KYKLOS, Vol. 62 – August 2009 – No.

3, 359–370

Intrinsic vs. Extrinsic Motivations to Volunteer and

Social Capital Formation

Giacomo Degli Antoni

I. INTRODUCTION

Although intrinsic motivations receive increasing attention in explaining

human actions (Titmuss, 1970, Deci, 1971, 1972, 1975, Frey, 1992, 1997),
our knowledge on their causes and effects is incomplete. Quite surprisingly, the
existing literature fails to consider the relationship between intrinsic motiva-
tions and social capital formation. This paper draws on an original dataset
developed by the author to investigate the effect of various motivations to
volunteer on the creation of volunteers’ social capital.1
Even in standard economic environments, people do not care about their
material payoffs alone or, more in general, about the consequences of actions;
they also have intrinsic reasons to act (Frey and Jegen, 2001). Following Deci,
we say that one ‘‘is intrinsically motivated to perform an activity when one
receives no apparent reward except the activity itself’’ (Deci, 1971, p.105).2
Specific conditions characterizing different situations may be more (or less)
congenial for intrinsic motivations to arise (Frey, 1997). The typical walrasian
Research fellow - EconomEtica, interuniversity centre of research, Bicocca University, Via Bicocca degli
Arcimboldi 8, 20126, Milano, Italy. Email: giacomo.degliantoni@unimib.it. The analysis presented in this
paper stems from research conducted within the joint project ‘‘The added value of volunteer work’’ carried
out by EconomEtica – University of Milano Bicocca, University of Parma, ‘‘Forum Solidarietà – centre
for voluntary work in Parma’’ and Cariparma Foundation. I would like to thank all the project’s
participants. I also wish to thank Marco Faillo, Lorenzo Sacconi, and two anonymous referees who
allowed me to improve the final version of this paper. Finally, I am deeply indebted to Gianluca Grimalda
for his accurate and precious notes and suggestions. Remaining errors are solely my own.
1
Even though literature has studied the role of intrinsic and extrinsic motivations in explaining the decision
to become a volunteer (see Hackl, Halla and Pruckner, 2007) we are not aware of previous studies focused
on the effect of these different motivations on social capital.
2
The literature stresses that the distinction between extrinsic and intrinsic motivations is not always clear
and that ‘‘in many cases, the two motivations come together’’ (Frey, 1997, p. 14). However, we share Frey’s
ideas (1997) that ‘‘for the purpose of explaining economically and socially relevant human behavior, it
suffices that it makes sense to distinguish activities which individuals (mainly) do just because they like
them, and others which they (mainly) do because they are induced to do so by monetary payment or by
command.’’ (Frey, 1997, p.14).

r 2009 Blackwell Publishing Ltd., 9600 Garsington Road,

Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA 359
 GIACOMO DEGLI ANTONI

market characterized by anonymous interactions restricted to the transmission

of coded information through the auctioneer’s agency seems not to leave room
for intrinsic motivations. And yet, intrinsic motivations may play an important
role in institutions characterized by personal relations and high participation in
decisions. This is, for example, the case of nonprofit organizations (Frey and
Goette, 1999), where the large presence of volunteer work may strengthen the
idea that people act because they obtain satisfaction from the action in itself.
Obviously, volunteers may also be extrinsically motivated. Firstly, they may
desire to improve their human capital through the activities carried out in the
association (Menchik and Weisbrod, 1987). Secondly, they may be attracted by
the social recognition that derives from volunteering (Schiff 1990, Meier and
Stutzer, 2008). Thirdly, people can volunteer in order to increase their social
network (Prouteau and Wolff, 2004, Meier and Stutzer, 2008).
This paper studies the influence of intrinsic and extrinsic motivations to
volunteer on three indices of social capital intended as networks of cooperative
relations.3
Several contributions on the effect of participation in voluntary associations
refer to notions usually associated with the concept of social capital. They
consider the effect of voluntary participation on civicness (Mayer, 2003,
Wollebæck and Selle, 2003), generalized trust (Brehm and Rahn, 1997, Stolle
and Rochon, 1998, Claibourn and Martin, 2000, Mayer, 2003, Wollebæck and
Selle, 2003), trust in public institutions (Brehm and Rahn, 1997, Stolle and
Rochon, 1998, Mayer, 2003, Wollebæck and Selle, 2003) and indicators of
tolerance, free riding and optimism (Stolle and Rochon, 1998). Whilst social
capital understood in terms of social norms has been thoroughly studied in
relation to associational membership, this is not the case of social capital
understood as a cooperative network of relations.4 Contributions in this regard
have been made by Wollebæck and Selle (2002) and Prouteau and Wolff (2004).
Wollebæck and Selle consider two proxies of social networks. Both are
computed by considering the groups outside the sphere of family and relations
which are evaluated to be important parts of one’s social circle. Five groups
were considered: ‘‘neighbours and local community’’ ‘‘current colleagues or

3
There are two main approaches to the concept of social capital. The first considers social capital in terms
of civicness and social norms of trust and reciprocity (e.g. Putnam et al., 1993, Knack and Keefer, 1997),
the second conceives social capital in terms of cooperative networks of relations (this is the approach
followed, for example, by Coleman 1988, 1990 and Burt, 1992, 2002).
4
It has to be stressed that, at a macro level, since the seminal work by Putnam et al. (1993), associational
membership in itself is often considered a proxy for social capital. In this perspective, for example, Sabatini
(2008) considers different indicators of voluntary participation to construct a proxy of social capital as
‘‘Voluntary Organizations’’ and Bjørnskov (2003) takes into account civic participation along with
generalized trust and perceived corruption to construct a synthetic measure of social capital. However, we
adopt a micro approach and investigate the relation between voluntary participation and relational
networks by analyzing how participation affects social network formation.

360 r 2009 Blackwell Publishing Ltd.

 INTRINSIC VS. EXTRINSIC MOTIVATIONS

fellow students’’ ‘‘former colleagues or fellow students’’ ‘‘friends from where

one grew up’’ and ‘‘others’’. The first measure of social networks is based on the
number of groups that a respondent declared to be part of his/her network. The
second proxy is a dummy variable that represents the presence or absence of
‘‘friends obtained in the current situation’’, that is, either current colleagues,
fellow students, or neighbours (Wollebæck and Selle, 2002, p.41). This second
indicator aims to measure the impact of involvement in associations on the
construction of new social networks. Both indicators are shown to be positively
correlated with membership. Prouteau and Wolff (2004) study the relationship
between consumption of relational goods and participation. They analyse the
correlation between associational participation and the number of informal
meetings with friends by using data at household level. The authors find that
the frequency of gatherings with friends per household is positively affected by
the participation in voluntary associations by at least one of the family
members.
In this paper we improve the understanding of the relationship between
associational membership and social network in three directions. Firstly, we
consider the role that different motivations to volunteer have on the creation of
social networks connected to participation in voluntary associations.5 It will be
shown that social capital formation is fostered by participation in voluntary
associations more for intrinsically motivated volunteers than for extrinsically
ones. Secondly, we use data at individual level which enable us directly to
connect participation and the creation of new social networks and to minimize
endogeneity problems (our questions are of the type: how many people met
since joining the association do you regard as friends?). Finally, we study the
effect of participation on social networks by looking both at the quantitative
side of networks (how many people met since joining the association do you
regard as friends?) and at the qualitative one (by considering the degree of
familiarity characterizing the relations started through the association).6
Section 2 presents the database, the social capital and the motivation indices
considered in the empirical analysis. Section 3 and 4 illustrate descriptive
evidence and econometric findings. Section 5 concludes.

5
Meier and Stutzer (2008) show that volunteering positively affects life satisfaction and, in particular, they
show that ‘‘people who are more extrinsically oriented benefit less from volunteering than people who put
more importance on intrinsic life goals.’’ (Meier and Stutzer, 2008, p.55). This result confirms that the
distinction between intrinsic and extrinsic motivations is important to explain the effect of volunteerism on
volunteers.
6
A recent study (Degli Antoni, 2009) has shown that the effect of social networks on economic variables
(specifically individual economic welfare) can only be fully understood by considering also the qualitative
aspect of social networks (which in Degli Antoni’s paper are considered by looking at the degree of
satisfaction with relations) whereas the literature usually focuses on the quantitative element.

r 2009 Blackwell Publishing Ltd. 361

 GIACOMO DEGLI ANTONI

II. DATABASE: SOCIAL CAPITAL AND MOTIVATION INDICES.

1. Database

The present analysis is based on an original database collected at the end of

2007 by the author by means of anonymous questionnaires filled in by 290
volunteers of 45 voluntary associations operating in Parma (the ninth Italian
province in terms of number of voluntary associations per inhabitants, with 7.3
voluntary associations per 10,000 inhabitants, Istat, 2003).7 The sample of
associations was a stratified random sample representing 10% of associations
operating in the province. The strata referred to the association’s activity8 and
to the district in which it operated.9 The number of volunteers per association
was 6.4 on average (minimum 2, maximum 11 and standard deviation 2.4). The
290 volunteers were randomly selected among the associations’ members and
they completed a structured questionnaire of 64 questions relative to their
experiences as volunteers. Compilation of the questionnaire lasted on average
45 minutes.
Information on various association’s characteristics, such as size, year of
foundation, operational characteristics etc., were also gathered by asking a
specific set of questions to a member of the association having detailed
knowledge of these aspects.10 In most cases this was the president of the
association. In ten cases these subjects also completed the questionnaire on
their experience as volunteers. We introduced a dummy variable in the
econometric analysis when this was the case.

2. Social Capital Indices

We considered three proxies for social capital. The first one (named network_
increase) measured the impact of participation on the increase in the member’s
7
In 1991, law no. 266 regulated voluntary associations in Italy by providing that, in order to access public
grants and to benefit from tax relief, they must be characterized by solidarity aims and a democratic
structure and their members must be for the most part volunteer workers. In this paper we focus only on
associations which fulfil these criteria and which are consequently registered in the public registers of
voluntary associations.
8
Seven activities are undertaken by the associations operating in Parma: Assistance, Health, Environ-
mental and Animal Conservation, Recreation and Culture, Civil Defence, Education, Civil Rights
Promotion and Preservation. We decided to stratify the sample by considering the activities because
volunteers’ motivation may significantly change in relation to the activity of the association.
9
Parma province is divided into four administrative districts which vary a lot for population density. In
order to have all the districts represented in our sample we decided to stratify the sample according to the
four districts.
10
The variables constructed by means of these questions concerned the organizational level and assumed
the same value for each volunteer belonging to the same association.

362 r 2009 Blackwell Publishing Ltd.

 INTRINSIC VS. EXTRINSIC MOTIVATIONS

social network. It was developed from the question: ‘‘As a whole, how many
people met since joining the association are now your friends?’’.
The second and third social capital proxies took explicit account of the
degree of familiarity characterizing the relations formed through the associa-
tion. The proxy named higher_familiarity was the standardized11 mean value of
the 4 answers to the following questions for each respondent: ‘‘How many
people met through the association would you:

a. talk to about family problems?

b. trust to look after your relatives (children/elderly persons)?
c. ask to take care of your home when you are on holiday?
d. give/ask for help in activities such as shopping, taking a child or elderly
persons to do different activities etc.?’’.

The proxy named lower_familiarity was the standardized mean value of the 3
answers to the question: ‘‘With how many people met through the association
have you started the following cooperative relations:

a. phone calls to ask for information or advice?

b. doing not very demanding errands?
c. asking for information about job opportunities?’’.

The activities considered in the higher_familiarity index directly concern the

family circle and consequently presumably imply a higher degree of confidence
than those considered in the lower_familiarity index. The aggregation into two
synthetic indices of social capital of the single questions was justified by their
high degree of correlation (all correlations are positive and statistically
significant at 1%).

3. Motivation Indices

Although volunteers may not be motivated by monetary incentives, it cannot

be ruled out that their decision to join a voluntary association was prompted by
other extrinsic motivations. We considered two extrinsic motivations to
volunteer: a desire to increase the social recognition which is usually associated
with volunteerism (Schiff, 1990, Meier and Stutzer, 2008) and the desire to
11
This proxy (and also the proxy lower_familiarity) was standardized with the following procedure:
xic
minðxi Þ
where: xic indicates the value i related to the association c. The standardization process
maxðxi Þ
minðxi Þ
generates standardized indicators with the same range of variation between 0 to 1, and it produces a more
robust trial in the presence of outliers (Saisana and Tarantola, 2002, p.11), which seem to characterize our
indicators.

r 2009 Blackwell Publishing Ltd. 363

 GIACOMO DEGLI ANTONI

increase the number of acquaintances or friends (Prouteau and Wolff, 2004,

Meier and Stutzer, 2008). With respect to the intrinsic motivations to volunteer,
we considered the desire to feel useful for others and ideal motivations.
Extrinsic and intrinsic motivations were measured by asking subjects: ‘‘With
respect to your decision to become a volunteer, how important were the
following aspects, from 1 (not at all) to 7 (entirely)?

| the desire to increase your number of acquaintances or friends (variable

named mot_network)
| the pursuit of social recognition (mot_socialrecognition)
| ideal motivations (mot_ideal)
| the desire to feel useful for others (mot_usefulness)’’

In the empirical analysis we consider both the single answers and two indices
developed by computing the arithmetic mean for each respondent of the replies
relative to the extrinsic and intrinsic motivations respectively. The two
aggregate indices are named mot_extrinsic and mot_intrinsic.12

III. DESCRIPTIVE FINDINGS

Descriptive statistics show that participation positively affects social networks

of cooperative relations. The average of persons met through the association
who have become part of the social network of volunteers (variable network_
increase) is equal to 7.48 (std.dev. 9.76; min 0; max 50),13 and 76.90% of
respondents declared a number greater than 0. At least 50% of respondents
declared to have met through the association at least one person they (higher_
familiarity index): would talk to about family problems (mean and median of
answers: 5,26 and 2 respectively); would trust to look after their relatives
(children/elderly persons) (3.24;1); would ask to take care of their home while
they are on holiday (2.88;1); would give/ask for help in activities such as
shopping, taking a child or elderly persons to do different activities etc.
(3.16;0.5). More than 50% of respondents declared to have started through
associations, at least with one person, the following cooperative relations
(lower_familiarity index): phone calls to ask for information or advice (5.05;3);
doing not very demanding errands (3.55;1); asking for information about job
12
In the empirical analysis, we assume that the number of social relations created through associations
does not affect the answers to the questions on motivations to become a volunteer. It would generate a
problem of endogeneity. According to our opinion, this risk is minimized both by the clear formulation of
the questions and by the use of different motivation and social capital indices.
13
The mean calculation did not include the 7 highest values declared by respondents, which ranged from 90
to 400. If these outliers are included, the mean is equal to 11.96 (std. dev. 34.31) and the median is 4.

364 r 2009 Blackwell Publishing Ltd.

 INTRINSIC VS. EXTRINSIC MOTIVATIONS

Figure 1
With respect to your decision to become a volunteer, how important were the following aspects,
from 1 (not at all) to 7 (entirely)?

opportunities (5.11;2). Note that the lowest mean and median concern
cooperation related to help with activities such as taking a child to do different
activities etc. which presumably implies a high degree of familiarity, while the
highest median (and the second highest mean) concerns the relation which
presumably involves the lowest degree of familiarity (phone calls to ask for
information or advice).
As for the motivations behind becoming a volunteer, intrinsic motivations
seem to be much more important than extrinsic motivations in explaining the
decision to engage in volunteer work (Fig. 1).
The desire to feel useful for others is the factor that most affects the decision
to become a volunteer, while the pursuit of social recognition is the least
important one. The desire to increase the number of acquaintances or friends
matters more than the pursuit of social recognition, but much less than the
intrinsic motivations.

IV. ECONOMETRIC FINDINGS

The empirical analysis used OLS estimates. In all the regressions we clustered
standard errors by considering to which associations the volunteers belonged.
We assumed that the observations were independent across groups, but not
necessarily between groups (volunteers belonging to the same association).

r 2009 Blackwell Publishing Ltd. 365

 GIACOMO DEGLI ANTONI

Table 1
Motivations and Creation of Social Capital
Equation 1 (OLS) 2(OLS) 3(OLS) 4(OLS) 5(OLS) 6(OLS)
Dependent variable
higher_familiarity lower_familiarity network_increase
Mot_intrinsic 0.009 0.018 1.348
(2.50) (2.67) (0.91)
Mot_ extrinsic 0.002 0.006 4.062
(0.53) (0.94) (2.15)
Mot_ideal 0.004 0.009 1.965
(1.79) (2.78) (2.11)
Mot_ usefulness 0.006 0.009 2 1.363
(1.15) (1.01) (2 0.79)
Mot_network 2 0.001 0.001 2.460
(2 0.33) (0.16) (1.93)
Mot_socialrecognition 0.003 0.006 1.472
(0.90) (0.83) (0.69)
Constant 2 0.232 2 0.242 2 0.301 2 0.311 2 115.728 2 112.389
(2 4.56) (2 3.91) (2 3.50) (2 2.93) (2 4.60) (2 3.87)
R2 0.319 0.325 0.276 0.278 0.408 0.414
Root MSE 0.052 0.052 0.088 0.088 30.325 30.398
Obs 165 165 164 164 163 163
t-statistics in brackets. Significant at 10%; significant at 5%; significant at 1%.

Equations 1 and 2 in Table 1 show the results relative to the index of social
capital referred to the relations started through associations characterized by a
high degree of familiarity (higher_familiarity). Equations 3 and 4 consider the
relations characterized by a low degree of familiarity (lower_familiarity).
Equations 5 and 6 focus on the quantitative aspect of social networks simply
by considering the number of persons met since joining the association whom
the respondent reports as friends (network_increase).
All the equations include an array of control variables:14

 demographic and occupational individual characteristics, i.e. age; sex;

education; condition of being unemployed;
 degree of involvement by the volunteer with the organization, i.e. number
of months the volunteer has worked for the association; number of hours
per week devoted to the association by the volunteer;

14
Descriptive statistics and estimates’ results related to these variables are omitted for reason of space and
available from the author upon request.

366 r 2009 Blackwell Publishing Ltd.

 INTRINSIC VS. EXTRINSIC MOTIVATIONS

 association’s characteristics, i.e. the sector of activity, the district in which

the association operated, number of the association’s volunteers; number
of members not actively participating in the association; number of paid
workers; number of recipients of the association’s services; numbers of
years in operation; association’s area of activity that varies from the city
(value 1) to the international level (value 6); how often discussion groups
aimed at raising members’ awareness of collective problems are organized
by the association, from 1 (never) to 6 (every week); how often informal
meetings to discuss the association’s activity are organized by the
association, from 1 (never) to 6 (every week).

Equation 1 shows that intrinsic motivations to volunteer positively affect the

formation of relational networks involving the volunteer and people met
through the association characterized by a high degree of familiarity. If we
consider the various motivations to volunteer separately, the more robust
effect seems to be the one generated by ideal motivations. From a quantitative
point of view, when mot_intrinsic increases by a standard deviation, the
effect on higher_familiarity is an increase of 0.177 in the standard deviation
(equation 1) and an increase of one standard deviation in mot_ideal increases
the higher_familiarity index of 0.117 standard deviation (equation 2).
Equation 3 shows that intrinsic (and not extrinsic) motivations affect also the
dependent variable lower_familiarity. The coefficient magnitude of the intrinsic
motivations index implies that the increase of standard deviation in mot_
intrinsic increases the lower_familiarity index of 0.215 standard deviation
(equation 3). In particular, mot_ideal instead of mot_ usefulness seems to
generate the effect on lower_familiarity. A standard deviation increase in
mot_ideal is associated with an increase in the lower_familiarity index of 0.158
standard deviation (equation 4).
These results suggest that, thanks to volunteer work, ideal motivated
volunteers are more likely to increase the network of social relations through
which they may ask or give cooperation in different activities. Neither
motivations associated with the desire to feel useful for others nor extrinsic
motivations have a similar role in increasing the social capital generated
through the experience as a volunteer.
The dependent variable of equations 5 and 6 focuses on the quantitative
aspect of social networks by considering the number of persons met since
joining the association whom the respondent reports as friends. The use of this
indicator of social capital is complementary to the ones considered in the
equations 1–4, and gives us important insights into the nature of the friendship
relationships started through association membership. If we look at the two
aggregate indices of motivation, extrinsic motivations seem to prevail over
intrinsic ones in their correlation with the dependent variable. In particular,

r 2009 Blackwell Publishing Ltd. 367

 GIACOMO DEGLI ANTONI

when mot_extrinsic increases by a standard deviation, the effect on network_

increase is an increase of 0.172 in the standard deviation (equation 5). When we
include the single motivations in the regression (equation 6), we find that both
ideal motivations and the desire to increase the number of acquaintances or
friends positively affect the creation of social networks at a quantitative level.
This result is interesting and suggests that extrinsic motivations, and in
particular the decision to join an association in order to increase the number of
acquaintances or friends, promote the creation of networks from a quantitative
point of view but they do not facilitate the creation of relations based on a
particular degree of familiarity. We may say that people who join associations
for instrumental reasons achieve their aim because they increase their social
network, but they are unable to make their new acquaintances part of their
more close and familiar friends.

V. CONCLUSIONS

Although intrinsic motivations receive increasing attention in explaining

human behavior, our knowledge on their causes and effects is incomplete. In
particular, the existing literature lacks to address the effects of different
motivations to act on social capital formation. The present paper focused on
motivations to volunteer and empirically investigated the role of intrinsic and
extrinsic motivations in promoting volunteers’ social capital. Our analysis
considered three indices of social capital, aimed at measuring both the
quantitative (number) and the qualitative (degree of familiarity and coopera-
tion) character of social relations, and intrinsic and extrinsic motivations to
volunteer (ideal motivations, the desire to feel useful to others, the pursuit of
social recognition and the desire to increase the number of acquaintances or
friends).
We found that the creation of social capital through participation in
voluntary associations is not indifferent to the motivations which induced
the volunteer to start his/her unpaid activity. Intrinsic motivations, in terms of
ideal motivations, enable people to create relations characterized by a
significant degree of familiarity where the latter is measured by considering
the activities which characterize the relations started through associations. By
contrast, extrinsic motivations, and in particular the decision to join an
association in order to increase the number of acquaintances or friends,
promote the creation of networks from a quantitative point of view, but they
do not facilitate the creation of relations based on a particular degree of
confidence.
Our findings open important questions for further research. We have not
investigated either the determinants of intrinsic motivations to volunteer or the

368 r 2009 Blackwell Publishing Ltd.

 INTRINSIC VS. EXTRINSIC MOTIVATIONS

relation between intrinsic motivations and operational characteristics of

associations. Suitable actions implemented by voluntary associations which
promote intrinsic motivations to volunteer may increase membership in
associations as well as the creation of social capital. By contrast, incentives
which crowd out intrinsic motivations of volunteers may produce a negative
effect on the creation of cooperative networks of relations generated through
the participation in voluntary associations.

REFERENCES

Bjørnskov, Christian (2003). The Happy Few: Cross-Country Evidence on Social Capital and Life
Satisfaction, Kyklos. 56(1): 3–16.
Brehm, John and Wendy Rahn (1997). Individual-Level Evidence for the Causes and Consequences
of Social Capital, American Journal of Political Science. 41(3): 999–1023.
Burt, Ronald S. (1992). Structural Holes. Cambridge (Mass): Harvard University Press.
Burt, Ronald S. (2002). The Social Capital of Structural Holes, in Mauro F. Guillen, Randall Collins,
Paula England and Marshall Meyer (eds.), The New Economic Sociology. New York: Russell
Sage Foundation.
Claibourn, Michel P. and Paul S. Martin (2000). Trusting and Joining? An Empirical Test of the
Reciprocal Nature of Social Capital, Political Behavior. 22(4): 267–291.
Coleman, James S. (1988). Social Capital in the Creation of Human Capital, American Journal of
Sociology. 94: 95–120.
Coleman, James S. (1990). Foundations of Social Theory. Cambridge: Harvard University Press.
Deci, Edward L. (1971). Effects of Externally Mediated Reward on Intrinsic Motivation, Journal of
Personality and Social Psychology. 18: 105–115.
Deci, Edward L. (1972). Intrinsic Motivation, Extrinsic Reinforcement and Inequity, Journal of
Personality and Social Psychology. 22(1): 113–120.
Deci, Edward L. (1975). Intrinsic Motivation. New York: Plenum Press.
Degli Antoni, Giacomo (2009). Does Satisfaction Matter? A Microeconomic Empirical Analysis on
the Effect of Social Relations on Economic Welfare, Journal of Socio-Economics. 38(2): 301–309.
Frey, Bruno S. (1992). Tertium Datur: Pricing, Regulating and Intrinsic Motivation, Kyklos. 45(2):
161–184.
Frey, Bruno S. (1997). Not Just for The Money. An Economic Theory of Personal Motivation.
Cheltenham, UK and Brookfield, USA: Edward Elgar.
Frey, Bruno S. and Lorenz Goette (1999). Does Pay Motivate Volunteers?, Unpublished Manu-
script. Institute for Empirical Economic Research. University of Zurich.
Frey, Bruno S. and Reto Jegen (2001). Motivation Crowding Theory: A Survey of Empirical
Evidence, Journal of Economic Surveys. 15(5): 589–611.
Hackl, Franz, Martin Halla and Gerald J. Pruckner (2007). Volunteering and Income – The Fallacy
of the Good Samaritan?, Kyklos. 60(1): 77–104.
Istat (2003). Le Organizzazioni di Volontariato in Italia. Rome: Istat.
Knack, Stephen and Philip Keefer (1997). Does Social Capital Have an Economic Payoff? A Cross
Country Investigation, The Quarterly Journal of Economics. 112: 1251–1287.
Mayer, Nonna (2003). Democracy in France: Do Associations Matter?, in Marc Hooghe and
Dietlind Stolle (eds) Generating Social Capital Civil Society and Institutions in Comparative
Perspective. Palgrave Macmillan.
Meier, Stephan and Alois Stutzer (2008). Is Volunteering Rewarding in Itself?, Economica. 75:
39–59.

r 2009 Blackwell Publishing Ltd. 369

 GIACOMO DEGLI ANTONI

Menchik, Paul L. and Burton A. Weisbrod (1987). Voluntary Labor Supply, Journal of Public
Economics. 32: 159–183.
Prouteau, Lionel and Francois-Charles Wolff (2004). Relational Goods and Associational Partici-
pation, Annals of Public & Cooperative Economics. 75(3): 431–463.
Putnam, Robert D., Robert Leonardi and Raffaella Y. Nanetti (1993). Making Democracy Work.
Princeton: Princeton University Press.
Sabatini, Fabio (2008). Social Capital and the Quality of Economic Development, Kyklos. 61(3):
466–499.
Saisana, Michaela and Stefano Tarantola (2002). State-of-the-Art Report on Current Methodologies
and Practices for Composite Indicator Development, European Commission Joint Research
Center, EUR20408EN, European Commission-Joint Research Centre, Ispra.
Schiff, Jerald A. (1990). Charitable Giving and Government Policy. An Economic Analysis. New-York:
Greenwood Press.
Stolle, Dietlind and Thomas R. Rochon (1998). Are all Associations Alike?: Member Diversity,
Associational Type and Creation of Social Capital, American Behavioral Scientist. 42: 47–65.
Titmuss, Richard M. (1970). The Gift Relationship. From Human Blood to Social Policy. London:
Allend and Unwin.
Wollebæck, Dag and Per Selle (2002). Does Participation in Voluntary Associations Contribute to
Social Capital? The Impact of Intensity, Scope, and Type, Nonprofit and Voluntary Sector
Quarterly. 31(1): 32–61.
Wollebæck, Dag and Per Selle (2003). The Importance of Passive Membership for Social Capital
Formation, in Marc Hooghe and Dietlind Stolle (eds), Generating Social Capital Civil Society and
Institutions in Comparative Perspective. Palgrave Macmillan.

SUMMARY

Although intrinsic motivations receive increasing attention in explaining human actions, our knowledge
on their causes and effects is incomplete. Quite surprisingly, the existing literature fails to consider the
relationship between intrinsic motivations and social capital formation. The present paper increases the
understanding on the effect of intrinsic motivations by studying the role that different motivations to
volunteer have on the creation of volunteers’ social capital which is intended as networks of cooperative
relations.
Our empirical analysis considers three indices of social capital, aimed at measuring both the quantitative
(number) and the qualitative (degree of familiarity and cooperation) character of social relations, and
intrinsic and extrinsic motivations to volunteer (ideal motivations, the desire to feel useful to others, the
pursuit of social recognition and the desire to increase the number of acquaintances or friends).
We find that the creation of social capital through participation in voluntary associations is not indifferent
to the motivations which induced the volunteer to start his/her unpaid activity. In particular, we show that
intrinsic motivations enable people to extend their social networks by creating relations characterized by a
significant degree of familiarity. By contrast, extrinsic motivations, and in particular the decision to join an
association in order to increase the number of acquaintances or friends, promote the creation of networks
from a quantitative point of view, but they do not facilitate the creation of relations based on a particular
degree of confidence.

370 r 2009 Blackwell Publishing Ltd.