family values and the regulation of labor

Introduction The model Family ties and the demand for labor market regulation The persistent effects of family ties Conclusion

Family Values and the Regulation of Labor

Alberto Alesina(Harvard University)

Yann Algan(Science Po, OFCE)

Pierre Cahuc(Polytechnique, CREST)

Paola Giuliano(UCLA)

December 2011

1 / 58


Introduction

Differences in labor market regulation across the world

Rigid labor markets survive

Why?

2 / 58


We argue that family values play a crucial role

What is the relation between family values and labormarket regulations?

Flexible labor markets requires mobile workers

Otherwise, firms can take advantage of the immobility ofworkers and extract monopsony rents

In cultures with strong family ties, moving away fromhome has utility costs

Thus, individuals with strong family ties support regulatedlabor markets to reduce the monopsony power of firms,even though they produce lower employment

3 / 58


There are complementarities between the strength offamily ties and the stringency of labor market regulation

Weak family ties � flexible labor markets

Strong family ties � rigid labor markets

This leads to two equilibria

Weak family ties and labor market flexibility

Strong family ties and stringent labor market regulation

4 / 58


What can we explain ?On the positive side

why certain countries have more regulated labor marketsthan others

strong inertia in labor market regulations due tointergenerational transmission of family values

On the normative side

why rigid labor markets with high unemployment are notnecessarily less effi cient than flexible labor markets withlow unemployment

why it is so diffi cult to reform labor markets in manycountries

5 / 58


Related research

Cultural values and economic outcomes (Alesina andGiuliano (2008, 2010), Algan and Cahuc (2010), Algan etal. (2010), Fernandez and Fogli (2009), Giuliano (2007),Guiso, Sapienza and Zingales (2006))

Cultural values and institutions (Tabellini 2008)

Our approach: about family values and labor marketregulation

6 / 58


Outline

1 The model2 Family ties and the demand for labor market regulation3 Persistent effects of family ties

7 / 58


Results

Individuals with strong family ties migrate less, face awage and employment penalty, and demand more labormarket regulations (within country and second generationimmigrants)

Family values inherited from the country of origin beforeWWII are positively correlated with the stringency oflabor market regulation today

Family structures in the Middle Ages are related tocurrent desire for labor market regulation

8 / 58


The modelThe setup

There are two goods: labor and a numeraire goodproduced with labor

Continuum of individuals of mass one

Individuals are uniformly located on the [0, 1] line

Identical, risk neutral and no preference for leisure

Utility = consumption + valuation of family relations

3 stage static model

9 / 58


Stage 1: At birth, each individual is located on the [0, 1] line,on a point where his parents live

Individuals "choose" family values which can be1 either with strong family ties2 or with weak family ties

The choice of family values is irreversible

10 / 58


The share of individuals with strong family ties is denotedby σ

Person with weak family ties are indifferent between livingin their location of birth or elsewhere

Strong family ties yield utility{∆(σ) > 0 if immobile−∆(σ) if mobile

∆(σ), the valuation of strong family ties, increases with σ(social norms more influent when more spread)

11 / 58


Stage 2: People vote to choose labor market regulationaccording to the majority rule

There are two possible types of labor market regulation1 Labor market flexibility (i.e. laissez-faire)2 Regulation which comprises

minimum wagejob protection

12 / 58


Stage 3: Firms offer labor contracts

When a worker is employed in his initial location, hisproductivity y is drawn in the uniform distribution on theinterval [0, 1]

All workers can find jobs with productivity 1 in locationsdifferent from their initial location

Job protection constrains firms to keep all employeeswhose productivity is above a threshold value denoted byR ∈ [0, 1].Job protection entails deadweight losses c, that is theproduction of a worker who draws the productivity y isequal to y − c.

13 / 58


The modelThe solution

Model solved backward

In stage 3, the labor market is either regulated orflexible, and the share of individuals with strong familyties is given

14 / 58


Flexible labor market

Individuals with weak family ties get a wage equal to 1because they are perfectly mobile

UWF = 1

Individuals with strong family ties get a wage equal to 1 ifthey leave their initial location, but the cost of moving is2∆(σ), so their reservation wage is equal to:

max[0, 1− 2∆(σ)]

15 / 58


If ∆(σ) > 1/2 they stay and their utility is ∆(σ)If ∆(σ) < 1/2]

1 If their productivity is larger than their reservation wage,they are hired and their utility is 1− ∆(σ)(1− 2∆(σ) + ∆(σ))

2 If their productivity is lower than their reservation wage:1− ∆(σ)

The utility of individuals with strong family ties is

USF = max[∆(σ), 1− ∆(σ)]

16 / 58


Rigid labor market

The government regulates:{minimum wage wjob protection individual gets a job if y ≥ R

Deadweight losses associated with job protection:productivity is y − c, c > 0, c small, instead of yy uniform distribution on [0, 1]

17 / 58


Individuals with weak family ties get the expected utility

UWR = (1− R)max(1− c,w) + R(1− c).

The expected utility of individuals with strong family tiesis

USR = (1− R)max[w + ∆(σ), 1− c − ∆(σ)] +Rmax[∆(σ), 1− c − ∆(σ)]

18 / 58


In stage 2, people vote to choose the labor marketpolicy according to the majority rule

The share of individuals with strong family ties, chosen instage one, is given

The median voter can have either strong family ties orweak family ties

2 policies: either regulation or flexibility

19 / 58


Individuals with weak family ties obtain{UWF = 1 under labor market flexibilityUWR < 1 under labor market regulation

Therefore, individuals with weak family ties always preferlabor market flexibility

Vote → labor market flexibility if the share of strongfamily ties σ < 1/2

20 / 58


Suppose σ > 1/2,The expected utility of individuals with strong family tieswith flexible labor market is

USF = ∆(σ)

When the labor market is rigid:

USR = (1− R)w + ∆(σ)

Thus, when σ > 1/2, a median voter with strong familyties prefers a rigid labor market

21 / 58


The optimal labor market regulation:

max(w ,R)

USR = (1− R)w + ∆(σ)

subject to the zero profit condition∫ 1

R(y − c − w)dy = 0

22 / 58


SolutionR = c and w =

1− c2

Workers with strong family ties get the expected utility

USR =(1− c)2

2+ ∆(σ)

which is larger than ∆(σ), the utility they would get if thelabor market was flexible

In conclusion: the outcome of the vote is{regulation if σ > 1/2flexibility otherwise

23 / 58


In stage 1 individuals choose their family values withperfect foresight

The utility gains of choosing strong family ties ratherthan weak family ties are:

Γ(σ) ={max[∆(σ), 1− ∆(σ)]− 1 if σ ≤ 1/2∆(σ)− 1−c2

2 if σ > 1/2

24 / 58


Properties of equilibriaUnder the assumptions that ∆(1/2) > 1/2 and∆(0) < 1 there are two stable Nash equilibriaThe equilibrium with strong family ties and rigid labormarket has lower employment and lower production thanthe equilibrium with weak family ties

0 1/2 1 σ

A∆(σ)−1

∆(σ)−(1/2)

B

Gains associated with strong family ties

Share of population with strong family ties

25 / 58


Wefare comparison

"Standard" result in economics, BUT the equilibriumwith strong family ties provides higher welfare than theequilibrium with weak family ties IFF ∆(1) is suffi cientlylarge

The economy can be coordinated on an equilibrium inwhich the labor market can be

either too rigid

or too flexible

26 / 58


The modelThe dynamics of family values

Paternalistic parents wish to transmit their own values totheir children (Bisin and Verdier, 2001)

Each individual lives for one period, and has payoffs asbefore

Children inherit family values with probability p and arefree to choose their family values with probability 1− p

27 / 58


Sequence of events repeated in each period with aninfinite horizon, with a fraction pσt−1 contrained to havestrong family ties and a fraction p(1− σt−1) to haveweak family ties.

If σ0 > 1/2p, then the median voter chooses to regulatethe labor market and every individual is better off withstrong family ties. σ1 = 1− p(1− σ0).

In period t the labor market is regulated and the share ofindividuals with strong family tiesσtt→∞

= 1− pt(1− σ0)→ 1

σ0 < 1− 1/2p the same type of reasoning shows thatthe economy has flexible labor market in period t andthat σt

t→∞→ 0

28 / 58


If p > 1/2 and if the initial share of individuals withstrong family ties is

large → rigid labor market

small → flexible labor markets

29 / 58


The model yields two main predictions:

1 Individuals with strong family ties prefer more stringentlabor market regulation, because they want to staygeographically immobile and they want to be protectedfrom the monopsony power of firms

2 The strength of family ties can persist over time and canhave persistent effects on labor market regulation iffamily values are transmitted across generations

30 / 58


Family ties and the demand for labor marketregulation

We document two facts

1 strong family ties predict strong demand for job securityand wage regulation

2 positive cross-country correlations between the strengthof family ties and labor market rigidity

31 / 58


Data

World Values Survey: 4 waves, 1981-2000: family ties,preference for job security, demand for regulation

Survey of Health, Ageing and Retirement in Europe(SHARE): family ties (child-parent geographicalproximity)

International Social Survey Program: demand for wageregulation

Word Bank: employment protection

Aghion, Algan, Cahuc (2008) and ILO: minimum wage

32 / 58


Family ties

1 Share of adult population in parental house2 How important is the family in one person’s life (with 4being very important and 1 not important at all)

3 1) One does not have the duty to respect and loveparents who have not earned it; 2) Regardless of what thequalities and faults of one’s parents are, one must alwayslove and respect them.

4 1) Parents have a life of their own; 2) It is the parents’duty to do their best for their children even at theexpense of their own well-being;

5 Importance of obedience as value trasmitted to children6 The goal of my life is to make my parents proud7 Fraction of adult children who live 5 Km or closer to theirfamily; average age at which the young adult left home,frequency of contacts with their parents

33 / 58


Demand for regulation

Preference for job security (WVS): “Here are some moreaspects of a job that people say are important. Pleaselook at them and tell me which ones you personally thinkare important in a job?: Good Job Security?”(1 if jobsecurity is mentioned and zero otherwise)

Demand for wage regulation (ISSP): "Here is a list ofpotential government actions for the economy: Controlwages by law?" (from strongly agree (4) to stronglydisagree (1))

34 / 58


Cross-country correlations

35 / 58


36 / 58


37 / 58


38 / 58


The persistent effects of family ties

We show that:

1 Second generation U.S. immigrants inherit the familyvalues and the behavior of their country of origin

2 The strength of family ties before WWII is correlated withlabor market rigidities at the beginning of the 21st century

3 Family structures in the Middle Ages are correlated tocurrent desire for regulation

39 / 58


The persistent effects of family tiesIntergenerational transmission of family values

General Social Survey to study the impact of familyvalues on attitudes of U.S. immigrants towards labormarket regulation (since 1977)

March Supplement of the Current Population Survey ofthe U.S. to study labor market outcomes of immigrants.(since 1994)

Both surveys provide information on the birth place andthe country of origin of the respondent

40 / 58


For both attitudes and labor market outcomes, we runthe following OLS or probit (depending on the nature ofthe left hand side variable) regressions:

Yic = α0 + α1family_tiesc + α2Xi + δs + εic

Yic is our variable of interest for an immigrant i whoseforbear was born in country c

Xi are individual controls

family_tiesk is the measure of strong family tiescalculated from the WVS in the country of origin

δs state or county dummies whenever possible

41 / 58


Evidence from the CPS

Second generation immigrants outcomes from the CurrentPopulation Survey

We also repeat the exercise using the Censuses 1940,1960 and 1970

42 / 58


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

Share adult pop. in parent. home 0.101***(0.017)

Respect parents 0.058**(0.029)

Obedience 0.071***(0.014)

Parents proud 0.029***(0.007)

Parents’ responsibility 0.039*(0.023)

Family important 0.077**(0.034)

Observations 74852 95936 76389 76157 95962 96140

43 / 58


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

Share adult pop. in parent. home 0.002(0.018)


Obedience 0.033***(0.011)


Parents’ responsibility 0.035***(0.012)

Family important 0.070***(0.019)

Observations 39707 51081 40559 40451 51073 51150

44 / 58


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


Respect parents 0.161***(0.057)

Obedience 0.159***(0.058)


Parents’ responsibility 0.096**(0.046)


Observations 47135 60658 48167 48060 60709 60807Rsquared 0.27 0.27 0.27 0.27 0.27 0.27

45 / 58


Evidence from the General Social Survey

Evidence on demand for regulations for second generationimmigrants

Three questions on labor market regulation:

1 Whether job security is the first, second, third, fourth orfifth most important thing in a job

2 Whether the government should regulate wages3 Whether the governement should support decliningindustries

Each regression controls for a quadratic in age, years ofeducation, gender, income, employment and maritalstatus, number of children abd region fixed effects.

46 / 58


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

Share adults pop. in parent. home 0.917***(0.117)


Obedience 0.432**(0.165)


Parents responsibility 0.628***(0.214)

Family important 0.722***(0.260)


47 / 58


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



Obedience 0.549**(0.218)


Parents responsibility 0.791**(0.305)



48 / 58


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



Obedience 0.421*(0.210)


Parents responsibility 0.609**(0.231)



49 / 58


The persistent effects of family ties

Family values are transmitted across generations

We show the persistent effect of family values

1 Countries with stronger family ties before WWII havemore rigid labor markets in the beginning of the 21stcentury

2 Exploit differences in family types across Europeanregions dating back to the Middle Ages, by using theclassification of Emmanuel Todd (1990)

50 / 58


Inherited values before WWII and labor marketregulation today

We detect family values before 1940 by looking at thefamily values inherited from their country of origin byU.S. immigrants whose forebears arrived in the U.S.before 1940 (second gen. born before 1940, third gen.born before 1965 and fourth gen. born 1990)

Country of origin fixed effects OLS regression for answersto the question about family ties in the GSS

51 / 58


The question is: “Do you spend social evening withrelatives?”

1 almost daily2 several times a week3 several times a month4 once a month5 several times a year6 once a year7 never

52 / 58


53 / 58


(1) (2)Dependent variable Firing costs State regulation of

minimum wageInherited family ties .554** .024**before 1940 (.206) (.010)Civil law origin .001 .005

(.142) (.009)Scandinavian origin .158 .021

(.179) (.013)German origin .058 .007

(.146) (.010)Ln(population) .002 .003

(.043) (.002)Observations 24 23Rsquared .48 .55

54 / 58


Medieval family structures and current desire forlabor market regulation

We document a correlation between medieval familystructures and current family arrangements

We show that medieval family structures are related tocurrent desire for regulation

55 / 58


Todd’s classification of family types

Todd (1990) provide a classification of family types alongtwo dimensions:

1 Vertical relationship between parents and children("liberal" versus "authoritarian" family)

2 Horizontal relationship between siblings (based oninheritance rules)

Four family types:

1 Absolute nuclear family : liberal and unequal2 Egalitarian nuclear family : liberal and equal3 Extended family : authoritarian and unequal4 Communitarian family : authoritarian and equal

56 / 58


(1) (3) (2)VARIABLES Living with

parents(adult children)

Most importantfor a job:

job security

Government:help to

protect jobs

Egalitarian Nuclear Family 0.090*** 0.090 0.12(0.029) (0.08) (0.09)

Extended Family 0.061** 0.14** 0.102(0.022) (0.06) (0.082)

Communitarian Family 0.092*** 0.15** 0.262**(0.028) (0.07) (0.102)

Observations 17854 19460 8659Rsquared 0.347 0.090 0.167

57 / 58


Conclusion

Countries with strong family ties favour a host of labormarket regulations.

Indivuals coming from strong family ties societies alsohave lower geographical mobility, wages and employment.

Low employment associated with labor market rigiditycould be the price that certain countries choose to pay inorder to enjoy the benefits of family ties

This conclusion illustrates the interest of accounting forcultural values in economics and their interaction withinstitutions

58 / 58

family values and the regulation of labor

Documents