The Gig Picture: Recent Trends in Worker Classification

Jun 12, 2019

An increasing proportion of US workers are receiving their income from gig work with the IRS reporting that at least 11.8% of US workers are now participating in part-time work facilitated by app-based providers like Uber and Grubhub.[1] This trend is driven almost exclusively by these platforms in what is now being called the Online Platform Economy. Government agencies like the Department of Labor (DOL) and the National Labor Relations Board (NLRB) have defined their positions with respect to such workers, declaring them independent contractors and the Third Circuit Court of Appeal is due to hand down reasons in Razak v. Uber Techs. Inc., which will determine whether UberBLACK drivers in Pennsylvania are independent contractors or employees under the Federal Labor Standards Act.[2]

The courts and agencies have looked to the common law within the labor context to define their positions, using a group of defining characteristics called the Donovan factors, which focus on the employer’s control over a worker’s duties and finances, as well as the parties’ relationship. Tax common law uses very similar factors and by looking at Razak v. Uber, employers and their advisors can use these recent developments to classify their workers for tax purposes.

This article will discuss the following:

  • The arguments in Razak and the case’s implications;
  • The worker classification tests under the common law; and
  • Available tools for professionals to classify their workers for tax purposes.

Under Razak, Workers are Independent Contractors

Razak v. Uber[3] was the first ruling under federal law regarding Uber drivers and it found them to be independent contractors. Judge Baylson of the US District Court for the Eastern District of Pennsylvania ruled in favor of Uber’s arguments that because workers were free to choose when, where and for whom they worked, Uber had insufficient control over them for employee status. This status would have entitled them to minimum wages and overtime under the Fair Labor Standards Act (FLSA). A further “leading indicator” weighing heavily against the driver’s claims was that they were free to work for Uber’s competitors, like Lyft.

Judge Baylson analyzed each of the six Donovan factors:

  1. the degree of the alleged employer’s right to control the manner in which the work is to be performed;
  2. the alleged employee’s opportunity for profit or loss depending upon his managerial skill;
  3. the alleged employee’s investment in equipment or materials required for his task, or his employment of helpers;
  4. whether the service rendered requires a special skill;
  5. the degree of permanence of the working relationship; and
  6. whether the service rendered is an integral part of the alleged employer’s business.[4]

The drivers’ arguments succeeded only on points four and six, and even then, the court considered them as only barely supporting employee classification. On all other points, the court assessed the asserted control Uber possessed over the drivers—imposing territorial limits, delimiting driver access to the app based on performance and safety, and deducting certain driver earnings—and balanced it against the amount of freedom the drivers had over when, where and for whom they could drive. Looking at both the contract between the parties and the drivers’ activities, the court agreed with Uber in its claims the drivers were, and always had been, independent contractors.

Before a panel of three judges on appeal to the Third Circuit, the drivers argued that the District Court had erred because a reasonable jury could have found for the drivers. Uber argued the plaintiffs had not met their burden and were therefore independent contractors as a matter of law. The judges were most concerned with the fact that the contract expressly stated workers were not employees, the written language explicitly meeting the Donovan factors and defining the parties’ relationship as conforming most closely to that of independent contractor. It quickly dismissed the drivers’ arguments that their reliance on the work as a primary source of income and their subjective interpretation of the contract as creating an employment-like relationship as antithetical to contract law in the employment context.

If the ultimate reasons are based on contract law, it is unlikely the drivers will succeed because the agreement is clear that workers are not employees. The court may also consider implications for the broader economy and the growth of gig work. A finding of independent contractor will allow companies to continue using this employment model that reduces their withholding tax responsibilities and minimizes their wage and overtime obligations. A finding of employee will drastically affect the Online Platform Economy and may even threaten its entire foundation, which relies on flexible independent contractors working during peak periods. At the very least, contracts would need to be redrafted, tax plans re-evaluated, and budget planning overhauled.

Tax Law and Labor Law Both Analyze Control and Party Relationships

Worker classification is relevant for tax purposes because employers are obligated to withhold taxes for employees. Independent contractors take care of their own taxes and the IRS reports that only one-third of Online Platform Economy workers are reporting their self-employment income.[5] In this respect, employers, workers and the IRS should be paying close attention to these developments.

The test for worker classification under tax common law contains up to twenty factors, which the IRS groups under three categories of evidence:

  1. behavioral control;
  2. financial control; and
  3. relationship of the parties. [6]

These twenty factors and the grouping of them accords with the six Donovan factors such that a finding of independent contractor under one test would almost certainly be found under the other as well. This means Razak v. Uber and the reasoning of agencies like the DOL and NLRB are relevant to accurately classifying workers for tax purposes. That being said, circumstances change from business to business and what may work for Uber may not work for other Online Platform Economy participants. Using the facts in Razak v. Uber as a baseline example, employers and their advisors can consider how changes in the contract with their independent contractors can help ensure their workers are correctly classified.

The Path Forward: A Dynamic Approach to Worker Classification

Businesses and their advisers are now being aided by machine learning in determining worker classification for tax purposes. For example, in Razak v. Uber, the court was heavily influenced by the degree of control drivers possessed over their working hours, duration and locations; the fact that they could work for competitors; and that they brought their own car to work. While these factors are central to the court’s decision in that particular case, what does a loss of control over working hours and location, or the employer’s supply of tools, mean to other employers using gig workers? For white-collar businesses using university-educated gig workers, what level of skill or experience converts an employee into an independent contractor or vice versa? What if the worker uses their own computer but software owned and managed by the employer? And what of contract work that continues for months or years?

Previously, tax specialists alone would think through these hypotheticals, guided by case law or static checklists, but now, software driven by statistical analysis into previous court decisions allows all manner of professional to enhance their informed opinion. Blue J Legal’s software does this by looking at the same factors considered by the DOL, NLRB and the Third Circuit Court of Appeal and provides a prediction of how other courts would decide. The above hypotheticals can all be run through software to observe how they have interacted in previous cases and how they might interact in future ones.

The appeal of Razak v. Uber will provide the most authoritative decision to date on the worker classification of Uber drivers with implications for all employers providing app-based facilities in the Online Platform Economy. Should the court side with Uber and declare these workers as independent contractors, it will largely be business as usual in the development of the gig economy. If, however, it decides the workers are employees, contracts will be rewritten and entire business models will be upended, including for tax planning. Employers and their advisors are already using machine learning to navigate a whole host of legal issues and worker classification can stand to benefit greatly from these insights as well.

If you’re interested in learning more about recent developments in worker classification and how machine learning can help practitioners validate their tax positions, join Blue J Legal COO Avi Brudner for a webinar on June 26, 2019.


[1] As reported by Brett Collins et al, ‘Is Gig Work Replacing Traditional Employment? Evidence from Two Decades of Tax Returns,’ IRS (March 25, 2019).

[2] Razak v. Uber Techs., Inc., docket number 18-1944 (January 15, 2019). A summary of the oral arguments can be found here. A recording of the arguments can be found here.

[3] Razak v. Uber, 16-cv-00573-MMB (E.D. Pa. April 11, 2018) (Razak v. Uber).

[4] The DOL in its opinion letter to an unnamed company engaged in what it calls the “virtual marketplace’ also referred to the Donovan factors, though it did not call them as such: Department of Labor Wage and Hour Division, “Opinion Letter FLSA2019-6”, Washington, DC (April 29, 2019).

[5] Brett Collins ibid at footnote 1.

[6] These and the twenty factors can be viewed in the IRS’ ‘Present Law and Background,’ [JCX-26-07]5; Pub. 15-A.