Skip to main content

The effect of church tax on church membership


In this study, we examine the effect of church tax on the church membership decision using Finnish data. We present both descriptive statistics from an opting-out website and econometric evidence exploiting the panel structure of a large individual-level data set. Our descriptive analysis shows that opting out is concentrated towards the last days of the year, i.e., the last chance to avoid paying church tax for the entire coming year. Our econometric evidence suggests that the average effect of tax incentives for the whole population is both statistically and economically significant. A 1 standard deviation increase in church tax leads to between 0.5 and 1 percentage point decline in the likelihood of church membership. In addition, we find that church membership dropped substantially when a law change made opting out significantly easier. This finding suggests that transaction costs play an important role in the membership decision.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8


  1. See Francis and Katz (2000) for an extensive overview of the study of joining and leaving religion in different disciplines.

  2. In Germany, roughly 67 % of the population of 82.1 million belonged to one or the other of the two state churches: the Protestant Church and the Catholic Church 27.2 million.

  3. In Finland, the church tax is a separate income tax. It is set on the municipality level by a democratically elected parish council and varies between 1.00 and 2.25 % across municipalities. Figure 1 displays a distribution of the church tax rates in 2006. For comparison, the tax is in Germany is 8 or 9 % of the income tax, i.e., a person paying an income tax of 20 % pays a church tax of 1.6 %.

  4. Heino et al. (1997) and Lumijärvi (1998) both find in their survey studies that the high cost is the main reason for people to leave church in Finland, 24 and 28 % of the surveyees, respectively, being of that opinion. In a survey from the UK by Richter and Francis (1998), only 1 % mention the high cost as the main reason for leaving church.

  5. Daniel Hamermesh writes “....religious organizations are public goods [in the United States]; it’s easy to enjoy the services offered (pun intended) without paying your fair share of the costs.”[The New York Times, Freakonomics Blog, October 14, 2008]

  6. Membership to state church is not a requirement for enjoying some of the services provided by church even in a pay-to-play system. Church’s approach tends to be rather inclusive with respect to, e.g., foreign nonmembers, and its charity extends to all people in need. See Appendix C in Supplementary material for a list of services that are exclusively for members only.

  7. Our data set contains 10,501,554 individual/year cells for 1,142,977 individuals over the years 1996–2006.

  8. Iannaccone (1998) discusses extensively the economic models of religiosity.

  9. The revised law of freedom of conviction (6.6.2003/453) was passed on June 6, 2003 and put into action on August 1, 2003.

  10. The law on electronic communication in the activities of public authorities (24.1 13/2003) was passed on January 24, 2003.

  11. The website conducts a voluntary online survey for a random sample of online leavers. In August 2010, roughly 2,446 individuals were shown the survey and 612 responses were received. Twenty-eight percent of the responses included the words tax or money.

  12. The tax proposal for the previous year’s taxes arrives by mail traditionally during the first weeks in April.

  13. The corresponding Figs. 1, 2, and 3 for years 2006 and 2008 are available from the authors.

  14. Of the 1,142,977 individuals in the data, 48,013 changed their membership status once, 4,883 did so twice, and 502 did so more than twice between 1996 and 2006.

  15. As church membership status can not be identified for individual/year cells with zero local income tax (and consequently also zero church tax), these cells are not included in our 10,501,554 individual/year cells.

  16. Since transfers and deduction are not available in FLEED, we obtain the correct tax base (\(\textrm{wage}-\textrm{deductions}+\textrm{transfers}\)) for each individual/year from the Tax Income Database at Statistics Finland.

  17. We link spouses income to each each individual/year cell through a spouse identifier (the social security number of the spouse). Since this identifier is nonmissing only for the married individuals or the ones cohabitating, the household income variable will be the sum of the individual’s own payroll and capital gains for all others, e.g., for singles.

  18. Although the church tax rate varies within municipalities over time in the data, the researcher only observes the county identifier (not a municipality identifier) for each individual. This way, Statistics Finland prevents the potential identification of individuals.

  19. The discussion on the choice between the LPM and its nonlinear alternatives draws on Wooldridge (2002, Chapter 15).

  20. It warrants mention that the component of the overall standard deviation that is generated by variation over time within individuals is roughly 62 euros. Thus, multiplying the obtained coefficients of church tax with this “within” standard deviation yields slightly less than half the size of the size effects reported for one overall standard deviation.

  21. We also test for forward looking behavior by including contemporaneous church tax instead of its first lag in the equivalent regression as the one reported in column (4) of Table 2. We find significantly weaker effects. The coefficient (t-statistic) for current church tax, reported equivalently as in Table 2, is −0.0015 (0.000098).

  22. The specification for the structural break test is as follows:\(\left(\text{ Member}\right) _{it}=\lambda _{0}+\lambda _{1}\left( \text{Church tax} \right) _{it-1}+\lambda _{2}t\)+λ 3Post\(^{2002}+\lambda _{4}t\ast \) Post2002 \( +\; b^{\prime }X_{it}+\delta _{i}+\theta _{\textrm{county}}+v_{it.}\) Here, the year trend is denoted by t.

  23. The number of members were 4,348,442 in 2006. A − 0.01 elasticity coefficient implies a 0.1 % decrease in membership from a 10 % increase in prices, i.e., a loss of 4,348 members. At the average church tax in 2006 (≈267 nominal euros), the loss through decreasing membership caused by a 10 % increase in tax rates implies ≈1.2 million euros.

  24. The tendency to adhere to the default option has been documented, for instance, in participation to pension saving programs (Madrian and Shea 2001).


  • Bar-El R, García-Muñoz T, Neuman S, Tobol Y (2012) The evolution of secularization: cultural transmission, religion and fertility-theory, simulations and evidence. J Popul Econ. doi:10.1007/s00148-011-0401-9

    Google Scholar 

  • Deheija R, DeLeire T, and Luttmer E (2007) Insuring consumption and happiness through religious organizations. J Public Econ 91:259–279

    Article  Google Scholar 

  • Francis LJ, Katz YJ (eds.) (2000) Joining and leaving religion: research perspectives. Leominster: Gracewing

    Google Scholar 

  • Gruber J (2005) Religious market structure, religious participation, and outcomes: is religion good for you? Adv Econ Analysis Policy 5(1):Article 5

    Google Scholar 

  • Gruber J (2004) Pay or pray? The impact of charitable subsidies on religious attendance. J Public Econ 88(12):2635–2655

    Article  Google Scholar 

  • Gruber J, Hungerman DM (2007) Faith-based charity and crowd out during the great depression. J Public Econ 91:1043–1069

    Article  Google Scholar 

  • Gruber J, Hungerman DM (2008) The church versus the mall: what happens when religion faces increased secular competition? Quart J Econ 123:831–862

    Article  Google Scholar 

  • Heino H, Salonen H, Rusama J, Ahonen R (1997) Suomen evankelis-luterilainen kirkko vuosina 1992–1995. (Engl. The Evangelical Lutheran Church of Finland between years 1992–1995). Kirkon tutkimuskeskus. Tampere

  • Hungerman DM (2005) Are church and state substitutes? Evidence from the 1996 welfare reform. J Public Econ 89:2245–2267

    Article  Google Scholar 

  • Iannaccone L (1998) Introduction to economics of religion. J Econ Lit 36:1465–1495

    Google Scholar 

  • Lumijärvi J (1998) Miksi kuulua kirkkoon? Espoolaisten kirkkoon situotuminen1990-luvulla. (Engl. Why belong to church? The commitment to church among people in Espoo county, Finland) Publications of the Department of Practical Theology, University of Helsinki, No. 91

  • Madrian BC, Shea DF (2001) The power of suggestion: Inertia in 401(k) participation and savings behavior. Quart J Econ 116:1149–1187

    Article  Google Scholar 

  • Richter P, Francis LJ (1998) Gone but not forgotten: church leaving and returning. Dartman, Longman & Todd: London

    Google Scholar 

  • Wooldridge JM (2002) Econometric analysis of cross section and panel data. The MIT Press

Download references


We thank Saku Aura, Pierre-André Chiappori, Kristiina Huttunen, Pekka Ilmakunnas, Mikael Lindahl, Daniele Paserman, Tuomas Pekkarinen, Otto Toivanen, Roope Uusitalo, and the seminar participants for their helpful comments. We are also indebted to Jori Mäntysalo and his colleagues at Free Thinkers’ Union of Finland for their generous data provision and assistance. Lyytikäinen gratefully acknowledges financial support from the Yrjö Jahnsson Foundation. Santavirta gratefully acknowledges financial support from the Yrjö Jahnsson Foundation and Eemil Aaltonen Foundation.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Torsten Santavirta.

Additional information

Responsible editor: Junsen Zhang

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

(PDF 43.5 KB)

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Lyytikäinen, T., Santavirta, T. The effect of church tax on church membership. J Popul Econ 26, 1175–1193 (2013).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Church tax
  • Church membership
  • Transaction cost

JEL Classification

  • H24
  • H31
  • Z12