RETURNS TO VOCATIONAL EDUCATION IN PORTUGALSofia Oliveira, Nova SBE

Variable Estimates

Exp0.0211***(0.0006)

Exp2 -0.0001**(0.0001)

Voc0.0190***(0.0018)

Voc*Exp-0.0025**(0.0010)

Voc*Exp2 -0.0001(0.0001)

No. of observations 1 509 643

No. of individuals 553 319

𝑅2 (overall) 0.54

𝜌 0.56Source: Quadros de PessoalNote: Dependent variable: logarithm of wage. Robust standard errorsin parentheses. Significant at ***1%, **5%, *10%.

3. Variables

Explanatory variables:

Worker-related variables:‐ Type of education (𝑣𝑜𝑐=1 if vocational)‐ Age‐ Gender‐ Hours worked per week‐ Tenure in years‐ Labor-market experience in years‐ Previous experience‐ Presence in the labor market with 18, 19, 20

and 21 years old‐ Hierarchy level within the firm

Firm-related variables:‐ Industry‐ Region‐ Log of sales‐ Number of branches‐ Number of employees‐ Shares of public and private capital stock

Explained variable: log of wage

5. Results

4. Econometric Model

Estimation method: random effects.

Plus worker’s and firm’s controls.

𝑙𝑤𝑎𝑔𝑒𝑖𝑡= 𝛽0 + 𝛽4𝑒𝑥𝑝𝑖𝑡 + 𝛽5𝑒𝑥𝑝𝑖𝑡

2 + 𝛽6𝑣𝑜𝑐𝑖 + 𝛽7𝑣𝑜𝑐𝑖 ∗ 𝑒𝑥𝑝𝑖𝑡+ 𝛽8𝑣𝑜𝑐𝑖 ∗ 𝑒𝑥𝑝𝑖𝑡

2 + 𝜇𝑖 + 𝜀𝑖𝑡

- Evaluate the regression at average values of allvariables but 𝑣𝑜𝑐, 𝑒𝑥𝑝, 𝑒𝑥𝑝2 and interaction𝑣𝑜𝑐 ∗ 𝑒𝑥𝑝

- Let 𝑒𝑥𝑝 to vary within its whole range (between0 and 16) and calculate the fitted regression forthe two groups of workers.

6. Wage-Experience Patterns

Investigating differences between job-orientedtracks:

7. Summing up

‐ Workers with vocational education earn 2%more at the beginning of the career, but theirwages grow at a slower rate.

‐ Generally educated workers catch up at 8 yearsof experience.

‐ Such wage-experience profile is morepronounced in the case of vocational coursesof level III when compared to other job-oriented tracks.

6,5

6,6

6,7

6,8

6,9

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Log

of w

age

Experience

Wage-Experience Profiles

General Education

Vocational Education

Source: Quadros de Pessoal

6.9

6.8

6.7

6.6

6.5 7.8

6,6

6,7

6,8

6,9

7

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Log

of w

age

Experience

Wage-Experience Profiles, disaggregated vocational education

Vocational courses of level IIIOther job-oriented tracksGeneral education

Source: Quadros de Pessoal

7

6.9

6.8

6.7

6.6 9.25.7

2. Dataset and Sample

Dataset: Quadros de Pessoal

Selected sample:

Individuals with upper secondary attainment

(12th grade)

Reasons for such sample:

Observations from 1993 to 2009

Individuals born between January 1974

and 1990

- Upper secondary was the main target of

vocational education reforms in Portugal

- Increases similarity between individuals

- These individuals experienced the

education system after the creation of

vocational schools in 1989

- The first cohort of individuals is expected

to be in the labor market by 1993, and

the last year with available data is 2009

General vs Vocational Education

‐ Is there a wage premium to vocationaleducation vis-à-vis general education whenentering the labor market?

‐ If so, does it dissipate over the worker’slifecycle?

Goal: to derive wage-experience patterns.

Research Questions

General education or academic track (“ensinoregular”) prepares students especially to pursuesuperior level studies.Vocational education or job-oriented track(“ensino profissional”) prepares studentsespecially for an occupation.

The sample is an unbalanced panel.# individuals: 633 549# observations: 1 794 272% of workers with vocational education: 13.37%

1. Literature Review

Empirical studies on the returns to differenttypes of education:- Main references:

Hanushek, Woessman and Zhang (2011)Cörvers et al. (2011)

‐ For the Portuguese labor market:Pereira and Martins (2001)