Challenges for Comparative Research on Philanthropy in Europe

1
Challenges for Comparative Research on
Philanthropy in Europe

• René Bekkers and Pamala Wiepking
• Department of Philanthropic Studies, VU
  University Amsterdam and ICS/Department of
  Sociology, Utrecht University
• The Netherlands
• R.Bekkers_at_fsw.vu.nl

2
Research Questions

• How large are differences in philanthropy between
  nations in Europe?
• How can these differences be explained?

3
What we have

• Lots of data on volunteering, but much less on
  charitable giving
• Several datasets on giving using
• Different definitions of philanthropy
• Different questionnaire modules to measure
  philanthropy
• Different survey methods

4
Prospects for Data Access

• Learning from the Civicus index
• Tax authorities and privacy issues
• Survey data on corporate philanthropy difficult
  to gather
• Foundations even more difficult to get access to
• Getting survey data on households least
  problematic ? lets do this!

5
What we need

• New data on giving, using
• A clear definition of philanthropy
• A validated, cross-nationally adequate instrument
  to measure philanthropy
• One single method of data collection

6
Definitions

• There is no true definition of philanthropy
• Definitions should identify a clearly delimited
  set of phenomena
• Issues
• Exclude informal giving?
• Exclude any benefits for donor?
• What is voluntary anyway?

7
Conceptual model
8
The questionnaire should identify

• Units of analysis individuals, AND/OR
  households, OR foundations, OR corporations
• Channels churches, charities, foundations, other
  nonprofit organizations
• Destinations causes and services
• Resources money, goods, labor

9
European Social Survey
E1-12 a) CARD 43 For each of the voluntary
organisations I will now mention, please use
this card to tell me whether any of these things
apply to you now or in the last 12 months, and,
if so, which. E1-12 b) Do you have personal
friends within this organisation?
10
WARNING

• DATA FROM THESE
• MEASURES MAY BE
• VERY FAR
• FROM THE LIKELY TRUE VALUES

11
Questionnaires on household giving

• The Gold Standard the Method Area Module
  (e.g., GINPS)
• Incomplete coverage Area (ESS2002, EB 62.2)
• Severely limited (Very) Short
• Methodology is Destiny shorter questionnaires
  yield (strong) underestimates of giving volume
  and bias parameter estimates
• Source Bekkers, R., Wiepking, P. (2006). To
  Give or Not to GiveThats the Question.
  Nonprofit Voluntary Sector Quarterly, 35 (3)
  533-540.

12
Donors per sector ()
13
Donors per sector in the Netherlands in two data
sources ()
14
PhilanthropyEuropean Social Survey, 2002
15
True giving likely to be underreported
Cross tabulation of ESS direct question on
political giving and marking donated for
political party in the matrix question (USNL)
16
Correlates of giving
17
What we dont know

• How are France, Germany, Italy, Denmark,
  Switzerland doing?
• How much is donated to charity?
• How do countries differ in the composition of
  philanthropy?
• How do countries differ in the characteristics of
  donors?
• Where do all these country differences come from?

18
PhilanthropyEuroBarometer 62.2, 2004
19
Donors per sector in the Netherlands in two data
sources ()
20
True giving likely to be overreported in EB

• 5 report giving to a political party or
  organization but only 2.5 is a member and a
  only a fraction donate
• 40 report giving to an environmental
  organization at best, 28 is a member
• Humanitarian aid (41) does not even include
  one-off Tsunami donations

21
Correlates of giving
22
What do we think we see?

• Philanthropy seems to be more widespread in
  countries with
• A Protestant background
• Higher levels of income
• Not so much in countries with
• Higher levels of education
• Stronger social networks

23
Legal Regulation and Other Context Correlates

• Philanthropy seems to be more widespread in
  countries with
• Independent monitoring system
• Non-communist past
• Social democratic past
• Larger government expenditures
• Lower income inequality
• Higher level of trust

24
Understanding philanthropy

• Bekkers Wiepking (2007)
• Literature review of all (550) empirical articles
  published on philanthropy
• Identifies 8 mechanisms driving giving, including
  solicitation, costs, values
• We should measure these context mechanisms

25
Comparative Research Questions

• There seem to be large differences between
  countries in philanthropy
• Are these differences the result of context or
  composition effects?
• Do conventional (US, UK, NL) explanations of
  philanthropy hold in other countries?

26
Solicitation

• Most giving occurs in response to solicitations
• Cf. discrimination on paid labor market
• How much solicitation occurs and what methods are
  used to raise funds shape level and composition
  of donor pool
• Likewise for volunteering

27
Giving in the Netherlands (GIN)

• Cross-sectional surveys since 1995
• Computer Assisted Self Interview (online survey)
  TNS/Nipo
• Easily extended to the other European countries
• Method-Area Modules on giving and volunteering
• Reward for completing survey
• Experiments with donating reward

28
GINPS Design

• Panel study 2002, 2004, 2006, 2008
• Panel started with representative sample
  (n1,707) and oversample of Protestant Christians
  (n257)
• Longitudinal cross-country study
• Internet use limits generalizability, but
  produces valid estimates

29
GINPS validation Catholics

• Estimated total value of donations to Catholic
  church based on self-reports is exactly the same
  as true amount donated
• Underreporting of incidence (27 vs 44) and
  overreporting of amount (81 vs 70) lead to
  underestimate (22 vs 31)

30
Experimental validation study

• Comparison of predictors of amount donated in
  past year and donation of reward for
  participation in survey
• Parameter signs are similar, except for church
  attendance (NS in experiment)
• Source Bekkers, R. (2007). Measuring Altruistic
  Behavior in Surveys The All-Or-Nothing Dictator
  Game. Survey Research Methods, 1(3) 139-144.

31
Cancer Fund validation study

• Amounts donated to Cancer Fund were matched to
  database for households that agreed to matching
  (95, n191)
• Reported amounts were slightly higher than
  recorded amounts median difference 2.32
  (mode0)
• Average correlation .853
• Catholics and (especially) Protestants
  underestimated giving

32
European Research Network on Philanthropy (ERNOP)

• Founded January 2008
• The State of Giving Research. Household
  donations to charitable organizations in twelve
  European countries (Amsterdam University Press,
  2009)

33
Further information

• Rene Bekkers, Head of Research, Department of
  Philanthropic Studies, VU University Amsterdam
• R.Bekkers_at_fsw.vu.nl
• www.giving.nl

34
PhilanthropyESS02 EB62.2
35
Volunteering European Social Survey, 2002
36
Correlates of giving