Better Reykjavik and My Neighborhood: a city’s standing channel for resident ideas

Better Reykjavik (Betri Reykjavík) is an online idea and debate platform, built by the nonprofit Citizens Foundation, through which Reykjavik residents propose and rank ideas, with the city committed since October 2011 to sending the top-ranked ideas each month to its committees for formal consideration.⁠¹ Its companion, My Neighborhood (Hverfið mitt), is a participatory budget in which residents aged 15 and older vote online on small neighborhood projects.⁠^2,3 The case is one of the longest-running links between online participation and city government, and its record includes both durable institutional adoption and persistent problems of low turnout, uneven participation and slow follow-through.

What happened

Better Reykjavik began outside government during a political crisis. Derek Lackaff’s peer-reviewed study in JeDEM (2015) describes how Iceland’s 2008 banking collapse produced a deep loss of institutional trust; citing annual Capacent Gallup surveys, he reports that trust in the Reykjavik City Council stood at 9% in 2008 and 22% in 2010.⁠¹ The OECD Observatory of Public Sector Innovation records that the platform was launched by the Citizens Foundation in May 2010 and “opened a week before elections in Reykjavik”.⁠⁴

The May 2010 municipal election changed the platform’s status. On May 29, 2010, the Best Party, a new party led by the actor and comedian Jón Gnarr, defeated the incumbent Independence Party and entered coalition talks with the Social Democrats; Lackaff reports that it used a website to solicit citizen opinion for the coalition platform.⁠¹ The new city government then formed a steering committee, chaired by the city’s web director, that spent about a year designing a process that required no legal change.⁠¹ Technical management stayed with the platform’s developers, Bjarnason and Grímsson, who founded the Citizens Foundation in December 2010, and a city project manager was assigned to collect the top ideas and route them to committees.⁠¹ In October 2011 the city signed a contract with the Citizens Foundation to provide the platform.⁠¹

The participatory budget grew alongside it. The OECD case entry states that the initiative began in 2011 as Better Neighborhoods and was later renamed My Neighborhood.⁠⁴ The city reports that the first vote took place in 2012.⁠⁵

The platform outlived the party that adopted it. In May 2014 Gnarr did not seek a second term and the Best Party was dissolved; a coalition led by the Social Democrat Dagur B. Eggertsson took office, created a standing Administration and Democracy Council, and placed Better Reykjavik under it.⁠¹ Lackaff describes this as further formalization of the project within the city’s structure.⁠¹

My Neighborhood has continued on a two-year cycle. The 2023 vote ran from September 14 to September 28, 2023.⁠³

How it worked

Better Reykjavik and My Neighborhood are two different mechanisms that share a platform, and they hand residents very different kinds of power.

Better Reykjavik: agenda-setting without a signature threshold

Proposals and debate. Registered users post ideas, vote for or against them, and add structured “points” for or against each idea; points are ranked by other users’ answers to the question “Is this helpful?”.⁠¹ The design separates arguments for and arguments against into two columns rather than a single comment thread.⁠¹

The monthly rule. The city’s published rules, quoted by Lackaff as of summer 2015, committed it to submitting the top five ideas by support each month for formal consideration, plus the top idea in each of thirteen categories (tourism, construction, leisure, sports, human rights, culture, education, transportation, planning, governance, environment, welfare and miscellaneous) to the matching advisory board.⁠¹ At noon on the last business day of each month, those ideas are frozen and sent on, and the rules say they should be considered “ideally within a month”.⁠¹ The rules also state plainly that “it is not possible to formally review all submitted ideas”.⁠¹

Design choices meant to build legitimacy. Lackaff reports three choices the city made deliberately: politicians and administrators did not take part in online debate before ideas reached committees; ideas were stripped of their authors’ names before submission, on the reasoning that a crowdsourced idea has no single owner; and there was no minimum vote threshold, so the top ideas move forward every month however few people voted.⁠¹ A comparative study cited by Lackaff (Åström and colleagues, 2013) singled out this last feature as unusual among e-petition systems.⁠¹

Advisory, not binding. Nothing in the rules obliges a committee to adopt an idea. The obligation is to consider it and respond.⁠¹

My Neighborhood: a binding vote on a capped budget

Idea collection and review. The city describes the process as idea collection, review of ideas, electronic voting and implementation.⁠² According to the OECD entry, ideas are gathered over about three weeks, after which a project team and political district committees decide which are reasonable and implementable.⁠⁴

The ballot. The city’s current rules put 25 ideas on the ballot in each neighborhood: the 15 most popular ideas advance automatically and the neighborhood council chooses 10 more.⁠² Voting is open to everyone who turns 15 in the year of the vote and is legally resident in Reykjavik.⁠²

The money. The city states that funding is 850 million krónur per cycle, divided between districts,⁠⁶ with each neighborhood receiving a fixed equal amount plus an additional amount according to its population.⁠² The OECD entry describes the budget as about 6% of the city’s capital investment budget, allocated across ten neighborhoods.⁠⁴ After the vote, winning ideas go through design, tendering and local consultation before construction.⁠²

Legal basis and oversight. The city’s Internal Audit and Advisory office states that My Neighborhood rests on Chapter X of the Local Government Act on consultation with residents, which allows residents’ influence to be secured through, among other things, resident votes.⁠³ Under the rules for the 2022 to 2023 round, Internal Audit supervises the conduct of the electronic vote and checks that documented procedures are followed.⁠³

Measured results

Reykjavik’s results come mostly from the city and the platform’s operator, so most figures below are self-reported; the 2023 audit is the main independent check within the city administration.

Better Reykjavik

• Lackaff, citing the platform’s own “Top Ideas” figures, reports that by July 2015 over 630 proposals had received formal consideration and 501 had been implemented or were being implemented since 2010.⁠¹
• From the site’s analytics, Lackaff reports that between October 2011 and September 2014 there were 11,390 registered users, 1,959 ideas, 3,339 points for and against and 2,706 comments, with more than 500 ideas in the first five weeks after the official launch, and “a modest amount of activity per week, with a few to a few dozen ideas posted” outside campaign spikes.⁠¹
• The Citizens Foundation stated in April 2014 that 70,000 of 120,000 inhabitants (58%) had used Better Reykjavik.⁠⁷ That figure counts any use, including viewing, and is the operator’s own claim. The OECD entry, also based on submitted material, lists 27,000 registered users, more than 8,900 ideas and close to 700 ideas realized.⁠⁴

The gap between “70,000 people used it” and “11,390 registered to participate” over a comparable period.⁠^1,7 is itself a finding: reach and active participation are different measures.

My Neighborhood

The city reports that 696 projects had been approved through the program between the first vote in 2012 and the 2018 round⁠⁵ In its review of the 2023 vote, Internal Audit confirmed that documented procedures had been followed in the main, judged that the deviations it identified did not affect the security or outcome of the vote, and confirmed that the published results matched the vote count.⁠³

The denominators matter. Even in the best year shown, more than 83% of eligible residents did not vote.⁠³ The existing overview at comparative cases discusses the 2018 round; the point here is the trend, which rose to 2021 and fell back in 2023.⁠³

Failures and criticisms

Better Reykjavik and My Neighborhood have critics inside and outside city government, and some of the sharpest criticism comes from officials who supported them.

Vague answers from the administration. The city’s web director told Lackaff that in the early period “the administration tended to answer very vaguely, unclear answers, answers like ‘we are going to look into this,’” when the aim was a clear yes or no; he described later improvement as “not a revolution, but it’s a step forward”.⁠¹ The OECD entry lists as a challenge that “administrative systems are resistant to change, even with good support from its officials”.⁠⁴

Ideas without champions. Lackaff reports that the “ideas committee” of representatives from standing committees stopped meeting “because people stopped coming to the meetings,” after which the project manager sent top ideas to committees by email; the manager said administrators “sometimes think the ideas need a representative or speaker in some way”.⁠¹

No threshold means weak ideas advance. Eggertsson, then mayor, argued that the absence of a minimum vote threshold lets “weaker” ideas move through, and that requiring people to “rally a bit to get their ideas on the table” would strengthen legitimacy and deliberation.⁠¹ He also worried that keeping officials out of the online discussion turned it into “just a voting system”.⁠¹ These are criticisms of the very design choices the city made to protect legitimacy, which shows a real tradeoff rather than a simple error.

Amplifying the already active. Eggertsson warned that the platform risked being connected “both in practice and in the public mind” with the “active” part of society.⁠¹ The turnout figures support the concern: a participatory budget decided by 12% to 16% of eligible residents.⁠^5,3 is decided by a self-selected minority, and the sources consulted do not show whether that minority resembles the city by age, income or district.

Language as a barrier. Lackaff notes that almost all user content is in Icelandic, which deters outside trolling but may exclude immigrants and refugees; Eggertsson said the weakness is that “you don’t get the marginalized groups” into the tool⁠¹

Partisan association. City staff told Lackaff that a project so closely tied to particular coalitions risked being replaced by a future government, and the project manager said opposition parties that had criticized it might “have a hard time continuing with it, as is”.⁠¹

Getting people to take part. The OECD entry’s own summary is that “the biggest challenge is in getting citizens to participate”.⁠⁴ A 2016 preprint by Bojic, Marra and Naydenova, based on interviews with Reykjavik officials, argued that budgets and staff for online engagement should be proportionate to the effort needed to act on what residents propose, so as not to “undermine their beliefs about their power to influence decisions”.⁠⁸ That paper is a preprint and was not peer reviewed as far as the source shows.

The case for the defense. Supporters point out that the monthly commitment has been honored across multiple administrations,⁠¹ that hundreds of ideas were reported implemented,⁠^1,4 and that the participatory budget vote is independently supervised and its results checked against the count.⁠³ A low-turnout process with real outcomes, they argue, may still be better than a consultation with none.

What transferred and what did not

Better Reykjavik’s software traveled more easily than its institutional commitments. The Citizens Foundation describes its open-source platform as Your Priorities,⁠⁷ and the OECD entry, drawing on the foundation’s submission, says the tools have been used by “almost 2 million citizens in over 20 countries”.⁠⁴ That is a self-reported reach figure, not a measure of policy impact.

What did not obviously transfer is the monthly obligation to consider top ideas without a threshold. Lackaff attributes Better Reykjavik’s survival to its escape from what Mahrer and Krimmer called the “middleman paradox”: the problem that the elected officials who would have to introduce citizen participation are often the ones who resist it.⁠¹ In Reykjavik, a crisis, a new party with a mandate for experimentation and a fast-moving grassroots tool lined up at the same moment.⁠¹ Those conditions are hard to copy. For contrast, see the Madrid and Paris cases, and the origins of participatory budgeting in Porto Alegre.

Lessons for My Vote Forward

These points are interpretation and design proposals, not results.

• Separate setting the agenda from deciding. Better Reykjavik ranks ideas for consideration; My Neighborhood decides a capped budget. My Vote Forward’s distinction between public priorities, voluntary opinions and private voting follows a similar logic: different outputs carry different authority.
• Promise a response, and define what a response is. The early “we are going to look into this” problem;⁠¹ suggests that any institution receiving published priorities should be expected to answer with a clear status. See proposals for status changes.
• Thresholds are a design choice with costs both ways. No threshold keeps ideas flowing even when few people take part; a threshold may raise quality but shut out small communities. The choice should be written down and explained; see governance.
• Report reach and participation separately, with denominators. The difference between “used” and “registered to participate”⁠^1,7 is the kind of gap a published release should never blur. See measurement.
• Independent checking of the count builds trust. Reykjavik’s Internal Audit review of the 2023 vote⁠³ is a model of a modest, concrete check; see audits and verification.
• Language access is not optional. A tool that works in one language will under-represent people who do not use it.

Unresolved questions

• Who votes in My Neighborhood by age, district, income and language background? The sources consulted give totals but no demographic breakdown.
• What share of ideas sent to committees were accepted, rejected or left without a clear answer in recent years? The figures found stop around 2015.⁠¹ or are undated⁠⁴
• Why did My Neighborhood turnout fall from 16.37% in 2021 to 12% in 2023.⁠³? No explanation was found.
• What happened to My Neighborhood after the 2022 to 2023 cycle? The city’s page found for this review describes the 2023 vote as completed and lists no newer round⁠²
• Did Better Reykjavik change how residents trust city government? Trust rose from 2012 to 2014,⁠¹ but the sources do not separate the platform’s effect from economic recovery and other causes.