Sect Influence: A Religious Group With Measurable Electoral Power

What We Found

• In municipalities where the Family Federation for World Peace and Unification (the former Unification Church) was active, a ruling-party candidate closely aligned with Shinzo Abe—Japan’s longest-serving prime minister—saw much stronger vote growth than in otherwise similar municipalities without a church presence.
• The estimated impact was about 17,000 votes, roughly one-fifth of the candidate’s previous total—enough to turn a likely loss into a win—and likely a minimum estimate.
• For politicians, the calculation was straightforward: despite the Church being widely criticised for aggressive fundraising practices, its ability to mobilize votes made the relationship politically valuable.

Why It Matters

For years, observers could only guess how many votes a religious group without its own candidates could deliver—or why politicians would court a group facing strong public backlash. This case offers rare, concrete numbers through a natural comparison: the same candidate, under the same electoral rules, running in two different elections as ties to the Church changed.

The findings feed into debates about campaign ethics, political mobilization, and where to draw the line between religious groups and ruling-party politicians—an issue that came under intense scrutiny after the 2022 assassination of former prime minister Abe by a man who said his family was ruined by the Church.

Visual Highlight

Vote share climbed much more in municipalities with a church facility than in carefully matched comparison municipalities.

How We Did It

Design

We compared municipalities that were very similar—in population, income, economic conditions, and more than 100 other factors—with only one major difference: whether they had a Church facility. This near “apples-to-apples” comparison allowed us to isolate the added votes linked to the Church’s presence.

Data & Methods

• Election results from the Ministry of Internal Affairs and Communications.
• Church facility locations downloaded from the group’s public website.
• Background indicators included demographics, urban–rural status, income levels, land area, and other municipal characteristics.
• Multiple matching approaches to pair similar municipalities and check that the estimated effect did not depend on a single method.

The size of the effect remained stable across different reasonable model choices.

My Role

• Led the data analysis for the project, designing the matching strategy used to estimate the electoral impact of the religious organization.
• Produced the causal effect estimates and built the full R workflow for the analysis.
• Reported and wrote the full newspaper article explaining the findings for a general audience.
• Collaborated with academic researchers to refine the study design and interpretation.
• Served as first author of the academic paper based on the investigation, which formalized the methodology and results.

A case where causal methods turn a politically sensitive question into publishable evidence.

Impact & Publication

• Initial results ran as a news investigation in The Asahi Shimbun.
• The work was later expanded into a peer-reviewed academic paper in Electoral Studies (選挙研究). Online publication will follow due to the publisher’s schedule; a PDF is available upon request.
• Findings led to an invitation to speak at a lawyers’ association supporting victims of the former Unification Church.