The Risk and Cost of Job Loss in Canada, The Risk and Cost of Job Loss in Canada, 1978-2008 René Morissette, Hanqing Qiu and Ping Ching Winnie Chan R é M i tt H i Qi d Pi Chi Wi i Ch Social Analysis Division Statistics Canada Statistics Canada 1
Plan of presentation Plan of presentation I MOTIVATION I. MOTIVATION II. DATA III. METHODS IV. RESULTS V. CONCLUSION 2
I. Motivation I. Motivation • Stiglitz (2009) report: “Steps should be taken to improve measures of people’s health, education, personal activities, political voice, social connections, environmental conditions and insecurity.” (p. 58) • Goal = contribute to the measurement of economic insecurity due to job loss due to job loss ― or job (in)security ― by analyzing 3 of its or job (in)security by analyzing 3 of its dimensions: – probability of job loss – Incidence of re ‐ employment following job loss I id f l t f ll i j b l – Short ‐ term + longer ‐ term earnings losses following job loss 3
I. Motivation I. Motivation • Assessing ∆ Probability of job loss may: – help uncover changes in the employer ‐ employee relationship – have implications for education/training policies concerned with workers’ adaptability to shocks • Documenting the evolution of re ‐ employment rates and longer ‐ term earnings losses informs discussions regarding: l i l i f di i di – Optimal design of Employment Insurance (EI) program – Policies, if any, aimed at compensating displaced workers 4
II. DATA II. DATA Longitudinal Worker File (LWF): 1978 ‐ 1989 + 1983 ‐ 2010 • • LWF: – Administrative data set – 10% random sample of Canadian workers – Includes: I l d • Age, sex, marital status (no info on education & occupation) • Industry, union status, firm size, pension coverage y, , , p g • *** Allows the identification of layoffs vs other separations (through the record of employment :ROE) • If laid ‐ off worker returns to the firm in year t or year t+1 == temporary layoff, otherwise permanent layoff. I 5
II. DATA II. DATA • Focus is on permanent layoffs among … • … Paid workers aged 25 to 54 6
II. DATA II. DATA • OUTCOMES CONSIDERED: OUTCOMES CONSIDERED: 1. Permanent layoff rate = – Number of permanent layoffs_t (from LWF) / / Average annual paid employment_t (from LFS) 2. Post ‐ layoff paid employment rate = – % of laid ‐ off workers with wages and salaries in the year following layoff (t+1) 7
II. DATA II. DATA 3. Short ‐ term earnings losses: = 1 year after layoff – Narrow sample = workers with wages from t ‐ 1 to t+1 – Broad sample = add workers with no wages in t+1 4 Long term earnings losses : 4. Long ‐ term earnings losses : = 5 years after layoff 5 years after layoff – Narrow sample = wages >= $10,000* in t ‐ 1 and >0 up to t+5 – Broad sample = add workers with no earnings in some post ‐ layoff years if they filled a tax form in t+5 (and still alive) y + – Sub ‐ samples = workers with “stable Labour market attachment” and “high ‐ seniority” workers: • • a) >= $10 000 from t 6 to t 4 b) > 0 from t 3 to t 1 and c) > 0 up to t+5 a) >= $10,000 from t ‐ 6 to t ‐ 4, b) > 0 from t ‐ 3 to t ‐ 1, and c) > 0 up to t+5 (narrow), or other restrictions (broad) * 2002 dollars 8
II. DATA II. DATA 5 Short ‐ term (measured) earnings declines: = 1 year after layoff 5. Short ‐ term (measured) earnings declines: = 1 year after layoff – Explain the worsening of earnings declines among displaced manufacturing workers from the late 1990s to the mid ‐ 2000s 9
III. METHODS III. METHODS • Short ‐ term earnings losses + Long ‐ term earnings losses: – Fixed effects models à la Jacobson, Lalonde and Sullivan (1993) – Control group = individuals who satisfy the earnings restrictions imposed on the narrow treatment group but who were not laid ‐ off in year t – For high ‐ seniority workers and workers with stable labour market attachment, we allow earnings to fall (up to) 3 years prior to h ll i f ll ( ) 3 i displacement – For narrow samples: both log Earnings and Earnings are used as a F l b th l E i d E i d dependent variable – For broad samples: Earnings For broad samples: Earnings 10
III. METHODS III. METHODS • 30 cohort ‐ specific models of short ‐ term losses 30 cohort specific models of short term losses • 26 cohort ‐ specific models of long ‐ term losses • 16 cohort ‐ specific models of long ‐ term losses for high ‐ seniority workers and workers with stable l.m.a. y • Worsening of measured short ‐ term earnings declines of displaced manufacturing workers 1998 99 vs 2005 06 displaced manufacturing workers, 1998 ‐ 99 vs 2005 ‐ 06: – Decomposition method of Firpo, Fortin and Lemieux (2009) 11
12 IV. RESULTS IV. RESULTS
Despite numerous changes in the Canadian labour market, the risk of job loss did not trend upwards over the last three j p decades PERMANENT LAYOFF RATES OF WORKERS AGED 25-54 % 16.0 14.0 12.0 12 0 10.0 8.0 6.0 4.0 2.0 0.0 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Men Women 13
This was true in manufacturing … g PERMANENT LAYOFF RATES IN MANUFACTURING % % 16.0 14.0 12.0 10.0 8.0 6 0 6.0 4.0 2.0 0.0 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Men Women 14
… as well as outside manufacturing PERMANENT LAYOFF RATES OUTSIDE MANUFACTURING % % 16.0 14.0 12.0 10.0 8.0 6 0 6.0 4.0 2.0 0.0 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Men Women 15
Yet a closer look at manufacturing reveals sharply diverging layoff rates across industries in recent years l ff t i d t i i t Permanent layoff rates in manufacturing, by 3-digit industry, 1998-1999 vs 2005-2006 Men Women 1998-99 2005-06 1998-99 2005-06 % % Manufacturing 6.1 5.1 6.5 5.6 Textile mills 4.0 6.7 4.0 9.0 Textile product mills 5.4 5.7 6.0 6.6 Clothing 7.4 8.4 6.6 9.3 Petroleum and coal products Petroleum and coal products 3.1 3.1 1.8 1.8 2.6 2.6 1.5 1.5 Non-metallic mineral products 8.6 6.3 7.9 4.3 16
Women’s post ‐ displacement paid employment rates (1 year after job loss) increased markedly j ) y Post-displacement paid employment rates in t+1,1978-2008 % 90.0 85.0 80.0 80 0 75.0 70.0 65.0 60.0 55.0 50.0 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Men Women 17
Although sensitive to labour market tightness, short ‐ term earnings losses displayed no trend in the aggregate g p y gg g Short ‐ term earnings losses, 1979 ‐ 2008 a) Men: All industries ) M All i d i b) W b) Women: All industries All i d i Narrow sample: log Y Narrow sample: log Y Narrow sample: Y Narrow sample: Y % % % Broad sample: Y Broad sample: Y 10 10 0 0 ‐ 10 ‐ 10 ‐ 20 ‐ 20 ‐ 30 ‐ 30 ‐ 40 ‐ 40 ‐ 50 ‐ 50 ‐ 60 ‐ 60 79 82 85 88 91 94 97 00 03 06 79 82 85 88 91 94 97 00 03 06 18
Despite falling aggregate unemployment rates, short ‐ term earnings losses of displaced manufacturing workers worsened from the late 1990s to the mid ‐ 2000s 2000s Short ‐ term earnings losses, 1979 ‐ 2008 c) Men: Manufacturing d) Women: Manufacturing 10 10 0 0 0 0 ‐ 10 ‐ 10 ‐ 20 ‐ 20 ‐ 30 ‐ 30 ‐ 40 ‐ 40 ‐ 50 ‐ 50 ‐ 60 ‐ 60 79 82 85 88 91 94 97 00 03 06 79 82 85 88 91 94 97 00 03 06 19
This deterioration in short ‐ term earnings losses was not observed outside manufacturing Short ‐ term earnings losses, 1979 ‐ 2008 e) Men: Outside manufacturing f) Women: Outside manufacturing 10 10 0 0 0 0 ‐ 10 ‐ 10 ‐ 20 ‐ 20 ‐ 30 ‐ 30 ‐ 40 ‐ 40 ‐ 50 ‐ 50 ‐ 60 ‐ 60 79 82 85 88 91 94 97 00 03 06 79 82 85 88 91 94 97 00 03 06 20
Like short ‐ term earnings losses, long ‐ term earnings losses responded to labour market tightness g Long ‐ term earnings losses of displaced workers earning at least $10,000 in year t ‐ 1, 1979 ‐ 2004. a) Men: All industries b) Women: All industries Narrow sample: log Y Narrow sample: log Y Narrow sample:Y Narrow sample:Y Narrow sample: Y Narrow sample: Y % % Broad sample:Y Broad sample: Y 20 20 0 0 ‐ 20 ‐ 20 ‐ 40 40 ‐ 40 40 ‐ 60 ‐ 60 79 82 85 88 91 94 97 00 03 79 82 85 88 91 94 97 00 03 21
Long ‐ term earnings losses of displaced workers with a stable labour market attachment exceeded those observed in the aggregate Long ‐ term earnings losses of displaced workers with stable labour market attachment, 1989 ‐ 2004 a) Men: stable labour market attachment a) Men: stable labour market attachment b) Women: stable labour market attachment b) Women: stable labour market attachment Narrow sample: log Y Narrow sample: log Y Narrow sample:Y Narrow sample: Y % % Broad sample:Y p Broad sample: Y p 20 20 10 10 0 0 ‐ 10 ‐ 10 ‐ 20 ‐ 20 ‐ 30 ‐ 30 ‐ 40 ‐ 40 ‐ 50 ‐ 50 ‐ 60 ‐ 60 89 91 93 95 97 99 01 03 89 91 93 95 97 99 01 03 22
Recommend
More recommend
