time for blood: the e⁄ect of a naturally-occurring ... · the e⁄ect of a naturally-occurring...

30
Time for blood: The e/ect of a naturally-occurring incentive for altruistic behavior Nicola Lacetera y University of Toronto Mario Macis z Johns Hopkins University November 24, 2010 Abstract This paper analyzes the e/ects of a legislative provision that grants a one-day paid leave of absence to blood donors who are employees in Italy, using a unique dataset with the complete donation histories of the blood donors in an Italian town. The across-donor variation in job market status, and within-donor job switching over time are the sources of variation that we employ to study whether the paid-day-o/ incentive a/ects the frequency of their donations. Our analysis indicates that the day-o/ privilege leads donors who are employees to make, on average, one extra donation per year, which represents an increase of around 40%. We discuss the implications of our ndings for policies aimed at reducing the shortages in the supply of blood and, more generally, for incentivizing pro-social behavior. Keywords: Incentives, Altruism, Public Good Provision, Pro-social Behavior, Public Health. JEL Codes: D12, D64, I18. This is a revised version of a paper previously circulated under the title "Motivating Altruism: A eld Study". We thank James Andreoni, Oriana Bandiera, Gary Becker, Alberto Bisin, Thomas Buchmueller, Christopher Carpenter, David Clingingsmith, Stefano DellaVigna, John DiNardo, Luigi Guiso, Daniel Hamermesh, Uri Gneezy, Antonio Lacetera, Francesca Mazzolari, Hugo Mialon, Felix Oberholzer-Gee, James Rebitzer, Heather Royer, Ariel Rubinstein, Jesse Shapiro, Joel Slemrod, Justin Sydnor, Lorenzo Zirulia, Ezra Zuckerman, and seminar participants at Indiana University, Northwestern University, MIT, UC Irvine, the University of Michigan, Cleveland State University, Case Western Reserve University, the University of Chicago, the 2008 ASSA Meetings, and the 2008 ASHE Conference for very useful comments. To protect the privacy of the donors in our database, we have agreed to keep the name of The Town (as well as any other identifying information) condential. We are extremely thankful to the President, Board, members, and sta/ of The Towns unit of the Italian Association of Blood Donors (Associazione Volontari Italiani del Sangue, AVIS), and to the Head and sta/ of the Transfusion Unit in The Towns Hospital, for their precious collaboration. Ra/aella Manna has o/ered an invaluable contribution to the data collection e/ort, and Hong Pum Chung has provided excellent research assistance. This research was conducted in part while Nicola Lacetera was at Case Western Reserve University, and Mario Macis was at the Ross School of Business of the University of Michigan. Financial support from the Ross School of Business Small Grants Research Fund is gratefully acknowledged. y Rotman School of Management, Department of Strategy, and UT-Mississauga, Department of Management. E-mail: [email protected] z Carey School of Business. E-mail: [email protected]

Upload: others

Post on 30-Apr-2020

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

Time for blood:The e¤ect of a naturally-occurring incentive for altruistic behavior�

Nicola Laceteray

University of TorontoMario Macisz

Johns Hopkins University

November 24, 2010

Abstract

This paper analyzes the e¤ects of a legislative provision that grants a one-day paid leave ofabsence to blood donors who are employees in Italy, using a unique dataset with the completedonation histories of the blood donors in an Italian town. The across-donor variation in jobmarket status, and within-donor job switching over time are the sources of variation that weemploy to study whether the paid-day-o¤ incentive a¤ects the frequency of their donations.Our analysis indicates that the day-o¤ privilege leads donors who are employees to make, onaverage, one extra donation per year, which represents an increase of around 40%. We discussthe implications of our �ndings for policies aimed at reducing the shortages in the supply ofblood and, more generally, for incentivizing pro-social behavior.Keywords: Incentives, Altruism, Public Good Provision, Pro-social Behavior, Public Health.JEL Codes: D12, D64, I18.

�This is a revised version of a paper previously circulated under the title "Motivating Altruism: A �eld Study". Wethank James Andreoni, Oriana Bandiera, Gary Becker, Alberto Bisin, Thomas Buchmueller, Christopher Carpenter,David Clingingsmith, Stefano DellaVigna, John DiNardo, Luigi Guiso, Daniel Hamermesh, Uri Gneezy, AntonioLacetera, Francesca Mazzolari, Hugo Mialon, Felix Oberholzer-Gee, James Rebitzer, Heather Royer, Ariel Rubinstein,Jesse Shapiro, Joel Slemrod, Justin Sydnor, Lorenzo Zirulia, Ezra Zuckerman, and seminar participants at IndianaUniversity, Northwestern University, MIT, UC Irvine, the University of Michigan, Cleveland State University, CaseWestern Reserve University, the University of Chicago, the 2008 ASSA Meetings, and the 2008 ASHE Conference forvery useful comments. To protect the privacy of the donors in our database, we have agreed to keep the name of TheTown (as well as any other identifying information) con�dential. We are extremely thankful to the President, Board,members, and sta¤ of The Town�s unit of the Italian Association of Blood Donors (Associazione Volontari Italianidel Sangue, AVIS), and to the Head and sta¤ of the Transfusion Unit in The Town�s Hospital, for their preciouscollaboration. Ra¤aella Manna has o¤ered an invaluable contribution to the data collection e¤ort, and Hong PumChung has provided excellent research assistance. This research was conducted in part while Nicola Lacetera was atCase Western Reserve University, and Mario Macis was at the Ross School of Business of the University of Michigan.Financial support from the Ross School of Business Small Grants Research Fund is gratefully acknowledged.

yRotman School of Management, Department of Strategy, and UT-Mississauga, Department of Management.E-mail: [email protected]

zCarey School of Business. E-mail: [email protected]

Page 2: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

1 Introduction

Every year, millions of people devote time, e¤ort, and money to voluntary, pro-social activities such

as donating cash for a cause or organization, helping the elderly and disabled, cleaning beaches, or

donating blood. Yet, for many of these activities, supply is still below societal needs. A notable

example is represented by blood donations. Blood transfusions are required in such situations as

blood loss due to trauma, blood replacement during surgeries, the treatment of premature babies,

and the treatment of several chronic diseases. There is no available substitute for human blood, and,

in recent years, the demand has increased dramatically due to, among other causes, an aging popu-

lation and new surgical procedures such as organ transplants (Di Rado, 2004; www.bloodbook.com).

However, blood supply frequently does not keep pace with demand.1 Neither eligibility criteria nor

a lack of information seem likely to explain these shortages. In fact, while on average, in the US for

example, about 50% of adults are eligible to donate blood, only between 5-10% do so at least once

a year. Also, information campaigns and communications about shortage periods are very often

conducted by such organizations as the Red Cross, government agencies, schools, hospitals, etc.

An alternative cause to be explored concerns people�s behaviors and incentives. Individuals might

simply not �nd it worthwhile to dedicate time to donate blood if the private bene�ts of donating

blood fall short of the opportunity costs. This implies that policies o¤ering explicit incentives to

donate might play a role in encouraging participation in activities that are, in most countries, based

on voluntary and unpaid contributions.

In this paper, we study the e¤ects of Law 584, a legislative provision that gives Italian blood

donors the right to a paid day o¤work on the same day that they donate blood or blood components.

The law applies to all donors who are employed at any private or public organization, and salary

and contributions are reimbursed by the state. Our goal is to evaluate whether this legislative

provision induces blood donors to make more donations, and to quantify this e¤ect.

One would not necessarily expect a paid day o¤ to increase blood donations. In fact, while

standard economic theory would deliver this prediction, research in psychology and more recently,

in economics, argues that extrinsic incentives might crowd out the quantity and quality of the

supplied altruistic activity. For instance, the introduction of economic rewards might create doubts

about the true reason behind pro-social behavior, thus potentially crowding out intrinsic motives.

Economic incentives, therefore, could back�re instead of reinforcing altruistic motivations (Bénabou

and Tirole, 2006; Deci, 1975).2 Given the presence of competing predictions on the impact of

economic incentives on pro-social behavior in general and on blood donation in particular, empirical

examinations are in order.1Shortages are frequent in most Western countries and even more severe in developing nations (World Health

Organization, 2009). It is estimated that, worldwide, there is currently a shortage of about 22 million units of blood(HemoBiotech, 2008).

2These views lend support to the early claims of Titmuss (1971) about the negative e¤ects of paying for blood, interms of both lower quality of blood and lower quantity because people would refrain from a "paid" prosocial activity.Titmuss�1971 book The Gift Relationship was very in�uential in marking the end of a paid blood supply system inthe U.S. in the 1970s (Healy, 2006; Shearmur, 2007).

2

Page 3: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

Our analysis is based on a unique, longitudinal dataset comprising the individual histories of

blood donations of the whole population of donors in an Italian mid-sized town ("The Town"

hereinafter). In addition to demographic information and the number and dates of donations

made by about 2,600 unique donors in the periods 1985-89 and 2002-06, the database includes

information on the donors�labor market status, and occupations. Because our data do not include

the years both before and after the introduction of the policy (Law 584 was passed in 1967), we

cannot evaluate the e¤ects of the policy on the extensive margin of donations (i.e., on inducing

more people to become blood donors). However, the data do allow us to study whether blood

donors are responsive to the paid-day-o¤ incentive. In particular, our empirical strategy exploits

cross-sectional variation in donors�labor market statuses, and, crucially, variation in job-switching

by donors over time.

First, we analyze the extent to which donors who are employees actually take advantage of the

day-o¤ incentive. We do so by looking at actual take-up rates of the bene�t (available only for the

year 2006), and at the choice of the donation day by the donors in our sample. Consistent with

abundant anecdotal evidence, we �nd that donors who are employees are signi�cantly more likely

to donate on Fridays compared to other categories of donors. Because donations can be made only

in the morning in The Town�s hospital, and because most employees do not work on Saturday,

donating on a Friday maximizes the number of consecutive days o¤ (2.5) a donor can enjoy if he

or she takes advantage of Law 584.3 In contrast, individuals who are out of the labor force (i.e.,

not a¤ected by the possibility of a day o¤) tend to distribute their donations uniformly during the

week. The "Friday e¤ect" we detected for employees suggests that a substantial share of this group

of donors exploits the full potential bene�t from the day-o¤ provision. As a further corroboration

of this view, we �nd that the take-up rates of the incentive, as represented by the percentage of

donors who request a document attesting their donation to be presented to their employer, average

70%, with spikes of 80% on Fridays and Mondays. Thus, the vast majority of donors who are

employees does take advantage of the incentive as opposed to, say, donating and returning to work.

Interestingly, about 30% of employees who donate on a Saturday request the day-o¤ document;

therefore, a substantial share of employees donating on Saturday are, indeed, also donating on their

closest workday to the statutory free day (i.e., Sunday).

While all this evidence is consistent with extrinsic motivations a¤ecting blood donation de-

cisions, this does not imply, per se, a positive e¤ect of the day-o¤ provision on the quantity of

supplied blood. In fact, employees might just shift their donation day, or simply take the day o¤

on days on which they would have donated anyway, while keeping the same number of donations

they would have made in the absence of the policy.

Thus, we turn to the core question of the paper: Does the day-o¤ provision of Law 584 induce

blood donors to make more donations? We exploit the facts that the incentive bene�ts only donors

who are employees, that our data include information on donors�occupations, and that a subset of

3A number of studies in human resource management show that workers, in fact, prefer work schedules that allowfor more consecutive days o¤ (Facer and Wadsworth, 2008; Moores, 1990; Pierce and Dunham, 1992).

3

Page 4: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

the donors in our sample change labor market status. The longitudinal nature of our data enables

us to perform regressions of each donor�s donation frequency on his or her labor market status with

individual �xed e¤ects, which absorb any heterogeneity in time-invariant personal attributes that

might be correlated with donation behavior (including "intrinsic" altruism). Our identi�cation

strategy, therefore, uses within-donor occupational shifts to determine whether changing labor

market status is associated with di¤erent donation frequencies. Our �ndings indicate that when

donors are in the "employee" status (i.e., when they are eligible to bene�t from the day-o¤ incentive)

they make on average one extra blood donation per year, a substantial e¤ect that represents a 40%

increase. Notably, our �ndings are unchanged if we focus on the transitions between being an

employee and being out of the labor force (excluding therefore self-employment). Thus we are

able to rule out a potential alternative explanation for our �ndings, i.e. that people donate more

when their opportunity cost of time is lower (as it would happen for an employee as compared

to a self-employed individual), because being out of the labor force arguably leads to even more

available time to donate. We also investigate whether ceasing to be an employee is associated

with a lower donation frequency. Although our point estimates are negative, thus indicating a

reduction in donations when donors cease to bene�t from the day-o¤ incentive, they are (in most

cases) small in magnitude and not statistically signi�cant. This �nding is consistent with some

form of "persistence" or "habit" in behavior whereby those with higher donation frequencies tend

to maintain a high frequency even after they have lower incentives to do so. Our �ndings, �nally,

are robust to a variety of speci�cations and sample selections, aimed at addressing concerns such

as measurement error and serial correlation in labor market status. Our robustness exercises also

limit the concern that changes in labor market status might be capturing other life changes which

might be correlated with donation frequencies.

In addition to providing novel �ndings from a new source of data, our study complements the

existing experimental literature on the impact of explicit incentives on the performance of altruistic

activities in at least two other ways.4 First, the fact that most industrialized countries often make

it illegal to reward blood donations makes it di¢ cult to analyze the issue empirically in the �eld.

To the best of our knowledge, our study is the �rst to analyze the actual behavior of an entire

population of blood donors in response to a naturally occurring incentive. Second, the individuals

in our �eld study are free to not enjoy the economic bene�ts (e.g., by choosing the day of the

week on which to donate, or by returning to work after donating). The fact that the incentive

4Early laboratory experiments by Deci and his collaborators found that adding explicit rewards for the performanceof activities that are motivated by intrinsic reasons leads to a reduction in the performance of those activities (Deci,1975). Similar �ndings have been obtained by, among others, Frey and Oberholzer-Gee (1997) and Gneezy andRustichini (2000). Gneezy and Rustichini, however, �nd that "large enough" incentives do stimulate pro-socialbehavior. With speci�c reference to blood donation, Mellstrom and Johannesson (2008) document that Swedishfemale college students are less willing to undertake a health test to be eligible to donate blood if o¤ered monetaryincentives. In partial contrast to these results, Goette and Stutzer (2008), in a �eld experiment in Switzerland, �ndthat o¤ering lottery tickets to promote donations increases turnout at blood drives. Other studies have focused onother extrinsic motives for pro-social activities, such as the quest for social recognition. See, for example, Andreoniand Bernheim (2009); Ariely, Bracha, and Meier (2008); Harbaugh (1998a, 1998b); Lacetera and Macis (2010a);Nowak and Sigmund (2000); Polborn (2007); Price (2003); and Wedekind (1998).

4

Page 5: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

is naturally occurring and that individuals are free to not bene�t from it reduces concerns about

social desirability bias and limited sorting in the experimental literature (Harrison and List, 2004;

Lazear, Malmendier, and Weber, 2010).

The remainder of the paper is organized as follows. Section 2 describes the institutional context

of this study and the data. Section 3 presents the empirical analysis and �ndings. Section 4 o¤ers

a summary of the �ndings and considerations on their policy implications.

2 Institutional background and data

The data used in this study originate from hand-collected information on the entire blood donation

histories of all donors in an Italian town located in the north-central part of the country.5 Before

describing the data in detail, we provide some background information on the blood donation

system in Italy and in The Town, and we describe the day-o¤ incentive introduced by Law 584.

2.1 Blood donation in Italy and in The Town

Blood donation in Italy is organized through blood banks run by volunteer donor associations. The

associations have a central headquarters as well as town-level units. To donate blood, an individual

is required to become a member of one of these associations. The three major associations, which

are present in di¤erent parts of the country and do not compete with one another, are Associazione

Volontari Italiani del Sangue (AVIS), with about 1.1 million members in 2007, Federazione Italiana

delle Associazioni Donatori di Sangue (FIDAS), with about 400,000 members (Caligaris 2007),

and Fratres, with 150,000 members (in 2000).6 The a¢ liation is to a local unit of the national

associations, and blood donors predominantly donate in the town where "their" unit is located.

In The Town, blood donation is managed by the largest blood donor association, AVIS, and

donations of either whole blood or blood components take place at The Town�s public hospital,

Monday through Saturday from 8 to 11 a.m. Donors do not make appointments, and donate on a

"�rst come, �rst served" basis.

In Italy, only individuals between 18 and 65 years of age are allowed to donate blood. Italian

law limits to the frequency of donations of blood and blood components. Whole blood can be

collected once every 90 days from male donors and (since 1991) once every 180 days from females.7

5The demographic, social, and economic characteristics of The Town�s population are highly representative ofthe overall Italian urban population. Statistics comparing The Town with other Italian towns under a number ofsocio-economic characteristics are available upon request.

6Blood donations run through blood banks and voluntary donor associations (which were present since the 1920s)have become the o¢ cial blood donation and collection system in Italy, after a brief period, following the end of WorldWar II, when the Red Cross played a prominent role. Similar blood bank systems exist in other countries, suchas Denmark, Greece, Norway, Portugal, and Spain. In the U.K., France, and Ireland, in contrast, the organizationof blood donation is run by the State. The Red Cross, �nally, is the dominant organization that manages blooddonation in such countries as Belgium, The Netherlands, Germany, and the U.S. In the U.S., however, the system ismore heterogeneous and competitive, comprising the Red Cross, blood banks, and hospitals directly managing blooddonations. See Healy (2006) on the di¤erent organizational modes of blood donations.

7Donors can give platelets once every 30 days and plasma once every 14 days. Conversations with doctors andAVIS o¢ cials in The Town revealed that the type of donation is "exogenous" to a donor�s choice. Donors, in general,

5

Page 6: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

The time required for a platelet or plasma donation is about one hour, compared to an average

of twenty minutes for a whole blood donation. Including the time to reach the donation site, the

waiting time before the donation and the resting time at the hospital after the donation (which is

higher for whole blood donation), on average, a blood donor should expect a commitment of about

two hours.

2.2 An explicit incentive to blood donors: A one-day paid leave of absence foremployees

According to the National Law 584 of 1967, all donors who are employed at any private or public

organization have the right to a paid day o¤ work on the same day that they donate blood or

blood components. Employers are refunded by the state for the related salary expenses they incur

(including social security and other contributions). From the employee�s standpoint, this provision

is equivalent to sick days in addition to those that he or she has by contract, with an important

di¤erence. Unlike an illness absence, when the employee is required to stay home (a medical

inspector can be sent to check for the sickness claim), this requirement does not hold in the case

of an absence for blood donation.8 Employees in other countries are typically allowed time during

work hours to donate blood, usually without deductions in salary or accrued leave. However, these

provisions just give donors the material time to make their donation, and, typically, the donor has

to return to work after donating. The bene�t from Law 584 goes well beyond giving some rest to

donors after their donation, and, therefore, it represents an explicit reward to blood donors.

The typical work week of an Italian employee is Monday to Friday. Most businesses, as well as

most public workplaces, do not operate on Saturdays. Some exceptions are represented by hospital

doctors, teachers (because in Italian public schools, attended by the near totality of students, only

Sundays are o¤), and some employees in public workplaces such as City Hall clerks in o¢ ces open

to the public. A further notable exception is given by stores, most of which are closed, by law, on

Monday mornings and are open on Saturdays.

2.3 The data

Using both AVIS�s and The Town�s hospital�s archives, we identi�ed all of the Association�s members

(and, therefore, all of The Town�s blood donors) from 1983 to 2006. For each individual donor, we

obtained the entire donation history over this 24-year period. Information on donors includes sex,

age, blood type, and the date when each individual became a donor. Following AVIS�s practices,

in our analysis we consider a donor to be "active" in each year if that donor has made at least one

join the Association to donate whole blood, and are assigned to donating blood components if they are not eligibleto donate whole blood (e.g., if they have insu¢ cient iron in their blood), or if there is some urgent need for a bloodcomponent.

8Also similar to the rules regarding sick days is the absence of a formal requirement to inform the employer inadvance of the intention to donate blood (and to not go to work) by an employee. The employer cannot refuse, inany case, to "concede" the one-day leave. This does not exclude that, on an informal basis, employees can give priornotice to their employers.

6

Page 7: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

donation in the previous two years. Crucially for the goal of this paper, we have information on the

donors�occupational status. AVIS does not update its members�occupational status on a regular

(e.g., yearly) basis. Within the period covered by our data, updates were made in 1985 and 2002.

We therefore limit our analysis to two subperiods, 1985-1989 and 2002-2006, attributing the same

occupational status in the initial year of each period to the following four years. This introduces

the possibility that, for some donors and in some years, occupation is recorded with error. We

show, however, that our results are essentially unchanged when we restrict the analysis only to the

years 1985 and 2002, when occupation is accurately measured.

Table 1 presents descriptive statistics on the donors. The left panel shows data on all active

donors, while the right panel focuses on donors who were active in both periods. The number of

active donors has increased over time, going from 845 in 1985 to 2,332 in 2006, and so has the

fraction of female donors, from 24% in the mid 1980s to 30% in more recent years. The pool of

donors has aged slightly over time, moving from an average age of about 37 to about 40 years. The

average number of donations per year is 1.9 during 1985-89 and 1.8 during 2002-06. In 1985-89,

68% of donors (62% in 2002-06) were employees, 9% (12% in 2002-06) were self-employed, and 23%

(25% in 2002-06) were out of the labor force. Of the 845 donors who were active in 1985-89, 338

(40%) were still active in 2002-06. Reaching the age limit is what mostly explains why donors cease

to donate and hence drop out of our database between 1985-89 and 2002-06. 62% of the donors

who were below the age of 40 in 1989 were still active in 2002 (67% for males and 46% for females).9

Compared to the overall sample of active donors, the 159 donors who were active in both periods

and for whom the data contains complete occupation information (the "panel" donors) have been

members of AVIS for a longer period (7.3 years vs. 5.1 years in 1985-89), and tend to make more

donations per year (2.5 vs. 1.9 in 1985-89). The distribution of these panel donors across labor

market statuses is not too dissimilar from the overall distribution. Finally, only 9% of the donors

in the panel are females, compared to 24% overall in 1985-89, but this discrepancy is mostly due

to patterns in missing occupation information, as shown in Appendix Table 1.10

3 Empirical Analysis

We perform our empirical analysis in two steps. First, we provide evidence consistent with the fact

that blood donors take into account the economic bene�ts and costs from their donation decisions,

9Linear probability and Probit regressions of the probability for a donor who was active in 1985-89 to be stillactive in 2002-06, shown in Appendix Table 3, con�rm this. In particular, controlling for the age of donors in 1989,employment status in 1985-89 does not have a signi�cant e¤ect on the probability of being still active in 2002-06(columns 3 and 6 in Appendix Table 3).10Appendix Table 1 shows descriptive statistics for all of the 338 donors in the panel, as well as separately for

the donors with missing occupation information in at least one of the two periods. The main di¤erence in terms ofobservable characteristics between the samples with complete and incomplete job information is that the latter sampleincludes a larger number of female donors (which explains also the smaller average number of donations per yearand the shorter tenure as blood donors). We conjecture that a majority of female donors who did not indicate theiroccupation might have simply skipped the question because they are are out of the labor force (e.g., housekeepers);however, we chose to conduct our analysis on the observations with hard information only.

7

Page 8: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

and, as standard theories would predict, they try to maximize the di¤erence between the two. We

do so by examining the choice of donation days by donors of di¤erent labor market statuses, and by

looking directly at take-up rates of the day-o¤ bene�t. Our second step, and the core of this study,

is to study whether the Law 584, by increasing the economic bene�ts of donating for employees,

leads to an increase in blood donations.

3.1 Economic considerations, Law 584 and the choice of donation days

Donating on di¤erent days of the week entails di¤erent costs and bene�ts for di¤erent categories

of workers. Donors who are employees would need to ask for time o¤, or they could donate on a

day when they do not work. Law 584, however, gives them the option of taking a day o¤ when

they make a blood donation. This creates di¤erential bene�ts on di¤erent days of the week. Since

donations can be made only in the morning in The Town�s hospital, and because most employees

do not work on Saturday, donating on a Friday (if leisure is taken as a good) maximizes the

number of consecutive days o¤ (2.5) that a donor-employee can enjoy. A number of studies show

that workers, in fact, favor work schedules with more consecutive days o¤ (Facer and Wadsworth,

2008; Moores, 1990; Pierce and Dunham, 1992). Therefore, one would expect donors-employees to

show a preference for donating on Friday if they respond to the incentive in the sense of making

the most out of it. The behavior of this group can be contrasted to that of other categories of

donors, who do not enjoy the day-o¤ bene�t. First, donors who are out of the labor force (e.g.,

students, housekeepers, retirees) should not display strong preferences over di¤erent days of the

week, because costs and bene�ts are very similar. Second, for donors who own and operate a

business ("self-employed" hereinafter), the cost of donating varies across days of the week; they

would incur a cost if they donated on a working day, but not on the day their business is closed.

In particular, also this group should not display any particular preference for donating on Friday.

Figure 1A reports the distribution of donations across the days of the week by donors� labor

market status, pooling the data from the two periods 1985-89 and 2002-06.11 Combined with

data on the actual take-up rates, described below and shown in Figure 1B, these distributions

are consistent with donors taking into account costs and bene�ts considerations when choosing

when to donate, and they indicate that a substantial fraction of donors who are employees takes

advantage of Law 584 to the fullest extent. Figure 1A shows that there are substantial di¤erences

in the distribution of donation days across job categories. For employees, a higher-than-average

number of donations, around 20%, take place on Fridays. The "excess fraction" of Friday donations

(fraction Friday minus 16:70%) is +3:3% for employees, with a statistically signi�cant t-ratio of

8:8, against �0:93% (not statistically di¤erent from zero) for donors out of the labor force, and

a statistically signi�cant �4:8% for the self-employed (t-ratio = �6:3). Donations by the self-11The distribution of donations across days of the week for all donors is virtually identical to that for donors with

non-missing labor market status. In the remainder of the paper, we will limit our attention to donors with non-missinglabor market status information. Also, we replicated all of the analysis performed in this section separately by period(1985-1989 and 2002-2006), and the results are very similar.

8

Page 9: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

employed are strongly clustered on Mondays and Saturdays. Speci�cally, we �nd that store owners�

donations are clustered on Mondays, when stores are closed, and the other employers�donations

are concentrated on Saturdays, when the other businesses are closed. Finally, the donations by

donors who are out of the labor force are essentially uniformly distributed across days of the week.

Our analysis of the choice of donation days by groups of donors based on their labor market

statuses indicates that all categories of donors take into account cost-bene�t considerations when

deciding when to donate. In particular, a substantial fraction of donors-employees shows a prefer-

ence for Friday, which, thanks to the day-o¤ incentive, allows them to maximize the consecutive

number of days o¤ work that they can obtain by donating.12 Together with Friday, Saturday also

emerges as a favorite day by employees. This is not surprising, for at least two reasons. First, it is

possible that for some employees taking a full day o¤ work is too costly (i.e., too disruptive given

their role and responsibilities in the workplace). Second, some donors might prefer not to take

advantage of the day-o¤ bene�t out of a concern that their blood donation might be misinterpreted

as a self-interested act rather than as a purely altruistic act, and therefore they might decide to

donate on a non-workday to keep their signal "clean" (Bénabou and Tirole 2006). However, we have

some evidence that a fraction of the Saturday donations are "equivalent" to the Friday donations,

as we interpreted them above. Such evidence comes from looking at actual take-up rates of the

day-o¤ bene�t. Upon request, The Town�s hospital provides the donors with o¢ cial documentation

that they can then present to their employer to prove that they donated blood so they can actually

enjoy the paid day o¤ (the employer then presents the same document to the state when asking

for reimbursement). The Town�s hospital, unfortunately, does not keep complete records of these

documents. We obtained partial records (about half a year) for 2006, and report the information

in Figure 1B. The �gure shows that, on average, Monday through Friday about 70% of donors-

employees requested o¢ cial documentation to prove that they donated. This indicates that they

actually took the day o¤ rather than donating and then returning to work. Spikes in the take-up

rate (about 80%) are found on Mondays and Fridays. Interestingly, the take-up rate is about 30%

on Saturday; therefore, a non-negligible fraction of donors-employees who choose Saturday also

enjoy the day o¤ and extend their weekend. As reported above, for some categories of workers

Saturday is not a free day, therefore in this sense it is "equivalent" to a Friday for the other salaried

workers. Between the Friday donors and the Saturday donors who request the doctor�s certi�cate,

it would appear that about a quarter of donations by employees result in an extended weekend.

12All of the above �ndings are robust to regression analyses. In regressions reported in Appendix Table 2, weestimate multinomial logit models for which the four outcomes are donating on (1) Monday, (2) Tuesday throughThursday, (3) Friday, and (4) Saturday. The explanatory variables include indicators for the labor market status ofthe donor, as well as controls for demographic characteristics and year e¤ects. The results from Panel A of Table3 con�rm that donors-employees are signi�cantly more likely to donate on Friday than on any other day of theweek (with the exception of Saturday) compared to donors who are out of the labor force as well as compared tothe self-employed. In particular, the odds of an employee donating on Friday rather than on Tuesday-Thursday areabout 1.5 times greater than for a donor who is non-employed and about 2.6 times higher than for donors who areself-employed. We obtain similar results (available upon request) when we look at the propensity of donors to donatethe day right before a series of national holidays (i.e., such days look like Fridays).

9

Page 10: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

An alternative interpretation of the above �ndings is that employees choose to donate on Fridays

(or Saturdays for those who work on that day) in order to reduce disruption for the employer. We

have no speci�c evidence to conclude that the last days of the week are "slower" work days, so that

an absence on those days would cause less disruption. Furthermore, if donors want to minimize the

disruption to their employers, they could donate and then go back to work; however, the high take-

up rates discussed above are inconsistent with this being the case. Finally, if causing less disruption

was the prevalent explanation for the choice of the donation day, rather than the willingness to

fully enjoy the day-o¤, we should not see an increase in the donation frequency, which is, instead,

what the next section of the paper shows.

3.2 Does the day-o¤ incentive lead to more donations?

The previous analysis presented suggestive evidence that donors-employees choose when to donate

in a way that is consistent with exploiting the full potential of the bene�t (i.e., maximizing the

number of consecutive days o¤). However, it is possible that the Law induces donors to change

their preferred donation day, or simply take the day o¤ on days on which they would have donated

anyway, while leaving their total number of donations unchanged. In the remainder of the paper,

we assess whether the incentive actually causes donors who are entitled to the bene�t to make more

donations than they would otherwise.

3.2.1 Descriptive evidence

We begin by asking whether donors who are employees donate more or less frequently, in general,

compared with other donors. Figure 2A shows the distribution of the number of donations per year

by labor market status: employees, self-employed and out-of-labor-force donors. The di¤erences

in the distribution of donation frequencies across these groups are substantial. The percentage of

active donors (i.e., donors who have donated at least once in the previous two years) not making

any donation in a given year is 15% on average, 13% among employees, almost 18% among the

self-employed, and 20% among the non-employed. Similarly, the share of active donors making one

donation is 23% among employees, 27% for the self-employed, and about 30% for the non-employed.

Conversely, while about 44% of employees make 3 or more donations a year, this fraction is just

37% among the self-employed and 35% for those out of the labor force. The �ndings are very

similar when we limit the analysis, in Figure 2B, to male donors only, in order to account for

the fact that female donors have tighter restrictions to the number of donations per year they

can make. Kolmogorov-Smirnov tests con�rm that the distribution of donation frequencies for

employees is signi�cantly di¤erent than that for the self-employed and that for the donors out

of the labor force (p-value < 0.001 for both the overall and the male distributions), while no

statistically signi�cant di¤erence is detected when the distributions of the self-employed and those

out of the labor force are compared to each other (p-value = 0.133 overall, and 0.260 for males).

This descriptive evidence is strongly suggestive that the paid-day-o¤ provision might be inducing

10

Page 11: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

donors-employees to donate more often than they would without the incentive. The regression

analyses that follow will determine the causal nature, and the size, of this e¤ect.

3.2.2 Identi�cation strategy

Assessing the causal relationship between the provisions of Law 584 and an increase in donation

frequency using di¤erent labor market statuses requires that we control for both observable charac-

teristics and unobservable heterogeneity (including "intrinsic" altruism) that might be correlated

with the propensity to donate blood. The richness of our data and, crucially, their longitudinal

nature allows us to do so. In particular, we have multiple observations for each donor; therefore,

we perform donor-�xed-e¤ects regressions, which absorb any unobservable and time-invariant het-

erogeneity. Further, we exploit the fact that a set of donors who were active in the 1985-89 period

are still active in the 2002-06 period, and a subset of these donors changed labor market status in

the second period relative to the �rst. Thus, we use within-donor occupational shifts to determine

whether changing labor market status is associated with di¤erent donation frequencies. In other

words, we implement a di¤erence-in-di¤erences methodology where we use donors who are active

in both periods and have not changed labor market status as the "control" group, and where the

"treatment" consists of donors who switch to and from paid employment. If the day-o¤ incentive

does play a role in the donors�choices of their frequency of donation, we expect to observe, ceteris

paribus, a change in the donation frequency of donors in the "treatment" group compared to donors

in the "control" group. In particular, one should expect to observe an increase in the donation

frequency for the donors who become employees (thus becoming eligible for the day-o¤ bene�t),

and a decrease in the donation frequency for the donors who cease to be employees (thus losing

the day-o¤ bene�t). As with any di¤erence-in-di¤erences estimation, our identifying assumption is

that, in the absence of the policy, the average outcomes for the "treated" (i.e., occupation switch-

ers) and "control" (i.e., those who do not change occupation) groups would have followed similar

paths over time. In the analysis below, we discuss whether this assumption is likely to be satis�ed

in our setting, and perform a series of robustness checks.

3.2.3 Regressions results

Main results The right-hand side panel of Table 1 reports descriptive statistics on the panel of

159 donors who were active in both periods 1985-89 and 2002-06, and for whom we have complete

labor market information. Of these donors, 59 changed labor market status between 1985-89 and

2002-06 (73% of the transitions were between being an employee and being out of the labor force).

As a �rst step toward establishing a causal relationship between the day-o¤ incentive and the

frequency of donations, in columns (1), (2) and (3) of Table 2 we report the results from the

estimates of the following model:

DONATIONSit = �i + �EMPLOY EEit + Xit + "it; (1)

11

Page 12: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

where the number of donations in year t by individual i is regressed on a dummy variable equal

to 1 if a donor is an employee at a given point in time, and 0 otherwise, controlling for period

e¤ects (1985-89 and 2002-06), and four age-group dummies (18-29, 30-39, 40-49 and 50+). All

columns in the table report estimates from linear regressions with individual �xed e¤ects (�i in

model (1) above). The individual �xed-e¤ects, as mentioned above, control for any unobservable,

time-invariant donor characteristics (including "intrinsic" altruism) possibly correlated with the

frequency of donations. The inclusion of period e¤ects ensures that the results are not driven by

common trends,13 and the age-group dummies control for possible e¤ects due to aging. In all of

our regressions, we cluster the standard errors at the level of the individual donor, to account

for potential heteroschedasticity and auto-correlation of the error term within individuals which

could cause the standard errors to be otherwise understated (Bertrand, Du�o and Mullainathan,

2004). The sample includes all donors with valid labor market information in column (1), while it

is limited to the panel of donors active in both periods in column (2). In column (3) and beyond,

we further restrict the analysis to the subsample of donors for whom we have complete information

about their transitions across occupational categories in the two periods. At the bottom of each

column, we report the mean of the dependent variable, for comparison. The coe¢ cient on the

EMPLOY EE indicator is the di¤erence-in-di¤erences estimator, identi�ed, in our �xed-e¤ects

regressions, out of the donors who switch labor market status, transitioning in or out of employee

status. In these speci�cations, the coe¢ cient estimate for � is positive and statistically signi�cant.

The point estimates indicate that when donors are in the "employee" status (i.e., when they are

eligible to bene�t from Law 584) they make, on average, about 0:7 extra donations per year. The

e¤ect is sizeable; it amounts to 30� 37% of the average number of yearly donations by the donors

in the sample.

As mentioned above, if the day-o¤ incentive does a¤ect donor�s behavior, one should expect

donors who become eligible to the day-o¤ bene�t (by becoming a paid employee) to increase their

donations, and donors who lose the bene�t (by ceasing to be employees) to reduce them. Model

(1) implicitly assumes these two e¤ects to be identical in absolute size. However, a reason why one

might not �nd those e¤ects to be symmetric is the possibility that, as found in other contexts (e.g.,

Charness and Gneezy 2009), some form of "persistence" in behavior might exist, whereby those

who were induced to donate more frequently by some incentive maintain a high frequency even

after they lose the incentive. To explore this, we estimate the following variant of model (1) above:

13All of our results are essentially unchanged if we include year �xed e¤ects among the regressors. We chose topresent results without year e¤ects because this makes the interpretation of the coe¢ cient on the 2002-06 perioddummy variable more straightforward, as described in the text.

12

Page 13: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

DONATIONSit = �i + �1NEV ER_EMPLOY EEit �D2002�2006 (2)

+�2EMPLOY EE_TO_OTHERit �D2002�2006+�3OTHER_TO_EMPLOY EEit �D2002�2006+ Xit + "it:

In this exercise, we divide donors into four categories: (a) donors who were employees in both

1985-89 and 2002-06 (the omitted category in model (2)); (b) donors who were non-employees in

both 1985-89 and 2002-06; (c) donors who were non-employees in 1985-89 and became employees in

2002-06; and (d) donors who were employees in 1985-89 and became non-employees in 2002-06. We

de�ne dummy variables for these groups and interact them with the dummy variable for period 2002-

06, D2002�2006 (the direct term D2002�2006 is included in the controls vector X). We perform the

analysis on male and female donors jointly (column 4), as well as separately for males (columns 5 and

6).14 The coe¢ cients of interest are those on the interactions of the 2002-06 dummy and the labor-

market-transition indicators (of course, because our speci�cations include person �xed e¤ects, the

main e¤ects of those dummies are not identi�ed). The coe¢ cient estimates on the 2002-06 dummy

are small, and not statistically di¤erent from zero, which indicates that there was no signi�cant

change in the yearly frequency of donations for the baseline group, which consists of the donors who

were employees in both periods. The coe¢ cient estimates on the (never-an-employee)*(2002-06)

interaction is also very small and not statistically di¤erent from zero. This indicates that donors

who were not employees in both periods also did not experience a change in the number of yearly

donations between the two periods. In both columns (4) and (5), however, the coe¢ cient on the

(other-to-employee)*(2002-06) interaction is estimated to be positive and statistically signi�cant,

indicating that donors who became employees increased their donations compared to donors who

did not change labor market status, net of period, age-group, and individual-speci�c �xed e¤ects.

The magnitude of the estimates indicates that donors who become employees (and hence become

eligible for the paid-day-o¤ incentive), on average, increase their donations by around one extra

donation per year �a 40�45% increase. As for the coe¢ cient on the (employee-to-other)*(2002-06)interaction, the point estimates have a negative sign (in the case of males, this coe¢ cient indicates

that donors who cease to be employees reduce their donations by 0.54 donations a year), although

they fail to attain conventional levels of statistical signi�cance. On the one hand, this result is

consistent with a negative impact of leaving the employee status, all else constant, and therefore

losing the possibility to enjoy the day o¤. On the other hand, the smaller size of the estimated

coe¢ cient can be seen as being consistent with some form of "persistence" or "habit" in behavior,

whereby those with a higher donation frequency tend to maintain it even after they have lower

incentives to do so. Because of the structure of our data, our identi�cation comes from changes in

14Because of the very small size of the female subsample (just 15 individuals), we are not able to obtain meaningfulestimates for this group of donors.

13

Page 14: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

employment status across two periods (1985-89 and 2002-06) that are more than a decade apart.

One might be concerned that any di¤erences across periods that we detect might be the result of

unobservable trends. However, the fact that there is no discernible trend in the donation frequency

of the groups of donors for which there was no change in the incentive (i.e., donors who were eligible

for the incentive throughout the periods and donors who never bene�ted from it) reassures us about

the validity of our identi�cation strategy.

In column (6) we report estimates from a further restricted sample, where we only consider

transitions between being an employee and being out of the labor force (as reported above, these

transitions are 73% of all transitions). We do so because a potential alternative explanation for

the higher donation frequency during the "employee" status is the consideration that people might

donate more when their opportunity cost of time is lower. This would be the case for transitions to

and from self-employment, where donors, plausibly, have a higher cost of time. However, this does

not apply to the transition between being an employee and being out of the labor force (student,

retired, and so on), since in this latter status, if anything, people would have a lower opportunity

cost of time. The estimates in column (6) dissipate the concern that this alternative explanation

explains our results. In fact, the coe¢ cient on the (other-to-employee)*(2002-06) dummy is posi-

tive, statistically signi�cant, and larger than in columns (4) and (5), indicating that becoming an

employee is associated with nearly 1.5 extra donations per year.

Robustness tests We perform a series of tests to corroborate the �ndings reported above. As a

�rst way to probe the robustness of our results, we have estimated model (2) after collapsing the

data into two periods: 1985-89 and 2002-06. In columns (1) and (2) of Table 3, each donor has

only two observations (one per period), and the dependent variable is therefore the average number

of yearly donations made by each donor in each period. We do so because our key explanatory

variables do not actually change from one year to the next but only from one period to the next,

and, as suggested by Bertrand et al. (2004), collapsing the data at the appropriate level helps

obtaining proper estimates and standard errors (clustering the standard errors, which we do above,

is another strategy recommended by Bertrand et al. 2004). Our results are strongly con�rmed in

this robustness exercise, in both magnitude and statistical signi�cance. In column (1), where we

include all donors in the panel, the coe¢ cient on the (other-to-employee)*(2002-06) interaction is

estimated to be about 1, and in column (2), where we limit the sample to males, it is roughly 1.6.

The comparison of these results with the corresponding regressions in Table 2 (i.e., columns (5)

and (6)) shows that the results we obtain are virtually identical.

Our second robustness test addresses potential errors in attributing the same job to donors

for each whole period, while having exact information only on the years 1985 and 2002. Thus, in

columns (2) and (3) of Table 3 (again, on males and females �rst, and then only on males), we use

information for the years 1985 and 2002 only, and the results still obtain. If anything, the e¤ect of

becoming an employee is estimated to be bigger, which might indicate that measurement error in

the occupation variable was biasing our results toward zero.

14

Page 15: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

Additional robustness checks concern the key identifying assumption that the average outcomes

for the "treated" (i.e., occupation switchers) and "control" (i.e., those who do not change occu-

pation) groups would have followed parallel paths over time in the absence of the policy. In fact,

our interpretation of the e¤ect of the day-o¤ law on the frequency of donation is valid provided

that no other factor that is associated with changing occupation is systematically associated with

donation frequency. In particular, although in our regressions we are always including age-group

dummies, it is possible that the dummies associated with "becoming an employee" and "ceasing to

be an employee" are capturing other changes that take place when individuals reach certain ages

(e.g., changes in family structure or in health). This is a possibility because donors who were out

of the labor force and then became employed were, in 1985-89, of relatively young age (29 years

old on average). Conversely, those who were employees and then left the labor force were relatively

older in the �rst period. To lessen this concern, we conducted two checks, reported in columns

(5)-to-(8) of Table 3. In columns (5) and (6), we restrict the sample to donors who were between

the ages of 18 and 32 in 1985-89. Although the sample size is substantially reduced, this restriction

ensures that donors in the "treatment" and "control" groups are relatively homogeneous in terms of

age and ageing. The results con�rm and even strengthen the previous �ndings; they indicate that

becoming an employee is associated with 1.6 additional donations per year (2 in the case of males).

Interestingly, within this sample we obtain that ceasing to be an employee is associated with about

1.5 fewer donations per year (both overall and for males). As a further test, we constructed dummy

variables for the three age categories (18-29, 30-39 and 40-49, setting the latter as the omitted

category) donors in the panel could belong to in 1985-89, we interacted those indicator variables

with the 2002-06 dummy, and included the interactions in the regressions (as before, because our

speci�cations include person �xed e¤ects, the main e¤ects of those age-group dummies are not

separately identi�able). The coe¢ cient on these interactions measure the change in yearly dona-

tion frequency experienced in the period 2002-06 by individuals who were of a certain age group

in the period 1985-89. As can be seen in columns (7) and (8) of Table 3, the coe¢ cient estimates

on these interaction terms, in the full sample as well as limited to males, are not signi�cant, and

they do not change the estimates on the (labor-market-transition)*(2002-06) dummies. This makes

us more con�dent that the coe¢ cients on the (labor-market-transition)*(2002-06) interactions are

capturing the e¤ect of being eligible for the day-o¤ bene�t rather than that of other life changes

associated with ageing.15

15 In the analyses reported above we have chosen to estimate linear regression models because we wanted to accountfor individual heterogeneity by performing �xed e¤ects speci�cations. In Appendix Tables 4A and 4B we check therobustness of our results to estimating Ordered Logit regressions. Column (1) in Table 4A (all donors) and 4B (maledonors) in the Appendix shows that our results are indeed robust to this alternative speci�cation. In columns (2)through (6) we report the estimated marginal e¤ects from the Ordered Logit regressions, which indicate a substantialand statistically signi�cant shift in the distribution of yearly donation frequencies away from low values (zero, oneand two donations) and toward higher values (three and four or more donations) for donors who become employees.

15

Page 16: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

4 Discussion and conclusion

We have analyzed the e¤ects of a legislative provision that grants a one-day paid leave of absence

to Italian blood donors. Our results indicate that the policy induces donors who are employees

to make, on average, one extra blood donation per year, which represents an increase of around

40%. Because we identi�ed this e¤ect by exploiting within-donor variation in occupation, our

�ndings cannot be attributed to unobservable heterogeneity across donors of di¤erent labor market

statuses. The e¤ect was found to be robust to a variety of speci�cations and sample selections,

which corroborate our causal interpretation. We did not �nd the e¤ect to be fully symmetric;

ceteris paribus, donors who become employees make about one extra donation per year, while the

reduction in donation frequency by donors who cease to be employees is small in magnitude and

typically not statistically signi�cant. This result is consistent with "habit formation" in behavior,

whereby those with a higher donation frequency tend to maintain a high frequency even after they

have lower incentives to do so (somewhat similar to what Charness and Gneezy (2009) �nd for

gym attendance). Further indication that the day-o¤ incentive a¤ects donor behavior comes from

our analysis of patterns in the choice of donation day, combined with actual take-up rates. We

documented that a substantial fraction of donors who are employees choose to donate on a day

that extends their weekend (notably Friday), whereas no such preference was found for donors of

a di¤erent labor market status, and we have argued that this is consistent with donors-employees

maximizing the extrinsic, economic bene�t from donating.

The incentive we have analyzed in this paper and its behavioral e¤ects have social welfare

implications in that they impact the voluntary supply of blood. The evidence we presented indicates

that the paid-day-o¤ provision does stimulate more donations. Therefore, removing this policy (a

measure, in fact, recently advocated by the Italian Employers�Association as well as by the Central

Government) would likely result in a reduced number of donations by existing donors. In fact, other

countries have recently introduced similar provisions in related contexts. In the United States, for

instance, a number of states allow certain categories of employees (e.g., public employees) to take

a paid leave of absence for the purpose of being bone marrow or organ donors. Although speci�c

empirical analyses are necessary to determine the e¤ects of those provisions, our results do indicate

that rewarding altruism with paid time o¤ work can be an e¤ective way of stimulating it.

Welfare comparisons of such provisions have to balance the gains from the policy with its costs

to taxpayers. We found that the policy leads to about one extra donation per donor per year. The

state, however, must �nance all donations made by donors-employees, i.e., about three donations

per donor per year. Evaluated at an average labor cost of 139 euros per day (inclusive of social

security and other contributions [ISTAT, 2007]), that one extra blood donation has a cost of around

400 euros related to the day-o¤ incentive. To that, one would then add the production costs for the

additional collected unit such as labor and equipment costs, and the cost incurred to separate the

di¤erent blood components. Figures published by the Italian Health Ministry set the production

costs of one additional unit of whole blood at around 250 euros. Thus, the incentive put in place by

16

Page 17: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

Law 584 would appear to be cost-e¤ective provided that the full social value of one unit of whole

blood is at least about 650 euros.16

This paper also relates to the larger debate on the role of extrinsic incentives in stimulating

pro-social behavior. Our evidence supports a positive role for material rewards, at least in the case

of blood donations, consistent with other studies such as Goette and Stutzer (2008) and Lacetera,

Macis, and Slonim (2009). Other studies �nd that at least some categories of potential donors,

such as females, might react negatively (Mellstrom and Johannesson, 2008). A question would be

whether the results depend on the type of reward. For example, cash was used in the experiment

of Mellstrom and Johannesson (2008), while Lacetera et al. (2009) consider a variety of in-kind

rewards and Goette and Stutzer (2008) use lottery tickets and cholesterol tests. The survey evidence

in Lacetera and Macis (2010b) shows, again in the case of blood donation, donors might be adverse

to cash payments but not to other forms. On the other hand, Heyman and Ariely (2004) �nd that,

while for very small amounts ($1 or less) in-kind payments appear to be preferred to cash, this does

not hold (or is reversed) as soon as the payments are a little larger (e.g. $5).17 In fact, also the

size of the reward might matter. In this sense, a paid day o¤ may be considered a "large enough"

reward (and a step removed from direct cash), and consistent with the evidence of Gneezy and

Rustichini (2000), it generates a positive response by donors. A further feature of the incentive

we analyzed is that, unlike other studies, donors can decide not to take advantage of it (e.g. by

returning to work after donating, or by donating on a non-working day). This allows for di¤erent

tastes and attitudes by di¤erent donors to be satis�ed, thus it can contribute to the success of

this provision. Finally, further research is needed to establish what kind of e¤ects these incentives

have on the extensive margin (i.e., in attracting new donors or inducing existing donors to cease to

donate).

References

[1] Andreoni, J., and Bernheim, B.D., 2009: "Social image and the 50-50 norm: A theoretical and experi-

mental analysis of audience e¤ects," Econometrica, 77, 1607-1636.

[2] Ariely, D., Bracha, A., Meier, S., 2008: "Doing good or doing well? Image motivation and monetary

incentives in behaving prosocially," American Economic Review, 99, 544-555.

[3] Bénabou, R. and Tirole, J., 2006: "Incentives and Prosocial Behavior," American Economic Review,

96, 5, 1652-1678.

16Estimating the social bene�t from one unit of blood is a much more complicated exercise to perform, however,and one that is beyond the scope of this paper. In fact, from one unit of blood collected, a full unit of red cellsand several partial units of plasma, platelets, and cryoprecipitate are typically derived, which can be used on oneor multiple patients. To obtain an estimate of the social bene�t one would need to estimate the expected impact ofthe transfusion of each blood component on the life expectancy of the patients multiplied by the value of those extrayears of life.17A related study by Kube, Maréchal and Puppe (2010) �nds that, while in-kind rewards seem to work better for

certain tasks, when the experimental subjects are given the choice between a cash payment and a cash-equivalentin-kind gift, they overwhelmingly choose cash.

17

Page 18: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

[4] Bertrand, M., Du�o, E. and Mullainathan, S., 2004: "How Much Should We Trust Di¤erences-in-

Di¤erences Estimates?," Quarterly Journal of Economics 119, 249�275.

[5] Caligaris, A.O., 2007: "Relazione del Presidente," 46th National Annual Conference, FIDAS.

[6] Charness, G. and Gneezy, U., 2009: "Incentives to Exercise," Econometrica, 77, 3, 909-931.

[7] Deci, E.L., 1975: Intrinsic Motivation, Plenum Press, New York.

[8] Di Rado, A., 2004: "All You Need Is Blood," USC Health Magazine, Fall issue.

[9] Facer, R.L, II, and Wadsworth, L., 2008: "Alternative Work Schedules and Work-Family Balance : A

Research Note," Review of Public Personnel Administration, 28, 166-177.

[10] Frey, B.S., and Oberholzer-Gee, F., 1997: "The Cost of Price Incentives: An Empirical Analysis of

Motivation Crowding-Out," American Economic Review, 87, 4, 746-755.

[11] Gneezy, U., and Rustichini, A., 2000: "Pay Enough or Don�t Pay at All," Quarterly Journal of Eco-

nomics, 776,791-810.

[12] Goette, L., and Stutzer, A., 2008: "Blood Donation and Incentives: Evidence from a Field Experiment,"

IZA Working Paper 3580.

[13] Harbaugh, W.T., 1998a: "What do Donations Buy? A Model of Philanthropy Based on Prestige and

Warm Glow," Journal of Public Economics, 67, 269-284.

[14] Harbaugh, W.T., 1998b: "The Prestige Motive for Charitable Transfers," American Economic Review,

88, 2, 277-282.

[15] Harrison, G. W., and List, J.A., 2004: "Naturally Occurring Markets and Exogenous Laboratory

Experiments: A Case Study of the Winner�s Curse," Economic Journal, 118, 822-843.

[16] Healy, K., 2006: Last Best Gifts. Altruism and the Market for Human Blood and Organs, University Of

Chicago Press, Chicago.

[17] Hemobiotech, 2008: http://www.hemobiotech.com, site last visited on October 7th, 2008.

[18] Heyman, J., and Ariely, D., 2004: "E¤ort for Payment: A Tale of Two Markets," Psychological Science,

15, 11, 787-793.

[19] ISTAT, 2007: "Struttura Costo del Lavoro in Italia," www.istat.it.

[20] Kube, S., Maréchal , M.A. and Puppe, C., 2010: "The Currency of Reciprocity. Gift-Exchange in the

Workplace," University of Zurich.

[21] Lacetera, N. and Macis, M., 2010a: "Social Image Concerns and Pro-Social Behavior: Field Evidence

from a Nonlinear Incentive Scheme," Journal of Economic Behavior and Organization 6, 225�237.

18

Page 19: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

[22] Lacetera, N., and Macis, M., 2010b: "Do All Material Incentives for Pro-Social Activities Back�re?

The Response to Cash and Non-Cash Incentives for Blood Donations, Journal of Economic Psychology

31, 738-748.

[23] Lacetera, N., Macis, M., and Slonim, R., 2009: "Will there be Blood? Incentives and Substitution

E¤ects in Pro-Social Behavior", IZA DP 4567.

[24] Lazear, E., Malmendier, U., and Weber, R., 2010: "Sorting, Prices, and Social Preferences," Working

Paper.

[25] Levitt, S.D., and List, A., 2007: "What Do Laboratory Experiments Measuring Social Preferences Tell

Us about the Real World," Journal of Economic Perspectives, 117, 522, 1051-1072.

[26] Mellstrom, C.,and Johannesson, M., 2008: "Crowding Out in Blood Donation: Was Titmuss Right?,"

Journal of the European Economic Association, 6, 4, 845-63.

[27] Moores, J.M., 1990: "A Meta-Analysis of the E¤ects of Compressed Work Weeks," Applied HRM

Research, 1, 1, 8-14.

[28] Nowak, M.A., and Sigmund, K., 2000: "Shrewd Investments," Science, 288, 5467, 819-820

[29] Polborn, M., 2007: "Competing for Recognition through Public Good Provision," CESifo working

paper.

[30] Pierce, J.L. and Dunham, R.B., 1992: "The 12-Hour Work Day: A 48-Hour, Eight-Day Week," The

Academy of Management Journal, 35, 5, 1086-98.

[31] Price, M.K., 2003: "Pro-Community Altruism and Social Status in a Shuar Village," Human Nature,

14, 2, 191-195.

[32] Shearmur, J., 2007: "The Real Body, Part I: Blood and Corpses," Policy, 23, 4, 28-32.

[33] Titmuss, R.M., 1971: The Gift Relationship, London: Allen and Unwin.

[34] Wedekind, K., 1998: "Give and Ye Shall Be Recognized," Science, 280, 5372, 2070-2071.

[35] World Health Organization, 2009: Global Blood Safety and Availability �Facts and Figures from the

2007 Blood Safety Survey.

19

Page 20: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

20

Table 1: Summary statistics

The left panel reports statistics on all active donors in “The Town”, where a donor is considered active at a given point in time if she

donated at least once in the previous two years. The right panel reports statistics on the donors who were active in both periods and

with complete labor market status information. When appropriate, standard deviations are reported below the corresponding mean in

parenthesis.

1985-89 2002-06 1985-89 2002-06

Mean Mean Mean Mean

Females 0.24 0.30 0.09 0.09

Age 36.87 39.5 35.18 50.06

(10.82) (10.74) (8.06) (8.09)

Blood type 0 0.52 0.49 0.50 0.51

Blood type A 0.38 0.39 0.45 0.43

Blood type B 0.07 0.09 0.05 0.05

Blood type AB 0.03 0.03 0.00 0.01

Years since joined AVIS 5.05 8.06 7.31 22.97

(6.28) (7.89) (5.84) (6.08)

N donations/year 1.91 1.76 2.54 2.42

(1.35) (1.62) (1.20) (1.84)

Employees 0.68 0.62 0.72 0.73

Self-employed 0.09 0.12 0.11 0.10

Out of the labor force 0.23 0.25 0.16 0.17

N donors 845 2,332 159 159

N donor-year observations 3,241 8,500 662 713

All active donorsPanel of donors

active in both periods

Page 21: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

21

Table 2: Labor market status and frequency of donation: Regressions results.

All the columns report results of individual fixed-effects regressions where an observation is a donor/year. In (1), the sample includes

all donors with valid labor market information. In (2) through (6), the sample includes donors who were active in both 1985-89 and

2002-06. Columns (3) through (6) include panel donors with complete labor market information in both periods. In these columns,

the omitted "job transition" category consists of donors who were employees in both 1985-89 and 2002-06. Controls (not reported in

the table) include age dummies (18-29, 30-39, 40-49 and 50+). Standard errors, clustered at the level of the individual donor, are

reported in parentheses. *** p<0.01, ** p<0.05, * p<0.1

Dependent variable: Number of donations/year

Sample:All active

donors

Subsample:

All with

complete

transition info

All with

complete

transition info

Males with

complete

transition info

Males with

complete

transition info

(1) (2) (3) (4) (5) (6)

2002-06 dummy 0.10 -0.14 -0.10 -0.19 0.01 -0.25

(0.17) (0.22) (0.24) (0.27) (0.28) (0.31)

Employee dummy 0.73*** 0.68*** 0.68***

(0.27) (0.25) (0.24)

(Never an Employee)*(2002-06) -0.18 -0.31 0.12

(0.34) (0.38) (0.60)

(Employee-to-Other)*(2002-06) -0.41 -0.54

(0.35) (0.35)

(Other-to-Employee)*(2002-06) 0.96** 1.09**

(0.45) (0.50)

(Employee-to-Out of Lab Force)*(2002-06) -0.35

(0.43)

(Out of Lab Force-to-Employee)*(2002-06) 1.46**

(0.66)

Mean of Dep. Var. 1.70 2.15 2.39 2.39 2.45 2.47

Donor-Year Observations 8,138 1,901 1,375 1,375 1,252 1,047

N of donors (FE) 1,879 289 159 159 144 120

Adj. R-squared 0.50 0.38 0.38 0.38 0.39 0.40

Donors active in both 1985-89 and 2002-06

Page 22: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

22

Table 3: Robustness tests

All columns report results of individual fixed-effects regressions. In (1) and (2), an observation is a donor-period and the dependent

variable is the average yearly number of donations made by each donor in each period (1985-89 and 2002-06). In the remaining

columns, an observation is a donor-year and the dependent variable is the number of donations per year. In columns (3) and (4), the

sample is limited to years 1985 and 2002 only. In the other columns, all years in the periods 1985-89 and 2002-06 are used. Controls

(not reported in the table except for columns (7) and (8)) include age dummies (18-29, 30-39, 40-49 and 50+). The omitted "job

transition" category consists of donors who were employees in both 1985-89 and 2002-06. Standard errors, clustered at the level of

the individual donor, are reported in parentheses. *** p<0.01, ** p<0.05, * p<0.1

Dependent variable:

Sample:

Subsample: All donors Males All donors Males All donors Males All donors Males

(1) (2) (3) (4) (5) (6) (7) (8)

2002-2006 dummy -1.07** -0.81 0.58 0.58 -0.27 -0.06 -0.36 -0.16

(0.50) (0.54) (0.75) (0.74) (0.41) (0.41) (0.358) (0.375)

(Never an Employee)*(2002-06) -0.07 -0.24 -0.35 -0.30 0.30 -0.12 -0.13 -0.31

(0.33) (0.36) (0.70) (0.71) (0.34) (0.36) (0.337) (0.380)

(Employee-to-Other)*(2002-06) -0.26 -0.43 0.10 -0.17 -1.48*** -1.52*** -0.22 -0.36

(0.36) (0.36) (0.73) (0.73) (0.29) (0.29) (0.357) (0.358)

(Other-to-Employee)*(2002-06) 1.01** 1.16** 2.21** 2.14** 1.62*** 2.04*** 0.93** 1.07**

(0.43) (0.48) (0.92) (0.95) (0.33) (0.36) (0.455) (0.504)

(Age 18-29 in 1985-89)*(2002-06) 0.52 0.41

(0.502) (0.532)

(Age 30-49 in 1985-89)*(2002-06) 0.41 0.36

(0.364) (0.385)

Mean of Dep. Var. 2.32 2.38 2.63 2.64 2.28 2.39 2.39 2.45

Donor-Year Observations 318 288 274 250 423 372 1,358 1,241

N of donors (FE) 159 144 158 143 52 45 156 142

Adj. R-squared 0.19 0.20 0.03 0.04 0.51 0.55 0.38 0.39

Donors active in both

1985-89 and 2002-06 1985 and 2002 onlyDonors active in both

1985-89 and 2002-06

Donors active in both

1985-89 and 2002-06

who were of ages

18-32 in 1985-89

Average N. of yearly

donations

in periods 1985-89

and 2002-06

Number of donations/year

Page 23: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

23

Figure 1A: Distribution of donations across days of the week, by labor market status.

Donations of all active donors with positive donations, pooled 1985-89 and 2002-06 years. Each line represents the distribution of

donations across days of the week, separately for each group of donors according to their labor market status (employee, self-

employed, out of labor force). N. of donations by employees = 11,358; N. of donations by self-employed = 1,766; N. of donations by

donors who are out of the labor force = 3,494.

Figure 1B: Take-up rates of the day-off benefit, by day of the week (year 2006)

The figure reports, for each day of the week, the percentage of employees who request the official document, signed by the doctor who performed the blood draw, to be presented to the employers as proof of the blood donation to justify the day off. N = 630.

0

5

10

15

20

25

30

Monday Tuesday Wednesday Thursday Friday Saturday

Fre

quen

cy d

istr

ibuti

on

(%

)

Self-Employed Employees Out of Labor Force

0

10

20

30

40

50

60

70

80

90

100

Monday Tuesday Wednesday Thursday Friday Saturday

%

Page 24: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

24

Figure 2A: Distribution of the number of donations per year, by labor market status.

All active donors, pooled 1985-89 and 2002-06 years. Each line represents the distribution of the number of donations per year,

separately for each group of donors according to their labor market status (employee, self-employed, out of labor force). N. of

person-year observations for employees = 5,197; N. of person-year observations for self-employed = 930; N. of person-year

observations for donors out of the labor force = 2,035.

Figure 2B: Distribution of the number of donations per year, by labor market status,

male donors only.

Same as Figure 2A, limited to the sample of male donors. N. of person-year observations for employees = 4,188; N. of person-year

observations for self-employed = 759; N. of person-year observations for donors out of the labor force = 1,052.

5

10

15

20

25

30

0 1 2 3 4 or more

Fre

quen

cy d

istr

ibuti

on

(%

)

Number of donations per year

Self-employed Out of Labor Force Employees

5

10

15

20

25

30

0 1 2 3 4 or more

Fre

quen

cy d

istr

ibuti

on

(%

)

Number of donations per year

Self-employed Out of Labor Force Employees

Page 25: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

Time for blood:

The effect of a naturally-occurring incentive for altruistic behavior

Appendix Tables

Page 26: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

i

Appendix Table 1: Comparison of full panel and panel of donors with missing occupation

The left panel reports statistics on donors who were active in both periods (1985-89 and 2002-06) whereas the right panel reports

statistics on donors who were active in both periods but with missing occupation. When appropriate, standard deviations are reported

below the corresponding mean in parenthesis.

1985-89 2002-06 1985-89 2002-06

Mean Mean Mean Mean

Females 0.16 0.16 0.22 0.22

Age 34.15 48.91 33.08 47.64

(8.24) (8.57) (8.30) (8.90)

Blood type 0 48.98 51.10 47.74 50.76

Blood type A 43.75 41.52 42.62 39.70

Blood type B 4.92 4.75 5.27 4.85

Blood type AB 2.35 2.63 4.37 4.70

Years since joined AVIS 6.19 21.78 4.82 20.40

(6.64) (6.68) (7.29) (7.07)

N donations/year 2.32 2.12 2.09 1.80

(1.24) (1.79) (1.23) (1.67)

N donors 338 338 179 179

N donor-year observations 1,331 1,378 669 665

Panel of donors

active in both periods

- Full sample -

Panel of donors

active in both periods

- Donors with missing

occupation information -

Page 27: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

ii

Appendix Table 2: Donation days and occupation: Multinomial Logits results

In each panel, the first column reports the raw coefficients, the second the z-score for the test of the hypothesis that the coefficient is equal to zero, the third the associated p-value. The fourth column reports the factor change in odds of alternative 1 relative to alternative 2 for a unit increase in the independent variable. Controls include sex, age dummies, period fixed effects and year fixed effects.

Dependent Variable: Day of donation

Employees vs. Out of Labor Force

Alternative 1 Alternative 2 Coefficient Z P>|z| e^b

Friday Monday 0.27 3.44 0.00 1.31

Friday Tuesday-Thursday 0.38 5.95 0.00 1.46

Friday Saturday -0.03 -0.38 0.70 0.97

Alternative 1 Alternative 2 Coefficient Z P>|z| e^b

Friday Monday 0.94 10.21 0.00 2.56

Friday Tuesday-Thursday 0.38 4.62 0.00 1.47

Friday Saturday 0.82 9.26 0.00 2.26

Observations: 16,578

Wald chi2(p>Chi2) 426.29 (0.000)

Odds comparing alternative 1 to alternative 2

Employees vs. Self-employed

Page 28: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

iii

Appendix Table 3: Predicting the survival of donors in the panel

The sample includes all donors who were active in 1985-89. The dependent variable is equal to 1 if a donor was still active in 2002-06

and 0 otherwise. For the Probit regressions, marginal effects are reported. Standard errors, clustered at the level of the individual

donor, are reported in parentheses. *** p<0.01, ** p<0.05, * p<0.1

Dependent Variable = 1 if donor still active in 2002-2006, 0 otherwise

Sample:

Method:

(1) (2) (3) (4) (5) (6)

Female -0.14*** -0.18*** -0.17*** -0.15*** -0.20*** -0.19***

(0.048) (0.056) (0.055) (0.054) (0.064) (0.065)

Age 18-29 0.10** 0.12** 0.10* 0.11** 0.13** 0.11*

(0.048) (0.056) (0.056) (0.052) (0.058) (0.060)

Age 30-49 -0.10* -0.09 -0.11* -0.10* -0.09 -0.12*

(0.055) (0.064) (0.064) (0.055) (0.066) (0.068)

Age 50+ -0.57*** -0.55*** -0.60*** -0.59***

(0.037) (0.039) (0.041) (0.030)

Employee -0.03 -0.05 -0.03 -0.05

(0.047) (0.047) (0.056) (0.056)

N. donations per year 0.09*** 0.10***

(0.021) (0.026)

Observations 596 459 459 596 412 412

Adjusted R-squared 0.15 0.16 0.19

Pseudo R-squared 0.14 0.04 0.07

Donors who were active in 1985-1989

OLS Probit

Page 29: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

iv

Appendix Table 4A: Labor market status and frequency of donation: Ordered Logits results

(Males and Females)

The table reports coefficients (column (1)) and marginal effects (columns 2 through 6) from an Ordered Logit regression. An

observation is a donor-year. The dependent variable is the number of donations per year, and can take five outcomes: 0 donations, 1

donation, 2 donations, 3 donations and 4 (or more) donations per year. Standard errors, clustered at the level of the individual donor,

are reported in parentheses. *** p<0.01, ** p<0.05, * p<0.1

Dependent Variable: Number of donations/year

Sample:

Subsample:

Method:

Coeffs.

N.don.=0 N.don.=1 N.don.=2 N.don.=3 N.don.=4+

(1) (2) (3) (4) (5) (6)

Female -0.65** 0.07* 0.07** 0.02*** -0.07* -0.09***

(0.301) (0.037) (0.033) (0.007) (0.034) (0.035)

2002-06 dummy -0.70 0.06 0.07 0.04 -0.06 -0.12

(0.448) (0.037) (0.046) (0.028) (0.035) (0.075)

Never an Employee -0.37 0.03 0.04 0.02* -0.04 -0.06

(0.318) (0.032) (0.035) (0.010) (0.034) (0.043)

Employee-to-Other -0.31 0.03 0.03 0.02 -0.03 -0.05

(0.372) (0.036) (0.040) (0.015) (0.037) (0.054)

Other-to-Employee -0.48 0.04 0.05 0.02** -0.05 -0.07*

(0.304) (0.032) (0.034) (0.009) (0.033) (0.041)

(Never an Employee)*(2002-06) 0.52 -0.03 -0.05 -0.04 0.03** 0.10

(0.427) (0.023) (0.038) (0.042) (0.015) (0.089)

(Employee-to-Other)*(2002-06) -0.24 0.02 0.03 0.01 -0.02 -0.04

(0.427) (0.040) (0.047) (0.018) (0.043) (0.062)

(Other-to-Employee)*(2002-06) 0.87** -0.05*** -0.08*** -0.08* 0.04*** 0.17*

(0.395) (0.019) (0.030) (0.041) (0.011) (0.090)

Age 18-29 in 1985-89 -0.06 0.00 0.01 0.00 -0.01 -0.01

(0.432) (0.035) (0.046) (0.027) (0.038) (0.070)

Age 30-49 in 1985-89 0.21 -0.02 -0.02 -0.01 0.02 0.04

(0.389) (0.030) (0.040) (0.026) (0.031) (0.066)

(Age 18-29 in 1985-89)*(2002-06) 0.03 -0.00 -0.00 -0.00 0.00 0.00

(0.555) (0.044) (0.058) (0.036) (0.046) (0.092)

(Age 30-49 in 1985-89)*(2002-06) -0.04 0.00 0.00 0.00 -0.00 -0.01

(0.440) (0.036) (0.047) (0.027) (0.038) (0.071)

Observations 1,358

Pseudo R-squared 0.016

Ordered Logit

Donors active in both 1985-89 and 2002-06 with complete labor market

transition information

Marginal effects

All donors

Page 30: Time for blood: The e⁄ect of a naturally-occurring ... · The e⁄ect of a naturally-occurring incentive for altruistic behavior Nicola Laceteray University of Toronto Mario Macisz

v

Appendix Table 4B: Labor market status and frequency of donation: Ordered Logits results

(Males)

The table reports coefficients (column (1)) and marginal effects (columns 2 through 6) from an Ordered Logit regression. An

observation is a donor-year. The dependent variable is the number of donations per year, and can take five outcomes: 0 donations, 1

donation, 2 donations, 3 donations and 4 (or more) donations per year. Standard errors, clustered at the level of the individual donor,

are reported in parentheses. *** p<0.01, ** p<0.05, * p<0.1

Dependent Variable: Number of donations/year

Sample:

Subsample:

Method:

Coeffs.

N.don.=0 N.don.=1 N.don.=2 N.don.=3 N.don.=4+

(1) (2) (3) (4) (5) (6)

2002-06 dummy -0.51 0.04 0.05 0.04 -0.04 -0.09

(0.465) (0.038) (0.045) (0.032) (0.036) (0.079)

Never an Employee -0.50 0.05 0.05 0.02** -0.05 -0.07*

(0.312) (0.035) (0.033) (0.009) (0.034) (0.041)

Employee-to-Other -0.22 0.02 0.02 0.01 -0.02 -0.03

(0.367) (0.033) (0.037) (0.021) (0.034) (0.057)

Other-to-Employee -0.57* 0.05 0.06 0.03*** -0.05 -0.09*

(0.336) (0.037) (0.035) (0.010) (0.036) (0.044)

(Never an Employee)*(2002-06) 0.37 -0.03 -0.03 -0.03 0.02 0.07

(0.504) (0.030) (0.044) (0.049) (0.022) (0.101)

(Employee-to-Other)*(2002-06) -0.40 0.04 0.04 0.02 -0.04 -0.06

(0.426) (0.044) (0.045) (0.016) (0.045) (0.059)

(Other-to-Employee)*(2002-06) 1.02** -0.06*** -0.08*** -0.10** 0.03* 0.21**

(0.428) (0.019) (0.028) (0.045) (0.018) (0.102)

Age 18-29 in 1985-89 0.18 -0.01 -0.02 -0.01 0.01 0.03

(0.450) (0.034) (0.044) (0.034) (0.033) (0.079)

Age 30-49 in 1985-89 0.30 -0.02 -0.03 -0.02 0.02 0.05

(0.399) (0.031) (0.039) (0.030) (0.030) (0.070)

(Age 18-29 in 1985-89)*(2002-06) -0.20 0.02 0.02 0.01 -0.02 -0.03

(0.584) (0.052) (0.060) (0.033) (0.054) (0.091)

(Age 30-49 in 1985-89)*(2002-06) -0.14 0.01 0.01 0.01 -0.01 -0.02

(0.464) (0.040) (0.047) (0.029) (0.041) (0.075)

Observations 1,241

Pseudo R-squared 0.014

Donors active in both 1985-89 and 2002-06 with complete labor market

transition information

Males

Ordered Logit

Marginal effects