Figure 2.1 The Longitudinal Employer-Household Dynamics Program

Transcription

1 Figure 2.1 The Longitudinal Employer-Household Dynamics Program Demographic Surveys Household Record Household-ID Data Integration Record Person-ID Employer-ID Data Economic Censuses and Surveys Census Entity-ID Data Person Record Household-ID Person-ID Data Business Register Employer-ID Census Entity-ID Data Source: Authors illustration.

2 Figure 2.2 Mean Annual Earnings by Firm- or Person-Effects Quintile, 1992 to 2003 Mean Annual Earnings (2003 Dollars) $80,000 $70,000 $60,000 $50,000 $40,000 $30,000 $20,000 $10, $0 Firm Effects Person Effects Quintile (1 = Highest) Source: Authors tabulations using microdata from U.S. Census Bureau ( ; 12 states).

3 Figure 2.3 Cumulative Net Employment Growth by Firm-Effects Quintile and by Year, 1992 to 2003 Cumulative Net Employment Growth from 1992 (in Millions) Firm-Effects Quintile (1 = Highest) Note: Each bar represents cumulative employment growth within a quintile for a year in the period of 1993 to 2003, relative to 1992.

4 Figure 2.4 Employment over Time Within Person-Effects Quintile 1, 1992 to 2003 Quintile Employment As Percentage of Total Employment Firm-Effects Quintile (1 = Highest)

5 Figure 2.5 Employment over Time Within Person-Effects Quintile 2, 1992 to 2003 Quintile Employment As Percentage of Total Employment Firm-Effects Quintile (1 = Highest) Figure 2.6 Employment over Time Within Person-Effects Quintile 3, 1992 to 2003 Quintile Employment As Percentage of Total Employment Firm-Effects Quintile (1 = Highest)

6 Figure 2.7 Employment over Time Within Person-Effects Quintile 4, 1992 to 2003 Quintile Employment As Percentage of Total Employment Firm-Effects Quintile (1 = Highest) Figure 2.8 Employment over Time Within Person-Effects Quintile 5, 1992 to Quintile Employment As Percentage of Total Employment Firm-Effects Quintile (1 = Highest)

7 Figure 2.9 Average Earnings Differential by Firm-Effects Quintile, 1992 Versus 2003 Average Firm Ratio of Ninetieth Percentile Earnings to Tenth Percentile Earnings Firm-Effects Quintile (1 = Highest)

8 Table 2.1 Distribution of Employment and Turnover Across Firm-Effects Quintiles, 1992 to 2003 Firm-Effects Quintile (1 = Highest) Industry Agriculture 2.4% 6.5% 10.2% 14.5% 66.4% Mining Utilities Construction Nondurable manufacturing Durable manufacturing Wholesale trade Retail trade Transportation Services Information Finance Real estate Professional Management Administrative Education Health care Entertainment Accommodation and food Other services Public administration Firm size (number of employees) 25 to to to ,000 to 9, ,000 or more Average annual turnover and churning Turnover 22.3% 21.5% 23.9% 29.4% 36.6% Churning Note: For employment distribution, rows sum to 100 percent. Annual turnover of a firm is defined as the number of job accessions and job separations between two consecutive years divided by the total firm employment in those two years. Annual churning is defined as the number of job accessions and job separations minus the absolute change in firm employment divided by the total firm employment. Turnover and churning are calculated for jobs with full-quarter employment only.

9 Table 2.2 Distribution of Employment Across Person-Effects Quintiles, 1992 to 2003 Person-Effects Quintile (1 = Highest) Age Eighteen to thirty-four 19.0% 18.4% 20.4% 20.9% 21.3% Thirty-five to forty-four Forty-five to sixty-four Race White Black Hispanic Asian Pacific Islander Native American Gender Female Male Foreign-born No Yes Educational attainment Less than high school High school Some college Associate s degree Bachelor s degree Master s degree Professional degree Doctoral degree Note: Rows sum to 100 percent. Data on education come from the long form of the 2000 Decennial Census of Population, which includes approximately one-sixth of the workers in our data. Education results are calculated solely for these workers and solely for the year 2000.

10 Table 2.3 Distribution of Employment Across Firm-Effects Quintiles, 1992 to 2003 Firm-Effects Quintile (1 = Highest) Person-effects quintile Quintile % 24.2% 7.0% 1.7% 0.4% Quintile Quintile Quintile Quintile Age Eighteen to thirty-four Thirty-five to forty-four Forty-five to sixty-four Race White Black Hispanic Asian Pacific Islander Native American Gender Female Male Foreign-born No Yes Educational attainment Less than high school High school Some college Associate s degree Bachelor s degree Master s degree Professional degree Doctoral degree Note: Rows sum to 100 percent. Data on education come from the long form of the 2000 Decennial Census of Population, which includes approximately one-sixth of the workers in our data. Education results are calculated solely for these workers and solely for the year 2000.

11 Table 2.4 Distribution of Employment Within Firm-Effects Quintiles, 1992 Versus Firm-Effects Quintile (1 = Highest) Firm-Effects Quintile (1 = Highest) Industry Agriculture 0.2% 0.6% 0.8% 1.2% 6.0% 0.2% 0.5% 0.7% 1.1% 5.2% Mining Utilities Construction Nondurable manufacturing Durable manufacturing Wholesale trade Retail trade Transportation Services Information Finance Real estate Professional services Management Administrative Education Health care Entertainment Accommodation and food Other Public administration Note: Columns sum to 100 percent.

12 Table 2.5 Distribution of Employment Within Person-Effects Quintiles, 1992 Versus Person-Effects Quintile (1 = Highest) Person-Effects Quintile (1 = Highest) Industry Agriculture 0.2% 0.2% 0.6% 1.5% 6.2% 0.2% 0.4% 0.6% 1.0% 5.4% Mining Utilities Construction Nondurable manufacturing Durable manufacturing Wholesale trade Retail trade Transportation Services Information Finance Real estate Professional services Management Administrative Education Health care Entertainment Accommodation and food Other Public administration Note: Columns sum to 100 percent.

13 Table 2.6 Distribution of Employment by Industry Within Person-Effects Quintiles at Firms Where Firm Fixed- Effects Quintile Is Better Than Person Fixed-Effects Quintile Person-Effects Quintile (2 = Highest) Person-Effects Quintile (2 = Highest) Industry Total Total Agriculture 0.2% 0.5% 1.2% 2.8% 1.3% 0.2% 0.3% 0.9% 2.0% 0.9% Mining Utilities Construction Nondurable manufacturing Durable manufacturing Wholesale trade Retail trade Transportation Services Information Finance Real estate Professional services Management Administrative Education Health care Entertainment Accommodation and food Other Public administration Note: Columns sum to 100 percent.

14 Table 2.7 Mean Firm Fixed Effects and Person Fixed Effects, by Industry Firm Fixed Effects Person Fixed Effects Industry Agriculture Mining Utilities Construction Nondurable manufacturing Durable manufacturing Wholesale trade Retail trade Transportation Services Information Finance Real estate Professional services Management Administrative Education Health care Entertainment Accommodation and food Other Public administration

15 Table 2.8 Distribution of Employment Across Firm-Effects Quintiles, 1992 Versus Firm-Effects Quintile (1 = Highest) Firm-Effects Quintile (1 = Highest) Person-Effects Quintile Quintile % 26.3% 8.0% 1.9% 0.3% 67.7% 22.4% 7.6% 1.8% 0.6% Quintile Quintile Quintile Quintile Note: Rows sum to 100 percent.

16 Figure 3.1 Rates of Job Change and Job Displacement, 1992 to % Average Annual Rate Job Change Displacement Job Change Displacement Any Tenure Tenure at Least Three Years

17 Figure 3.2 Rates of Job Change and Job Displacement over Time, 1992 to % 30 Annual Rate Job Change Displacement Job Change Displacement Any Tenure Tenure at Least Three Years Note: Bars represent the rate of job change or displacement for each year from 1992 to 2002.

18 Figure 3.3 Rates of Job Displacement over Time by Firm-Effects Quintile, 1992 to % Annual Rate of Job Displacement Any Tenure Tenure at Least Three Years Firm-Effects Quintiles (1 = Highest) Note: Bars represent the rate of job displacement for each year from 1992 to Figure 3.4 Rates of Job Displacement over Time by Person-Effects Quintile, 1992 to % Annual Rate of Job Displacement Any Tenure Tenure at Least Three Years Person-Effects Quintiles (1 = Highest) Source: Author s tabulations using microdata from U.S. Bureau of the Census ( ). Note: Bars represent the rate of job displacement for each year from 1992 to 2002.

19 Table 3.1 Rates of Job Change and Job Displacement, 1992 to 2003 Industry Job Change Any Tenure Tenure at Least Three Years Job Change Displacement Displacement Agriculture 38.3% 12.5% 18.5% 4.3% Mining Utilities Construction Nondurable manufacturing Durable manufacturing Wholesale trade Retail trade Transportation Services Information Finance Real estate Professional services Management Administrative Education Health care Entertainment Accommodation and food Other services Public administration Firm-effects quintile Quintile Quintile Quintile Quintile Quintile Firm size 25 to to to ,000 to 9, ,000 or more Note: Job change is defined as a change in a worker s primary employer from year t to year t + 1 (including non-employment in year t + 1). Displacement is defined as a job change where a displacement event (a decline of 30 percent or more in quarterly employment) occurred at the firm that is the primary employer in year t, and additionally the worker had positive earnings at the firm in the quarter of the displacement event, had positive earnings at the firm for at least one quarter prior to the event, and had zero earnings at the firm for at least two quarters after the event.

20 Table 3.2 Rates of Job Change and Job Displacement, 1992 to 2003 Tenure at Least Any Tenure Three Years Job Change Job Change Displacement Displacement Person-effects quintile Quintile % 1.7% 16.9% 1.3% Quintile Quintile Quintile Quintile Age Eighteen to thirty-four Thirty-five to forty-four Forty-five to sixty-four Race White Black Hispanic Asian Pacific Islander Native American Gender Female Male Educational attainment Less than high school High school Some college Associate s degree Bachelor s degree Master s degree Professional degree Doctoral degree Note: Data on education come from the long form of the 2000 Decennial Census of Population, which includes approximately one-sixth of the workers in our data. Education results are calculated solely for these workers and solely for the year 2000.

21 Table 3.3 Distribution of Displaced Workers Across New Firm-Effects Quintiles, 1992 to 2003 and 1995 to 2003 Any Tenure, 1992 to 2003 Tenure at Least Three Years, 1992 to Total Total Original firm-effects quintile changed to a firm in... Quintile % 23.4% 20.5% 17.2% 13.5% 22.5% N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Any Tenure, 1995 to 2003 Tenure at Least Three Years, 1995 to Total Total Original firm-effects quintile changed to a firm in... Quintile % 23.8% 18.2% 16.1% 10.6% 38.2% Quintile Quintile Quintile Quintile Note: Columns sum to 100 percent. The sample is restricted to workers employed at the beginning and end of each period. Displacements occurred in 1992 or 1995, while final job is observed in The table presents the distribution of workers across firm-effects quintiles in 2003 for each quintile in which workers started in 1992 or 1995.

22 Table 3.4 Earnings Growth of Workers by Displacement, Firm-Effects Quintile, and Tenure, 1992 to 2003 and 1995 to 2003 Any Tenure, 1992 to 2003 Firm-Effects Quintile in 1992 (1 = Highest) Tenure at Least Three Years, 1992 to 2003 Firm-Effects Quintile in 1992 (1 = Highest) Total Total Workers who remained in the same firm 36.2% 28.7% 23.1% 29.3% 29.4% 29.7% N/A N/A N/A N/A N/A N/A Workers who were displaced to a firm... In a better firm-effects quintile N/A N/A N/A N/A N/A N/A N/A In the same firm-effects quintile N/A N/A N/A N/A N/A N/A In a worse firm-effects quintile N/A 14.4 N/A N/A N/A N/A N/A N/A Total N/A N/A N/A N/A N/A N/A Any Tenure, 1995 to 2003 Firm-Effects Quintile in 1995 (1 = Highest) Tenure at Least Three Years, 1995 to 2003 Firm-Effects Quintile in 1995 (1 = Highest) Total Total Workers who remained in the same firm % 19.7% 16.7% 19.3% 20.1% 20.3% Workers who were displaced to a firm... In a better firm-effects quintile N/A N/A In the same firm-effects quintile In a worse firm-effects quintile N/A N/A 8.7 Total Note: Columns sum to 100 percent. The sample is restricted to workers employed at the beginning and end of each period. Displacements occurred in 1992 or 1995, while final job is observed in The table presents the distribution of workers across firm-effects quintiles in 2003 for each quintile in which workers started in 1992 or 1995.

23 Table 3.5 Distribution of Displaced Workers Across New Firm-Effects Quintiles, Three Subperiods Any Tenure, 1992 to 1995 Tenure at Least Three Years, 1992 to Total Total Original firm-effects quintile changed to a firm in... Quintile % 22.2% 19.0% 11.9% 8.5% 20.6% N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Any Tenure, 1996 to 1999 Tenure at Least Three Years, 1996 to Total Total Original firm-effects quintile changed to a firm in... Quintile % 27.2% 11.6% 10.0% 7.7% 22.8% Quintile Quintile

24 Quintile Quintile Original firm-effects quintile changed to a firm in... Any Tenure, 2000 to 2003 Tenure at Least Three Years, 2000 to Total Total Quintile Quintile Quintile Quintile Quintile Note: Columns sum to 100 percent. The sample is restricted to workers employed at the beginning and end of each period. Displacements occurred in the first year of each subperiod, while the final job is observed in the last year. The table presents the distribution of workers across firm-effects quintiles in the last year for each quintile in which the workers started in the first year.

25 Table 3.6 Earnings Growth of Workers by Job Change, Firm-Effects Quintile, and Tenure, Three Subperiods Any Tenure Firm-Effects Quintile in 1992 (1 = Highest) Tenure at Least Three Years Firm-Effects Quintile in 1992 (1 = Highest) Total Total 1992 to 1995 Workers who remained in the same firm 9.2% 5.2% 4.8% 6.2% 5.6% 6.6% N/A N/A N/A N/A N/A N/A Workers who were displaced to a firm... In a better firm-effects quintile N/A N/A N/A N/A N/A N/A In the same firm-effects quintile N/A N/A N/A N/A N/A N/A In a worse firm-effects quintile N/A 8.0 N/A N/A N/A N/A N/A N/A Total N/A N/A N/A N/A N/A N/A Any Tenure Firm-Effects Quintile in 1996 (1 = Highest) Tenure at Least Three Years Firm-Effects Quintile in 1996 (1 = Highest) Total Total 1996 to 1999 Workers who remained in the same firm % 10.9% 9.2% 11.5% 10.0% 13.6% Workers who were displaced to a firm... In a better firm-effects quintile N/A N/A

26 In the same firm-effects quintile In a worse firm-effects quintile N/A N/A 0.4 Total Any Tenure Firm-Effects Quintile in 2000 (1 = Highest) Tenure at Least 3 Years Firm-Effects Quintile in 2000 (1 = Highest) Total Total 2000 to 2003 Workers who remained in the same firm Workers who were displaced to a firm... In a better firm-effects quintile N/A N/A In the same firm-effects quintile In a worse firm-effects quintile N/A N/A 14.1 Total

27 Appendix 3A.1 Distribution of Workers Who Changed Firms Across New Firm-Effects Quintiles, 1992 to 2003 and 1995 to 2003 Any Tenure, 1992 to 2003 Tenure at Least Three Years, 1992 to Total Total Original firm-effects quintile changed to a firm in... Quintile % 25.5% 18.6% 17.7% 14.9% 24.0% N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Any Tenure, 1995 to 2003 Tenure at Least Three Years, 1995 to Total Total Original firm-effects quintile changed to a firm in... Quintile % 26.5% 16.1% 15.0% 11.3% 28.8% Quintile Quintile Quintile Quintile Note: Columns sum to 100 percent. The sample is restricted to workers employed at the beginning and end of each period. Job changes occurred in 1992 or 1995, while the final job is observed in The table presents the distribution of workers across firm-effects quintiles in 2003 for each quintile in which workers started in 1992 or 1995.

28 Appendix 3A.2 Earnings Growth by Job Change, Firm-Effects Quintile, and Tenure, 1992 to 2003 and 1995 to 2003 Any Tenure, 1992 to 2003 Firm-Effects Quintile in 1992 (1 = Highest) Tenure at Least Three Years, 1992 to 2003 Firm-Effects Quintile in 1992 (1 = Highest) Total Total Workers who remained in the same firm 36.2% 28.7% 23.1% 29.3% 29.4% 29.7% N/A N/A N/A N/A N/A N/A Workers who changed to a firm... In a better firm-effects quintile N/A N/A N/A N/A N/A N/A N/A In the same firm-effects quintile N/A N/A N/A N/A N/A N/A In a worse firm-effects quintile N/A 26.0 N/A N/A N/A N/A N/A N/A Total N/A N/A N/A N/A N/A N/A Any Tenure, 1995 to 2003 Firm-Effects Quintile in 1995 (1 = Highest) Tenure at Least Three Years, 1995 to 2003 Firm-Effects Quintile in 1995 (1 = Highest) Total Total Workers who remained in the same firm % 19.7% 16.7% 19.3% 20.1% 20.3% Workers who changed to a firm... In a better firm-effects quintile N/A N/A In the same firm-effects quintile In a worse firm-effects quintile N/A N/A 2.9 Total Note: Job changes occurred in 1992 or Tenure is measured as of the beginning of the period. Better/same/worse quintile represents a comparison of the quintile of the worker s primary employer in the end year to the quintile of the worker s primary employer in the start year.

29 Appendix 3A.3 Distribution of Workers Who Changed Firms Across New Firm-Effects Quintiles, Three Subperiods Any Tenure, 1992 to 1995 Tenure at Least Three Years, 1992 to Total Total All workers who changed firms Original firm-effects quintile changed to a firm in... Quintile % 25.1% 14.7% 12.9% 9.8% 22.5% N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Quintile N/A N/A N/A N/A N/A N/A Any Tenure, 1996 to 1999 Tenure at Least Three Years, 1996 to Total Total All workers who changed firms Original firm-effects quintile changed to a firm in... Quintile % 37.6% 14.3% 11.7% 9.3% 30.2% Quintile (Table continues on p. 88.)

30 Appendix 3A.3 (Continued) Any Tenure, 1992 to 1995 Tenure at Least Three Years, 1992 to Total Total Quintile Quintile Quintile Any Tenure, 2000 to 2003 Tenure at Least Three Years, 2000 to Total Total All workers who changed firms Original firm-effects quintile changed to a firm in... Quintile Quintile Quintile Quintile Quintile Note: Columns sum to 100 percent. The sample is restricted to workers employed at the beginning and end of each period. Job changes occurred in the first year of each subperiod, while the final job is observed in the last year. The table presents the distribution of workers across firm effects quintiles in the last year for each quintile in which the workers started in the first year.

31 Appendix 3A.4 Earnings Growth by Job Change, Firm-Effects Quintile, and Tenure, Three Subperiods Any Tenure Firm-Effects Quintile in 1992 (1 = Highest) Tenure at Least Three Years Firm-Effects Quintile in 1992 (1 = Highest) Total Total 1992 to 1995 Workers who remained in the same firm 9.2% 5.2% 4.8% 6.2% 5.6% 6.6% N/A N/A N/A N/A N/A N/A Workers who changed to a firm... In a better firm-effects quintile N/A N/A N/A N/A N/A N/A N/A In the same firm-effects quintile N/A N/A N/A N/A N/A N/A In a worse firm-effects quintile N/A N/A N/A N/A N/A N/A Total N/A N/A N/A N/A N/A N/A Any Tenure Firm-Effects Quintile in 1996 (1 = Highest) Tenure at Least Three Years Firm-Effects Quintile in 1996 (1 = Highest) Total Total 1996 to 1999 Workers who remained in the same firm % 10.9% 9.2% 11.5% 10.0% 13.6% Workers who changed to a firm... In a better firm-effects quintile N/A N/A (Table continues on p. 90.)

32 Appendix 3A.4 (Continued) Any Tenure Firm-Effects Quintile in 1992 (1 = Highest) Tenure at Least Three Years Firm-Effects Quintile in 1992 (1 = Highest) Total Total In the same firm-effects quintile In a worse firm-effects quintile N/A N/A 2.9 Total Any Tenure Firm-Effects Quintile in 2000 (1 = Highest) Tenure at Least Three Years Firm-Effects Quintile in 2000 (1 = Highest) Total Total 2000 to 2003 Workers who remained in the same firm Workers who changed to a firm... In a better firm-effects quintile N/A N/A In the same firm-effects quintile In a worse firm-effects quintile N/A N/A 12.9 Total Note: Job change occurs at beginning of the period. Tenure is measured as of the beginning of the period. Better/same/worse quintile represents a comparison of the quintile of the worker s primary employer in the end year to the quintile of the worker s primary employer in the start year.

33 Appendix 3A.5 Regressions on Log Change in Earnings Model 1 Model 2 Model 3 Coeffi- Coeffi- Coefficient t-value cient t-value cient t-value 1992 to 2003, all workers Intercept Displaced Changed jobs, not displaced Age in Person effect Change in firm effect Change in tenure R-squared to 2003, all workers Intercept Displaced Changed jobs, not displaced Age in Person effect Change in firm effect Change in tenure R-squared to 2003, tenure at least three years Intercept Displaced Changed jobs, not displaced Age in Person effect Change in firm effect Change in tenure R-squared to 1995, all workers Intercept Displaced Changed jobs, not displaced Age in Person effect Change in firm effect Change in tenure R-squared to 1999, all workers Intercept Displaced (Table continues on p. 92.)

34 Appendix 3A.5 (Continued) Model 1 Model 2 Model 3 Coeffi- Coeffi- Coefficient t-value cient t-value cient t-value Changed jobs, not displaced Age in Person effect Change in firm effect Change in tenure R-squared to 1999, tenure at least three years Intercept Displaced Changed jobs, not displaced Age in Person effect Change in firm effect Change in tenure R-squared to 2003, all workers Intercept Displaced Changed jobs, not displaced Age in Person effect Change in firm effect Change in tenure R-squared to 2003, tenure at least three years Intercept Displaced Changed jobs, not displaced Age in Person effect Change in firm effect Change in tenure R-squared Note: Control variables used in all three models are the worker s age in years as of the beginning of 1992 and the worker s person fixed effect calculated over the period 1992 to Models 2 and 3 include an additional variable for the change in the firm fixed effect from the worker s employer in the start year to the employer in the end year. Model 3 includes an additional variable for change in the worker s tenure from the start year to the end year.

35 Figure 4.1 Earnings, Large Versus Small MSAs Annualized Earnings (2003 dollars) $50,000 $45,000 $40,000 $35,000 $30,000 $25,000 $20,000 $15,000 $10,000 $5,000 $0 Large Small Mean Earnings Median Earnings

36 Figure 4.2 Employment and Earnings Change, Large Versus Small MSAs 25% Percentage Change in Employment or Earnings Large Small to to to to to to 2003 Employment Mean Earnings Median Earnings

37 Figure 4.3 Employment Change by Firm-Effects Quintile, Large Versus Small MSAs Percentage Change in Employment 35% Large Small Large Small to to 2003

38 Figure 4.4 Employment Change by Person-Effects Quintile, Large Versus Small MSAs Percentage Change in Employment 50% Large Small Large Small to to 2003 Figure 4.5 Mean Earnings Change by Firm-Effects Quintile, Large Versus Small MSAs Percentage Change in Mean Earnings 20% Large Small Large Small to to 2003

39 Figure 4.6 Mean Earnings Change by Person-Effects Quintile, Large Versus Small MSAs Percentage Change in Mean Earnings 30% Large Small Large Small 1992 to to 2003 Figure 4.7 Median Earnings Change by Firm-Effects Quintile, Large Versus Small MSAs Percentage Change in Median Earnings 12% Large Small Large Small 1992 to to 2003

40 Figure 4.8 Median Earnings Change by Person-Effects Quintile, Large Versus Small MSAs Percentage Change in Median Earnings 20% Large Small Large Small 1992 to to

41 Figure 4.9 Rates of Job Displacement, 1992 to 2003, Large Versus Small MSAs 2.5% 2.0 Mean Annual Rate Small Large Small Large Small Large Small Large Manufacturing Manufacturing Nonmanufacturing Nonmanufacturing Any Tenure Tenure at Least Three Years Note: Job change is defined as a change in a worker s primary employer from year t to year t + 1 (including non-employment in year t + 1). Displacement is defined as a job change where a displacement event (a decline of 30 percent or more in quarterly employment) occurred at the firm that is the primary employer in year t, and additionally the worker had positive earnings at the firm in the quarter of the displacement event, had positive earnings at the firm for at least one quarter prior to the event, and had zero earnings at the firm for at least two quarters after the event. For workers with tenure of at least three years, the period is 1995 to 2003.

42 Figure 4.10 Change in Mean Earnings of Displaced Workers by MSA Group, Starting Industry, Period, and Change in Firm- Effects Quintile 80% 70 Change in Mean Earnings to to to to to to to to 2003 Small Large Small Large Manufacturing Nonmanufacturing Better Quintile Same Quintile Worse Quintile Total Note: Workers are those with at least three years of tenure as of 1996.

43 Table 4.1 Employment Shares by MSA Group and Industry Small MSAs Large MSAs Construction 3.01% 3.77% 3.71% 3.38% 4.36% 4.35% Nondurable manufacturing Durable manufacturing Wholesale trade Retail trade Transportation Information Finance Professional services Administrative Health care Entertainment Accommodation and food Other services Other industries Note: Other industries includes agriculture, mining, utilities, real estate and leasing, management of companies, and the remainder of education.

44 Table 4.2 Correlations Between Earnings and Employment Growth Across MSAs, 1992 to 2000: Correlations with Overall Employment Growth, Earnings Growth, and Job Creation/Destruction All MSAs Small MSAs Large MSAs Mean earnings growth and... Employment growth Median earnings growth Job creation Job destruction Median earnings growth and... Employment growth Mean earnings growth Job creation Job destruction

45 Table 4.3 Correlations Between Earnings and Employment Growth Across MSAs, 1992 to 2000: Correlations with Employment Growth by Firm-Effects Quintile All MSAs Small MSAs Large MSAs Mean earnings growth and... Firm-effects quintile 1 employment growth Firm-effects quintile 2 employment growth Firm-effects quintile 3 employment growth Firm-effects quintile 4 employment growth Firm-effects quintile 5 employment growth Median earnings growth and... Firm-effects quintile 1 employment growth Firm-effects quintile 2 employment growth Firm-effects quintile 3 employment growth Firm-effects quintile 4 employment growth Firm-effects quintile 5 employment growth

46 Table 4.4 Correlations Between Earnings and Employment Growth Across MSAs, 1992 to 2000: Correlations with Employment Growth by Industry All MSAs Small MSAs Large MSAs Mean earnings growth and... Agriculture employment growth Mining employment growth Utilities employment growth Construction employment growth Manufacturing employment growth Wholesale employment growth Retail employment growth Transportation employment growth Information employment growth Finance employment growth Real estate employment growth Professional employment growth Management employment growth Administrative employment growth Education employment growth Health care employment growth Entertainment employment growth Food and hotels employment growth Other services employment growth Median earnings growth and... Agriculture employment growth Mining employment growth Utilities employment growth Construction employment growth Manufacturing employment growth Wholesale employment growth Retail employment growth Transportation employment growth Information employment growth Finance employment growth Real estate employment growth Professional employment growth Management employment growth Administrative employment growth Education employment growth Health care employment growth Entertainment employment growth Food and hotels employment growth Other services employment growth

47 Table 4.5 Industry Distribution of 2003 Employment for Workers Displaced in 1996 Original Industry New Industry Small MSAs Other Large MSAs Manufacturing Manufacturing Other Construction Nondurable manufacturing Durable manufacturing Wholesale trade Retail trade Transportation Information Finance Professional services Administrative Health care Entertainment Accommodation and food Other services Other industries Note: Columns sum to 100 percent. Other industries includes agriculture, mining, utilities, real estate and leasing, management of companies, and the remainder of education. Workers are those with at least three years of tenure as of 1996.

48 Table 4.6 Fraction of Displaced Workers Who Moved into Jobs of Similar or Changing Quality Original Industry Manufacturing Non-Manufacturing 1996 to to to to 2003 Small MSA Large MSA Small MSA Large MSA Small MSA Large MSA Small MSA Large MSA Better 23.3% 14.3% 22.9% 15.2% 30.4% 30.0% 33.4% 33.1% Same Worse Note: Workers are those with at least three years of tenure as of Better/same/worse quintile represents a comparison of the quintile of the worker s primary employer in the end year to the quintile of the worker s primary employer in the start year.

49 Table 4A.1 percentage Change in Employment or Earnings by MSA and by Period 1992 to 1996 Small MSAs 1996 to to to 1996 Large MSAs 1996 to to 2003 Change in employment Total 10.8% 8.6% 7.0% 9.4% 10.5% 4.2% Firm-effects quintile Person-effects quintile Change in mean earnings Total Firm-effects quintile Person-effects quintile Change in median earnings Total Firm-effects quintile Person-effects quintile

50 Table 4A.2 Correlations Between Earnings and Employment Growth Across MSAs, 1992 to 2003: Correlations with Overall Employment Growth, Earnings Growth, and Job Creation and Destruction All MSAs Small MSAs Large MSAs Mean earnings growth and... Employment growth Median earnings growth Job creation Job destruction Median earnings growth and... Employment growth Mean earnings growth Job creation Job destruction Table 4A.3 Correlations Between Earnings and Employment Growth Across MSAs, 1992 to 2003: Correlations with Employment Growth by Firm-Effects Quintile All MSAs Small MSAs Large MSAs Mean earnings growth and... Firm-effects quintile 1 employment growth Firm-effects quintile 2 employment growth Firm-effects quintile 3 employment growth Firm-effects quintile 4 employment growth Firm-effects quintile 5 employment growth Median earnings growth and... Firm-effects quintile 1 employment growth Firm-effects quintile 2 employment growth Firm-effects quintile 3 employment growth Firm-effects quintile 4 employment growth Firm-effects quintile 5 employment growth

51 Table 4A.4 Correlations Between Earnings and Employment Growth Across MSAs, 1992 to 2003: Correlations with Employment Growth by Industry All MSAs Small MSAs Large MSAs Mean earnings growth and... Agriculture employment growth Mining employment growth Utilities employment growth Construction employment growth Manufacturing employment growth Wholesale employment growth Retail employment growth Transportation employment growth Information employment growth Finance employment growth Real estate employment growth Professional employment growth Management employment growth Administrative employment growth Education employment growth Health care employment growth Entertainment employment growth Food and hotels employment growth Other services employment growth Median earnings growth and... Agriculture employment growth Mining employment growth Utilities employment growth Construction employment growth Manufacturing employment growth Wholesale employment growth Retail employment growth Transportation employment growth Information employment growth Finance employment growth Real estate employment growth Professional employment growth Management employment growth Administrative employment growth Education employment growth Health care employment growth Entertainment employment growth Food and hotels employment growth Other services employment growth

52 Table 4A.5 Regressions on Log Change in Earnings for Workers with Tenure at Least Three Years Model 1 Model 2 Model 3 Coeffi- Coeffi- Coefficient t-value cient t-value cient t-value 1996 to 2000, small MSAs, nonmanufacturing workers Intercept Displaced Changed jobs, not displaced Age in 1992 (years) Person fixed effect MSA employment change (per 100 percent change) Displaced*employment change Job change*employment change R-squared to 2000, small MSAs, manufacturing workers Intercept Displaced Changed jobs, not displaced Age in 1992 (years) Person fixed effect MSA employment change (per 100 percent change) Displaced*employment change Job change*employment change R-squared to 2000, large MSAs, nonmanufacturing workers Intercept Displaced Changed jobs, not displaced Age in 1992 (years) Person fixed effect MSA employment change (per 100 percent change) Displaced*employment change Job change*employment change R-squared to 2000, large MSAs, manufacturing workers Intercept Displaced (Table continues on p. 126.)