Lethal Layoffs: Understanding the Mortality Effects of Job Loss - - PowerPoint PPT Presentation

lethal layoffs understanding the mortality effects of job
SMART_READER_LITE
LIVE PREVIEW

Lethal Layoffs: Understanding the Mortality Effects of Job Loss - - PowerPoint PPT Presentation

Lethal Layoffs: Understanding the Mortality Effects of Job Loss Christina Gathmann, Kristiina Huttunen and Robin Stitzing Aalto University May 17, 2013 Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the


slide-1
SLIDE 1

Lethal Layoffs: Understanding the Mortality Effects of Job Loss

Christina Gathmann, Kristiina Huttunen and Robin Stitzing

Aalto University

May 17, 2013

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-2
SLIDE 2

Motivation

Strong correlation between socioeconomic status and health Contradictory evidence from studies on macroeconomic conditions and micro-level studies

Recessions are good for your health (Ruhm, 2000) Job displacement increases mortality especially for prime-age male workers (Eliason and Storrie, 2007, Sullivan and von Wachter, 2009, Browning and Heinesen, 2012 ).

Martikainen, M¨ aki and J¨ antti (2007) find no mortality effects for working in downsizing establishments in Finland

Methods and time periods vary We still know little about the mechanism beyond job loss and mortality

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-3
SLIDE 3

This Study

Examine the effect of job displacement on cause-specific mortality in Finland during a period of large business cycle-fluctations (1988-2009)

Investigate how effect of job displacement vary with business cycle conditions.

1991-1993: Deepest economic contraction in an industrialized country since the 1930s (see Gorodnichenko et al. AER 2012)

Estimate the effect on various different outcomes: earnings, joint family income, employment, divorce Make distinction between own and spousal job loss Aim to produce comparable estimates to previous studies

Today’s talk focuses on recession period

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-4
SLIDE 4

Why Does Job Loss Affect Mortality?

Lower income (+)

Large literature has documented that displaced workers suffer long-lasting income reductions (e.g. Sullivan, Jaconson and Lalonde, 1994). Worse nutrition, limited access to health care

Lower employment (+/-)

More time to exercise, less work-induced stress, loss of self-esteem

Increased risk of marital dissolution (+)

Kerwin and Stephens (2004), Eliason (2012)

Change in health behaviour (+)

Recent study by Black, Devereux and Salvanes (2013) documents that job displacement increases smoking

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-5
SLIDE 5

Data

1/3 random sample of Finnish longitudinal employer-employee data for 1988-2009 All 16 - 74 year old Finnish residents Worker, plant, firm and spouse codes We can follow individuals in and out of labor force Merge with cause-of-death records

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-6
SLIDE 6

Plant Closure Sample

20-49 year old workers working in plants with at least 10 workers in base year t (1991-1993) Displaced worker: workers whose plant closed down between t and t+1 (and early-leavers) Non-displaced worker: workers whose plant did not closed down between t and t+1 Follow each worker (and his/her spouse) at least 3 years before base year t until 15th year after job loss

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-7
SLIDE 7

Mass Layoff Sample

20-49 year old workers working in plants with at least 50 workers in base year t (1991-1993) Displaced worker: workers who separated from plants that closed down or downsized more than 30 percent between t and t+1 (and early-leavers) Non-displaced worker: workers whose plant did not closed down or who did not separate from donsizing plants between t and t+1

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-8
SLIDE 8

Annual Income

10 20 30 40 Annual Income in 1000 euros

  • 3
  • 2
  • 1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time since Displacement Displaced Nondisplaced

Males

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-9
SLIDE 9

Annual Income

10 20 30 Annual Income in 1000 euros

  • 3
  • 2
  • 1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time since Displacement Displaced Nondisplaced

Females

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-10
SLIDE 10

Sample Means of Pre-Displacement Characteristics: Plant Closures

Males Females Variable Non- Displaced Displaced Non- Displaced Disp. Age 36.76 36.34 36.38 36.42 Education Compulsory .26 .26 .32 .30 Secondary .45 .40 .39 .33 Tertiary .29 .34 .28 .35 Annual Inc. t=0 32923 33979 23683 24312 Annual Inc. t=-3 30358 31549 21156 21853 Plant Size 521 137 527 115 Tenure 2.96 2.52 2.84 2.57 Has a spouse .77 .75 .74 .74 Outcomes Dead by t+2 .0017 .0042 .0007 .0009 Dead by t+15 .0249 .0254 .0127 .0143 Observations 295208 1648 200683 1047

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-11
SLIDE 11

Empirical Specification I

ln( pibt 1 − pibt ) = Xibβ + Dibtδ + τbt + ǫit (1) where pitb is the probability of worker i in base year sample b dying by the year t Dibt is a dummy for displacement prior to year t Xib is a vector of worker, spouse or region (predisplacement) characteristics τtb is base-year dummy

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-12
SLIDE 12

Empirical Specification II

Yibt = Xibβ +

15

  • j=−3

Dibt−jδj + τbt(+αib) + ǫibt (2) where Yibt is the outcome (e.g. earnings, family income, probability of dying, cumulative mortality) for individual i in base year sample b at time (since base year) t Dibt−j is a dummy for loosing a job j periods ago Xib is a vector of worker, spouse or region (predisplacement) characteristics τtb is baseyear-time dummies αib is baseyear-specific individual fixed-effect.

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-13
SLIDE 13

Effect of Job Displacement on Income

  • 6000
  • 4000
  • 2000

2000 Annual Income 1000 euros

  • 3
  • 2
  • 1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time Since Displacement 95 % Conf. Int. OLS Coef.

Males

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-14
SLIDE 14

Effect of Job Displacement on Income

  • 6000
  • 4000
  • 2000

2000 Annual Income 1000 euros

  • 3
  • 2
  • 1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time Since Displacement 95 % Conf. Int. OLS Coef.

Females

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-15
SLIDE 15

Effect of Job Displacement on Mortality: Logistic Model, High Tenure Workers

Dependent Variable: Dead by t+15

Male Female Variable

  • Spec. 1
  • Spec. 2

Spec1 Spec2. Mass Layoffs 0.169* 0.131 0.043 0.057 (0 .080) (0.081) (0.143) (0.117) Observations 132034 131260 83712 83388 Plant Closures 0.230* 0.228

  • 0.213
  • 0.279

(0 .136) (0.136) (0.295) (0.307) Observations 193159 191962 134383 133829

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-16
SLIDE 16

Effect of Job Displacement on Mortality: Logistic Model, Mass Layoffs

Dependent Variable: Dead by the year t+2 or t+15

High Tenure All Variable t+2 t+15 t+2 t+15 Males Mass Layoffs 0.406 0.131 0.271 0.086 (0.265) (0.081) (0.231) (0.067) Observations 124843 131260 181295 83388 Females Mass Layoffs

  • 0.053

0.057

  • 0.318

0.0084 (0.515) (0.117) (0.503) (0.117) Observations 59583 83388 103179 127449

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-17
SLIDE 17

Effect of Job Displacement on Mortality: Logistic Model, Plant Closures

Dependent Variable: Dead by the year t+2 or t+15

Males Females Variable t+2 t+15 t+2 t+15 Males Plant Closures 0.094 0.184

  • 0.304
  • 0.005

(0.425) (0.109) (1.013) (0.205) Observations 281768 278531 191082 208471

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-18
SLIDE 18

Effect of Job Displacement on Cause-Specific Mortality High Tenure Males

Dependent Variable: Cause-specific death by t+15

Variable Alcohol Heart Accidents Suicides Mass Layoffs

  • 0.0053

0.0598 0.157

  • 0.005

(0.239) (0.155) (0.225) (0.239) Plant Closures 0.256 0.355 0.132 0.256 (0.340) (0.250) (0.379) (0.340)

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-19
SLIDE 19

Next

Include older workers Extend to other periods Investigate the effect of spousal job loss

Disentangle the income effect from employment ending

Other outcomes: Family Income, Divorce Probability, Employment

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss

slide-20
SLIDE 20

Conclusions

Long-lasting severe earning losses for workers displaced in deep recession Modest mortality effects for male workers Workers displaced in recession years very similar to other workers

Christina Gathmann, Kristiina Huttunen and Robin Stitzing Lethal Layoffs: Understanding the Mortality Effects of Job Loss