Algorithmic Management, Monitoring, and Control: Workforce Planning in the Digital Age
These days, it’s hard to learn anything about workplace politics without running into “algorithmic management.” We are told that companies are increasingly controlling the workforce through an array of digital “tricks”. These companies record our keystrokes, track our locations, and watch us through our web cameras. We hear this same story in academic journals, government reports, and popular newspapers. In fact, this issue has even made its way into federal law—specifically, the current US Department of Labor law regarding independent contractors. Like most popular accounts, this rule assumes that algorithmic management is in full swing. And it takes practice as a form of “control.”
There is only one problem: algorithmic management is not real. A term used to describe business processes that are so old and commonplace that they would be boring if they weren’t tagged with the dreaded label. These processes include monitoring performance, providing incentives, and tracking performance. Critics of these practices have successfully marketed them as something new and bad for American workers. And for several years now, that misleading narrative has been shaping policy at the Department of Labor.
Fortunately, the narrative may be over. The Department of Labor recently proposed to eliminate this concept from its regulations and return to the general definition of “control” in the workplace. This definition would recognize, at least implicitly, that incentives and monitoring are not mechanisms of control in themselves. Properly understood, they are ways in which businesses treat their suppliers over which they have no control. That is, it is not proof of control, but its opposite. And that’s true even when packaged under the dreaded label of “algorithmic management.”
Algorithmic Management: The Empty Set
The term algorithmic management has always been vague. It was founded in a 2015 academic paper focusing on rideshare platforms of the time. The paper studied how these platforms use digital tools to connect dozens of independent drivers. The paper focuses on these three tools: matching passengers and drivers, customer rating systems, and surge pricing. The paper found that these tools not only helped platforms balance driver supply and passenger demand, but also kept drivers happier. Algorithmic management, it turned out, wasn’t such a bad thing.
But since then, the word has taken on a life of its own. Now it’s being used in tools ranging from software programming to chatbots to AI, the last of which has raised the alarm. Writers such as Noam Scheiber of the New York Times, Sam Levine of the Federal Trade Commission, and Veena Dubal of UC Irvine have described algorithmic management in dangerous terms, arguing that companies are using it to trick people into working long hours for less money. Dubal has even compared it to a modern day Jim Crow.
These arguments are not just rhetoric. They have begun to shape public policy. In 2024, the Department of Labor adopted a regulation defining the difference between employees and independent contractors. That distinction is important because under the Fair Labor Standards Act (FLSA), the nation’s primary wage and hour law, only workers are entitled to minimum wage and overtime. The FLSA distinguishes between these categories of workers based on a number of factors, including who “controls” the work. The 2024 Act defined this type of control to include certain types of algorithmic management. In particular, it showed “technical” monitoring, such as GPS tracking. That type of monitoring, the regulation says, counts as control even if it is not combined with other measures. Technology surveillance itself was a way to control someone else’s work.
That way of doing it is a deviation from the traditional rules of the department. Traditionally, control meant some kind of affirmative action: the business had to order, prevent, or punish the employee for doing something. But the 2024 rule expanded the concept to include mere glances. If a business simply collects information about a job, it can be considered to control the job—as long as it does so “professionally.”
Analog Logic for a Digital World
This kind of thinking takes the issue backwards. Incentives, monitoring, and similar techniques are not signs of control: they are signs of your absence. In economic terms, they are the means by which businesses solve the “principle-agent problem.” The basic problem is this: when a business hires a contractor, the two have different interests. While the business wants the best service at the lowest price, the contractor wants the highest value for the least effort. Therefore, in order to protect its interests, the business must use safeguards against slow operations. If it used its own workers, it could control that risk by telling the workers what to do. But it can’t do that when it uses contractors, often because it doesn’t know enough about the job itself. (That’s why it hired a contractor in the first place.) So instead, it may require the contractor to report when it reaches certain milestones (monitoring). Or it may offer the contractor additional payment for higher quality deliverables (incentives).
This type of indirect influence has never been considered “control” for the purpose of separating employees—nor should it. If that were the case, it would be difficult to classify any work as independent. Because all principals monitor their agents to some degree, monitoring itself tells us nothing about how to classify the relationship. For example, if a business ships a package and pays for delivery confirmation, it is monitoring the transaction in some way. But no one thinks that the law should treat every supplier as an employee of all its customers.
None of this changes with modern technology. Even though businesses and contractors use technology, they still face agent problems. And the best tools to deal with those problems are still monitoring, incentives, and similar indirect methods. While these methods may be working faster in the digital world, the dynamics are the same—even wrapped in the dreaded label of “algorithmic management.”
Of course, this great efficiency of algorithmic tools is exactly what makes people uncomfortable with them. People worry that if an employer can monitor their keystrokes or access their webcams, the employer may use the information to control their work. But the important point is not whether the employer collects the information: it is how the employer uses the information. If an employer uses keystroke data to evaluate, compensate, or penalize an employee, the use instead of data counts as control. Legally, the question is whether the principal controls the work—not whether the principal knows how the work is done. That is true whether the data is collected through technology, delivery receipt, or personal observation.
Fortunately, the Department of Labor is coming around to that idea. At the end of February 2026, it published a new law to distinguish between contractors and employees. This new law does not say anything about algorithmic management or any other type of “technical” management. Instead, it goes back to the original principles: if a person is in control of his work, he is probably a contractor. If he doesn’t, he’s probably an employee. The point is the same even when work enters the digital world.
