Measuring Heterogeneity in Job Finding Rates among the Non-Employed Using Labor Force Status Histories

Transcription

1 FEDERAL RESERVE BANK OF SAN FRANCISCO WORKING PAPER SERIES Measuring Heterogeneity in Job Finding Rates among the Non-Employed Using Labor Force Status Histories Marianna Kudlyak Federal Reserve Bank of San Francisco Fabian Lange McGill University IZA September 2017 Working Paper Suggested citation: Kudlyak, Marianna, Fabian Lange Measuring Heterogeneity in Job Finding Rates among the Non-Employed Using Labor Force Status Histories Federal Reserve Bank of San Francisco Working Paper The views in this paper are solely the responsibility of the authors and should not be interpreted as reflecting the views of the Federal Reserve Bank of San Francisco or the Board of Governors of the Federal Reserve System.

2 Measuring Heterogeneity in Job Finding Rates among the Non-Employed Using Labor Force Status Histories Marianna Kudlyak Federal Reserve Bank of San Francisco Fabian Lange McGill University and IZA This Draft: September 19 th, 2017 First Draft: May 20 th, 2014 Abstract: We introduce a novel approach to studying heterogeneity in job finding rates by classifying the non-employed, the unemployed and those out of the labor force (OLF) according to their labor force status (LFS) histories using four-month panels in the CPS. Time since last employment as measured using the LFS histories is the best predictor of future employment for both the unemployed and those OLF; in predicting future employment, it outperforms currentmonth responses to survey questions about duration and reason for unemployment, desire to work, or reasons for not searching. Those OLF who were recently employed are twice as likely to find a job as those who report that they want a job. For the unemployed, the duration since last employment and the self-reported durations of unemployment often disagree. In these cases, the duration since last employment is a better predictor of re-employment, that is, the unemployed who report long durations after recent employment have similar job finding rates as those who report short durations. We present evidence that the disagreement between the duration of joblessness and self-reported duration of unemployment is not caused by classification error. Instead, the respondents report durations of looking for work and they often disregard short-term jobs and include periods when they continued searching while employed. Using our proposed approach, we reexamine the unemployment duration distribution and current approach to misclassification error in the CPS. JEL: E24, E32, J30, J41, J63, J64 Keywords: Job Finding Rate. OLF. Unemployment. Duration Dependence. Heterogeneity. Misclassification Error. For their useful comments, we thank our discussants Bart Hobijn and David Wiczer and the seminar and conference participants at the 2014 System Microeconomics Meeting at FRB Minneapolis, the 2015 FRB Cleveland-University of Kentucky Labor Workshop, the 2015 Society of Economics Dynamics Meeting in Warsaw, the 2015 Econometric Society Meeting in Montreal, the 2015 Society of Labor Economists Meeting in Montreal, the 2015 CADRE Workshop at FRB Kansas City, the 2015 System Microeconomics Meeting at FRB Dallas, the University of Montreal, the FRB Chicago, the FRB SF, the FRB, and the BLS. The authors thank Theodore Naff for expert editorial assistance. The views expressed here are those of the authors and do not necessarily reflect those of the Federal Reserve Bank of San Francisco, the Federal Reserve System, or any other institution with which the authors are affiliated. All errors are ours. 0

3 1. Introduction Characterizing the heterogeneity that leads to finding employment and measuring employment potential of the non-employed population are central to labor economics. For these purposes, economists predominantly rely on the monthly cross-sections of the Current Population Survey (CPS) and specifically the survey questions designed to measure how closely attached the nonemployed are to the labor market. Much of the literature has focused on the distinction between the unemployed (i.e., those non-employed who report actively searching and available for work) and those OLF, using the former as its main measure of the population of job seekers. However, the basic distinction based on the cross-sectional CPS falls short of describing transitions in the labor market. Even though unemployment does on average predict re-employment, the majority of transitions to employment are by those OLF. Even though the OLF population constitutes about 90% of the total non-employed population, the factors that predict employment transitions of the OLF are understudied. For the unemployed, the CPS collects information on self-reported duration of unemployment which has been shown to strongly predict future transitions to employment. No such information is collected for the OLF and instead the standard way to classify the OLF by their labor market attachment is based on their self-reported desire to work and reasons for not searching. 1 However, it seems likely that the same factors that correlate with future transitions to employment among the unemployed, namely durations, also correlate with transitions into employment among the OLF since the distinction between unemployment and OLF is not clear-cut but a matter of degree. 2 Theoretically, as much as active search (unemployment), passive search or even waiting (OLF) can be a productive activity in obtaining a job. 3 In addition, search activity of those who want and are available to work might not be characterized by the same intensity in every month. Thus, the lack of active search by the OLF in the current period might not be informative about job search activities in the preceding periods. Empirically, the evidence from the CPS Re-Interview Survey suggests that the distinction between unemployment and OLF is not sharp (Jones and Riddell, 1999). In the course of this survey a subsample of CPS respondents were interviewed again a week 1 See Jones and Riddell (1999, 2006) and Flinn and Heckman (1983). 2 See Hall (1970) and Clark and Summers (1979) for the discussion of the ambiguity of the distinction. 3 See, for example, the stock-flow search model of Coles and Smith (1998) or the waiting at the airport example of Hall (1983). 1

4 after the original interview about their labor market activities during the original reference week. The re-interview leads to substantial differences in classification (Abowd and Zellner, 1985, and Poterba and Summers, 1986). Relatedly, the unemployed who were employed or OLF in the previous month often report durations of unemployment longer than one month. Consequently, current-month information collected from the non-employed respondents poses a few crucial puzzles. We set out to solve these by examining the information leading up to the current month that is available in the CPS. In this paper, we propose a novel approach for studying attachment to the labor market among the non-employed. Our approach exploits the short four-month panels available in the CPS. Using the first three months of each panel, we generate three-month sequences of labor force statuses ending in non-employment (unemployment and OLF) in the third month. We then examine transitions from non-employment in month three to employment in month four conditional on the three-month histories of the non-employed ( LFS history ). Our main finding is that the transitions from non-employment to employment decline with the duration since last employment, regardless whether we consider the unemployed or the OLF, whether for the unemployed or OLF. Thus, the LFS histories which can inform us about the duration since last employment (or duration of joblessness ) offer a unique perspective that cannot be obtained from current-month information. For the unemployed, we show that the data on self-reported unemployment duration and joblessness durations in the LFS histories often disagree. In particular, 20% of those who transition from employment and 50% of those who transition from OLF into unemployment report durations of more than 5 weeks during their first month unemployed. We find that there is negative duration dependence in job finding rates with duration of joblessness but no negative duration dependence with reported duration of unemployment once the duration of joblessness is controlled for. That is, conditional on having been recently employed, the unemployed reporting durations of unemployment longer than a month find jobs at similar rates as those reporting durations shorter than a month. We find that respondents are more likely to report longer unemployment durations after recent employment if the employment was short-term or if they started on-the-job search during the employment spell. Their duration reports are not arbitrary and are consistent with the concept of the duration of looking for work, which is the actual question in the CPS used by analysts to code unemployment 2

5 durations. 4 While recent employment breaks the negative duration dependence in unemploymentto-employment exits, this is not the case if longer durations of unemployment are reported after OLF. Among those OLF, the duration since last employment is the single most powerful predictor of future employment. Critically, the duration measure of joblessness for the OLF can only be obtained using the LFS histories. The average OLF individual who was recently employed has, regardless of the self-reported desire for work, twice as high a job finding rate as the average OLF individual who reports wanting a job. The majority of those wanting a job have histories with no recent employment and thus low employment transition rates. We also find that not only the duration since last employment but also the duration (continuity) of the last employment matters: longer spells are associated with higher employment transition rates. We identify a large group of OLF those with recent continuous employment - with a high employment transition rate of 0.40, second only to the rate of unemployed with recent continuous employment, 0.46, but much higher than the rate of other unemployed. Finally, conditional on the labor force status in the past two months, the unemployed have higher employment transition rate in the following month than the OLF. The documented regularities hold by detailed demographic categories by age, gender and education. To illustrate the predictive ability of the LFS histories, we run a horse-race between the 18 LFS histories and the current-month detailed classification by duration and reason for the unemployed as well as the desire to work, reason for not actively looking for work, school attendance, retirement, disability for the OLF. The results of this horse-race confirm that LFS histories explain a much larger fraction of the variation in the employment transition rates than do the current month classifications. For example, among those OLF, two months of the past labor force status explain four times more of the variation of the transitions into employment in the next month than the current-month information designed to capture labor market attachment. The main explanatory 4 Upon reentering unemployment after a short period of employment respondents often report durations that are consistent with the durations prior to the interrupting employment spell. The data are consistent with some respondents stopping their clock during the short-term employment when reporting the number of weeks of looking for a job afterwards (the discrepancy of 0), while many others continue the search throughout the short-term employment period (the discrepancy of up to two months). 3

6 power of the LFS histories comes from the information on the duration since and the continuity of the most recent employment among the non-employed. Using our proposed approach, we examine two applications for its use in the labor literature. First, reported unemployment durations are used in a variety of applications as if they were in fact distributions of unemployment or more broadly joblessness. 5 Our findings strongly suggest that they are not. Thus, we propose a correction of the distribution of reported unemployment durations using the observed joblessness duration based on the LFS histories rather than the reported durations whenever possible. Despite the short panels in the CPS, correcting the distribution of reported unemployment goes a long way towards obtaining an estimate of the duration of joblessness among the unemployed. The resulting distribution has a 5 pp higher mass on onemonth durations than the standard one. Second, we challenge a common practice in the literature that treats transition reversals between unemployment and OLF (nonparticipation) as a classification error, i.e., a DeNUNification procedure which recodes the unemployment-nonparticipation cyclers into one of the two continuous non-employment statuses (Elsby, Hobijn, and Sahin, 2015). We test the hypothesis that frequent switching between labor force statuses represents pure classification error by comparing the job finding rates of the non-employed with different LFS histories and wages upon employment. We find that the OLF with recent unemployment have five times higher job finding rate than those OLF for three consecutive months. The unemployed with recent OLF and the unemployed for three consecutive months have similar job finding rates but the former have lower wages upon reemployment. The differences in outcomes rule out pure classification error as an explanation behind the observed histories. Rather, frequent status changes between unemployment and OLF are informative about how attached to the labor market a non-employed person is. Our work contributes to a long strand of the literature that focuses on non-employment and factors associated with transitions from non-employment to employment. Following Flinn and Heckman (1983), most of the literature focuses on unemployment. Blanchard and Diamond (1990), Fallick and Fleischman (2004), Kudlyak and Schwartzman (2012), and Elsby, Hobijn, and Sahin (2015) 5 For a list of applications see literature on Ins and Outs of unemployment, or on ex-ante heterogeneity versus true negative duration dependence in exit rates from unemployment (Hall, 2005, Fujita and Ramey 2009, Elsby, Michaels, and Solon, 2009, Shimer, 2012, Hornstein, 2012, and Ahn and Hamilton, 2016). 4

7 discuss the transitions to and from OLF. 6 However, these studies do not explore four-month LFS histories. Similarly, Jones and Riddell (2006), Krueger, Cramer, and Cho (2014), and Hall and Schulhofer-Wohl (2017) explore the panel dimension of the labor force status surveys but limit themselves to transitions from the current-month LFS into employment over a few subsequent months. Our work is also related to the literature on negative duration dependence in exit from unemployment (Nickell, 1979; Hornstein, 2012; Ahn and Hamilton, 2014; Alvarez, Borovikova, and Shimer, 2014; Kroft, Lange, Notowidigdo, and Katz, 2016; and Fujita and Moscarini, 2017). We extend that literature to document the negative duration dependence not only for the unemployed but also for those OLF. The prevailing approach to the puzzles that motivate our work centers on classification error. 7 The classification error approach implies there exists a sharp distinction between the unemployed and OLF and that it can be captured by the current-month information. In our work, we substitute LFS histories for the sharp distinction between search activity of the unemployed and those OLF. We find that supplementing current-month information of the non-employed with just two months of LFS, readily available in the CPS, uncovers a large and crucial dimension of heterogeneity of the non-employed in terms of their labor market attachment. The rest of the paper is structured as follows. Section 2 describes our construction of the LFS histories. Section 3 describes findings on the employment transition rates. Section 4 applies the LFS histories approach to examine labor market attachment of unemployment-nonparticipation cyclers. Section 5 runs the horse race between the LFS histories and the detailed information from the current month survey. Section 6 examines the self-reported labor market attachment and LFS 6 Central to models in this literature is the theoretical construct of job seekers. Quantitative or empirical research has typically used official unemployment as the empirical counterpart for this construct (with the notable exception of Veracietro, 2008). Only recently has the literature started to consider those OLF when exploring the dynamics of job finding rates in the U.S. labor market (Krusell, Mukoyama, Rogerson, and Sahin, 2012; Elsby, Hobijn, and Sahin, 2015; Elsby, Hobijn, Sahin, and Valletta, 2012; Hornstein, 2012; Diamond, 2013; Farber and Valletta, 2015; Kroft, Lange, Notowidigdo, and Katz, 2016; Rothstein, 2012; and Van Zandweghe, 2017). 7 For example, Abowd and Zellner (1985) and Poterba and Summers (1986) estimate the classification error based on the assumption that the Reconciled Subsample of the CPS Re-interview Survey contains true labor force status. Feng and Hu (2013) restrict measurement error structure to a Markovian one. Elsby, Hobijn, and Sahin (2015) use a DeNUNification procedure. 5

8 histories of those OLF. Section 7 studies reported duration of unemployment versus duration of joblessness in the labor force status histories. Section 8 concludes. 2. Constructing Labor Force Status Histories in the CPS 2.1. The Non-Employment Concepts in the Current Population Survey The data in the analysis are from the Current Population Survey (CPS) basic monthly files from January 1976 to March The CPS distinguishes between two groups of non-employed. According to the CPS Manual, the unemployed are those aged 16+ who did not work at all during the reference week, who were not absent from a job, who actively looked for work during the past four weeks and were available for work during the reference week. Persons who were on layoff from a job to which they expect to return and were available for work during the reference week are also classified as unemployed, even if they did not actively look for work. 8 The persons not in the labor force are those who did not work last week, was not temporarily absent from a job, did not actively look for work in the previous four weeks, or looked but was unavailable for work during the reference week; in other words, a person who was neither employed nor unemployed. (The CPS Manual) 2.2. The Labor Force Status Histories in the CPS We exploit the panel structure of the CPS to classify the non-employed based on their LFS history. In the CPS, respondents are interviewed for 4 consecutive months, then they are not interviewed for 8 months, and then they are interviewed again for 4 consecutive months. The interview months are labeled from 1 to 8, and are referred to as month-in sample (MIS, hereafter). The monthly CPS 8 The unemployed who are expected to return to a job are on layoff. The definition of layoff unemployment was tightened during the 1994 CPS redesign. After 1994, those on layoff must expect to be recalled to the job within 6 months or the employer must have given the person a specific date upon which they would be recalled in order to be counted as unemployed without actively searching for work. 6

9 file thus contains data from respondents in any of the eight interview months. We match the respondent s records across month-in-sample to obtain short four-month panels. 9 We focus on the LFS histories of the non-employed individuals - unemployed or OLF - in month three of the panels (i.e., in MIS-3 and MIS-7). We then study the employment transition rate from non-employment in month three, conditional on the three-month LFS histories, to employment in month four. 10 In order to arrive at population-representative samples, we weigh the data using the average of the CPS sampling weights in the third and fourth months of the four-month panels. There are 18 possible LFS histories that have either U or OLF in the third month and we refer to the histories by the sequences of statuses from t-2 to t, i.e., NEU is a history with OLF in t-2 (N denotes being OLF), employment in t-1 and unemployment in t. There are six different subpopulations of the non-employed based on the LFS histories: (1) unemployed, recently employed (EEU, EUU, UEU, ENU, NEU); (2) unemployed, not recently employed except UUU (NNU, NUU, UNU); (3) unemployed in the three consecutive months (UUU); (4) OLF, recently employed (EEN, ENN, NEN, EUN, UEN); (5) OLF, not recently employed except NNN (UUN, UNN, NUN); (6) OLF in the three consecutive months (NNN). 3. Heterogeneity in Employment Transition Rates by Labor Force Status History In this section, we study how monthly transition rates from non-employment to employment vary with the LFS histories. We document that both for the unemployed and those OLF the rates decline with the length of time since last employment and that, conditional on the duration since last employment, the duration of last employment matters The LFS Histories of the Non-Employed 9 To match the individual records month-to-month, we follow Madrian and Lefgren (1999) and Shimer (2012) and match individuals by race, age and sex besides individual and household ID. This approach minimizes errors in matching across months that arises due to the fact that the CPS uses a sample of addresses. Following the BLS approach, we do not impute missing observations or address the issue of possible varying responses conditional on the month in sample interview. We leave these questions for further research. 10 Hereafter, we will treat MIS 5 through MIS-8 in the same manner as MIS-1 through 4. 7

10 There is substantial heterogeneity in the previous two-month labor force status among the nonemployed (Table 3.1). On average, during , among the unemployed, 35% were recently employed in at least one of the two prior months, 36% were continuously unemployed for at least three months, and the rest were reporting some combination of unemployment and out of the labor force. Among those OLF, 87% were continuously OLF for at least three months, 8% were employed in at least one of the two prior months, and the rest were reporting some combination of unemployment and OLF. Some combination of the past labor force status are more prevalent than the others; however, the relative ranking of the LFS histories by their prevalence in the population persists over time. 11 We now turn to examining employment transition rates. Figure 3.1 shows average nonemployment to employment transition rates by LFS history (left axis) and their respective population shares (right axis). All 18 histories are labelled on the x-axis and ranked by their transition rates from the highest to the lowest. The histories of those OLF in the third month are in red and those of the unemployed - in black, a color scheme we maintain throughout the paper. Figure 3.1 shows that job finding rates are much higher among those who were recently employed, regardless whether they are currently unemployed or OLF. 12 Furthermore, job finding rates are highest among those who are newly non-employed and beforehand were continuously employed for at least two months. This finding suggests that it is important to explore both duration since recent employment and continuity of recent employment as predictors of future transitions to employment Duration since Recent Employment It is well known that, for the unemployed, job finding rates decline with the reported duration of unemployment. In this Section, we present evidence that job finding rates also decline in the duration of joblessness (i.e., the time since last recorded employment in the LFS histories). The non-employed, either unemployed or OLF, who were employed last month have higher 11 See Appendix Figure A3.1 for the time series of the shares of all 18 LFS histories in the civilian noninstitutionalized population 16 years or older. 12 See also Table 3.1. The ranking of the histories by the transition rates persists over time as can be seen in Appendix Figure A3.2 that shows the time series of the annual averages of monthly transition rates from non-employment to employment by detailed LFS history from 1976 to

11 employment transition rates than those who were last employed two months ago, and those in turn have higher employment transition rates than those with no employment in the past two months. Table 3.2 reports estimates from a simple employment transition linear probability model, estimated separately for the unemployed, for those OLF, and for the pooled sample of the unemployed and OLF. 13 Among the unemployed, those who have been jobless for only one month are 1.6 times as likely to transition to employment the next month as those who have been jobless for two months, and they are more than three times as likely to transition to employment as those who have been jobless for at least two months. Among those OLF, those who have been jobless for only one month are on average twice as likely to transition to employment the next month as those who have been jobless for two months, and they are ten times as likely to transition to employment as those who have been jobless for at least two months. From a one-month CPS data, it is not possible to construct a duration measure of non-employment for those OLF. Our findings based on the short panels of the CPS are therefore the first to document duration dependence among the OLF. For the unemployed, the CPS collects information on the duration of unemployment and these data have been extensively analyzed. However, in Section 7, we document that the reported duration of unemployment and duration of joblessness often disagree and that the rates decline with duration of joblessness but not necessarily with reported duration of unemployment, conditional on joblessness. Our findings thus point to an urgent need to revisit the research on duration dependence in unemployment and how it treats the available data on unemployment durations Duration of the Recent Employment We find that not only the duration since recent employment but also the duration of recent employment matters for future transitions to employment. 14 Those who were employed more continuously are more likely to transition back into employment than those who were employed 13 Table 3.2, Columns 1-3 contain the results from a model with three dummies that indicate the duration since the most recent employment an indicator for the most recent employment in month t-1, an indicator for the most recent employment in month t-2, and an indicator for no employment in month t-1 and t-2 (additional controls are age, gender, education, year and seasonal dummies). 14 We also find that past continuous employment predicts future continuous employment. These results are available upon request. 9

12 for a short period. 15 That is, EEU and EEN histories have higher employment transition rates than UEU, NEU, UEN, or NEN. Since 1976, 40% of those with the EEN histories transition back into employment within the next month compared to just 25% of those with the UEN histories. These results are shown in Table 3.2, Columns 4-6, which report estimates of a simple linear probability model with the indicators of the duration since and the duration of the most recent employment Unemployment versus OLF, Conditional on the Prior LFS History Our final results in this Section relate to how job finding rates vary between those who are currently unemployed and those who are OLF: conditional on their LFS histories, those who are currently unemployed find employment at higher rates than those who are currently OLF (Table 3.3 and Appendix Figure A3.3). If we do not condition on prior LFS histories, the average unemployed have a higher employment transition rate than the average OLF. This is a composition effect, however, because, as we have shown above, the OLF individuals with recent employment have much higher employment transition rates than the unemployed with no recent employment. Much of the unconditional difference is driven by a very large share of those who are OLF in the three consecutive months they are disproportionally retired and disabled (50% are retired and 13% are disabled in three consecutive months) with very low job finding rates (0.006 and 0.007, respectively) Robustness In this subsection, we present robustness checks with respect to frequent measurement or conceptual concerns. 15 Note that continuity of the employment does not necessarily imply employment with the same employer. 16 Table 3.2, Columns 4-6 shows the results from an employment linear probability model with four dummies that represent the nature of the previous employment captured by the LFS histories an indicator for employment in month t-1 and t-2 (i.e., most recent and continuous employment), an indicator for employment in t-1 only, an indicator for employment in t-2 only, and an indicator for no employment in neither t-1 nor t-2. That is, we disentangle the first dummy in columns 1-3 into two: one representing at least two months of employment and the other one representing only one month of employment. 17 Even after excluding these groups, the employment transition rate of those OLF in three consecutive months is still the lowest among all the 18 LFS histories, (Appendix Table A3.1 and Figure A3.5). 10

13 To begin, we note that the differences in the employment transition rates by LFS history are not driven by age, gender, or education. 18 This can be seen in Figure 3.2 that shows the average employment transition rates by LFS history with and without controls for demographics, where the histories are ranked by the employment transition rates Temporary Layoffs in Unemployment We next examine whether temporary layoffs can account for our results on negative dependence of transition rates on duration of joblessness for the unemployed. Specifically, Fujita and Moscarini (2017) argue that once recalls by prior employers are taken into account, the exit rate from unemployment does not exhibit negative duration dependence. While the CPS data do not have information for actual recalls in the CPS, we use temporary layoff to account for a least a portion of expected recall. Focusing on the LFS histories that end with unemployment (EEU, UEU, NEU, EUU, and UUU), we split each of these histories into two groups by reason for unemployment being temporary layoff (L) or other (O) as follows: the histories with temporary layoffs are EEL, UEL, NEL, ELL, LLL, and the histories with unemployment for reasons other than temporary layoff are EEO, UEO, NEO, EOO, OOO, respectfully. Using only those histories without temporary layoffs, we find that all documented regularities carry through (See Appendix Figure A3.4). Consistently with the earlier literature, temporary layoff is associated with higher transition probability to employment than other reasons for unemployment Waiting for a New Job to Begin One hypothesis for the high employment transition rate of some of those OFL (especially those with recent employment, EEN) is that these individuals already have a job lined up. However, this hypothesis is not consistent with the way the CPS classifies individuals into OLF or unemployment. Specifically, the CPS asks two different questions that contain information about waiting for job to begin. The individuals who indicate that they are waiting for a job to begin are not classified as OFL. First, they are classified as employed: waiting for a new job to begin is one of the possible reasons for those who state that they are employed but absent from job. 18 This is consistent with the finding by Shimer (2012). The population shares of different histories among different demographic groups, however, differ. 11

14 Second, they can be classified as unemployed. Specifically, those who are looking for work and answer that they are unavailable to start the job last week if the job had been offered, are asked why they are unavailable. The respondents who chose option waiting for new job to begin are classified as unemployed (The CPS Manual). Thus, waiting for a new job to begin does not account for the high employment transition rates observed among the OLF. 4. Are Unemployment-Nonparticipation Cyclers Misclassified? We have shown that LFS histories contain important information about future employment transition rates of the non-employed. In this section, we challenge a practice in the literature that treats frequent changes between unemployment and OLF (nonparticipation) as classification error Employment Transition Rates and Wages of Those Consistently Unemployed, Consistently OLF, and Unemployment-OLF Cyclers In two widely cited papers Abowd and Zellner (1985) and Poterba and Summers (1986) compared responses in the CPS 1981 Re-interview Survey with those in the original survey. They noted that many respondents who were classified as unemployed during the original interview were classified as employed or OLF based on the re-interview. 19 Abowd and Zellner (1985) and Poterba and Summers (1986) estimated the extend of the classification error by comparing original survey responses with those in the Re-interview survey assuming that responses to the Re-interview Survey were error-free. Unfortunately, the BLS has not conducted a re-interview survey since 1981 thus no more recent data on this issue is available. A concern arising when classification error is present is that the error induces spurious transitions between labor force states. To address this concern, Elsby, Hobijn, Sahin (2015) in a recent paper propose to treat transition reversals between unemployment and nonparticipation as classification error, a practice known as DeNUNification. Specifically, the NUN labor force status histories are recoded into NNN and UNU into UUU. The authors show that this practice substantially 19 During the Re-Interview Survey, a small subset of the individuals was contacted a week after their initial CPS interview and asked questions regarding their labor market-related activities in the initial reference week. 12

15 reduces estimated transitions in and out of the labor force and makes the transition rate between OLF and unemployment less counter-cyclical. The correction is based on the hypothesis that reversals between unemployment and nonparticipation represent spurious transitions between labor statuses. An alternative hypothesis is that the reversals are genuine and that respondents reporting UNU or NUN differ in how attached they are to the labor market from those reporting UUU or NNN, respectively. These hypotheses can easily be tested by comparing the job finding rates of those OLF with histories NUN and NNN and of those unemployed with histories UNU and UUU. The left-hand side of Panel A in Figure 4.1 shows monthly employment transition rates of those NUN and NNN. Those with NUN histories are five times more likely to transition to employment than those with NNN histories (0.10 versus 0.02, Table 3.1), even after controlling for demographics (Table 3.3 and Figure 5.1 below). In addition, the employment transition rate of the NUN histories is also much more cyclically volatile and declines more in recessions than the rate of the NNN histories. 20 Consequently, the non-employed with the NUN LFS histories are much more attached to the labor market than the non-employed with the NNN histories. They appear less attached than those with the UUU histories (the transition rate of 0.10 versus 0.15, Table 3.1) but clearly these individuals have not completely left the labor force. Figure 4.1, the left-hand side of Panel B shows employment transition rates of those UNU and UUU and the right-hand side shows the population shares. The transition rate of UUU histories is somewhat higher but the difference is statistically significant only in some years. 21 This observation supports the idea that at least for some non-employed there is little distinction between unemployment and out of the labor force and for the purposes of accounting for transitions between non-employment and employment, these histories are alike. 20 The right-hand side of Panel A in Figure 4.1 shows how prevalent the NUN and NNN histories are in the working age population. During , those NNN account for, on average, 31% of the working-age population. The group with NNN also constitutes almost 90% of all OLF so that movements in the aggregate labor force participation rate are largely accounted for by movements in the prevalence of this group. For instance, the U-shape of the population share of the NNN inversely tracks the increase and the post-2000 decline of the aggregate labor force participation rate. The population share of NNN shows a clear trend and no cyclical pattern. In contrast, the population share of NUN hovers around a quarter percent of the population and shows a clear countercyclical pattern. 21 The same holds if we control for age, gender, education, and seasonals (available upon request). 13

16 However, even though UNU (or NNU or NUU) have employment transition rates similar to UUU (Table 3.3), we find that those who find employment from UUU have higher wages than those who find employment from U but after circling between U and OLF, even after controlling for demographics (Table 4.1). That is, a unemployed continuously reporting active search (UUU) potentially signals a higher reservation wage than an unemployed who circles between unemployment and OLF or even an unemployed who takes on a short-term job (UEU) Discussion Our results on how employment transition rates differ with LFS histories challenge a practice in the literature that treats frequent changes between unemployment and OLF as a measurement error. We find that the non-employed with histories NUN have five times higher employment transition rate than those with NNN histories. The UNU and UUU histories have similar employment transition rates but the unemployment-nonparticipation cyclers have lower wage upon reemployment. Thus, there are important differences in outcomes for these different histories that rule out pure classification error to explain the observed histories. Consequently, NUN and UNU do not appear to be erroneous versions of NNN and UUU, respectively. The data support the interpretation that those who frequently changes status between OLF and unemployment are more closely attached to the labor market than those consistently OLF and less closely attached than those who are consistently unemployed. 5. A Horse Race: Labor Force Status History versus Detailed Current-Month Information In this section, we examine whether LFS histories are more predictive of future job finding rates than the information based on current-month survey responses only. For the unemployed, the most important variables based on current month survey responses in the literature are reported durations of unemployment and the reason of unemployment. For those OLF, the most important variables related to labor force attachment are self-reports on the desire for work and reasons for not actively looking for work. We therefore consider a horse-race between these current-month responses and the labor force histories as a way of illustrating how informative the labor force histories are for predicting transitions into employment. 14

17 5.1. Measuring Labor Force Attachment Using Current-Month Information The CPS contains a set of questions about search activities, desire to work and other activities of the non-employed that are used to classify the non-employed into unemployed versus OLF and also allow distinguishing within OLF between groups with different degree of labor market attachment. The OLF are asked a sequence of questions to determine how closely they are attached to the labor market. To begin, they are asked whether they currently want a job. If a person indicates that she wants a job, the person is asked about the main reason for not looking for work during the last 4 weeks. Respondents who want a job are also asked about their search behavior in the last 4 weeks and the last 12 months. Based on their responses, the Bureau of Labor Statistics assigns to the OLF a label indicating how attached they are to the labor market. Those who want a job, are available for work, who have looked for a job sometime in the prior 12 months (or since the end of their last job if they held one within the past 12 months), but were not counted as unemployed because they had not searched for work in the 4 weeks preceding the survey are referred to as marginally attached. Among the marginally attached, the BLS distinguishes between those who gave an economic-related reason for not looking for work (referred to as Want job, marginally attached, discouraged ), those who gave a non-economic-related reason for not looking for work (referred to as Want job, marginally attached, other ), and those who want a job but are neither in the first nor in the second category (referred to Want job, other ). Finally, the BLS subdivides those who do not want a job into the retired, disabled, those in school, and other. In summary, the BLS classifies those OLF into seven groups: (1) want job, marginally attached, discouraged, (2) want job, marginally attached, other, (3) want job, other, (4) do not want job, retired, (5) do not want job, disabled, (6) do not want job, in school (16-24 years old), and (7) do not want job, other. We distinguish among the unemployed by self-reported duration: the short-term unemployed the unemployed whose duration of unemployed is less than 5 weeks, medium-term unemployed (5-27 weeks of unemployment), and long-term unemployed (unemployed for 27 weeks or longer), and by reason for unemployment (on temporary layoff, on permanent layoff, quit, temporary job ended, new entry, re-entry) Horse-Race between the LFS History and Current-Month Classification 15

18 We now run a horse race between the three-month labor force status history and the detailed current-month classification in explaining variation in monthly employment transition rates of the non-employed. Since the detailed current-month classifications for the unemployed and the OLF are mutually exclusive by duration and reason for the former and by desire to work and other labor market activities for the latter, we first analyze the histories for the unemployed and for those OLF separately before pooling the histories of the unemployed and those OLF. Table 5.1 presents estimates of a linear employment transition probability model for the unemployed. The sample consists of the unemployed in month three of our four-month panels for whom we can construct the three-month LFS history and who we can classify by duration and by reason for unemployment. The right-hand side variables of interest are sets of dummy variables representing alternative classifications of the unemployed. The first set represents the nine LFS histories ending in unemployment. The other two sets are dummies for each duration category (short-, medium- and long-term unemployed) and dummies representing reason for unemployment (six categories). We also control for age, gender, education, and dummies for year and for month. In Columns 1-4 in Table 5.1, we examine specifications that include either only the dummies representing the LFS histories or dummies representing duration and/or reason of unemployment. Comparing the R-squares across Columns 1-4, we find that the classification by LFS history explains more variation in the employment transition rates of the unemployed than the classification by reported unemployment duration or by reason for unemployment, and about the same proportion of the variation in the employment transition rate as do duration and reason of unemployment together. Adding LFS history classification to the duration classification or to the classification by reason substantially improves explanatory power. In contrast, adding duration classification to the LFS history classification does not add much to explaining the variability; adding the reason for unemployment to the LFS history classification goes a bit further in explaining the variation. Overall, we find that the LFS history classification included by itself explains as much as simultaneously including classifications by duration and reason of the unemployed, but adding the 16

19 LFS history to these two classifications provides additional explanatory power (Column 4 versus Column 7). 22 We now run a similar horse race for those OLF (Table 5.2). We again use sets of dummies to represent the nine LFS histories of the OLF and a set of seven dummies for the desire for work and other labor market attachment groupings described above. Additional controls in the regressions are age, gender, education, dummies for year and for month, and a constant. Column 1 shows the results from a regression with only the LFS histories classification dummies, Column 2 shows the results from a regression with only the seven current-month categories, and Column 3 shows the results from a regression with both sets of dummies included. As can be seen, the R 2 from the regression on the set of the LFS dummies is almost four times as large as the R 2 from the regression on the set of the current-month dummies (Column 1 versus 2). Adding the currentmonth classification to the LFS history classification does not add to the explanatory power (Column 1 versus 3). Finally, we estimate a linear probability model of employment transitions estimated on the pooled sample of the unemployed and those OLF (Table A5.1). When we use only a simple dichotomous classification of the non-employed a dummy for unemployment versus for out of the labor force, the R 2 is When we add a set of dummies for unemployment duration, a set of dummies for reason for unemployment, a set of seven dummies for desire to work and labor market activities of those OLF, the R 2 goes up to When we further add the set of dummies for 18 LFS histories, the R 2 goes up to Finally, omitting the set of seven dummies for desire to work and labor market activities of those OLF leaves the R 2 almost unchanged at These results clearly demonstrate that LFS histories contain much more information about the employment transition propensity of the non-employed than the one-month LFS. Equally important, the LFS histories contain information beyond what can be extracted from the currentmonth variables on duration and reason for unemployment and the desire for work and labormarket activities of those OLF. Finally, most of the explanatory power is driven by the LFS histories of those OLF. For these individuals, the LFS histories contain information about the 22 The likelihood ratio test rejects Model in Column 4 in favor of Model in Column 7. 17

20 duration since the most recent employment, which is a crucial factor for predicting future employment and which is not extracted from the current-month CPS questions asked of those OLF Detailed Labor Force Status History versus Sequence of Past Statuses In Tables we considered how well a model with a full set of dummies for the 18 different LFS-histories leading up to U or N performed in predicting employment transition. Next, we consider its performance relative to a simpler, more parsimonious specification relying simply on a set of non-interacted indicator variables for LFS in each of the current and past two months. After omitting a base LFS category in each of these three months, this simpler model only requires estimating 5 coefficients as opposed to the 17 required for the specification that is fully saturated for LFS-histories. This simpler model restricts the effects of the different LFS status in different months to be the same regardless of the status in the other months. For example, employment as opposed to OLF in month t-1 is associated with the same change in the probability of reemployment regardless whether the individual is unemployed or OLF in month t; and it is associated with the same change in the probability of reemployment regardless of the LFS in month t-2. We find that the R-squares from the two models are similar, but the likelihood ratio test rejects the parsimonious model in favor of the full 18 histories model. 23 Among the five coefficients on the LFS, the largest positive coefficient is on the dummy for employment in period t-1, the second largest coefficient is on the dummy for employment in period t-2, and finally the third largest coefficient is on the dummy for unemployment in period t. That is, first, for the non-employed, duration since the most recent employment is the most powerful predictor of employment in the future. Second, unemployment versus OLF in the history predicts higher employment transition rate, conditional on the rest of the history. However, in contrast to the LFS histories model, the simple model of the past labor force status does not allow distinguishing between short- versus long-duration of the previous employment, which we find being an important factor in transitions to employment. 23 The estimates from the full 18 LFS histories model and the simple model of past labor force statuses are in Appendix Table A

21 6. Labor Force Status History and Self-Reported Labor Attachment of the OLF In the previous Section, we demonstrated that labor market attachment based on self-reported desire to work and reason for not looking explains little of the variation in employment transition rates among the OLF compared to the LFS histories. In this Section, we examine employment transition rates by self-reported labor market attachment, conditional on the LFS history. We find that recent employment is a much more powerful indicator of the future employment than the selfreported labor market attachment. Figure 6.1 shows the estimates from the linear probability model of the individual OLF-toemployment transition rate on the full set of interactions of the seven self-reported labor market attachment dummies with nine LFS histories of the OLF (colored lines). For comparison, the figure also shows the estimates from the model with the seven dummies representing self-reported labor market attachment only (black line). The estimates of the transition rates across LFS histories reveal much larger differences associated with time since recent employment than by self-reported desire of work. An average OLF individual who reports wanting a job has a much lower job finding rate than an average OLF individual who was recently employed, regardless of her self-reported desire to work. In particular, those who were employed in the two months just prior to the current OLF (i.e., EEN) have 2-4 times higher employment transition rates than all other OLF. Even if they do not want a job and are retired or disabled, their job finding rate is higher than of the rate of those OLF want a job but have no recent employment. Those without recent employment have low employment transition rates, regardless of the self-reported labor market attachment. Examining the composition of the OLF by self-reported labor market attachment and LFS history, we find that the majority of those wanting a job have LFS histories with no recent employment (Table 6.1, Panel B) and thus low employment transition rates. On a flip side, those OLF with recent employment are most likely to report not wanting a job and being in school or retired (Table 6.1, Panel A). Wanting a job and being discouraged is most likely to be reported by those without recent employment (except for those NNN, who are predominantly retired). 19

22 7. Reported Duration of Unemployment versus Duration of Joblessness in the Labor Force Status Histories Above, we showed that for those OLF the duration since recent employment obtained from the LFS histories is the most important predictor of future employment. For the unemployed, we can also obtain durations of unemployment based on CPS survey questions about the duration. In this Section we show that respondents actually do not report their unemployment duration, but rather their duration of job search, which is the actual question in the CPS. It is however common practice to treat these durations as durations of reported unemployment. In what follows, we will refer to these durations as reported unemployment duration. In this section, we focus on the unemployed and examine how reported unemployment durations and observed joblessness durations interact in predicting transitions to employment. We first show that the duration of joblessness as measured in the LFS histories and the reported unemployment durations often disagree. More than 20% of those currently unemployed who were employed last month and more than 50% of those currently unemployed who were OLF last month report unemployment durations of more than one month (5 weeks or longer). Second, we find that the duration of joblessness is a better predictor of employment transitions than is the reported duration of unemployment. Those employed last month and reporting 5+ weeks of unemployment are almost as likely to find a job as those reporting less than 5 weeks of unemployment. Third, we find that reported durations of 5+ weeks are more common after short-term employment or on-the job search. This is consistent with respondents including on-the-job search and possibly omitting short interruptions due to stop-gap jobs in their answers regarding the duration of looking for work. Finally, based on our findings, we construct a new distribution of joblessness for the unemployed that combines the observed duration of joblessness for those who we observe transitioning from employment to unemployment in the LFS histories with the reported duration of unemployment for the rest for the unemployed. The distribution constructed in such a way has a higher mass on shorter durations than the distribution of reported durations, which is typically used in the literature. 20

23 7.1. Reported Duration of Unemployment and Previous Labor Force Status It is important to understand how the CPS collects the information that eventually is reported as duration of unemployment. The CPS collects this information from those classified as unemployed, i.e., the non-employed who are available for work and actively searching for work during the reference week. To determine their duration of unemployment, these respondents are asked how long they have been looking for work. It is important to realize therefore that the unemployed are asked about the duration of looking for work, not the duration of the periods that they satisfied all the necessary criteria to be counted as unemployed. Consequently, the variable does not necessarily capture the duration of unemployment in the sense of the duration of the official CPS definition of unemployment. Nevertheless, this reported duration is typically used in the literature as duration of unemployment and referred to as such. As we demonstrate next, this practice of coding durations of unemployment based on reported durations of looking for work is inconsistent with the data. Table 7.1, Panel A shows the distribution of the reported duration of unemployment for the unemployed in month t conditional on the labor force status in month t-1 (for and ). If the reported duration is consistent with the past labor force status, the unemployed whose labor force status in the previous month is employment or OLF should report unemployment duration less than 5 weeks. In contrast, we find that during , on average, 20% of the unemployed who were employed last month report duration of 5 weeks or longer and approximately 4% report durations longer than 26 weeks. Among the unemployed who were OLF last month, approximately 50% report duration of 5 weeks or longer and approximately 13% report durations longer than 26 weeks. The percentage of these seemingly inconsistent reports of long unemployment durations varies systematically over time (Figure 7.1). The share of such reports appears to be countercyclical, i.e., increasing during recessions and declining during recoveries. After 2007, the reports of durations 53 weeks or longer increased substantially among the new unemployed, reaching 5% of all new transitions from employment and 25% of all new transitions from OLF. The cyclical behavior of reports of long (5-26 weeks) versus very long (53 weeks or longer) durations also differs. The difference is especially striking after the recession. While the reports of 5-26 weeks of 21

24 duration declined rapidly to the pre-recession level in 2011, the reports of 53 weeks of longer durations continued to grow reaching its peak in An alternative way to see the discrepancy between the reported duration and previous month labor force status is to examine the distribution of the previous month status by reported unemployment duration of the unemployed (Table 7.1, Panel B). During , among the unemployed with reported durations 5-26 weeks, on average, 8.6% were employed last month and 16.6% were OLF. Among the unemployed with reported durations weeks, 4.9% were employed last month and 13.5% were OLF Reported Duration of Unemployment versus Duration of Unemployment in the LFS Histories and the Subsequent Employment Transition Rates We now examine how reported unemployment durations and observed joblessness durations interact in predicting transitions to employment. This analysis sheds light on the underlying behavior that gives rise to discrepancies between the two measures of durations. In this Section, we will argue that duration of joblessness and the reported duration of unemployment (or, precisely, looking for job) are economically distinct measures and capture different labor market states. This is not inconsistent with Section 5, where we showed that reported unemployment durations capture almost as much variation in the employment transition rates of the unemployed as do the LFS histories. Table 7.2 helps explain why the two classifications capture a similar fraction of the variation in the employment transition rates of the unemployed. It shows the employment transition rates for the unemployed grouped in three alternative ways: by reported duration (less than 5 weeks, 5-8 weeks, 9-14 weeks, weeks, weeks and 53 weeks and longer), by the LFS in the previous month (employment, unemployment or OLF), and by reported duration and the LFS in the previous month. The table also shows the share of each group among all unemployed. During , the employment transition rate of those reporting short durations (less than 5 weeks) is 0.35 and of those employed during the previous month is 0.44, a similar order of magnitude. These high-transition rate groups on average make up 34% and 23% of unemployment, respectively. The unemployed with longer reported durations or the unemployed without recent employment (i.e., recently unemployed or OLF) all have employment transition rates around 0.20 and make up the lion-share of the unemployed. 22

25 Consequently, either using the reported durations or the actual LFS histories captures relatively similar variation in the overall employment transition rates as we find in Section 5. While the fractions of variation is the same, it does not mean that the information captured by the reported unemployment duration and by the LFS history classification is the same. We now proceed examining what the cases of discrepancy between reported duration of unemployment and duration of the observed unemployment status in the LFS history signify. We will show that employment prior to unemployment resets the actual reported duration in terms of affecting job finding rates, regardless of reported duration. This is not true for OLF prior to unemployment. Figure 7.2 contains one of the most important results of the paper. It displays four estimated unemployment-to-unemployment transition rate profiles by reported duration: for all unemployed, those employed in the previous month, those unemployed in the previous month, and those OLF in the previous month. The estimates of a linear probability model of the employment transition rate on the set of dummy variables that represent interactions between the six reported duration categories (<5, 5-8, 9-14, 15-26, and 53+ weeks) and previous-month labor force status (employed, unemployed, OLF) with controls for age, gender, education, annual and seasonal dummies are given in Table 7.3. The regressions are estimated separately for the and for the periods. First, we find that conditional on employment in the previous month, there is little negative dependence of unemployment-to-employment exit on the reported duration. Specifically, conditional on being employed last month, during the average employment transition rate of the unemployed with reported duration under 5 weeks is 0.45, of the unemployed with reported duration of weeks , and of the unemployed with the reported durations of weeks The decline in the employment transition rate between reported duration of less than 5 weeks and of up to 15 weeks is not statically significant, and the estimated decline for those reporting durations of 53+ weeks is only six percentage points. Second, we do find economically and statistically significant negative dependence on the reported duration of the unemployment-to-employment exit rates of those who were unemployed or OLF in the previous month. For those OLF in the previous month, during the average 23

26 employment transition rate of the unemployed with reported duration under 5 weeks is 0.23, of the unemployed with reported duration of weeks , and of the unemployed with the reported durations of weeks (Table 7.2). Interestingly, for the longer reported durations, the exit by reported duration is very similar for those who were unemployed or OLF in the previous month, especially during sample period (Figure 7.2). However, for the reported durations of less than 9 weeks, the exit rate for those who were OLF in the previous month is lower than the exit rate for those who were unemployed in the previous month. Additionally, we find that among those who transition to U from OLF, the employment transition rate is 1.5 as large among those who report short durations (as consistent with their LFS history) versus among those who report long duration (5 weeks or longer), during on average 0.23 and 0.15, respectively. This suggests that those who transition to U from OLF and report that they are newly unemployed are in fact closer to short-term unemployed and likely just starting their search. By contrast, those who transition to U from OLF and report longer durations might indeed have been unemployed for long time. Consequently, reported duration is only a weak predictor of transitions into employment for those recently employed, whereas it is an important factor for those being unemployed or OLF in the previous month. Third, most striking differences in the employment transition rates of the unemployed are not by reported duration but by employment versus non-employment in the previous month, even after controlling for the reported duration. The largest differences are within duration categories between those unemployed who recently reported employment and those who reported unemployment or OLF. We find substantially larger employment transition rates among those who were recently employed regardless of their reported duration. In that sense, observed joblessness is more important predictor of future employment than the reported unemployment duration. Figure 7.2 thus shows that the negative duration dependence profile of the aggregate unemployment-to-employment exit rate is a combination of the following three phenomena: 1. High exit rate of recently employed, who constitute 60% of all reporting durations less than 5 weeks; 24

27 2. Much lower exit rate of recently unemployed or OLF, who constitute more than 90% of those reporting 5+ weeks; and 3. Negative duration dependence at longer durations of those reporting 5+ weeks who were recently unemployed or OLF. Importantly, the number of new transitions into unemployment that report longer than one month durations is large. 24 The shares of the new transitions into unemployment that report longer durations are countercyclical, both from employment and from OLF (Appendix Figure 7.3). This suggests that the spell of unemployment or looking for work is more likely to be interrupted by the spell of employment or OLF during recessions. The share of the new transitions from OLF into unemployment that report longer durations exhibits a noticeable upward trend. Prior to 2009, the largest share of new transitions into unemployed was of those transitioning from employment and reporting less than 5 weeks of unemployment. During , however, the share of new transitions into unemployment from OLF reporting longer than one month durations exceeded the share of new transitions from employment reporting less than 5 weeks durations Long Reported Durations, Short-Term Jobs and On-the-Job Search We have shown above that the duration of actual joblessness that can be constructed from the LFS histories is a more important factor in predicting the employment transition rate than the reported duration of unemployment. In this section, we specifically focus on the unemployed, who were employed last month, and report unemployment durations of 5 weeks or longer. We show that such reports are not reflecting a reporting error, instead they contain important information on the type of labor market activity and the type of jobs these individuals were engaged in prior to transitioning to unemployment. First, we find that those who transition into unemployment from short-term employment (less than one month) are more likely to report long durations of unemployment (5+ weeks) than those whose prior jobs lasted at least two months. Their reported unemployment duration is not an error: it 24 One category that is likely a true reporting error are the unemployed reporting durations of less than 5 weeks who were unemployed in the previous month as well. After the 1994 CPS redesign, this category declined to 1.8% of all unemployed. See Table

28 either reflects the short-term employment as a continuation of the spell of looking for a job that started prior to the short-term employment, or it reflects the duration of the spell of looking for a job just prior to the short-term job. Such short spells of employment reset the negative dependence of the job fining rate on reported duration. Second, for some of the new transitions from employment to unemployment, the reported longer duration of unemployment indicates that the individuals started on-the-job search during the preceding spell of employment. In such cases, the individuals who transitioned from employment to unemployment and report durations of 5+ weeks of looking for work in fact have higher employment transition rates than those who report starting of looking for a job. Finally, we find that the incidence of reported longer durations of new employment to unemployment transitions is countercyclical Not a Reporting Error Consider a case of an unemployed individual who was employed last month and reports longer than one-month unemployment duration in the current month. There are three possible hypothesis concerning the reporting behavior of such respondents. First, the previous labor force status might have been reported erroneously and the individual was in fact unemployed and looking for a job. Second, the current unemployment duration is reported erroneously and the individual s unemployment duration should be reported as less than one month. Finally, it is possible that neither the previous month employment nor the current month unemployment duration was erroneously reported. Instead, the individual was indeed employed last month but still considered herself as looking for a job and thus accurately reported a long duration. It is straightforward to statistically reject the first hypothesis based on the estimates of the employment transition rates (see Table 7.3). Specifically, the employment transition rate of the unemployed reporting longer durations, who were employed last month, are much higher than the transition rates of the unemployed reporting longer durations, who were non-employed last month. Thus, last-month employment of the unemployed reporting long durations does not represent reporting error. The observation that employment transition rates of those unemployed who were employed last month do not decline with reported duration as the rates of those unemployed who were nonemployed last month do is potentially consistent with the notion that durations are indeed 26

29 misreported. It is however also consistent with the possibility that the previous month employment and the reported duration of looking for job are correct but that recent employment resets the negative dependence of the job finding rate on reported duration. To understand whether long reported durations following employment are just erroneous or whether they represent real differences in search behavior across individuals, we examine the data further. We find that such discrepancies between observed joblessness and reported unemployment durations do not simply reflect erroneous reporting of long durations. Instead, being employed and looking for work are not mutually exclusive states and the respondents reported duration reflects this, resulting in perceived inconsistencies between reported durations and observed LFS histories. For example, before transitioning into unemployment the individual might have already started looking for a new job while employed, i.e., was engaged in on-the-job search. Alternatively, the individual might have taken a short-term job while continuing looking for a better job. We find that reported durations often reflect these periods of search while employed. To make our case for this interpretation, we now specifically examine situations when unemployed individuals who were employed last month report long durations of unemployment Short-Term Employment and On-the-Job Search We start the analysis by examining the duration of prior employment spells. First, we find that when prior employment spells were short, then new unemployed are more likely to report long unemployment durations and the average length of reported durations is higher. Specifically, Table 7.4 shows the average incidence of reported durations of less than 5 weeks, 5-26 weeks, weeks and 53 weeks or longer, by the length of previous employment. Figure 7.3 shows the time series of the incidence 25 Among the unemployed, who were employed in the previous month but non-employed two month ago, during , on average 40% report 25 We construct Table 7.4 and Figure 7.4 using the duration information of the unemployed in month four of our four-month panels (to increase the observed length of the LFS history prior to unemployment) who were classified as employed in month three. The four-month length of the panel allows distinguishing among the unemployed in month four who were employed in at least three consecutive months prior to unemployment, employed in two consecutive months and non-employed three month ago prior to unemployment, and finally employment one month prior to unemployment and non-employed two months prior to unemployment. 27

30 durations of 5 weeks or longer. The median duration of these longer reported durations is 19 weeks (see Appendix Figure A7.2 for the mean and median). The incidence and the reported durations are higher during the period. In contrast, among the unemployed, who were employed previously in the three consecutive months, on average 15% report durations of 5 weeks or longer. The median duration of these longer reported durations is 8.5 weeks. 55% of the unemployed reporting durations 5+ weeks, however, is from the LFS histories with three months of consecutive employment because the fraction of the unemployed with at least three month of employment is higher than the fraction with 1-month of employment (72 versus 14%). Second, we find that reported long durations after the short-term employment are consistent with the respondent s reported duration of unemployment prior to the short-term employment, i.e., the longer reported durations are not random. This strongly suggests that the reported unemployment durations, even if inconsistent with observed employment spells, do not tend to represent a reporting error. Specifically, for the individuals who were unemployed just prior to the short-term employment and right after (i.e., the LFS histories UEU), we can measure the difference between reported durations prior to and after the short-term employment spell. Among those reporting long durations subsequent to a short-term employment spell, the median difference is 6 weeks, while the 25 th percentile is 0 and the 75 th percentile is 11 weeks. These data are consistent with some respondents stopping their clock during the short-term employment when reporting the number of weeks of looking for a job afterwards (the discrepancy of 0), while many continue the search throughout the short-term employment period (the discrepancy of up to two months). In contrast, for the unemployed individuals who report short (less than 5 weeks) durations after the short-term employment, the distribution has a median of -2 weeks, with the 25 th percentile at - 10 and the 75 th percentile at 0, indicating that they have started a new unemployment spell. Next, we examine the subsequent transitions into employment conditional on reported duration and duration of the prior employment. Table 7.5 shows the incidence of reported short (<5 weeks) and long (5+ weeks) durations conditional on the duration of prior employment for the new transitions into unemployment. First, the unemployment with longer previous employment have higher employment transition rates. Second, conditional on short-term employment prior to unemployment, there is some negative duration dependence between subsequent transitions to employment and reported duration: the employment transition rate is 0.40 for the unemployed who 28

31 report duration under 5 weeks and for the unemployed who report durations of 5 weeks and longer. However, the negative duration dependence disappears or is overturned for the unemployed who had at least two consecutive months of prior employment. An average employment transition rate for these unemployed reporting duration under 5 weeks is while the average employment transition rate for these unemployed reporting duration of 5+ weeks is (Table 7.5). This suggests that the unemployed who start looking for a job while still employed have at least as high or higher employment transition rates than the unemployed who started looking for a job after separation with the employer. Finally, the incidence of reported longer durations of those unemployed who were employed last month is countercyclical (Figure 7.3). This is consistent with the emerging evidence that on-thejob search is countercyclical and might be driven by precautionary motives (see Ahn and Shao, 2017) Duration Distribution Corrected for Observed Joblessness and Implications From our discussion in Section 7.1, there are two different distributions by duration of unemployment: one by reported duration, which is typically used in the literature, and the other based on the duration of unemployment observed in the LFS histories. Since the unemployed that were employed or OLF in the previous month often report longer durations, these two distributions differ. Reported unemployment durations are, however, used in a variety of applications as if they were in fact distributions of unemployment or more broadly joblessness. We show that previous employment breaks the negative duration dependence of the unemployment exit rates. That is, those unemployed who were employed last month and who report durations of 5+ weeks of looking for work have employment transition rate similar to those of the short-term unemployed. This is not the case if the longer duration is reported by the unemployed who were previously OLF. Using this information, we construct a novel distribution of unemployment durations that uses the observed duration of joblessness in the LFS histories for those unemployed who transition from employment and the reported duration of unemployment for the rest for the unemployed. Specifically, we construct the duration distribution for the unemployed in month four in our fourmonth panels. The four-month panels allow constructing the distribution across four observed 29

32 duration bins: (1) less than 5 weeks (U4), (2) 5-8 weeks (U5.8), (3) 9-14 weeks (U9.14), and (4) 15 weeks or longer (U15+). If the unemployed in month four was employed in month three, we assign the duration of less than 5 weeks. If the unemployed in month four was unemployed in month three and employed in month two, we assign duration 5-8 weeks, if the unemployed in month four was also unemployed in month three and two and employed in month one, we assign duration of 9-15 weeks. For the rest of the unemployed in month four, we keep their reported durations. For comparison, we also construct the distribution of reported durations and the distribution of the duration of unemployment observed in the LFS histories. To construct the distribution from the LFS histories, we assign the unemployment duration for an unemployed individual in month four by directly observing how many consecutive months the individual is unemployed without taking into account the reported duration. 26 The averages of the three distributions are in Table 7.6. The duration distribution corrected for observed joblessness has a higher mass in short durations than the distribution of reported durations does, as anticipated from the results in Section 7.1. We find that during , on average, 36.8% of the unemployed were unemployed for less than 5 weeks using the duration distribution corrected for observed joblessness in the LFS histories and 34% of the unemployed were unemployed for less than 5 weeks using the reported durations. During , these numbers are 32.5% and 29.1%, respectively. In the distribution of observed unemployment in the LFS, this share jumps to 46% during , and 46.1% during Figure 7.6 shows the duration distributions over time. The distribution of duration of unemployed constructed from the LFS histories (red line) is relatively stable over time. In contrast, the distribution of reported durations (black) shows increasingly higher mass on longer durations. To understand the discrepancy, Figure 7.7 shows the distribution of reported durations of newly unemployed: from employment, from OLF, and from both employment and OLF. The unemployed from OLF increasingly report longer durations; there is also an increase in reported durations for 26 We construct the distribution across four observed duration bins as follows: (1) less than 5 weeks, when the individual is unemployed in month four but employed or out of the labor force in month three; (2) 5 weeks or longer but less than 9 weeks, when the individual is unemployed in months four and three but employed or out of the labor force in month two; (3) 9-14 weeks, when the individual is unemployed in months four, three and two but employed or out of the labor force in month one; and, finally, (4) 15 weeks or longer, when the individual is unemployed in all four consecutive months. 30

33 those from employment after recession. Some part of the discrepancy between the reported durations and the LFS histories after 1994 is due to the CPS redesign. In 1994, the CPS switched to the dependent interviewing whereby the duration for an unemployed individual is increased by one month if he was also unemployed in the previous month. Any initial longer duration of new transitions into unemployment is carried over to the next month if the individual remains unemployed. The distribution that corrects reported durations for the observed duration of joblessness in the LFS histories lies between the distribution of reported durations and the distribution that corrects for both employment and OLF in the LFS histories (blue line in Figure 7.6). This distribution shows a decreasing mass at durations of less than 5 weeks and an increasing mass at durations of 15 weeks and longer. In 2010, the share of short-term (less than 5 weeks) unemployment reached its minimum: at 20% under the corrected distribution and at 15% under the distribution of reported durations. In 2010, the share of long-term (15 weeks and longer) unemployment reached its maximum: at 55% under the corrected distribution and at 65% under the distribution of reported durations. The distribution of the duration of unemployment has important implications for the analysis of the inflows and outflows of unemployment and for the estimates of the models of ex-ante heterogeneity and true negative duration dependence in exit rates from unemployment, the central questions in the macro-labor literature (see Hall, 2005; Fujita and Ramey, 2009; Elsby, Michaels, and Solon, 2009; Shimer, 2012; Hornstein, 2012; and Ahn and Hamilton, 2016, among others). For example, the share of short-term unemployment plays the key role in the construction of the job finding rate proposed by Shimer (2012) and subsequently in the inflow-outflow analysis. 8. Conclusion We propose a novel approach to studying heterogeneity in transitions from non-employment (unemployment and OLF) to employment using labor force status history that be easily constructed from the publicly available four-month panels of the CPS data. 31

34 Using this novel approach, we characterize new important factors that are associated with higher transitions to employment that cannot be extracted from current-month information. Specifically, we find that among the OLF, information on recent employment from the LFS history explains four times more variation in the employment transition rate than the respondents reported desire and reason for not looking for work. Not only the duration since the recent employment but also the duration of the recent employment matters. For the unemployed, we are able to combine the information on the reported duration of unemployment with the duration of joblessness observed in the labor force status histories and study the employment transition rates and the incidence of short-term employment. Finally, based on our findings, we construct the distribution of unemployment durations that corrects the reported durations for the observed employment in the labor force status histories. The resulting distribution has a higher mass on short durations and lower mass on 15+ weeks durations, as compared to the distribution of reported durations which is typically used in the literature. Much is left for future work. One such direction is a distinction between ex ante heterogeneity versus causal effect of the behavior of the non-employed (for example, continuous active search versus circling between unemployment and OLF) on the employment transition probability. An extension of the analysis to the 8-month histories as well as a detailed information from the eight months surveys beyond the labor force status might be fruitful. Another direction for future research is measurement of full potential employment and resource utilization in the labor market and we have started the work in Horsntein, Kudlyak and Lange (2014b). 32

35 References Abowd, John M., and Arnold Zellner "Estimating Gross Labor-Force Flows," Journal of Business and Economic Statistics, Vol. 3 (3): Ahn, Hie Joo, and James D. Hamilton Heterogeneity and Unemployment Dynamics, University of San Diego, mimeo. Accessed at Ahn, Hie Joo, and Ling Shao Precautionary On-the-Job Search over the Business Cycle, mimeo. Accessed at Alvarez, Fernando, Katarina Borovickova, and Robert Shimer Decomposing Duration Dependence in a Stopping Time Model, mimeo. Blanchard, Olivier, and Peter Diamond The Cyclical Behavior of the Gross Flows of U.S. Workers, Brookings Papers on Economic Activity, Vol. 2: Clark, Kim B. and Lawrence H. Summers "Labor Market Dynamics and Unemployment: A Reconsideration," Brookings Papers on Economic Activity, Vol. 1: Coles, Melvyn G., and Eric Smith Marketplaces and Matching, International Economic Review, Vol. 39: Current Population Survey Interviewing Manual U.S. Department of the Census. Diamond, Peter Cyclical Unemployment, Structural Unemployment, IMF Economic Review, Vol. 61: Elsby, Michael W. L., Bart Hobijn, and Aysegul Sahin "On the Importance of the Participation Margin for Labor Market Fluctuations," Journal of monetary Economics, Vol. 72: Elsby, Michael W. L., Bart Hobijn, Aysegul Sahin, and Rob Valletta "The Labor Market in the Great Recession: an Update to September 2011," Brookings Papers on Economic Activity 103: Elsby, Michael W. L., Ryan Michaels, and Gary Solon "The Ins and Outs of Cyclical Unemployment," AEJ: Macro, Vol. 1 (1): Fallick, Bruce C., and Charles A. Fleischman Employer-to-Employer Flows in the U.S. Labor Market: the Complete Picture of Gross Worker Flows, FEDS Working Paper No

36 Farber, Henry S., and Rob Valletta Do Extended Unemployment Benefits Lengthen Unemployment Spells? Journal of Human Resources, Vol. 50(4): Feng, Shuaizhang, and Yingyao Hu Misclassification Errors and the Underestimation of the US Unemployment Rate, American Economic Review, Vol. 103(2): Flinn, Christopher J., and Heckman, James J "Are Unemployment and Out of the Labor Force Behaviorally Distinct Labor Force States?" Journal of Labor Economics, Vol. 1(1): Fujita, Shigeru, and Garey Ramey The Cyclicality of Separation and Job Finding Rates, International Economic Review, Vol. 50(2): Fujita, Shigeru, and Giuseppe Moscarini Recall and Unemployment, American Economic Review. Hall, Robert E Why Is the Unemployment Rate So High at Full Employment? Brookings Papers on Economic Activity, Vol. 3: Hall, Robert E Is Unemployment a Macroeconomic Problem? American Economic Review Papers and Proceedings, 1983, Vol. 73 (2): Hall, Robert E Employment Fluctuations with Equilibrium Wage Stickiness, American Economic review, Vol. 95 (1): Hall, Robert E., and Sam Schulhofer-Wohl Measuring Job-Finding Rates and Matching Efficiency with Heterogeneous Jobseekers, AEJ Macro. Hornstein, Andreas Accounting for Unemployment: The Long and Short of It, Federal Reserve Bank of Richmond Working Paper No Hornstein, Andreas, Marianna Kudlyak, and Fabian Lange. 2014a. Measuring Resource Utilization in the Labor Market. FRB Richmond, Economic Quarterly, Q Hornstein, Andreas, Marianna Kudlyak, and Fabian Lange. 2014b. A New Measure of Resource Utilization in the Labor Market. Federal Reserve Bank of Richmond, mimeo, April Jones, Stephen R. G., and W. Craig Riddell Unemployment and Nonemployment: Heterogeneities in Labor Market States, Review of Economics and Statistics, Vol. 88(2): Jones, Stephen R. G., and W. Craig Riddell "The Measurement of Unemployment: An Empirical Approach," Econometrica, Vol. 67(1):

37 Kroft, Kory, Fabian Lange, Matthew J. Notowidigdo and Lawrence F. Katz Long-Term Unemployment and the Great Recession: The Role of Composition, Duration Dependence and Non- Participation, Journal of Labor Economics, Vol. 34 (S1, Part 2): S7-S54. Krueger, Alan B., Judd Cramer and David Cho Are the Long-Term Unemployed on the Margins of the Labor Market? BPEA, Spring. Krusell, Per, Toshihiko Mukoyama, Richard Rogerson, and Ayşegül Şahin "Is Labor Supply Important for Business Cycles?" NBER Working Papers Kudlyak, Marianna, and Felipe Schwartzman Accounting for Unemployment in the Great Recession: Nonparticipation Matters, Working Paper 12(4) (Federal Reserve Bank of Richmond). Madrian, Brigitte C., and Lars John Lefgren A Note on Longitudinally Matching Current Population Survey (CPS) Respondents, NBER Working Paper No. t0247. Nickell, Stephen Estimating the Probability of Leaving Unemployment, Econometrica, Vol. 47(5): Polivka, Anne E., and Jennifer M. Rothgeb Overhauling the Current Population Survey: Redesigning the CPS Questionnaire. Monthly Labor Review, Vol. 116(10): Poterba, James M., and Lawrence H. Summers "Reporting Errors and Labor Market Dynamics," Econometrica, Vol. 54(6): Rothstein, Jesse "Unemployment Insurance and Job Search in the Great Recession," Brookings Papers on Economic Activity, Fall: Shimer, Robert Reassessing the Ins and Outs of Unemployment, Review of Economic Dynamics 15, Veracierto, Marcelo Worker Flows and Matching Efficiency, Federal Reserve Bank of Chicago Economic Perspectives Vol. 35 4th Quarter: Van Zandweghe, Willem The Changing Cyclicality of Labor Force Participation, Economic Review, Q3:

38 Table 3.1: Employment Transition Rates and Population Shares, by Labor Force Status History Employment transition rate Population share LFS History EEU % 0.59% EEN % 1.11% UEU % 0.11% NEU % 0.05% EUU % 0.33% ENU % 0.11% UEN % 0.07% NEN % 0.38% EUN % 0.12% NNU % 0.47% UUU % 1.29% ENN % 1.15% NUU % 0.35% UNU % 0.24% UUN % 0.35% NUN % 0.35% UNN % 0.58% NNN % 30.65% Note: The sample consists of the four-month CPS panels, restricted to the panels with non-employment (unemployment or OLF) in month three. The tables shows the sample mean of annual averages of monthly (discrete) rates and shares. E in the histories denotes employment, U unemployment, and N OLF. All calculations use the CPS sampling weights as described in the text. The histories are ranked by the average employment transition rate over , the first column. 36

39 Table 3.2: Employment Transition Rates: Duration since and Duration of the Recent Employment Unemployed OLF U+OLF Unemployed OLF U+OLF Indicators for the most recent employment: Employment in t *** 0.344*** 0.370*** x x x ( ) ( ) ( ) E in t-1, E in t-2 x x x 0.419*** 0.385*** 0.405*** ( ) ( ) ( ) E in t-1, non-e in t-2 x x x 0.341*** 0.240*** 0.272*** ( ) ( ) ( ) Employment in t *** 0.171*** 0.198*** 0.256*** 0.171*** 0.198*** ( ) ( ) ( ) ( ) ( ) ( ) No employment in t-1,t *** *** *** 0.122*** *** *** ( ) ( ) ( ) ( ) ( ) ( ) Age, gender, educ, year, seasonls yes yes yes yes yes yes Observations 191,738 1,996,728 2,188, ,738 1,996,728 2,188,466 F-stat/p-value Adjusted R-squared Note: Table contains estimates from a linear probability model of non-employment to employment transitions. Columns 1-3 show the results from a model with three dummies that indicate the duration since the most recent employment an indicator for the most recent employment in month t-1, an indicator for the most recent employment in month t-2, and an indicator for no employment in month t-1 and t-2. Columns 4-6 shows the results from an employment linear probability model with four dummies that represent the nature of the previous employment captured by the LFS histories an indicator for employment in month t-1 and t-2 (i.e., most recent and continuous employment), an indicator for employment in t-1 only, an indicator for employment in t-2 only, and an indicator for no employment in neither t-1 nor t-2. Essentially, we disentangle the first dummy in columns 1-3 into two: one representing at least two months of employment and the other one representing only 1 month of employment. Additional controls are age, gender, education, year and seasonal dummies. The omitted category for demographics is year old female with high school education. The sample consists of the four-month CPS panels, restricted to the panels with non-employment in month three, The standard errors are in parentheses. The estimation employs the CPS sampling weights. 37

40 Table 3.3: Employment Transition Rates of the Unemployed and OLF, Conditional on Prior Two- Month Labor Force Status All non-employed in t Indicators for LFS(t-2)LFS(t-1) EE 0.383*** ( ) UE 0.272*** ( ) EU 0.237*** ( ) NE 0.231*** ( ) EN 0.161*** ( ) UU *** ( ) NU *** ( ) UN *** ( ) NN *** ( ) Indicators for interactions of LFS(t-2)LFS(t-1) and U(t) EEU *** ( ) UEU 0.101*** ( ) EUU *** ( ) NEU 0.124*** ( ) ENU 0.121*** ( ) UUU *** ( ) NUU *** ( ) UNU *** ( ) NNU 0.127*** ( ) Age, gender, education, year, seasonls yes Observations 2,188,466 Adjusted R-squared F Note: The table shows the estimates from the linear probability model of individual transitions from non-employment to employment. The sample consists of the four-month CPS panels, restricted to the panels with non-employment (unemployment or OLF) in month three, The regression with no constant. The omitted category for demographics is year old female with high school education. The standard errors are in parentheses. The estimation employs the CPS sampling weights. 38

41 Table 4.1: Real Wages of Newly Employed Workers, by LFS History UUU omitted EEU *** ( ) UEU (0.0126) NEU * (0.0180) EUU *** ( ) ENU *** (0.0141) UNU *** (0.0134) NNU *** ( ) NUU *** (0.0110) EEN *** ( ) UEN *** (0.0198) NEN *** (0.0113) EUN (0.0158) ENN ( ) UNN *** (0.0125) NNN *** ( ) NUN *** (0.0144) UUN *** Age, gender, education, (0.0133) yes year, seasonals, constant Observations 57,340 Adjusted R-squared r F df_m 60 Note: The table shows the coefficients estimates from a regression of log real hourly wages of new hires from non-employment with different LFS histories. The sample is restricted to the new hires only. The standard errors are in parentheses. The regression is estimated using the CPS sampling weights. 39

42 Table 5.1: Employment Transition Rate of the Unemployed: Labor Force Status History versus Current-Month Information LFS Histories Classification of the Unemployed LFS Histories + Duration Duration Reason Duration + Reason LFS Histories + Reason LFS Histories + Duration + Reason (1) (2) (3) (4) (5) (6) (7) LFS Histories (9 groups) yes x x x yes yes yes Duration of Ut (3 gr) x yes x yes yes x yes Reason for Ut (6 gr) x x yes yes x yes yes Age, gender, educ, year, seasonals, cons yes yes yes yes yes yes yes Observations 191, , , , , , ,738 Adjusted R-squared F-stat LR chi2(8) Model 4 vs Model 7 = Note: The sample consists of the four-month CPS panels, restricted to the panels with unemployment in month three, The regressions are estimated with a constant. When including sets of dummies for LFS histories, duration or reason for unemployment, we omit one category from each classification to avoid multicollinearity. The classification by reported duration of unemployment consists of the three categories: less than 5 weeks, 5-26 weeks, and 27+ weeks. The classification by reason for unemployment consist of six categories: temporary layoff, permanent layoff, temporary job ended, quit, re-entry, and new entry. The demographics controls are a set of 2 dummies for gender, a set of 7 dummies for age, and a set of 4 dummies for education; we omit one category from each set to avoid multicollinearity. The estimation employs the CPS sampling weights. 40

43 Table 5.2: Employment Transition Rate of the OLF: Labor Force Status History versus Current- Month Information Classification of the OLF LFS Histories Currentmonth attachment LFS History + Currentmonth attachment (1) (2) (3) LFS Histories (9 groups) yes x yes Desire to work, labor force attachment, retired, in school, disabled (7 gr) x yes yes Age, gender, educ, year, seasonals, cons yes yes yes Observations 1,996,728 1,996,728 1,996,728 Adjusted R-squared F-stat LR chi2(8) Model2 vs Model3 = Note: The sample consists of the four-month CPS panels, restricted to the panels with OLF in month three, The regressions are estimated with a constant. When including sets of dummies for LFS histories or the current-month classification, we omit one category from each classification to avoid multicollinearity. The classification by current-month labor market attachment consists of the seven categories: do not want job retired, disabled, in school, or other, and want job discouraged, marginally attached but not discouraged, and other. The demographics controls are a set of 2 dummies for gender, a set of 7 dummies for age, and a set of 4 dummies for education; we omit one category from each set to avoid multicollinearity. The estimation employs the CPS sampling weights. 41

44 Table 6.1: Self-Reported Labor Market Attachment and LFS Histories of the OLF, Average Shares in Panel A. Self-Reported Labor Market Attachment by LFS Histories LFS History EEN UEN NEN EUN ENN NUN UNN UUN NNN Want job Marg. attach., discouraged 1.2% 5.0% 0.8% 7.0% 0.9% 6.2% 5.1% 10.6% 0.3% Marg. attach., other 1.8% 6.6% 1.7% 9.9% 1.6% 11.3% 7.7% 14.5% 0.5% Other 16.6% 26.9% 10.2% 26.9% 8.2% 22.0% 16.5% 25.4% 2.9% Do not want job Other 36.5% 32.8% 35.2% 32.4% 31.5% 29.3% 33.0% 28.7% 18.3% In school, y o 19.1% 19.2% 24.3% 14.5% 21.8% 21.1% 20.5% 12.6% 10.3% Disabled 6.3% 3.8% 6.4% 3.9% 9.9% 4.0% 9.4% 4.3% 15.3% Retired 18.4% 5.7% 21.3% 5.5% 26.1% 6.0% 7.7% 3.9% 52.5% All 100% 100% 100% 100% 100% 100% 100% 100% 100% Panel B. Composition of Self-Reported Labor Market Attachment by LFS Histories LFS Histories Marg. attach., discoura ged Self-Reported Labor Market Attachment Want job Do not want job Marg. Want Disabled attach., job, other other Not in school, retired or disabled In school y o Retired Recently employed EEN 6.6% 5.7% 12.1% 5.8% 5.8% 1.4% 1.2% ENN 5.2% 5.1% 6.2% 5.2% 6.5% 2.3% 1.8% NEN 1.5% 1.9% 2.6% 1.9% 2.5% 0.5% 0.5% EUN 4.2% 3.3% 2.2% 0.6% 0.5% 0.1% 0.0% UEN 1.6% 1.3% 1.2% 0.3% 0.4% 0.0% 0.0% No recent employment UNN 14.7% 12.2% 6.3% 2.8% 3.0% 1.1% 0.3% UUN 18.1% 13.7% 5.8% 1.4% 1.1% 0.3% 0.1% NUN 10.4% 11.0% 5.0% 1.5% 1.9% 0.3% 0.1% NNN 37.6% 45.8% 58.6% 80.6% 78.2% 94.1% 96.0% All 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Note: See Section

45 Table 7.1: Reported Unemployment Durations and Previous Month Labor Force Status of the Unemployed Panel A. Distribution of the Reported Unemployment Durations, by Previous LFS Reported duration in month t, weeks LFS in month t Total E 79.2% 17.1% 2.0% 1.7% 100.0% N 52.0% 34.4% 5.8% 7.7% 100.0% U 10.2% 62.5% 13.7% 13.6% 100.0% E 80.6% 15.1% 2.1% 2.2% 100.0% N 45.2% 36.8% 7.4% 10.6% 100.0% U 3.5% 63.2% 15.6% 17.6% 100.0% Panel B. Distribution of the Previous LFS, by Reported Unemployment Duration Reported duration in month t, weeks LFS in month t E 53.3% 8.6% 4.9% 4.2% N 32.7% 16.6% 13.5% 17.0% U 14.0% 74.8% 81.6% 78.8% Total 100.0% 100.0% 100.0% 100.0% E 61.1% 7.3% 4.2% 3.6% N 32.9% 17.4% 14.8% 18.1% U 6.0% 75.3% 81.0% 78.3% Total 100.0% 100.0% 100.0% 100.0% Note: The sample consists of the four-month CPS panels, restricted to the panels with Unemployment in month three, sample mean of annual averages. All calculations use the CPS sampling weights. 43

46 Table 7.2: Employment Transition Rates and Shares of the Unemployed by Reported Unemployment Duration and Previous Month Labor Force Status, Averages of Monthly Series Labor Force Status in month t-1 < 5 weeks Reported unemployment duration in t Al 5+ weeks All unemployed E % 2.0% 1.0% 0.9% 0.7% 0.3% 5.0% 22.5% U % 10.5% 11.4% 11.6% 8.5% 7.4% 49.3% 54.6% OLF % 3.3% 2.3% 2.5% 2.4% 1.5% 12.0% 23.0% All unemployed % 15.7% 14.7% 15.0% 11.6% 9.3% 66.3% 100% E % 1.6% 1.0% 0.9% 0.8% 0.4% 4.7% 22.2% U % 9.8% 12.2% 11.7% 9.0% 9.7% 52.3% 54.1% OLF % 3.2% 2.5% 3.2% 3.2% 2.1% 14.2% 23.7% All unemployed % 14.6% 15.6% 15.8% 13.0% 12.2% 71.2% 100% Note: Table shows the employment transition rates of the unemployed in month three of our four-month CPS panels, sample mean of annual averages of monthly series. Table also shows the shares (in %) of the respective groups in the entire pool of the unemployed in month three of our four-month CPS panels, the sample mean of annual averages of monthly series. All calculations use the CPS sampling weights. 44

47 Table 7.3: Employment Transition Rates by Reported Unemployment Duration and Previous Month Labor Force Status Reported duration and (1) (2) previous month LFS < 5, employed in t-1 omitted omitted 5-8, employed in t *** ( ) ( ) 9-14, employed in t ( ) ( ) 15-26, employed in t *** *** ( ) ( ) 27-52, employed in t *** *** ( ) (0.0102) 53+, employed in t *** *** (0.0110) (0.0137) < 5, unemployed in t *** *** ( ) ( ) 5-8, unemployed in t *** *** ( ) ( ) 9-14, unemployed in t *** *** ( ) ( ) 15-26, unemployed in t *** *** ( ) ( ) 26-52, unemployed in t *** *** ( ) ( ) 53+, unemployed in t *** *** ( ) ( ) < 5, OLF in t *** *** ( ) ( ) 5-8, OLF in t *** *** ( ) ( ) 9-14, OLF in t *** *** ( ) ( ) 15-26, OLF in t *** *** ( ) ( ) 26-52, OLF in t *** *** ( ) ( ) 53+, OLF in t *** *** ( ) ( ) Age, gender, educ, year, yes yes seasonals, cons Observations 362, ,738 Adjusted R-squared F Note: The table shows the estimates of a linear probability model of unemployment-to-employment transition rate. Additional controls are age, gender, education, seasonal and year dummies. The omitted category is year old female with high school education. The estimation uses the CPS sampling weights. 45

48 Table 7.4: Reported Duration of Newly Unemployed, by Duration of Previous Employment Previous Labor Force Status Employed in t-1, t-2, t-3 Employed in t-1, t-2, non-employed in t-3 Employed in t-1, non-employed in t-2 Employed in t-1, t-2, t-3 Employed in t-1, t-2, non-employed in t-3 Employed in t-1, non-employed in t-2 Share in E to U transitions Distribution of reported unemployment duration in t < Mean duration, 5+ Median duration, 5+ Share in E to U transitions reporting 5+ weeks % 85.5% 12.8% 0.8% 0.8% % 11.2% 77.4% 17.7% 2.7% 2.2% % 14.3% 59.9% 28.7% 6.3% 5.0% % % 86.7% 11.4% 0.9% 1.0% % 10.5% 78.2% 15.8% 3.1% 2.8% % 13.2% 61.6% 26.2% 6.3% 5.9% % Note: The table shows the incidence of the reported durations of the new employment to unemployment transitions in month four of the four-month CPS panels, by the length of previous employment. The tables also shows the mean and median of the reported longer durations (5+ weeks) of the new employment to unemployment transitions in month four of the four-month CPS panels, by the length of previous employment. We use the CPS sampling weights. The shaded areas show the NBER recessions. 46

49 Table 7.5: Employment Transition Rates and Shares of the Unemployed by Reported Unemployment Duration and Duration of Previous Employment Previous Labor Force Status Reported unemployment duration in t < 5 weeks 5+ weeks All Employed in t-1 and t % 13.5% 78.3% Employed in t-1, non-employed in t % 9.3% 21.7% All employed in t % 22.8% 100.0% Employed in t-1 and t % 13.0% 79.0% Employed in t-1, non-employed in t % 8.9% 21.0% All employed in t % 21.9% 100.0% Note: The table shows the employment transition rates and shares (in %) for the unemployed in month three of our four-month CPS panels who were employed in the previous month. We use the CPS sampling weights. 47

50 Table 7.6: Unemployment Duration Distribution, from Reported Durations versus from LFS Histories Duration of unemployment <5 weeks 5-8 weeks 9-14 weeks 15+ weeks Total Reported duration of unemployment 34.0% 15.3% 13.1% 37.7% 100.0% Reported, corrected for Employment in LFS 36.8% 16.6% 13.3% 33.3% 100.0% Duration of unemployment in LFS histories 46.2% 18.8% 10.8% 24.2% 100.0% Reported duration of unemployment 29.1% 14.3% 13.6% 43.0% 100.0% Reported, corrected for Employment in LFS 32.5% 15.4% 13.3% 38.8% 100.0% Duration of unemployment in LFS histories 46.1% 18.8% 10.9% 24.1% 100.0% Note: The table shows the duration distribution for the unemployed in month four in the CPS four-month panels. The reported duration of unemployment is based on survey responses to the question of unemployment duration in the CPS. The reported duration of unemployment, corrected for employment in the LFS histories corrects for employment observed within the four month panel. The duration of unemployment in the LFS histories is based on observed histories of unemployment in the four month panel and reported durations if the observed unemployment spell exceeds the panel. Statistics are based on weighted data using the CPS sampling weights. 48

51 Figure 3.1: LFS Histories of the Non-Employed: Employment Transition Rates and Population Shares, Note: The figure shows average employment transition rates (black dots for U-histories and red dots for N- histories, left axis) and pop. shares (gray bars, right axis) by labor force status history of the non-employed, The pop. share of NNN is 31.1.% (not shown). The LFS histories on the x-axis are ranked by the employment transition rates. 49

52 Figure 3.2: Employment Transition Rates by Labor Force Status History, with and without Demographic Controls Note: The figure shows estimates from two regressions, The regression without demographic controls includes annual dummies and seasonal dummies. The regression with demographic controls in addition includes dummies for age, gender and education. The estimation employs the CPS sampling weights. 50

53 Figure 4.1: Unemployment-Nonparticipation Cyclers versus Continuously Unemployed or OLF Panel A. NUN and NNN Employment Transition Rates Population Shares Panel B. UNU and UUU Employment Transition Rates Population Shares Note: The figure shows annual averages of monthly series. Population shares are the shares in the civilian noninstitutionalized population 16 years old and older. The vertical line shows the 1994 CPS redesign year. We use the CPS sampling weights. 51

54 Figure 6.1: Employment Transition Rates of the OLF by Self-Reported Labor Market Attachment, All OLF and by LFS History Note: The figure shows the coefficient estimates and 95% confidence intervals from two linear probability models of the individual OLF-to-employment transition rate. The first model (black line) includes seven dummies representing self-reported labor market attachment: (1) WJ,MA Disc : want job, marginally attached, discouraged, (2) WJ,MA Other : want job, marginally attached, other, (3) WJ,Other : want job, other, (4) DNWJ,ONS : do not want job, not retired, not disabled, not in school, (5) DNWJ,OS :do not want job, in school, y. o., (6) DNWJ,Disabled :do not want job, disabled, (7) DNWJ,Retired : do not want job, retired. The second model includes a full set of interactions of the seven self-reported labor market attachment dummies with nine LFS histories of the OLF (colored lines). The regressions are estimated without a constant, additional controls are age, gender, education, and seasonals (the omitted category is year old female with high school education). The estimates are similar with or without annual controls. The estimation employs the CPS sampling weights. 52

55 Figure 7.1: Inconsistency in Reported Unemployment Durations and Observed LFS Panel A. Share of Unemployed Reporting Durations Longer Than 5 Weeks, Unemployed Who Were Employed in the Previous Month Panel B. Share of Unemployed Reporting Durations Longer Than 5 Weeks, Unemployed Who Were OLF in the Previous Month Note: The sample consists of the four-month CPS panels, restricted to the panels with Unemployment in month three. We use the CPS sampling weights. 53

56 Figure 7.2: Unemployment-to-Employment Transition Rate, by Reported Duration and Previous Month Labor Force Status Panel A Panel B Note: The figure shows the coefficient estimates from two linear probability models of the individual unemployment-to-employment transition rate: (1) the first model includes six dummies represented reported duration (gray line), (2) the second model includes dummies representing interactions of reported duration with the previous month LFS. The regressions are estimated without a constant, additional controls are age, gender, education, seasonal and year dummies (the omitted category is year old female with high school education). The coefficient estimates and the 95% CI from the regression in Table

57 Figure 7.3: Incidence of Reported Durations of 5+ Weeks among New Transitions from Employment to Unemployment, by Length of Previous Employment Note: The figure shows the incidence of reported long durations of the new employment to unemployment transitions in month four of the four-month CPS panels, by the length of previous employment. We use the CPS sampling weights. The shaded areas show the NBER recessions. 55

58 Figure 7.4: Unemployment Duration Distribution from Reported Data and from Actual LFS Histories Note: The figure shows the duration distribution for the unemployed in month four of the four-month CPS panels. The distribution of reported duration is constricted directly from the reports (black lines). The distribution from the observed LFS (red lines) is constructed as follows: less than 5 weeks, when the individual is unemployed in month four but employed or OLF in month three; 5-8 weeks, when the individual is unemployed in months four and three but employed or OLF in month two; 9-14 weeks, when the individual is unemployed in months four, three and two but employed or OLF in month one; and, finally, 15 weeks or longer, when the individual is unemployed in all four consecutive months. The distribution from the LFS histories with the correction for previous employment (blue lines) is constructed as follows: for the unemployed with employment in month three we assign duration of less than 5 weeks; for the unemployed with unemployment in month three and employment in month two, we assign duration 5-8 weeks; for the unemployed with unemployment in months three and two and employment in month one we assign duration of 9-15 weeks; and for the rest of the unemployed, we keep their reported durations. We use the CPS sampling weights. The shaded areas show the NBER recessions. All four panels share the common y-axis. 56

59 Figure 7.5: Distribution of reported Durations of Newly Unemployed from Employment and from OLF Note: The figure shows the duration distribution for the newly unemployed in month four of the four-month CPS panels, i.e., those who were employed or OLF in month three. The shaded areas show the NBER recessions. All four panels share the common y-axis. 57

60 APPENDIX Table A3.1: Employment Transition Rates and Population Shares of Retired, Disabled and Others, OLF for Three Consecutive Months Employment transition rate Population share LFS History NNN, retired in three consecutive months % NNN, disabled in three consecutive months % The rest of NNN histories % Note: The table shows the sample mean of the annual averages of monthly (discrete) rates and population shares. The sample consists of the four-month CPS panels, restricted to the panels with OLF in months 1, 2 and 3, N in the histories denote OLF status. All calculations use the CPS sampling weights as described in the text. 58

61 Table A5.1: Employment Transition Rate of the Unemployed and OLF: Labor Force Status History versus Current-Month Information Alternative Classifications of the Non-Employed (1) (2) (3) (4) (5) (6) Current LFS (2 gr) yes yes yes yes yes yes Duration of unemployment (3gr) x yes yes yes yes yes Reasons for unemployment (6 gr) x yes yes yes yes yes LM attachment of the OLF (7 gr) x x yes yes yes x LFS histories that end in U (9 gr) x x x yes yes yes LFS histories that end in OLF (9 gr) x x x x yes yes Age, education, gender, year, seasonals, constant yes yes yes yes yes yes Observations 2,188,466 2,188,466 2,188,466 2,188,466 2,188,466 2,188,466 Adjusted R-squared F stat Note: The sample consists of the four-month CPS panels, restricted to the panels with OLF in month three, See the notes to Tables 4.1 and 4.2. The estimation employs the CPS sampling weights. 59

62 Table A5.2: Employment Transition Rate of the Non-Employed: Detailed Labor Force Status History versus Past LFS Status Unemployed + OLF Past LFS LFS Histories 1 2 LFS Histories (18 gr) x yes LFS in t U *** x (0.0007) OLF omitted x LFS in t-1 E 0.212*** x (0.0008) U *** x (0.0008) OLF omitted x LFS in t-2 E 0.142*** x (0.0007) U *** x (0.0007) OLF omitted x Age, gender, educ, year, yes yes seasonals, cons Observations 2,188,466 2,188,466 Adjusted R-squared F stat LR chi2(12) Model1 vs Model2 = Note: The sample consists of the four-month CPS panels, restricted to the panels with OLF in month three, The estimation employs the CPS sampling weights. 60

63 Figure A3.1: Population Shares of the Non-Employed, by Labor Force Status History, Annual Averages of Monthly Shares Panel A. LFS Histories with Recent Employment Panel B. LFS Histories with No Recent Employment, Except NNN Note: The figure shows shares in the civilian noninstitutionalized population 16 years and older. The sample consists of the four-month CPS panels, restricted to the panels with non-employment (unemployment or OLF) in month three. We use the CPS sampling weights. The vertical line shows the 1994 CPS redesign year. 61

64 Figure A3.2: Employment Transition Rates of the Non-Employed, by Labor Force Status History, Annual Averages of Monthly Rates Panel A. LFS Histories with Recent Employment (Selected histories) Panel B. LFS Histories with No Recent Employment Note: The sample consists of the four-month CPS panels, restricted to the panels with non-employment (unemployment or OLF) in month three. E in the histories denotes employment, U unemployment, and N OLF. All calculations use the CPS sampling weights as described in the text. The vertical line shows the 1994 CPS redesign year. 62

65 Figure A3.3: Employment Transition Rates of the Unemployed and OLF, by the LFS in the Last Two Months Note: The figure shows the coefficient estimates and the 95% confidence intervals from the regression in Table 3.3. Figure A3.4: Employment Transition Rates, with and without Unemployed of Temporary Layoffs Note: The figure shows the sample mean of annual averages of monthly (discrete) rates. The sample consists of the four-month CPS panels, restricted to the panels with non-employment (unemployment or OLF) in month three, E in the histories denotes employment, U unemployment, N OLF, and O denotes the unemployed for reasons other than temporary layoffs. All calculations use the CPS sampling weights as described in the text. 63

66 Figure A3.5: Employment Transition Rates and Population Shares of Retired, Disabled and Others OLF for Three Consecutive Months A. Employment Transition rates B. Population shares Note: The figure shows the sample mean of the annual averages of monthly (discrete) rates and population shares. The sample consists of the four-month CPS panels, restricted to the panels with non-employment (unemployment or OLF) in month three, N in the histories denote OLF status. All calculations use the CPS sampling weights as described in the text. 64

67 Figure A6.1: Share of New Transitions into Unemployment, by Reported Unemployment Duration and Previous Month Labor Force Status Notes: See Note to Table

68 Figure A6.2: Mean and Median of Reported Durations among New Transitions from Employment to Unemployment with Reported Durations 5+ Weeks, by Length of Previous Employment Panel A. Mean Panel B. Median Note: The figure shows the mean and median of the reported longer durations (5+ weeks) of the new employment to unemployment transitions in month four of the four-month CPS panels, by the length of previous employment. We use the CPS sampling weights. The shaded areas show the NBER recessions. 66