Mothers who work and the human capital of children
Mothers who work and the human capital of children
In almost all developed economies there has been an increasing presence of mothers in the employment market, including those with young children. This leads to a higher level of family income, part of which can be spent on the children, but also less time available. What is the result of these changes? How does the greater availability of consumer goods and the lesser time spent with children influence the human capital of children?
it's a real pleasure for me to introduce christopher flynn christopher flynn is a u.s teacher professor at the department of university mou and after getting a degree in social science he's also got a phd at the chicago university so he combines two sets of skills he published widely and he's also very familiar with our country he has been often to our country he spent times teaching and he also married an italian economist daniela del bocca who's an excellent economist and she also deals with a topic so closely related to his he has dealt with italy extensively analyzing the infrastructures of the italian system and comparing them to the american system he is also actively involved in the child center for household labor and demographic economics and he's very good in this field and i find this combination very interesting as i said it was a pleasure for me to greet him here for a good reason not just for him being a male economist having doubt for so many years so with women related topics and that's a rare occurrence per se in addition to that his professional and a personal marriage with daniela de bocca has been very fruitful they also work together on a family related issues so in my opinion but there is to be confirmed by him also then drawing inspiration and ideas from their day-to-day marriage married life i'm a working mother myself so when i was asked to chair this panel i felt very much called upon because we all know how difficult it is combine the two realms so your working and personal life so that's why i think that is a topic that concerns all of us closely enough and the same thing should be true to our partners but it's not always that true but we get there eventually before turning the microphone over to christopher i wanted to give you some information so background italy combines two negative records italy has won one of the lowest birth rates in europa 1.34 and at the same time also one of the lowest women employment rate 45.3 so we combine these two negative records in denmark women employment rate is 71 the birth rate is 1.8 in the uk employment rate is 65 for women a fertility rate of 1.8 and france has an employment rate of 58 percent then almost a 2 percent birth rate so italy has now entered the vicious circle we have fewer offspring because we have to work in a way but indeed we work lesser than our european counterparts so that's a background i would like to lay for the foundation and for christopher flynn's remarks speak italian for a while but first of all i wouldn't want to torture your beautiful language and then when i talk about economics i tend to think in english too and i'd like to be clear in my presentation there are these very uh subtle concepts i'd like to outline more clearly that's why i now switch to italian to english so the discussion today it's a i'm going to talk about some empirical results as different from a number of the lectures here when we talk about human capital social capital in terms of my own research i tend not to focus too much on those concepts because they're so hard to measure i don't think that when you talk about child development however you can escape thinking about these things so part of the part of the talk will be discussing some empirical results part of the talk will be criticizing a large number of these empirical results part of the talk will be laying out a framework part of which was initiated by gary bechter who starts speaking at 5 30 so i'll try to be brief and i'll try to provide some framework in which i think we should think about these questions and what the difficulties are in terms of coming up with with precise answers about the relationship between working mothers and child outcomes would say in english so a number of things is first of all one theme and this is going to be when you talk about human capital uh whenever you're talking about capital whether it be physical capital or human capital you're speaking of investment investment is a dynamic process you put out some money today or time today and you only get the advantages from this the returns from this in the future what else do we know they're highly uncertain what those returns will be typically same is true as investing in financial instruments or investing in children to some extent so it's a dynamic phenomenon when we talk about the relationship between mothers working in child outcomes how do we want to measure at what point in time mothers working when the child is two the child outcomes when the child is 15 what's the relationship between working at one period of time and child outcomes at some other point in time there's so one thing that people may not be used to that was possibly initiated with gary becker is talking about quality of children we'll talk about children like investment goods in some way so as economists we don't deny the role of sentiments and so on we all hopefully love our children and so on but we talk about trying to improve the quality of our children we try to improve it for reasons altruistic reasons we'd like them to be better in their life and for selfish reasons we get more utility more pleasure when our kids are better in some sense of the word what better means is controversial but in particular when we're talking about child outcomes there's so many things that determine child outcomes how can we isolate the role of the mother working right i mean it's a difficult point it's involved with us with the children the friends of our children with the schools they go to with so many factors how can we isolate this one factor of the mother working that typically is not an exogenous event in any event right it's determined inside the household and what do we mean by quality of children so as economists we want to try to at least give some concrete sense of of what we mean so this will show you what an early stage we are at in terms of analyzing these kinds of questions even though as as microeconomists household economists labor economists we've been thinking about these issues now for at least 50 years so measures of quality typically the things i'm going to relate to you in terms of test results well in terms of outcomes that are measurable are a very primitive things things like how a child performs on a standardized test a test that is administered to the child it takes something like 15 minutes to do so how they do in school measured at some point in time how many years of school do they end completing uh in italia what kind of schools do they go to classic uh licheo scientifico what behavioral outcomes do they have problems with drugs with relating to other people they smoke they drink alcohol sexual activity do they become teenage mothers and when mothers and what is their success in the market so if we measure the outcomes of these children 20 years after their mother possibly worked when they were three years old can we measure any discernible effect 20 years later in terms of these outcomes that we're going to see labor market and so on so why should we be concerned the state invests a lot in education unfortunately of children they invest a lot more unfortunately possibly in pensions and a number of other things i can only speak for the united states i won't generate any controversy with respect to italy but old people in the united states are certainly treated much better than young people um nonetheless here's a little table of some results um compiled by hanyushek i believe it was fairly recently um maybe five years ago these are expenditures and dollars as of let's say five years ago um on primary education and secondary education so you can see this is all in dollars so italy even though the per capita income in italy is a fair amount lower than the united states italy is spending about as much per student as the united states is spending significantly more than is the uk and about the same as norway another fairly well-known welfare state where there's lots of state expenditures on medicine pensions education and so on so in italy spend a lot of money on education we had alan krueger yesterday talked about what's the return from the investment that's a different story i'm not going to go into that so much why is it important to invest in children why does society care about both the fertility rate why should they care about the fertility rate and why should they care about the quality of the children being produced now here we might speak of quality from the society's point of view in many dimensions earning capacity because when you look at the budget in the end of the day the government looks at the budget how much money is being reported that's taxable that taxable income funds pension payments funds the health system fund school instruction so we governments are naturally concerned with both the number uh the size of the population and italy we know is strongly decreasing native-born population and the quality of the population are concerned about both where they should be in studying educational outcomes studying what we call an economics the life cycle how things generate over time one thing will be led back to over and over again here is how economists in the end say there's so much we can't explain that's just explained by what we call initial conditions how people started out the families they were born into other factors that happen to them matter but these initial conditions matter the most so it's important to have children especially start at a good point if they start at a good point they end at a good point therefore if we create an adult generation that's of high quality we're likely going to have the next generation of high quality as well so i argue there's a big multiplier effect expenditures now that increase the quality of the population increase the quality in the future to this multiplier effect another theme i'm going to come back to is one of data we're talking about extremely complicated phenomena extremely complicated much more complicated than physics people often say oh these economists are just trying they're frustrated physicists what we're trying to understand is much more complicated than what physicists are trying to understand as we're trying to understand people who have their own programs their own objectives who are constantly changing it's very complicated we can't hope to ever understand it without measuring it unfortunately there's only a few societies in the world right now where there's important efforts devoted to data collection italy unfortunately is not one of them when i talk about there's very few italy's in a big club but most of these results i'll talk about are pertain to the united states situation or the united kingdom while there are some detailed household surveys in italy that are done that are high quality the bank of the italia household survey it's a high quality survey but they're not focused on children or development or child development another theme of the talk is going to be this is such a complicated phenomena um anyone who knows my work knows i tend to work with complicated models and that's a lot of controversy about that in the profession uh i would say how can you understand complicated behavior without complicated models in the end i'm going to show you a very simple version of one and argue why it's really not that complicated and it's certainly not complicated compared to the reality that all of you make choices in every day in your household in your work life so i'm going to say that the associations that have been found between mothers working when the child's young and certain kinds of outcomes are just associations for the most part they tell you nothing about for example what would happen in italy i know this is going to be very frustrating but it tells you nothing about what's going to happen in italy if the participation rate of married women goes from 40 to 80 percent we just don't know we can't extrapolate these results to that situation i'm going to try to say what we might have to know or understand to be able to do that so i'm going to just very briefly summarize some of these results which i want you to think of as associations so these would be these would be results obtained from surveys sometimes panel data surveys that's data collected in the same households over time in which there'd be some measures of what mothers were doing when the child was young and then some repeated measures they keep following this household and they would see what would happen to this child 10 years later 15 years later four years later what have you and you're looking at the relationship what happened when the mother when the child was young mother working or not working and then what happened in terms of these child measurements later on so in general they find they try to hold constant a number of factors and then just add in whether the mother worked you know father's education mother's education and so on and then just put in there's a mother work or not now what's the association between that holding all these other things constant and some child outcome well this is probably this strategy is probably most often done using test score data how do these children maybe when they're six or seven years old maybe four years after you're measuring whether their mother worked or not when they were four years old you're trying to see how they scored in some standardized tests relative to another similar household in terms of measured characteristics in which the mother didn't work so in general in terms of test score results they tend to find that if the mother worked full time there was some minor negative impact on test scores of the child let's say three four years later that's if the mother worked now these effects often as in most social science tend to wear off over time so if we go back and measure this kid 10 years later maybe we don't find much of a difference if you look at women working part-time instead you tend to see some weak but positive effects on test scores um so seems to be full-time work has some cost measured in terms of a test score in terms of the short run at least part-time work doesn't seem to have any cost if anything possibly it helps a little bit um how about success in school so this would be a case in which once again we're measuring whether the mother works when a child was four years old or three years old and then now we're looking further down the road to the child finish the trail if we're talking about italy unfortunately we don't have the data for italy but we're talking about the united states did the child finish high school how many years of school were completed in the end there there's almost no effect these are longer term effects obviously this is 15 years after the mother was working while the kid was young some studies show positive effects they show positive effects this is an important point this study of o'brien and james for example when we talk in social science when we talk about what's the effect of a mother working full-time versus part-time versus not working on child outcomes the first question you should ask is conditional on what what are the mothers characteristics what are the father's characteristics what are the characteristics of the child care system what are the characteristics of the labor market because the answer will depend on all those things so what they found was o'brien james is if you just look at mothers who have a higher education versus those that have this is for britain to study versus those that have a lower education the ones that have the higher education their working had a beneficial effect on the probability that their children attained a higher level of education mothers who had a lower edge level of education it had no effect or a negative effect now why would that be that would be because of the kind of work they're doing what have you right the reasons why they're working whether the mother is serving as a role model for example and working in a very professional job or the mother just has to supplement low household income so it's frustrating but as economists we always have to say it depends other effects are on unemployment probabilities actually here tend to find that having a working mother when the child was young decreases the probability of unemployment in the future this can be because you know they just have more labor market skills they see how them how their mother worked in a labor market how she managed her time they're more likely to be able to find employment have positive employment skills also there's some findings that children of households in which the mothers worked when they were young had lower levels of psychological stress and so on bad psychological outcomes down the road take all these studies together all these different kinds of measures of child results outcomes what's the effect you find on early mother of the employment of the mother when the child was young nothing nothing constant nothing regular no regular effects does that mean that mother working doesn't have any impact on child quality of course it does of course it means there still is in a a very strong likelihood that mother's working has an effect in some way but in economics we're used to there being a number of different kinds of effects i just want to spend one minute on a study i think is particularly good that's done in this kind of literature by haverman and wolfe these used to be colleagues of mine a long time ago frantianifa at wisconsin they did a study using so this is something kind of unique to the united states there's a panel data study called the panel study of income dynamics that's been in a uh been run since 1968. so they followed mothers they've seen these mothers for 40 years and they've also followed their children and they see child outcomes so these children that were four years old they're now 40 years old they see how the entire employment history of the mother impacts than what happened to the children when they're adults young adults and they found that there were positive if anything positive effects on child outcomes of the mother working when they were young they found they attributed these effects to the additional income that the mother working provided and the mother serving as a role model in terms of labor market behavior in the interest of time there's some i'm going to just skip over these graphs because they're comparative analyses and i would argue as ellen kruger did yesterday we can't learn anything from this so i was just going to make a point we can't learn anything from these because as i said before the effect of changing the labor market structure or what have you on outcomes in a society are so dependent on societal institutions we have such crude measures that we can use to compare across societies that's rather useless to try to draw any uh inferences from the cross-cultural across national studies so what's the economist what do i view the economist possible contribution to looking at this to be well it's to supply an analytical framework to think about these questions that's the first thing by supplying an analytical framework we then raise the question of well here's the analytical framework which data should we have access to to be informative about these processes that we view as important in determining child quality so we want a little bit of theory so as i say gave a tribute to gary becker who's here so a lot of the contributions of gary becker in particular and others but gary becker in particular we're trying to uh we're applying concepts relatively standard concepts from the theory of the firm how to the household viewing the household as really well first of all let's say you can say what's the what's the firm have to do with the household well a couple of things are comparable a couple of things aren't first of all obviously at least in my household we don't maximize profit i don't know what we maximize to be honest but it's not profit typically theory the firm that's a nice thing about the firm when you study microeconomics the firm's complicated strategic behavior but they just care about money obviously that's not what the household cares about only the number of complicated questions that arise in looking at household behavior maybe the most fundamental of which is the household as we all know is comprised of a number of different individuals unless we live alone and all these individuals have different tastes different preferences and we have to find some kind of way to balance all these things to come to some kind of decision right that's the complicated thing firms behave as one maximize or maximize profits at least that's the standard story households don't so that's a big difference between households and firms what are the constraints that households operate under well they operate under income constraints obviously and they under they operate under wage constraints right we all know that there's some wage that someone like with us with our skills or what have you can command in the market that's acts as a constraint on our decision-making process in particular our time allocation process we have limitations strong limitation in terms of time we have a time endowment it's a very binding thing for most of us more than our income then we have environmental constraints and this is going to be very important by environmental i mean everything outside the household basically and particularly talking about child development we mean the schools we mean even the labor market that our children are going to enter into when they leave the school uh the structure of society number of things everything that in particular can affect children then we have to specify the choices available to the household what does this house will decide what doesn't it decide the main point of uh comparison between the firm and the household are production technologies these are very important so firms we know we say yeah what's a firm a firm is a production function it buys a bunch of stuff it combines it in a way and it produces something and we have the advantage in that case that the things that buys are observable typically if the firm lets us see it and the thing it produces or is also observable households much more complicated but the household we know is a system of production technologies there's a production this is a gary becker example there's a production technology to produce dinner at night right somebody does the spaza usually the same person the moya there's the space that does the cooking or whatever and out comes the dinner more importantly that's very important all these other production processes but we want to focus on the production process that produces these quality characteristics of children okay so here's a function here's a production function so q is some one of these measures let's say of child quality something available to us it's possibly measurable x are inputs just like a firm decides the raw inputs to buy to put in the production process the household decides how many toys to buy whether to pay for private school for the children whether to pay for a tutor um whether to pay for a certain kind of babysitter toys what have you um s are inputs provided by the environment the school system the labor market to some extent can affect the growth of the child quality in some way most importantly unfortunately for us is this variable e which is a child endowment children as we all are are very heterogeneous they start off with different endowments they start off at age zero with different endowments different capacities these capacities strongly influence what they're eventually going to become maybe more than a lot of these inputs x and s how can we measure e how can we measure e and in particular recognizing that the choices that parents make in terms of x they're deciding these investments the parents know something about e as economists we don't know much about e we see we infer e we try to infer e but this is one of the biggest problems for us of talking about value added what's the what's the value added of the mothers staying home with a child when the child's four years old as opposed to going to the labor market well we don't know exactly because the mother going to work when the child's four years old could have a lot to do with that child endowment e i could say my kid's fine learning fine i don't have any problem i go out into the labor market now i have a good babysitter i have good child care no problem or we could say no the child has lots of problems i feel like i should stay home with the child so that influences your choice of going to the labor market or not how can we as economists try to separate the value-added of those things when both are strongly driven possibly by that unobservable variable to us the child endowment i don't have an answer but that's the problem uh so i've already said what some of the what some of the inputs in are into this process uh for the for the family uh those things they said private child care um for the environmental inputs public child care um teachers etc etc etc etc this is just a quick thing to illustrate yes households have different sets of preferences um let m be the mother so utility of the mother could depend on the mother's own consumption this is how we think about as economists mother gets a utility what's her utility depend on stuff she likes going to the park this and that thing she likes buying dresses that's c sub m q is a child quality everybody likes to have good child outcome z are other things that the house the household buys that affect everybody the house that's a public good everybody benefits from that father use of f has different preferences so when we're looking at household behavior and trying to figure out these answers any deep questions we have to think about well what are households doing if we want to be able to figure out what's the value added what difference does it make whether the mother works when the child is four years old or not we have to first figure out what are households doing why would the mother go to work instead of staying home so so what's the household as a firm we say maximizes profits what is a household doing so the household is maximizing something subject to a number of constraints that's what economics is trade-offs you're trying to get to some objective you have all these constraints that limit your choices what's the best place you can get subject to all the constraints you face so this is just a this is a way a much simpler model you'd have to see so we'd say here the formally let's say this is a hotly debated question among economists of the household let's say here the households are maximizing the sum of the utilities of the mother and the father they're going to maximize that subject to choosing what well their choices are constrained they're constrained by the household by the child quality production function they can't consume everything themselves for example they can't be totally selfish if they consume everything themselves that won't leave anything to invest in the child will have a bad child outcome right so we're trading off utility i get from buying a new suit versus spending money on the child we're choosing things like investments in the children private consumption of the adults and the child public goods and most importantly when it comes to the question we're addressing the distribution of time that's another fundamental contribution of becker economics of time much more important in some ways in economics of how we distribute our income how we distribute our time so in this case we can think about a mother in particular at least at least deciding time for our own leisure time for sleeping time for working time for investing in the child time for playing with the child time for cooking time for doing other household production technology this is this is what we're faced with every day this is the complicated problem people are faced with now the the family isn't the only one that's making choices that affect the outcome of the children uh we already said why it's in the strong interest of the society to get good child outcomes and to have absolutely a fairly large number of citizens as well so the school can choose investments in the children and and they do and there's another question of whether these investments are efficient or not for the same level of money could we spend in a different way and get better outcomes that's a different story the main point is that the choices in the public sector and the household interact to produce child outcomes neither one is indeterminate you know you can do so much in the household but if the child goes to a terrible school a child's not going to learn anything if the child goes to a wonderful school and comes home to a terrible household which the parents fight all day whatever it's not going to be a good outcome there either as hillary clinton wrote a book once it takes a village it takes everybody working together to produce a good child so for example one study is a large-scale program in the united states called head start uh it's one of the largest public federal government interventions in child development they spent over seven billion dollars in 2005 and they enrolled 1 million children in this program so these are children of usually low-income households children at risk head start significant that they're starting behind they need some help to catch up so that they can do well in school so it's a preschool intervention to try to get these students ready to compete what do you find in these social programs typically in the united states or anywhere you typically find maybe some short run effects the question is do these short run effects persist over time um what they've what they've there's a lot of debate on this i'm not an expert in these kinds of programs there's a lot of debate but the certainly consensus is that without parent involvement these programs can't be successful for the reasons i said we can take the child child in a terrible household take them out expose them to a nice environment for four hours a day send them back to the terrible household for 10 hours it's not going to work so they found that so that's the issue there's the environment the head start program there's the household they're trying to modify the household to reinforce the intervention that they're getting in the actual formal schooling program the head start program so one thing that they're doing one interpretation of that is they're trying to change they're taking people uh in trying to also they work with the parents in this case they actually they parents come in or they go to the household of the students and they actually intervene with the parents as well so the parents are part of the program and what they're trying to do how i would interpret it in terms of our little model they're trying to change the production technology for child quality they're trying to teach parents there's a better way to increase child quality that they may not know about because they may come from dysfunctional households themselves that's the cycle of poverty um you know i'm better speed up so they basically so let's talk about the basic trade-off basic trade-off of mother working is what mother spends more time we said mothers and people forget about fathers they're also important components here but we're concentrating the mother i hope at least we are mothers are deciding basically time to work let's just say in our very simple model time to work leisure time to spend with the children which we think is also having this productive effect if the mother she's mother has a fixed time endowment if she spends more time in the market now the time left over is reduced obviously but what's the advantage well the advantage now of course is the household has more income than it did before she doesn't work this income can be spent on a number of things can be spent on a new suit or something but part of it can be spent on the children so we have these two countervailing effects the mother in general is going to she's going to reorganize her time it's not that she's not going to spend any time with her child of course she'll spend time with her child but in general let's say she spent a little bit less time with the child than she would if she doesn't work or much less time but she can offset that with getting other kinds of substitutes for her reduced time with the child the question is what substitutes are available this is the main question what substitutes are available to offset this probable loss of time spent with a mother so the point is how the substitutes that are available the substitutes so for example if there's high quality child care i can send the child to high child quality child care maybe it's a little bit less effective in producing child quality but maybe than personally spending time with the child as a mother but maybe not much less and it has so many advantages added advantages right of socialization other children and so on now if there isn't that if there's only low quality let's say private child care which is often the case in the united states at least available to low-income women who essentially have to work then that can be a very negative then we'd say there's no good substitutes for the for the mother's time unfortunately the mother is compelled to work because the family is a very low income they don't eat if she doesn't work so the investment in child quality is somewhat a secondary consideration so if you're close to your you know nona or whatever right you know you can substitute then that's a good alternative at least for part of the time um i'll plug my wife's work a little bit so in italy for example if you're going to talk about italy in italy for example there does exist a high quality child care system unfortunately it's not aligned with the labor market in the following sense this is work my wife's done i have to plug her work as we say um you find that you find that you know the public child care is open is high quality in most regions but is open for a limited amount of time the time it's open doesn't correspond to full-time work so you may have the child care may end at four o'clock or so and then your job may end at six o'clock you have to then find some substitute for two hours somewhere you have to make some logistic arrangement the whole question of writing that kind of formal model down is all these constraints that people face these logistic constraints just complicate everyone's life there's a very effective child quality child care system in italy already it just has to be redesigned a little bit to make it an incredibly uh incredible resource for promoting uh more entering the labor market of of married women and mothers so what's the effect of reduced child time it's you know reduced time with the mother as i say difficult to difficult to estimate difficult to estimate because it's estimating a production function in which unlike in the case of the firm we don't measure all the inputs and we don't measure very well the output we have these crude measures of child quality important issue of causality which way does it go do as i say do child outcomes i gave that example before do bad child outcomes compel the mother to stay at home uh or does the mother just go to work doesn't care about what happens to children then there's a bad child outcome because she's not there we don't know that is we would have to do a much more careful analysis to distinguish between those two hypotheses so the point is in the literature we have a incredible number of studies uh that are mainly almost always associations they're just associations they don't tell us anything about causality unfortunately i'm going to tell you about a couple of studies and then i'll you know wrap up uh a couple of studies that take an approach i favor which is building a relatively complicated model but to answer hopefully an interesting question this is a paper by a former colleague of mine wilbur vander klow and two co-authors lou and roz they did something very interesting there's a big debate we heard alan krueger talk a little bit about it yesterday what are the effects of school inputs on child on on child outcomes right on educational outcomes and what have you number of students per class and so on when i wrote down that production function i said okay the family is in this environment and this environment is characterized by this vector s and the family then given those characteristics of the environment chooses the inputs in the united states there's a lot of mobility there's a lot of geographic mobility much more so than italia they looked at the very interesting question of what if the household through mobility decisions was simultaneously choosing their inputs and the school inputs this is a worry you didn't drink too much wine last night this is out of focus totally out of focus it's just meant to illustrate in the united states these are spent expenditures by state per pupil in the united states in the northeast up here in new york and this you see the black those are the highest expenditures that's my state where we pay a lot of tax but taxes schooling is mainly funded by state tax in the united states right local tax not federal tax so you have a lot of heterogeneity and also within states you have a lot of heterogeneity in taxes for school and how much is spent on the students in the school when you and there's been a number of studies done of this when you buy a house you buy a school district in the united states it's very clear if you look at how houses for sale on the internet the united states will say in this school district that has a big effect in the housing price they made the point of what if these local states or even smaller areas urban areas they have different labor markets and they have different schools what if households choose a geographic location through a migration decision by doing that they're choosing the school they're choosing the school inputs and they're able to choose the labor market that they live in as well by doing this you can choose for example you could say i want to work i'll go to an area where they spend a lot of money on schools that'll substitute for my being away from the home more okay and they actually then estimated this very complicated model i won't go into it but they could then look at these effects and say taking into account we call the endogeneity the people are making these decisions that the school characteristics you face are not something predetermined or something you're choosing basically taking that into account they found that the effect of school characteristics on child outcomes was reduced tremendously didn't have much of an effect they found that the effect of mothers working on child outcomes went from being negative to being positive and basically that was all these reversal of effects was due to the simultaneous choice of the school inputs the public inputs the environment and the household inputs so that's a very interesting study there where they specify you know forms of preferences you know lots of things you can take issue with but they get these strong results that are very suggestive of things we should look at now i'll basically conclude uh plugging my own research this is uh some work done with me to brown so this is a little bit um foreign but it's about the environment in the united states you have to look at mechanisms as i say many of the results we have that i've been referring to come from the united states what's the main thing that impacts child outcomes in the united states household structure in the united states about one half of marriages end in divorce if you look at marriages with children it would be less than that okay maybe one-third of marriages with children would end in divorce let's say children that come from divorced households or possibly unwed mothers mothers that were never married have significantly lower outcomes controlling for lots of other factors than children who don't so we want to look at what the effect is in this case we're looking at the state the state we also the emphasis of this research is how the state in this case that if you raise the participation rate in italy may well be the case that the divorce rate will increase as well one will have more money to have more opportunities they'll have more chances to move to change region to take advantage of these opportunities there'll be more stress in the house so there may be more divorces so this is a thing specific to the united states but it may be a concern in the future also in a country like italy saying in this case marriage laws divorce laws have a strong effect on investment that's our model how divorce laws affect not only after a family after a couple gets divorced how it affects expenditures on the child but also how the state law that can either favor divorce or not can affect investment incentives in the child even while the couple are still married taking a step further back can even affect the decision of how many children to have so i won't really go through this in any detail except to say that you know we do find that divorce laws that are state-specific in the united states right they're not federal they're state-specific there's a lot of variation in the nature of the divorce laws in terms of how much the father has to pay the mother and all these things child support payments custody arrangements and so on these things do have a strong effect on investments in children and can't explain our measure is test score state law variation can affect divorce rates can infect investment incentives in children both after divorce and before divorce and can affect in fact the decision to divorce um so i just want to make a strong claim that only by articulating more models like this that specify a mechanism that would relate things like child employ the mother's employment to child outcomes how does it work what's the mechanism what's the production technology what are the constraints only by doing that it's going to involve more theory more data more econometric work but only by doing that can we advance our knowledge where we can really say something where i can tell you that if the participation rate in italy goes up of married women by 30 percent this is what you may expect to see for the children i can't do that now all we have is some correlations some associations we have some studies are trying to look at this in a more serious way with models and it's such a complicated phenomenon i would argue that's the only way you can understand a phenomenon like this thank you thank you very much for giving us a lot of food for thought one of the last few slides you showed headed brown and flynn will make the lots of italian women happy because it seems so that children of divorced families have a worse outcome but before opening the q a to the floor i've got a couple of very quick questions we said in italy women work less than elsewhere for many reasons for instance as many observers and economists stay the lack of child care services in italy especially for children younger than three a new york times study just pointed out then the children who go to chelsea services when they are younger than three might have some focusing issues in the future and they do not tend to go so much with child care services in italy first of all because there are not many but also because italian mothers do not think a lot about these czech services so does it mean that there is not such a dire relationship between a crash i didn't talk about social capital law or or or jim heckman's research program on these companies i'm just saying because it relates to that we tend to concentrate on cognitive outcomes uh outcomes uh how smart are you in terms of doing a test uh how successful are you in the labor market that we think is related to cognitive outcomes well it's also the claim is certainly among developmental psychologists is strongly related if you came from let's say in the old days let's say in italy if you had a family of three you're talking about a fair amount of interaction at least among the few right but now if this family size is one or two staying at home you're typically staying at home without other children around there's a lack of possibility for socialization so i think high quality child care is definitely at least in at least a few hours a day is definitely indicated for almost any child uh as absolutely increasing their quality in the long run right through developing interactive skills with their children hopefully other children of different backgrounds to some extent these are all important things in life both in terms of just being a citizen but also in terms of labor market success or productivity it's not just how smart you are on a test it's how you work with other people and so on these are developed only when you start socializing and these skills are learned at early ages there's a lot of evidence that that if you take a kid and isolate that child until they're four years old and then put them in it's too late they have to there's has to be strong socialization early on in life let me ask you another question we said the italian birth rate is quite low but what's worse we tend to have many families with silver children it is now a ruler most of the women in my age range between 30 and 40 tend to have an only child but there is another interesting observation because women interviewed say 18 21 months after their first child when the time has come for the second child most of them say they would not be at the burden whereas if you interview them before they get married and have a child and ask them how many would you like to have 61 percent say two one quarter says three children only 12 percent says one child only so as it turns out our aspiration for motherhood is great but then it just clashes against the practical obstacles the paradox of italy basically right and this mediterranean pair is also through in spain also through in in greece you have relatively low married participation rates and yet you have low fertility rates as well usually the theory of the household fertility would say okay you make one choice of the other specialize in the labor market specialize in producing children you don't specialize in anything so this is it's kind of a paradox why is that um i don't know it's i i don't have a good response i have a child with my wife here that's uh it was went to italian schools and so on costs of having children high in italy it could be before marriage people don't really appreciate the cost or see the constraints but having children that live at home until you're until they're 30 years old median age of leaving home 28 years old something like that certainly seems strange than american but this is related i'm primarily a labor economist and so i tend to think of all a lot of these issues stemming from labor market inflexibility housing market inflexibility and so on i think a lot of these issues if you free up things particularly the labor market some of these things will tend to take care of themselves people won't be young people be getting jobs earlier they'll be able to support themselves leave home earlier that'll be less of a weight on the yep not too much well but it depends you know you have to really go to a freer labor market and then you know yeah these are extrapolations difficult to do extrapolations when something like that hasn't existed at least for a long time but certainly one would think that a lot of these problems uh are related to other rigidities in the italian environment that impact the family strongly the mother being able to find work you know the mothers can't find part-time work they would like to find part-time work like it's not available um the in the in the now some of the corporations are providing uh uh a zealand needle right so at firms and these are also in the united states proving to be very popular and very successful they the firm doesn't cost the firm anything the employees are willing to pay for it all themselves and it has all the benefits for the firm reducing turnover and so on it's a compromise if a firm wants full-time workers you say at least put my child there for part of the day i can interact with the child i can monitor the child i can monitor what's happening to the child this is a good compromise so we need innovations like this the one of the arguments is we need innovations to change the choice set of people another thing is tele telling the water you know working at home long distance commuting working on the phone you know working over the internet stay at home sometime with a child to the extent you can have tasks at work that allow you to do some work at home with your child that's also something that can promote growth and fertility but we need to change the set of options available to people a problem what about the weight of the cultural problem based on instant is that data we know that fathers contribute very little to household work also when mothers work outside their home so when we ask working mothers whether they are helped by their husbands or not 63 percent of them declare to be helped uh not to be helped at all and this is a household economist as i am part-time would argue that this is related to the labor market uh clearly the late the way we view household bargaining which i didn't talk about here obviously was that you know we come in you have different diversity different preferences in your husband and so on i have different one than my wife how do we resolve these issues well we bargain at least implicitly the theory of bargaining in economics says basically well who has bargaining power you have this preferred outcome i have this preferred outcome where's the compromise well the compromise is related to your outside options outside options are in large part related to work you can find so if for example the women don't have any labor market experience or whatever husband says you don't do what i want by via you know a woman has few options you tend to think in a situation where the women has experience and and this is documented to some to some extent using the consumption data in households if the woman's contributing more or has more of an income endowment than the man the consumption for the same total household income keep household income fixed fifty thousand euro a year give thirty thousand to the woman in one household twenty thousand to the man and one and switch that in another household the consumption patterns are different one reason could be because they have different levels of bargaining power the one with thirty thousand euro has more bargaining power when the women start to enter the labor market more they will have other sources of income other options other alternatives even if they don't exercise them it's not that everybody's going to get divorced and run off it's just to have that option even if it's not exercised that you bargain against so right now because the participation rates are so low and so on this is one reason that uh theoretically why you would say there's women have relatively low power in the household about the policies which have to be deployed in order to improve uh female labor lesina and kino to economists whom you certainly know proposed to levy fewer taxes on female labor of course finding measures so that you can get those taxes as in another way are you convinced about the goodness of that or not no it's a it's a viscous side i've talked to both of them about it it's a thing that doesn't convince me it's a there's lots of reasons let me just put it the most basically the most basic reason is what i said before there's a lot of problems with the italian labor market this is trying to go to a sort of second best solution so there's all these problems forget about them let's not deal with them let's have different tax rates on husbands and wives but certainly this is going to change the balance of bargaining power in the household that's one thing right this changes has a huge welfare effect on single people in the population here's a poor guy who's unmarried and now he just loses income he doesn't have a wife to offset that right um so it's it's a kind of thing that you know i just don't think is that well thought out it's it's you want to really free up the labor market give everyone equal opportunity there's nothing that says women should be in the labor market they said oh one of their one of their rationales for doing this is well this is going to increase the participation rate we don't know whether increasing the participation rate is a good thing or not we don't have the the estimates we don't know people's preferences we can't say what the right level participation should be in italy we have to do a lot more research to figure that out now if there tends to be a difference between what the right level would be and what the current level is then we can talk about policies to change that but certainly policies of differential taxation based on gender that would be about the 300th policy down the list for me there's lots of other more direct policies that make everything more open for everyone that's what you want to do you don't want to have a more of a sexual warfare right you want to have policies that are clear that promote you know equal access to everybody issued by the ministry for labor 10 days ago state that mothers aged 30 to 40 years old make out 36 percent of cases of discrimination on the on their job meaning that when they go on parental live and then they go back to their job but they are discriminated against due to the fact that they left their job force when they talk about their policy proposal they talk about it as changing prices it doesn't change prices only the market can set a price unless you're going to have price control unless you're going to have the government set prices that doesn't work too well so they can set tax policy but then the market is going so if there is discrimination that's not going to solve the discrimination problem all employers will do will adjust their wage offers to men and women so in the end your take-home pay can be the same as it is before you know what i mean that doesn't change the off they can change the offer this is equilibrium effects that you have to consider the employers if they want to discriminate they can undo this effect if they want to okay uh there should be a roving mic which can be used for that purpose scene that in your study you always made a reference to children aged three or more or older and i was surprised by the fact that no reference was made to the working situation of women with children ages 0 to 3. i just didn't point that out because i went through summer so there is that's a real feature as looking at what happens when i just kept mentioning that because it got stuck in my mind but most of the focus is on zero to one then there's a one to two if you have the whole history you can look at every year and the impact of every year if you want so there is that so these effects are there they're probably a little bit stronger for zero to one than they are for three um but they also tend to be so some of these results i was showing actually refer to zero to one you would think the most important age so even there effects negative effects of full-time employment of the mother tend to wear off over time so there's not even a strong indication there as i said before i would always say that that these are just associations so almost anything can happen but there's no strong association there even for very young children my wife remained at home for two or three for a couple of years so she she remained at home for two for three years with the first child and then the second and third child where they attended nursery school and my my wife continued working and i have not found an analysis about the results on children of working mothers related to the psychological condition of the mother who remains at home to take care of that her child and suffers socially because she does not have any social life because she does not work either she feels isolated have you considered this i should be clear i don't really work on that problem per se i don't really work on that particular problem but these are very important that's that's related to the point i was making when you say what's the impact of the mother were staying at home with the child instead of working the other side depends on the conditions under which she's staying at home right if the mother's staying at home because they can't afford child care it's not available if the mother's depressed or whatever that's going to have a very negative effect on the child so it's important these are all important mediating factors in determining these things these are all important unfortunately there's not one data set around that would study both the employment history of the mother the psychological comport of the behavior the stress of the mother on the mother whether she's depressed or not there are psychologists do studies of that unfortunately they're not typically integrated with data that economists would use as measures of outcomes in the end about like unemployment or or wages or what have you but these are all extremely important factors then you have to ask though if the mother's depressed how did that affect in the child staying home how did that affect the decision to have the child stay at home i mean it's complicated but yeah people people certainly are aware of this it's a little bit it's a question a little bit of discipline boundaries there's psychology there's economics they're both important but there's no one data set for example that combines both these kinds of things but they're extremely important i also have a question you talked about the need to have so you said that we need to have a free labor market as if it were a panacea but actually it is not easy to achieve that in italy is because if you have a free labor market a totally free labor market then you have other problems and of course you cannot change things completely instead of making a huge leap you should simply advance by adjustments i'm german i live in italy so i know the nordic situations that to say and also the southern one and that is important i believe in order to understand fully a culture so leaving aside a theory i would say that what you said could be applied specifically our labor market is very complicated uh if you change one thing completely then perhaps this is detrimental for something else it is not politically possible clearly it is clearly so you should tell proudly yes but italian politics is very complicated some hope i mean i think if you look at the case if you look at the case of spain spain has made significant progress italy some spain has made significant progress you can always debate okay these are fixed term contracts whatever they're not like having a permanent job forever okay but almost nobody except me because i have tenure but almost nobody has a permanent job anymore if people if everyone starts to understand that everyone it'll be more accepted accepted it'll be easier to actually implement labor market reforms than going from this transition period we say oh this guy he's got this job he can never be fired no matter what he does well these kinds of jobs are eventually they're being eliminated you don't read about it every day but they are being eliminated in the public sector as well as in the private sector for sure so it takes time i'm not saying you can do this overnight but but it's going to happen it's going to happen eventually and what it does thank you all for being here and see you again at 5 30.