For Gig Workers, Portable Benefits Are Only Half the Solution

When I worked on gig economy policy in Southeast Asia for Grab, the Singaporean super-app for ride-hailing and other services, and later at Grubhub in the United States, I understood why working on platforms like these mattered to people.

They provided a real source of income and flexibility and widened access to earning opportunities in a way that traditional labor markets simply did not. People came into this line of work from all walks of life. We saw veterans, single parents and people with disabilities all join these companies’ workforce. One’s background generally did not matter.

It’s well-known that work in the US is structurally different. Because the social safety net is tied to traditional employment, gig workers, many of whom are contractors, fall outside this paradigm, leaving them exposed in the case of a health emergency or if work disappears.

Portable benefits — those that workers can utilize as they move from one gig to another — seem like an attractive idea to address these issues.

Instead of tying benefits to a single employer, the portable benefit model aims to attach them to the worker and allow contributions to follow them across platforms and gigs. The Aspen Institute’s Future of Work work has been a key reference point on this concept for years, both for the design principles and for how portable benefits can be implemented in practice.

In Congress, proposals like Unlocking Benefits for Independent Workers Act (S. 2210) try to make the concept easier to operationalize by clarifying that providing portable benefits should not, by itself, be treated as evidence in federal worker-classification determinations. That has been a worry for many platforms in the past. The logic is to sidestep the classification question and expand access to benefits.

However, portable benefits by themselves cannot solve an even more basic problem.

Gig work on tech platforms is largely governed by computer programs and software, and the rules that matter most to a worker’s stability can feel opaque. This is where the portable benefits conversation can become too narrow.

Benefits can be portable in a legal sense, but if a worker’s access to earnings remains fragile in a practical sense it might not be enough to serve the purpose. For instance, a food delivery driver’s app could stop offering them orders without explanation, or a worker could be deplatformed by an automated fraud detection system with no way to appeal. These are real stories, and because these platforms operate on a massive scale, serving millions of people, the rules on who gets a customer order, pay calculations, performance scoring, fraud detection and deactivation decisions are frequently mediated through software systems that workers may not see, comprehend or even meaningfully contest.

Human Rights Watch’s 2025 report on gig labor, The Gig Trap, documents how major US platforms use algorithmic systems to shape pay and working conditions. The report is not subtle about the power imbalance: workers can face sudden drops in income, and they often do not have clear explanations for why their work volume changes or why they are removed from platforms.

The underlying governance issue is hard to dismiss: a portable benefit system might not help much if a worker loses their access to work overnight through automated processes they couldn’t appeal.

In the gig economy, a platform deactivating a worker’s account can function like an ordinary termination, but without the procedural norms people assume exist in the formal labor market, such as access to advance notice, protections against wrongful dismissal or even eligibility for unemployment insurance.

That is precisely why some cities have begun to treat algorithmic decisions on a worker’s account status as a due process issue rather than a customer support workflow.

Seattle’s App-Based Worker Deactivation Rights Ordinance is a good example: it establishes notice and challenge processes, allowing gig workers to understand why they were deactivated and how they could challenge their deactivation. New York City has also moved in this direction through its delivery-worker rules on pay transparency measures.

These are not portable benefits laws per se, but they are responding to a core reality: gig work is deeply shaped by systems that workers experience as black boxes.

If portable benefits are meant to reduce precarity, then algorithmic opacity is a direct threat to that goal.

Consider what happens if a benefits program is funded as a percentage of earnings, yet the definition of earnings is itself unclear to workers because pay is shaped by dynamic pricing, work incentives and time definitions that not only vary by platform but change instantly.

Consider the worker suddenly receiving fewer orders because of automated routing decisions, and as a result cannot hit the earnings threshold that makes them eligible for a benefit. Or the worker flagged by systems designed to catch fraud and removed from the platform, losing income and, in practice, the ability to sustain contributions.

In each case, benefits may be technically available, but the worker’s ability to rely on them is undermined by the instability created elsewhere.

A portable benefits agenda needs a companion pillar grounded in algorithmic transparency and due process.

Importantly, transparency does not have to mean that companies should be forced to publish trade secrets or give away fraud controls. The standard should be more practical where workers should be able to understand, at a minimum, how their pay is calculated, what the main factors are that affect access to offers, what behaviors can trigger enforcement actions and what steps exist to contest a high-impact decision like deactivation.

When enforcement actions are taken, including issuance of warnings, restrictions, or deactivations, platforms should be required to provide a written explanation citing the specific rule or signal that triggered the decision, the timeframe in question and the process available to contest it.

This is not an extraordinary ask. It is roughly analogous to what we expect of employers in traditional labor settings. These should be the norm on a national level, rather than leaving it to the local level where rules can change depending on a worker’s location.

The second pillar is auditability. A portable benefits system is only as good as the data underlying it. If earnings definitions vary across platforms or certain payments can be classified outside the contribution base, the system becomes easy to game and hard to enforce. Policymakers should require platforms to maintain standardized records of worker earnings, offer rates and account status changes that are accessible to workers and, where appropriate, to oversight bodies.

Third-party auditing should be a baseline requirement. Without this infrastructure, portable benefits risk becoming a paper promise — technically available but practically inaccessible.

Portable benefits are worth pursuing, and the US needs creative approaches because the safety net is so entangled with employment status. Yet we should not pretend portable benefits are sufficient on their own.

The daily reality of gig work is that the boss is often an algorithm, and the most important decisions about stability are frequently opaque. Policymakers must pair portable benefits with algorithmic transparency in order to produce real worker security. Anything else means we are building a modern benefits infrastructure on top of an invisible system that can still pull the floor out from under workers at any time.

This will take serious work. According to Deloitte, 57 million Americans participate in the gig economy, accounting for 37 percent of the US workforce, so the cost of inaction is not abstract.

New York City's Black Car Fund, a workers' compensation insurance program for taxi and app-based drivers funded by a small fee added to each fare, shows it can be done: a simple surcharge mechanism, a defined eligibility structure and a clear collection process have delivered real benefits to drivers for years.

That model will not translate directly to modern app-based work, but it proves the structure is buildable. The question is whether policymakers are willing to build it before the workforce that needs it grows any larger.