Education–Occupation Mismatch and Its Wage - Ricerca sull'intelligenza artificiale specializzata al MIT

Education–Occupation Mismatch and Its Wage
Penalties in Informal Employment in Thailand
Tanthaka Vivatsurakit and Jessica Vechbanyongratana∗

This study examines the incidence of vertical mismatch among formal and
informal workers in Thailand. Using the 2011, 2013, E 2015 Thailand
the study analyzes the relationship
Household Socio-economic Surveys,
between vertical mismatch and wage penalties and premiums across four types
of workers: formal government, formal private ﬁrm, informal private ﬁrm,
and informal own-account workers. The incidence of overeducation is modest
among the oldest cohort (8.7%) but prevalent among the youngest cohort
(29.3%). Government employees face the highest overeducation wage penalties
(28.2%) compared to matched workers, while in private ﬁrms, informal workers
have consistently higher overeducation wage penalties than formal workers.
Educated young workers are increasingly absorbed into low-skill informal work
in private ﬁrms and face large overeducation wage penalties. The inability of
many young workers to capitalize on their educational investments in Thailand’s
formal labor market is a concern for future education and employment policy
development in Thailand.

Keywords: informality, overeducation, returns to education, Thailand, vertical
mismatch
JEL codes: A20, E26, I26, J01

IO. introduzione

Over the past several decades, developing economies have emphasized
the expansion of education and increasing educational attainment for their
citizens as a means to achieve economic development. Despite rapidly increasing
educational attainment, subsequent skilled job growth has often lagged behind.
The combination of a rapidly growing educated workforce and slow growth of
skilled employment can lead to a problem of “overeducation”—also called vertical
mismatch—in developing countries, meaning that educated workers engage in
employment that requires less formal education than they have acquired.

∗Tanthaka Vivatsurakit: Faculty of Economics, Chulalongkorn University, Thailand. E-mail: tanthaka@gmail.com;
Jessica Vechbanyongratana (corresponding author): Faculty of Economics, Chulalongkorn University, Thailand.
E-mail: jessica.v@chula.ac.th. An earlier version of this paper was presented at the 15th Western Economic
Association International conference at Keio University, Tokyo. In addition to conference participants, we thank
Pundpond Rukumnuaykit, Nuarpear Lekfuangfu, Sasiwimon Warunsiri Paweenawat, Yong Yoon, the managing
editor, and two anonymous referees for helpful comments and suggestions. The usual Asian Development Bank
disclaimer applies.

Asian Development Review, vol. 38, NO. 1, pag. 119–141
https://doi.org/10.1162/adev_a_00160

© 2021 Asian Development Bank and
Asian Development Bank Institute.
Pubblicato sotto Creative Commons
Attribuzione 3.0 Internazionale (CC BY 3.0) licenza.

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

120 Asian Development Review

The existence of widespread informal employment in developing economies
adds a layer of concern against increasing rates of overeducation. According to
the International Labour Organization (ILO), own-account workers working in
informal enterprises, as well as employees whose “employment relationships [are],
in law or in practice, not subject to national labor legislation, income taxation, social
protection or entitlement to certain employment beneﬁts,” are considered informally
employed (International Labour Organization 2003). Informal employment
È
generally associated with low skill and low pay. Così, in a developing country
informal
context where formal employment growth is often slow,
employment may need to absorb a growing educated workforce, potentially
exacerbating overeducation wage penalties.

low-skill

This paper evaluates the incidence of vertical mismatch and associated
wage penalties and premiums across formal and informal employment in Thailand.
Thailand is a representative case of a developing country with a rapidly expanding
educated workforce alongside high rates of informal employment and slow formal
employment growth. Since the government’s supply of education and compulsory
education laws vary across different generations of workers, we analyze the
incidence of vertical mismatch and associated wage penalties across age cohorts. In
aggiunta, this paper analyzes the relationship between vertical mismatch and wage
penalties and premiums across four types of workers, including formal government,
formal private ﬁrm, informal private ﬁrm, and informal own-account workers.

We hypothesize that the incidence of overeducation will be higher among
younger cohorts due to rapid increases in compulsory education relative to skilled
job growth. Likewise, we expect the incidence of overeducation to be higher in
informal employment because the average skill level for informal jobs is low while
informal work has increasingly absorbed Thailand’s young, educated workforce.
We hypothesize that overeducation wage penalties are relatively high for formal
government employees compared to other types of workers because of the rigid
compensation system that sets pay based on occupation and experience, but gives
little additional reward for education completed beyond what is required for the
position. By contrast, the private sector is more ﬂexible in allowing overeducated
employees to fully utilize their abilities and is more likely to pay based on
capabilities (Dolton and Vignoles 2000). By extending the same logic, we expect
workers in informal private ﬁrm employment and particularly in informal own-
account work to have lower overeducation wage penalties than formal government
workers. Tuttavia, it is an empirical question whether formal or informal workers
in private ﬁrms have higher overeducation wage penalties.

The analysis uses individual-level data from the 2011, 2013, E 2015 rounds
of the Thai Socio-economic Survey (SES). Consistent with our hypothesis, we ﬁnd
that the incidence of overeducation is most prevalent (29.3%) among the youngest
cohort born between 1981 E 1990 and least prevalent (8.7%) among the oldest
cohort born between 1951 E 1960. We also ﬁnd high rates of overeducation in

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Education–Occupation Mismatch and Its Wage Penalties in Thailand 121

informal employment. This is particularly the situation among the youngest cohort,
Dove 37.3% of informal workers in private ﬁrms and 50.1% of informal own-
account workers are overeducated.

Using an augmented Mincerian wage regression, we ﬁnd that the overall
overeducation wage penalty is 20.9%, while the undereducation wage premium is
10.2%. Generalmente, we ﬁnd that overeducation wage penalties are higher in older
cohorts, suggesting that these penalties become larger later in one’s career. IL
penalties and premiums are similar across men and women. As expected, wage
penalties for government employees are relatively high at 28.2%, while the lowest
penalties belong to informal own-account workers at 3.9%. As for employees
in private ﬁrms, informal workers have consistently higher overeducation wage
penalties than formal workers across all age cohorts. Educated young workers
are increasingly absorbed into low-skill informal work in private ﬁrms and face
large overeducation wage penalties. The inability of many young workers to
capitalize on their educational investments in Thailand’s formal labor market is a
concern for future education and employment policy development in Thailand.

This paper is organized as follows. Section II provides a background on
Thailand’s education policies since the 1970s, its rising educational attainment, E
the growth of its formal workforce. Section III gives a brief review of the literature
on measuring overeducation and its wage penalties. This is followed by a description
of the data used in the analysis in section IV and the methodology in section V.
Section VI presents the empirical results followed by a discussion and conclusion
in section VII.

II. Thailand’s Rising Educational Attainment, Structural Change,

and Formalization of Work

As is the case with many developing countries, Thailand has prioritized
the expansion of education as a means to achieve economic development. Since
the 1970s, Thailand has increased compulsory levels of schooling from 4 years
A 9 years and initiated a large expansion of secondary and tertiary education.1
IL 1980 National Primary Education Act mandated that all villages should be
equipped with schools. One of the major changes in the Thai education system
was the increase in government-mandated compulsory education from 4 years to 6
years in the 1970s and from 6 years to 9 years implemented in 2002. Consequently,
the share of workers who have completed upper secondary school went up from
17% In 1990 A 25% In 2 decades (Aemkulwat 2010). Over the same period, IL
number of workers with vocational qualiﬁcations increased from 1.8 million to

1The Thai education system is split into primary education (grades 1–6), lower secondary education (grades
7–9), and upper secondary/lower vocational education (grades 10–12). Tertiary education includes postsecondary
upper vocational training, 4-year university education, and higher level degrees.

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

122 Asian Development Review

Figura 1. Gross Enrollment Rates in Thailand, 1971–2013

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Note: Gaps are due to missing data for some years.
Fonte: World Bank. World Development Indicators. https://databank.worldbank.org/source/world-development-
indicators (accessed May 2019).

3 million (Aemkulwat 2010). Thailand also saw a signiﬁcant increase in the number
of educational institutions at all levels, especially at the secondary and tertiary
levels. Per esempio, the number of higher education institutions rose from a handful
In 1970 A 185 institutions in 2014 (Paweenawat and Vechbanyongratana 2015). IL
expansion of schools combined with changes in the compulsory education laws led
to a steady increase in primary, secondary, and tertiary gross enrollment rates from
1971 A 2013, as shown in Figure 1. Primary education enrollment became universal
in the 1980s, while secondary enrollment increased from 18% A 82%, and tertiary
education from 3% A 50% since 1970.

In the past, Thailand’s economy was based primarily on agriculture. Thailand
has undergone a signiﬁcant economic transformation that started in the 1970s.
It experienced a rapid demographic transition, encouraged investment to develop
its manufacturing sector, and saw people move out of rural agriculture and into
work in urban areas (Baker and Phongpaichit 2009). Following the world’s oil
crisis in 1973 and other external factors, Thailand shifted toward export-oriented
manufacturing in the 1980s,
increasing exports of primarily labor-intensive
products by approximately 24% per year during 1984–1989 (Baker and
Phongpaichit 2009). From the 1990s onward, the tourism and service sectors
experienced growth in part due to the government’s promotion of Thailand as a
tourist destination (Kaosa-ard 2002). Figura 2 shows the contributions of each
sector to total employment in Thailand between 1991 E 2018.

Education–Occupation Mismatch and Its Wage Penalties in Thailand 123

Figura 2. Thailand’s Sectoral Employment Shares, 1991–2018

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Fonte: World Bank. World Development Indicators. https://databank.worldbank.org/source/world-development-
indicators (accessed May 2019).

Figura 2 demonstrates that the agriculture sector saw a rapid decline in its
contribution to employment, dropping from 60% A 32% over the 28-year period.
Allo stesso tempo, employment in the service sector rose rapidly from 22% A
45%, while the share of workers in manufacturing continued to rise during this
period, albeit more slowly, from 18% A 23%. Thailand’s smallholder agricultural
past means that most employment was traditionally considered informal. Since
the 1990s, a signiﬁcant number of workers moved from a work status of “unpaid
family worker” to “employees of private companies” (Aemkulwat 2010). Tuttavia,
even though Thailand experienced a major transformation of its economy over the
past 4 decades, the country largely did not experience concurrent formalization
of employment. The Thai government deﬁnes formal workers as employees who
are covered by employer-provided social insurance (such as the Civil Servants’
Welfare Scheme or protection under the Social Security Act B.E. 2533 [1990])
and protection under the labor law.2 The growth of formal private ﬁrm employment
through the expansion of social security has been slow, but it has picked up in recent

2The Thai government deﬁnes formal workers as follows: all government ofﬁcers and employees; all state
enterprise employees; all teachers in private schools, according to the Private School Act; government ofﬁcers and
employees of other countries or those who work in international organizations; all employees who have protection
under labor legislation; and workers who have social security according to Social Security Act B.E. 2533 (1990)
(Ministry of Labour n.d.).

124 Asian Development Review

Figura 3. Distribution of Formal and Informal Workers across Occupational
Categories, 2015

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

ISCO-08 = 2008 International Standard Classiﬁcation of Occupations.
Notes: The tabulation includes only workers who receive labor income, including government workers, private ﬁrm
employees, and own-account workers. The tabulations exclude employers and unpaid family workers.
Fonte: Authors’ calculations from the 2015 Thailand Socio-economic Survey.

years. The number of private ﬁrm workers covered by Section 33 of the Social
Security Act has grown from 8.6 million workers in 2008 A 10.8 million workers in
2017, which represents an increase from 23% A 29% of the total workforce. Despite
efforts to expand formal employment, Thailand’s informal workers continue to
make signiﬁcant contributions to the country’s economy, with ofﬁcial ﬁgures putting
the share of informal workers in the total workforce at 55% In 2018 (National
Statistical Ofﬁce 2018).

Informal employment is not distributed evenly across all occupations. Figura
3 shows the distribution of formal workers (government workers and formal
private ﬁrm employees) and informal workers (informal private ﬁrm employees and
own-account workers) who receive cash remuneration across occupational
categories based on the 1-digit 2008 International Standard Classiﬁcation of
Occupations (ISCO-08). Occupations requiring the highest levels of education and
skill are located toward the left side of Figure 3, including managers, professionals,
and technicians and associate professionals. These categories largely encompass
civil servants and highly skilled workers in larger private ﬁrms, and thus workers
employed in these occupations are generally formal. The occupational categories
that require the least education and skills are located toward the right side of the

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Education–Occupation Mismatch and Its Wage Penalties in Thailand 125

ﬁgure, including craft and related trades workers, plant and machine operators and
assemblers, and elementary occupations. Informal workers are disproportionately
represented in the occupational groups located on the right side of the ﬁgure. IL
one exception is the high number of formal workers in occupation category 8—plant
and machine operators and assemblers—which encompasses lower skilled factory
lavoro. The government’s push to develop the manufacturing sector during the 1980s
and 1990s attracted larger ﬁrms such that the government subsequently required
them to register for tax (including employment tax) purposes, which explains why
workers in occupation category 8 are largely formal. Despite the government’s
mandate that all ﬁrms hiring one or more workers must register their employees for
social security, many smaller enterprises remain unregistered, often intentionally
to avoid taxation and social security contributions. Own-account workers—who
are informally employed by deﬁnition—generally work in lower skill occupational
categorie. Infatti, approximately 95% of own-account workers are classiﬁed as
working in occupation categories 5 through 9, making up a signiﬁcant proportion
of workers in these categories.

Although one ﬁnds that informal workers are disproportionately represented
in occupations requiring lower skill and education, it is important to note that within
the Thai context, one ﬁnds both formal and informal workers often performing the
same jobs. Per esempio, according to the 2016 Thai Labor Force Survey Informal
Supplement, informal workers engaged in food, beverages, textile, and wearing
apparel manufacturing constituted 38%, 32%, 32%, E 47% of the workers in these
manufacturing subcategories, rispettivamente (Vechbanyongratana et al. 2021).

III. Related Literature on Education–Occupation Mismatch

With the growth in educated workforces around the world and the unintended
consequences of vertical education–occupation mismatch, several empirical studies
on the incidence and implications of a mismatch between attained and required
levels of education have been published in recent years. One of the challenges in
studying the wage impacts of vertical mismatch is how to quantify it. Hartog (2000)
summarizes three possible options as follows:

io.

systematic evaluation by
Job analysis. This method follows
job analysts such as the Dictionary of Occupational
professional
Titles published by the United States (US) Department of Labor or
recommendations of minimum required degrees by Thailand’s Ministry
of Labor (per esempio., Paweenawat and Vechbanyongratana 2015).

ii. Worker self-assessment. Mismatch is directly evaluated by workers
themselves. Surveys ask workers their opinion on the minimum

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

126 Asian Development Review

education needed to perform their jobs (per esempio., Duncan and Hoffman
1981, Sicherman 1991, Dolton and Vignoles 2000).

iii. Realized matches. This method was introduced by Verdugo and
Verdugo (1989). This study used the mean education level plus 1
standard deviation to determine the required level of education needed
to perform a job. This is then compared with the actual level of
education attained by each worker, which determines whether a worker
has the education that matches the required education for employment.
Other studies apply this method but use a modal value instead of the
mean (per esempio., Mendes de Oliveira, Santos, and Kiker 2000). Our paper
uses the modal method described here.

Duncan and Hoffman (1981) made signiﬁcant contributions to empirically
measuring the impact of overeducation on wages by introducing the overeducation,
required education, and undereducation (ORU) modello. In this model, overeducation
or undereducation is determined by the difference in attained and required
formazione scolastica. Earnings are regressed on required years of education, years of
overeducation, and years of undereducation. Using the US 1976 Panel Study of
Income Dynamics, Duncan and Hoffman (1981) ﬁnd that 46% of individuals are
perfectly matched, while 42% of workers receive higher levels of education than
required for their jobs. Inoltre, the results show that wages are determined
mainly by the required education level, and the coefﬁcient of surplus education
(overeducation) is positive and signiﬁcant. This method has been used by scholars
in several country contexts to estimate wage impacts of vertical mismatch, including
Dolton and Vignoles (2000) using British data; Hartog (2000) on the Netherlands,
Portugal, Spain, the United Kingdom, and the US; and Johansson and Katz (2007)
and Korpi and Tåhlin (2009) using Swedish data. All of these studies ﬁnd that
returns to required levels of schooling are higher than returns to surplus education,
which is consistent with the original ﬁndings by Duncan and Hoffman (1981).

Several studies regress the natural log of wages on a series of dummy
identify workers as overeducated, undereducated, or matched
variables that
educato. The expected sign on the overeducation dummy variable is negative
since it is expected that workers who are overeducated for their job would earn
less than a matched-educated worker (excluded category) with the same amount
of education. Verdugo and Verdugo (1989) pioneered this approach and found a
13% wage penalty among workers in the US. A study using Australian data by
Mavromaras et al. (2013) shows a 21.5% penalty among male workers aged 16–64
with a university degree or equivalent. Allo stesso modo, a study using data from the United
Kingdom by McGuinness and Sloane (2011) estimates a 31% A 39% wage penalty
among early career university graduates.

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Education–Occupation Mismatch and Its Wage Penalties in Thailand 127

There are two recent studies on overeducation wage penalties speciﬁc to
Thailand. The ﬁrst by Paweenawat and Vechbanyongratana (2015) analyze wage
penalties among male university graduates. The average wage penalty was found
to be 19%, but when stratiﬁed by cohort, younger workers were found to have
higher overeducation wage penalties that can be explained by an increasing supply
of young university graduates and a dearth of commensurate jobs in the market.
Another study by Pholphirul (2017) estimates both vertical and horizontal mismatch
(cioè., a mismatch between job and ﬁeld of study) using Thailand’s 2008 Labor
Force Survey. For vertical mismatches, the author uses the modal value method
to determine education–occupation matches for each worker. The author ﬁnds that
overeducated workers who completed compulsory lower secondary education or
above face on average an 18.6% wage penalty.

Despite the existence of recent studies on Thailand, no study, to date, ha
taken into consideration potential systematic differences in the incidence and
wage impacts of undereducation and overeducation across formal and informal
workers. This is important to consider since a signiﬁcant proportion of workers
in Thailand’s economy—and developing economies more generally—are in
fact informally employed and not covered by relevant labor regulations. Questo
paper adds to the literature by determining the incidence of undereducation and
overeducation and estimating wage premiums and penalties associated with vertical
informal workers.
between
education–occupation mismatch
Inoltre, this study considers the incidence of vertical mismatch and the
associated penalties and premiums across four cohorts of workers who were
exposed to different education policies and early career labor market opportunities
in Thailand’s rapidly changing economy.

formal

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

IV. Data

This study uses the Thailand SES, a nationally representative household
survey collected by the National Statistical Ofﬁce, for the years 2011, 2013,
E 2015 (National Statistical Ofﬁce 2011, 2013, E 2015). We deﬁne formal
employees as government and private ﬁrm workers who are covered by the Civil
Service Welfare Scheme, Sezione 33 under the Social Security Act (1990), O
other employer-provided welfare program.3 Informal workers are deﬁned as those
in private ﬁrm employment without employer-provided social welfare, anche
as those engaged in own-account work.4 The dataset includes observations on

3There are three schemes under the Social Security Act (1990), including Section 33, Sezione 39, and Section

40. Sezione 33 refers to employer-provided social security, while Sections 39 E 40 are voluntary schemes.

4We deviate slightly from the government’s deﬁnition of informal employment by deﬁning all own-account
workers as informally employed even if they are coded as being covered by social security (less than 4% Di
own-account workers). We are interested in workers with employer-provided protections. Own-account workers with

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

128 Asian Development Review

Tavolo 1. Thai Education Classiﬁcations Harmonized with ISCO-08 Skill
Level Classiﬁcations

Level

Description

0
1
2

Completed less than primary education
Completion of primary education or the ﬁrst stage of basic education
Lower secondary education, upper secondary education, postsecondary,

nontertiary Education (Por Wor Chor)

Higher educational institution following completion of secondary

education for a period of 1–3 years (Por Wor Sor and Por Wor Tor)
Higher educational institution for a period of 3–6 years leading to the

award of a ﬁrst degree or higher qualiﬁcation

ISCO-08 = 2008 International Standard Classiﬁcation of Occupations.
Sources: International Labour Organization (2012) and National Statistical Ofﬁce (2010).

104,137 workers who report labor income.5 A total of 53,206 workers are classiﬁed
as informally employed, of which 27,481 work in private ﬁrms and 25,725 are
own-account workers.

The workers are coded into ﬁve education classiﬁcations that are harmonized
with the ISCO-08 skill level classiﬁcations (International Labour Organization
2012). Tavolo 1 shows the National Statistical Ofﬁce’s harmonization of Thai
education levels with the ISCO-08 skill level classiﬁcations. The classiﬁcation
of overeducation, undereducation, and matched education for each individual is
based on realized matches suggested by Verdugo and Verdugo (1989) and Mendes
de Oliveira, Santos, and Kiker (2000). Following Mendes de Oliveira, Santos,
and Kiker (2000), the modal educational category (0–4) within each occupation
is used to determine “required education.” After ﬁnding the modal educational
category within each ISCO-08 occupation code at the 3-digit level, each worker’s
education level is then compared to the modal education level for their occupation
to determine whether the worker is overeducated, undereducated, or matched
educated.6 For example, if a worker completed an upper secondary diploma
(categoria 3) but works in a job that primarily employs workers with primary
formazione scolastica (categoria 1), this worker would be considered overeducated for their
current job. Tavolo 2 reports summary statistics for the sample used in this study.

Informal private ﬁrm employees and own-account workers on average have
lower levels of education, con 62.6% E 53% having completed primary school
or less, rispettivamente. This is in contrast to formal government workers of which only

social security coverage are most likely registered for one of the voluntary social security schemes (Sezione 39 O
40). The coding does not impact the results.

5For own-account workers, we use business income instead of labor income. Since own-account workers are

self-employed and do not have other employees, business income is comparable to labor income in this case.

6If there is more than one modal value, the smaller value is selected. Also, the estimations are not sensitive
to the method of constructing the vertical mismatch dummy variables. Using the median level of education in each
occupational category yields qualitatively similar results to the modal method.

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Education–Occupation Mismatch and Its Wage Penalties in Thailand 129

T
N
tu
o
C
C
UN
–
N
w
O

)
5
2
7
5
2
=
N

(

R
e
k
R
o
W

e
e
sì
o
l
P
M
E

l
UN
M
R
o
F
N
IO

)
1
8
4
,
7
2
=
N

(

M
R
io
F
e
T
UN
v
io
R
P

l
UN
M
R
o
F

)
0
9
7
,
7
2
=
N

(

e
e
sì
o
l
P
M
E

T
N
e
M
N
R
e
v
o
G

l
UN
M
R
o
F

)
1
4
1
,
3
2
=
N

(

)
7
3
1
,
4
0
1
=
N

(

l
l

S
C
io
T
S
io
T
UN
T
S
sì
R
UN
M
M
tu
S

.
2

e
l
B
UN
T

v
e
D
D
T
S

N
UN
e
M

v
e
D
D
T
S

6
1
7
1
1

8
4
4
3
1

1
1
7
,
5

0
8
4
0

4
5
4
0

6
7
4
0

6
5
4
0

4
2
4
0

9
6
4
0

9
2
2
0

5
8
2
0

2
6
3
0

1
9
2
0

8
4
3
0

5
9
2
0

5
3
2
0

6
2
3
0

5
5
0
0

9
8
0
0

5
9
4
.
0

7
3
4
.
0

3
6
4
.
0

2
5
4
.
0

4
7
4
.
0

8
5
4
.
0

7
6
1
.
0

1
1
2
.
0

N
UN
e
M

9
5
7
,
7

1
3
4
.
0

7
5
2
.
0

2
1
3
.
0

6
8
2
.
0

0
4
3
.
0

9
9
2
.
0

9
2
0
.
0

7
4
0
.
0

v
e
D
D
T
S

N
UN
e
M

v
e
D
D
T
S

N
UN
e
M

v
e
D
D
T
S

N
UN
e
M

e
l
B
UN
io
R
UN
V

9
7
2
,
0
1

0
1
8
,
4
1

6
0
2
,
5
1

5
5
8
,
1
2

5
9
0
,
2
1

8
7
1
,
4
1

l
UN
T
o
T

sì
l
H
T
N
o
M

l
UN
e
R

7
9
4
.
0

9
9
3
.
0

3
3
4
.
0

6
7
2
.
0

0
6
3
.
0

3
9
4
.
0

3
1
3
.
0

6
2
4
.
0

1
5
5
.
0

9
9
1
.
0

0
5
2
.
0

3
8
0
.
0

3
5
1
.
0

6
1
4
.
0

0
1
1
.
0

8
3
2
.
0

7
5
4
.
0

8
2
3
.
0

9
7
3
.
0

9
2
2
.
0

6
2
2
.
0

2
2
4
.
0

2
6
2
.
0

3
9
4
.
0

3
0
7
.
0

3
2
1
.
0

4
7
1
.
0

6
5
0
.
0

4
5
0
.
0

2
3
2
.
0

4
7
0
.
0

4
8
5
.
0

0
0
5
.
0

4
1
4
.
0

6
4
4
.
0

6
8
3
.
0

1
0
4
.
0

7
6
4
.
0

0
5
2
.
0

9
1
4
.
0

6
0
5
.
0

0
2
2
.
0

4
7
2
.
0

3
8
1
.
0

1
0
2
.
0

2
2
3
.
0

7
6
0
.
0

8
2
2
.
0

)
T
H
UN
B

5
1
0
2
(

S
G
N
io
N
R
UN
e

)
1
/
0
(
D
e
T
UN
C
tu
D
e

D
e
H
C
T
UN

)
1
/
0
(
sì
R
UN
M

io
R
P

N
UN
H
T

S
S
e
l

)
1
/
0
(

D
e
T
UN
C
tu
D
e
R
e
D
N
U

)
1
/
0
(

D
e
T
UN
C
tu
D
e
R
e
v
O

)
1
/
0
(

sì
R
UN
D
N
o
C
e
S

R
e
w
o
l

)
1
/
0
(

sì
R
UN
D
N
o
C
e
S

R
e
P
P
U

)
1
/
0
(

sì
R
UN
M

io
R
P

)
1
/
0
(
sì
R
UN
io
T
R
e
T

9
6
8
1
1

0
9
0
1
3

1
6
4
.
2
1

1
1
4
.
7
2

9
2
1
.
1
1

0
0
5
.
9
1

4
0
4
.
1
1

9
6
3
.
3
2

9
2
5
.
2
1

1
1
3
.
5
2

e
C
N
e
io
R
e
P
X
e

l
UN
io
T
N
e
T
o
P

8
9
4
0

7
3
4
0

7
4
4
0

4
5
2
0

7
5
4
0

5
2
4
0

7
2
4
0

4
6
3
0

0
5
5
0

3
4
7
0

3
2
7
0

0
7
0
0

7
9
2
0

7
3
2
0

9
3
2
0

8
5
1
0

3
9
4
.
0

5
6
4
.
0

4
9
4
.
0

6
1
2
.
0

0
6
4
.
0

3
9
3
.
0

2
2
4
.
0

7
1
4
.
0

6
1
4
.
0

4
8
6
.
0

6
7
5
.
0

9
4
0
.
0

5
0
3
.
0

0
9
1
.
0

1
3
2
.
0

4
2
2
.
0

0
0
5
.
0

6
7
4
.
0

9
5
4
.
0

8
8
3
.
0

0
0
5
.
0

4
1
3
.
0

1
9
2
.
0

6
0
3
.
0

3
9
4
.
0

3
5
6
.
0

8
9
6
.
0

4
8
1
.
0

8
0
5
.
0

1
1
1
.
0

3
9
0
.
0

4
0
1
.
0

0
0
5
.
0

7
4
4
.
0

1
4
4
.
0

7
0
2
.
0

1
3
4
.
0

7
2
4
.
0

5
6
4
.
0

1
6
3
.
0

0
8
4
.
0

3
2
7
.
0

6
3
7
.
0

5
4
0
.
0

6
4
2
.
0

0
4
2
.
0

5
1
3
.
0

4
5
1
.
0

0
0
5
.
0

9
5
4
.
0

6
6
4
.
0

5
8
2
.
0

5
7
4
.
0

4
9
3
.
0

1
1
4
.
0

7
6
3
.
0

4
8
4
.
0

9
9
6
.
0

0
8
6
.
0

9
8
0
.
0

4
4
3
.
0

2
9
1
.
0

5
1
2
.
0

0
6
1
.
0

)
1
/
0
(

e
l
UN
M
e
F

)
1
/
0
(

D
e
io
R
R
UN

)
1
/
0
(

N
UN
B
R
U

)
1
/
0
(

k
o
k
G
N
UN
B

)
1
/
0
(

l
UN
R
T
N
e
C

)
1
/
0
(

H
T
R
o
N

)
1
/
0
(

T
S
UN
e
H
T
R
o
N

)
1
/
0
(

H
T
tu
o
S

.
S
sì
e
v
R
tu
S
C
io
M
o
N
o
C
e
–
o
io
C
o
S
D
N
UN
l
io
UN
H
T
5
1
0
2
D
N
UN

,
3
1
0
2

,
1
1
0
2
M
o
R
F

S
N
o
io
T
UN
l
tu
C
l
UN
C

’
S
R
o
H
T
tu
UN

:
e
C
R
tu
o
S

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

130 Asian Development Review

Figura 4. Educational Attainment by Birth Cohort

Fonte: Authors’ calculations from 2011, 2013, E 2015 Thailand Socio-economic Surveys.

11% have completed primary school or less. Formal government workers are also
signiﬁcantly more likely to have completed higher education with 58% completing a
bachelor’s degree or higher compared to only 5% of informal private ﬁrm employees
E 9% of own-account workers. Così, it is not surprising that real monthly earnings
for formal workers are on average signiﬁcantly higher than for informal workers.
Formal government employees and formal private ﬁrm employees earn on average
21,855 baht (B) and B14,810 compared to B7,759 and B13,448 for informal
employees and own-account workers, rispettivamente.

Given generational differences in access to education and early career
labor market opportunities, it is instructive to see the differences in completed
education and the incidence of formal and informal employment stratiﬁed by
birth cohort shown in Figures 4 E 5. The overall picture in Figure 4 is one
of increasing educational attainment across successive birth cohorts. Among the
oldest cohort, more than half of workers completed less than primary education
E 39% completed lower secondary education or more. Among the youngest
cohort, only 2% completed less than primary education, while 85% completed lower
secondary education or higher. Figura 5 indicates that there is declining informality
across successive birth cohorts. The incidence of informality among employees
and own-account workers is highest among the oldest cohort at 61%. Tuttavia,
despite rapid industrialization and structural change in the Thai economy, the rate of
informal employment is still high among the youngest cohort at 40%. È interessante notare,
individuals in the youngest cohort are much less likely to be own-account workers
and government employees than previous generations. The youngest workers are

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Education–Occupation Mismatch and Its Wage Penalties in Thailand 131

Figura 5. Employment Sector by Birth Cohort

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

Fonte: Authors’ calculations from 2011, 2013, E 2015 Thailand Socio-economic Surveys.

Figura 6. Undereducated Workers by Birth Cohort and Employment Sector

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Fonte: Authors’ calculations from 2011, 2013, E 2015 Thailand Socio-economic Surveys.

much more likely to be employed by private ﬁrms, but the incidence of informality
among young workers in private ﬁrms is 40%.

The incidence of undereducation and overeducation for the entire sample
stands at 27.4% E 22%, rispettivamente, but differs across birth cohorts and
employment sector, as illustrated in Figures 6 E 7.

132 Asian Development Review

Figura 7. Overeducated Workers by Birth Cohort and Employment Sector

Fonte: Authors’ calculations from 2011, 2013, E 2015 Thailand Socio-economic Surveys.

The proportion of undereducated workers has declined over successive birth
cohorts for every work status, particularly for formal private ﬁrm employees,
own-account workers, and informal private ﬁrm employees. This pattern is
consistent with increasing educational attainment among the younger cohorts due to
more compulsory education and increased opportunities to complete secondary and
tertiary education. The proportion of overeducated formal government workers is
similar across cohorts. Tuttavia, the incidence of overeducated formal and informal
private ﬁrm employees and own-account workers has increased over successive
cohorts, which is consistent with increasing levels of education. Although the
youngest cohort is the least likely to be engaged in own-account work, the incidence
of overeducation among those in this group is high at 50%. Likewise, among the
30% of the youngest cohort employed informally by private ﬁrms, the incidence of
overeducation is 37%.

V. Methodology

We use an augmented Mincerian wage regression model to estimate the
overeducation and undereducation wage penalties and premiums, rispettivamente.
includes dummy variables for
We run an ordinary least squares model
overeducation and undereducation with matched education as the excluded
categoria.
ln wi = α + β1OverEdi + β2UnderEdi + X (cid:3)
The dependent variable, ln wi, is the natural log of real monthly earnings, Xi
including potential work experience
is a vector of individual characteristics,

γ + εi

Quello

(1)

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

Education–Occupation Mismatch and Its Wage Penalties in Thailand 133

(age − years of schooling − 6) and potential work experience squared; E
dummy variables for level of education completed (primary, lower secondary,
upper secondary, and tertiary); married; female; urban area; region (central, north,
northeast, and south); and survey year. OverEdi is a dummy variable that indicates
that an individual’s educational attainment is greater than the modal value of
education found in their occupation, and UnderEdi is a dummy variable that
indicates that an individual’s level of education is lower than the modal value for
their occupation.

We ﬁrst run regression (1) using the pooled sample from 2011, 2013, E
2015, and then we run it separately by employment sector. We then repeat the
analysis stratiﬁed by male and female to see whether there are any gendered
differences in overeducation wage penalties and undereducation wage premiums.
The ﬁnal analysis is stratiﬁed by birth cohorts and employment sector to see if
the overeducation wage penalties and undereducation wage premiums diverge for
individuals facing different compulsory education policies, educational access, E
early career labor markets.

VI. Empirical Findings

The empirical results for the baseline pooled regression and regressions
stratiﬁed by sector of employment are reported in Table 3. The average
overeducation wage penalty and undereducation wage premium are 20.9% E
10.2%, rispettivamente. IL 20.9% wage penalty is comparable to the previous estimate
Di 19% in the study by Pholphirul (2017) using the 2008 Labor Force Survey.
The overeducation wage penalties differ across employment sectors. The largest
overeducation wage penalty is in the formal government sector at 28.2%. The high
penalty may reﬂect the rigidity of the Thai civil service system where remuneration
is strictly tied to occupation and experience. A government worker with high
levels of education would be paid similarly with a government worker with lower
academic credentials working in the same position. A 21.8%, informal private ﬁrm
workers have higher overeducation wage penalties than formal private ﬁrm workers
(17.9%). È interessante notare, own-account workers have the lowest overeducation wage
penalties at 3.9%. This may reﬂect the nature of own-account work in which workers
are their “own bosses,” allowing them ﬂexibility to work according to their own
productivity regardless of occupation.

Tavolo 3 indicates that on average—after controlling for a full set of
covariates—women earn 19.2% less than men. The results stratiﬁed by employment
sector show that the gender wage differentials are smaller within formal work
(15.7%–17.5%) compared to informal work (22.1%–22.2%). Given that women
appear to be at a wage disadvantage compared to men, it is of interest to know
whether women and men experience different overeducation wage penalties and

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

134 Asian Development Review

Tavolo 3. Overeducation Wage Penalties and Undereducation Wage Premiums in Thailand,
Ordinary Least Squares Regressions
(1)

(2)

(4)

(5)

(3)

Dependent Variable: Ln(Monthly Labor Income)

Pooled
Linea di base
−0.209***
(0.005)
0.102***
(0.006)
0.175***
(0.008)
0.495***
(0.009)
0.866***
(0.012)
1.264***
(0.011)
−0.015***
(0.005)
−0.352***
(0.006)
−0.050***
(0.007)
0.040***
(0.001)
−0.006***
(0.000)
−0.192***
(0.004)
0.086***
(0.004)
−0.167***
(0.006)
−0.358***
(0.007)
−0.316***
(0.007)
−0.209***
(0.007)
0.096***
(0.004)
0.123***
(0.004)
0.145***
(0.004)
8.411***
(0.015)

Formal Employment
Private
Firm
Employee
−0.179***
(0.007)
0.096***
(0.008)
0.096***
(0.012)
0.447***
(0.015)
0.838***
(0.017)
1.134***
(0.016)

Government
Employee
−0.282***
(0.011)
0.133***
(0.014)
0.268***
(0.027)
0.910***
(0.024)
1.418***
(0.026)
1.907***
(0.025)

Informal Employment
Private
Firm
Employee
−0.218***
(0.008)
0.180***
(0.008)
0.186***
(0.010)
0.464***
(0.013)
0.723***
(0.024)
1.153***
(0.022)

Own-
Account
Worker
−0.039**
(0.015)
0.096***
(0.015)
0.064***
(0.019)
0.237***
(0.022)
0.376***
(0.031)
0.571***
(0.029)

D
o
w
N
o
UN
D
e
D

F
R
o
M
H

T
T

:
/
/

D
io
R
e
C
T
.

io
T
.

e
D
tu
UN
D
e
v
/
UN
R
T
io
C
e
–
P
D

0.050***
(0.001)
−0.004***
(0.000)
−0.157***
(0.007)
0.040***
(0.008)
−0.139***
(0.015)
−0.193***
(0.015)
−0.200***
(0.015)
−0.121***
(0.016)
0.169***
(0.008)
0.075***
(0.008)
0.110***
(0.008)
7.436***
(0.030)

0.028***
(0.001)
−0.004***
(0.000)
−0.175***
(0.005)
0.049***
(0.005)
−0.158***
(0.007)
−0.399***
(0.010)
−0.397***
(0.010)
−0.276***
(0.010)
0.007
(0.005)
0.184***
(0.006)
0.201***
(0.006)
8.640***
(0.019)

F
/

3
8
1
1
1
9
1
8
9
7
6
9
4
UN
D
e
v
_
UN
_
0
0
1
6
0
P
D

B
sì
G
tu
e
S
T

o
N
0
9
S
e
P
e
M
B
e
R
2
0
2
3

0.022***
(0.001)
−0.004***
(0.000)
−0.222***
(0.006)
0.083***
(0.007)
−0.309***
(0.014)
−0.495***
(0.014)
−0.486***
(0.014)
−0.326***
(0.014)
0.053***
(0.006)
0.154***
(0.007)
0.182***
(0.007)
8.504***
(0.024)
516.355
0.281
27,481

0.024***
(0.002)
−0.004***
(0.000)
−0.221***
(0.010)
0.137***
(0.012)
−0.177***
(0.019)
−0.414***
(0.020)
−0.318***
(0.020)
−0.207***
(0.022)
0.174***
(0.012)
0.084***
(0.012)
0.124***
(0.012)
8.835***
(0.043)
192.903
0.111
25,725

Overeducated

Undereducated

Elementary

Lower secondary

Upper secondary

Tertiary

Formal employee

Informal employee

Own-account workers

Potential experience

Potential experience2

Female indicator

Married

Central

North

Northeast

South

Municipal area

Survey year 2013

Survey year 2015

Constant

F-statistic
Adjusted R2
Observations

4,585.75
0.427
104,137

1,804.62
0.559
23,141

1,241.08
0.458
27,790

Notes: Robust standard errors in parentheses. ***P < 0.01 **p < 0.05 *p < 0.1. Source: Authors’ calculations from 2011, 2013, and 2015 Thailand Socio-economic Surveys. Education–Occupation Mismatch and Its Wage Penalties in Thailand 135 Table 4. Overeducation Wage Penalties and Undereducation Wage Premiums in Thailand by Gender, Ordinary Least Squares Regressions (1) (2) (3) (4) (5) Dependent Variable: Ln(Monthly Labor Income) Formal Employment Informal Employment Men Overeducated Undereducated F-statistic Adjusted R2 Observations Women Overeducated Undereducated Pooled Baseline −0.197*** (0.007) 0.125*** (0.008) 2,439.89 0.423 53,735 −0.219*** (0.008) 0.081*** (0.008) F-statistic Adjusted R2 Observations 2,471.88 0.433 50,402 Government Employee −0.257*** (0.014) 0.177*** (0.017) 1,065.54 0.544 12,025 −0.275*** (0.022) 0.106*** (0.027) 904.934 0.587 11,116 Private Firm Employee −0.212*** (0.011) 0.170*** (0.014) 636.293 0.448 14,077 −0.180*** (0.010) 0.048*** (0.010) 686.816 0.472 13,713 Private Firm Employee −0.209*** (0.010) 0.189*** (0.010) 265.740 0.250 16,056 −0.221*** (0.013) 0.162*** (0.012) 256.025 0.290 11,425 Own- Account Worker −0.033 (0.021) 0.078*** (0.023) 104.933 0.121 11,577 −0.051** (0.022) 0.109*** (0.021) 69.467 0.074 14,148 Notes: Robust standard errors in parentheses. Other controls: education, potential experience, potential experience2, married, urban, region, and survey year. ***p < 0.01 **p < 0.05 *p < 0.1. Source: Authors’ calculations from 2011, 2013, and 2015 Thailand Socio-economic Surveys. undereducation wage premiums. Table 4 reports the regression results stratiﬁed by gender. Despite the fact that women have a wage disadvantage when controlling for personal characteristics, women experience similar wage penalties and premiums as men. Overall, the wage penalty for men is 19.7% compared to 21.9% for women, while the undereducation wage premiums are 12.5% and 8.1% for men and women, respectively. The wage penalties are also similar across all four employment sectors. The similarities in overeducation wage penalties may be due in part to the fact that men and women in the Thai labor market have similar worker characteristics, including labor force participation and educational attainment. As mentioned previously, many of the oldest workers were required to complete only 4 years of compulsory schooling and entered the labor market when Thailand was just beginning its structural transformation, and it was still primarily an agricultural economy. In contrast, the youngest cohort in the sample was required to complete 6–9 years of compulsory education and had access to free education through secondary school and expanded tertiary education opportunities. Moreover, younger workers entered the job market in an economy that was much more diversiﬁed with a broader range of occupations requiring various skill levels. Because the oldest and youngest workers faced very different l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a d e v / a r t i c e - p d l f / / / / 3 8 1 1 1 9 1 8 9 7 6 9 4 a d e v _ a _ 0 0 1 6 0 p d / . f b y g u e s t t o n 0 9 S e p e m b e r 2 0 2 3 136 Asian Development Review education policies and labor market conditions, which resulted in lower incidences of undereducation and higher incidences of overeducation in younger cohorts, it is of interest to see whether older and younger workers face different undereducation wage premiums and overeducation wage penalties. Table 5 reports regression results across employment sectors and birth cohorts. Columns (1) and (2) in Table 5 show results across four birth cohorts in formal government and formal private ﬁrm employment, while columns (3) and (4) show the results for informal workers in private ﬁrms and own-account work. The results show that along with the decrease in the incidence of undereducation, the undereducation wage premium is lower for formally employed workers in younger cohorts. Similar to workers in formal employment, informally employed private ﬁrm workers and own-account workers generally have decreasing undereducation wage premiums across successive birth cohorts. The youngest generation of workers born in the 1980s, for which undereducation is rare, have no undereducation wage premiums with the exception of a small premium in informal private ﬁrm work. Despite the increase in the incidence of overeducation over successive birth cohorts, the overeducation wage penalty is lower for younger workers in formal government employment, formal private ﬁrm employment, and informal private ﬁrm employment. Since the survey data used for the analysis was collected between 2011 and 2015, we observe wages for each of the cohorts at different points within their careers. The high overeducation wage penalties in the oldest cohort and relatively low wage penalties in the youngest cohort likely reﬂect different earnings trajectories for overeducated versus matched-educated workers. For example, a university graduate who spends their career in restaurant service (overeducated) will likely have a shallower earnings trajectory than a university graduate who works as an accountant (matched educated) throughout their career. This scenario would result in larger overeducation wage penalties later in one’s career. For the youngest cohort of formal workers, the overeducation wage penalty is relatively modest at around 15%. However, the wage penalties within each cohort are higher for informally employed private ﬁrm workers than for formally employed private ﬁrm workers. This is an important observation considering that informal work in private ﬁrms continues to absorb a large number of younger workers (see Figure 5) who are more likely to be overeducated than in previous generations (see Figure 7). As for informal own-account work, there is no clear pattern across generations. Most own-account workers are employed in services and crafts and related trades (ISCO-08 occupational categories 5 and 7). Although the overeducation wage penalty is 14.5% among the oldest cohort born in the 1950s, cohorts born in the 1960s and 1970s face no overeducation wage penalties. Although only 11% of the youngest cohort is employed as own-account workers, 50% are overeducated and face a wage penalty of 9.3%. l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a d e v / a r t i c e - p d l f / / / / 3 8 1 1 1 9 1 8 9 7 6 9 4 a d e v _ a _ 0 0 1 6 0 p d . / f b y g u e s t t o n 0 9 S e p e m b e r 2 0 2 3 Education–Occupation Mismatch and Its Wage Penalties in Thailand 137 Table 5. Overeducation Wage Penalties and Undereducation Wage Premiums in Thailand by Birth Cohorts, Ordinary Least Squares Regressions (1) (4) Dependent Variable: Ln(Monthly Labor Income) (2) (3) Formal Employment Informal Employment Government Employee −0.282*** (0.011) 0.133*** (0.014) 1,804.62 Private Firm Employee −0.179*** (0.007) 0.096*** (0.008) 1,241.08 0.559 23,141 −0.456*** (0.040) 0.276*** (0.030) 377.877 0.596 5,048 −0.358*** (0.020) 0.108*** (0.023) 572.910 0.556 7,645 −0.234*** (0.017) 0.031 (0.026) 284.011 0.445 6,237 −0.151*** (0.021) −0.019 (0.026) 115.388 0.322 4,211 0.458 27,790 −0.322*** (0.055) 0.236*** (0.032) 103.247 0.535 1,562 −0.217*** (0.020) 0.122*** (0.016) 380.047 0.534 5,598 −0.175*** (0.012) 0.103*** (0.013) 475.371 0.461 10,138 −0.150*** (0.010) −0.007 (0.012) 345.815 0.388 10,492 Private Firm Employee −0.218*** (0.008) 0.180*** (0.008) 516.355 0.281 27,481 −0.338*** (0.045) 0.221*** (0.019) 91.303 0.318 4,045 −0.243*** (0.018) 0.166*** (0.013) 158.260 0.290 8,113 −0.211*** (0.013) 0.181*** (0.015) 156.470 0.272 8,231 −0.191*** (0.013) 0.131*** (0.017) 109.281 0.228 7,092 Own- Account Worker −0.039** (0.015) 0.096*** (0.015) 192.903 0.111 25,725 −0.145*** (0.045) 0.120*** (0.027) 38.519 0.096 6,348 −0.027 (0.026) 0.061** (0.024) 81.911 0.112 9,631 0.004 (0.026) 0.153*** (0.033) 48.318 0.096 7,012 −0.093** (0.038) −0.012 (0.058) 13.920 0.070 2,734 All workers Overeducated Undereducated F-statistic Adjusted R2 Observations Born 1951–1960 Overeducated Undereducated F-statistic Adjusted R2 Observations Born 1961–1970 Overeducated Undereducated F-statistic Adjusted R2 Observations Born 1971–1980 Overeducated Undereducated F-statistic Adjusted R2 Observations Born 1981–1990 Overeducated Undereducated F-statistic Adjusted R2 Observations Notes: Robust standard errors in parentheses. Other controls: education, potential experience, potential experience squared, married, urban, region, and survey year. ***p < 0.01 **p < 0.05 *p < 0.1. Source: Authors’ calculations from 2011, 2013, and 2015 Thailand Socio-economic Surveys. We acknowledge that workers are not randomly assigned to be overeducated, matched educated, or undereducated for their jobs, which could bias the coefﬁcient estimates. There are relevant unobservable factors, such as low ability or degree l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a d e v / a r t i c e - p d l f / / / / 3 8 1 1 1 9 1 8 9 7 6 9 4 a d e v _ a _ 0 0 1 6 0 p d / . f b y g u e s t t o n 0 9 S e p e m b e r 2 0 2 3 138 Asian Development Review completion from low-quality institutions, that cannot be corrected for using the existing data, potentially leading to overestimated wage penalties for overeducated persons who in fact work at their correct level of productivity. Although we cannot directly solve the ability bias in this present study, previous work on overeducation wage penalties shows that even when taking into account unobserved individual heterogeneity, the negative impact of overeducation on wages generally does not disappear. For example, Korpi and Tåhlin (2009) employ a ﬁxed-effect approach using panel data from Sweden. Their results suggest that even after accounting for unobservable personal characteristics, returns to years of education beyond what is required for the job are positive and signiﬁcant, suggesting that the ordinary least squares estimates are not merely capturing differences in unobserved ability. A study by Mavromaras et al. (2013) employs ﬁxed effect and random effect models to panel data and ﬁnds that unobservable individual heterogeneity cannot explain all of the negative impact of overeducation and overskilling among working-age Australian men. Papers by McGuinness and Bennett (2007) and Paweenawat and Vechbanyongratana (2015) use a quantile approach to show that overeducation occurs at all points along the wage and ability distribution, which suggests that overeducation is not synonymous with low ability in Northern Ireland and Thailand, respectively. Speciﬁcally in the case of Thailand, overeducated male university graduates born between 1966 and 1985 face large overeducation wage penalties at all points along the ability distribution, which is consistent with an imbalance between the number of university graduates and jobs available in the economy (Paweenawat and Vechbanyongratana 2015). Results from previous related studies give us some level of conﬁdence that our estimated coefﬁcients on the undereducation and overeducation variables are not entirely driven by the ability bias and do in fact capture in part the relationship between vertical education–occupation mismatch and wages in formal and informal employment. VII. Discussion and Conclusions Since the 1970s, Thailand has enacted a variety of policies to pursue economic development. These policies include increasing compulsory education from 4 years to 9 years, providing free education through upper secondary school and expanding higher education opportunities. The government also worked to change the structure of the economy, transforming it from a largely informal agriculture-based economy to a formalized industrial and service-based economy. While the former has resulted in dramatic increases in the average educational attainment of the populace, the latter, while diversifying job opportunities, has failed to fully formalize work, leaving the majority of Thailand’s workers still engaged in informal employment. This paper estimates the incidence of vertical education–occupation mismatch and its associated wage premiums and penalties across formal and l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a d e v / a r t i c e - p d l f / / / / 3 8 1 1 1 9 1 8 9 7 6 9 4 a d e v _ a _ 0 0 1 6 0 p d . / f b y g u e s t t o n 0 9 S e p e m b e r 2 0 2 3 Education–Occupation Mismatch and Its Wage Penalties in Thailand 139 informal employment over four cohorts of workers. It adds to the existing literature by considering the consequences of vertical mismatch in a developing country context where the labor force is largely informal. The paper also extends Pholphirul’s (2017) earlier work on Thailand by going beyond the mean wage impact of vertical mismatch on wages by taking into consideration informality and generational differences in education and early career labor market conditions. Informal workers continue to make large contributions to the Thai economy, thus understanding the interaction of vertical mismatch and its consequences within formal and informal employment is important for pinpointing potential inefﬁciencies in education and labor market policies and helping to develop potential solutions. This paper has shown that the Thai government’s education and economic policies have led to an increase in the incidence of overeducation among younger cohorts of workers, which is especially pronounced among informal workers. This implies that employment opportunities in Thailand do not match with its increasingly educated populace. Although the youngest cohort born between 1981 and 1990 is more likely to be formally employed than in previous generations, 40% of this cohort is still absorbed into informal employment, of which 41% are classiﬁed as overeducated. Overeducated informal workers in private ﬁrms face the highest overeducation wage penalties within the youngest birth cohort. Dissonance between formal job development and government education policies is an issue that policy makers in developing economies need to heed. Thailand’s current approach to education that encourages students to complete high levels of general education without the promise of formal employment commensurate with their educational qualiﬁcations incurs costs to both individuals (i.e., time costs, wage penalties, and potentially forced entry into informal employment) and society (i.e., inefﬁcient education spending and potential losses of tax revenues from unregistered employees). The government may want to consider better aligning its curriculum and degree offerings with formal job development. At present, the Thai government is focused on increasing high-skilled job opportunities. Thailand has introduced the “Thailand 4.0” policy, which is aimed at advancing the development of the country through innovation (Royal Thai Embassy 2018). As part of its strategy, the government has identiﬁed 10 target industries for development.7 One of the government’s current target industries, for example, is automobile manufacturing. The development of vocational education aimed at ﬁlling formal technical jobs within automobile manufacturing would (i) better target the amount of education an individual needs to complete, thus minimizing time and monetary costs of education and (ii) channel young workers into well-matched formal employment. If the government is successful in moving Thailand 4.0 forward 7Eastern Economic Corridor Ofﬁce of Thailand. https://www.eeco.or.th/en/content/targeted-industries. l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a d e v / a r t i c e - p d l f / / / / 3 8 1 1 1 9 1 8 9 7 6 9 4 a d e v _ a _ 0 0 1 6 0 p d . / f b y g u e s t t o n 0 9 S e p e m b e r 2 0 2 3 140 Asian Development Review and creating more high-skilled, formal employment that is commensurate with academic credentials, vertical education–occupation mismatch and its penalties would be expected to decline. Time will tell whether this or other government policies to develop more formal sector high-skill jobs will help alleviate the high incidence of informality among younger workers and allow them to earn at their potential. Finally, we acknowledge the limitations of the above analysis given the use of cross-sectional data. However, given the results from previous related research using panel data, particularly the research by Paweenawat and Vechbanyongratana (2015) that shows overeducation occurs across the entire ability distribution in Thailand, we believe our results are not entirely driven by the ability bias. In the future, we hope to extend this work and better control for individual heterogeneity by using panel data. Future work will also include an analysis by level of education, particularly differences in penalties between vocational and general education. References Aemkulwat, Chairat. 2010. “Labor Force Structure Change and Thai Labor Market, 1990–2008.” Paper presented at the Second ANDA International Seminar on Skills Development for the Emerging New Dynamism in Asian Developing Countries under Globalisation, Phnom Penh. Baker, Chris, and Pasuk Phongpaichit. 2009. A History of Thailand (2nd edition). Cambridge: Cambridge University Press. Dolton, Peter, and Anna Vignoles. 2000. “The Incidence and Effects of Overeducation in the UK Graduate Labour Market.” Economics of Education Review 19 (2): 179–98. Duncan, Greg, and Saul Hoffman. 1981. “The Incidence and Wage Effects of Overeducation.” Economics of Education Review 1 (1): 75–86. Eastern Economic Corridor Ofﬁce of Thailand. https://www.eeco.or.th/en/content/targeted- industries. Hartog, Joop. 2000. “Over-Education and Earnings: Where Are We, Where Should We Go?” Economics of Education Review 19 (2): 131–47. International Labour Organization. 2003. Guidelines Concerning a Statistical Deﬁnition of https://ilo.org/global/statistics-and-databases/standards -and-guidelines/guidelines-adopted-by-international-conferences-of-labour-statisticians/ WCMS_087622 (accessed December 2019). Employment. Informal ______. 2012. International Standard Classiﬁcation of Occupations: Structure, Group Deﬁnitions and Correspondence Tables. Geneva. Johansson, Mats, and Katarina Katz. 2007. “Wage Differences between Women and Men in Sweden—The Impact of Skill Mismatch.” Institute for Labour Market Policy Evaluation Working Paper No. 2007:13. Kaosa-ard, Mingsarn. 2002. “Development and Management of Tourism Products: The Thai Experience.” Chiang Mai University Journal 1 (3): 289–302. Korpi, Tomas, and Michael Tåhlin. 2009. “Educational Mismatch, Wages, and Wage Growth: Overeducation in Sweden, 1974–2000.” Labour Economics 16 (2): 183–93. l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a d e v / a r t i c e - p d l f / / / / 3 8 1 1 1 9 1 8 9 7 6 9 4 a d e v _ a _ 0 0 1 6 0 p d . / f b y g u e s t t o n 0 9 S e p e m b e r 2 0 2 3 Education–Occupation Mismatch and Its Wage Penalties in Thailand 141 Mavromaras, Kostas, Seamus McGuinness, Nigel O’Leary, Peter Sloane, and Zhang Wei. 2013. “Job Mismatches and Labour Market Outcomes: Panel Evidence on University Graduates.” Economic Record 89 (286): 382–95. McGuinness, Seamus, and Jessica Bennett. 2007. “Overeducation in the Graduate Labour Market: A Quantile Regression Approach.” Economics of Education Review 26 (5): 521–31. McGuinness, Seamus, and Peter J. Sloane. 2011. “Labour Market Mismatch among UK Graduates: An Analysis Using REFLEX Data.” Economics of Education Review 30 (1): 130–45. Mendes de Oliveira, Manuel, Maria Clementina Santos, and B.F. Kiker. 2000. “The Role of Human Capital and Technological Change in Overeducation.” Economics of Education Review 19 (2): 199–206. Ministry of Labour. (n.d.). Social Security Act, B.E. 2533 (1990). https://www.mol.go.th/ wp-content/uploads/sites/2/2019/07/social_security_act_2533_sso_1.pdf. (accessed April 20, 2019). National Statistical Ofﬁce. 2010. Translation of International Standard Classiﬁcation: ISCO-08. Bangkok: Sri Mueang Printing. ______. 2011, 2013, and 2015. Thailand Socio-economic Surveys. http://www.nso.go.th/sites/ 2014en/Pages/survey/Social/Household/The-2017-Household-Socio-Economic-Survey. aspx (accessed April 20, 2019). ______. 2018. The Informal Employment Survey 2018. http://www.nso.go.th/sites/2014/ /Informal_work_force/ _2561/2561_Full_Rep / DocLib13/ ort.pdf (accessed April 20, 2019). Paweenawat, Sasiwimol, and Jessica Vechbanyongratana. 2015. “Wage Consequences of Rapid Tertiary Education Expansion in a Developing Economy: The Case of Thailand.” The Developing Economies 53 (3): 218–31. Pholphirul, Piriya. 2017. “Educational Mismatches and Labor Market Outcomes: Evidence from Both Vertical and Horizontal Mismatches in Thailand.” Education + Training 59 (5): 534– 46. Royal Thai Embassy, Washington, DC. 2018. National Strategy Thailand 4.0 Ofﬁcially Launched. https://thaiembdc.org/2018/10/22/national-strategy-thailand-4-0-ofﬁcially-launched/ (accessed April 19, 2019). Sicherman, Nachum. 1991. “Overeducation in the Labor Market.” Journal of Labor Economics 9 (2): 101–22. Vechbanyongratana, Jessica, Yong Yoon, Nuarpear Lekfuangfu, and Peera Tangtammaruk. 2021. “The Socio-economic Complexities of Formalisation of Women’s Employment in Thailand.” In Informal Women Workers in the Global South: Policies and Practices for the Formalisation of Women’s Employment in Developing Economies, edited by Jayati Ghosh, 178–214. London and New York: Routledge. Verdugo, Richard R., and Naomi T. Verdugo. 1989. “The Impact of Surplus Schooling on Earnings.” Journal of Human Resources 24 (4): 629–43. World Bank. World Development Indicators. https://databank.worldbank.org/source/world- development-indicators (accessed May 2019). l D o w n o a d e d f r o m h t t p : / / d i r e c t . m i t . / e d u a d e v / a r t i c e - p d l f / / / / 3 8 1 1 1 9 1 8 9 7 6 9 4 a d e v _ a _ 0 0 1 6 0 p d . / f b y g u e s t t o n 0 9 S e p e m b e r 2 0 2 3
Scarica il pdf