Public sector innovation’s greatest enemy

Breaking down silos in public sector organisation is an essential step in spreading innovations across the system.

The public sector is siloed. Collaboration between different public entities is close to impossible. Pockets of innovation exist, but there is no system-level diffusion. Replicability and scale-ups are often too complex to see the light of day.

These are common sources of worries for many public servants across national public sectors – shared almost universally.

Public administrators are constrained by institutions that allow for little change. These institutions in turn maintain both systematic and physical divisions between agencies, departments and teams. Coordination between different fields – a necessary condition for the multidisciplinary nature of public sector innovation – is marginal.

Theorising Non-coordination

This is an excellent real-life reflection of the game theory behind the non-coordination problem, whereby individual preferences (or that of small teams), when aggregated at the level of larger groups, do not translate into similarly ordered preferences. This is in essence Mancur Olson’s argument who, in what remains a key piece in the field of strategic coordination, writes that “rational, self-interested individuals will not act to achieve their common or group interests” (1965, p.2).

A simple logical game supports this thesis. Imagine 99 public servants must vote on a new project. An absolute majority (50) is required for the project to be adopted. The main question to ask here is: will there be a majority in favour of adoption?

To better investigate the issue at hand, the five conditions below must hold true for the example to make sense:

  1. Public servants can abstain. There is no pre-existing law or norm forcing participation;
  2. Should the project not be implemented, and the status quo remain, all public servants are indifferent and receive a payoff of 0;
  3. Participation to the voting process incurs a cost. This cost takes a value of c, and c > 0. Abstention, logically, incurs no cost;
  4. All public servants are better off with the new project in place, for which they receive a payoff of B. Thus it must be true that B > c > 0;
  5. Finally, there is complete information asymmetry: no one has any information on others’ possible set of actions.

If this seems a little complex, let’s now walk through all possible outcomes this game could provide.

If 98 vote in favour (with final payoff of Bc, or the full benefit minus the cost of participation), then the one remaining public servant is likely to abstain such that no participation cost would be incurred, and she would benefit fully from the project. She would receive a payoff of B, and would be better off than all of her colleagues – a flagrant case of free-riding. The exact same logic applies for any instance where an absolute majority is reached.

Should 50 abstain however, the game changes. A majority no longer holds, and the project is not adopted. Abstainees’ payoff is equal to zero, and they remain indifferent. The other 49 feel differently. They participated but received no benefit. Their final payoff is thus equal to –c; that is, the participation cost. They are worse off than before the vote even took place (and probably a little frustrated).

Because this game does not allow for full information, no public servant will rationally run the risk of participating and not receiving the benefit. Logically, they will never participate, such that they receive B (which is larger than B – c) if the project is adopted, and 0 (which is larger than –c) if it is not. This leaves us with a strange case of public servants asking for the implementation of a project but the group’s outcome resulting in… no project. Welcome to the beautiful world of non-coordination.


What This Mean for Public Sector Innovation

Let’s turn back to the public sector. Given bureaucracy-heavy institutions where costs of change are high and success rates are low, incentives to innovate, or the mere provision of some form of system change, are minimal. Information asymmetry runs high, thus preventing innovators in one department from anticipating how their colleagues and managers across other departments may behave. Innovation remains highly local and contextual, and scale-ups are  rare. In a world of non-coordination, no one wants to act. No information: no systemic change.

Do not get me wrong: this does not mean that innovation is non-existent. The OECD’s Observatory of Public Sector Innovation (OPSI) has kept track of thousands of initiatives across the world. While some are context-specific and never transcend the walls of one office, others have spread across all public institutions, resulting in true institutional changes. Innovation hubs spring up like mushrooms, both independent and state-powered. More and more public servants are brought together around innovative projects.

This is not enough. A new narrative must see the light of day in the public sector, for people to become less risk-averse and institutions to become more prone to change. As Elinor Ostrom puts it in her famous Crowding Out Citizenship (2000), we need more “conditional co-operators”. If more information about others’ behaviours is accessible, then coordination is made easier and some public servants will be rationally willing to share the costs of participation for a better-than-today outcome. Mancur Olsen can be proven wrong.

Fostering connections and better communication within and across departments will allow best practices to be identified and scaled up across the organisation and beyond.


Where To From Here?

Information asymmetry, then, is innovation’s greatest enemy. It is now a necessity to make information across public institutions more accessible, user-friendly and thus less siloed. More must be done for conditional cooperators to proliferate and inter-department coordination to be strengthened. Public servants’ ability to know what their counterparts want and do in other parts of the public service may be enough for coordination to occur. We can no longer do without digestible information.

With this in mind, the use of new technologies and advanced data collection techniques must be brought to the fore. Data sharing – through digitalisation, for example – on departments’ strategic plans, projects, teams etc. is crucial. As the latest OPSI report puts it, “[…] data, information and knowledge are building blocks for innovation, and their free flow within and across public sector organisations in an important condition for building individual and organisational capacity to innovate” (2017, p.199). At the same time, inter-department workshops, perhaps through innovation hubs, must be the new norm for colleagues to meet, understand what obstacles they face, and ultimately find ways in which they can advance their interests through the means of successful coordination.

With more information, however, comes the risk of information overload. This must not be overlooked. An outrageous amount of data and observations can blur one’s ability to make informed and rational decisions. Too much information at once is impossible to process, and can delay decision-making. Consequentially, the public sector must not only provide data – an  open-by-default rule on non-sensitive data is becoming the norm in many democracies –but must also provide ways to digest it. Open data portal and in-house sharing platforms such as GCTools Canada or Finland’s Place to Experiment are of paramount importance because they allow for tailored access to information. While data must be out there, its use must be structured, logical, personalisable hence personalised, and digestible. Information has no value when it overwhelms, and only becomes a powerful tool when it starts to make sense.

What long-run results can we expect from a reduction in information asymmetry on the one hand, and the structuring of multilevel data and observations on the other? Two outcomes can reasonably be hoped for:

  1. Non-coordination patterns would end for good, and innovation would sprout to holistic levels;
  2. Public institutions would foster the necessary dynamics to create their own “conditional co-operators”. In addition to innovative projects, an innovation narrative would also be established across the public sector. What would result is true system change, and a systematically-flexible administration. The beginning of a new era for the public sphere.

Théo Bourgery is a second-year MPA student at LSE, originally from France. He received an Honours BA in Sociology and Economics from McGill University. Prior to LSE, Théo was a political appointee to a French MP, in charge of parliamentary affairs. He is now interning at the OECD’s Observatory of Public Sector Innovation (OPSI) with a focus on data management and blockchain technology. Théo is the lead editor of the Public Sphere’s print journal

Leave a Reply

Your email address will not be published. Required fields are marked *