Making Work Pay: An Assessment of the Experience with Action Emploi - - PowerPoint PPT Presentation

▶
making work pay an assessment of the experience with
SMART_READER_LITE
LIVE PREVIEW

Making Work Pay: An Assessment of the Experience with Action Emploi - - PowerPoint PPT Presentation

Jul 30, 2023 15 likes •256 views

The AE Program Empirical Evidence Econometric Model Results Conclusion Making Work Pay: An Assessment of the Experience with Action Emploi Guy Lacroix Dpartement dconomique Universit Laval Joint HRSDCCLSRN Meeting,

slide-1
SLIDE 1

The AE Program Empirical Evidence Econometric Model Results Conclusion

Making Work Pay: An Assessment of the Experience with “Action Emploi”

Guy Lacroix†

†Département d’économique

Université Laval

Joint HRSDC–CLSRN Meeting, Toronto, 18–19 November 2008

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-2
SLIDE 2

The AE Program Empirical Evidence Econometric Model Results Conclusion

Outline

The AE program Empirical Evidence Empirical Strategy Estimation Results Conclusion

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-3
SLIDE 3

The AE Program Empirical Evidence Econometric Model Results Conclusion

Origin of the program

Major Policy Initiatives: Family Assistance Programs

Social assistance (1996) “Individual trajectories”, parcours individualisés 5$/day Daycare (1997) Benefited single parents [B. Fortin & M. Blouin (2005)] Increased women’s participation [P . Fortin (2008), P . Lefebvre (2008)] Integrated child benefit & drug insurance to low-income parents, rather than SA recipients.

What more can be done ? (André Boisclair)

Meeting with André Boisclair SSP AE announced in 2001–2002 budget (for a single year) Permanent program, announced in March 2008 budget. Very little empirical evidence.

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-4
SLIDE 4

The AE Program Empirical Evidence Econometric Model Results Conclusion

Origin of the program

Major Policy Initiatives: Family Assistance Programs

Social assistance (1996) “Individual trajectories”, parcours individualisés 5$/day Daycare (1997) Benefited single parents [B. Fortin & M. Blouin (2005)] Increased women’s participation [P . Fortin (2008), P . Lefebvre (2008)] Integrated child benefit & drug insurance to low-income parents, rather than SA recipients.

What more can be done ? (André Boisclair)

Meeting with André Boisclair SSP AE announced in 2001–2002 budget (for a single year) Permanent program, announced in March 2008 budget. Very little empirical evidence.

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-5
SLIDE 5

The AE Program Empirical Evidence Econometric Model Results Conclusion

Policy relevance

Full-scale SSP-like experiment Targeted at poorly educated individuals with little labour market attachment Training programs have limited success. Wage subsidies believed to be more efficient (Phelps, Friedman, 1960’s) Tax Credits: EITC (U.S.), WFTC (U.K.), WITB (Canada, 2007), “Prime au Travail” (2005, Québec)

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-6
SLIDE 6

The AE Program Empirical Evidence Econometric Model Results Conclusion

The AE program

Program parameters

Implemented on December 1st 2001. SA recipients need to find full-time job within 12 months. Benefits paid out as early as January 1st 2002. Must have claimed benefits 36 out of last 45 months. Benefits not related to income, available for up to three years (3$/hour, 2$/hour, 1$/hour) Contacted by mail and caseload managers.

Timeline

Eligibility (36 months) Qualification (12 months) 01−12−2001 30−11−2002 30−11−2004 31−12−2005

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-7
SLIDE 7

The AE Program Empirical Evidence Econometric Model Results Conclusion

AE population

Type of household # individuals Proportion Singles 5 984 45.2 Single parents 4 118 31.3 Couples, no children 727 5.5 Couples, children 2 271 17.1 Unknown 144 1.1 Schooling Primary 660 5.0 Secondary (incomplete) 6 167 46.6 Secondary (completed) 1 302 26.4 CEGEP 1 302 9.8 University 792 6.0 Unknown 828 6.2 Cumulative months on SA 36–47 921 7.0 48–119 5847 44.1 120 + 6342 47.9 Unknown 134 1.0

Source: Direction générale de la recherche, 2003. Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-8
SLIDE 8

The AE Program Empirical Evidence Econometric Model Results Conclusion

Descriptive statistics

Variable AE Non-AE Mean Std Err Mean Std Err Age 33.712 7.772 35.107 9.647 Education 11.054 2.236 10.374 2.590 Number of children 1.586 0.870 1.520 0.871 Gender (1=Female) 0.879 0.326 0.870 0.337 Born in Canada 0.816 0.387 0.816 0.387 Region of residence Bas St-Laurent 0.025 0.019 Saguenay - Lac St-Jean 0.040 0.039 Capitale-Nationale 0.093 0.070 Mauricie 0.078 0.067 Estrie 0.049 0.033 Montréal 0.179 0.255 Outaouais 0.038 0.044 Abitibi-Témiscamingue 0.022 0.022 Côte-Nord 0.018 0.015 Nord du Québec 0.002 0.003 Gaspésie - Iles de la Madeleine 0.029 0.020 Chaudière -Appalaches 0.029 0.027 Laval 0.024 0.030 Lanaudière 0.062 0.046 Laurentides 0.071 0.052 Montérégie 0.150 0.141 Centre du Québec 0.011 0.007 Montréal banlieue 0.083 0.109 Number of observations 3 807 47 311 Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-9
SLIDE 9

The AE Program Empirical Evidence Econometric Model Results Conclusion

Probit Regression

Variable Marginal T-Stat P-Value Effect (∂Φ/∂X) Age

  • 0.001
  • 9.06

0.000 Education 0.008 16.48 0.000 Number of Children 0.007 5.69 0.000 Gender (1=Female) 0.005 1.58 0.114 Born in Canada

  • 0.003
  • 0.91

0.362 Region of residence (Montreal omitted) Bas St-Laurent 0.060 6.00 0.000 Saguenay - Lac St-Jean 0.032 4.53 0.000 Capitale-Nationale 0.053 9.38 0.000 Mauricie 0.042 7.40 0.000 Estrie 0.063 8.24 0.000 Outaouais 0.017 2.55 0.011 Abitibi-Témiscamingue 0.029 3.19 0.001 Côte-Nord 0.050 4.46 0.000 Nord du Québec

  • 0.000
  • 0.01

0.995 Gaspésie - Iles de la Madeleine 0.069 7.09 0.000 Chaudière -Appalaches 0.036 4.34 0.000 Laval 0.008 1.03 0.305 Lanaudière 0.056 8.43 0.000 Laurentides 0.056 8.84 0.000 Montérégie 0.032 7.48 0.000 Centre du Québec 0.060 3.96 0.000 Montréal banlieue 0.004 0.88 0.379 Log likelihood

  • 13230.965

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-10
SLIDE 10

The AE Program Empirical Evidence Econometric Model Results Conclusion

Examples of Transitions: Non-Participants

!"" #"$ %"& #"' () () *++,() %-.,/01234350.26

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-11
SLIDE 11

The AE Program Empirical Evidence Econometric Model Results Conclusion

Examples of Transitions: Participants

!"" #"$ %"& #"' ()*+,-,.)/+01 23 34 34523 677523

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-12
SLIDE 12

The AE Program Empirical Evidence Econometric Model Results Conclusion

Transitions: Non-Participants

Sample Transition Frequencies : Non-Participants

(a) Initial spell Destination state State of origin Off-SA SA Cen. Marginal Duration† Off-SA — 6,176 6,176 12.9 SA 22,359 — 18,776 41,135 39.6 Marginal 22,359 6,176 18,776 47,311 (b) Subsequent spells Destination state State of origin Off-SA SA Cen. Marginal Duration† Off-SA — 27,289 16,122 43,411 5.9 SA 21,052 — 12,413 33,465 13.0 Marginal 21,052 27,289 28,535 76,876

† Censoring not accounted for Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-13
SLIDE 13

The AE Program Empirical Evidence Econometric Model Results Conclusion

Transitions: Participants

Sample Transition Frequencies : Participants (a) Initial spell Destination state State of origin Off-SA SA AE AE-SA Cen Marginal Duration† Off-SA — 165 2 167 3.8 SA 2,470 — 967 203 3,640 22.6 Marginal 2,470 165 969 203 3,807 (b) Subsequent spells Destination state State of origin Off-SA SA AE AE-SA Cen Marginal Duration† Off-SA — 3,214 2,080 30 2,776 8,100 8.9 SA 2,567 — 435 90 830 3,922 9.7 AE 2,989 71 — 807 197 4,064 18.4 AE-SA 74 472 580 — 4 1,130 2.4 Marginal 5,630 3,757 3,095 927 3807 17,216

† Censoring not accounted for

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-14
SLIDE 14

The AE Program Empirical Evidence Econometric Model Results Conclusion

Smoothed Hazard Rates

Transitions between Off-SA and SA

.01 .01 .01 .02 .02 .02 .03 .03 .03 .04 .04 .04 Hazard Rate Hazard Rate Hazard Rate 20 20 20 40 40 40 60 60 60 Months Months Months part = 0 part = 0 part = 0 part = 1 part = 1 part = 1 (Intermediate Spells) (Intermediate Spells) (Intermediate Spells)

Transitions: Off-SA -> SA Transitions: Off-SA -> SA Transitions: Off-SA -> SA

.01 .01 .01 .02 .02 .02 .03 .03 .03 .04 .04 .04 .05 .05 .05 .06 .06 .06 Hazard Rate Hazard Rate Hazard Rate 20 20 20 40 40 40 60 60 60 80 80 80 Months Months Months part = 0 part = 0 part = 0 part = 1 part = 1 part = 1 (Intermediate Spells) (Intermediate Spells) (Intermediate Spells)

Transitions: SA -> Off-SA Transitions: SA -> Off-SA Transitions: SA -> Off-SA

Evaluation Two M.A. Theses have found the differences to be statistically significant using DD estimators. Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-15
SLIDE 15

The AE Program Empirical Evidence Econometric Model Results Conclusion

Proportional Hazard Transition Model

Basic Model Let λijk[t|AEi, Xi] be the hazard rate between states j and k for individual i: λijk[t|AEi, Xi] = exp[βkAEi + Xiγk] × λ0k(t) × θj Estimation Strategy Carefully track individual histories Allow endogenous AE participation Account for 4-month “grace period” Account for unobserved heterogeneity

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-16
SLIDE 16

The AE Program Empirical Evidence Econometric Model Results Conclusion

Proportional Hazard Transition Model

Basic Model Let λijk[t|AEi, Xi] be the hazard rate between states j and k for individual i: λijk[t|AEi, Xi] = exp[βkAEi + Xiγk] × λ0k(t) × θj Estimation Strategy Carefully track individual histories Allow endogenous AE participation Account for 4-month “grace period” Account for unobserved heterogeneity

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-17
SLIDE 17

The AE Program Empirical Evidence Econometric Model Results Conclusion

Covariates: Destination-Specific

Parameter estimates Weibull Duration Model Logit First Spell Subsequent Spells SA Off-SA AE Off-SA SA AE AE-SA Age

  • 3.843
  • 0.746
  • 7.454
  • 3.205
  • 1.939
  • 5.108
  • 4.654
  • 4.468

(0.065) (0.106) (0.399) (0.065) (0.058) (0.193) (0.364) (1.208) Schooling

  • 1.514
  • 1.015
  • 2.639
  • 0.163
  • 0.392
  • 0.897
  • 1.632

1.436 (0.021) (0.040) (0.137) (0.023) (0.021) (0.067) (0.131) (0.427) Gender

  • 0.522
  • 0.492
  • 0.861
  • 0.396
  • 0.150
  • 0.585
  • 0.244

0.709 (0.016) (0.029) (0.093) (0.019) (0.017) (0.054) (0.091) (0.294) # Children

  • 0.260
  • 0.110
  • 0.569
  • 0.096
  • 0.068
  • 0.293
  • 0.384
  • 0.612

(0.008) (0.014) (0.039) (0.006) (0.005) (0.019) (0.034) (0.099) Born Canada

  • 0.285
  • 0.286
  • 0.383
  • 0.195

0.011

  • 0.297
  • 0.294

0.194 (0.015) (0.028) (0.085) (0.018) (0.016) (0.047) (0.083) (0.312) Montreal

  • 0.084
  • 0.049
  • 0.519
  • 0.273
  • 0.058
  • 0.403
  • 0.837
  • 0.492

(0.015) (0.031) (0.084) (0.016) (0.015) (0.051) (0.097) (0.338) AE 0.059

  • 0.339

(0.026) (0.024) Completed AE

  • 0.848
  • 0.344

(0.056) (0.065) Unobserved heterogeneity (Origin-Specific)

  • 1.238
  • 1.977
  • 0.788

0.285

  • 0.080

(0.012) (0.011) (0.046) (0.100) (0.260) Probability of type II 0.837 (0.010) Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-18
SLIDE 18

The AE Program Empirical Evidence Econometric Model Results Conclusion

Transition Intensities Parameters

(a) Initial spell (Weibull parameters) Destination state State of origin Off-SA SA AE Off-SA 1.225 (0.012) SA 1.583 1.569 (0.007) (0.050) (b) Subsequent spells (Weibull parameters) Destination state State of origin Off-SA SA AE AE-SA Off-SA 0.691 0.666 (0.005) (0.018) SA 0.699 0.285 (0.005) (0.012) AE 0.946 1.055 (0.013) (0.033) (c) Subsequent spells: AE-SA spline parameters Duration (month) State of origin 1 month 2 months 3-4 months AE 1.022 0.753 0.652 (4.040) (0.273) (0.203) Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-19
SLIDE 19

The AE Program Empirical Evidence Econometric Model Results Conclusion

Simulation Results

Simulation results, mean heterogeneity Age 20 30 35 40 50 Mean duration Off-SA 7.7 10.9 11.8 12.9 15.6 Mean duration SA 14.9 18.3 19.6 21.2 24.6 % Participants 4.6 3.5 2.5 1.4 1.3 % time off SA 28.9 28.6 30.1 29.7 29.5 % time on SA 70.5 70.8 69.3 69.9 70.2 Average # Spells 5.9 4.7 4.3 3.9 3.3 Education 10 12 14 16 18 Mean duration Off-SA 9.1 10.3 10.6 11.3 11.5 Mean duration SA 16.3 17.4 18.9 20.5 22.4 % Participants 3.8 3.8 2.6 2.7 2.6 % time off SA 29.9 30.7 29.3 28.7 27.1 % time on SA 69.7 68.7 70.3 70.8 72.5 Average # Spells 5.2 4.8 4.5 4.2 3.9 # of Children 1 2 3 4 5 Mean duration Off-SA 10.5 10.9 12.3 12.7 12.8 Mean duration SA 18.8 20.1 22.0 24.3 26.2 % Participants 2.5 1.8 2.2 2.1 0.3 % time off SA 28.6 28.2 27.8 25.5 24.5 % time on SA 70.9 71.4 71.8 73.9 75.4 Average # Spells 4.5 4.3 3.9 3.6 3.4

† Based on 1,000 replications, average-modal individual

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-20
SLIDE 20

The AE Program Empirical Evidence Econometric Model Results Conclusion

Simulated Transitions: Fresh Spells

f

.05 .05 .05 .1 .1 .1 .15 .15 .15 .2 .2 .2 Transition Intensities Transition Intensities Transition Intensities 50 50 50 100 100 100 150 150 150 Months Months Months Non-Participants Non-Participants Non-Participants Participants - Incomplete Participants - Incomplete Participants - Incomplete Participants - Complete Participants - Complete Participants - Complete (Fresh Spells, Type I individuals) (Fresh Spells, Type I individuals) (Fresh Spells, Type I individuals)

Simulated Transition Intensities: Off-SA -> SA Simulated Transition Intensities: Off-SA -> SA Simulated Transition Intensities: Off-SA -> SA

.01 .01 .01 .02 .02 .02 .03 .03 .03 .04 .04 .04 .05 .05 .05 Transition Intensities Transition Intensities Transition Intensities 50 50 50 100 100 100 150 150 150 Months Months Months Non-Participants Non-Participants Non-Participants Participants - Incomplete Participants - Incomplete Participants - Incomplete Participants - Complete Participants - Complete Participants - Complete (Fresh Spells, Type II individuals) (Fresh Spells, Type II individuals) (Fresh Spells, Type II individuals)

Simulated Transition Intensities: Off-SA -> SA Simulated Transition Intensities: Off-SA -> SA Simulated Transition Intensities: Off-SA -> SA

.02 .02 .02 .04 .04 .04 .06 .06 .06 .08 .08 .08 .1 .1 .1 Transition Intensities Transition Intensities Transition Intensities 50 50 50 100 100 100 150 150 150 Months Months Months Non-Participants Non-Participants Non-Participants Participants - Incomplete Participants - Incomplete Participants - Incomplete Participants - Complete Participants - Complete Participants - Complete (Fresh Spells, Type I individuals) (Fresh Spells, Type I individuals) (Fresh Spells, Type I individuals)

Simulated Transition Intensities: SA -> Off-SA Simulated Transition Intensities: SA -> Off-SA Simulated Transition Intensities: SA -> Off-SA

.005 .005 .005 .01 .01 .01 .015 .015 .015 Transition Intensities Transition Intensities Transition Intensities 50 50 50 100 100 100 150 150 150 Months Months Months Non-Participants Non-Participants Non-Participants Participants - Incomplete Participants - Incomplete Participants - Incomplete Participants - Complete Participants - Complete Participants - Complete (Fresh Spells, Type II individuals) (Fresh Spells, Type II individuals) (Fresh Spells, Type II individuals)

Simulated Transition Intensities: SA -> Off-SA Simulated Transition Intensities: SA -> Off-SA Simulated Transition Intensities: SA -> Off-SA

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-21
SLIDE 21

The AE Program Empirical Evidence Econometric Model Results Conclusion

Expected Durations: Fresh Spells

Expected Durations, by type and AE status Type I Het Type II Het Non Part Incomp Comp Non Part Incomp Comp Mean duration Off-SA 3.953 6.440 10.204 17.728 21.173 22.468 Mean duration SA 9.590 8.885 19.910 19.291 19.795 11.894 % time off SA 0.292 0.420 0.339 0.479 0.517 0.654

† Based on 1000 replications, average–modal individual

Simulation Results: AE vs No AE With AE No Type AE I II Mean Mean duration Off-SA 8.6 10.8 8.7 10.6 Mean duration SA 16.7 19.3 16.9 26.4 % Participants 2.2 2.2 3.3 – % time off SA 28.0 28.2 27.7 20.3 % time on SA 71.7 71.3 71.8 79.7 Average # Spells 5.3 4.4 5.2 3.5

† Based on 1000 replications, average–modal individual

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-22
SLIDE 22

The AE Program Empirical Evidence Econometric Model Results Conclusion

Expected Durations: Fresh Spells

Expected Durations, by type and AE status Type I Het Type II Het Non Part Incomp Comp Non Part Incomp Comp Mean duration Off-SA 3.953 6.440 10.204 17.728 21.173 22.468 Mean duration SA 9.590 8.885 19.910 19.291 19.795 11.894 % time off SA 0.292 0.420 0.339 0.479 0.517 0.654

† Based on 1000 replications, average–modal individual

Simulation Results: AE vs No AE With AE No Type AE I II Mean Mean duration Off-SA 8.6 10.8 8.7 10.6 Mean duration SA 16.7 19.3 16.9 26.4 % Participants 2.2 2.2 3.3 – % time off SA 28.0 28.2 27.7 20.3 % time on SA 71.7 71.3 71.8 79.7 Average # Spells 5.3 4.4 5.2 3.5

† Based on 1000 replications, average–modal individual

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-23
SLIDE 23

The AE Program Empirical Evidence Econometric Model Results Conclusion

Conclusion

1

Results qualitatively similar to those obtained from simple difference-in-differences estimator.

2

Government probably justified, ex post, to implement the program on a permanent basis. But,

3

Very low participation rate. What will happen if it increases (i.e. general equilibrium effects) ?

4

Off-SA can mean On-EI . . .

5

Need to carefully account for interactions between Provincial - Federal programs.

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”

slide-24
SLIDE 24

The AE Program Empirical Evidence Econometric Model Results Conclusion

Conclusion

1

Results qualitatively similar to those obtained from simple difference-in-differences estimator.

2

Government probably justified, ex post, to implement the program on a permanent basis. But,

3

Very low participation rate. What will happen if it increases (i.e. general equilibrium effects) ?

4

Off-SA can mean On-EI . . .

5

Need to carefully account for interactions between Provincial - Federal programs.

Guy Lacroix Making Work Pay: An Assessment of “Action Emploi”