Scoial Mobility

8/14/2019 Scoial Mobility

1/19

Princeton University

Intergenerational Social Mobility: The United States in Comparative PerspectiveAuthor(s): Emily Beller and Michael HoutSource: The Future of Children, Vol. 16, No. 2, Opportunity in America (Fall, 2006), pp. 19-36Published by: Princeton UniversityStable URL: http://www.jstor.org/stable/3844789

Accessed: 06/01/2010 15:56

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

http://www.jstor.org/action/showPublisher?publisherCode=princetonu.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Princeton University is collaborating with JSTOR to digitize, preserve and extend access to The Future of

Children.

http://www.jstor.org
http://www.jstor.org/stable/3844789?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=princetonuhttp://www.jstor.org/action/showPublisher?publisherCode=princetonuhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/3844789?origin=JSTOR-pdf


2/19

Intergenerationalo c i a l M o b i l i t y :T h e U n i t e d S t a t e s i n ComparativePerspective

Emily Bellerand Michael HoutSummaryEmily Beller and Michael Hout examine trends in U.S. social mobility, especially as it relates tothe degree to which a person's income or occupation depends on his or her parents' back-ground and to the independent contribution of economic growth. They also compare U.S.social mobility with that in other countries. They conclude that slower economic growth since1975 and the concentration of that growth among the wealthy have slowed the pace of U.S.social mobility.In measuring mobility, economists tend to look at income and sociologists, occupation. The con-sensus among those measuring occupational mobility is that the average correlation between theoccupations of fathers and sons today ranges from 0.30 to 0.40, meaning that most variation inthe ranking of occupations is independent of social origins. Those measuring income mobilitytend to agree that the elasticity between fathers' and sons' earnings in the United States today isabout 0.4, meaning that 40 percent of the difference in incomes between families in the parents'generation also shows up in differences in incomes in the sons' generation.Beller and Hout show that occupational mobility increased during the 1970s, compared withthe 1940s-1960s, but there is some evidence to suggest that by the 1980s and 1990s it had de-clined to past levels. Existing data on income mobility show no clear trends over time, but in-creases in economic inequality during the 1980s made mobility more consequential by makingeconomic differences between families persist for a longer time.In international comparisons, the United States occupies a middle ground in occupationalmobility but ranks lower in income mobility. Researchers have used the variation in mobility tostudy whether aspects of a country'spolicy regime, such as the educational or social welfare sys-tems, might be driving these results. There is as yet, however, no scholarly consensus about thesources of cross-national differences in mobility.

www.futureofchildren.orgEmilyBellers a Ph.D.candidatensociology t the Universityf California,erkeley.MichaelHout s a professor f sociology nddemog-raphy t the Universityf California,erkeley.

VOL. 16 / NO. 2 / FALL2006 19


3/19

Emily Beller and Michael Hout

bM~ bostAmericanshink t un-fair when things they can-not control limit theirchances to succeed in life.Particularly un-Americanis the notion that circumstances of birth set

life on a course that may be hard to alterthrough one's own efforts. So, rags-to-richesstories are popular, and crowds cheer for theunderdog. Academic research on social mo-bility goes beyond the stories and the dramato quantify the link between circumstances ofbirth and economic success, both for thepopulation as a whole and for important andinteresting groups within it. Sociologists andeconomists put numbers to patterns by com-paring the social and economic success ofAmericans with an absolute standard that iscompletely free of traces of birth and with arelative standard that is based on recent ex-perience or the current experience of othercountries. On the absolute standard, Ameri-cans' occupations and incomes are tied muchmore closely to their parents' occupationsand incomes than they would be in a worldwhere circumstances of birth were irrelevantfor adult success.1 On the relative scale, tiesbetween people's current occupations and in-comes and those of their parents are aboutwhat they have been over the past twenty-five years, but substantially weaker than theywere in the early 1960s.2Social mobility from one generation to thenext is the difference between a person's cur-rent income, wealth, or occupation and thatof the family that raised her.3 An opportunitystructure promotes social mobility if it allowspeople to escape poverty while limiting thedegree to which those who grow up in privi-leged homes get advantages throughout theirlives. Growth promotes mobility, too, by rais-ing everyone, regardless of background,above the level of that background. In this ar-

ticle we will focus most on the opportunitystructure because scholars have written moreabout it. But it is important to keep in mindhow important growth can be. Nearly every-one who grew up in the Great Depression ex-perienced substantial upward mobility inadulthood. It was not that America was moreequitable when the children of the Great De-pression grew up than it was before or hasbeen since; it was that the nation recoveredfrom its economic collapse and thereforemost people were much better off. Socialmobility should not be confused with in-equality, which refers to differences amongpeople in wealth, income, and occupationalstatus at any point in time. Social mobilitywould not matter in a society in which therewas no inequality. Parents would have no ad-vantages to bequeath to their children, andno one would care where they ended up. Butwhen inequality is great, social mobility mat-ters a lot. The advantages of rising to the topare large, and the consequences of remainingstuck at the bottom are much more serious.Social mobility is high if the opportunitystructure is open-that is, if the barriers andadvantages associated with a person's back-ground are few. But openness of that sort isnot the only way to spur intergenerationalmobility. Mobility is also high if growth isstrong and widespread enough to makeeveryone better off. The opportunity struc-ture, in the form of barriers and advantages,is symmetrical in the sense that in the ab-sence of growth, removing a barrier thatmight block a person who starts low also im-plies removing an advantage from a personwho starts high. Growth, on the other hand,can-in President John F. Kennedy's famousphrase-lift all boats. If growth is wide-spread, it creates new opportunities that canlift a person who starts low without knockingdown a person who starts high. But slow

20 THE FUTURE OF CHILDREN


4/19

Intergenerational Social Mobility: The United States in Comparative Perspective

growth reduces social mobility, as does aclosed opportunity tructure.4Growing inequalitydoes not necessarilyin-crease or decrease the prevalence of socialmobility,but it does increase the differencebetween the upwardlymobile andthe down-wardly mobile. When inequality increases,extreme incomes, occupations,and amountsof wealth(highandlow)become morepreva-lent, and fewer people occupythe middle ofthe distribution.So an upwardlymobile per-son hasfarther o rise and a downwardlymo-bile person has fartherto fall in a more un-equal society. Also (and this is a little lessintuitive)an increase n inequalityover a per-son's lifetime increases the probabilitythatsomeone who startslife in extremeprivilegewill staythere and (simultaneously)ncreasesthe probability hat someone whose parentswere poor will also be poor.Those increasesin immobility are offset, though, by a de-creasein the probability hatsomeone whoseparentswere aboutaveragewill end up nearthe average(because risinginequalityelimi-nates positions near the average). The in-creasedimmobilityat the extremesand mo-bilityin the center do not implya strongerorweakercorrelationbetween circumstancesofbirth and adulthood;they follow from thedefinition of inequality-more extreme out-comes, fewer averageones.It is possible to talk about social mobilityingeneralterms,but most researchers ocus onone of five specific formsof mobility:educa-tional mobility, occupationalmobility,wagemobility, amilyincome mobility,and wealthmobility.Eachhas its owninterestingproper-ties. We focus on two types: family incomemobility and occupational mobility. Thefirst-typically the domain of economists-isthe extent to whichan adult's orfamily's) el-ative income or rank n the income distribu-

tion is similar o his or her father's orfather'sfamily's)relative income or rank. The sec-ond-most often the province of sociolo-gists-is the extent to which the status ortype of job a personwindsup with resemblesthatof his or her fatheror mother.We review research on income and occupa-tionalmobility,examiningchangesin the op-portunitystructureandgrowth,as well as the

Growing inequality does notnecessarilyincrease ordecreasethe prevalenceofsocial mobility,but it doesincreasethe differencebetween the upwardly mobileand the downwardlymobile.effects of inequality. We first try to quantifythe extent of intergenerational occupationaland income mobility in the United States. Wethen compare estimates of mobility in theUnited States today with evidence both fromthe American past and from cross-nationalcomparisons. Where possible, we discuss theintergenerational persistence of wealth andproperty as well. Intergenerational educa-tional mobility is another fascinating topic,but it is beyond our scope in this review.

Measuring IntergenerationalSocial MobilityImportant differences in the concepts of oc-cupational and income mobility can help toexplain how it is possible that mobility in onedomain might be greater than mobility in an-other. People's incomes vary significantlyeven if their jobs share the same occupational

VOL. 16 / NO. 2 / FALL 2006 21


5/19


category. Analyses of occupational mobilityand analyses of income mobility provide dif-ferent pictures of people's prospects, becausethey ask different questions. Intergenera-tional persistence in occupational status isnot a good proxy for persistence in income,and vice versa; a person who is upwardly mo-bile occupationally does not necessarily enjoya higher relative income than his or her par-ents (and vice versa).5

Analysesof occupationalmobility and analysesofincomemobilityprovidedifferentpictures of people'sprospects,becausethey askdifferentquestions.In addition, analysts investigating occupa-tional and income mobility ace different im-itations and use differentmethodologies.Onthe one hand,occupation s easier to measurethan income becausepeople remember theirparents'occupationsreliablyandwith a highdegree of accuracy,whereas dollar amountsare much harder to recall, and most peopleplainly do not know their parents'incomes.Inflation erodes the value of the dollarovertime, too, further complicatingthe task ofevaluatingparents' ncomes, even if they areknown. On the other hand, occupationscanbe hardto rank,whereas income is straight-forwardlyscored in dollars (or the relevantlocal currency).In addition, researchersin-terested in occupationalmobilityoften wantto measure the componentof mobility hat isindependentof growth,whereasincome mo-bilityresearchersdo not typicallydistinguishbetween the two.

Researchers interested in occupational mo-bility must first come to grips with the prob-lem of how to rank occupations, getting be-yond the qualitative detail of specific jobdescriptions to arrive at useful categories orscores. Some solve the problem by groupingoccupations into relatively large classes. Oth-ers rank them on a scale from 0 to 100.6 Inthe first approach, researchers gather occu-pations into several broad classes, such asprofessionals (for example, doctors andlawyers), skilled trade workers (for example,electricians and carpenters), or the self-employed, and then create a matrix that al-lows them to compare each person's occupa-tion with his or her father's occupation. Whilethis approach reveals details of which occu-pations are linked across the generations andwhich are not, its results are hard to summa-rize unless the categories are clearly ranked.

Ranking allows the straightforward estima-tion of an overall intergenerational correla-tion between the ranking of a person's occu-pation and that of his or her father. Acorrelation of 0 implies that a person's occu-pational rank is completely independent ofthat of his or her parents, and therefore thatthere is perfect mobility between ranksacross generations. A correlation of 1 impliesthat ranks do not change from generation togeneration. The correlation that a researchercalculates for a real society places that societyon the continuum from perfect mobility tocomplete rigidity.In principle, one could use an intergenera-tional income correlation to measure incomemobility as well as occupational mobility, butin practice researchers (usually economists)typically measure income mobility slightlydifferently. They look at the strength (persis-tence) of the relationship between parents'and children's income in percentage terms;



6/19


thatis, they ask how much (whatpercentage)of the income differencebetween familiesinone generationpersistsinto the next genera-tion. This estimate is called the intergenera-tional elasticity.If the elasticity s 0.4, for ex-ample,theywould concludethata 10percentdifference in parents' ncome would lead to a4 percent difference in offspring's ncomes.The advantageof usingthe intergenerationalelasticity, romthe researcher'spoint of view,is that it cancapturethe amplifying ffects onmobility of rising income inequality,or thedampeningeffects of fallingincome inequal-ity (the formula or the intergenerational or-relationdiscards his useful information).Onthe low extreme,an elasticityof 0 describes asocietyinwhichfamilyeconomicbackgroundis not at all relatedto adultincome, whereasan elasticityof 1 describesa societyin whicheach personends up in exactly he same eco-nomicpositionas her or his parents(justlikethe correlation).But unlike the correlation,the elasticity s unbounded,so one could, inprinciple, discover that two people whostarted ife in families10percentapartendedup 15 percentapart(if the elasticitywas 1.5).Mobility is the complement of the elastic-ity-a low intergenerational lasticity trans-lates to a high mobilityrate,and a high elas-ticitytranslates o a low mobilityrate.

SocialMobility n theContemporaryUnited StatesHaving defined our terms and introducedsome of the analyticaldistinctions that re-searchersuse,we turnnow to the heart of thematter:how much mobilityAmericanshaveexperienced from their youth till now. Wediscuss occupationalmobilityfirst, and thenturnto income andwealthmobility.Intergenerational Occupational MobilityOne way to assess occupationalmobility inthe United States s to categorizeoccupations

into a few classes and then measure the ex-tent of class immobility,downwardmobility,and upward mobility between generations.Using this technique,we analyzednationallyrepresentativedata on men and women bornafter 1950.7We distinguishedsix generaloc-cupational categories in descending order:upperprofessionalor manager, owerprofes-sionalor clerical,self-employed,technicalorskilledtrade, farm,and unskilledand serviceworkers.8Among men, 32 percent were im-mobile (theiroccupationwas in the samecat-egory as their father's), 37 percent wereupwardlymobile,and 32 percentwere down-wardlymobile. Fifteen percent of the mobil-ity was driven by structuralchange in theeconomy,or economicgrowth-more profes-sional jobs and fewer farmjobs were avail-able to sonsthan to theirfathers; hat also ac-counts for why upward mobility was morecommon than downwardmobility.Women'smobility patternsreflect the gender segrega-tion of the laborforce, as well as opportunityandgrowth.Amongwomen, 27 percentwereimmobile,46 percentwere upwardlymobile,and 28 percent were downwardlymobile.MostAmericansregardsalesandclericaljobsas better than most blue-collarjobs, so themillions of blue-collar men's daughterswhowork in stores and offices are upwardlymo-bile (justnotverymuch).Thatparticularypeof short-rangeupwardmobilityaccounts forthe fact that more American women thanmen areupwardlymobile.Table 1 shows the data for men from whichthe aboveestimateswere generated.It showsthe outflow of sons from each class back-groundcategoryto currentoccupationalcat-egories (in percentages).The bold diagonalentries show the percentage of men fromeach class backgroundwho stay where theybegan; this "stickiness" s greatest for themost and least advantagedclass background

VOL. 16 / NO. 2 / FALL 2006 23


7/19

Emily Beller and Michael HoutTable 1. Intergenerational Occupational Mobility of Men Born between 1950 and 1979Percent

Destination:on'soccupationOrigin: Upper Lower rofessional Self- Technical Farm Unskilledfather'soccupation professional andclerical employed and skilled sector and service TotalUpperprofessional 42 24 7 12 0 15 100Lower rofessionalandclerical 29 27 7 17 0 20 100Self-employed 29 18 16 19 0 18 100Technical ndskilled 17 19 6 30 1 26 100Farmector 14 11 8 17 13 37 100Unskilled ndservice 16 17 6 22 1 38 100Source:General ocialSurveys, 988-2004.

categories. If we consider the column per-centages instead (that is, the share in eachclass from each backgroundcategory [datanot shown]),it is striking hat the proportionof immobile incumbents is almost alwayshigher than the proportiondrawnfrom anyother classcategory.The most extreme exam-ple is that 66 percentof men in the farmclasscame from a farmbackground.Another way to assess occupational persis-tence is to examineintergenerational ccupa-tional correlations.As noted, these correla-tionsdifferdependingon whichcharacteristicof occupationsis the focus of research.Forexample,the intergenerational orrelationofthe prestigeof fathers'and sons'occupationsis lowerthan the correlationof the educationlevel associatedwith their occupations.9Oneof the most commonlyused scales for meas-uring occupations s the socioeconomicstatusindex (SEI), which providesa rank for eachoccupation.Average ntergenerationalather-son correlations n the SEI and similar in-dexesare in the neighborhoodof 0.35 to 0.45,implyingthat some 12 to 21 percent of thevariation in sons' occupations can be ac-counted for by fathers' occupations.10Thelarger estimates are mostly from the early

1960s; the smaller ones are from the 1980sand 1990s.11For the men in table 1 we calcu-late the correlation to be 0.32.12

Assessing whether a given intergenerationalcorrelation or mobility rate reflects a low orhigh degree of occupational mobility requiresdetermining an appropriate reference forcomparison. Complete mobility is neitherplausible nor, arguably, desirable, given thatsome of the factors leading to the intergener-ational persistence of social position, such ascognitive ability or work effort, seem accept-able-that is, fair.13 Complete immobility isalso implausible. In the absence of accepteddefinitions of what constitutes low or highmobility, one strategy is to contrast the U.S.estimates with those from a range of compa-rable countries. Comparisons with other in-dustrialized countries (to which we turn later)support the prevailing idea that occupationalmobility in the United States is reasonablyhigh, as does the finding that U.S. occupa-tional persistence does not extend past twogenerations.14But one complication in analyzing occupa-tional mobility using either SEI correlationsor class mobility tables such as table 1 is that



8/19


there is no straightforwardway to incorpo-rate two parents'occupations nto the inter-generationalcorrelationsor classbackgroundcategories. Thus occupational mobility re-search s limited,forthe mostpart,to studiesof father-child orhouseholdhead-child) oc-cupationalpersistence. The case of incomemobility, o which we turnnext,is instructive:intergenerationalassociationsappear to beweaker when calculations do not includeboth parents' earningsand other sources offamily income. Of course, occupationalsta-tuses do not add together the way incomesdo, so we use multivariateregressionto cal-culate the total associationbetween familybackgroundand occupational tatus. For themen in table 1 we find the multiplecorrela-tion is 0.38.Intergenerational Income MobilityThe currentconsensusamong researchers sthat intergenerationalpersistence,or elastic-ity,between fathers'and sons'earnings n theUnited States lies at about 0.4 on the 0-1scale describedabove.15The persistencebe-tween total childhood family income andadult sons' family income or personal earn-ings is even greater, n the range of 0.54 to0.6.16An elasticityof 0.54 meansthat,for ex-ample, a 10 percent differencebetween twofamilies' ncomes is associatedwith a 5.4 per-cent difference in their sons' earnings.Thecorrespondingelasticity between family in-come and daughters' earnings is lower, at0.43. When analysts focus on marriedwomen, the elasticity between total child-hood family income and adult daughters'total family ncome is 0.39. The same elastic-ity for married sons is 0.58. These gender-specific patterns occur because men con-tribute about70 percentof family ncome,onaverage,and because there is an associationbetween childhood family income andspouses' income.17

The conclusion that the intergenerationalelasticity between father's and son's earningsin the United States is as high as 0.4 wasreached only recently, and these estimatesmay understate the true income persistence,as more recent research has tended to raiseestimates of the elasticity. Early estimatesplaced the father-son earnings elasticity at 0.2or lower-indicating substantially more eco-nomic mobility than an estimate of 0.4 wouldimply. 18

The upward trend in estimates reflectsmethodological improvements, probably notreal-life trends. In the 1970s researchers hadto estimate the size of intergenerational elas-ticities from one year of data about fathersand one year of data about sons. The newer,higher estimates accumulate income overfive or more years for both fathers and sons.19Another improvement has been the recogni-tion that a person's age affects his or herearnings. Calculations based on young peo-ple's earnings understate the persistence thatis seen when we observe people during theirtop-earning years.20 New, logically similarcorrections are resulting in a further increasein the estimated elasticity to 0.6.21 We havedoubts about this higher estimate for father-son earnings persistence. The theory behindaccumulating data is that each family or per-son has a "true"income level but minor upsand downs (and measurement errors) pro-duce variations around the true value thatlower the elasticity. In the short run, this the-ory is credible. Over longer and longer spans,it becomes harder to believe that there is justone true value.Elasticities are good indicators of a society'saverage level of intergenerational economicpersistence, but they do not provide much in-formation about mobility patterns. Mobilitymatrices that give the probability of chil-

VOL. 16 / NO. 2 / FALL 2006 25


9/19

Emily Beller and Michael HoutTable 2. Intergenerational IncomeMobility: Probability of Son's QuartileGiven Parent's QuartilePercent

Destination:on's incomequartileOrigin:arent'sincomequartile First Second Third FourthFirst 42 28 19 12Second 26 29 27 19Third 18 24 29 29Fourth 15 18 25 40Source:AdaptedromH.Elizabetheters,"Patternsf Intergener-ationalMobilityn Incomeand Earnings," eviewof EconomicsandStatistics74, no. 3 (1992): 460.

dren's economic position conditionalon fa-thers' or familyposition providea more de-tailed picture of intergenerationalmobility.Similar o the patternof occupationalmobil-ity shownin table 1, the income mobilityma-trixin table 2 shows that economic immobil-ity is highest among children whose familyincomes fall in the top or the bottom quar-tiles of the earningsdistribution.22 his pat-tern is consistent with other U.S. economicmobility matrices, which show the greatestrigidityat the extremesof the distribution.23That overallmobilityrates are higher in themiddle of the income distributiondoes notnecessarilymeanthatthe impactof family n-come is weaker n the middle thanit is at thetop and bottom of the distribution-by defi-nition, people at the bottom of the distribu-tion can experience only upward mobility,and the reverse s true atthe top of the distri-bution. People in the middle have theprospectof movingeitherup or down.Besides lookingat descriptive ncome mobil-itymatrixes,anotherwaythat researcherscanlearn more about mobilitypatternsthan theaverage intergenerationalelasticitycan pro-vide is to calculate separate estimates for

people who start ife at low, middle,andhighpoints on the income distribution for theirparents' generation.Some evidence suggeststhat the effect of childhoodfamily ncome onadult income is strongerat the high end ofthe father'searningsdistribution han at thelow end.24A differentquestion s how the effect of fam-ily backgrounddiffers along the son's earn-ings distributionrather than that of the fa-ther. Such analyses suggest that father'sincome is more persistent among sons withlow earnings than among sons with highearnings.25This implies that opportunity orupward mobilityis more equal than the op-portunity for downwardmobility-presum-ably,advantagedparents are able to protecttheir children from downwardmobility,butchildren from more disadvantaged back-groundsdo have a greaterchance of upwardmobilitythan the intergenerational lasticity(which,as noted, describes the average evelof mobility)wouldsuggest.Intergenerational Wealth MobilityFinally,how does wealthmobilitycompare ooccupationand income mobility?First,thereis substantialintergenerational ersistenceinfamilywealth;the correlation s in the neigh-borhood of 0.50.26Wealth is importantbe-cause its distribution is far more unequalthan the distribution of family income andbecause it seems to have greatereffects onother aspectsof familywell-being, especiallyhomeownershipand investmentin children'seducation.27The disparity n wealth not only persistsbe-tween the generations, t mushrooms.With-out a cushion of inheritedwealth, emergen-cies hit harder;andpeople who have no nestegg have to let opportunitiespass by.Becauseof a wealth deficit, African Americans are



10/19

Intergenerational Social Mobility: The United States in Comparative PerspectiveTable 3. Intergenerational Wealth Mobility, 1979-2000Percent

DestinationuintileOriginuintile Poorest Second Third Fourth Richest TotalPoorest 45 27 11 9 9 100Second 24 35 20 14 7 100Third 11 20 35 21 13 100Fourth 7 11 23 33 25 100Richest 5 6 9 25 55 100Source:LisaA.Keister,GettingRich:America's ewRichandHowTheyGotThatWay Cambridgeniversityress,2005), table2.10.

more vulnerable to shocks and less able tocapitalize on breaks than whites with thesame income, so the next generation will in-herit less too.28The wealth gap will not closeany time soon; wealthy people's assets grow ata rate that approximates that of the New YorkStock Exchange.29 Furthermore, inheritedwealth can put families in better neighbor-hoods and school districts than they could af-ford if they had to rely exclusively on theirincomes.At the very top of the wealth distribution, in-novations in computer and telecommunica-tions technologies created new fortunes inthe 1980s and 1990s and pushed new peopleto the top of lists like the Fortune 400. As in-teresting as the extremely high tail of thewealth distribution is, however, those 400wealthy people are not, by definition, repre-sentative of their 300 million fellow citizens.Thus most analyses of wealth mobility focuson the wealth differentials in representativesamples of American families and house-holds.30Wealth mobility in the United Statesresembles occupational and income mobilityin a few key respects.A number of familiar features show up in thewealth mobility matrix in table 3. First, ineach row, the main diagonal entries are the

largest, indicating the relative strength ofpersistence and mobility. They are somewhathigher than comparable figures for family in-come. Second, the richest and poorest quin-tiles are less mobile than the middle groups,as was true for income mobility.Contemporary and PastU.S. Mobility RatesThe best way to evaluate contemporary mo-bility is to compare it with the past. Strongevidence from several approaches shows thatbarriers to mobility weakened substantiallyfor American men from the 1960s throughthe mid-1980s, thus increasing opportunity.Trends since that time are not yet clear, sincethe children born from the late 1980s onwardhave not yet entered the labor force. How-ever, assuming there has been no change inmobility patterns, the rise in income inequal-ity over the last few decades means that eco-nomic differences between families will per-sist for a longer time.

Occupational MobilityBarriers to mobility weakened substantiallyfor American men during the 1960s; the in-tergenerational correlation in occupationsfell from roughly 0.4 to 0.3 on the scale from0 to 1.31 These trends continued for menuntil the mid-1980s; the correlation declined

VOL. 16 / NO. 2 / FALL 2006 27


11/19


another 30 percent in ten years.32 Data setsfor women are available only from the 1970s;they show that women's and men's intergen-erational occupational mobility differed be-cause the occupational distributions differed,but that the occupational mobility rate, ad-justed for differences in distribution, did notdiffer for men and women. Trends forwomen resembled those for men as well.33

Mobilitythat results rom amore open opportunitystructure-a decreasedadvantageto upper-statusbackground-leads to bothdownward and upwardmovement.One majorreason for the declining correla-tion between fathers' and sons' occupationswas the climbing share of men with collegedegrees-occupational opportunity for col-lege graduates s quitehigh, thoughof coursethe likelihood of college graduation s itselfhighly dependent on class background.34State interventions on behalf of disadvan-taged groups may also have contributedtothe increased mobility.35By itself, the declining effect of fathers' occu-pation should have increased overall occupa-tional mobility in the United States between1972 and 1985, but for these later years theincrease in opportunity was counteracted bya slowdown in economic growth. (Con-versely, the growth slowdown would have re-sulted in less social mobility in the UnitedStates were it not for the opposite trend inopportunity.) The combined effect of in-

creased opportunityand slower growth keptoverallmobilityrelativelyhigh through1985.Some evidence suggests that the trend to-wardgreater opportunity lowed or reversedfor men born after 1970, but datalimitspre-vent firmconclusionsat this time.36Historically, ver the courseof the twentiethcenturyeconomicgrowthproducedmoreup-ward than downward movement. Mobilitythat results from a more open opportunitystructure-a decreased advantageto upper-status background-leads, however, to bothdownwardand upwardmovement.37So thechanges in the American mobility patternsince the early 1970s have resulted in moredownward mobility, especially for the off-spring of the most privileged classes, andsomewhat less upward mobility. Table 4shows our calculationsof the amount anddi-rection of men's occupational mobility;thatis, the shareof men upwardlymobile,down-wardlymobile, or immobileby yearof birth.The earliest cohort (bornin the 1930s) firstentered the labor force in the 1950s andreached its top earning potential around1980; the latest cohort (born in the 1970s)firstentered the laborforce in the 1990s andwill reach its top earning potential around2020. Almosthalf of the cohort born in the1930s was upwardlymobile;only one-fourthof thatcohortwas downwardlymobile.Sincethen, fewer men havebeen upwardlymobileand more have been downwardlymobile.Among men born in the 1960s and 1970s,downwardmobilityis almost as prevalentasupward mobility. Immobilityrose acrossco-hortsfromone-fourth o one-third.Income MobilityLess is known about change over time inintergenerational income mobility. Someresearch has suggested a decline in theintergenerational elasticity among recent



12/19

Intergenerational Social Mobility: The United States in Comparative PerspectiveTable 4. Amount and Direction of Men'sOccupational Mobility, by Year of BirthPercent

Upwardly DownwardlyYear f birth Immobile mobile mobile Total1930-39 26 49 25 1001940-49 28 45 27 1001950-59 31 39 30 1001960-69 31 35 33 1001970-79 33 35 32 100Source:General ocialSurveys, 988-2004.

cohorts-indicating higher mobility-but thedecline is not statistically significant (that is,it might appear by chance because of insuffi-cient data), and using alternate data gener-ates the conflicting finding that the elasticitymay have increased.38 The most convincingfinding is that there has been no change overthe past century in intergenerational incomemobility.39Because the data for the analysis of trendsover time in income mobility are limited, it ispossible that trends did change but that datacould not detect them. One way around thatdifficulty is to study change over time in howa set of family background indicators broaderthan father's earnings or family income mayaffect adult income. One such study usedparent income, parent education, parent oc-cupation, family race and ethnicity, familystructure, number of siblings, and region toinvestigate how background affected eco-nomic outcomes over time. It found that theeffect of family background on men's eco-nomic outcomes declined during the 1960s,then remained constant during the 1970s,1980s, and 1990s. Although the effect of fam-ily background remained constant over thosethree decades, the economic gap betweenadvantaged and disadvantaged men in-creased because economic inequality in-

creased during this period. On the otherhand, the effect of family background onwomen's outcomes declined between the1970s and 1990s. Coupled with the increasedeconomic inequality, this meant that the gapsin women's outcomes remained constant overthe period.40A yet broader approach investigates the jointimpact of state residence, ancestry, and fam-ily income on men's economic outcomes andfinds a down-up cycle. The intergenerationalcorrelation between social origins and adultincomes was fairly constant from 1940 until1960, fell substantially in 1970s (indicatingincreased mobility), and then returned toprevious levels in the 1980s and 1990s. Theintergenerational elasticity, on the otherhand, declined between 1940 and 1980 butincreased during the 1980s and 1990s. Themeasures are different because the elasticityis sensitive to income inequality, which fol-lowed the same trend over time-decliningfrom 1940 to 1980 and then rising during thenext two decades. The trends in the intergen-erational correlation considered togetherwith the elasticity suggest that income mobil-ity was not abnormally low in the 1980s. Butbecause economic inequality increased, theconsequences of a historically normal degreeof mobility were greater, and a greater shareof the economic differences between familiescould persist for a longer time.41U.S. Social Mobility Rates inInternational ComparisonDirect comparisons of intergenerational so-cial mobility in different countries are diffi-cult to make, because both data availabilityand research methodologies differ fromcountry to country. Until recently it has beenhard to compare occupational mobility in theUnited States with that in other countries be-cause of differences in occupational coding,

VOL. 16 / NO. 2 / FALL 2006 29


13/19


but new research using comparable codingshows thatthe United States is at the medianin terms of opportunity:owerthan the mostopen nations, such as Sweden, Canada,andNorway;but higher than the more rigid na-tions, such as WestGermany, reland,or Por-tugal.42Other research suggests that Italy,France, and Great Britain are among theother societies that now display the lowestcomparativemobilityrates.43Recentresearchattributes hese internationaldifferences n occupationalmobility o two di-mensions of educational inequality-theshareof adultswho attendcollege andequal-ityof educationalopportunity the strengthofthe effect of family backgroundon educa-tional attainment).44Opportunityis muchgreateramongcollege-educatedadults of dif-ferent class backgroundsthan it is amongadultswith less education.The United Stateshas one of the highestlevels of college atten-dance,but also a relatively ow level of equal-ity in overalleducationalopportunity.Althoughthe United States occupies a mid-dle ground in internationalcomparisonsofoccupationalmobility, ts ranking n terms ofincome mobility is lower. Both the UnitedStates and Great Britain have significantlyless economic mobility than Canada, Fin-land, Sweden, Norway, and possibly Ger-many;and the United States may be a lesseconomically mobile society than GreatBritain.45Much of the higher intergenera-tionalelasticity n the United Statesis due togreater income immobility at the top andbottom of the earningsdistribution; he mo-bilityof middle earners ooksmoresimilar othat in the other countries.46Two explanations or these internationaldif-ferences in income mobility appearparticu-larlycompelling.First,it seemsplausiblethat

high income inequalityat a given time couldcause a high intergenerationalpersistenceofeconomic status. The United States andGreat Britain have high income inequalitycoupled with low income mobility,whereasScandinavian ountries displaythe oppositepattern.Canada,however,casts doubton thisexplanation,because it has relativelyhigh in-come inequality coupled with high incomemobility.Second, given the limited ability of low-incomeparents o invest n their children's d-ucation,it is possible that progressivepublicpolicies towardeducationfinancingcould ex-plainwhy some countrieshavehigherrates ofeconomic mobility.Research shows that dif-ferences in educationfinancingalone do notexplain mobility differences between coun-tries,but educationfinancing s an importantpart of the explanation, ogetherwith otherfactorsthat differ between countries,such asthe earningsreturn to education(how muchanotheryearof education ncreasesone'searn-ings)and the heritabilityeithergeneticor en-vironmental) f income-predictiveraits.47Higher economic returns to education andlower levels of public financingof educationdecrease intergenerationalmobilitybecausewhen incomedependson education,childrenfrom low-income families need to go to col-lege to be upwardlymobile. But with lesspublic financing of education, fewer low-income children can go to college. Both fac-tors also increase ncomeinequalityat a giventime, because lowerpublic financingof edu-cation owersequalityof educationalopportu-nity, while higher returns to education in-crease the earnings gap between more andless educatedpeople. These patternsmayex-plain why most countries either have low in-come inequalityandhigh income mobilityorhigh income inequalityand low income mo-



14/19


bility. The economic returns to education arehigher in the United States than they are inother countries, which may explain thestronger intergenerational income persis-tence. The role of heritability also implies thatdiffering degrees of assortative marriage in acountry-differing rates of couples from simi-lar economic backgrounds marrying-will af-fect intergenerational mobility. Marital sort-ing increases intergenerational inequality.48ConsequencesandPolicyDiscussionThe research literature-and by necessityour review of it-focuses on the way theeconomy affects mobility. To us, though, thatleaves unexplored the most profoundchanges affecting families and their potentialto promote or hinder their children'sprospects: the way family structure itself af-fects both income and occupational mobility.Sophisticated mobility studies came of age inthe era when most people grew up in a rela-tively stable family structure, anchored bythe earnings of a paternal breadwinner. Tyingthe circumstances of birth to the income ofthe family breadwinner greatly simplifies thetask of quantifying social mobility. And aslong as that was an appropriate simplifica-tion, researchers made significant progress.But changes in family structure since the1970s have contributed to growing economicinequality. Two-earner families have signifi-cantly higher standards of living than single-parent families.49At least part of the connec-tion between parents' incomes and thesuccess of their adult children is presumablydue to the disrupting effects of familybreakup. To be sure, researchers have consid-ered family structure in important papers andbooks over the past forty years.50 But so far

they have not been able to take fully into ac-count what they know about family structurewhen they measure social origins. This gap inthe research reduces our confidence in cur-rent estimates of social mobility. In particular,it appears that a father's absence from hisfamily can reduce the correlation between hisoccupation and the success of his children.He is more able to pass on the advantages ofhis accomplishments if he lives with his fam-ily.51Until this issue gets sorted out, it will behard to say what family policy is most appro-priate for promoting social mobility.What we can say is that greater opportunityand increased growth promoted social mobil-ity during the 1960s and 1970s. The impor-tance of socioeconomic background for adultsuccess declined during those decades, whileeconomic growth further boosted all job seek-ers and earners. Research has tied the declin-ing importance of socioeconomic backgroundto better educational opportunities and equalopportunity legislation and its enforcement.52Each seems to be a potential tool for levelingdifferences in the American opportunitystructure. Institutions need to compensate forthe ways that family differences lead to dif-ferences in achievement-a point made byJames Coleman twenty years ago.53 Educa-tional opportunity promotes social mobilitynot only by distributing human capital inmany ways that are independent of social ori-gins, but also by loosening the ties betweenoccupational and income origins and destina-tions among college graduates.54Establishingnorms of fairness and enforcing them seemslike a particularlysixties-style idealist solution.But establishing and enforcing those normsduring the 1960s improved the life chances ofdisadvantaged people during the 1960s and1970s and could, in principle, do so again.

VOL. 16 / NO. 2 / FALL 2006 31


15/19


16/19

Intergenerational Social Mobility: The United States in Comparative Perspective15. Most of the recent estimates pinpoint the father-son earnings elasticity between 0.35 and 0.5; see Gary

Solon, "IntergenerationalMobilityin the LaborMarket,"n Handbookof LaborEconomics, vol. 3A, editedby Orley Ashenfelter and David Card (Amsterdam:North Holland, 1999), pp. 1761-800; GarySolon, "In-tergenerational Income Mobility in the United States,"American Economic Review 82, no. 3 (1992):393-408; David J. Zimmerman,"Regressiontoward Mediocrityin Economic Stature,"American EconomicReview 82, no. 3 (1992): 409-29. These estimates exclude non-full-time workers. Estimates of persistencethat include all sons with incomes are lower; see Joseph G. Altonji and Thomas A. Dunn, "Relationshipsbetween the Family Incomes and LaborMarket Outcomes of Relatives," n Researchin LaborEconomics,edited by Ronald G. Ehrenberg (Greenwich, Conn.:JAIPress, 1991): 269-310; Nathan D. Grawe,"Inter-generational Mobility for Whom? The Experience of High and Low Earning Sons in International Per-spective,"in Generational Income Mobility in North America and Europe, edited by Corak,pp. 58-89 (seenote 13); H. Elizabeth Peters, "Patternsof IntergenerationalMobility in Income and Earnings,"Review ofEconomicsand Statistics74, no. 3 (1992): 456-66.

16. LauraChadwick and GarySolon, "IntergenerationalIncome Mobility among Daughters,"American Eco-nomic Review 92, no. 1 (2002): 335-44; BhaskharMazumder,"The Apple Falls Even Closer to the Treethan We Thought: New and Revised Estimates of the IntergenerationalInheritance of Earnings," n Un-equal Chances, edited by Bowles, Gintis, and Groves,pp. 80-99 (see note 2); Peters, "Patternsof Intergen-erationalMobility in Income and Earnings"(see note 15).

17. Chadwickand Solon, "Intergenerational ncome Mobility among Daughters"(see note 16).18. GaryBecker and Nigel Tomes, "HumanCapitaland the Rise and Fall of Families,"Journal of Labor Eco-

nomics 4, no 3, pt. 2: "TheFamily and the Distribution of Economic Rewards"(1986): S1-39.19. Gary Solon, "Biasesin the Estimation of IntergenerationalEarnings Correlations,"Review of Economics

and Statistics 71, no. 1 (1989): 172-74; Solon, "IntergenerationalIncome Mobility in the United States,"(see note 15); Zimmerman,"Regressiontoward Mediocrityin Economic Stature"(see note 15).

20. Steven Haider and Gary Solon, "Life Cycle Variationin the Association between Current and LifetimeEarnings,"Working Paper 11943 (Cambridge,Mass.: National Bureau of Economic Research, 2006).

21. Mazumder,"TheApple Falls Even Closer to the Tree" (see note 16).22. For example, lower overall mobility on the low end of the income distributionmay be due to factors such

as the lower abilityof poor parents to invest in children'seducation, the debilitatingeffects of livingin poorneighborhoods,or the employment effects of high rates of incarceration.

23. Grawe, "IntergenerationalMobility for Whom?" (see note 15); MarkusJanttiand others, "AmericanEx-ceptionalism in a New Light:A Comparison of Intergenerational Earnings Mobility in the Nordic Coun-tries, the United Kingdomand the United States,"Discussion Paper 1938 (Bonn, Germany:IZA [Institutefor the Study of Labor],2006); Mazumder,"TheApple Falls Even Closer to the Tree"(see note 16).

24. Kenneth A. Couch and Dean R. Lillard,"Non-LinearPatternsof IntergenerationalMobility in Germanyand the United States," n Generational Income Mobility in North America and Europe, edited by Corak,pp. 190-206 (see note 13).

25. Eric R. Eide and Mark H. Showalter,"FactorsAffecting the Transmissionof Earningsacross Generations:A Quantile Regression Approach,"Journal of Human Resources34, no. 2 (1999): 253-67; Grawe, "Inter-

VOL. 16 / NO. 2 / FALL 2006 33


17/19


generational Mobility for Whom?"(see note 15). There is some conflicting evidence on this issue when adifferent data source is used, but this may be due to a problem with some of the data collection.

26. Casey B. Mulligan,"Galtonversus the Human Capital Approachto Inheritance,"Journalof Political Econ-omy 107, no. 6, pt. 2: "Symposiumon the Economic Analysis of Social Behavior in Honor of Gary S.Becker"(1999): S184-224.

27. Ibid.28. Thomas M. Shapiro, The Hidden Cost of Being African American (Oxford University Press, 2004), pp.

60-85.29. Lester C. Thurow, Generating Inequality: The Distributional Mechanisms of the Economy (Springfield,

Va.:National Technical InformationService, 1975).30. Dalton Conley, Being Black, Living in the Red:Race, Wealth,and Social Policy in America (Universityof

California Press, 1999); Lisa A. Keister,Getting Rich: America'sNew Rich and How They Got That Way(Cambridge University Press, 2005); Melvin Oliver andThomas M. Shapiro,Black Wealth/WhiteWealth:ANew Perspectiveon Racial Inequality (New York:Routledge, 1995).

31. Featherman and Hauser,Opportunityand Change (see note 2).32. DiPrete and Grusky,"StructureandTrendin the Process of Stratification"see note 11); Hout, "More Uni-

versalism, Less StructuralMobility" see note 2).33. There is limited informationon trends before this time, and firm conclusions on occupationalmobilityafter

the mid-1980s are hardto come by because of data limitations. TimothyJ. Biblarz,VemnL. Bengston, andAlexanderBucur,"SocialMobilityacross Three Generations," ournalof Marriageand the Family58, no. 1(1996): 188-200; DiPrete and Grusky,"Structureand Trend in the Process of Stratification"see note 11);Michal Hout, "Status,AutonomyandTraining n OccupationalMobility,"AmericanJournalof Sociology89,no. 6 (1984): 1379-409; Hout, "MoreUniversalism,Less StructuralMobility" see note 2).

34. Claude S. Fischer and others, Inequality by Design: Crackingthe Bell Curve Myth (PrincetonUniversityPress, 1996), pp. 152-55; Hout, "Status,Autonomy and Training n Occupational Mobility"(see note 33).

35. DiPrete and Grusky,"Structureand Trend in the Process of Stratification"see note 11).36. Emily Beller and Michael Hout, "Income Inequality and Intergenerational Mobility: Change across Co-

horts,"unpublished Working Paper (University of California, Berkeley,2005). The extension of the meas-urement of occupational class background to include mothers' as well as fathers' occupation will alsochange the estimated trends in mobility rates over time, because the impact of including mothers'occupa-tion varies depending on the period (since, for example, mothers' labor force participationand parents'maritalsortingby class also varies by period).

37. Michael E. Sobel, Michael Hout, and Otis D. Duncan, "Exchange,Structure and Symmetryin Occupa-tional Mobility,"AmericanJournalof Sociology 91, no. 2 (1985): 359-72.

38. David I. Levine and BhashkarMazumder,"Choosingthe Right Parents:Changes in the IntergenerationalTransmission of Inequality between 1980 and the Early 1990s,"WorkingPaper WP-02-08 (Federal Re-serve Bank of Chicago, 2002); SusanE. Mayerand Leonard M. Lopoo, "WhatDo Trendsin the Intergen-



18/19

Intergenerational Social Mobility: The United States in Comparative Perspectiveerational Economic Mobility of Sons and Daughters in the United States Mean?" n GenerationalIncomeMobility in North Americaand Europe, edited by Corak,pp. 90-121 (see note 13).

39. Jo Blandon,Paul Gregg, and Stephen Machin,"IntergenerationalMobility in Europe and North America,"A Report Supportedby the Sutton Trust(London:LSE Centre for Economic Performance, 2005); Chul-InLee and Gary Solon, "Trends n IntergenerationalIncome Mobility,"WorkingPaper 12007 (Cambridge,Mass.:National Bureau of Economic Research,2006).

40. Hardingand others, "TheChangingEffects of Family Background" see note 2).41. Daniel Aaronson and BhashkarMazumder,"IntergenerationalEconomic Mobility in the United States:

1940-2000," Working Paper 05-12 (Federal Reserve Bank of Chicago, 2005).42. Emily Beller and Michael Hout, "WelfareStates and Social Mobility,"Research in Social Stratificationand

Mobility (forthcoming).43. Richard Breen and Jan 0. Jonsson, "Inequalityof Opportunityin Comparative Perspective: Recent Re-

search on Educational Attainment and Social Mobility,"Annual Review of Sociology 31 (2005): 223-43;RichardBreen, ed., Social Mobility in Europe (OxfordUniversity Press, 2004), pp. 37-76.

44. Beller and Hout, "Welfare States and Social Mobility"(see note 42); Richard Breen and Ruud Luijkx,"Conclusions," n Social Mobility in Europe, edited by Breen, pp. 383-410; Michael Hout, "MaximallyMaintainedInequalityRevisited: Irish EducationalMobility in Comparative Perspective," n Changing Ire-land, 1989-2003, edited by Maire Nic Ghiolla Phadraig and Betty Hilliard (University College DublinPress, forthcoming).

45. Anders Bjorklundand MarkusJantti,"IntergenerationalIncome Mobility in Sweden Compared with theUnited States,"American Economic Review 87, no. 5 (1997): 1009-18; Blandon, Gregg, and Machine, "In-tergenerationalMobility in Europe and North America"(seenote 39); Miles Corak,"GenerationalIncomeMobility in North America and Europe: An Introduction,"in Generational Income Mobility in NorthAmerica and Europe, edited by Corak,pp. 1-37 (see note 13); Janttiand others, "AmericanExceptionalismin a New Light"(see note 23); Solon, "CrossCountryDifferences in Intergenerational Earnings Mobility"(see note 1).

46. Janttiand others, "AmericanExceptionalismin a New Light"(see note 23).47. Casey B. Mulligan and Song Han, "HumanCapital, Heterogeneity, and Estimated Degrees of Intergener-

ational Mobility," Working Paper W7678 (Cambridge, Mass.: National Bureau of Economic Research,2000); Solon, "CrossCountryDifferences in Intergenerational Earnings Mobility" see note 1).

48. Raquel Fernandez and Richard Rogerson, "Sorting and Long-Run Inequality,"Working Paper W7508(Cambridge,Mass.: National Bureau of Economic Research,2000).

49. Sara McLanahanand GarySandefur,Growing Upwith a SingleParent: What Hurts, What Helps (HarvardUniversityPress, 1994).

50. Timothy J. Biblarz and AdrianE. Raftery,"FamilyStructure,EducationalAttainment,and SocioeconomicSuccess: Rethinking the Pathology of Matriarchy,"American Journal of Sociology 105, no. 3 (1999):321-65; Beverly Duncan and Otis Dudley Duncan, "Family Stability and Occupational Success," SocialProblems16, no. 3 (1969): 273-85; Sara McLanahanand Larry Bumpass, "IntergenerationalConsequences

VOL. 16 / NO. 2 / FALL 2006 35


19/19

Emily Beller and Michael Houtof Family Disruption," American Journal of Sociology 94, no. 1 (1988): 130-52; Kelly Musick and RobertD. Mare, "Family Structure, Intergenerational Mobility, and the Reproduction of Poverty: Evidence forIncreasingPolarization?"Demography 41, no. 4 (2004): 629-48.

51. Susan E. Mayer, What Money Can't Buy: Family Income and Children's Life Chances (Harvard UniversityPress, 1997).

52. DiPrete and Grusky, "Structure and Trend in the Process of Stratification" (see note 11); Hout, "More Uni-versalism, Less StructuralMobility"(see note 2).

53. James S. Coleman, "SocialCapital in the Creationof Human Capital,"AmericanJournalof Sociology 94,supplement (1988): S95-120.

54. Blau and Duncan, The American Occupational Structure, chap. 5 (see note 11); Hout, "More Universal-ism, Less StructuralMobility"(see note 2).


Scoial Mobility

Documents

Transcript of Scoial Mobility