Executive Summary
By 2030, the United Nations projects that 60% of the global population — approximately 5 billion people — will live in urban areas. Cities are simultaneously the engines of economic growth (generating over 80% of global GDP according to the World Bank) and the frontlines of governance challenges: congestion, pollution, housing affordability, public safety, and equitable service delivery. Digital technology offers transformative tools for addressing these challenges, but the "smart city" concept remains contested, with fundamentally different visions competing for implementation.
This policy brief examines smart city governance through the lens of platform economics and public choice theory. We analyse three distinct governance models — Singapore's technocratic optimisation model, Barcelona's digital sovereignty model, and Seoul's participatory data commons model — and identify the critical governance trade-offs that determine whether urban digital infrastructure serves democratic objectives or undermines them. The global smart city technology market, valued at $685 billion in 2025 by McKinsey, is growing at 18% annually — making governance frameworks urgent rather than aspirational.
The Platform Dynamics of Urban Data
Smart cities are, at their core, data platforms. Sensors embedded in infrastructure — traffic cameras, air quality monitors, energy meters, waste management systems, public transport networks — generate vast quantities of urban data. The aggregation, analysis, and application of this data produces the "intelligence" in smart city operations: predictive traffic management, responsive energy grids, optimised waste collection, and targeted public service delivery.
Platform economics provides a framework for understanding the governance implications. Urban data platforms exhibit the same network effects, multi-sidedness, and winner-take-all dynamics as commercial platforms. The more data a city platform collects, the better its analytics; the better its analytics, the more services it can provide; the more services it provides, the more citizens interact with it, generating more data. This positive feedback loop creates powerful centralisation tendencies that raise fundamental questions about who controls urban data and for whose benefit.
The OECD's Smart City Governance Programme identifies three critical governance dimensions for urban data platforms: access (who can use urban data and under what conditions), control (who determines what data is collected and how it is processed), and accountability (how are data-driven decisions challenged and by whom). The answers to these questions differ dramatically across the three models we examine.
Singapore: Technocratic Optimisation
Singapore's Smart Nation initiative, launched in 2014 and now in its second decade, represents the most comprehensive national smart city programme globally. The city-state deploys an integrated sensor network comprising over 110,000 devices across transport, environment, buildings, and public spaces, feeding data into a centralised national digital twin — Virtual Singapore — that models the entire urban environment in three dimensions.
The governance model is technocratic: data collection and analysis are centralised under the Government Technology Agency (GovTech), with policy decisions informed by analytics but ultimately made by appointed officials operating within Singapore's authoritarian-parliamentary system. The model prioritises operational efficiency: traffic optimisation has reduced average commute times by 12%; predictive maintenance of public infrastructure has cut repair costs by 30%; and environmental monitoring has enabled targeted interventions that improved air quality by 15% in high-density districts.
However, the efficiency gains come at a governance cost. Singapore's comprehensive surveillance infrastructure — including facial recognition cameras, movement tracking through the TraceTogether system (originally deployed for COVID-19 contact tracing), and integrated databases linking transport, health, and financial records — enables a degree of population monitoring that most democratic societies would find unacceptable. The absence of comprehensive data protection legislation (Singapore's PDPA exempts government agencies) means that there are few legal constraints on how urban data is used.
Barcelona: Digital Sovereignty
Barcelona's smart city model, developed under the Digital City Plan (2015–2023), represents a deliberate counter-model to technocratic surveillance. The city's approach centres on the concept of "technological sovereignty" — the principle that urban data should be controlled by citizens and their democratic representatives rather than by technology vendors or central government agencies.
In practice, this translates into several distinctive governance innovations. Barcelona's DECODE project developed open-source tools for privacy-preserving data sharing, enabling citizens to share specific data attributes (e.g., "I am a resident of this neighbourhood") without revealing their identity. The city's procurement policies mandate that data generated by smart city systems remain under municipal ownership, preventing vendor lock-in and ensuring that urban data is a public asset rather than a private one. The Decidim platform enables participatory budgeting and policy deliberation, with over 400,000 registered users contributing to decisions about how urban data is collected and used.
The sovereignty model has limitations. By prioritising citizen control over operational efficiency, Barcelona's smart city systems are less technically integrated than Singapore's. Data sharing across municipal departments is constrained by the same privacy protections that prevent surveillance, resulting in analytical silos that reduce the potential for cross-domain optimisation. The participatory governance mechanisms, while democratically admirable, slow decision-making and can be captured by organised interest groups — a well-documented phenomenon in public choice theory.
Seoul: Participatory Data Commons
Seoul's Digital Mayor's Office and its associated Smart Seoul platform represent a hybrid model that attempts to combine Singapore's analytical ambition with Barcelona's democratic values. The centrepiece is the Seoul Open Data Plaza, which makes over 8,400 datasets publicly available, and the citizen participation platform (mVoting), which has facilitated over 7,000 policy consultations since its launch.
Seoul's distinctive innovation is the concept of the "urban data commons" — a framework that treats urban data as a shared resource governed by commons management principles. Citizens contribute data voluntarily through mobility apps, environmental sensors, and service feedback mechanisms, with clear governance rules about how contributed data may be used. The data commons is governed by a multi-stakeholder board comprising city officials, civil society representatives, academic researchers, and private sector participants, with decisions made through deliberative processes.
The data commons model addresses the principal limitation of both the Singapore and Barcelona approaches. Unlike Singapore, data governance is subject to democratic oversight and consent. Unlike Barcelona, the commons framework enables data integration and cross-domain analytics, because the governance rules — rather than technical silos — manage access and use. Seoul's approach demonstrates that operational efficiency and democratic legitimacy are not inherently incompatible, though achieving both requires sophisticated institutional design.
The Surveillance–Service Trade-off: A False Dichotomy?
The conventional framing of smart city governance as a binary choice between surveillance efficiency and democratic privacy is, we argue, overly simplistic. Advances in privacy-enhancing technologies (PETs) — including differential privacy, federated learning, secure multi-party computation, and zero-knowledge proofs — increasingly enable analytics on urban data without exposing individual-level information.
Differential privacy, for example, enables statistical analysis of aggregate patterns (traffic flows, energy consumption, disease incidence) while providing mathematical guarantees that individual records cannot be reverse-engineered. Google's deployment of differential privacy in Chrome's data collection and Apple's use of the technology in iOS analytics demonstrate that privacy-preserving analytics are commercially viable at scale.
Applied to smart cities, PETs could enable Singapore-level analytical capability with Barcelona-level privacy protections. However, PET adoption in urban governance remains nascent: the OECD reports that fewer than 8% of smart city deployments incorporate privacy-enhancing technologies, largely because procurement specifications do not require them and municipal IT teams lack the expertise to evaluate them. Closing this adoption gap is a priority governance intervention.
Governance Recommendations
Drawing on the comparative analysis and the emerging PET landscape, we propose five governance principles for smart city digital infrastructure:
- Data as public infrastructure: Urban data should be treated as public infrastructure — publicly owned, governed by transparent rules, and accessible for public benefit — rather than as a by-product of service delivery to be exploited commercially.
- Privacy by architecture: Smart city systems should be designed with privacy-enhancing technologies as default components, not optional add-ons, ensuring that surveillance is architecturally difficult rather than merely legally prohibited.
- Participatory governance: Decisions about what data is collected, how it is processed, and for what purposes should involve structured citizen participation, drawing on Seoul's data commons model.
- Interoperability and vendor independence: Open standards and data portability requirements should prevent vendor lock-in, ensuring that cities retain sovereignty over their digital infrastructure.
- Equity audits: Smart city deployments should be subject to regular equity audits assessing whether digital services reach all communities or exacerbate existing spatial and socioeconomic inequalities.
Implications for GDEF's Technology & Transformation Working Group
With $685 billion flowing into smart city technology annually, the governance frameworks adopted today will shape urban life for billions of people over the coming decades. The choice is not whether cities become "smart" but how — and for whom. GDEF's Technology & Transformation Working Group will advance urban digital governance principles in its programme of work, with particular focus on privacy-enhancing technology adoption and participatory data governance frameworks.
References & Sources
- UN-Habitat, World Cities Report 2024: Urbanization and Climate Change. unhabitat.org/world-cities-report
- OECD, Smart City Governance Programme: Framework and Case Studies. oecd.org/smart-cities
- McKinsey Global Institute, Smart Cities: Digital Solutions for a More Livable Future, 2025 Update. mckinsey.com/mgi
- ITU, Key Performance Indicators for Smart Sustainable Cities. itu.int/ssc
- Government Technology Agency of Singapore, Smart Nation: The Way Forward. smartnation.gov.sg
- Barcelona City Council, Barcelona Digital City Plan 2020–2023. ajuntament.barcelona.cat/digital
- Seoul Metropolitan Government, Smart Seoul 2025 Master Plan. smart.seoul.go.kr
- Ostrom, E. (1990). Governing the Commons. Cambridge University Press. doi.org/10.1017/CBO9780511807763
- Dwork, C. (2006). "Differential Privacy." Proceedings of the 33rd International Colloquium on Automata, Languages and Programming. doi.org/10.1007/11787006_1