’l Emp. L. Project (May 18, 2018), https://www.nelp.org/publication/marketplace-platforms-employers-state-law-reject-corporate-solutions-support-worker-led-innovation. Most, but not all, of these laws explicitly apply to workers’ compensation claims. Applicability of the laws to workers’ compensation is noted parenthetically throughout the notes.

18. Id.

19. Ariz. Rev. Stat. Ann. § 23-1603 (applies to workers’ compensation laws).

20. Fla. Stat. ch. 451. The law was passed allegedly after a mere 10 minutes of discussion. Lawrence Mower, Florida Lawmakers Approve Last-Minute Change on Behalf of Powerful Lobbyist, Tampa Bay Times (Mar. 12, 2018), https://www.tampabay.com/florida-politics/buzz/2018/03/12/florida-lawmakers-approve-last-minute-change-on-behalf-of-powerful-lobbyist. Florida’s law appears to only apply to “platforms” that accept “service requests from the public only through its online-enabled technology application service, website, or system.” Fla. Stat. § 451.01(3)(b) (emphasis added). The qualifier “only” may limit the universe of putative employers covered by the law. “A marketplace contractor must be treated as an independent contractor, and not as an employee, of the marketplace platform for all purposes under state and local laws, regulations, and ordinances, including, but not limited to, [the workers’ compensation and ‘reemployment’ statutes].” Id. § 451.02(1).

21. Ind. Code § 22-1-6-3 (“treated as an independent contractor for all purposes under state law”).

22. Iowa Code ch. 93 (appears not to apply to workers’ compensation).

23. Ky. Rev. Stat. Ann. § 336.137 (applies to workers’ compensation).

24. Tenn. Code Ann. § 50-8-102 (“for all purposes under state and local laws, rules, ordinances, and resolutions”).

25. Utah Code Ann. § 34-53-102 (limited to building services/janitorial contexts).

26. The Handy lobbyist is not exactly coy about strategy:

“If starting with the harder states failed, we’re taking a shot at something[] a little faster,” said venture capitalist and political strategist Bradley Tusk, whose firm Tusk Ventures ran Uber’s state legislative campaign and now represents Handy. “What is ultimately a better business decision? To try to change the law in a way that you think works for your platform, or to make sure your platform fits into the existing law?”

Lydia DePillis, For Gig Economy Workers in These States, Rights Are at Risk, CNN Bus. (Mar. 14, 2018), https://money.cnn.com/2018/03/14/news/economy/handy-gig-economy-workers/index.html.

27. Id. The Alabama situation, however, is a bit hard to evaluate from a distance. SB 363 apparently failed to come out of committee in March 2018. Alabama Senate Bill 363, LegiScan (2018), https://legiscan.com/AL/text/SB363/id/1753340.

28. See, e.g., Fla. Stat. § 451.02(1); Tenn. Code Ann. § 50-8-102. Interestingly, most of the laws exempt package delivery and government employers.

29. See Marketplace Contractor Legislation Updated, NDWA Labs, https://www.ndwalabs.org/copy-of-marketplace-contractor-legislation-updated (last updated May 14, 2018).

30. As an aside, these arguments bear a resemblance to those that have arisen under the infamous section 230 of the Communications Decency Act of 1996: “No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.” As one commentator has noted:

One thing that is interesting about the law, especially as we talk about it as this liability shield, is that it’s all-powerful and lets companies ignore everything that’s happening on their platforms. It also doesn’t require platforms to be hands off. They won’t be treated as publishers or take on more liability if they decide to intervene or take things down because it’s offensive or it’s misinformation.

The Law That Built the Internet Economy Is Going International, Marketplace Tech (Dec. 13, 2019), https://www.marketplace.org/shows/marketplace-tech/the-law-that-built-the-internet-economy-is-going-international. Here, the same liability shield is evident; it is the identity of the companies as tech companies that is in question, so this is an interesting case of bootstrapping.

31. Tenn. Code Ann. § 50-8-102 (emphasis added).

32. A.B. 5, 2019–2020 Leg. (Cal. 2019) (codifying Dynamex Operations W., Inc. v. Superior Court, 416 P.3d 1 (Cal. 2018)); see supra note 2. California ballot initiative Proposition 22 exempts “app-based” transportation and delivery companies from AB 5. Ryan Byrne, With Funding from Uber, Lyft, and Doordash, Campaign Behind California Proposition 22 Tops $180 Million, Ballotpedia News (Sept. 9, 2020), https://news.ballotpedia.org/2020/09/09/with-funding-from-uber-lyft-and-doordash-campaign-behind-california-proposition-22-tops-180-million. Following the ballot initiative, in early January 2021, Albertsons announced that in California it was replacing delivery employees with gig and contract workers. Eli Rosenberg, Albertsons Is Laying Off Employees and Replacing Them with Gig Workers as App Platforms Rise, Wash. Post (Jan. 6, 2021), https://www.washingtonpost.com/business/2021/01/06/vons-albertsons-doordash-prop-22-layoffs.

33. See Restatement (Second) of Agency § 220(2)(b) (Am. Law Inst. 1958) (whether or not the one employed is engaged in a distinct occupation or business).

34. See id. § 220(2)(e) (whether the employer or the workman supplies the instrumentalities, tools, and the place of work for the person doing the work); § 220(2)(f) (the length of time for which the person is employed).

35. Tenn. Code Ann. § 50-8-101(1) (emphasis added).

36. Id. § 50-8-101(2) (emphasis added).

37. The Texas market contractor administrative rule applicable solely to unemployment cases appears to at least recognize these criticisms. That rule applies a modified control test to platforms that “accept[] service requests from the public (including third-party individuals and entities) only through [their] digital network[s], and do[] not accept service requests by telephone, by facsimile, or in person at physical retail locations.” 40 Tex. Admin. Code § 815.134(b)(1)(B)(ii).

38. Gobeille v. Liberty Mut. Ins. Co., 136 S. Ct. 936, 944 (2016) (alterations in original).

39. Camargo’s Case, 479 Mass. 492 (2018) (noting varying employee and independent contractor definitions under Massachusetts law).

40. Even online tabulators do not seem able to keep up with the schemata. See Employee or Contractor? The Complete List of Worker Classification Tests by State, Wrapbook (Oct. 25, 2019), https://www.wrapbook.com/worker-classification-tests-by-state (seeming unaware of marketplace platform developments).