Turkey gig worker platform rights

Turkey's Courier Insurance Bill Fixes a Real Coverage Gap, but a Blanket 'Very Hazardous' Label Is the Blunter Tool

DEM MP Cengiz Çiçek's bill would insure all motorcycle couriers through platforms. The coverage goal is sound, but the design needs sharper targeting.

Motorcycle Courier Safety in Turkey People of Internet Research · Turkey 44+ Courier deaths in 2025 At least 44 couriers died at work,… 63 Deaths in 2024 (recorded) The 2025 fall does not show risk i… 61% Victims aged 18–28 27 of the 2025 victims were in thi… ~50% Victims on digital platforms Half of couriers who died in 2025 … peopleofinternet.com
Motorcycle Courier Safety in Turkey People of Internet Research · Turkey 44+ Courier deaths in 2025 63 Deaths in 2024 (recorded) 61% Victims aged 18–28 ~50% Victims on digital platforms peopleofinternet.com

Key Takeaways

On 26 October 2026, Evrensel reported that DEM Party Istanbul MP Cengiz Çiçek had submitted a bill to the Grand National Assembly (TBMM) on motorcycle couriers. The bill is an opposition proposal, not a government one, and the report does not give the date of formal filing. The mechanics matter, so this piece sticks to what the report states.

What the bill would do

Per Evrensel, the bill has four moving parts:

An earlier DEM proposal covers similar ground. Bianet reported in January 2026 on a bill by DEM's Diyarbakır MP Ceylan Akça Cupolo. It also sought the "very hazardous" class, employer duties on protective equipment and training, and oversight of algorithmic speed pressure. Whether the two texts are linked is not clear from the reporting, and neither has reportedly advanced.

The strongest case for the bill

The argument for reform is serious. The Courier Rights Association counted at least 44 motorcycle couriers killed at work in 2025, including four minors. Half of those who died were working for digital platforms. The association warns that the figure may be higher, because no official data tracks courier deaths. It also says the fall from 63 recorded deaths in 2024 does not show that risk is declining.

When a rider is classed as an independent "partner", the platform can capture the benefit of fast delivery while the injured rider carries the cost. A rider with lapsed premiums can lose benefits at the moment of greatest need. The bill targets exactly that, and removing premium-debt conditions from incapacity and survivor benefits is a defensible way to stop the penalty falling on the injured worker or their family. Insurance coverage that follows the work, not the contract label, is a proportionate principle.

Where the design is blunter than it needs to be

The coverage goal deserves support. Two design choices deserve scrutiny.

First, the hazard class. Turkey's occupational safety framework sorts workplaces into hazard classes, and the class drives compliance duties and costs. Bianet notes that a Resmî Gazete decision of 21 May 2025 had already moved courier work from "less hazardous" to "hazardous". Moving it again by statute would take the call away from the expert commission that sets classes from accident data. A fixed legislative label also cannot tell a rider in dense Istanbul traffic from a part-time rider in a small town. Pro-innovation regulation should price risk where it occurs, using accident data and employer track records. A blanket top-tier label across every platform and restaurant that dispatches a rider is a cruder instrument.

Second, making platforms pay for self-employed work they only partly control. Treating the organising platform as the reporting employer makes sense where it dispatches, times and rates the delivery. It is harder to defend for a neighbourhood restaurant that uses its own rider or a freelancer, and the small business may lack the administrative capacity to file premiums correctly. A workable version would key liability to the degree of platform control, such as assignment, time targets and ratings, and would give small merchants a simple payment route. Treating all courier arrangements the same would push the smallest operators into informality, which is the opposite of the bill's aim.

The pressure that drives fatalities is also upstream of insurance. Insurance pays after a crash, and it does nothing to prevent one. Cupolo's earlier proposal went after algorithmic time pressure. Transparency duties on delivery-time targets and dispatch logic, with independent audit, would address accident causes more directly than a hazard label alone.

A wider debate

Turkey is not alone in this argument. ERR News reported that Estonia has passed a Platform Work Act strengthening rights for Bolt and Wolt drivers and couriers. In Europe, a Digital Omnibus proposal now before the EU Parliament is drawing criticism from unions. Netzpolitik reports that a European Trade Union Institute study finds it could make workplace data abuses harder to prove, including hidden algorithmic metrics. The common thread is that platform labour rules are shifting from status labels to data and accountability, and Turkey's debate should look there too.

What to watch

As an opposition bill, it faces a hard path in a parliament where the governing majority sets the agenda. Its likeliest effect is on the debate. The government can adopt the coverage principle, meaning accident and disease insurance that does not depend on contract labels or the worker's premium record. It can pair it with data-driven hazard classification and platform duties scaled to actual control over the rider. That path would protect couriers while keeping delivery platforms viable, and it would give Turkish regulators the accident data they currently lack. The 2025 toll shows the gap is real. The remedy should be matched to where the risk comes from.

Sources & Citations

  1. Evrensel: DEM's Çiçek courier bill
  2. Bianet: Cupolo courier bill (Jan 2026)
  3. Bianet: 44 couriers killed in 2025
  4. Law 5510, Social Insurance and General Health Insurance (mevzuat.gov.tr)
  5. Law 6331, Occupational Health and Safety (mevzuat.gov.tr)
  6. ERR News: Estonia Platform Work Act
  7. Netzpolitik: Digital Omnibus and workplace data