Four scenarios for the future of #opengov


(cross posted from Joinup)

In the context of the study on the new generation of eGovernment services , we have produced a definition and taxonomy of “Open Government Services”, gathered 180 initiatives and carried out a cost-benefit analysis of 10 cases (not yet public). We have also carried out a survey of stakeholders (ongoing) and will host a workshop on May 31st.

This post spells out 4 initial scenarios to be discussed at the workshop, based on 4 levels of ambition, from the most to the least ambitious:

  • open, participatory policy-making
  • collaborative delivery of human services
  • automated, federated delivery of administrative services
  • failure of open government and back to traditional e-government

In each scenario, participants will be asked to enrich it with more details; identify its drivers and bottlenecks; suggest action to make it happen (or avoid it to happen).

Scenario 1: open policy making

The main policy decisions are taken with the fundamental input of citizens. Online discussions become the norm, and part of the policy cycle.

Any decision which is deemed “of public interest” is published for open consultation of at least one month, both in the executive and legislative branch. These open consultations allow anyone to present comments and comment and vote on the proposals of others.

On average, each consultation involves 1000 participants, but some of them reach 100.000 and the majority of them has less than 500 people. Many citizens who were never involved in politics take part in these decision, bringing their specific knowledge. Additional live and email contact are carried out for specific segments of the population under-represented in the consultation.

The quality of the input is high, and many proposals by citizens are directly included in the final decisions, and citizens also help identifying the top quality proposals.

The huge amount of qualitative and quantitative data is analysed by governments using advanced text analytics software.

The government reports back to citizens about what has been done and how their comments have been used, leading to a virtuous learning cycle.

During the implementation of the policy, open data are published about its implementation and citizens can comment and add proposals to improve the quality of implementation.

Citizens also provide open input in the final evaluation of the implemented policy, which is published online for public commenting.

Government partially reduce their spending in scientific support to policy, thanks to the open intelligence brought by citizens, and in receiving lobbyists. Lobbying mostly happen through the platform rather than in dedicated meetings.

As a result, citizens trust government more, are more willing to pay taxes and less likely to vote for populist parties. Public policies are more effective as all stakeholders feel ownership and collaborate to its success.

CASES: Tartu Participatory Budgeting, Parlement et Citoyens, IoPartecipo.


  • Can you add more details to this scenario?
  • What are the bottlenecks to get there?
  • What are the main policy actions to overcome these bottlenecks?

Scenario 2: collaborative human services

Public services of genuine added value (human services) are systematically designed and implemented with the involvement of citizens and business.

Any function of government providing services to users has to run a systematic “crowdsourcing test” to assess how citizens can contribute by brining specific knowledge and their experience as users, at least for some services considered as “core service”. Any service not being co-produced should duly justify the reasons.

By default, these “core services”:

  • are co-designed with users (citizens and business) and are only developed by government where there are no existing services run by users.
  • provides open data and API for users to build added-value services and integrate with existing services.
  • provide ways for users to collaborate and support each other in the delivery of the service, leveraging their competences.
  • enable open feedback by users on the quality of the service.

Citizens are used to provide specific input and feedback in the delivery of these services. The majority of citizens provides feedback on the status and quality of the services, report problems and provide input for improvement. The feedback is constructive, even when critical. They also perceive are their public duty to regularly support other fellow citizens and business using the same service.

Civil servant continue their work by monitoring citizens input and ensuring that it is well balanced, and acting in issues not covered enough (such as specific services or some specific areas of the city). Rather than controlling the status of the street, they monitor citizens input on it and ensure that all streets are adequately monitored.

Public spending remains on similar level, but the quality of services is much higher and citizens satisfaction is high. There are less mistakes being done and money wasted in delivering services, and trust in public  service delivery increases.

CASES: PatientOpinion. FixMyStreet, Torrelodones.


  • Can you add more details to this scenario?
  • What are the bottlenecks to get there?
  • What are the main policy actions to overcome these bottlenecks?

Scenario 3: federated administrative services

All the core administrative services are tightly integrated across government, and provided through composable modules that are re-used and integrated automatically. Any such service provide API access for additional integration without any additional software development.

Services are provided automatically and proactively to citizens, and the once only data provision principle is enforced throughout all levels of government and also on cross-border services.

Citizens and business have predictable times for the delivery of their administrative documents, and can monitor their progress online.

Clear rules about access to data are provided, and users can see which data are owned by which agencies, and grant permission for access.

Administrative services are open for integration with services provided by business. For instance, government electronic ID is used by banks, and social media authentication is used by government for some of their activities (e.g. online discussions). Government provide open data and open API systematically for third parties.

Citizens carry out their transaction largely online, in an automated way directly or through dedicated commercial services. Online is the default way to interact with government, but

They are not particularly engaged in public issues, but they can monitor any decision taken by government through specific dashboards produced by newspapers and NGOs.

Government spending is significantly reduced because of savings in service delivery and greatly reduced rate of mistakes. Moreover, government spend less in developing customised software, but reuse software built by other government agencies and “off-the-shelf” apps by commercial players.

Companies and business save time and money thanks to automated, proactive services.

There is a thriving market of business built online services based on and integrated with government-built software components.

CASES: Di@vgeia, NemID, Interoperable data gathering for e-social security


  • Can you add more details to this scenario?
  • What are the bottlenecks to get there?
  • What are the main policy actions to overcome these bottlenecks?


Scenario 4: the end of open government

20 years after the Obama memo on open government, it becomes clear that transparency, collaboration and participation are not delivering on their promises.

The open data portals have been closed because of lack of usage. The promised large scale economic gains from reusing open data never materialized.

Citizens are not interested in monitoring themselves how government works, and much less in taking part in service delivery and public policy.

Building composable services turned out to be immensely time-consuming and difficult to orchestrate across governments. Many service failures took place and it proved impossible to understand ownership and ultimate responsibility for the quality of service.

Citizens and business simply want government to do the basic service delivery, limiting the costs to the minimum.

Public policies are designed top-down, in a technocratic way, based on the available scientific evidence. Human services are delivered by expert civil servants or outsourced to the private sector. Administrative services are delivered by large, centralised organisational units, supported by software built on demand by large IT corporations.

Government costs and benefits remain stable, but at least there are no major disasters happening. Public Sector Innovation disappears from the policy agenda.


  • Can you add more details to this scenario?
  • What are the drivers that generate it?
  • What are the main policy actions to avoid these drivers?

Next steps

These scenarios will be discussed online and at the workshop. Based on the stakeholders input, initial policy recommendations will be developed.

They will be published online for public commenting.

Based on the feedback received, a final version will then be developed.


Four scenarios for the future of #opengov

What has #opengov achieved so far? An initial checklist

cross-posted from Joinup

When thinking about future scenarios for the forthcoming workshop, it can be useful to look back at what we promised, as open government advocates, and what was actually achieved. What worked, and what didn’t? Using an Hegelian metaphor: what is alive and what is dead of open government?

I refer mostly to what I see among my clients, that is EU, national and regional organisations. Hence, I don’t refer to the majority of organisations, but probably to the most advanced ones.


  • ACHIEVED: The nature of policy discussions has certainly  become more open in the government domain. Civil servants are now frequently participating in twitter, facebook and Linkedin. Policy workshops frequently integrate dynamic interaction methods, such as World Cafe methodologies, rather than traditional powerpoint followed up by scant discussions. The manifest goal of many policy discussions launched by government is to reach out beyond the “usual suspects”: the novelty of the people involved is becoming one of the quality criteria of online engagement.

NOT ACHIEVED: The link between open engagement and policy remains unresolved. The final loop of participation, the modification of the policy and the feedback to the participants, remain an exception rather than the rule. As a good practice, consider the charter of “Parlement et citoyens”, which engages MPs to provide a video feedback to citizens about what they did.


  • ACHIEVED: Open data have become the default option, not only because of the revised directive, but most importantly in the mindset and expectations. The trend continues, with for instance research data funded through H2020 being public by default by 2017. The main benefit remains in term of transparency and accountability: for instance, recently Open Corporates was instrumental in the resignation of a Spanish Minister.

NOT ACHIEVED: uptake of open data remains minor. A minority of citizens download datasets; and the economic benefits by startups reusing public data have most probably being overestimated.


  • ACHIEVED: a culture of co-design of public services and public sector innovation is becoming widespread. Many government agencies have created a “Lab” designed to promote innovative methodologies, and there are many specialised provider in service design, agile methods, innovation without permission. The benefits of co-design public services appear clear, also in economic terms.

NOT ACHIEVED: co-design of public policy is much more challenging, as Beth Noveck show in her book. The questions are more specialised, require greater engagement and it is difficult to gain meaningful insight. Collaboration in policy design remains more a goal in itself than an actual effective way to design better public policies.

NOT ACHIEVED: Integration of offline and online collaboration. Successful co-creation requires live interaction. Online tools are important before and after (through open discussion and open feedback), but actual concrete co-creation requires live collaboration. In this sense, we could say that pure online collaboration is no longer a viable option: online and offline have to be considered as integrated tools, but too, in  practice, they still aren´t.


  • ACHIEVED: metrics for participation and collaboration are becoming more and more common. Most online tools show the numbers openly in the homepage (e.g. Uptake is now, finally, considered a key performance indicator, while this was not the case at the early stage of e-government and e-participation.

NOT ACHIEVED: The focus is more on the quantity of participation than in the quality of the contribution. The success of online engagement is too often measured only in terms of “number of tweets”.

What are your views? What do you consider as the main achievement and shortcomings of open government so far?

What has #opengov achieved so far? An initial checklist

Who hijacked my liberal agenda?

I am, in many ways, a liberal.

In particular, I favour free trade. I think that, by default, free trade agreement are good. And my argument falls flatly with neo Ricardian approaches, such as the classical Krugman. Free trade has, historically, always be supported by multinationals, but this is not enough to make it wrong. We know that for all the past free trade agreements.

But I have a problem:

  • In today’s debate, free trade is the establishment; it’s defending the interest of big multinationals which are holding the real power and manoeuvring politicians and governments. In today’s debate, Bernie Sanders is the honest guy, Hillary Clinton is the establishment. TTIP is the big companies agenda, anti-TTIP are honest. There is no room for a genuine, pro free market argument. When I hear Bernie Sanders arguments, that free trade takes away american jobs, I think that’s not too far away from Donald Trump. That is incorrect, it’s manipulative, it’s populist.
  • The “new Labour” wave of politicians (Clinton, Blair, Schroeder…) have shown low moral standards and do not have the credibility to defend genuine liberalism.
  • The content of TTIP is being overshadowed by its secrecy and the ISDS.

My plea is this. There are genuine liberals who believe in free trade: not because they are spokespeople of multinationals; not out of cynical realism about “that’s the way the world goes”; but because free trade is the best antidote against war, and because protectionism is worse.

We desperately need credible, charismatic, authoritative left wing free trade advocates, that do not eat into the hands of multinationals, that don’t use the revolving door, that don’t have questionable moral habits.

Who hijacked my liberal agenda?

To rank or not to rank, this is the question cc @bettermeasured

I have always been wary of rankings in policy analysis. Benchmarking and histograms seem to over-simplify reality, trivialize discussions and encourage “me-too” policy competition.

This is why our policy dashboards (see here, here and here) do not produce rankings, but visual summaries which emphasize the diversity and richness of information.

So when we decided to elaborate, together with the European Digital Forum (i.e. the Lisbon Council and Nesta), a report presenting the results of our startup manifesto policy tracker, our first version contained few rankings and lots of qualitative information: it was designed as a “policy map”. After several iterations and discussions, we accepted to pivot it towards a “policy scoreboard” with plenty of rankings: which countries do more to support startups?

The methodology, previously discussed on this blog, was finally mutuated from the OECD Going for Growth: a simple percentage of implementation of the recommendations contained in the original Startup Manifesto. Every recommendation has the same weight, even if some are crucial (e.g. legislation on second chance for entrepreneurs) and some are of dubious effectiveness (e.g. having a national champion).

I am pleased to communicate that the report has just been published and presented to Commissioner Oettinger and widely discussed on Twitter.

I have to admit that the online discussions always started by an assessment of how a country has scored. The rankings where the single “point of entry” into the discussions, that then evolved into insightful quali-quantitative reflections.

For all their limitations, rankings were clearly necessary for communication and, most importantly, in order to kickstart meaningful discussions.

In conclusions, rankings are nor good neither bad. They are just an important tool in policy analysis, that has to be used appropriately.



To rank or not to rank, this is the question cc @bettermeasured

#OpenGov future scenarios: triumph or defeat?

Cross-posted from 

It’s a defining moment for open government. more than seven years after Obama came to power, many experts are voicing disappointments over its achievement. While the normative importance of the goal is not under discussion, the concrete achievements seem not to have lived up to the expectations.

At least, that is the tone of the posts such as this one by Alberto Cottica, or the recent book by Beth Noveck quoted in the same post.

Yet on the other hand, there is very little evidence behind such statements. Is it the opinion of some experts or is there tangible evidence? And if so, what is the reason and most importantly, what should government do next? Is it a problem of design, or one of implementation?

Precisely these are the question that our new study for the European Commission, OGS, is trying to answer through desk research, case studies, surveys and interviews. On April 28th, we will present the results in a highly interactive workshop that will aim at co-designing the future scenarios of open government.

Let me take sides in this debate. I think open government has not achieved its promises, and that’s ok. There are much more data available than people are actually using, and that’s ok. Perez taught us that over-investment is a necessary part of going from installation to deployment of a new paradigm.

We should remember the fundamental assumption that openness and transparency are elements of good governance in themselves. They are goals, not means.

Yet, these goals come at a cost, and they must be feasible. It’s a problem of managing expectations and updating plans, in an agile manner. Not everything went according to plan, but we know that you can’t plan this ex ante and you should dynamically adapt during implementation.

What matters now is to maximize learning from experience so that we can update our plans and focus on what to do well, and set the right expectations for the future.

It’s a matter of calculating costs and benefit:  not in absolute terms but in comparison with what is traditionally done by government. We know open government hasn’t work perfectly but is there genuinely an alternative? It’s not like government worked perfectly before and then open gov came to create problems.

In broad terms, my feeling is that open government did good for society at relatively moderate costs. In the future, it should focus on some specific activities that worked – I will write on this in my next post.

Over the next days, I will start launching ideas on what is dead and what is alive for open government.

We want to develop together future scenarios for open government in order to set the scene for the workshop.

The final goal of this study is to provide concrete recommendations for the next eGovernment action plan of the European Commission. Alberto Cottica says its all about embracing complexity, Beth Noveck that the focus should be on identifying the right expertise in citizens.

What is your idea? We need all the help we can get out there!

Feel free to comment here, or to provide your answers to the online survey here.

#OpenGov future scenarios: triumph or defeat?

Is mainstreaming the perfect killing? For policies, I mean

In our work of policy evaluation, many times I’ve come across policy changes.

A specific policy priority, such as open policy, foresight, social innovation becomes important. It has now a dedicated measure, a funding system or an organisational unit.

Then after a while, the funding or the unit disappears. The official explanation is that it achieved its goal and it is now being mainstreamed across all government policies.

But maybe this is just a nice way to kill a policy that didn’t achieve its goals?


Is mainstreaming the perfect killing? For policies, I mean

Qualitative policy indicators cc @bettermeasured

One of our main activities at Open Evidence is to build policy scoreboards, such as this this and this.

I like this visual representation because it is not a ranking and it illustrates the full complexity of the issue. You can see who’s good and bad on a specific issue, but you don’t have a ranking which encourages “me too” behaviour and institutional isomorphism.

Screen Shot 2015-12-01 at 16.18.44.jpg
Screencap of the Startup Manifesto Policy Tracker

However, it is clear that rankings are an effective way to channel the results of an analysis. If you want to be heard, you need to tell a story. And the large dashboard tells many, not one story. I could bore you to death with the danger of a single story, but in my job, it helps to communicate.

So how can we build a qualitative policy indicator, that we could use for the startup manifesto?

The most interesting example so far is the OECD Reform Responsiveness index, included in the Going for Growth analysis (full details in the 2010 edition, page 79) and used in the Lisbon Council Euro Plus Monitor.

But still, it’s a simple percentage of reforms carried out over reform recommended. Not so sophisticated after all and without any weighting.

Do you know other good examples?

Qualitative policy indicators cc @bettermeasured