The Quiet Erosion: Why Labour Rights Backsliding Belongs in the Democratic Decline Conversation

The Quiet Erosion: Why Labour Rights Backsliding Belongs in the Democratic Decline Conversation

A new CEPS analysis argues that the erosion of collective labour rights across Europe constitutes a form of democratic backsliding that rarely makes headlines but fundamentally reshapes how citizens can contest power. With the ITUC Global Rights Index showing Europe's worst scores since 2014, the question is whether policymakers are paying attention to the right indicators.

The intersection of labour rights, democratic governance, and technological transformation is precisely the kind of structural question that deserves sustained attention. Human x AI Europe on May 19 in Vienna brings together the ecosystem to work through these tensions directly.

Three Strikes, Three Democracies

Belgium, 1893. Italy, 1943. Poland, 1980. A new CEPS analysis opens with these three moments when workers' collective action fundamentally expanded democratic rights. The Belgian general strike won suffrage extension. The FIAT Mirafiori walkouts marked the first mass resistance against fascism. Solidarność helped collapse Soviet-backed rule.

The argument is straightforward: when discussions of democratic backsliding focus exclusively on rule of law, judicial independence, media freedom, and corruption, they miss a crucial dimension. The right to strike, protest, and organise is under growing pressure across Europe. This pressure rarely makes headlines, but it quietly reshapes the mechanisms through which citizens can contest power.

Is this framing correct? The question deserves disaggregation.

What the Numbers Actually Show

The 2025 ITUC Global Rights Index provides the empirical foundation. Europe and the Americas recorded their worst scores since the Index began in 2014. Europe's average rating has deteriorated from 1.84 in 2014 to 2.78 in 2025, representing the largest decline of any region worldwide over the past decade.

The specific violations are worth examining:

• 87% of countries violated the right to strike
• 80% violated the right to collective bargaining
• 72% of countries restricted workers' access to justice, the worst figure ever recorded
• 45% of countries restricted freedom of speech and assembly

According to La Strada International's analysis, the right to strike was restricted or denied in 73% of European countries specifically. Workers in over half of European countries had limited or no access to justice, a sharp increase from less than a third in 2024.

The ILO notes that Georgia's rating fell from 3 to 4 due to systematic violations, including a "foreign influence" law introduced without trade union consultation. Italy also lost its top-tier rating in 2025.

The Disagreement Worth Having

The CEPS argument connects labour rights erosion to democratic backsliding. But this connection can be understood in at least three different ways, and they have different implications:

The instrumental view: Collective action rights are tools that enable citizens to contest government policy. When these tools are weakened, democratic accountability suffers regardless of whether other institutions remain intact. This is the strongest version of the argument.

The symptomatic view: Labour rights erosion tends to correlate with broader authoritarian tendencies. The same governments that restrict strikes often also target media and courts. Labour rights are an early warning indicator rather than a separate phenomenon.

The structural view: The capacity for collective action is constitutive of democracy itself, not merely instrumental to it. A democracy where workers cannot organise is not fully democratic, even if elections remain free.

These three framings suggest different policy responses. The instrumental view calls for protecting specific legal rights. The symptomatic view calls for monitoring labour rights as part of broader rule-of-law assessments. The structural view calls for reconceptualising what democratic health means.

The Technology Dimension

The CEPS analysis does not extensively address automation and AI, but the connection is increasingly unavoidable. Research from the UC Berkeley Labor Center shows that collective bargaining is already being used to give workers voice over automated management systems. Unions are negotiating who gets notified before scheduling algorithms go live, who sits on tech committees, and what limits exist on electronic monitoring.

Partnership on AI has documented how unions in Ireland, Italy, and the United States have secured AI protections for roughly 30,000 workers through collective bargaining. The Financial Services Union of Ireland, SLC-CGIL of Italy, and SEIU Local 668 in Pennsylvania all negotiated specific provisions around AI deployment.

Recent analysis from Felhaber Larson notes that the International Longshoreman's Association secured a contract provision prohibiting all fully automated technology, while a Las Vegas Culinary union agreement contractually obligates employers to bargain over any decision to implement AI.

The question this raises: if collective bargaining becomes a primary mechanism for governing AI deployment in workplaces, what happens when collective bargaining rights themselves are weakened?

The Legal Landscape Is Shifting

The picture is not uniformly negative. The UK's Employment Rights Act 2025 removed minimum service level rules for strikes, increased dismissal protection for industrial action, reduced notice requirements from 14 to 10 days, and extended industrial action mandates from 6 to 12 months.

Meanwhile, the European Committee of Social Rights found in March 2026 that Italy violated Article 6§4 of the European Social Charter because its definition of "essential public services" was too broad, effectively constraining the right to strike beyond what international standards permit.

These developments point in opposite directions. Some jurisdictions are strengthening labour rights while others are weakening them. The aggregate trend, as measured by the ITUC Index, is negative. But the variation matters for understanding what policy levers exist.

What Would Have to Be True

For the CEPS argument to hold, several claims need to be defensible:

First, that labour rights erosion is not simply a byproduct of economic pressures but reflects deliberate policy choices. CEPA's analysis of democratic backsliding suggests that attacking constraints on power, including labour rights, is a "fast route to power" that politicians across Europe are discovering.

Second, that the erosion is systematic rather than episodic. The ITUC data showing consistent decline over a decade supports this.

Third, that the connection to democratic health is causal rather than merely correlational. This is harder to establish empirically, but the historical examples (Belgium, Italy, Poland) suggest that collective action capacity has been constitutive of democratic expansion.

Fourth, that existing democratic backsliding frameworks are incomplete without labour rights. Carnegie Endowment's work on European democratic backsliding focuses on media freedom, individual rights, and rule of law. The CEPS argument is that this framework misses something important.

The Question That Remains

The CEPS analysis ends with a provocation: labour rights backsliding might not make headlines, but it is quietly reshaping European democracies. The question is whether this framing changes anything.

If policymakers accept that labour rights belong in democratic health assessments, what follows? The EU's Rule of Law reports could incorporate labour rights indicators. The European Social Charter's enforcement mechanisms could be strengthened. AI governance frameworks could explicitly protect collective bargaining rights.

If policymakers reject the framing, they need to explain why the right to organise, bargain collectively, and strike is categorically different from other democratic rights. That explanation would itself be clarifying.

The debate is worth having explicitly rather than by default. When 87% of countries violate the right to strike and Europe records its worst labour rights scores in a decade, the question of whether this constitutes democratic backsliding is not academic. It determines what gets measured, what gets monitored, and what gets defended.