Discussion Paper
No.1778
June 2021
Marriage as insurance:
job protection and job
insecurity in France
Andrew E. Clark
Conchita D'Ambrosio
Anthony Lepinteur
ISSN 2042-2695
Abstract
Job insecurity is one of the risks that workers face on the labour market. As with any risk, individuals
can choose to insure against it. We here consider marriage as a way of insuring against labour-market
risk. The 1999 rise in the French Delalande tax, paid by large private firms when they laid off workers
aged 50 or over, led to an exogenous rise in job insecurity for the uncovered (younger workers) in the
affected firms. A difference-in-differences analysis using French panel data reveals that this greater job
insecurity for the under-50s led to a significant rise in their probability of marriage, and especially when
the partner had greater job security, consistent with marriage providing insurance against labour-market
risk.
Key words: marriage, insurance, employment protection, perceived job security, difference-in-differences
JEL codes: I38; J13; J18
This paper was produced as part of the Centre’s Community Wellbeing Programme. The Centre for
Economic Performance is financed by the Economic and Social Research Council.
We would like to thank Petri Böckerman, Arnaud Chevalier, Thomas Dohmen, Markus Gebauer, Kevin
Lang, Marion Leturcq, Giorgia Menta, Anne Solaz and seminar participants at the IZA/HSE seminar
for their help and feedback. We gratefully acknowledge financial support from CEPREMAP, the Fonds
National de la Recherche Luxembourg (Grant C18/SC/12677653) and EUR grant ANR-17-EURE-
0001.
Andrew E. Clark, Paris School of Economics and Centre for Economic Performance, London
School of Economics. Conchita D’Ambrosio, University of Luxembourg. Anthony Lepinteur,
University of Luxembourg
Published by
Centre for Economic Performance
London School of Economics and Political Science
Houghton Street
London WC2A 2AE
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or
transmitted in any form or by any means without the prior permission in writing of the publisher nor be
issued to the public or circulated in any form other than that in which it is published.
Requests for permission to reproduce any article or part of the Working Paper should be sent to the
editor at the above address.
A.E. Clark, C. D’Ambrosio and A. Lepinteur, submitted 2021.
1
1. Introduction
Why do people get married? Weiss (1997) suggests that marriage comes with a number of
economic advantages. The first reflects the benefits of specialisation between spouses (Becker,
1973 and 1981). Second, in a context of imperfect credit markets, marriage may also relax
credit constraints via implicit credit arrangements within households (Borenstein and Courant,
1989) and enhance investment (for example, one partner may work while the other is in
education investing in their human capital). Collective and non-rival goods are also jointly
produced and consumed within partnerships, with common examples being children or
housework (Chiappori, 1992, and Van Klaveren et al., 2008).
Stevenson and Wolfers (2007), in their empirical review of changing trends in marriage
and divorce in the US, highlight the roles of pre-marital cohabitation (which has risen),
specialisation in marriage (now argued to be less important), the tax implications of
partnership, birth control, changes in relative wages, Divorce Laws, and the marriage
“matching function” (via education, the workplace and the internet).
The benefit from marriage that we will address here, which appears in both Weiss (1997)
and Stevenson and Wolfers (2007), is that of risk-sharing. As noted by Hess (2004) and Shore
(2010), partnerships provide insurance by allowing couples to diversify risk, if the exogenous
income shocks that the two partners receive are not perfectly (positively) correlated. Couples
can in addition adjust their relative labour supply to reduce the impact of shocks.
A number of contributions have provided indirect evidence that is consistent with the
insurance role of marriage. In Rosenzweig and Stark (1989), arranged marriages between
families from different villages in South India significantly reduced the variability in food
consumption. Farm households with greater income risks were more likely to engage in
arranged marriages at longer distances. In US data, Halla and Scharler (2011) find that the
influence of idiosyncratic output-growth shocks on consumption is smaller in States where the
2
percentage of married is higher. Bertocchi et al. (2011) consider household investment
decisions, and conclude from their analysis of 14 years of Italian SHIW data that the married
invest more in risky assets (as marriage is considered to be a safe asset). Anderson and Ray
(2019) show that marriage protects against the risk of death, especially for women. In a similar
vein, Van den Berg and Gupta (2015) use individual data from Dutch registers (from 1815 to
2000) and find a protective effect of marriage against mortality for men.
With respect to the labour market, there is a considerable literature on marriage and the
business cycle. Schaller (2013) analyses 32 years of US State-level panel data: marriage is
shown to be pro-cyclical, with unemployment being associated with less marriage. Lichter et
al. (2006) appeal to individual-level NLSY-79 data, and conclude that the probability of
transition from cohabitation to marriage rises with partner’s education and the partner working;
it is lower for the unemployed. Education and employment are found to have a similar influence
on marriage probability in Chinese data (Yu and Xie, 2015). Early labour-market experiences
also seem to matter. Ekert-Jaffé and Solaz (2001 and 2002) and Landaud (2019) analyse
different French datasets to show that early-career unemployment and temporary jobs reduce
the probability of forming a couple; De La Rica and Iza (2005) come to similar conclusions
using Spanish data. One interpretation is that unemployment provides a negative signal about
the potential partner’s unobserved characteristics. Consistent with this interpretation, in
Charles and Stephens (2004), job loss increased the risk of divorce when resulting from layoff
but not as a result of plant closings: they note that the former may convey information about
the partner’s noneconomic suitability as a mate.
Some work has explicitly looked not at events that have already occurred, but rather future
risk on the labour market. Schneider et al. (2018) discuss the role of economic resources in
marriage, which they extend to include wealth and expected future earnings. Schneider and
Reich (2018) continue in the same line, and find that union membership (as an indicator of
3
economic security) is a predictor of marriage in NLSY-79 data. Xie et al. (2003) are also
individual-level and forward-looking, and calculate five different measures of current and
future “earnings potential”: they show that all five are positively correlated with the transition
from cohabitation to marriage using US Census and cohort data.
We also consider labour-market risk, but here with respect to job loss. The role of forward-
looking job insecurity in partnership has already been mentioned: Schneider et al. (2018)
explicitly cite job security as a component of future earnings, but note that it is difficult to pin
down exogenous changes in resources, and in particular that job security is not observed in
their PSID data. However, we are not aware of any work that has been able to appeal to
exogenous variation in job insecurity at the individual level in this context. We here propose a
difference-in-differences analysis from a quasi-natural experiment to establish a causal effect
of individual job insecurity on the probability of marriage. As opposed to past individual
unemployment or current macroeconomic shocks, we are able to identify a plausible exogenous
change in job insecurity: our analysis sample consists of workers, some of whom became more
at risk than others of future job loss following a French labour-market reform.
As in Clark and Lepinteur (2021), we use 1994-2001 French data from the European
Household Community Panel (ECHP) to show that the 1999 rise in the French Delalande tax
on the layoffs of those aged 50+ in large firms increased job insecurity for the under-50s in
these firms (as compared to the under-50s in small firms, where the Delalande tax did not
change in 1999). We are interested in the way in which this exogenous greater risk affects
behaviour. Clark and Lepinteur (2021) suggest that one reaction is to reduce risk exposure, via
lower fertility; we instead here ask whether individuals will take out more insurance against
risk, via their marital status.
1
1
The analysis in Clark and Lepinteur (2020) does not use the same sample as we do here. Their fertility analysis
is carried out on the sample of workers who were already married before the reform and continued to be married
after it. On the contrary, we explicitly model moves between marital statuses.
4
We conclude that greater job insecurity amongst French workers increased women’s
probability of marriage by four percentage points; there is no effect for men. This is consistent
with evidence on gender differences in preferences (Croson and Gneezy, 2009): women are
generally less willing to take risks in the context of lotteries (Hartog et al., 2002; Holt and
Laury, 2002; Fehr-Duda et al., 2006) and portfolio selection (Sunden and Surette, 1998;
Finucane et al., 2000; Charness and Gneezy, 2012). Dohmen et al. (2018) measure risk
aversion by both a self-assessed measure and revealed preference from a series of binary
choices between a lottery and a sure return. Women were found to be more risk averse than
men in almost all of the 76 countries for which they collected data (see their Figure III). As
such women may react more to the threat of future job loss than do men.
We check that our estimates are robust to a number of potential confounding factors, such
as European macroeconomic trends and the French 35-hour workweek that was announced in
1998 and introduced in 2000; they are also qualitatively similar using a number of different
estimation methods. While we do identify an insecurity effect on marriage, there is no change
in the probability of entering a partnership in general, or indeed of leaving one: the greater
probability of marriage for women then mostly reflects a shift into marriage from pre-reform
cohabitation. The effect of job insecurity on women’s marriage probability is the same by age,
education and wage, but larger for women who were already mothers before the reform. Last,
as predicted by risk-sharing, the probability of marriage only rises when the partner is
employed and did not experience greater job insecurity due to the layoff-tax rise.
The remainder of the paper is organised as follows. Section 2 describes the institutional
background and the identification strategy for our analysis of individual-level insecurity, and
Section 3 the ECHP data and the estimation sample. The main results then appear in Section 4,
and the robustness checks and heterogeneity analyses in Section 5. Section 6 discusses the
5
importance of partner characteristics in the light of risk-sharing theory. Last, Section 7
concludes.
2. Institutional Background and Identification
a. The Advantages of Marriage in France
Marriage in France brings a number of financial advantages, some of which are
particularly attractive for workers facing economic insecurity. First, French legislation
considers cohabiting couples as two separate tax units but married couples as a single tax unit.
As shown in Leturcq (2012), the formula used by the French government to calculate annual
income tax in the early 2000s yields a considerably lower income tax rate for the married as
compared to the cohabiting. These gains from marriage are particularly high when the income
gap between the spouses is large. According to Echevin (2003), around 50% of married couples
in 1999 benefited from a lower yearly income tax figure of at least 1000 Euros. Buffeteau and
Echevin (2004) provide additional simulations of the financial gains from marriage in France
in the early 2000s.
There are a number of marital-property regimes in France. Up until the end of the 1990s,
the most common form of property regime was that of the “regime de communauté de biens
réduite aux acquêts” (Fremeaux and Leturcq, 2013), in which any property owned by one of
the spouses before marriage continues to be treated as belonging only to that spouse in the case
of divorce. However, all property, assets and income acquired during the marriage are treated
as common property and are subject to division in the case of divorce.
Article 212 of the French Civil Code states that “les époux se doivent mutuellement […]
secours”. The “devoir de secours” (obligation of assistance) rarely takes the form of a legal
and formal transfer of resources during marriage; it is supposed to come about naturally via
resource pooling and solidarity between the two spouses. However, in the case of divorce or
6
even during the divorce proceedings, the spouse with the most-favourable financial situation
can be asked to transfer resources to the other spouse.
Last, unemployed married individuals can more easily benefit from various social benefits
(such as social security and health insurance) when their partner is employed. In the case of the
death of a partner, marriage also simplifies the inheritance procedures for the surviving spouse
and gives the survivor the right to receive a part of the pension of the dead spouse (“pension
de reversion”).
b. The Delalande Tax
The Delalande tax was introduced into French Law in 1987 to tackle the rise in layoffs
among older workers. Despite a number of changes over time, the experience-rating principle
of the tax remained unchanged: firms laying off workers of over a certain age have to pay the
Delalande tax to the government. This tax, which was also meant to help balance the
unemployment-insurance system, is proportional to the laid-off worker's gross wage and is
applied to private-sector workers with permanent contracts. From its introduction in 1987 up
to 1992, this tax was three months of gross wages for all workers aged 55 and over who were
laid off. Major changes to this tax were then introduced in July 1992, January 1993 and January
1999; in particular, starting in 1992 the layoff tax applied to all workers aged 50 or above. The
Delalande tax was finally abolished in 2008.
Table 1 shows how the Delalande tax changed over time as a function of the age of the
laid-off worker and firm size. From 1993 to December 1998, the tax was a function of worker
age only, and independent of firm size. The tax then rose in January 1999, but only for firms
with 50 or more employees.
As predicted by the theoretical model in Behaghel (2007), a higher Delalande tax reduces
the separation rate of those workers covered by the reform (those aged 50 or more) but increases
the separation rate of the younger workers (aged under 50) in the same firms. Using the same
7
ECHP data as we do here, Georgieff and Lepinteur (2018) empirically confirm this prediction:
the higher Delalande tax had perverse effects on younger workers in larger French firms by
increasing both their perceived job insecurity and their actual risk of layoff.
We here wish to assess the causal effect of this exogenous change in job security from the
tax reform on the marriage probability of younger workers. We do so by exploiting the firm-
size discontinuity and the resulting rise in job insecurity for younger workers in larger firms.
The reform then provides a natural quasi-experimental design for a difference-in-differences
(D-i-D) estimation, in which younger workers in large private firms are the treatment group
and younger workers in smaller private firms the control group.
The 1999 reform to the Delalande tax was announced by the French government one year
beforehand (in early 1998), and the reintroduction of the firm-size discontinuity was also public
knowledge by the end of 1998 (the ECHP interviews took place in November and December
of each year). As such, employers may have been able to strategically adjust their labour
demand before the change in the Law. Georgieff and Lepinteur (2018) find evidence of such
anticipation effects, as employers in (to-be) treated firms laid off relatively more workers just
after the announcement of the reform but prior to its implementation. Higher layoffs are found
for both those aged over 50 and younger workers, so that the tax rise brought about some
restructuring in larger firms. Given this anticipation, job insecurity for the under-50s in large
firms may well have started to rise before the reform’s official implementation. We take this
possibility into account by estimating the following D-i-D equation, including both a post-1999
treatment effect and a 1998 announcement effect of the reform:
(1)
This equation is estimated only for workers under the age of 50. Here
first represents
the perceived job security of worker i in year t and then a dummy for worker i being married
8
in year t.
2
The variable Treat
it
is the treatment dummy, and is one for younger private-sector
workers in large firms (50+ employees) and zero for younger private-sector workers in smaller
firms. The variable Post1998
t
is a dummy for observations after January 1999 (the
implementation date of the higher Delalande tax), λ
t
are year fixed effects (so that λ
1998
refers
to the policy-announcement year of 1998) and X
it
is a vector of standard individual socio-
demographic controls. This includes age dummies (in five-year bands), the lagged number of
children in the household,
3
health status, weekly working hours, the log of the monthly wage,
and occupation and region dummies. This equation also controls for individual fixed effects µ
i
.
The year dummies in equation (1) entirely subsume the main effects of λ
1998
and Post1998
t
.
The coefficients of interest in the above equation are α
2
and α
3
, which respectively capture the
effects of the reform’s announcement in 1998 and its implementation starting in 1999 on first
job security and then marital status.
The risk-sharing hypothesis predicts positive and significant α
2
and α
3
coefficients for
workers in the marriage-probability regression, reflecting their rise in risk from job insecurity
and the subsequent greater demand for insurance. However, it is also possible that the treatment
reduce the probability of marriage, as workers who have become more at risk of layoff are now
seen to be “riskier” partners.
We expect the effect of the reform’s announcement and implementation to be of the same
sign, so that we can also estimate the following D-i-D regression:
(2)
2
We also consider in Online Appendix Table A2 the following types of marital outcomes: partnered, partnered
but not married, divorced or separated, and never in a relationship.
3
It is not clear whether children should be included in the list of control variables. Marriage may encourage
parenthood, or alternatively those who already have children may be more likely to marry. Including the lagged
number of children in the household as a control somewhat attenuates concerns about reverse causality. Dropping
the lagged children variable changes our key estimated coefficient of the effect of job insecurity on marriage only
very little.
9
where Post1997
t
is a dummy for observations after 1998 (the announcement year of the higher
Delalande tax). The coefficient β
2
in equation (2) now captures what we will call the total effect
of the reform, starting from its announcement (β
2
thus corresponds to equation (1) when α
2
=
α
3
). Both Equations (1) and (2) will be estimated using linear techniques (although we will also
estimate non-linear regressions in the robustness checks).
3. Data and Estimation Sample
The data we use here come from the European Community Household Panel (ECHP), a
nationally-representative longitudinal survey of households that covered 15 European
countries (including France) between 1994 and 2001. The sample size in France was on average
12,000 adults per wave. The interviews mainly took place between November and December
of each year. The ECHP contains detailed information on respondents’ socio-economic
characteristics, income, employment conditions, social relations and so on.
4
Respondents
report their marital status in each ECHP wave.
The 1999 rise in the Delalande tax only applied to firms with 50 or more employees. The
ECHP records the number of employees in the firm in which the respondent works using the
following categories: “None”, “1 to 4”, “5 to 19”, “20 to 49”, “50 to 99”, “100 to 499” and
“500 or more”. This is the variable that we will use to assign younger workers to the treatment
or control groups.
In the first part of our analysis, we show that the higher layoff tax for those aged 50 or
over in larger firms increased job insecurity for younger (aged under 50) workers in these same
firms; the main analysis will then address the question of whether this same tax rise affected
marriage.
Regarding job insecurity, we consider the following ECHP question on perceived job
security:
4
More details are available at http://ec.europa.eu/eurostat/web/microdata/european-community-household-panel.
10
“How satisfied are you with your present job in terms of job security?”
The answers to this satisfaction question were given on a 6-point scale, with 1 referring to “Not
Satisfied” and 6 “Fully Satisfied”.
5
This measure of perceived job security is a strong predictor
of individual choices such as future job quits (Clark, 2001), and reflects objective changes in
layoff and hiring rates (Böckerman et al., 2011).
6
It has also been shown to be related to
permanent contracts, unemployment-insurance benefits and employment protection legislation
(Clark and Postel-Vinay, 2009).
We consider respondents between the ages of 20 and 49 working in the private sector with
permanent contracts, and with valid information on the sociodemographic variables, job
characteristics and perceived job security. As the reform was announced many months before
its implementation, we last may suspect that firms strategically adjusted their workforce. To
rule out concerns regarding self-selection into the treatment, we will analyse only workers who
reported no change in their firm’s size from 1997 onwards. This ends up dropping only around
six per cent of the baseline sample (the resulting estimated coefficients for the effect of job
insecurity on the probability of being married are actually not materially affected by this
restriction). Note that we do not consider here the workers who were protected by the reform,
i.e. workers aged 50+. Almost 80% of respondents in this group were already married, and
there was little subsequent movement out of marriage (97.5% of those aged 50+ reported the
same marital status during the years before and after the reform’s announcement). This limited
within-variation considerably attenuates the statistical power for this older group.
Our analysis sample consists of 10,371 observations on 2,797 different individuals from
1995 to 2001 (we cannot use the first 1994 wave of ECHP here, as the information on whether
the firm is public or private is missing) between the ages of 20 and 49. Figure A1 in the Online
5
Figure A2 in the Online Appendix shows the distribution of self-reported job security. It can be seen that 70%
of responses on the 1-6 scale were 4 or 5. This negative skewness is commonly found for subjective measures.
6
We can replicate this kind of analysis in the ECHP data: the probability of both the end of the job match, and of
layoffs in particular, at time t is negatively correlated with self-reported job security at time t-1.
11
Appendix shows the number of observations at each stage of our sample selection.
7
The
descriptive statistics for the analysis sample appear in Table 2. Just under half (47%) of the
sample are in the treated group. With respect to our two dependent variables, 57% of
observations come from individuals who report being married, and average job security on the
one-to-six scale is a little over four.
Although we focus on a specific part of the French population, i.e. young private-sector
workers with a permanent contract, the share of married workers in our estimation sample is
similar to the national value (see Figure 1). The share of unmarried partnered workers in our
sample is slightly higher than the national figure, but this is very likely because we exclude
individuals above age 49. The marriage share in the French adult population has steadily
decreased from 1990 to 2009 in Figure 1, falling from 56% to 52%. Over the same period, the
share of unmarried partnered individuals rose markedly by 9.5 percentage points (from 6.7%
to 16.2%).
4. Main Results
Table 3 first shows how reported job security for younger workers changes with the
employment protection of older workers in the French sample of ECHP panel data. We here
use the analysis sample described in Table 2. Columns (1), (3) and (5) of Table 3 refer to the
D-i-D estimates of Equation (1), where we allow for a post-1999 treatment effect as well as an
announcement effect of the reform in 1998; the results in columns (2), (4) and (6) come from
the estimation of Equation (2), where the announcement and implementation dummies are
7
The fall in the number of observations due to missing values (from 13,966 to 10,936 observations) is mostly due
to missing firm size. As firm size is essential in defining treatment status, we do not impute missing values here.
Balance tests confirm that the observable characteristics of the observations dropped due to missing firm size are
not significantly different from those with valid information. Around 1,000 observations are also lost because of
missing values for weekly working hours. Again, these dropped observations have characteristics similar to those
in the estimation sample.
12
combined into one “total” effect. Columns (1) and (2) of Table 3 show the results for the whole
sample, while columns (3) - (4) and (5) - (6) respectively refer to those for women and men.
As can be seen in column (1), the 1999 rise in the layoff tax for older workers in larger
firms significantly reduced the perceived job security of younger workers in these larger firms
(as compared to younger workers in smaller firms, where there was no change in the layoff
tax). Job security fell significantly after both the announcement and the implementation of the
rise in the Delalande tax. The magnitude of the estimates is in line with the findings in Georgieff
and Lepinteur (2018) and Clark and Lepinteur (2021).
8
Pairwise Wald-tests confirm that the
estimated announcement and implementation effects of the reform in rows 1 and 2 are not
statistically different from each other. Column (2) then combines these into the total effect of
the reform, which is naturally negative and significant and of a similar size to the figures in
column (1). We estimate that the higher Delalande tax reduced the perceived job security of
younger workers in treated firms by 0.2 points, which from Table 2 is around one-sixth of a
standard deviation. We find similar estimates for women and men in the remaining columns of
Panel A. Although the reform effects look larger for men, given the associated standard errors
these gender differences are not significant.
9
We can replicate these results using a simple
binary variable for job security (comparing values 1-4 against 5-6): all of the treatment
estimates continue to be significant, with an effect that is again larger, but not significantly so,
for men.
8
The figures are not exactly the same as the analysis samples differ across the papers. In Georgieff and Lepinteur
(2018), the ECHP analysis sample pools together workers above and below age 50, and the authors estimate a
triple difference-in-differences to separately identify the effect of the reform on workers below and above age 50.
Clark and Lepinteur (2021) analyse the same age group as we do here, namely the under-50’s, but focus on the
fertility behaviour of those who were already married before the reform and continued to be so afterwards.
9
Using an estimation sample with the same characteristics as in this paper and the French Labour Force Survey,
Georgieff and Lepinteur (2018) show that the actual layoff probability of the treated workers below age 50
significantly increased after the announcement of the reform. We have replicated their analysis separately for men
and women and find no significant gender differences. These results are available upon request.
13
Panel B of Table 3 then estimates the relationship between job insecurity and the
probability of being married in the same analysis sample.
10
The estimated coefficients for the
whole sample in columns (1) and (2) are positive but not significant at conventional levels. As
it has been suggested that women are on average more risk averse than men (Sunden and
Surette, 1998; Finucane et al., 2000; Hartog et al., 2002; Holt and Laury, 2002; Fehr-Duda et
al., 2006; Croson and Gneezy, 2009; Charness and Gneezy, 2012), we might expect these
estimates to differ by gender. The results in columns (3) and (4) show that women are
significantly more likely to be married following the implementation of the reform. The
pairwise Wald-tests again reveal that the anticipation and implementation effects of the reform
are statistically identical. The results combining these two effects in column (4) reveal that the
higher Delalande tax increased the probability of being married for women by just over four
percentage points.
11
This corresponds to 7.5% of the pre-reform share of married women in the
treated group (57%). It is half of the observed difference in female marriage rates between the
early and late 20’s (see Table A1).
There is no significant marriage effect for the corresponding sample of men in columns
(5) and (6). A first possibility is that prime-age women who worked in permanent private-
sector jobs in the 1990s in France were differentially selected than their male counterparts. The
comparison of the pre-reform characteristics of men and women who do and do not appear in
our estimation sample actually reveals only small differences: there is more positive selection
of women into the estimation sample by education than there is for men, and less selection by
hours of work. However, we control for both of these variables in our estimations (and will, in
10
The full set of marriage results, including the estimated coefficients on all of the control variables, appears in
Online Appendix Table A1.
11
These regressions include a number of sociodemographic controls, as noted at the foot of the table. The inclusion
of these controls has no effect on our treatment effects. This is to be expected: if the assignment to the treatment
is random, the inclusion of additional controls will not affect the point estimates but should increase their
precision. The results without these control variables are available upon request.
14
addition, show below that there is no differential effect of the reform treatment on marriage by
either education or hours of work).
Second, this may reflect that men are on average less risk-averse: the rise in job insecurity
from the 1999 rise in the Delalande tax may not have had much effect on their demand for
insurance. Last, men who looked for insurance through marriage may not have been able to
find it: their greater job insecurity may make them less attractive. This is consistent with Ekert-
Jaffé and Solaz (2001 and 2002) and Landaud (2019), where men without a job or in temporary
employment were less likely to be partnered.
One of the requirements for D-i-D estimation to produce causal effects is that there would
have been a common trend in the dependent variable (here perceived job security and the
probability of being married) in the control and treatment groups in the absence of the policy
reform. Figure 2 thus plots the estimated yearly effects of being in the treatment (as opposed
to the control) group on perceived job security (in Panel A) and the probability of being married
(Panel B). The left-hand side shows the results for the whole sample and the right-hand figure
those for men and women separately.
12
None of the pre-reform announcement estimates (that
can be considered as placebos) in Figure 2 are significantly different from zero, providing
evidence in favour of the common-trend assumption.
13
We also need to make sure our reform effect is not influenced by changes in other
individual characteristics. For most of the observable characteristics, we find no significant
differences in the gaps between the treatment and control groups before and after the reform’s
announcement, as shown in Tables A2 and A3. However, we do see that the weekly working
12
In these latter figures we have slightly horizontally-shifted the curves for men and women so that the confidence
intervals around the estimates can be seen clearly.
13
Figure A3 in the Online Appendix plots the time profile of average perceived job security and the probability
of being married in the treatment and control groups, providing additional supporting evidence for the parallel-
trend assumption. In the bottom-right panel of Figure A3, the marriage rate of women in the treatment group is
lower than that of women in the control group before the reform. Table A3 reveals that women in the treatment
group are more educated and work more hours, both of which are associated with less marriage in this age group.
15
hours in the treatment groups dropped significantly more (in the whole and female sample only)
than in the control groups. This is not surprising: in 2000, standard weekly hours in France
dropped from 39 to 35 for workers in firms with 20 employees and more. While we do not
worry about the whole sample (because we do not find a significant effect of the 1999 rise of
the Delalande tax on the probability to be married for the whole sample), our treatment
estimates for the sample of women might capture the influence of the mandatory 2000 working-
time reduction. We will rule out this possibility in our robustness section by excluding workers
from companies with fewer than 20 employees from our estimation sample.
Last, we ask in Table A4 whether the total effect of the rise in the Delalande Tax led to
other changes in terms of marital status and couple formation. The first column of this table
reproduces the estimated marriage coefficients from columns (2), (4) and (6) of Table 3 for
comparison purposes. We then look in column (2) at the probability of being in a partnership,
irrespective of legal marital status: the estimates show that the greater job insecurity from the
reform had no effect on being in a couple in general. Combined with the results for marriage,
we thus expect (for women) a negative reform effect on cohabitation. This is indeed what we
find in column (3) of Panel B: women whose job insecurity rose following the change in the
Delalande tax are less likely to be in a non-married relationship. As the ECHP is a household
panel, where all adults are interviewed, and as we here focus on the transition from cohabitation
to marriage, we might expect the estimation results here for men and women to be identical (as
we are looking at either side of a couple). This turns out not to be the case in Table A4. The
explanation is that the men and women in this table do not come from the same sample of
cohabitees: fully one half of treated women who we observe switching from cohabitation to
marriage marry men who are not in the sample in Panel C (as the jobs of the latter are temporary
or in the public sector, they are not active in the labour force, or they are aged 50 or over).
16
Last, Columns (4) and (5) of Table A4 look in turn at being divorced/separated or never
having been in a relationship, finding no significant effect for men or women, or for the whole
sample.
14
5. Robustness Checks and Heterogeneity
The results in Table 3 refer to the effect of the reform on first job security and then marital
status for the whole analysis sample. We now turn to a number of robustness tests, and then
ask whether the effects of job insecurity on the probability of being married are larger for
certain types of workers. The analyses that we will carry out here refer to the total effect of the
reform, as in Equation (2).
15
We have also estimated the analogous effect of all of our
robustness and heterogeneity tests for the estimated effect of the reform on perceived job
security; these results appear in Online Appendix Table A5.
a. Robustness Checks
i. Ruling out confounding shocks
As noted above, the 2000 working-time reduction was another notable labour-market
reform around the time of the change in the Delalande tax. In 1998, the French Ministry of
Labour announced a reduction in the standard workweek from 39 to 35 hours in companies
with more than 20 employees. This could potentially have affected the perceived job security
of workers in those firms, and hence our estimated coefficients. We check that our main
estimates do not pick up the effect of the 35-hour week by re-running our baseline regression
excluding workers in firms with under 20 employees (which ensures that all of the individuals
in our estimation sample were equally-affected by the 35-hour week): this drops around 15%
14
The marital-status categories in Table A4 are not mutually-exclusive: an individual can be both divorced and
in a new relationship. We find similar results to those in Table A4 when we estimate a multinomial-logit model
on the following mutually-exclusive categories: “Married in a partnership”, “Non-married in a partnership”,
“Divorced or separated with no partner” and “Never in a relationship” (we exclude widows here, as there are too
few observations). These results are available upon request.
15
The results of the robustness and heterogeneity analyses when we split the reform up into announcement and
implementation effects are very similar.
17
of the initial estimation sample. The estimated reform coefficients in column (1) of Table 4 are
similar to those in the baseline (in columns (2), (4) and (6) of Panel B in Table 3). Note that
when we exclude workers in firms with under 20 employees from our estimation sample, the
differences in weekly working hours observed in Tables A2 and A3 are no longer statistically
different from zero.
As part of our identification relies on changes over time, we would like to be sure that
these changes reflect the evolution in the French layoff tax, rather than some broader macro-
economic developments. One way of testing for the latter is to re-estimate our baseline
regressions on similar samples of workers in neighbouring and arguably similar countries
(where of course the French layoff tax did not apply), as the ECHP is harmonised across
European countries. Data limitations restrict this comparison to Spain, Italy and Denmark.
16
We present the resulting D-i-D estimates for these three countries in columns (2), (3) and (4)
of Table 4, none of which is significantly different from zero. Macroeconomic trends do not
then seem to lie behind our main result of greater job insecurity increasing the share of married
women.
ii. The estimation method
Our main results above come from fixed-effect analyses, comparing the same individual
before and after the labour-market reform. We expect fixed-effects and OLS analyses to
produce different estimates, as the former introduce attenuation bias when there is
measurement error (with the resulting OLS estimates being larger than their FE counterparts in
absolute terms) and for omitted-variable reasons, where the OLS estimates will be biased if the
treatment is correlated with unobserved individual time-invariant characteristics. The baseline
16
The information in the final waves of the ECHP in Belgium and Germany does not allow us to accurately
distinguish the public from the private sector, or to measure perceived job security.
18
results without fixed effects in column (5) of Table 4 are slightly larger than the baseline fixed-
effect estimates, but not significantly so.
We also ask whether these results are affected by the way in which we define the dependent
variable. Our baseline regressions treat the probability of being married as a cardinal variable
(and are thus fixed-effect linear-probability models). However, it can be argued that non-linear
estimation is more suitable for this dummy dependent variable. Column (6) of Table 4 thus re-
estimates our main regression using a conditional fixed-effect logit model; the results here
continue to produce a positive significant estimated coefficient for women, with job insecurity
increasing the probability of being married.
iii. Sample composition
Firm size is reported by the respondent and may not be accurate. In this case, individuals
can be mis-allocated to firm-size groups, and hence to the control or treatment groups. This
mis-reporting may be random, in which case we estimate a lower bound of the treatment (as
some of the control group are treated, and some of the treated group are not). A potentially
more serious problem arises if the mis-allocation is not random, conditional on the control
variables in our regressions. We address mis-reporting using a test inspired by the donut
regression-discontinuity design, and drop workers who are close to the firm-size treatment
threshold of 50 employees. From the firm-size categories listed in Section 3, this implies re-
estimating the treatment without respondents who say they work in firms with “20 to 49
employees” or “50 to 99 employees”. Intuitively, mis-judgement may cause workers to
erroneously report a firm-size category above or below the correct value, but not to jump three
categories (so that they report a firm size of under 20 employees when the real value is over
100 employees, for example).
The estimated treatment coefficients from the baseline analysis (in panel B of Table 3) and
these “donut” regressions (in the last column of Table 4) are of the same size, although the
19
estimate of the latter for women is no longer significant (due to the smaller sample size when
we drop two firm-size categories, producing larger standard errors). That the estimated
coefficient does not change between the two specifications helps to dispel any worries about
the systematic mis-reporting of firm size.
17
We last consider attrition. From the announcement of the reform onwards, 16% of our
treatment group left the estimation sample before the last wave (this figure is the same in the
control group, with no noticeable differences between men and women). We may worry that
attrition is not independent of the reform’s implementation so that our estimates are biased,
especially if the marriage probability of leavers decreases after they leave the sample. A first
point is that the ECHP does supply two weights, both of which are called attrition weights (our
main regression results do not use weights). Applying these attrition weights does not change
the nature of our results.
We can also calibrate the “unobserved” difference-in-difference in the attrition group that
would be required produce a main estimate of zero: this is 0.21 (= 0.04*84%/16%). Our main
result is that the difference in the probability of being married between women in large and
small firms rose by 4 percentage points post-reform. To cancel this figure out, we would require
the analogous figure for the 16% of women in the attrition group to be a fall of 21 percentage
points after the reform. This figure is more than double the largest age effect on marriage in
column (4) of Table A1, and does not seem plausible.
b. Heterogeneity
The D-i-D estimates in Panel B of Table 3 show the average treatment effect for workers
in large firms. In Table 5 we consider whether these effects might differ across groups. We
17
As in Clark and Lepinteur (2021), we can consider public-sector workers as an alternative control group but
none of our treatment estimates was significantly different from zero at conventional levels. Note however that
the marriage probability pre-trends differ significantly between the treatment group and public-sector workers,
which cast doubts about the validity of the estimation strategy in this particular sample. All of these results are
available upon request.
20
first consider age, and interact the total effect of the reform with a dummy for being born in
1963 or before. The resulting estimates appear in columns (1) and (6) of Table 5 respectively
for women and men: neither interaction term here is significant.
18
We second know that the relationship between fertility decisions and job insecurity likely
depends on education and earnings (Chevalier and Marie, 2017; Clark and Lepinteur, 2021).
We here ask whether an analogous relationship is found for marriage. We thus interact the
reform with “High-education” and “High-wage” dummies, corresponding respectively to
workers with post-secondary education and those with above-median wage: both of these
interaction variables are measured in the pre-reform years only.
19
The resulting estimates on
these interactions in columns (2), (3), (7) and (8) of Table 5 indicate no significant difference
in the effect of job insecurity on marriage by education or wages.
20
Our last interactions concern pre-reform family characteristics. Columns (4) and (9) show
the estimated coefficient on the interaction with a dummy for the treated workers having
children before the higher Delalande tax: this is positive and significant only for women. One
natural interpretation is that women with children are exposed to greater risk than men with
children, as in France the former ended up with sole custody of the children in 80% of
separations in the early 2000s.
21
The figures in columns (5) and (10) refer to the interactions
with the number of children the respondent had at the time of the reform. The resulting
estimated coefficients are positive and significant (and of the same size) only for women with
one or two children (although that for women with three or more children is not statistically
different from that for mothers with one or two children).
18
1963 is the median birth year in our estimation sample. The interaction terms from different birth-year thresholds
continue to be insignificant.
19
We also considered an interaction between the treatment and a continuous measure of monthly wages (in logs),
which produces similar results. Using household income rather than the monthly wage also makes no difference.
20
With respect to hours of work, interactions with pre-reform hours either as a continuous variable or as a dummy
for part-time work (between 1 and 29 hours per week) yielded no significant estimates.
21
See Table 2 in http://www.justice.gouv.fr/art_pix/stat_Infostat%20132%20def.pdf.
21
Why does the effect of job insecurity differ across some groups of workers? One possibility
is that the relationship between the layoff tax and job insecurity is more pronounced for some
workers, and especially those who already had children. We look for evidence of this in Online
Appendix Table A6, where the dependent variable is now perceived job security. None of the
interaction terms in this table is significantly different from zero: as such, all of the different
groups of employees in this table react similarly to the reform in terms of their perceived job
security.
A second possibility is that the link between perceived job security and marital status may
differ across groups, for reasons of risk-aversion. In Görlitz and Tamm (2015), the higher risk-
aversion due to parenthood is larger for women. As such, mothers arguably constitute the most
risk-averse group of workers in our estimation sample, which might explain why they are the
most likely to get married after a rise in their job insecurity.
Risk-Sharing Theory
Our results above are in line with risk sharing theory, as the rise in marriage after the 1999
rise in the Delalande tax appears only for the arguably most risk-averse workers (i.e. women
with children), although we have not yet provided any explicit test of this theory. In Hess (2004)
and Shore (2010), couples manage income risk by trying to ensure that the two partners’
exogenous income shocks are not positively perfectly correlated. In the context of the reform
we analyse here, we expect a stronger treatment effect for women whose partner has a stable
job.
This is what we test in Table 6, where we interact the reform effect with dummy variables
for different types of partners. Columns (1) and (4) show the average effect of the reform for
women and men, as in Panel B of Table 3, while columns (2) and (5) show the estimated
coefficients of the interaction of the reform with a dummy for “Employed Partner”. As risk-
sharing theory would predict, the shift from cohabitation to marriage (following the results in
22
Table A4) is only higher for women whose partners are currently working.
22
We find no
significant results for any of the groups of men.
We then take the treatment status of the partner into account. Under risk-sharing, household
members try to avoid correlated risks. The marriage incentives from job insecurity should then
be weaker for women whose partner is also affected by the reform. The estimates in column
(3) are in line with this prediction: marriage only rises significantly for women whose partner
does not currently suffer from exogenously-higher job insecurity. We found no significant
differences in this respect for the partner working in the public vs. private sector or having a
permanent vs. temporary contract (although this latter might reflect a lack of statistical power,
as only around 3% of partners in our estimation sample had a temporary contract). We continue
to find no significant marriage results for any of the groups of men (in the last column of Table
6).
6. Conclusion
Job insecurity increases the probability of marriage for women. The 1999 layoff-tax reform
in the French labour market protected older workers, but to the detriment of younger workers.
As this reform was applied only in larger firms, we can carry out a difference-in-differences
analysis. The exogenous change in the future probability of job loss affected perceived job
security; it also produced a robust significant rise in the probability of being married for
women. Our identification strategy here evaluates the effect of job insecurity on the probability
of being married for workers who were employed both before and after the reform (and not
those who changed their labour-force status): this is probably what most people will understand
by job insecurity, in the sense that insecurity is forward-looking.
22
The interactions in Table 6 refer to the partner’s current labour-market position. As such, the employed in row
2 may have started work after their partner was treated. Equally, in rows 3 and 4, partner treatment may have
caused individuals to switch into more secure jobs (i.e. those that are not affected by the higher Delalande tax). If
we only consider the partner’s pre-reform employment status, we do not allow for these behavioural reactions (if
we do so we actually find similar estimates, although the coefficients are less precisely-estimated).
23
Our results are novel as they are, to the best of our knowledge, the first that appeal to a
natural labour-market experiment to show that risk-sharing is one of the causes of marriage.
Job insecurity increased the probability of marriage for women, and more so for those who are
probably more risk-averse (mothers). In line with risk-sharing theory, we show that this
marriage effect was not found for couples in which the layoff risk for both partners rose after
the 1999 tax rise (i.e. couples where both members worked in treated firms). The lack of any
effect for men may reveal their lower risk-aversion with respect to job insecurity, or the greater
difficulty that insecure men face on the marriage market.
Part of the attraction of marriage then seems to be the risk-sharing it provides. Why then
don’t all couples get married? There are a wide variety of factors at play here, including cultural
norms. Some of these can be argued to be economic. In the same way that employment
protection legislation might discourage hiring due to the costs of firing, more rigid or expensive
divorce procedures may discourage some couples from marrying. In this case, more flexible
divorce procedures may lead to more people getting married, as would a more advantageous
tax treatment of the married relative to the cohabiting. The flexibility of marriage and the labour
market can thus be considered as intertwined.
24
References
Anderson, S., and Ray, D. (2019). “Missing Unmarried Women.” Journal of the European
Economic Association, 17, 1585-1616.
Becker, G. S. (1973). “A theory of marriage: Part I.” Journal of Political Economy, 81, 813-
846.
Becker, G. S. (1981). “Altruism in the family and selfishness in the market place.” Economica,
48, 1-15.
Behaghel, L. (2007). “La protection de l'emploi des travailleurs âgés en France : une évaluation
ex ante de la contribution Delalande.” Annales d'Economie et de Statistique, 85, 41-80.
Bertocchi, G., Brunetti, M., and Torricelli, C. (2011). “Marriage and other risky assets: A
portfolio approach.” Journal of Banking and Finance, 35, 2902-2915.
Böckerman, P., Ilmakunnas, P., and Johansson, E. (2011). “Job security and employee well-
being: Evidence from matched survey and register data.” Labour Economics, 18, 547-554.
Borenstein, S., and Courant, P. N. (1989). “How to carve a medical degree: human capital
assets in divorce settlements.” American Economic Review, 79, 992-1009.
Buffeteau, S., and Echevin, D. (2004). “Fiscalité et mariage.” Economie Publique, 13, 3-28.
Charles, K.K., and Stephens, M. (2004). “Job Displacement, Disability, and Divorce.” Journal
of Labor Economics, 22, 489-522.
Charness, G., and Gneezy, U. (2012). “Strong evidence for gender differences in risk taking.”
Journal of Economic Behavior and Organization, 83, 50-58.
Chevalier, A., and Marie, O. (2017). “Economic uncertainty, parental selection, and children’s
educational outcomes.” Journal of Political Economy, 125, 393-430
Chiappori, P. A. (1992). “Collective labor supply and welfare.” Journal of Political Economy,
100, 437-467.
Clark, A. E. (2001). “What really matters in a job? Hedonic measurement using quit data.”
Labour Economics, 8, 223-242.
25
Clark, A. E. (2003). “Unemployment as a social norm: Psychological evidence from panel
data.” Journal of Labor Economics, 21, 323-351.
Clark, A. E., and Postel-Vinay, F. (2009). “Job security and job protection.” Oxford Economic
Papers, 61, 207-239.
Clark, A. E., and Lepinteur, A. (2021). “A Natural Experiment on Job Insecurity and Fertility
in France.” Review of Economics and Statistics, forthcoming.
Croson, R., and Gneezy, U. (2009). “Gender differences in preferences.” Journal of Economic
Literature, 47, 448-74.
De la Rica, S., and Iza, A. (2005). “Career planning in Spain: Do fixed-term contracts delay
marriage and parenthood?” Review of Economics of the Household, 3, 49-73.
Échevin, D. (2003). “L'individualisation de l'impôt sur le revenu : équitable ou pas ?” Economie
et Prévision, 4, 149-165.
Ekert-Jaffé, O., and Solaz, A. (2001). “Unemployment, marriage, and cohabitation in France.”
Journal of Socio-Economics, 30, 75-98.
Ekert-Jaffé, O., and Solaz, A. (2002). “Couple formation in France: the changing importance
of labor market early career path.” Journal of Bioeconomics, 4, 223-239.
Falk, A., Becker, A., Dohmen, T., Enke, B., Huffman, D., and Sunde, U. (2018). “Global
Evidence on Economic Preferences.” Quarterly Journal of Economics, 133, 1645-
1692.
Fehr-Duda, H., De Gennaro, M., and Schubert, R. (2006). “Gender, financial risk, and
probability weights.” Theory and Decision, 60, 283-313.
Finucane, M. L., Alhakami, A., Slovic, P., and Johnson, S. M. (2000). “The affect heuristic in
judgments of risks and benefits.” Journal of Behavioral Decision Making, 13, 1-17.
Frémeaux, N., and Leturcq, M. (2013). “Plus ou moins mariés : l'évolution du mariage et des
régimes matrimoniaux en France.” Économie et Statistique, 462, 125-151.
Georgieff, A., and Lepinteur, A. (2018). “Partial employment protection and perceived job
security: evidence from France.” Oxford Economic Papers, 70, 846-867.
26
Görlitz, K., and Tamm, M. (2015). “Parenthood and risk preferences.” Ruhr Economics Papers
No. 552.
Halla, M., and Scharler, J. (2012). “Marriage, divorce, and interstate risk sharing.”
Scandinavian Journal of Economics, 114, 55-78.
Hartog, J., Ferrer‐i‐Carbonell, A., and Jonker, N. (2002). “Linking measured risk aversion to
individual characteristics.” Kyklos, 55, 3-26.
Hess, G. D. (2004). “Marriage and consumption insurance: What’s love got to do with it?”
Journal of Political Economy, 112, 290-318.
Holt, C. A., and Laury, S. K. (2002). “Risk aversion and incentive effects.” American Economic
Review, 92, 1644-1655.
Landaud, F. (2019). “From Employment to Engagement? Stable Jobs, Temporary Jobs, and
Cohabiting Relationships.” NHH Department of Economics Discussion Paper No.
10/2019.
Lepinteur, A. (2019). “Working time mismatches and self-assessed health of married couples:
Evidence from Germany.” Social Science & Medicine, 235, 112410.
Leturcq, M. (2012). “Will you civil union me? Taxation and civil unions in France.” Journal
of Public Economics, 96, 541-552.
Rosenzweig, M. R., and Stark, O. (1989). “Consumption smoothing, migration, and marriage:
Evidence from rural India.” Journal of Political Economy, 97, 905-926.
Schaller, J. (2013). “For richer, if not for poorer? Marriage and divorce over the business
cycle.” Journal of Population Economics, 26, 1007-1033.
Schneider, D., Harknett, K., and Stimpson, M. (2018). “What Explains the Decline in First
Marriage in the United States? Evidence from the Panel Study of Income Dynamics,
1969 to 2013.” Journal of Marriage and Family, 80, 791-811.
Schneider, D., and Reich, A. (2018). “Marrying Ain’t Hard When You Got A Union Card?
Labor Union Membership and First Marriage.” Social Problems, 61, 625-643.
27
Shore, S. H. (2010). “For better, for worse: intrahousehold risk-sharing over the business
cycle.” Review of Economics and Statistics, 92, 536-548.
Stevenson, B., and Wolfers, J. (2007). “Marriage and Divorce: Changes and Their Driving
Forces.” Journal of Economic Perspectives, 21, 27-52.
Sunden, A. E., and Surette, B. J. (1998). “Gender differences in the allocation of assets in
retirement savings plans.” American Economic Review, 88, 207-211.
Van den Berg, G. J., and Gupta, S. (2015). “The role of marriage in the causal pathway from
economic conditions early in life to mortality.” Journal of Health Economics, 40, 141-
158.
Van Klaveren, C., Van Praag, B., and Van den Brink, H. M. (2008). “A public good version of
the collective household model: an empirical approach with an application to British
household data.” Review of Economics of the Household, 6, 169-191.
Weiss, Y. (1997). “The formation and dissolution of families. Why marry? Who marries
whom? And what happens upon divorce.” Handbook of Population and Family
Economics, 1, 81-123.
Xie, Y., Raymo, J., Goyette, K., and Thornton, A. (2003). “Economic potential and entry into
marriage and cohabitation.” Demography, 40, 351-367.
Yu, J., and Xie, Y. (2015). “Changes in the determinants of marriage entry in post-reform urban
China.” Demography, 52, 1869-1892.
28
Figures and Tables
Figure 1: Marriage and Cohabitation Rates in France from 1990 to 2009
Notes: The marriage and cohabitation rates correspond respectively to the number of married
and cohabiting individuals divided by the size of the adult population. These rates were
calculated using data from the INED time series on the number of married and cohabiting
couples (https://www.ined.fr/fr/tout-savoir-population/chiffres/france/couples-menages-
familles/couples_menages_familles/) and from the UN website
(http://data.un.org/Data.aspx?d=POP&f=tableCode%3A22) for the adult population.
29
Figure 2: Parallel-Trend Assumption – Panel Results
Panel A: Perceived Job Security
Panel B: Probability of being married
Notes: The sample here is permanent-contract private-sector workers in the French ECHP between the ages of 20
and 49. There are 10371 observations in the whole sample (2797 individuals), 4248 observations in the sample of
women (1175 individuals) and 6123 observations in the sample of men (1622 individuals). Each point shows the gap
in the outcome between the treatment group (i.e. being a younger worker in a firm with 50 or more employees) and
the control group (being a younger worker in a firm with fewer than 50 employees) in that year, as compared to the
same gap in the omitted year (1995). These numbers come from regression analyses that include individual and year
fixed effects, as well as age dummies (in five-year bands), the number of children in the household (lagged), health
status, weekly working hours, the log of the monthly wage, and occupation and region dummies. The error bars
represent the 90% confidence intervals. The dotted vertical line indicates the date at which the rise in the Delalande
tax was announced, while the dashed vertical line shows the date of its implementation. Men and women were
interviewed at the same date; we have slightly left-shifted the points for men so that the confidence intervals can be
seen.
30
Table 1: The Evolution of the Delalande Layoff Tax over Time
Worker’s age
50
51
52
53
54
55
56-57
58
59
July 1987-June 1992
All firm sizes
3
3
3
3
July 1992 - Dec. 1992
20 or more employees
1
1
2
2
4
5
6
6
6
Under 20 employees
0.5
0.5
1
1
2
2.5
3
3
3
Jan 1993-Dec 1998
All firm sizes
1
1
2
2
4
5
6
6
6
January 1999-2008
50 or more employees
2
3
5
6
8
10
12
10
8
Under 50 employees
1
1
2
2
4
5
6
6
6
Source: Legislative texts.
Notes: For each age group, the table shows the tax to be paid by the firm to the unemployment-insurance system if it lays off a worker
of that age. The tax is a function of the worker’s wages, and is stated in months of gross wage. This tax applies to private-sector workers
with permanent contracts only.
Table 2: Descriptive Statistics of the ECHP Analysis Sample
Mean
SD
Min
Max
Dependent Variable:
Perceived Job Security [1-6]
4.13
1.17
1
6
Married
0.57
0
1
Partnered
0.77
0
1
Partnered but not Married
0.19
0
1
Divorced or Separated
0.04
0
1
Never in a Relationship
0.20
0
1
Difference-in-differences Variables:
Treatment Group
0.47
0
1
Post Period
0.37
0
1
Individual Characteristics:
Age
34.60
6.85
20
49
Female
0.41
0
1
High Education (Post-Secondary)
0.24
0
1
Number of Children in Household (lagged)
0.95
0.97
0
8
Health Status [1-5]
3.88
0.72
1
5
Job Characteristics:
Weekly Working Hours
39.39
7.86
2
96
Monthly Wage (log)
8.98
0.48
5.08
11.60
Observations
10371
Individuals
2797
Note: These numbers refer to permanent-contract private-sector workers in the French ECHP
between the ages of 20 and 49.
31
Table 3: The Delalande Tax, Job Security and the Probability of Being Married – Panel Results
Panel A: Job Security
Whole Sample
Women
Men
(1-6)
(1)
(2)
(3)
(4)
(5)
(6)
Reform Announcement
-0.188
***
-0.149
*
-0.225
***
(0.057)
(0.090)
(0.075)
Reform Implementation
-0.202
***
-0.159
*
-0.240
***
(0.053)
(0.086)
(0.067)
Total Effect of the Reform
-0.198
***
-0.156
**
-0.236
***
(0.048)
(0.077)
(0.062)
Panel B: Married
Whole Sample
Women
Men
(Dummy)
(1)
(2)
(3)
(4)
(5)
(6)
Reform Announcement
0.018
0.031
0.010
(0.012)
(0.020)
(0.016)
Reform Implementation
0.015
0.047
*
-0.007
(0.015)
(0.025)
(0.019)
Total Effect of the Reform
0.016
0.042
**
-0.002
(0.013)
(0.021)
(0.017)
Notes: These are linear regressions. The sample here is permanent-contract private-sector workers in the French ECHP
between the ages of 20 and 49. There are 10371 observations in the whole sample (2797 individuals), 4248 observations in
the sample of women (1175 individuals) and 6123 observations in the sample of men (1622 individuals). The announcement
effect of the reform refers to the treatment from the beginning of 1998, when the reform to the Delalande tax was announced,
up to its implementation on January 1
st
1999; the reform-implementation effect considers the implementation treatment
starting on January 1
st
1999. These two effects correspond to α
2
and α
3
in Equation (1). The total effect of the reform
corresponds to β
2
in Equation (2). Standard errors in parentheses are clustered at the individual level. All of the regressions
include individual and year fixed effects, as well as age dummies (in five-year bands), the number of children in the household
(lagged), health status, weekly working hours, the log of the monthly wage, and occupation and region dummies. *, ** and
*** indicate significance at the 10%, 5% and 1% levels respectively.
32
Table 4: The Rise in the Delalande Tax and the Probability of being Married – Robustness Checks
20+
employees
Spanish
sample
Italian
sample
Danish
sample
OLS
Conditional
FE Logit
Donut
DiD
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Panel A: Whole Sample
Total Effect of the reform
0.009
-0.003
0.011
0.025
0.033
0.193
0.014
(0.018)
(0.013)
(0.013)
(0.026)
(0.020)
(0.370)
(0.015)
Individual Time-Invariant Controls
.
.
.
.
Yes
.
.
Individual Fixed Effects
Yes
Yes
Yes
Yes
.
Yes
Yes
Observations
8897
8938
11373
4682
10371
10371
7523
Individuals
2360
3127
3411
1432
2797
2797
2067
Panel B: Women
Total Effect of the reform
0.050
**
-0.040
0.022
-0.014
0.058
*
1.530
**
0.043
(0.024)
(0.025)
(0.023)
(0.041)
(0.033)
(0.772)
(0.028)
Individual Time-Invariant Controls
.
.
.
.
Yes
.
.
Individual Fixed Effects
Yes
Yes
Yes
Yes
.
Yes
Yes
Observations
3494
3084
4265
1739
4248
4248
3018
Individuals
944
1134
1340
572
1175
1175
860
Panel C: Men
Total Effect of the reform
-0.019
0.010
0.006
0.055
0.022
-0.176
-0.005
(0.025)
(0.016)
(0.015)
(0.033)
(0.025)
(0.498)
(0.019)
Individual Time-Invariant Controls
.
.
.
.
Yes
.
.
Individual Fixed Effects
Yes
Yes
Yes
Yes
.
Yes
Yes
Observations
5403
5584
7108
2943
6123
6123
4505
Individuals
1416
1993
2071
860
1622
1622
1207
Notes: These are linear regressions, except in column (6). The sample here is permanent-contract private-sector workers in the French ECHP between the ages of 20
and 49 (except in columns (2), (3) and (4), where it is their Spanish, Italian and Danish counterparts). The total effect of the reform corresponds to β
2
in Equation (2).
Standard errors in parentheses are clustered at the individual level, except in columns (6). The conditional FE logit coefficients in column (6) refer to the log of the
odds ratio. All of the regressions include year fixed effects, age dummies (in five-year bands), the number of children in the household (lagged), health status, weekly
working hours, the log of the monthly wage, and occupation and region dummies. The individual time-invariant controls are gender and education dummies. *, **
and *** indicate significance at the 10%, 5% and 1% levels respectively.
33
Table 5: The Rise in the Delalande Tax and the Probability of being Married – Heterogeneity Analysis
Women
Men
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
Total Effect of the Reform
0.062
0.014
0.019
-0.040
-0.040
0.018
-0.002
-0.012
0.014
0.014
(0.039)
(0.022)
(0.023)
(0.033)
(0.033)
(0.031)
(0.018)
(0.029)
(0.024)
(0.024)
Interacted with:
Born before 1963
-0.031
-0.033
(0.045)
(0.036)
High Education
0.080
-0.018
(0.055)
(0.049)
High Monthly Wage
0.039
0.014
(0.039)
(0.036)
Parent beforehand
0.114
***
-0.023
(0.042)
(0.032)
1 child beforehand
0.113
**
-0.015
(0.053)
(0.050)
2 children beforehand
0.124
**
-0.047
(0.052)
(0.037)
3+ children beforehand
0.065
0.022
(0.046)
(0.041)
Notes: These are linear regressions. The sample here is permanent-contract private-sector workers in the French ECHP between the ages of 20 and 49. There are 4248
observations in the sample of women (1175 individuals) and 6123 observations in the sample of men (1622 individuals). The total effect of the reform corresponds to
β
2
in Equation (2). Standard errors in parentheses are clustered at the individual level. All of the regressions include individual and year fixed effects, as well as age
dummies (in five-year bands), the number of children in the household (lagged), health status, weekly working hours, the log of the monthly wage, and occupation
and region dummies. *, ** and *** indicate significance at the 10%, 5% and 1% levels respectively.
34
Table 6: The Rise in the Delalande Tax and the Probability of being Married by Partner’s Employment Status – Panel Results
Women
Men
(1)
(2)
(3)
(4)
(5)
(6)
Total Effect of the Reform
0.042
**
-0.016
-0.015
-0.002
-0.018
-0.020
(0.021)
(0.031)
(0.031)
(0.017)
(0.021)
(0.021)
Interacted with:
Employed Partner
0.093
**
0.032
(0.036)
(0.027)
Employed Partner Not Currently Affected by the Higher Delalande Tax
0.101
**
0.029
(0.039)
(0.028)
Employed Partner Currently Affected by the Higher Delalande Tax
-0.017
-0.008
(0.042)
(0.055)
Notes: These are linear regressions. The sample here is permanent-contract private-sector workers in the French ECHP between the ages of 20 and 49. There are 4248
observations in the sample of women (1175 individuals) and 6123 observations in the sample of men (1622 individuals). The total effect of the reform corresponds to β
2
in
Equation (2). The standard errors in parentheses are clustered at the individual level. All of the regressions include individual and year fixed effects, as well as age dummies
(in five-year bands), the number of children in the household (lagged), health status, weekly working hours, the log of the monthly wage, and occupation and region dummies.
*, ** and *** indicate significance at the 10%, 5% and 1% levels respectively.
35
Online Appendix
Figure A1: Sample Selection
Note: Each box shows the number of observations in the French ECHP sample, from the raw data (95 171 observations) to the
sample that is used in our empirical analyses (10 371 observations).
36
Figure A2: The Distribution of Perceived Job Security in the ECHP
Note: This figure refers to permanent-contract private-sector workers in the French ECHP
between the ages of 20 and 49.
37
Figure A3: Average Perceived Job Security and the Probability of being Married by Treatment Group
Panel A: Perceived Job Security
v
Panel B: Probability of being married
Note: The dotted vertical line indicates the date at which the rise in the Delalande tax was announced, while the dashed vertical
line shows the date of its implementation.
38
Table A1: The Delalande Tax and Marriage – Panel results with the full set of controls
Whole Sample
Women
Men
(1)
(2)
(3)
(4)
(5)
(6)
Reform Announcement
0.018
0.031
0.010
(0.012)
(0.020)
(0.016)
Reform Implementation
0.015
0.047
*
-0.007
(0.015)
(0.025)
(0.019)
Total Effect of the reform
0.016
0.042
**
-0.002
(0.013)
(0.021)
(0.017)
Age: 26-30
0.109
***
0.109
***
0.090
***
0.089
***
0.124
***
0.123
***
(0.020)
(0.020)
(0.031)
(0.031)
(0.027)
(0.027)
Age: 31-35
0.113
***
0.113
***
0.088
**
0.087
**
0.132
***
0.132
***
(0.024)
(0.024)
(0.034)
(0.034)
(0.032)
(0.032)
Age: 36-40
0.097
***
0.097
***
0.063
*
0.062
*
0.122
***
0.122
***
(0.026)
(0.026)
(0.036)
(0.036)
(0.035)
(0.035)
Age: 41-45
0.061
**
0.061
**
0.003
0.002
0.101
***
0.100
***
(0.028)
(0.028)
(0.042)
(0.042)
(0.038)
(0.038)
Age: 46-49
0.006
0.006
-0.080
-0.081
0.065
0.064
(0.032)
(0.032)
(0.051)
(0.050)
(0.042)
(0.042)
Monthly Wage (in log)
0.021
0.021
-0.007
-0.007
0.054
**
0.054
**
(0.016)
(0.016)
(0.023)
(0.023)
(0.021)
(0.021)
Weekly Working Hours
-0.001
**
-0.001
**
-0.000
-0.000
-0.001
**
-0.001
**
(0.000)
(0.000)
(0.001)
(0.001)
(0.001)
(0.001)
Self-Assessed Health [1-5]
-0.001
-0.001
-0.006
-0.006
0.004
0.004
(0.004)
(0.004)
(0.007)
(0.007)
(0.005)
(0.005)
Number of Children in the Household (lagged)
0.054
***
0.054
***
0.057
***
0.057
***
0.052
***
0.052
***
(0.007)
(0.007)
(0.012)
(0.012)
(0.009)
(0.009)
Notes: These are linear regressions. The sample here is permanent-contract private-sector workers in the French ECHP between the ages of 20 and
49. There are 10371 observations in the whole sample (2797 individuals), 4248 observations in the sample of women (1175 individuals) and 6123
observations in the sample of men (1622 individuals). The announcement effect of the reform refers to the treatment from the beginning of 1998,
when the reform to the Delalande tax was announced, up to its implementation on January 1
st
1999; the reform-implementation effect considers
the implementation treatment starting on January 1
st
1999. These two effects correspond to α
2
and α
3
in Equation (1). The total effect of the reform
corresponds to β
2
in Equation (2). Standard errors in parentheses are clustered at the individual level. All of the regressions include individual fixed
effects, as well as wave, occupation and region dummies. *, ** and *** indicate significance at the 10%, 5% and 1% levels respectively.
39
Table A2: Differences in observable characteristics between the treatment and control group before and after the reform’s announcement
Before the announcement of the
1999 rise in the Delalande tax
After the announcement of the
1999 rise in the Delalande tax
Difference-in-
Differences
Treatment
Control
Difference
Treatment
Control
Difference
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Age
34.128
33.092
1.036
***
36.140
35.143
0.997
**
-0.040
[6.082]
[6.517]
(0.175)
[7.024]
[7.302]
(0.199)
(0.266)
Female
0.386
0.441
-0.056
***
0.380
0.425
-0.045
***
0.011
[0.487]
[0.497]
(0.014)
[0.486]
[0.494]
(0.014)
(0.019)
High Education
0.266
0.187
0.079
***
0.298
0.217
0.081
***
0.002
[0.442]
[0.390]
(0.012)
[0.458]
[0.412]
(0.012)
(0.017)
No. Children in Household (lagged)
0.943
0.943
-0.000
0.958
0.959
-0.000
-0.000
[0.984]
[0.969]
(0.027)
[0.970]
[0.954]
(0.027)
(0.038)
Health Status
3.901
3.916
-0.015
3.839
3.851
-0.013
0.002
[0.740]
[0.752]
(0.021)
[0.706]
[0.673]
(0.019)
(0.028)
Weekly Working Hours
40.108
39.516
0.592
***
38.851
39.097
-0.246
-0.838
***
[6.898]
[9.151]
(0.218)
[6.557]
[8.285]
(0.218)
(0.309)
Monthly Wage (log)
9.004
8.818
0.227
***
9.143
8.930
0.214
***
-0.013
[0.440]
[0.471]
(0.013)
[0.472]
[0.482]
(0.013)
(0.018)
Notes: There are 10371 observations in the whole sample (2797 individuals). Standard errors are in parentheses and standard deviations are in square brackets.
*, ** and *** indicate significance at the 10%, 5% and 1% levels respectively.
40
Table A3: Differences in characteristics between the treatment and control group before and after the reform’s announcement by gender
Before the announcement of the
1999 rise in the Delalande tax
After the announcement of the
1999 rise in the Delalande tax
Difference-in-
Differences
Treatment
Control
Difference
Treatment
Control
Difference
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Panel A: Women
Age
33.925
32.902
1.024
***
35.712
35.117
0.595
*
-0.428
[6.117]
[6.617]
(0.278)
[7.105]
[7.327]
(0.318)
(0.422)
High Education
0.312
0.219
0.093
***
0.338
0.255
0.082
***
-0.011
[0.464]
[0.414]
(0.019)
[0.473]
[0.436]
(0.020)
(0.027)
No. Children in Household (lagged)
0.821
0.915
-0.093
**
0.804
0.930
-0.126
***
-0.033
[0.886]
[0.859]
(0.038)
[0.85]
[0.873]
(0.038)
(0.054)
Health Status
3.888
3.876
0.012
3.801
3.835
-0.034
0.045
[0.718]
[0.754]
(0.032)
[0.710]
[0.678]
(0.030)
(0.044)
Weekly Working Hours
37.701
35.646
2.054
***
36.755
35.635
1.119
***
-0.935
*
[6.725]
[9.122]
(0.355)
[6.220]
[8.023]
(0.319)
(0.478)
Monthly Wage (log)
8.901
8.625
0.277
***
8.993
8.759
0.235
***
-0.042
[0.447]
[0.468]
(0.020)
[0.472]
[0.493]
(0.021)
(0.029)
Panel B: Men
Age
34.257
33.243
1.014
***
36.402
35.162
1.240
***
0.226
[6.058]
[6.434]
(0.227)
[6.963]
[7.286]
(0.256)
(0.343)
High Education
0.236
0.161
0.075
***
0.274
0.189
0.085
***
0.010
[0.425]
[0.368]
(0.014)
[0.446]
[0.392]
(0.015)
(0.021)
No. Children in Household (lagged)
1.019
0.966
0.054
1.053
0.980
0.073
**
0.020
[1.034]
[1.047]
(0.037)
[1.022]
[1.010]
(0.037)
(0.053)
Health Status
3.910
3.948
-0.038
3.862
3.864
-0.002
0.036
[0.753]
[0.748]
(0.026)
[0.703]
[0.669]
(0.026)
(0.037)
Weekly Working Hours
41.619
42.573
-0.954
***
40.138
41.661
-1.123
***
-0.569
[6.571]
[7.940]
(0.260)
[6.428]
[7.508]
(0.257)
(0.365)
Monthly Wage (log)
9.134
8.970
0.164
***
9.235
9.056
0.179
***
0.015
[0.411]
[0.414]
(0.015)
[0.448]
[0.434]
(0.016)
(0.022)
Notes: There are 4248 observations in the sample of women (1175 individuals) and 6123 observations in the sample of men (1622 individuals). Standard errors
are in parentheses and standard deviations are in square brackets. *, ** and *** indicate significance at the 10%, 5% and 1% levels respectively.
41
Table A4: The Delalande Tax and Other Types of Partnerships – Panel Results
Panel A: Whole Sample
Married
Partnered
Partnered
but not
married
Divorced
or
Separated
Never in a
relationship
(1)
(2)
(3)
(4)
(5)
Total Effect of the Reform
0.016
-0.023
-0.009
-0.004
0.013
(0.013)
(0.015)
(0.013)
(0.007)
(0.011)
Panel B: Women
Married
Partnered
Partnered
but not
married
Divorced
or
Separated
Never in a
relationship
(1)
(2)
(3)
(4)
(5)
Total Effect of the Reform
0.042
**
-0.013
-0.052
**
0.006
0.010
(0.021)
(0.023)
(0.025)
(0.013)
(0.020)
Panel C: Men
Married
Partnered
Partnered
but not
married
Divorced
or
Separated
Never in a
relationship
(1)
(2)
(3)
(4)
(5)
Total Effect of the Reform
-0.002
-0.007
-0.002
-0.009
0.013
(0.017)
(0.014)
(0.018)
(0.009)
(0.012)
Notes: These are linear regressions. The sample here is permanent-contract private-sector workers in the French
ECHP between the ages of 20 and 49. There are 10371 observations in the whole sample (2797 individuals), 4248
observations in the sample of women (1175 individuals) and 6123 observations in the sample of men (1622
individuals). The total effect of the reform corresponds to β
2
in Equation (2). Standard errors in parentheses are
clustered at the individual level. All of the regressions include individual and year fixed effects, as well as age
dummies (in five-year bands), the number of children in the household (lagged), health status, weekly working
hours, the log of the monthly wage, and occupation and region dummies. *, ** and *** indicate significance at
the 10%, 5% and 1% levels respectively.
42
Table A5: The Rise in the Delalande Tax and Perceived Job Security – Robustness Checks
20+
employees
Spanish
sample
Italian
sample
Danish
sample
OLS
BUC Ordered
Logit
Donut
DiD
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Panel A: Whole Sample
Total Effect of the reform
-0.167
**
-0.018
-0.049
0.032
-0.110
***
-0.475
***
-0.142
***
(0.069)
(0.064)
(0.058)
(0.077)
(0.042)
(0.114)
(0.055)
Individual Time-Invariant Controls
.
.
.
.
Yes
.
.
Individual Fixed Effects
Yes
Yes
Yes
Yes
.
Yes
Yes
Observations
6657
8938
11373
4682
10371
16479
7523
Individuals
1740
3127
3411
1432
2797
2797
2067
Panel B: Women
Total Effect of the reform
-0.202
*
0.143
-0.065
0.167
-0.131
*
-0.379
**
-0.095
(0.109)
(0.103)
(0.105)
(0.145)
(0.070)
(0.187)
(0.087)
Individual Time-Invariant Controls
.
.
.
.
Yes
.
.
Individual Fixed Effects
Yes
Yes
Yes
Yes
.
Yes
Yes
Observations
2599
3084
4265
1739
4248
6546
3018
Individuals
691
1134
1340
572
1175
1175
860
Panel C: Men
Total Effect of the reform
-0.150
*
-0.092
-0.037
-0.019
-0.103
**
-0.549
***
-0.192
***
(0.090)
(0.081)
(0.069)
(0.092)
(0.052)
(0.144)
(0.073)
Individual Time-Invariant Controls
.
.
.
.
Yes
.
.
Individual Fixed Effects
Yes
Yes
Yes
Yes
.
Yes
Yes
Observations
4058
5854
7108
2934
6123
9933
4505
Individuals
1049
1993
2071
860
1622
1622
1207
Notes: These are linear regressions, except in column (6). The sample here is permanent-contract private-sector workers in the French ECHP between the ages of 20 and 49 (except in
columns (2), (3) and (4), where it is their Spanish, Italian and Danish counterparts). The total effect of the reform corresponds to β
2
in Equation (2). Standard errors in parentheses are
clustered at the individual level. The BUC ordered logit coefficients in column (6) refer to the log of the odds ratio and the number of observations is artificially higher due to the
estimation method. All of the regressions include year fixed effects, age dummies (in five-year bands), the number of children in the household (lagged), health status, weekly working
hours, the log of the monthly wage, and occupation and region dummies. The individual time-invariant controls are gender and education dummies. *, ** and *** indicate significance
at the 10%, 5% and 1% levels respectively.
43
Table A6: The Rise in the Delalande Tax and Perceived Job Security – Heterogeneity Analysis
Women
Men
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
Total Effect of the Reform
-0.150
-0.181
**
-0.203
-0.036
-0.034
-0.248
***
-0.219
***
-0.302
***
-0.137
-0.138
(0.117)
(0.089)
(0.135)
(0.138)
(0.138)
(0.091)
(0.071)
(0.089)
(0.106)
(0.106)
Interacted with:
Born after 1963
-0.018
0.015
(0.154)
(0.123)
High Education
0.116
-0.061
(0.170)
(0.138)
High Monthly Wage
0.071
0.194
(0.167)
(0.125)
Parent beforehand
-0.163
-0.137
(0.165)
(0.129)
1 child beforehand
-0.192
-0.148
(0.193)
(0.156)
2 children beforehand
-0.036
-0.109
(0.184)
(0.156)
3+ children beforehand
-0.688
-0.170
(0.433)
(0.200)
Notes: These are linear regressions. The sample here is permanent-contract private-sector workers in the French ECHP between the ages of 20 and 49. There are 10371
observations in the whole sample (2797 individuals), 4248 observations in the sample of women (1175 individuals) and 6123 observations in the sample of men (1622
individuals). The total effect of the reform corresponds to β
2
in Equation (2). Standard errors in parentheses are clustered at the individual level. All of the regressions
include individual and year fixed effects, as well as age dummies (in five-year bands), the number of children in the household (lagged), health status, weekly working
hours, the log of the monthly wage, and occupation and region dummies. *, ** and *** indicate significance at the 10%, 5% and 1% levels respectively.
CENTRE FOR ECONOMIC PERFORMANCE
Recent Discussion Papers
1777
Marc J. Melitz
Stephen J. Redding
Trade and innovation
1776
Holger Breinlich
Valentina Corradi
Nadia Rocha
Michele Ruta
J.M.C. Santos Silva
Tom Zylkin
Machine learning in international trade
research – evaluating the impact of trade
agreements
1775
Giuseppe Berlingieri
Luca Marcolin
Emanuel Ornelas
Service offshoring and export experience
1774
Facundo Albornoz
Héctor F. Calvo Pardo
Gregory Corcos
Emanuel Ornelas
Sequential exporting across countries and
products
1773
Nicholas Stern
Anna Valero
Innovation, growth and the transition to net-
zero emissions
1772
Paul Dolan
Christian Krekel
Ganga Shreedhar
Helen Lee
Claire Marshall
Allison Smith
Happy to help: The welfare effects of a
nationwide micro-volunteering programme
1771
Xuepeng Liu
Emanuel Ornelas
Huimin Shi
The trade impact of the Covid-19 pandemic
1770
Tito Boeri
Edoardo di Porto
Paolo Naticchioni
Vincenzo Scrutinio
Friday morning fever. Evidence from a
randomized experiment on sick leave
monitoring in the public sector
1769
Andrés Barrios-Fernández
Jorge García-Hombrados
Recidivism and neighborhood institutions:
evidence from the rise of the evangelical
church in Chile
1768
Stephen J. Redding
Suburbanization in the United States 1970-
2010
1767
Anna Valero
Education and management practices
1766
Piero Montebruno
Olmo Silva
Nikodem Szumilo
Court severity, repossession risk and demand
in mortgage and housing markets
1765
Ghazala Azmat
Katja Kaufmann
Formation of college plans: expected returns,
preferences and adjustment process
1764
Anna Valero
Education and economic growth
1763
John Van Reenen
Innovation and human capital policy
1762
Sarah Flèche
Anthony Lepinteur
Nattavudh Powdthavee
The importance of capital in closing the
entrepreneurial gender gap: a longitudinal
study of lottery wins
1761
Elodie Djemaï
Andrew E. Clark
Conchita D'Ambrosio
Take the highway? Paved roads and well-
being in Africa
1760
Sabrina T. Howell
Jason Rathje
John Van Reenen
Jun Wong
Opening up military innovation: causal effects
of ‘bottom-up’ reforms to U.S. defense
research
1759
Marcus Biermann
Remote talks: changes to economics seminars
during Covid-19
1758
Yatang Lin
Thomas K.J. McDermott
Guy Michaels
Cities and the sea level
The Centre for Economic Performance Publications Unit
Tel: +44 (0)20 7955 7673 Email info@cep.lse.ac.uk
Website: http://cep.lse.ac.uk Twitter: @CEP_LSE
