Indonesia's loose partnership model leaves ride-hailing drivers without decent work protections; Spain's Riders' Law, which presumes employment and curbs algorithmic management, provides a practical template for stronger labour safeguards.

Main Finding

Indonesia’s sectoral, non-binding regulation of ride-hailing (centered on Minister of Transportation Regulation Permenhub 12/2019 and a “partnership” model) fails to satisfy core decent-work standards under the Fairwork framework (fair pay, conditions, contracts, management, representation). Spain’s Riders’ Law (Real Decreto‑Ley 9/2021), by contrast, provides a more coherent and enforceable model—presuming employment status for drivers and regulating algorithmic management—offering practical regulatory elements Indonesia could adopt to reduce platform-driven precarity.

Key Points

• Problem framing
  • Platform capitalism concentrates control of data and access with platforms, enabling algorithmic management that displaces formal employer responsibilities and creates worker precarity.
  • Algorithmic management functions as de facto employer control (fare-setting, task allocation, sanctions), undermining the autonomy rationale used to classify drivers as independent partners.
• Comparative assessment (using Fairwork indicators: fair pay, fair conditions, fair contracts, fair management, fair representation)
  • Indonesia
    • Regulatory design: fragmented, sectoral administrative rules; partnership model institutionalized by Permenhub 12/2019.
    • Shortcomings: inadequate income security, weak social protection, contractual mismatch with factual working conditions, limited procedural safeguards around algorithmic decisions, and weak collective representation mechanisms.
    • Specific gaps identified in Permenhub 12/2019 (examples): service-fee mechanics tied to app display but no wage-floor enforceability; partnership classification; complaint centers and limited procedural protections but no strong collective-bargaining anchors.
    • Social response: driver mobilizations (e.g., May 20, 2025 protests) indicate widespread dissatisfaction.
  • Spain
    • Regulatory design: systemic integration of platform work into general labour law via the Riders’ Law.
    • Key reforms: presumption of employment (shifts burden of proof), recognition of algorithmic management as employer conduct, mandatory algorithmic transparency and procedural guarantees.
    • Outcome: a clearer legal pathway to social protection, contributions, and collective bargaining for platform drivers.
• Recommendations (from the article)
  • Normative reclassification of drivers toward employment where factual control exists.
  • Procedural and transparency guarantees for algorithmic management (disclosure, auditability, due process for sanctions).
  • Legal support for collective representation and bargaining for platform workers.
  • Legislative anchoring of platform work into general labour law, replacing sectoral administrative patchworks.

Data & Methods

• Methodological approach
  • Normative juridical research (legal doctrinal analysis) with legislative, conceptual, and comparative methods.
  • Vertical normative review: assessed Permenhub 12/2019 against the Indonesian Constitution, Manpower Law, and international labour/human-rights obligations.
  • Comparative legal analysis: contrasted Indonesia’s sector-specific administrative approach with Spain’s systemic labour-law integration.
  • Substantive-equality lens: focused on whether law addresses structural vulnerabilities (economic dependency, algorithmic control, unequal bargaining power).
• Analytical instrument
  • Applied the five Fairwork indicators (Heeks et al.) as legal benchmarks and organized evaluation hierarchically (basic to advanced indicators).
• Sources
  • Primary statutes/regulations (Permenhub 12/2019; Spain’s Riders’ Law / Real Decreto‑Ley 9/2021), judicial rulings, international ILO materials, doctrinal and policy literature, and reported worker mobilizations.
• Presentation
  • Descriptive, evaluative, and argumentative synthesis; tabled mapping of Permenhub provisions to Fairwork indicators.

Implications for AI Economics

• Algorithmic management as an economic and regulatory object
  • Treat platform algorithms not just as optimisation tools but as instruments of labour governance with distributive consequences (wage determination, task allocation, sanctions).
  • Regulation that presumes employment status and requires algorithmic transparency shifts legal and economic incentives: platforms internalize labour costs (wages, social contributions) and face constraints on opaque automated control.
• Data and rent extraction
  • Platforms’ control of user/worker data is a source of rent; legal reclassification and transparency measures can alter data governance incentives and the value extraction model.
  • Mandated disclosures and audits reduce information asymmetries, raising compliance costs and potentially reducing surplus capture from drivers.
• Labour supply, wages, and firm responses
  • Enforceable wage floors, social security obligations, or collective bargaining can raise driver compensation but may lead platforms to adjust prices, reduce margins, reduce driver hours, or accelerate automation.
  • Platforms may respond by: (a) changing contractual forms (more subcontracting), (b) redesigning algorithms to reduce labor intensity, or (c) automating services where feasible—each with distinct welfare and market-structure effects.
• Market structure and competition
  • Stricter labour and algorithmic rules may raise entry costs and alter competitive dynamics—larger incumbents may be better positioned to absorb compliance costs, potentially increasing concentration.
• Policy trade-offs and design
  • Trade-off between worker protection and platform-driven service affordability/accessibility; policy must design enforceable but flexible rules (procedural algorithmic constraints, auditing standards, transitional supports).
• Research agenda for AI economics
  • Empirical evaluation strategies:
    • Difference‑in‑differences or synthetic control analyses using Spain’s Riders’ Law (pre/post) to estimate impacts on wages, hours, platform prices, driver welfare, and supply.
    • Natural experiments exploiting staggered enforcement or municipal-level variation in Indonesia.
    • Platform-level data collection: scraping, APIs, partnerships, admin/social-security records to measure algorithmic outputs (dispatching patterns, dynamic fares), driver earnings, and churn.
    • Algorithmic audits and black‑box testing to quantify opacity, bias, or incentive effects on driver behavior.
    • Qualitative interviews and surveys for worker bargaining power, perceived autonomy, and enforcement experiences.
  • Model development:
    • General-equilibrium or structural models capturing interactions among platform pricing, algorithmic assignment, driver labor supply, and regulatory constraints.
    • Models of firm response to reclassification: endogenous decisions about automation, subcontracting, or geographic/service scaling.
• Policy research and evaluation priorities
  • Cost–benefit analysis of mandating employment presumption vs. sectoral regulation.
  • Design and impact of procedural algorithmic governance (what level of transparency/auditability actually reduces harm without exposing proprietary IP).
  • Mechanisms to support enforcement (data-sharing mandates, fines, independent algorithmic auditors).
  • Distributional outcomes—who gains/loses across drivers, consumers, and shareholders.

Key takeaway for AI economists and policymakers: regulating algorithmic management and reclassifying employment status are not just legal remedies but economic interventions that reshape incentives, data governance, firm strategies, and welfare outcomes. Comparative legal reforms—like Spain’s Riders’ Law—offer testable policy treatments whose market impacts should be rigorously evaluated with mixed quantitative and qualitative methods.