Příspěvky nebo odpovědi uživatele Rupx

Příspěvky nebo odpovědi uživatele Rupx. Výpis příspěvků 201 až 220 (263)

(0) 06.12.14 15:58 Rupx (169801)

Debt Jobs or Housing: What’s Keeping Millennials at Home?

013 period may be leading to extended coresidence with parents for the most recent youthcohorts, and that the challenging youth job markets of the recent recession further obstructedyoung workers’ path to independence.

V. Discussion and Conclusions

This paper investigates young people’s parental coresidence rates in the CCP, and therelationships among coresidence decisions and local house prices, local employment conditions,and the student debt reliance of local college students. Evidence from the CCP shows thatcoresidence with parents has been persistently increasing for 25- and 30-year-olds since 1999,while the number of 25- and 30-year-olds living alone or with more than one non-parent hasdeclined (defining parents as people 15-45 years older than the youth). This trend is corroborated by similar analysis in the CPS. Simultaneously, homeownership has decreased for both agegroups. Both the fraction of individuals who have student debt and those individuals’ average balances have steadily increased over the same period.Panel estimates relying on geographic variation in economic conditions at the zip code-,county-, and state-level reveal mixed effects of local economic growth on young Americans’ propensity to live independently. While a one percentage point drop in state youthunemployment is estimated to increase the two-year rate of moving away from parents by 0.2 percentage points, a one standard deviation increase in house price gains over the two year period increases the probability that an independent youth moves home by 0.37 percentage points. On net, then, it appears to be not the overall strength of the local economy, but therelative circumstances of youth labor markets and goods markets where middle-aged parentstend to live and where independent youth tend to live, that shapes the trend in coresidence with parents.Finally, we find that a high state-level student loan balance per college graduate among ayoung person’s cohort both significantly increases the rate at which independent young peopletransition to living with their parents and significantly slows the rate at which dependent youthtransition away from their parents. Estimates indicate that a $10,000 increase in average studentdebt among a youth’s state cohort leads to a 0.81 percentage point increase in the likelihood ofmoving home to parents, and to a 2.63 percentage point

decrease

in the likelihood of moving outof one’s parents’ household. Given that student debt has been increasing since 1999, this

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:58 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

30in the probability of leaving home. The intuition behind this relationship is straightforward, asindependent living is costly, and typically demands a stable income source. Moreover, the problem of mechanical endogeneity arising from location choices is particularly weak in thiscase, as youth unemployment is measured at the state level, so that in order to influence theunemployment calculation the youth would have to cross state borders.Student loans are, once again, estimated to encourage coresidence with parents. The Table 4,column (4), student loan coefficient is large, negative, and highly significant. It indicates that a$10,000 increase in average student debt per graduate in the youth’s state-cohort decreases the probability of moving out of her parent’s home over the two years by 2.63 percentage points.The student loan estimate is not sensitive to the inclusion of the state-cohort graduation rate,which, perhaps surprisingly, is estimated to be strongly negatively associated with moving awayfrom home.

As before, the aggregated state-cohort student debt reliance is purged of features ofthe individual student’s situation, or her parents’ level of supportiveness. As before, given thestate fixed effects included in the estimation, the effect of state-cohort student debt is identifiedusing variation between cohorts in a given state in student debt reliance, and to the specificationaccounts for time-fixed regional variation in degree of support for youth and education.Again, total unemployment and house prices can be taken as indicators of broader economicconditions, this time in the parent’s location. Here we find very modest estimates of the effect oflocal economic conditions on youths’ propensity to leave home, with modest and insignificantcoefficients on changes in county total unemployment, county median income, and zip codehouse prices over the period. This may be the net result of ambivalent effects of strengtheninglocal economic conditions on youths’ capacity for independence. While strengthening economicconditions may improve the ability of youth to secure employment and fund independenthouseholds, and of parents to bankroll moves away from home, strengthening local conditionsmay also give rise to increasing local prices, particularly housing prices, which encouragecontinued coresidence.In sum, estimates from the model of the flow away from parents paint a picture ofstagnation in response to weakening labor market opportunities and growing student debt burdens. They provide evidence that the escalating student debt we’ve observed over the 2003-

Recall that the combination of stock and flow model estimates reflect that state-cohorts with higher graduationrates move away from parents considerably more slowly, and also experience less churning in their locations

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:56 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

increases prices, particularly house prices, and these increased prices drive the youth home to her parents.The reverse causality generated mechanically by the shifting populations in this case worksagainst our finding a positive effect of house prices on moving home, implying that this pointestimate of a 0.011 percentage point increase in the rate of moving home for every one percentincrease in local house prices over the January 2000 base prices is a lower bound estimate of thetrue effect. Further, the extent of this reverse causality problem may be large given that house prices are measured at the zip code level, and many young people may cross zip codes to returnto parents. Note that the 0.011 percentage point increase is reasonably substantial when oneconsiders the magnitude of the swings in house prices over the period. As noted, the standarddeviation of the zip code-level house price index over the period in our sample is 44.7.

Theaverage homeowner in the CCP experienced roughly a 50 percentage point increase in house price over the boom, and roughly a 25 percentage point decrease over the housing market bust.Additional variation in these experiences at the regional level enhances the effect of house priceson coresidence with parents estimated for this sample.On the other hand, the reverse causality problem actually works toward finding a spuriousnegative effect of total unemployment on moving in with parents, as unemployed youth leavingthe county at higher rates exert negative pressure on the unemployment coefficient. Further, thetotal unemployment coefficient becomes smaller and insignificant with the inclusion of income,graduation rate, and student debt. Together these mechanical endogeneity concerns lead us toinfer a substantial positive effect of housing costs in the independent location on the odds that achild moves home to parents, and a modest, possibly insignificant negative effect of both totalunemployment and youth unemployment in the independent location on moves home.The estimates in column (4) demonstrate a significantly higher rate of moving home instates with higher debt cost of a college degree in the youth’s graduating cohort. An increase of$10,000 in the debt cost of a degree is associated with a 0.8 percentage point increase in the probability of an independent young resident of the state moving home over the course of twoyears.

This effect is robust to controlling for the share of college graduates among the youth’s

Our effect is about twice the size of that estimated by Dettling and Hsu (2014), who study the effect of individual-level student debt on parental coresidence. Two factors may contribute to the difference in estimates. First, Dettlingand Hsu’s population is individuals age 18 to 30, which will lead their independent youth sample to be older thanours (since the likelihood of living with ones’ parents substantially decreases in age) and thus perhaps less sensitive

29cohort in the state, which, unsurprisingly, is associated with a lower propensity to move home.

These estimates provide further, and perhaps more credible, evidence of a positive effect ofgrowing student deb ...

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:55 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

test the robustness of our student debt estimate by instrumenting student debt per graduate withthe four-year average of state education spending when the individual was age 18-21. Thisinstrument is extremely powerful, with a first-stage F-statistic of over 100,000, and plausiblyaffects parental coresidence through the intensive margin alone, particularly when considered atthe aggregate level. Two-stage least squares estimates of the model are presented in column (8)of Table 2; the estimated coefficient on student debt per graduate is nearly double its un-instrumented value and is highly statistically significant, suggesting that the extensive marginwas indeed biasing the effect of student debt on parental coresidence downward.

b. Flow home to parents from independent living

Table 3 reports the coefficient estimates for the moving home model in expression (2). Therelevant scales of measurement are as follows: The dependent variable is set to 100 for a youthwho moves and 0 for a youth who does not. Unemployment measures and the graduation rateamong current 24 year olds in the state, similarly, range from 0 to 100. Recall that the CoreLogichouse price index takes a value of 100 in January 2000 for each zip code, and hence measuresgrowth in house prices relative to January 2000.Our preferred specification appears in column (4) of Table 3. Here we find that an increasein local house prices has a positive and highly significant effect on the probability that the youngadult moves home. In fact, we estimate that a one standard deviation increase in the house pricechange over the two years leads to a 0.37 percentage point increase in the probability of a youthmoving home over two years. Further, looking across the Table 3 specifications, we see that thehouse price coefficient is particularly robust to the inclusion (or exclusion) of other regressors,such as local income, student debt reliance, and graduation rate.Total unemployment and the local house price index, taken together, provide a measureof the strength of local demand. As demand increases in the youth’s independent location, thatlocation’s total unemployment decreases and its house prices increase. We observe a negative, insome cases significant, coefficient on total unemployment and a positive and consistentlysignificant coefficient on house price index in the moving home regression. Together, theseresults suggest that strengthening local demand conditions in the youth’s independent locationincrease the likelihood that the youth moves back home. One interpretation of these results isthat, conditional on the youth labor market, stronger demand in the youth’s independent location

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:52 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

coresidence among 25-year-olds.

Beginning with the red line, which depicts the predictedcoresidence rates from the baseline regression with time-varying characteristics, we fixcharacteristics at their 2000 (2003) levels, adding fixed characteristics in the order: totalunemployment, youth unemployment, house price index, household income, student debt pergraduate, and, finally, graduation rate. The top panel shows that changes in the overallunemployment rate, household income, and graduation rate have only very modest associationswith the rate of parental coresidence. The bottom panel of Figure 8 shows the additive proportionof the variation of parental coresidence that the stock model attributes to each covariate; itsuggests that house prices accounted for roughly 20% of the increase during the mid-2000s,while youth unemployment accounted for as much as 25% of the increase during the GreatRecession. However, student debt per graduate claims the largest share of the increase,accounting for as much as 50% of the increase in the likelihood of a young person’s living with parents.One challenge to the interpretation of the relationship between student debt per graduateand the rate of parental coresidence presented above is the possible existence of a secularunobserved trend in parental coresidence, perhaps related to a cultural change regarding young people’s living with their parents. If there exists such a secular cultural trend, then the correlation between parental coresidence and student debt per graduate (which has increased steadily,though at a slightly decreasing rate, every year on average across states) may be incidental. Wetest this conservative hypothesis by including a national linear time trend in our stockspecification, shown in column (6) of Table 2. Including a time trend attenuates all of theestimated coefficients, including the coefficient on student debt per graduate, but the student debtestimate remains positive and statistically significant at the 1% level, and both the youthunemployment rate and the house price index also maintain their positive and statisticallysignificant coefficients at the 10% level. Figure 9 shows a decomposition of these conservativeestimates, suggesting that, conditional on a secular cultural trend unrelated to student debt,student debt only explains about 10% of the increase in parental coresidence since 2004, withanother 10% being explained by house prices during the mid-2000s. An alternative specificationincluding year dummies instead of a linear time trend, a more flexible control for unobserved

Note that student debt is first measured in our data in 2003, and hence we delay the student debt and graduation baselines until data permit the addition of the student debt control. The estimates include an indicator for missingeducation measures, which is set to one during the earliest years of the sample.

26national culture shocks, yields very similar results. However, given that student debt is measuredat the state level (implying that, at the individual level, it is measured with extreme imprecision),this model is extremel ...

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:51 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

increases the unemployment measures in the origin. Assuming movers are more likely to beunemployed than the population of young people overall, this mechanical source of endogeneitycould bias coefficients on the unemployment measures downward. Of course, to the extent thatthe unemployment measure responds with a lag rather than contemporaneously, the estimateswill not be biased. This seems to be a likely possibility. Further, this source of reverse causalityof youth residence choices is not a concern if moves to and from parents’ households all occurwithin the relevant location. For example, to the extent that all moves home or away occur withinthe same county, no unemployment coefficient will be affected.Assuming youth in our model influence the housing market as well, a similar type of reversecausality could affect the coefficients on house prices. Assuming that a child living in a parent’s basement does not lead the parent to demand more housing, the departure of a youth who lives athome has no effect on house prices. Hence we are less concerned about the effect of reversecausality on the house price coefficient in the moving out regression. A youth who has beenliving independently and returns home, however, leaves the origin housing market and thereforedecreases total housing demand in the origin. As each youth in the moving home regressionexerts this influence, this source of mechanical reverse causality could bias the estimatedcoefficient on house prices in the moving-home regression downward. As in the case ofunemployment, to the extent that the resulting effects on house prices appear with a lag, or thatmoves between parent and independent youth locations fail to cross locations, this is not aconcern. However, house prices in the estimation are measured at the zip code level, and so it isreasonably likely that youth moving back home will cross zip code lines and exert spuriousdownward pressure on house price effects in the moving-home regression.Standard endogeneity concerns deriving from observable and unobservable individual andlocal characteristics that are fixed over the two year window are accounted for by the transitionapproach we take to estimation. Obvious examples include child ability, parent generosity, and persistent regional characteristics. Remaining major concerns regarding the endogeneity ofchanges in local characteristics to youths’ transitions home and away seem most likely to arisefrom third factors determining both changes in local characteristics from

1 and youths’interest in living with parents. An immediate example is changing local economic conditions.Their effect is likely to be picked up by some combination of total employment and house pricemeasures. Given this, we interpret total employment and house price coefficients as though they

24contain both direct effects of employment and house prices, and indirect effects of localeconomic conditions. So far we have not found that this changes our inferences based on theestimates substantially. Concerns regarding third factors influencing local levels of student debtamong recent graduates are more relevant, and we discuss them along with the model results inthe following section, providing first-pass instrumental analysis to allay these concerns.

...

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:50 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

Endogeneity concerns in the flow models

The most obvious endogeneity concerns in the context of our broader youth co-residencequestion involve individual student debt. Were we to estimate expressions (2) and (3) usingindividual student debt, the debt the young person has accumulated, and has not yet repaid,would presumably be endogenous to parental co-residence for a host of reasons. First, whether ayoung person is currently accumulating or paying down student debt presumably reflectsmeasured and unmeasured aspects of the youth’s broader financial and life circumstances, andthese, in turn, influence co-residence. Second, the literature has documented substantialheterogeneity in parents’ financial generosity.

At the individual level, parental generosity mayact as a third factor, determining both parental contributions for college, hence student debt, and parents’ willingness to house their adult children. Third, individual heterogeneity in features suchas debt-aversion, risk tolerance, and diligence may drive both the level or change in individualstudent debt and the decision to move home or away. Finally, individual ability and educationalattainment are likely both to be correlated with the student debt level and to influence co-residence with parents.Given the many sources of endogeneity of individual student loan level and growth to youthresidence outcomes, we estimate the dependence of co-residence with parents on student debt by proxying for student debt with the mean student debt cost of a degree in the youth’s state-cohort,as described above. We estimate both flow models including state fixed effects, and hence thecoefficient on state-cohort student debt is identified by changes across cohorts within a state inthe mean debt price of a degree. Such differences are relatively free of confounding familycharacteristics like generosity and debt aversion, and instead are influenced by changes (withinstate) from cohort to cohort in factors including tuition at state colleges and the generosity offederal, state, and institution-level grant and loan aid.

One possible concern regarding reverse causality arises from the individual youth’sinfluence on labor and housing markets in the location she leaves, should she decide to move. Ayouth who moves away from one location to another, if unemployed, can be expected todecrease the total and youth unemployment rates in the original location. If employed, she

Brown, Scholz, and Seshadri (2012), McGarry and Schoeni (1995).

Note that we control for state economic conditions that might otherwise appear in the mean state-cohort studentdebt price of a degree through youth unemployment, total unemployment, and house prices

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:49 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

a. Stock of young people living with parents

Figure 7 shows that the trend in 25-year-old parental coresidence appears more similar tothe trend in student debt than to unemployment rates or the house price index at the nationallevel, but does not capture geographic variation in these relationships or allow us to weigh therelative contributions of each feature of the environment to a young person’s residence choice.As a first, descriptive, step, we estimate a model of the likelihood that, at a given time, a child isliving with his or her parents as a function of local socioeconomic conditions. In anticipation ofthe flow model to come, we consider individuals at two ages, 23 and 25. Define

as anindicator for whether individual

living in location

at time

coresides with her parents. Wemodel the likelihood that an individual lives with her parents as a function of the conditions inher locality one year earlier, including fixed effects by state to control for unobserved differencesin culture and policy.

We thus estimate the following linear probability model:

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:48 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

18independence, that we observe for recent cohorts of young adults. Motivated by this potential inthe CCP, and by the questions raised by the relationships among the national trends, in thefollowing section we present an empirical model of parental coresidence. First, we describe alagged stock model explaining the revealed coresidence decisions of 23- and 25-year-olds as afunction of local unemployment, youth unemployment, house prices, and student debt per recentgraduate, a linear decomposition of which provides a simplified visualization of the conditionsassociated with parental coresidence. Then, to account for differences in the economiccharacteristics of regions where young people live alone, and where their parents live, weseparately model the flows of young CCP consumers into and out of parents’ households overtime as a function of changes in the economic and social conditions of young consumers’ initiallocations.

III. Empirical model

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:45 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

aged 18-30, based on our own calculations in the CPS, followed a similar modest decline in the boom and then increased even more sharply, from 6.8 percent in late 2006 to a maximum of 14.8 percent in early 2010. Its recent recovery involves a decline from 14.8 percent at its 2010maximum to 11.3 percent in the end of 2012. The monthly CoreLogic house price index, hererepresented by the purple line, increased from a normalized value of 100 in January 2000 to a peak of 200 in 2006, fell to a 2011 trough of 134, and since then has moved through an unsteadyrecovery to reach 165 by late 2013. In sum, while aggregate student debts have followed a steep,unbroken upward trajectory since 2003, both employment and house prices have experienced a pronounced boom, bust, and recovery.One question, then, is to what degree the residence choices of the young track the recent, pronounced business cycle. To the extent that they move with the boom and bust, residencechoices may appear to be driven by economic conditions, such as youth labor markets and thecost of housing. To the extent that their changes are gradual and persist throughout the boom, bust, and recovery, however, they may appear to be driven instead by young consumers’ recent,unprecedented accumulation of student debt.In Figure 7, as before, the upward trend in coresidence with parents appears steady, andsuggests little direct relationship to broad economic indicators such as unemployment measuresand the house price index. This would seem to suggest that the decision to stay home with parents, or to move back in, relates more closely to the recent changes in the debt burden ofhigher education than to swings in youth labor markets and the cost of housing.However, the analysis presented in Figure 7 is unsophisticated, and, as such, poses morequestions than it resolves. In terms of the aggregate trends, the steady increase in coresidencewith parents may reflect not a failure to respond to aggregate conditions, but offsetting effects of,for example, job and housing markets on residence decisions among the young. The failure of allyouth residence decisions to reflect the recent recovery in employment and house prices remainsa mystery.Finally, these aggregate trends, while informative, may mask evolving local relationshipsamong housing cost, labor markets, and youth residence choices. The fine geographic data andlong panel of the CCP allow us to exploit time variation in local economic conditions and studentdebt reliance to learn more about the contributions of jobs, housing costs, and student debt at thelocal level to the decisive aggregate trends toward parents, and away from economic

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:44 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

162006 and continuing to decline until it reached 12.1 percent in 2013. Thus the drop inhomeownership among 25 year olds led the downturn in the housing market by roughly a year,and peak homeownership at 30, traditionally the median age of first home purchase in the U.S.,was reached a full two years later.

Clearly the youngest homeowners had a differentrelationship to the housing boom and bust than traditional first time buyers.Figure 6 relates the timing of the homeownership decline to that of the increase in the rateof living with parents for each age group. Declines in homeownership coincide with increases inliving with parents for both age groups. At 25, homeownership drops steadily from 2005 through2013, and the rate of living with parents shows a steady increase throughout the 2003 to 2013window. At 30, living with parents increases throughout, but appears to accelerate after 2006.Shortly after, in 2007, homeownership reaches its peak and then declines steeply until 2013, atwhich point 30 year olds in the CCP are equally likely to be living with a parent or to own ahome.

c. Prevailing macroeconomic conditions and youth residence choices

So far we have seen approximately unbroken upward trends in coresidence with parentsamong 25 and 30 years over the years from 2003 to 2012, and a substantial change in aggregatehomeownership at 30 around the Great Recession. All of this raises questions regarding therelationships among youth residence choices and the prevailing economic conditions underwhich these choices are made.Figure 7 represents trends in broader economic conditions and youth residence choices ina common space. Aggregate U.S. student debt, as measured in the CCP, is represented by theyellow line, and follows a steep upward path, without wavering around the recession. Totalunemployment for the U.S. is drawn from the Bureau of Labor Statistics (BLS) monthly LaborForce Statistics for 1999-2014. Represented by the dark red line, it showed a modest declineduring the economic boom of the mid-2000s, followed by a steep increase from 5.0 percent in2008 to 10.0 percent in late 2009, and a subsequent recovery. The recent recovery inunemployment has been gradual but substantial: total unemployment among the adult population participating in the workforce, seasonally adjusted, fell from 10.0 percent in late 2009 to 5.8 percent in the October 2014 BLS Employment Situation release. Unemployment among youth

See Lautz (2011) for recent median ages at first purchase

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:43 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

15debt, as a homeowner.

The presence of home-secured debt on the credit report is a particularlyreliable proxy for homeownership at young ages, and its absence a reliable proxy for non-homeownership, as very few 30 year old homeowners in the U.S. own their homes outright. Inthe figures, we trace the proportions of 25 and 30 year olds who have owned homes over the past4 years, inferred based on the preceding four years of linked data in the panel. The object ofinterest is whether the individual currently owns or has ever owned a home, and four years ofhistory is a reasonably good proxy for ever owning at these very young ages. Similar resultsobtain where we track the rate of current homeownership and the rate of ever owning over thefull course of the panel. The potential difficulty with the latter measure is that the look-backwindow available in the CCP lengthens as the panel progresses, creating time dependence in thequality of the measure of homeownership. However, homeownership measures based on all threeapproaches are very similar for these young age groups.We find that homeownership among 30 year olds grew modestly from 43.2 in 2003 to44.4 percent in 2006-7. After its peak at 44.4 percent, it dropped off dramatically following thehousing market crisis, to 42.6 in 2008 and 31.9 percent by 2013. Hence we observe more than a10 percentage point drop in the share of 30 year olds who have owned their own homes over thecourse of five short years. This hump-shape in homeownership corresponds to the cohabitation pattern discussed earlier. However, the decline in homeownership for 30 year olds in recent yearshas been much larger than that for cohabitation.Homeownership rates among 25 year olds are, as expected, substantially lower. Perhapsmore surprising is the timing of their growth and decline relative to the housing market boomand bust. Speculation regarding the source of the boom and bust, and its relationship to sub- prime lending and easy credit for buyers with limited funds for down payments, suggests ahousing market that grew to reach younger and younger consumers. The CCP time trends onearly homeownership appear to tell a different story. Homeownership among 25 year olds grewfrom 22.7 percent in 2003 to a peak of 22.9 percent in 2005. This increase appears reasonablymodest, in the face of softening sub-prime lending standards and historically low down payments. From its peak in 2005, homeownership among 25 year olds fell to 22.5 percent in

Evidence from our sample indicates a sharp change in the rate at which young couples include both members in amortgage following the Great Recession. This is discussed in a companion paper on student debt andhomeownership. Whether we credit couples’ mortgages to each other or not. we find that homeownership at age 30declines markedly after 2007

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:42 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

14only recently beginning to recover. MeasureOne reports a growth of total private student loan balances among the seven leading lenders currently in the market from $44 billion in 2008Q3 to$63 billion in 2013Q3 (MeasureOne, 2013).

In sum, student borrowing has changedsubstantially from 2003 to 2013.All of the above, however, describes the student loan market as a whole, including parentand student borrowers of all ages. More relevant to the residence choices of young Americansmay be the trends in student debt among recent graduates (and dropouts). Figure 4 depicts the proportion of CCP 25 year olds participating in the student debt market, along with the meanstudent loan balance of 25 year olds in each year, among those 25 year olds who have studentdebt. Once again, the only financial variables appearing in this paper pertain to student loans.They are reported in 2012 U.S. dollars. We observe an increase from 25 percent in 2003 to 45 percent in 2013 of 25 year olds with positive student debt, for an 80 percent growth in 25 yearolds’ rate of participation in student debt markets over the decade. Mean student loan balances at25 among those with positive student debt balances between the ages of 22 and 25 nearlydoubled over the period, from $10,649 in 2003 to $20,932 in 2013. In sum, the remarkableaggregate growth in student debt over the decade from 2003 to 2013 is more than reflected in thestudent debt growth we observe for 25 year olds in the CCP. More students are enrolling incollege, and students in college are borrowing more to fund their educations. As speculated bythe NAR, the CFPB, and various arms of government, we might expect the burden of increasingeducational debt to delay standard life-cycle economic milestones, such as living independentlyand the purchase of first homes. Next we turn to trends in early homeownership in the CCP. Figure 5 depicts the trend inhomeownership among all 25 and 30 year olds represented in the CCP.

We inferhomeownership based on the presence of home-secured debt, whether mortgage or home equity- based loans, on the sample member’s credit report. Further, as some couples may include onlyone member’s name on the mortgage, we record an individual with no home-secured debt on heror his own credit report, but who is living with one adult of similar age who holds home-secured

Some of this growth in private student loans among the seven leading lenders may reflect buying debt from otherlenders.

Note that Figure 5 reflects all youth in the CCP as well as the small portion of youth represented in the Census butnot in the CCP. We assume that 25 and 30 year olds not covered by the CCP (and who thus do not have Equifaxcredit reports) are not homeowners, as (we infer) they almost certainly cannot have mortgages. The qualitativefindings all persist in the CCP-only sampl

13holder, and with two or more adults of similar age.

The latter category we interpret asroommates. We find a consistent pattern across the two age groups, though, of course, the leveland growth of coresidence with parents is much greater for 25 than for 30 year olds. At each age,the growth we observe in coresidence with parents appears to come at the cost of fewer young people living alone, and fewer young people living with young roommates. The rate of livingalone, for example, falls from just above 26 percent for each age group in 1999 to 15 percent for25 year olds, and 17 percent for 30 year o ...

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(2) 06.12.14 15:41 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

12Given general agreement that young Americans are staying home with parents at anincreasing rate, what alternative living arrangements are they forsaking? Popular speculationsuggests declining rates of first marriage among young people in the wake of the recession. Afterthe release of the 2009 American Community Survey, Mather and Lavery (2010) noted arecession-era decline in the share of young people who had ever been married. Shortly after,Wolfers (2010) countered that this data artifact represented not a meaningful decline in stablerelationships, but an ongoing increase in the age at first marriage in the U.S., coupled with anincrease in cohabitation during the recession, which may have been motivated by a desire to cutliving expenses. The relevant question for the current study, then, may be whether youngAmericans are choosing extended adolescence at home with parents in place of independentadulthood and marriage.Our CCP measures do not allow us to measure the rates at which CCP sample membersare marrying before and after the recession. They do not even allow us to measure cohabitingrelationships, whether or not they involve marriage. What we can do, however, is look at trendsin the rate at which young Americans coreside with one other adult of a similar age. The benefitof this approach is that it includes marriage with both opposite sex and same sex cohabitation,yielding a broader picture of trends in coresiding relationships over the period. The obviousdrawback, however, is that it includes roommate pairs whose relationships are platonic. Ouranalysis of CPS household characteristics suggests that this later group is reasonably rare from atleast the age of 30 onward.

Interpretation of trends in living with a single adult roommate ofcomparable age should, however, bear this inclusion in mind.We categorize individuals who are not coresiding with parents into three types. Anindividual is defined as living alone if she is the only (Equifax-covered) resident at her streetaddress. We then divide the remaining individuals into those who live with only one other personand those who live with more than one other person, excluding households with more than 10 people and individuals whose report lists a post office box address.Figure 3 panels (a) and (b) show CCP trends from 1999 to 2013 in the rates at which 25and 30 year olds, respectively, appear alone, with parents, with one adult of similar age to the file

Calculations available from the author

Dosavadní doporučení: přečíst (2)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:40 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

11show an increase in coresidence with parents from 22.8 percent to 40.9 percent of 25 year olds,and from 16.5 to 28.4 percent of 30 year olds. Their slopes are quite similar to the CCP-onlytrends, while their levels are roughly four to six percentage points lower for the 25 year olds andtwo percentage points lower for the 30 year olds (for whom coverage in the CCP is fairlycomplete).As a final check of our coresidence results using the CCP, we turn to the 1999-2012waves of the CPS, and create coresidence measures designed to be similar to our CCP measuresusing the CPS. We construct U.S.-representative samples of 25 and 30 year olds in the CPS,using the CPS individual weights. From there, we create an indicator of coresidence with parentsthat equals one for any youth living in the same household with one or more individuals who are15 to 45 years older.

The purple coresidence curves in Figure 2 panels (a) and (b) represent ourcoresidence calculations using these criteria in the CPS.Coresidence rates measured in this manner in the CPS are similar to those based on theCCP and assuming Census youth not represented in the CCP live away from home, though theslope of the CPS coresidence curve is somewhat less steep. The inference that coresidence with parents rises markedly from 1999 to 2012 is accurate to both the CCP and the CPS series. For 25year olds, the CPS coresidence curve lies quite close to the CCP curve that assumes youthwithout credit reports live away. Coresidence for 25 year olds in the CPS grows from 27 to 37 percent, for a 36 percent increase over the period. For 30 year olds, the CPS curve lies three to 10 percentage points below the CCP lower estimate, which in turn is quite close to the CCP-onlyestimate. The CPS curve depicts a steady growth in coresidence with parents from 1999 to 2010,followed by a very modest decline in coresidence with parents for 30 year olds in 2011 and 2012.In sum, we observe a steady growth in coresidence with parents among U.S. youth. Whilethe level of coresidence rates may be sensitive to measurement choices, the levels we obtain aresimilar enough across alternative methods and sources to suggest that our CCP measures areinformative, and the marked upward trend in coresidence with parents is robust to all sources andmethods discussed in this paper.

b. Trends in other living arrangements

As on the CCP, we assume that any individual in a household of 10 or more persons is living away from parents.

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:40 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

measure of coresidence relative to survey-based measures is the possibility that mailing addresscorrespondence, even with the requirement that apartment numbers match, may not perfectlymeasure residence in a shared housing unit.In addition, credit report coverage of younger U.S. individuals is extensive but notcomplete, as described above. To the extent that the five to 14 percent of the youth population ofthe U.S. that is not represented in the trends in Figure 1 lives with parents, these trends willreflect underestimates of the true underlying rate of coresidence with parents among youngAmericans. To the extent that the small unrepresented share of youth do not live with parents, theFigure 1 trends will be overestimates of the true rate of coresidence.In order to address this concern, Figure 2 depicts the coresidence trends for 25 and 30year olds when one assumes that all 25 and 30 year olds represented in the Census but not in theCCP (that is, individuals with no active credit history) live with parents, and then when oneassumes that they live away from parents. This creates an upper and a lower bound on estimatesof the coresidence rates of U.S. youth based on the CCP.The more plausible assumption may be that Census youth not represented in the CCP liveaway from parents. Reasons behind this include institutional populations, such as military and prison populations, who generally live away from home and, we infer, have limited credit reportcoverage. Such populations tend to be young, and hence their credit report coverage andresidence status are particularly relevant for this study of youth residence. According to the U.S.Bureau of Justice Statistics, 0.94 percent of U.S. resident adults were incarcerated at the end of2011. Presumably the incarcerated shares of 25 and 30 year olds are greater. Similarly, as of2010, 2.28 million U.S. adults were active duty or reserve members of the armed forces.

Thisrepresents 1.2 percent of adults 18-64 years of age. Again, shares of the population in themilitary are likely much larger at ages 25 and 30. Though prison and military populations mayhave actively updating credit files, they are presumably more likely to be among the small shareof 25 and 30 year olds without active credit files, and are of course substantially more likely thanother young consumers to live away from parents.The estimated trend in coresidence with parents in which we assume all youthrepresented in the Census but not in the CCP live away from parents is represented in Figure 2 by the series with long dashes, lying in each case below the CCP-only trend. These lower trends

See the National Defense Authorization Act for Fiscal Year 2013, H.R. 4310

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(3) 06.12.14 15:39 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

9law. Hence from this point we refer to this living arrangement simply as “coresidence with parents”.For 30 year old CCP sample members, we observe an increase in the rate of coresidencewith parents from 18.7 percent in 1999 to 31.5 percent in 2013. Note that this pattern is free oflife-cycle effects, as we measure coresidence with parents for the cross-section of CCP samplemembers who are 30 years old in each year. This substantial growth in living with parents isapproximately monotonic over the period, and proceeds at a steady pace.Among 25 year olds, the rate of growth is similar, though the levels, as expected, arehigher. Coresidence with parents for 25 year olds grows from 28.3 percent to 48.8 percent between 1999 and 2013. As with 30 year olds, the trend for 25 year olds is approximatelymonotonic and the growth in coresidence is steady, though coresidence declines slightly in 2013.Overall, the rate of coresidence with parents observed in the CCP grows by 68.4 percent for 30year olds, and by 72.4 percent for 25 year olds, from 1999 to 2013. A striking change appears tohave occurred since 1999 in the living arrangements of young consumers.

There are several reasons why the coresidence rate measured using the CCP might differfrom measures in other relevant sources. For example, a 2013 report from the Pew ResearchCenter, based on their own analysis of the March Current Population Survey (CPS), reported that32 percent of 18-31 year olds in 2007, 34 percent in 2009, and 36 percent in 2012 live with parents. However, the Pew analysis defines an individual as living with a parent only if she liveswith a parent or step-parent, not a parent-in-law, and is not her- (or him-) self a head ofhousehold. Clearly, this narrows the definition of living with parents from the one used in ouranalysis.Further, the CCP is unable to identify family relationships, though it does provide the agedistribution of individuals living at the same address. Hence, our measure of coresidence with parents is an overestimate, as it contains the small minority of U.S. adults, described above,living with significantly older non-relatives or spouses. This should also increase our measure ofcoresidence relative to calculations like the Pew results. Another factor that might increase our

This trend may be determined in part by social or demographic phenomena, rather than economic pressures.While the number of Americans aged 45-64 increased by 24 percent from 2002 to 2012 (U.S. Department of Healthand Human Services Administration on Aging,www.aoa.acl.gov/Aging_Statistics/Profile/2013/3.aspx), thelifetime number of children per woman remained near two and, if anything, was very slightly increasing from 1970to 2010 (Population Reference Bureau 2012). It is unclear, then, the extent to which changing demographics on theirown can be expected to generate changes in the rate of coresidence with parents. In the interest of accounting for possible social and demographic changes, we allow for a time trend as we model the stock of co-residence below

Dosavadní doporučení: přečíst (3)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:38 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

number).

These data lead us to define coresidence (with parents) to be the circumstance inwhich a young person (here either a 25- or 30-year-old) resides at the same street address as atleast one (Equifax-covered) individual who is between 15 and 45 years older than her, withoutregard to household head status or the relationship between the household members.

Data fromthe Center for Disease Control and Prevention’s (CDC) National Vital Statistics System showthat, for children born in 2012, 99.8% of mothers and 84.7% of documented fathers were withinthis age range (15-45).

Moreover, we define individuals who live in households of more than10 people (3.7% of 25-year-olds and 3.6% of 30-year-olds) as

not

coresiding, because mostsituations in which one would live in such a large household (prison, military, trailer park) arenot such that the individual is in their parents’ household.

Note that our definition mightoverestimate the aggregate rate of coresidence with parents due to a possible lag between ayoung person’s switching their home address and updating their credit report address (asreported by financial institutions), which might bias the aggregate coresidence rate upwards.Figure 1 depicts the proportion of U.S. 25 and 30 year olds living with “parents” in theCCP from 1999-2013.

As above, we define living with parents as sharing an address, includingan apartment number if one exists, with at least one household member who is 15-45 years older. Note that this includes a range of coresidence circumstances, including coresidence with a parentor parents in which the child is the economic dependent, coresidence with a parent or parents inwhich the parent is the economic dependent, coresidence with a spouse or partner’s parents,coresidence with a grandparent, and even rooming with an older spouse or non-relative.Evidence from the CPS, available upon request from the authors, suggests that the overwhelmingmajority of households with this age profile consist of children living with parents or parents-in-

See Avery et al. (2003) for a detailed discussion of the contents, sources, and quality of credit report data.

We exclude household members with empty credit files, as those individuals’ addresses may no longer beaccurately recorded by their creditors, or thereby by Equifax itself.

Analysis of the extent to which this criterion includes non-parent relatives (grandparents, uncles, and aunts) andnon-relatives using the CPS is available from the authors. We find that we may capture a fair number of coresidentgrandmothers and aunts. Romantic cohabitation or marriage with partners 15 or more years older, however, is quiterare in the CPS. Hence the overwhelming majority of coresident households captured by this criterion involve ayoung person living with an older relative, most often parents or step-parents.

We also assume that individuals whose address is listed as a post office box do not coreside (4% of 25-year-olds,and 5% of 30-year-olds).

The median individual born in a year turns 25 around July 1

25 years later. In order to capture the average ...

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(0) 06.12.14 15:37 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

100, and it is the most comprehensive monthly house price index available. We aggregate anannual index to avoid seasonal variation. The CoreLogic data cover a total of 6739 zip codes(representing 58% of the total U.S. population, and 63% of the 25- and 30-year-olds observed inour sample) in all 50 states and the District of Columbia.Annual county-level income data for 3,142 counties are drawn from the Internal RevenueService’s (IRS) Statistics of Income (SOI) program, which annually aggregates household-leveladjusted gross income as reported on US tax forms. Average aggregate household income acrossour population is $53,200, and is higher within the coresiding population.Using the CCP’s loan-level student loan balance data, we calculate the average studentdebt burden per graduate as the gross third-quarter student debt held by 24-year-olds in thatstate-year over the total number of college graduates from universities in that state-year, chained by the CPI to 2012 dollars. We calculate the total number of graduates using the IntegratedPostsecondary Education Data System (IPEDS), summing over the number of graduates of four-year

and two-year institutions who receive degrees within 150% of the normal completion time

in that state-year; the IPEDS data are available every year after 2002. We also calculate theaverage graduation rate as the ratio of the total number of graduates to the total number of 24-year-olds in the state, as estimated by the US Census. The average graduation rate across statesover our sample is 29.1%, and the average state-level per-graduate student debt burden is$20,100.Finally, as one predictor of states’ student debt reliance, we use state-level primary andsecondary school spending data collected at the school district level for the US Census’s PublicEducation Finances Report, measuring total current spending per enrolled student (as of fall ofthat school year). Spending per student averages $7,900.

II. Aggregate trends in the economic conditions facing young consumers and theirresidence choices

Coresidence with parents: measurement and trends

Each observation in the CCP includes the (anonymized) information in an individual’scredit report at the end of that quarter (e.g. zipcode, birth year, total balances of 10 types ofconsumer debt, etc.) as well as the information in the credit reports of all members of thatindividual’s household, where households are defined by street address (down to an apartment

Dosavadní doporučení: přečíst (0)

Vaše doporučení:

Téma: Akcie ERSTE BANK

(2) 06.12.14 15:37 Rupx (169801)

Zprávy o akcii Odpovědět

Debt Jobs or Housing: What’s Keeping Millennials at Home?

We construct a cohort-level dataset from the CCP by extracting a panel of all individualswho turn 25 or 30 years old in each year between 1999 and 2013. Because the time-series aspectof our study drastically increases the number of observations, we only pull a random 1% sampleof the covered U.S. population, instead of the full CCP 5%. There are 567,932 25-year-olds and598,455 30-year-olds in the dataset, on whom we have 11.9 million and 17.1 millionobservations, respectively.In order to calculate bounds on our coresidence estimates, for some applications (which will be clear below) we balance our panel by including null observations in all quarters in whichEquifax provides no credit report for an individual (starting at age 18), as well as including nullobservations for individuals whom we do not observe as having a credit report at age 25 or 30(imputing the number of such individuals at the state level from the U.S. Census). Our final, balanced dataset includes a total of 30.5 million observations.

b. Other data sources

Columns (1)-(4) of Table 1 summarize the additional data that we use in our aggregateanalysis of parental coresidence. They provide average stock values of each characteristic acrossall individuals and years in our sample, and compare the levels of individuals who coreside with parents with those of individuals who live independently.Annual county-level unemployment data are drawn from the Bureau of Labor Statistics’s(BLS) Local Area Unemployment Statistics (LAUS) program. The unemployment data arereported on a monthly basis, and they cover a total of 3,145 counties. We calculate the youthunemployment rate at the state level using employment data from 18- to 30-year-old individualsin the CPS, aggregated from months to quarters.

The average youth unemployment rate acrossstates over our sample is 9.7%, ranging from 1.8% in 2000 Connecticut to 22.1% in 2010 WestVirginia.House price appreciation values are calculated at the zip code level using data from theCoreLogic housing price index (HPI). The CoreLogic HPI uses repeat sales transactions to trackchanges in sale prices for homes over time, with the January 2000 baseline receiving a value of

households in 2007 include no member with a credit report. They also find a proportion of household heads underage 35 of 21.7 percent in the 2007 SCF, 20.64 in the 2007Q3 CCP, and 20.70 from Census 2007 projections,suggesting good representation of younger households in the CCP.

CPS youth unemployment data is only available from 1999-2012

Dosavadní doporučení: spíše nečíst (2)

Vaše doporučení:

« Předchozí | 9 | 10 | 11 | 12 | 13 | 14 | Další »
stránkovat diskuzi po 10 | 20 | 50

Názory a diskuze

Název	Kurz	Změna
Burza Prime 16:27
COLT CZ GROUP SE	680.00	+0.44%
ČEZ	874.00	-0.68%
ERSTE GROUP BANK A	1 129.00	+2.26%
GEN DIGITAL	573.00	+0.53%
GEVORKYAN	256.00	-0.78%
KOFOLA CS	304.00	0.00%
KOMERČNÍ BANKA	783.00	+0.06%
MONETA MONEY BANK	102.20	0.00%
PHILIP MORRIS ČR A	15 400.00	+0.79%
PHOTON ENERGY	43.20	+0.47%
PILULKA LÉKÁRNY	128.00	-0.39%
PRIMOCO UAV SE	930.00	-0.53%
VIG	769.00	+1.59%

Kurzy měn
Kurzovní lístek ČNB pro 01.07.2024
USD americký dolar	23.315
AUD australský dolar	15.549
GBP britská libra	29.541
BGN bulharský lev	12.808
EUR euro	25.050
HUF maďarský forint	6.352
NOK norská koruna	2.192
PLN polský zlotý	5.824
CHF švýcarský frank	25.860
TRY turecká lira	71.351