Problems and Purpose

Singapore is currently facing demographic shifts. In 1970, non-residents comprised 2.9% of the total population [1]; by June 2025 this had grown to about 31% (1.91 million of 6.11 million residents).[2] At the same time, the resident total fertility rate fell to 0.87 in 2025, and far below the replacement rate of 2.1.[3] Studying these shifts, IPS’ social capital research shows that the average number of close friends among Singaporeans declined from 10.67 in 2018 to 6.49 in 2024, with growing preferences for interaction within one’s socioeconomic strata and a small decline in cross-racial close friendships.[4] [5] A 2025 IPS “Faultlines” study found that if immigration were mismanaged, 37.1% of respondents expected a fall in trust in government, 37.5% expected anger against particular communities, and 35.5% expected decreased national identity or sense of belonging.[6]

While standard survey methods used in the above studies can help to reveal patterns, they cannot show how residents might reason together across citizenship and residency lines when given time, balanced information, and facilitation. The IPS–REACH Consensus Conference was specifically designed to address this gap by asking whether a structured mini-public could enable residents to:

(a) articulate where they can and cannot “live with” one other’s positions on local–foreign issues; and

(b) build bridging social capital, which are cross-cutting ties that connect people across social divides. [7] [8] [9]

The research team also wanted to test whether such deliberation could catalyse collaborative action post-process, rather than conclude with a report alone. [9] [10]

Participant Recruitment and Selection

Participant selection followed principles of deliberative representativeness rather than statistical sampling. [10] [11] Recruitment materials with online sign-up links were disseminated publicly via social media, messaging apps, existing organizational networks, as well as through flyer distribution in the locality where the Consensus Conference was conducted. Interested residents completed a screening questionnaire to provide demographic information.

The final sample of 24 residents fulfilled the quota sampling to ensure meaningful representation across residency statuses (i.e., citizens, permanent residents, and non residents), age groups, genders, and ethnic backgrounds.

Attendance at all full-day sessions was strong with all 24 participants attending Sessions 1, 2, and 4, and 23 participants attending Session 3 (where the last participant left mid-session due to illness).

Methods and Tools Used

Adapting the Consensus Conference Model

The process drew on the Danish consensus conference model. In addition, it introduced several innovations tailored to Singapore’s context and for the goal of building bridging social capital. [12] [13]

Core design features included:

• Multiple sessions over time: There were four in-person, full-day sessions, preceded by a briefing and pre-deliberation survey and followed by post-deliberation surveys.

• Expert input and Q&A: There was an introductory briefing by a subject matter expert where participants were presented with key data on immigration policy, labour markets, education, and public sentiment, with extended Q&A that allowed participants to interrogate statistics and correct misperceptions.

• Professional facilitation: Seven experienced facilitators, each with at least five years’ experience and more than 10 sessions, were trained to uphold inclusive dialogue, manage power dynamics, and treat emotions and personal stories as legitimate deliberative inputs. [14]

• Participant-generated statements: Rather than voting on pre-written propositions, the participants drafted, revised, and combined statements in their own words, which then became the objects of consensus testing.

What Went On: Process, Interaction, and Participation

Deliberative Process

Deliberation followed a structured five-step process for each topic.

1. Initial positions and reasons: In small groups, participants stated their positions on an agreement scale and explained their reasons.

2. Bidirectional movement and condition mapping: Participants were asked to take a “half-step” toward both stronger disagreement and stronger agreement, and to specify what conditions would move them to these positions. This surfaced conditional preferences that are often implicit in ordinary discussion.

3. Statement drafting: Participants then drafted statements they considered fair, reasoned, and potentially supportable by the group, drawing on the reasons, conditions, and perspectives surfaced in steps 1 and 2.

4. Small-group validation and revision: The participants sought consensus from one another through round-robin questions, inviting revisions when participants could not reach consensus and recording statements as “no go” if final unanimity was not possible or if original proposers could not accept the changes.

5. Plenary validation: Statements that reached unanimous small-group consensus were brought to the large group plenary, where consensus was sought again using a movement-based process. This protocol combined cognitive perspective taking with embodied movement to make agreement and dissent visible and to normalise shifts in view.

Three-Tier Consensus Architecture and Safeguards

A key methodological innovation was a three-tier architecture for classifying outcomes, coupled with safeguards against false consensus.

• Common Ground: Statements that 100% of participants said they could “live with”.

• Near Consensus: Statements with at least 85% “can live with” responses.

• No Go: Statements with less than 85% support.

“No go” statements were not softened into vague compromising language. Instead, they were documented explicitly in the Residents’ Report as boundaries where expectations about identity, fairness, or responsibility remained too contested to be treated as shared. To guard against false consensus, the facilitators revisited earlier expressed views when the participants appeared to have shifted suddenly to reassure and normalise dissent. Post-session evaluations on the session process were also administered to participants to check whether anyone felt pressured to conform so that adjustments to facilitation can immediately be made the following session.

Measuring Social Capital and Attitudinal Change

To assess whether deliberation changed social capital, the research team administered pre- and post-deliberation surveys measuring people’s willingness to consider other viewpoints, intellectual humility, comfort interacting with people of different backgrounds, trust in government, and perceived civic efficacy. The research team also coded transcripts and observations to track how the participants challenged stereotypes, revised their own views, and responded to others’ experiences. The close to 50-page Residents’ Report, co-authored by the participants themselves, provided additional textual data on how the participants chose to frame community norms, fairness, and identity.

Linking Deliberation to Collaborative Action

The organisers also sought to test whether mini-publics could catalyse collaborative action. During Session 3, the participants used Open Space Technology to brainstorm community initiatives. [15] This led to the proposal of “Triad Trails / Food Heritage Trails”, a participant-driven heritage walk to bring local and foreign residents together around food and history. Among the participants, a cross-residency working group of seven citizens and three foreigners was formed to carry the project forward.

Deliberative Quality

Process evaluations showed high satisfaction, with 95.8% reporting a positive experience, 91.6% finding it meaningful, and 87.5% feeling empowered. All participants agreed that the facilitators recorded views respectfully even when disagreement occurred.

Influence, Outcomes, and Effects

Attitudinal Shifts

Pre- and post-deliberation surveys documented meaningful changes. Singapore citizens showed the largest gains in comfort engaging people whose backgrounds or perspectives differed from their own, while non-resident participants registered the largest reductions in certainty that their own views on local–foreign integration were fully correct, signalling increased intellectual humility. Agreement that both locals and foreigners should make an effort to know each other rose from 83.3% to 95.8%, and participants reported greater willingness to participate in future engagement exercises. Trust in government responsiveness increased with respondents, indicating more likelihood to believe that citizens have a say in what government does and that agencies seriously consider public input.

Together, these shifts suggest gains in both bridging and linking social capital. [7] [16]

Analysis and Lessons Learned

This case offers several lessons for deliberative practice and democratic innovation in diverse societies.

First, consensus proved to be domain-dependent. Aspirational norms of mutual respect and reciprocal effort in everyday life achieved high consensus, while distributive questions about jobs and education could be handled through conditional formulations. However, identity-laden questions about multiculturalism and belonging proved more resistant to deliberative resolution. [9] [17] [18]

Second, conditional formulations, especially such qualifiers as “when qualifications are equal”, functioned as a consensus mechanism. These formulations allowed participants to affirm both citizen priority and meritocracy, and citizens themselves often insisted on such qualifiers, reflecting an internalised commitment to meritocratic ideals.

Third, the case demonstrated that emotions and narratives can enrich, rather than derail, deliberation. Treating personal stories and emotional responses as legitimate inputs brought the recognition of anxieties into the open and helped participants understand one another’s stakes, consistent with critical approaches to deliberation. [14] [17] The bidirectional movement protocol, strict “can live with” threshold, explicit protection of “no go” zones, and participant-authored report together constitute a distinctive adaptation of the consensus conference model to an Asian, policy-linked setting. [9] [12]

Notwithstanding the innovations to consensus conference, this study recognises its limitations, among which are that the 24 participants were self-selected and skewed older and more educated than the general population;, the pilot was conducted in a single constituency; and the voices of jobless or struggling citizens, work permit holders, and domestic workers were largely absent. These groups likely hold stronger views and face different constraints. [9] [11]

Even with these constraints, the pilot shows that an adapted consensus conference is a viable mechanism for building and studying bridging social capital in a diverse society. It demonstrates that carefully structured deliberation can shift participants’ positions, sustain respectful dialogue across residency lines, and generate a replicable framework that links discussion to collaborative action. Future research could prioritise recruiting more under represented groups, extending the model across multiple constituencies, and testing whether the gains observed here hold under more varied conditions.

See Also

• Research Report: Building Bridges Across Differences: Piloting a Consensus Conference on Local-Foreign Integration in Singapore https://lkyspp.nus.edu.sg/docs/default-source/ips/project-report_ips-reach-building-bridges-across-differences_piloting-a-consensus-conference-on-localforeign-integration-in-singapore.pdf

• Residents’ Report: Consensus Conference on Local-Foreign Integration in Singapore https://lkyspp.nus.edu.sg/docs/default-source/ips/residents'-report_consensus-conference.pdf

• Video: Consensus Conference on Local-Foreign Integration in Singapore https://www.youtube.com/watch?v=U_MptqiZ3nI&t=50s

References

[1] Pang, A. (ed.) (2010) Phases of Singapore’s Demographic Development Post World War II. ETHOS Issue 07. https://knowledge.csc.gov.sg/ethos-issue-07/phases-of-singapores-demographic-development-post-world-war-ii/

[2] National Population and Talent Division, Strategy Group, & Ministry of Manpower. (2025). Population in brief 2025. Government of Singapore.

[3] Singapore Department of Statistics. (2026). Births and fertility. https://www.singstat.gov.sg/find-data/explore-data-themes/population/births-and-fertility/latest-news-data

[4] Mathew, M., Teo, K. K., & Tay, M. (2025). Friendships in flux: Generational and socio economic divides in Singapore (IPS Working Paper No. 62). Institute of Policy Studies.

[5] Mathew, M., Teo, K. K., Poh, R., & Tay, M. (2025). Results from the IPS OnePeople.sg indicators of racial and religious harmony 2024 (IPS Working Paper No. 59). Institute of Policy Studies.

[6] Mathew, M., Teo, K. K., Tay, M., & Poh, R. (2025). Faultlines in Singapore: Perceptions and management with a focus on race and religion (IPS Working Paper No. 60). Institute of Policy Studies.

[7] Putnam, R. D. (2000). Bowling alone: The collapse and revival of American community. Simon & Schuster.

[8] Putnam, R. D. (2007). E pluribus unum: Diversity and community in the twenty first century. Scandinavian Political Studies, 30(2), 137–174.

[9] Thomas, N., Lee, J., Goh, W., Najib, A., & Wong, K. L. (2026). Building bridges across differences: Piloting a consensus conference on local–foreign integration in Singapore (Project Report). Institute of Policy Studies.

[10] Fishkin, J. S. (2009). When the people speak: Deliberative democracy and public consultation. Oxford University Press.

[11] Jacquet, V. (2017). Explaining non participation in deliberative mini publics. European Journal of Political Research, 56(3), 640–659.

[12] Grundahl, J. (1995). The Danish consensus conference model. In S. Joss & J. Durant (Eds.), Public participation in science: The role of consensus conferences in Europe (pp. 31–40). Science Museum.

[13] Joss, S., & Durant, J. (Eds.). (1995). Public participation in science: The role of consensus conferences in Europe. Science Museum.

[14] Young, I. M. (2000). Inclusion and democracy. Oxford University Press.

[15] Owen, H. (2008). Open Space Technology: A User's Guide (3rd ed.). Berrett-Koehler Publishers.

[16] Szreter, S., & Woolcock, M. (2004). Health by association? Social capital, social theory, and the political economy of public health. International Journal of Epidemiology, 33(4), 650–667.

[17] Dryzek, J. S. (2005). Deliberative democracy in divided societies: Alternatives to agonism and analgesia. Political Theory, 33(2), 218–242.

[18] Gutmann, A., & Thompson, D. (1996). Democracy and disagreement. Harvard University Press.

Contributors

Policy Lab (Institute of Policy Studies): [email protected]

Nicholas Thomas: [email protected]

Justin Lee: [email protected]

Wilson Goh: [email protected]