Trade unions in the south and co-operation between unions in the South and in the North: A survey of the economics literature

285
Trade unions in
the south and co-
operation between
unions in the
South and in the
North: A survey
of the economics
literature
Mari Kangasniemi
Jukka Pirttilä

¡ ¢ £ ¡ ¤ ¥ ¡ ¡ ¦ § ¨ ¤ © © £ § ¥ ¢ ¡ § © ¥ © ¡ ¨ ¨ § © ¡
! # $ % # ! ' ( ' ) $ 0 # ) ' % ' ) 1 2 3 # % ) ! % % # $ 4 4 ' %
5 6 7 8 9 @ A 9 B C D E
F G H I P P Q R S T U V P U V P W X Y Y I P P I X ` a S b T Q H a G c V S a d ` I X ` e X f I S V c I Q T g a ` Q c a X h i I ` f V ` S p
q
a Q H V ` r Q H a c a s c a P a ` Q V Q I t a P X h Q H a g a ` Q c a u a P s a W I V f f T v I I Q Q V e X t a c I u h X c R P a h R f P R w w a P Q I X ` P p
G H a t I a U P a x s c a P P a S V c a Q H X P a X h Q H a V R Q H X c P V ` S S X ` X Q ` a W a P P V c I f T c a y a W Q Q H a t I a U P X h
Q H a G c V S a d ` I X ` e X f I S V c I Q T g a ` Q c a X h i I ` f V ` S p € ` T a c c X c P c a Y V I ` X R c c a P s X ` P I b I f I Q T p
F F V c I ‚ V ` w V P ` I a Y I ƒ „ V b X R c … ` P Q I Q R Q a h X c † W X ` X Y I W v a P a V c W H ‡
F F F ˆ R r r V ‰ I c Q Q I f ƒ d ` I t a c P I Q T X h G V Y s a c a ‡

‘ ’ “ ” • ” – — – ˜ ‘ • ™ d e
f ˜ g h i g g ˜ j k ” • ” – — g ™ d e
” l m n l o p l l q r s o t u t n r v u p m l r t w p
” r t n x o p r m m l o y l o t l z • { | | e z | } s m p r o n r
” u ~ | € ™ e z e ‚ z z |
g x ~ n ƒ „ w p t r … s t u o r v r p u n u o r v r † m l ‡ w u y ˆ
‰ l ‡ w u y ˜ o p t r t u t s Š w y – ‹ w o w v r ‹ — s p s l y ‹ ~
” r t n x o p r m m l o y l o t l z • { Œ ˜ | | e z | } s m p r o n r { Œ r o m l o Ž
‘ s m s „ ~ w o s z e d € ™ e z e ‚ z z |
– v l r m … ˆ y p t o l v s p u y o l v s † m l ‡ w u y ˆ
˜ g k € ‚ d ‘ € e ™ ‘ ™ | € ‘ ’ ’ e ‘ “ ” „ Ž Š •
˜ g g k ’ ‚ € e ‘ ’ d | ’ ” „ Ž Š •

1
TIIVISTELMÄ
Tässä kirjallisuuskatsauksessa tarkastellaan viimeaikaista taloustieteellistä tutkimusta ammattiliitto-
jen roolista kehitysmaissa ja pyritään selvittämään tulevaisuuden tutkimustarpeita. Ammatillisen
järjestäytymisen vaikutuksesta palkkaan ja työllisyyteen on paljon sekä teoreettista että empiiristä
tutkimusta kehittyneiden maisen osalta, ja jonkin verran myös kehitysmaista. Tavanomaisten
palkka-, työllisyys- ja tulonjakovaikutusten lisäksi järjestäytymisellä voi kehitysmaissa olla kuiten-
kin vaikutuksia myös esimerkiksi työoloihin, erityisesti jos valtion mahdollisuudet valvoa niitä ovat
heikot. Kehitysapua kanavoidaan monissa maissa myös ammattijärjestöjen kautta. Tämän toimin-
nan vaikuttavuudesta ei ole juurikaan järjestelmällisiä tilastoaineistoon perustuvia evaluaatioita.
Tutkimuksen tulisikin jatkossa pyrkiä täyttämään näitä aukkoja tietämyksessä.
ABSTRACT
This paper offers a survey of the recent economic literature on the effects of unions in developing
countries, with a particular emphasis on the impact of unionisation on labour standards. The paper
also discusses the studies – or, more aptly, the lack thereof – of the impact of development aid that
is channelled via unions. The observations point towards important areas for rigorous further
empirical research.
Key words: trade unions, development, civil society, labour standards, development aid.
JEL classification: J51, O15, O22.
1 INTRODUCTION
There is a considerable degree of variation in the institutional set up of labour markets between
countries. For several decades, studies of the impact of these institutions on issues such as wages,
productivity, macroeconomic performance, unemployment, and wage dispersion have been popular
topics in mainstream economics. There is, however, much less research on similar linkages in the
context of developing countries. Institutions and their quality in general are, however, central to the
debate on what the most important barriers are to unleashing countries’ development potential.
Labour markets are not irrelevant to economic development. Not only are they essential for the
majority of people who draw most of their income from work, but also the right to decent work is

2
seen as crucial for human well-being as well as dignity. The most recent World Development
Report (2012), for example, is centred on jobs and their instrumental role in development. The
World Bank also emphasises in the Report that while excessive regulation of labour markets (by
union power, for example) is detrimental to development there is a wide ‘plateau’ of different
labour market arrangements that can all be compatible with economic growth. Therefore, the
impacts of unionisation as such are not necessarily harmful or beneficial; we need to understand the
specific circumstances to know what works and what does not work in a specific country and
institutional setting.
On the other hand, it is widely recognised that promoting civil society organisations is essential for
enhancing the quality of governance so badly needed in many developing countries. Trade unions
are a form of civil society organisations, and indeed the rights to associate and bargain collectively
are considered as core labour standards along with the prohibition of forced labour and child labour.
Therefore, the impact of collective bargaining is also in a sense an inevitable consequence when the
core labour standards are recognised. This further emphasises the importance of studying how the
impact arises and how the positive influence of labour standards could be enhanced.
Many unions in the developed world support trade unions in developing countries. Governmental
aid agencies also often provide funds for this work. Like development aid in general, even in the
presence of potentially beneficial effects, supporting labour unions in developing countries gives
rise to the question as to how effective a way it is to promote the goals of international aid. In
addition to the impacts (that may be negative) on employment, economic growth, trade and
investment unions can, however, serve as a means of promoting well-being directly. This could
happen through better working conditions, improvements in social dialogue or strengthening the
civil society. The total effect or the means to maximise the benefits and minimise the harm requires
further information on the impacts of unions in the conditions that prevail in developing countries.
The aim of this survey is to establish the current state of knowledge on the role that trade unions
play in developing countries and to discuss how they could be used to promote positive
developments. The emphasis is strongly but not exclusively on the economic impacts or impacts
that can be measured. This naturally leaves out some of the potential effects of unions (such as the
intrinsic value of the right to form and join unions and much of the dynamism that a vibrant civil
society can create).

3
There is a large literature on the economic impacts of collective bargaining in industrialised
countries. Most of the earlier literature is summarised in Aidt and Tzannatos (2002) and we briefly
touch upon it. However, the main focus of this survey is in the more recent literature and what is
known about the impacts in developing countries. In writing this report, we have relied extensively
on the excellent survey by Freeman (2010) and one of the background papers for the World
Development Report of 2013, Betcherman (2012), which also draws heavily on Freeman’s work.
Our novelty is that we (i) focus somewhat more on the notion of ‘decent work’ and labour standards
and, in particular, (ii) we survey the small amount of research work that has been conducted about
the development collaboration between unions in the developing countries and in the rich countries.
Before setting out to review the literature in detail, we have to discuss how we can classify the
impacts as “good” or “bad” from the point of view of the countries themselves or the aid providers.
Generally, developing countries need economic growth to improve their standards of living and
reduce poverty. Factors such as efficiency and productivity contribute directly to the level of
income in the country. Employment, however, is the most obvious way out of poverty and generally
should not be considered just as a consequence of growth (World Bank 2012). In the globalised
world exports and foreign direct investment are also dependent on the quality and costs of
production that a country can provide.
However, distributional effects are also of major importance. In many developing countries, the
distribution of income is very uneven, and the high poverty rates are partly a result of this. There is
an ongoing debate on the implications of the overall distribution of income: in general it is widely
accepted, however, that high poverty rates are undesirable for many reasons. The ultimate goal of
development aid as well as policymaking in general is to improve the well-being of the citizens of
developing countries. Both income and enhancing labour standards contribute to this aim. Similarly,
promoting civil society, democracy and institutional quality are necessary, both because they can
accelerate economic growth and because they directly improve well-being.
Naturally, in many cases it is not possible to improve one aspect of the economy or society without
simultaneous damage somewhere else. For example, a more even distribution of income may
require the causing of distortions through higher taxation. Given that we cannot calculate the total
sum of well-being directly, if we can only attain one goal at the expense of the other, the burden of
weighing costs and benefits observed falls to the policy-maker. The dynamic effects, however, may
also differ from the immediate effects.

4
We approach the issue from the economics point of view: even if the goal itself is not completely
measurable, our focus is on finding and reviewing empirical evidence on links that are observable
and significant in data. Even when such associations are found in the data, proving causality can be
problematic. In many of the macro-level studies we review, it remains doubtful as to which one of
the factors causes the other to a larger extent, or whether they are both a result of some underlying
process. This problem is particularly prominent when one is studying phenomena where also the
outcome itself is difficult to measure or even to define in measurable terms. Thus, studying topics
such as the relationship between social capital, the strength of civil society, participation and
economic growth, especially establishing causality, is difficult if these methods are used. Studying
or drawing inference on the magnitude of more complex long-term influences such as the impact of
unions on the economy through the fortification of civil society would be even more difficult. In
this review we have not attempted to cover work on all these phenomena but to concentrate on the
influence through simpler mechanisms.
The experimental approach described in Section 5 does not suffer from these problems. However, it
is not possible to apply this approach to all issues, and the results cannot necessarily be generalised
for other contexts. The strengths and weaknesses of this approach are discussed, and we consider
how experiments could be used to examine the effectiveness of development assistance given
through trade unions.
The structure of this survey is as follows. In Section 2 we discuss the impacts of collective
bargaining implied by the economic theory, which has been mainly written from a developed
country perspective. In Section 3, we discuss the differences that a developing country context
presents for this theoretical analysis. Section 4 reviews the existing empirical evidence on the
impact of unions in developing countries on various outcomes, including wages, unemployment,
working conditions and also institutional development at the macro level. Section 5 presents the
current knowledge on the impacts of development collaboration between unions in the north and
south. Section 6 concludes.

5
2 THE ECONOMIC IMPACTS OF COLLECTIVE BARGAINING
2.1. Theory
In the standard economic textbook model, a trade union acts as an entity that bargains over wages
on behalf of the workers (see e.g. Boeri and Van Ours 2008, Naylor 2003 or any textbook on labour
economics). In the simplest model, the union is a monopoly that can set a wage unilaterally.
Effectively, the union sets the wage at the level where the marginal benefit to the union of raising
the wage is equal to the marginal loss. The marginal benefit and loss depend on the preferences of
the union and the wage elasticity of demand for labour. Given the wage, the firm decides how many
workers it wants to employ. The supply of labour is assumed to be flat: as long as the wage exceeds
the reservation wage or the wage required to accept work, each firm can hire as many workers as
they like, or in a competitive equilibrium up to the point where the value of the marginal product of
a worker equals their wage which maximises the profit of the firm. The union, on the other hand,
considers how many individuals in total are employed (and, subsequently, unemployed) and how
high their wage is compared with the reservation wage. The more elastic the demand for labour is,
the less the union can afford to raise the wage, as higher elasticity will result in a larger decline in
employment and higher unemployment.
Usually, it is unrealistic to assume that a union unilaterally sets the wage. Normally, unions and
employers’ organisations bargain over it (the right-to-manage model). Individual employers then
take the wage as given. The two parties have different degrees of bargaining power. When unions
have all the power, the result is equal to that obtained in the monopoly union model. Generally,
bargaining power, however, is more evenly split between employers and unions. As in the
monopoly situation, the resulting wage is higher than that attained in the competitive market but
lower than the wage in the monopoly union model. The amount of relative bargaining power
influences the outcome: the higher the bargaining power of the unions is, the closer is the outcome
to that of the monopoly union model; and the higher the bargaining power of firms is, the more it
resembles the outcome of the competitive labour market.
The model can be extended so that the unions and employers bargain over both wages and
employment; this is so-called efficient bargaining. In this case, the profit and union utility levels
can, in fact, be superior to those in the models where only wages are bargained over. The outcome
depends again on the relative bargaining power of the parties, and the stronger the competition in
the product market is, the less scope there is for bargaining.

6
The bargaining model can be extended, for example, by assuming that union membership depends
on the characteristics of the workers and the individual gains from membership. Similarly, the
preferences of the union become more complex, if we do not assume that workers are homogenous
and that decisions within the union are made by means of a ballot box, for example. Under certain
assumptions the union’s objectives coincide with the median member’s preferences. Unions may
also provide insurance for example against income fluctuations, and protection for low skilled
workers. In the case of developing countries where the enforcement of legislation (concerning, for
example, the minimum wage) is weak and social security largely absent we can hypothesise that the
insurance function of trade unions is important. On the other hand, as will be discussed in the next
section, organisations called trade unions can also engage in actions that have little to do with the
protection of workers.
As summarised by Kaufmann and Bennett (2007), the costs of unionisation in the monopoly model
compared with a competitive outcome arise from allocative inefficiency, technical inefficiency,
reduced capital investment and higher firm deaths. However, the impact of unions also depends on
the alternative or the degree of competition without unions. The assumption of a non-competitive
labour market or monopsony, where there is only one buyer of labour (for example, a single large
factory in a small town) would actually imply that a union increases allocative efficiency.
A key point to remember is that these effects of the unions are compared with the outcome of a
perfectly competitive labour market. However, employers also have market power and are to some
degree monopsonists (Manning 2003). The presence of a monopsonistic firm reduces wages below
their competitive level, and in such a situation a minimum wage, negotiated for example by a union,
can raise the wage closer to the competitive one. The evidence in Brummond (2012) suggests that
in one particular developing country, namely Indonesia, firms have a significant degree of market
power in the labour market. Monopsonists may, therefore, be much more prevalent in developing
country contexts.
Furthermore, there may be other sources of imperfections in the labour market. These include the
lack of information, mobility costs or incomplete contracts. Though undoubtedly the aim of trade
unions, in most cases, is to push wages higher, their activities may have other goals or
consequences which may reverse the conclusion about their impact on efficiency. In their seminal
work (1984) Freeman and Medoff presented an alternative view. They emphasised the two faces of
unions. In addition to the efficiency-reducing monopoly face the union ‘voice’ is an important
factor that, in fact, enhances efficiency, presuming there are imperfections such as imperfect

7
information or mobility costs. The union provides an opportunity for individual workers to express
their concerns about workplace issues without the risk of losing their job or being mistreated. As a
consequence, not only do workers benefit but also the employee does. Without the union as a
discussion platform, unhappy workers may quit their job, whereas with the union voice the
employer will be better informed about worker preferences and avoid unnecessary and costly
turnover. Similarly, through the union the employer may receive information about how the
production process could be organised more efficiently. At best, unions may also provide labour
market information, monitor and evaluate. It can also be argued that unions facilitate the drawing up
of contracts that are more efficient, and the reduction of transaction costs compared with the
situation where bargaining takes place at the individual level. The “worker voice” may thus imply
higher overall productivity.
Labour standards, in particular, can be seen as a public good which can be promoted by using the
‘voice face’ of unions, as discussed above. Public good, by definition, is a good that provides utility
for a large number of people; however, the cost of obtaining the good is higher than the benefit for
any individual person, so obtaining the good requires aggregating preferences through an
institution. For an individual, the marginal cost of demanding higher standards is too high compared
to the individual benefit. Thus, attaining efficiency requires aggregation of individual preferences
through a trade union. In the context of developing countries the voice face may be particularly
significant, as the enforcement of labour standards or worker rights by the authorities is weak.
In the spirit of Freeman and Medoff, Donado and Wälde (2010, 2012) emphasise the role that
worker movements have played in improving occupational safety to the levels that are now
generally accepted in high income countries but would not have been attained without worker
movements. For an individual worker, it is impossible to recognise whether an illness is work-
induced, but a group like a trade union can gather information on a large number of cases and
establish a connection. In their model, higher levels of occupational health also ultimately provide
better returns on investment. According to their analysis, improving occupational health in both
developed and developing countries is welfare-enhancing and mutually beneficial.
Freeman and Medoff found that their theory was corroborated by empirical facts in the US
economy at the time. Unionised workplaces indeed operated under different and more explicit rules.
The threat effect improved wages and conditions in non-union firms too. Unions also increase the
share of compensation that goesto employee benefits, and shift the composition towards deferred
benefits, and reduce wage inequality. Unionised workers are less likely to quit. During recessions

8
unionised firms more often use temporary layoffs and sometimes agree on wage cuts. Union
workers are, however, more dissatisfied, although unionised workplaces also have higher
productivity. Unionised firms provide lower return on capital and reduced profits, but reduced
profits mainly occur in the concentrated sectors of the economy, where profits are generally above
the normal level.
On the other hand, there are several reasons why trade unions may fall short of the optimal outcome
even when there are potential gains from unionisation. Unions may have other objectives that
override the improvement of labour standards or productivity, or they may only serve the interests
of insiders at the cost of outsiders. Unions may be highly adversarial and have high transaction
costs, as there are limited channels for worker voices outside formal union channels in the
traditional form of unionism, which ultimately leads to a rejection of unionism. This has happened,
for example, in the United States (Hirsch 2007). The distributional effects of unions may mainly
benefit the unionised workers, unless the ‘threat’ effect of unionisation is very strong. In a
developing country context this problem is acute, as there is typically a large informal labour
market which is not covered by regulations or formal institutions.
It can also be argued that the positive voice effects cannot exist without the negative ‘monopoly
face’ (Kaufman and Bennett 2007). In order for a union to effectively promote positive objectives it
will have to possess some monopoly power. Even if a positive effect of unions is observed in the
empirical data, it may hide a large negative effect. The institution that serves as the voice of the
workers does not necessarily need to be a union in the traditional sense. There may be alternative
institutional forms that are capable of attaining similar positive outcomes. As pointed out above, the
modes of union behaviour as well as the degree of co-operativeness in industrial relations vary a lot.
Globalisation has further heightened the worries regarding the impact of wage pressures arising
from union activity. In an open economy, higher wages and inflationary pressures may, in
particular, impact on trade. For foreign direct investment, in addition to cost structures, conflict-
prone industrial relations may serve as a deterrent for investors looking for a predictable
environment for production.
On the other hand, some elements of the role that trade unions play in the light of economic theory
may be more relevant in developing countries. As pointed out above, the voice effect may be
particularly important because of the low general enforcement of labour standards, whereas the
protection provided by unions can serve as insurance when overall social security or social

9
protection policies are largely missing or largely incomplete. It has to be noted that this does not
necessarily imply that overall unions in developing countries are optimal labour market institutions.
Union membership may be the cause of another source of inequality if it is not accessible to
everyone and, on the other hand, other policies or forms of labour market institutions may provide
even better outcomes.
Though the economic effects are important, it has to be noted that the freedom of association and
collective bargaining are recognised as basic rights and as one of the core labour standards. As such,
it is hardly an option to banish unions in the developing world, even if they have negative effects on
trade or growth. This is linked to the general question on implementation and enforcement of labour
standards. The literature on labour standards is discussed further below. In a broader sense, the
freedom of association is also a characteristic of a well-functioning civil society and democracy.
These are overarching institutions whose importance in development has long been recognised (see,
for example, Rodrik et al, 2004 and Hoff and Stiglitz, 2001).
2.2. Empirical evidence from developed countries
As mentioned in the introduction, there is a large amount of research on the general impacts of
collective bargaining. In their extensive survey, Aidt and Tzannatos (2002) review this evidence,
which largely concentrates on studies on industrialised countries. Their main observations are that
there is little systematic difference in economic performance between countries with difference
forms of collective bargaining. Micro studies are often context-specific, whereas the
macroeconomic effects are hard to disentangle. According to Aidt and Tzannatos the existing
studies suggest that there is an observable wage mark-up and compression of wage distribution, and
the return to skill is reduced. The turnover is lower and tenure, on average, is longer in unionised
firms, and there is more training. On the downside, unionisation lowers profits, employment growth
and investment. The impact on productivity is indeterminate.
At the macro level, Aidt and Tzannatos conclude that bargaining coverage is generally related to
higher real wage growth, lower employment, higher unemployment and higher inflation and there is
a negative association with earnings inequality. However, the degree of co-ordination has often
been stated as a significant determinant of the overall effect. High union density and bargaining
coverage do not have as negative an impact when bargaining is highly co-ordinated. Betcherman
(2012) presents a shorter review including some more recent work, and argues that there is no

10
consistent evidence on whether different bargaining structures help or hurt employment. The lack of
consensus may be partly due to difficulties in measuring structures and practices.
Similar arguments have been presented by others. Freeman (2007) points out that the main channels
through which labour market institutions have an impact on economic outcomes are incentives,
efficient bargaining, information, communication and trust. According to Freeman the main impact
they have is on the distribution of income; the other effects are much less clear. Similarly, Traxler
and Brandl (2011) study the economic impact of collective bargaining coverage with OECD data.
The results suggest that collective bargaining has little implication for macroeconomic outcomes,
except for income inequality in the case of centralised bargaining.
We are interested in whether similar effects can be found in developing countries. In the following
we first discuss some special features of labour markets and the state of collective bargaining in
developing countries. Later on, we present some existing evidence from studies of developing
countries.
3 SPECIAL FEATURES OF DEVELOPING COUNTRY LABOUR
MARKETS
Informality
Informal labour markets are a distinctive feature of developing economies. The definition of the
informal sector is subject to discussion (see, for example, Kanbur 2009), but in general the informal
sector means activities that are outside formal regulation, either because of their small scale or the
casual nature of production or because of intentional evasion of regulation.
The role that the informal labour market plays is also subject to theoretical and empirical research.
While informality undoubtedly allows, for example, the avoidance of even reasonable forms of
regulation, it may be that informality only represents a transient period on the workers’ and firms’
ladder to the formal economy. Informality is also a heterogeneous phenomenon, including both
voluntary and involuntary informality, and diverse types of workers, such as workers working on
their own account, unpaid family workers and people working as employees for informal companies
and informal employees of formal companies. The share of voluntary and involuntary informal
employment varies across countries.

11
It can be argued that informality is caused by regulation, for example labour market institutions,
which is too rigid and cumbersome for the conditions prevailing in developing countries: the costs
of following the rules can exceed the potential benefits of formalisation and the cost related to the
risk of not abiding by all regulation. Freeman (2009), however, points out that the share of workers
in the informal sector increased in the 1990s and the first decade of the 21st century irrespective of
the strength of economic growth or the degree of regulation. This alone suggests that the underlying
process is not related to regulation alone. There is a large literature on labour market informality,
but we shall only review a few studies to present an overall picture of the issues.
LaPorta and Shleifer (2004) summarise the usual perceptions of the informal sector in three views:
the romantic view, the parasitic view and the dual economy. In the romantic view the informal
economy consists of productive or potentially productive activities that are held back by the burden
of bureaucracy. The parasitic view emphasises their illegal nature and the unfair advantage they
gain by avoiding taxes and regulation, and sees them as hindering economic growth. The dual view
considers the informal firms as the less productive sector of the economy that does not really
compete with the formal sector and, by and large, serves different customers and hires different
employees. The authors find that informal firms are actually small and unproductive and the growth
comes from the creation of productive, formal firms. Therefore, they consider the dual view as the
most realistic perception of the informal sector. Though their work concerns companies, a similar
pattern seems to be prevalent in labour markets overall: informal work is linked to poverty and in
many instances has low productivity (OECD 2009). Kucera and Roncolato (2008) point out that the
distinction between involuntary and voluntary informal employment is not meaningful as long as
informal workers live in poverty.
In the case of labour standards, informality can be a hindrance to enforcing them. Galli and Kucera
(2004) state that, for example in Latin America, countries with higher labour standards tend to have
higher shares of formal employment, given the other explanatory factors. This is not necessarily a
causal relationship, and does not imply that exogenous changes in civic rights automatically
increase the share of formal employment. It may also highlight the fact that attaining high labour
standards (or social security coverage) may be obstructed by high levels of informality.
In line with the “parasitic view” the best way to reduce informality is to enforce regulation more
efficiently, whereas if regulation is perceived as the main cause of informality, stricter enforcement
may only lead to more sluggish economic growth. Brazil is one of the few countries where there has
indeed been a significant reduction in the share of informal labour. According to Almeida and

12
Carneiro (2011), stricter enforcement in Brazil led to an increase in formal sector employment and
non-employment and a decrease in employment in the informal sector. This is mainly due to
reduction in self-employment. The World Bank (2012) emphasises that various means have to be
used to combat informality, as the causes for informality are also diverse, ranging from better
enforcement to simplifying regulation and strengthening the social contract.
As far as distributional effects are concerned, it is clear that whatever benefits may be obtained
through trade unions, they are largely beyond the reach of informal workers and firms. As the
evidence shows, rather than seeing informality as a result of rigid labour market institutions, the aim
should be to provide the informal sector with institutions that can provide a “voice” for similar to
what the unions in the formal sector provide in order to reap the benefits of “worker voice”. A well
known example of this is the Self-Employed Women’s Association (SEWA) in India, which has
operated since 1972. SEWA represents its members at various levels (local council, municipal, state
and national planning bodies, advisory boards etc.) and also provides training, social services and
collective bargaining (The World Bank 2012). In a recent overview report to the Solidarity Center
(Schurman and Eaton 2012) it has been found that there are many ways of organising informal
workers: in addition to traditional trade unions there are membership-based organisations,
community-based organisations, co-operatives, NGOs, networks and alliances.
Urbanisation
Freeman (2009) points out that earlier “the urban bias” was seen as a major problem in developing
countries. Institutionally determined urban wages that were too high drew population from the
countryside, resulting in high unemployment in urban areas (the Harris-Todaro model). Labour
market institutions that set the wage in the formal urban sector above the market-clearing level were
often seen as a hindrance to attaining full employment and sustainable development1
. The original
model did not specify exactly which institutions set the urban sector wage. For example, Calvo
(1978) formulated a model based on the Harris-Todaro model where trade unions act as the wage-
setting or bargaining agent for the urban sector.
However, in modern times the views are more varied and the impact of labour market institutions is
known to be dependent on the context as well as on the exact features of the particular institutions.
Fields (2003) finds that, even in the original Harris-Todaro model, restraining the modern sector
1
Though simplified, the model also highlights the importance of paying attention to all sectors of the economy, which
is clearly still of major importance (Fields 2007).

13
wage does not unambiguously increase welfare. In addition to the rural-urban divide, there are other
characteristics of developing country labour markets that may have strong implications for the
impact of institutions. One of them is the presence of informal labour markets. This is linked to the
lack of official social security systems as well as to the lack of the enforcement of policies and
regulation such as the minimum wage. In general, we also have to consider what the role and the
objectives of trade unions are in developing countries.
Poverty alleviation
The first of the Millennium Development Goals is poverty eradication. What are the likely impacts
of unionization on poverty? While standard economics models would predict that union presence
led to involuntary unemployment, unions also tend to raise wages, and therefore the net impact of
unions on the extent of poverty is ambiguous. Fields and Kanbur (2007) examine the impacts of
minimum wages on poverty, and although they do not discuss the unions’ role directly, one could
think that unions could be one of the reasons behind the minimum wage. They study both
conventional labour markets and cases with the presence of income sharing between those who
work and those who end up unemployed because of the minimum wage. Interestingly, unions could
be one means to administer such income sharing. The authors demonstrate that minimum wages can
either decrease or increase overall poverty and they also present the precise conditions –which
depend on the level of the minimum wage relative to the poverty line, the elasticity of labour
demand, the extent of income sharing between the employed and the unemployed and the poverty
measure used -- under which these effects materialise.
How different are unions in developing countries?
In addition to the existence of informal labour markets, the existing bargaining institutions in
general are weaker in the developing world (Freeman 2010). According to several indices
developing countries have a lower reliance on collective bargaining (World Bank 2012). Data on
collective bargaining suffers from the lack of standardisation, although ILO has put some effort into
computing consistent estimates of union density and collective bargaining coverage (Hayter and
Stoyevska 2010). The figures calculated by ILO for a number of countries indicate that the coverage
of collective bargaining of total employment is as low as 1.1% in countries such as Bangladesh and
the Philippines, whereas in Argentina and Brazil it is 60% of wage and salary earners. Thus, there is
a large variation in the strength of collective bargaining.

14
In an overview (2008) of the application and realisation of rights concerning the freedom of
association and collective bargaining, the ILO emphasises that there is still a need to increase efforts
to promote respect for these rights. According to the ILO there have been positive developments,
but also an increase in complaints concerning anti-union discrimination and interference, and
government intervention is a recurrent problem. Several countries continue to exclude categories of
workers from the right to bargain collectively. According to the ILO, corporate social responsibility
initiatives need to be developed with the participation of trade unions, and effort needs to be put
into organising informal workers. The ILO World of Work Report (2012) also states that in many
middle-income economies industrial relations have recently been reinforced, although overall there
is also evidence of the decline of unionisation. In countries such as Argentina, South Africa and
Viet Nam dialogue has helped with dealing with the crisis.
The character of unions and the role they play varies widely from country to country. In some cases
unions are major players in the field of national politics,; in others they are little more than
extensions of the company management and do not constitute what is normally perceived as a trade
union. Visser (2003) notes that for the unions’ bargaining power it matters whether they are
concentrated or fragmented, united or rivals, subservient to other interests, such as employers or
political parties, or independent, undemocratic or democratic, and what the managerial quality of
the leadership is. The composition of the membership is naturally important. Finally, the labour
laws and other institutions within the country as well as conditions in the product market are highly
significant in determining the strength and the characteristics of a union.
It is not possible within the scope of this paper to provide a full description and classification of
trade unions in all developing countries. However, some general features of unions in various parts
of the world can be presented.
In Asia there is great variety in the activities that unions perform and the power they hold, largely
due to historical reasons. Unions in China, Viet Nam and Singapore have a strong political voice
but little influence at the workplace. In China and Viet Nam the unions are effectively monopolised
by central bodies related to the Communist parties of each country, respectively. For example, in
China the unions are largely a result of a top-down approach to unionisation, where the All China
Federation of Trade Unions serves as a link between the party-state and trade unions. According to
Liu (2011) trade unions in China have not been able to prevent severe breaches of occupational
health and safety regulations. Liu suggests reforms to the trade union law to improve unions’
possibilities to protect workers’ basic rights. Lee and Liu (2011) argue that change is leading to

15
more democratic unions. In Singapore, on the other hand, the ruling People’s Action Party has a
partnership with the National Trade Union Congress, and in this way trade unions have a voice in
the policy-making process, but at the workplace level the influence of trade unions is severely
limited due to legal restrictions. (Yoon 2009.)
In Cambodia, Indonesia, the Philippines and Thailand trade unions are highly fragmented. In
Indonesia total union membership is less than 4%. The Philippines, on the other hand, has 145
competing labour federations that are engaged in unionism across sections. Cambodia recognised
basic human rights in 1993, including the freedom of association. There were no trade unions when
the civil war ended in the mid-1990s. The garment sector, however, now has unionisation rate of
45% (Yoon 2009). Unions have played an active role in ensuring compliance with the ILO core
labour standards. The number of unions has increased over time but the number of strikes has not
(Neak and Robertson 2009). In India, trade unions have a long tradition dating back to colonial
times. They tend to focus, however, on advancing the interests of employees in narrow formal
sectors and are linked to political parties (Visser 2003).
In Africa the attitude of the colonial regimes towards unionism varied. Unions were heavily
involved in the struggles for independence and later in the politics for democratisation, thus perhaps
representing the type of union tightly linked to national politics. Currently, informality is a major
challenge in Africa, and some manufacturing sectors face stiff competition from Asia. Trade unions
in Africa have been losing members, partly due to the privatisation of public enterprises (Kocer and
Hayter 2011). The collective bargaining coverage rate is as high as 17% in South Africa, but in
many African countries it is only a few per cent (Hayter and Stoevska 2011).
As far as Latin America is concerned, collective bargaining coverage rates for salaried workers are
relatively high in some parts of it, whereas in others it is very low. The power of unions has
suffered setbacks in some countries, due to neo-liberal economic policies. Literature on union
effects is perhaps slightly more extensive for Latin America than for other parts of the developing
world, and some of these studies are reviewed below.
An interim conclusion
In addition to the factors mentioned above, it is clear that workers in the developing countries have
typically had less education and they may suffer more from bad health than workers in the
developed world. Unemployed workers in developing countries also have little or no inactivity

16
benefits. Together with the presence of the informal sector, what do these observations imply for
the expected impacts of unionisation in poor countries? First, it is likely that the ‘voice’ effect could
be more important in the context where knowledge levels and firm productivity are relatively low.
Second, the weakness of the official safety net means that workers will put a large value on job
protection. Third, increasing low pay can have a direct impact on the incidence of poverty:
employed individuals that get a higher wage will gain, whereas there is a worry that the overall
number of jobs can go down and wages in the informal sector will remain depressed. The empirical
literature, reviewed below, attempts to answer at least some of these questions.
4 THE IMPACT OF UNIONISATION IN DEVELOPING COUNTRIES:
EMPIRICAL EVIDENCE
As in the case of higher income countries, there seem to be two views of the role that labour market
institutions play in determining macroeconomic outcomes such as unemployment and growth in
developing countries. To present it simply, the more negative view perceives them as fatally
reducing the flexibility of the labour market; the more positive view perceives them as crucial for
issues such as equality. Collective bargaining can also be contrasted with other means of providing
income security or poverty reduction, although, as explained below, labour market institutions and
social policies are not independent of each other.
Labour market institutions naturally include several factors, in addition to union density and the
coverage of collective bargaining issues such as employment protection, unemployment benefits,
and other laws regulating employment contracts. Extracting the impacts of individual institutions at
the country level is difficult, as they tend to be correlated and the institutions may also have
interaction effects that cannot be distinguished, however, from the total impact of the institutional
“package”. In the following we concentrate on literature where collective bargaining or industrial
relations are specifically mentioned: however, in the analysis it is not always clearly separated from
other measures of labour market rigidity. The emphasis is on literature published during the last
decade.
Globalisation has also boosted discussion on the impact of domestic regulation, as it is more likely
to have a major impact if the country is heavily dependent on international competition. Martin
Rama (2003) represents the more negative view of the role of institutions based on collective
bargaining and worker rights: he argues that the coverage of collective bargaining is correlated with

17
the income share of the fourth quintile. Therefore, the core labour standards are relatively
ineffective in reducing inequality, and universal social protection is a better way of achieving this
goal. Rama agrees that rigid labour markets do not change long-term economic efficiency but states
that they may affect the impact of reforms and slower recovery after them, for which public sector
employment and unions are largely responsible. Since the early 2000s a number of papers have
been published on the topic, based on a variety of methods.
Botero et al (2004) conclude in their study that in a panel of countries, including developing
countries, heavier regulation is associated with lower male labour force participation and higher
unemployment. Their measures of regulation include an index of labour union power and collective
disputes. The collective relations index, unlike measures of employment law and social security, is
positively related to the size of the unofficial economy. Thus, in their view, the results do not
support the hypothesis of labour market institutions as an efficiency-enhancing instrument.
Besley and Burgess (2004) study Indian states. The states have partly similar regulatory frameworks
but their laws differ in the ways in which they concern industrial relations. In particular, there is a
piece of central legislation, the Industrial Disputes Act, which regulates, for example, the right to
strike. The act has been amended by the states individually. The act sets out the procedures to be
followed in the case of an industrial dispute, but the states have diverged from the same starting
point in different directions. The authors have coded the amendments as pro-worker, neutral and
pro-employer. Cumulated scores serve as a measure of regulation which correlates with the number
of strikes and lockouts in the state. The paper shows that states which amended the Industrial
Disputes Act in a more pro-worker direction suffer from a stagnating manufacturing sector. The
reforms have also led to increases in the informal sector activities and urban poverty. The study has
been widely cited, but it has also been criticised on the grounds that the data has been simply
miscoded and that the definition of pro-worker reforms is not consistent. Bhattacharya (2006)
argues that the coding is misleading and largely arbitrary, and the results thus fragile. Neither is the
result easy to generalise, as it concerns a very specific situation and context.
Empirical evidence, however, does not only support the negative view. For example, Vandenberg
(2010) finds that collective bargaining has no effect on unemployment in a 3-year panel of 90
countries. Based on his own analysis, Freeman (2010) reiterates for developing countries what has
been previously said by Aidt and Tzannatos’s review about mainly higher income countries: there is
no strong evidence that labour market institutions affect economic growth. Freeman also argues that
in times of crises and major change collaborative industrial relations may be highly beneficial.

18
Thus, what is significant is not just the existence of labour market institutions or collective
bargaining systems but also their finer characteristics and overall quality. Feldmann (2009)
measures the quality of industrial relations. The measure is based on a survey of CEOs of a
representative sample of companies, in which where they are asked about the nature of the relations
between managers and employees. The survey results of the survey together with harmonised
unemployment figures are used to investigate the link between unemployment and industrial
relations. In his analysis, factors such as geographical variables and ethnic fractionalisation are
controlled for as well as the strictness of labour and business regulations. The results show that high
quality industrial relations are likely to lower unemployment in developing countries, though the
magnitude of the effect is moderate. Justino (2006) analyses data from Kerala, and uses the
numbers of strikes as one measure of activism. She shows that excessive activism may harm state
income, but collective action in the context of responsible bargaining systems may contribute
toward increased economic development.
Micro-level impact of studies on the impact of trade unions in developing countries
The above studies are mainly based on country-level or regional aggregate variables. As pointed out
in the introduction, there are severe problems in establishing causal relationships with data across
countries or establishments where either union density or status is fixed over time or changes as the
result of the same processes that impact on the outcomes.
Freeman (2010) points out that in the 1990s more evidence in the developing countries was gained
with pseudo-experiments, for example country-wide changes in union activities as a result of
changes in legislation. However, he states that it seems to be very difficult to find general rules
about the effects. Findings from a number of studies (the majority of which concern either Latin
America or Africa) seem to imply that unions are associated with higher wages, the positive union
premium varying between 3% and 20%. There are also results suggesting higher non-wage shares
of compensation, lower turnover and less dispersion of pay, but these do not concern a large number
of countries. Below we summarise a few relatively recent country-specific studies in more detail.
Many of the studies on union effects use Latin American data. Kuhn and Marques (2005) have
edited a collection of studies on Latin America. Johnson (2005) analyses union membership in
several countries and compares North American and Latin American countries. Her results show
that low unionisation rates in Latin America are partly a result of structural factors such as the age
of the workforce or the types of enterprises they work at. However, in Latin America highly

19
educated workers are more likely to be unionised, unlike those in North America. However, as
shown by the results, legislation as well as product competition also plays a role.
Saavedra and Torero (2005) study unionisation as well as union impacts on firm performance,
labour productivity and profits in Peru. The probability to be a member of a union has declined,
which is probably largely due to legislative changes that restricted union power. Detailed
econometric estimations show that unions have a negative impact on profits, but the relationship
with productivity is less clear. The effect seems to be negative but disappears when a large number
of control variables are added. The authors argue that the possible negative relationship may be
related to contentious labour management relations. Capital intensity increases with unionisation.
In the same volume, Menenzes-Filho et al (2005) examine the relationship between unions and
economic performance in Brazilian establishments. Brazil has undergone significant trade
liberalisation, which may have an effect on the role of unions, as there are fewer rents to be
extracted after being exposed to international competition. Brazilian unions are regulated, but
before 1988 the Labour Code restricted union activity. A new constitution in 1988 changed the
situation significantly and since then unions have enjoyed a guarantee of monopolistic
representation, but they do not represent workers at the plant level. Bargaining outcomes are
extended to all workers in the industry. At the firm or plant level, unions negotiate on working
conditions and practices, employment levels and other issues. The econometric estimates show that
unionisation reduces plant profitability but up to a certain level raises productivity and employment.
Beyond certain levels of union density the impact is, however, less positive or negative.
Cassoni et al (2005) study unions in Uruguay where, too, there has been a decline in union density.
There are agreements at the plant or firm level and industry level; the last typically covers small
firms. The authors find that unions raise productivity, employment and investment. Firms seem to
substitute labour for capital and move to more capital-intensive production. The change in the level
of bargaining may have changed the preferences of unions, and they negotiate over employment in
exchange for smaller wage increases.
A different result is obtained from Guatemalan coffee farms (Urizar and Lee, 2005); unions are
associated with lower productivity. This may reflect the practices of the unions at coffee farms
(which are generally perceived to be productivity-reducing) or the nature of the relations between
management and unions. When only the changes in productivity following a change in union status
(fixed-effects estimation) are considered, the results do not show significant negative effects.

20
Teachers are, in general, a fairly unionised occupation, and in developing countries teachers are of
special interest because of the large role of the quality of and access to education in development.
Zegara and Ravina (2005) show that in Peru union status itself has no impact on the effort of
teachers or their output (maths and language scores). Murillo et al. (2005) show that in Argentina
teachers’ unions have mixed effects on the outcomes: union fragmentation and adversarial political
alignments decrease the effective number of class days, and unionisation has a negative impact on
job satisfaction (and therefore on pupil performance), but on the other hand unions have a positive
impact on student outcomes through tenure and smaller class size. No impact of teacher
unionisation on resources is found.
Overall, the most robust results obtained from this set of studies are that unions reduce profits and
increase employment. There are a number of topics that have been studied in some industrialised
countries and would also be of interest in the context of developing countries. These include, for
example, the impact of unions on working time and employment security. However, for developing
countries these studies are largely missing, in addition to the fact that the number of studies
concerning even the basic effects in developing countries is fairly small.
The evidence for high-income countries suggests that the main impact of unionisation is
redistributive. For developing countries the lower coverage of unions as well as the existence of the
informal sector implies that the overall effect may be very different, although the existence of union
premium suggests that, at least to some extent, unions have similar effects on the wages of
unionised workers. Hayter and Weinberg (2011) review the evidence for inequality impacts of
collective bargaining in both developed and developing countries. Below, we present the main
points of their summary as well as some individual studies in detail.
Hayter and Weinberg do not find any convincing evidence from the existing literature that in
developing countries collective bargaining emphasises the gap between the insiders and outsiders.
On the other hand, unlike that for high income countries, recent evidence on unions actually
narrowing the overall wage or income inequality in developing countries is not strong or
comprehensive, either, although in most cases unions do seem to reduce wage dispersion within the
unionised sectors. Some such evidence does exist for Mexico and Argentina. Fairris (2003) shows
that unions were a strong equalising force that affected the dispersion of wages in 1984, but were
not as effective in 1996. Unionisation has declined and obviously unions have also lost power
between these periods.

21
Overall, it is hardly surprising that the total effect is small, given the low coverage of collective
bargaining or the lack thereof in developing countries. As far as overall income distribution is
concerned, other factors also play a role. However, in the absence of universal or generous social
security systems in developing countries, the lowest income groups in particular obtain most of
their income from work.
Reinecke and Valenzuela (2011) discuss the developments in Chile. The article highlights many of
the issues that limit the possibilities for collective bargaining institutions to impact on the
distribution of income in developing countries. Collective bargaining institutions are very weak in
Chile, partly due to legislation, and the wage distribution has become more unequal over the years,
especially at the bottom of the income distribution. There have been modest improvements overall
in income distribution but these have been due to increased social spending. Strikingly, labour
productivity has risen faster than wages. Collective bargaining in Chile is characterised by a lack of
trust between employers and workers, and the unionisation rate and coverage are low by regional
standards. According to the little data that is available, unionisation is also biased towards better-
earning workers. Rios-Avila and Hirsch (2012) compare union effects on wage distribution in
Chile and Bolivia, two neighbouring countries of which Chile is economically the more developed.
They find that unions do reduce wage dispersion in both countries, although there is indeed less
impact on the lower half of the distribution. The results are surprisingly similar to those obtained in
the US.
In South Africa the union premium is found to be exceptionally high, and a common finding in the
earlier literature is that they reduce wage dispersion (Casale and Posel 2011). Racial wage
differentials have typically been found to be smaller in the unionised sector (Azam and Rospabe
2007), although gender wage differentials may even be larger (Casale and Posel 2011). Kerr and
Teal (2012) study the determinants of earnings inequalities by using a comprehensive panel data
set. They find that individual ability and human capital explain a part of the earnings differentials
including the union premium. This would suggest that firms are responding to unionisation, by
increasing their capital stock and lowering their labour usage but hiring higher quality workers.
Calderon and Chong (2009) use country level panel data to disentangle the links between labour
market institutions and income inequality. They study the impact of labour regulations on income
inequality in a large panel of countries (including developing countries). Union density is included
in the index and as a separate regressor. They find that both de jure and de facto regulations
improve income distribution.

22
However, contrasting evidence also exists. Hasan, Mitra and Ural (2006) look at the impact of trade
liberalisation on poverty reduction and at the way in which this varies with labour market
institutions. The results indicate that delicensing in Indian states has had a more beneficial impact
on poverty reduction in states with flexible labour market institutions. The states are classified as
flexible or inflexible based on the Besley and Burgess data, but some corrections have been made.
Baccaro (2008), on the other hand, finds that in a panel of countries, including developing countries
and covering the years 1989-2005, changes in trade union density and bargaining structure are not
related to income inequality within countries over time, with the exception of Central and Eastern
European countries.
Batt and Nohara (2009) compare outcomes in similar establishments across countries with different
institutions. Their sample also includes a few developing countries and consists of data from call
centres. Though the amount of data is naturally much smaller than in studies using data sets
covering the whole economy, the similarity of the settings allows a fairly straightforward
comparison of wage structures. Indeed the outcomes, in particular wage dispersion, by country
resemble those predicted by theories and other studies. Emerging markets resemble more liberal
market economies, as the unions are weak and they also have very high levels of wage dispersion.
Trade unions as a tool for implementing labour standards
Economic effects aside, the theory of a union “voice” suggests, and the conceptualisation of labour
standards as a public good suggests, that a trade union can be particularly effective in improving
welfare through its impact on working conditions.
Labour standards in general have been subject to a heated debate, especially in the context of
globalisation. As countries have become more interdependent and production can now be easily
moved to another country or split into several countries, the “race to the bottom” has become a
concern, especially for countries that rely on bulk production whose mainly competitive advantage
is low costs. In order to avoid the race to the bottom, it is often argued that labour standards need to
be agreed on at the international level (Palley 2004). Effective enforcement would, however, still
pose a problem.
There is a diversity of views on how labour standards are best promoted: in addition to those
arguing for the use of trade unions or governmental regulations there is broad support for voluntary
corporate self-regulation, for example through certification. All in all, the core issue with labour

23
standards seems to be analogous to that of collective bargaining: they may increase welfare and
even productivity, but in a global environment that may happen at a cost. Singh and Zammit (2004)
argue that labour standards are best achieved in a non-coercive and supportive international
environment. They also emphasise that the conventions related to collective bargaining should be
revised in order to include the provision of a voice for informal sector workers as well as the
promotion of co-operation between employers, employees and those working in the informal sector
rather than assume that the infrastructure for dispute settlement systems exists.
Not surprisingly, in the empirical economic literature on labour standards the links between gaining
growth, trade and investment by lowering labour standards are not clear. Though some authors
argue that the level of development needs to be high enough for labour standards to be introduced
and that therefore it is too early to improve in the developing countries (Hall and Leeson 2007),
there is evidence that labour standards go hand in hand with improved economic performance.
Rene Bazillier (2010) investigates the core labour standards and development, in particular the
impact of labour standards on long-term income. He argues that countries with higher labour
standards have a higher long-term GDP per capita, as enforcement decreases discrimination and
therefore adds incentives to obtain education, and unionisation makes investments in on-the-job
training more likely, thus increasing human capital in the long run. In his empirical analysis,
Bazillier uses an aggregate index of labour standard enforcement as an explanatory variable in a
growth model augmented with human capital. He also uses a method whereby an attempt is made to
control for the fact that per capita income also impacts on labour standards. His results imply that
good enforcement of labour standards has indeed a positive effect on the long-term GDP per capita.
Busse (2004) studies the determinants of core labour standards in developing countries. The results
show that income levels, openness to trade and human capital levels are important in explaining
adherence to the core labour standards. The coefficient of openness to trade is, however, negative
for civil rights (union rights). However, another study by Busse, Nunnenkamp and Spatareanu
(2011) shows that investments by multinationals are significantly higher in countries that adhere to
labour rights. They use a panel of developing countries and also try to establish causality rather than
plain statistical association. Bonnal (2010a and 2010b) shows that in a panel of countries de facto
labour standards (measured by the rate of work injuries and the rate of strikes and lockouts) have a
positive impact on economic growth and, together with the quality of institutions, are associated
with better export performance.

24
Mosley and Uno (2007) use a panel of countries and construct a labour rights index that measures
the legal rights of workers to freedom of association and collective bargaining, key elements of core
labour standards and respect for these rights in practice. The results show that foreign direct
investment is positively and trade negatively related to labour rights. The level of regional
competition and democracy are positively related to labour rights. For the level of GDP per capita
(among developing countries) the association is negative, in other words richer developing
countries have worse records in labour rights. According to the authors, this may be a result of their
being more industrialised rather than of higher income causing the deterioration of labour rights.
Greenhill, Mosley and Prakash (2009) find that strong legal protections of collective labour rights in
export destinations are associated with more stringent labour laws in the exporting country, but the
effect is weaker in the context of labour practices. Neumayer and DeSoysa (2007) focus on the
impact of globalisation on the right to associate. They find that more open countries have fewer
violations that more closed ones. Thus, the evidence suggests that, to an extent, globalisation can
also promote the right for collective bargaining, or at least it does not cause major deterioration.
As pointed out above, globalisation has been a major trigger for concerns for the implications of
trade union activity for competitiveness. Kucera and Sarna from ILO (2006) study how trade union
rights affect trade competitiveness and classify the sectors on the basis of labour intensity. The
authors, however, find that the relationship between trade union rights and exports is highly
sensitive to the classification of labour-intensive sectors.
More broadly speaking, labour standards can also include health and safety at work. Pouliakas and
Theodossiou (2008) review the literature on the economics of occupational health and safety and
find that there is some evidence that the presence of trade unions improves occupational health and
safety. There is also evidence of the opposite: it is however, difficult to establish whether this is
because of endogeneity, or dangerous conditions that make it more likely for the workplace to
become unionised. They also point out that markets are unlikely to provide an efficient and
equitable level of occupational health and safety due to factors such as moral hazard, asymmetric
information and externalities, and thus some form of government or other intervention is necessary.
Enforcement of labour standards
Given the number of initiatives and the amount of publicity the topic has gained, there is
surprisingly little literature on how labour standards are effectively attained. In particular, published

25
rigorous statistical evaluations of the initiatives are scarce. Several authors have, however,
discussed the topic in a broader sense. Graham and Woods (2006) argue that government action is
also necessary, and support from international organisations and legal instruments may be needed to
assist. Compa (2004) argues that corporate codes of conduct may create tensions between unions
and NGOs. Sarovilla and Verma (2006) argue for a new role for soft regulation. In their view, the
ILO does not have punitive power. Corporate codes of conduct, on the other hand, are often
problematic, as reporting initiatives have limited participation and there is no effective monitoring.
The authors propose that national governments should initiate a process by which a minimum
standard is set for participating companies, and perhaps provide a tax incentive for companies that
participate.
One of the most extensive efforts to monitor and enforce labour standards in a developing country
has been the Better Factories Cambodia programme and a more recent multi-country programme
Better Work2
. The programmes are run by ILO and (in the case of Better Work) the International
Finance Corporation but rely on co-operation with governments, employers’ and workers’
organizations, and other parties. The Better Factories Cambodia project was initiated by a trade
agreement between the United States and Cambodia, whereby the US promised Cambodia better
access to US markets in exchange for improved working conditions in the garment sector. BFC
assesses and reports on working conditions in Cambodian garment factories and aims to help them
to improve working conditions and productivity. Better Factories Cambodia conducts unannounced
factory visits to check on working conditions. Evaluation includes interviews with workers’
management and union representatives. The results are reported back to managers together with
suggestions for improvement. The programme also offers a range of training opportunities and
resources to the industry. Polaski (2006) shows that a majority of factories remedied problems
found as a result of inspections and that the programme has been highly cost effective, although she
finds that international buyers received great benefits from the programme without contributing to
it.
Better Work was established in 2006 and launched in Jordan in 2008, and a number of countries
have joined the programme since then. In addition to ILO and IFC, it relies on donor and buyer
support. Both programmes include extensive data collection and regular synthesis reports. The data
produced by these programmes has been used to study a number of issues, in addition to evaluating
improvements in the actual labour standards. The aim of Better Work is to assess its impact
2
The websites for the programmes and the source of the information presented here are http://www.betterfactories.org/
and http://betterwork.com/global/.

26
carefully and the programme’s objective is to construct randomized controlled trials to be able to
clearly identify which changes can be directly attributed to its operations.
A number of studies are available from the programmes’ websites. Rossi and Robertson (2011)
describe the state of and changes in industrial relations during BFC, and argue that, although there
have been significant improvements, the atmosphere is still challenging and further work needs to
be done.
There are a few examples of earlier quantitative evaluations of smaller-scale programmes published
in international journals. They may include some data analysis but are generally a combination of
qualitative and quantitative analysis. Locke, Qin and Rause (2007) and Locke, Kochan, Romis and
Qin (2007) assess a labour standards monitoring programme by Nike. Critics have argued that
private, voluntary monitoring crowds out union or government actions. The auditing programme by
Nike for its suppliers seems to have produced mixed results, as revealed by a comparison of two
plants with different practices and productivities but similar compliance scores. The authors
conclude that firm codes of conduct need to be complemented by the effective enforcement of
national laws and unions or other institutions that provide employees with a voice.
Bulut and Lane (2011) present case studies on Turkey and China, which are mainly of a qualitative
nature. According to them, the success of private regulations (multi-stakeholder initiatives which
may include unions) has been only partial. MSIs often have weak connections with workers and co-
operation between participants does not work well. Buyers often tolerate partial incompliance or
effectively cause incompliance by setting unrealistic demands for delivery times etc.
Barrientos and Smith (2007) conducted an impact assessment of Ethical Trading Initiative codes.
They surveyed 29 ETI companies and evaluated the impact that company codes had on workers in
the value chains of a few companies in South Africa, India and Vietnam, and further conducted two
in-depth company studies on Costa Rican bananas and horticulture. The collection of data is not
based on a random sample or a sophisticated experimental set-up but yet they attempt to establish
some statistical evidence in addition to the qualitative assessment. Some improvements are found in
health and safety and reductions in working hours. The impact on wages is less clear; the initiative
has not led to a substantial increase in income and the codes did not benefit contract workers. The
impacts on “process rights” are found to be minimal, discrimination had not been reduced and there
was no impact on the freedom of association and collective bargaining.

27
5 EVIDENCE ON THE EFFECTIVENESS OF AID/PROJECTS
CHANNELLED VIA UNIONS
Given that high-income countries have devoted a considerable amount of resources to foreign aid, it
is hardly surprising that there is a fair amount of research studying the implications of aid3
. Most of
this work concerns intergovernmental aid. Non-governmental and civil society organisations,
however, are currently important players in the field of development aid (for an overview of NGOs,
see Werker and Ahmed 2008). It is often stated that channelling aid through NGOs or other civil
society organisations may get around the fungibility problem or the use of money on secondary
purposes. Strengthening civil society organisations is also perceived to be an important goal per se,
and crucial for long-term development. There is, however, little evidence on the overall relative
effectiveness of the different channels. The few existing examples include Masud and Yontcheva
(2005), who show that NGO aid reduces infant mortality more effectively than official bilateral aid.
Nancy and Yontcheva (2006) show that poverty is the main determinant of the allocation of NGO
aid.
To our knowledge, no rigorous empirical work on aid given through trade unions exists, although in
many countries governments assist trade unions’ work in developing countries. The Trade Union
Development Cooperation Network (2012) has explored support mechanisms for trade unions’
development aid in 18 countries: the United States, Belgium, Denmark, France, the UK, the
Netherlands, Sweden, Italy, Finland, Norway, Spain, Luxemburg, Switzerland, Australia, Ireland,
Germany, Japan and Portugal. In these countries governments are or have been funding trade
unions’ development activities and have indeed been their largest funders in many cases. The share
of government funding is 100% in some cases and 10 of the trade studied receive at least two-thirds
of their aid budget from the government. In some cases the more political activities are separated
from the government-funded ones and are fully self-funded.
The report stresses that “the quest for results” emphasised by government development aid agencies
is highly problematic for trade unions’ aid, as the impact of their activities is not easily measurable
because the goals include phenomena such as democratisation, the strengthening of civil society or
political projects. In addition, macro-level effects are hard to attribute to a single cause because of
the presence of other confounding factors. The report also states that there are many external factors
that influence the outcome, and showing a genuine impact can be impossible. Undoubtedly, these
3
Extensive reviews of studies on the macroeconomic effects of aid and good practices are, for example, Collier and
Dollar (2004) and Bourguignon and Sundberg (2007), Easterly and Pfutze (2008), Minoiu and Reddy (2009).

28
concerns are well founded. However, in some instances the project evaluation methods described
below could provide valuable additional information on the effectiveness of aid. For example, it is
mentioned that the number of collective labour agreements, a quantitative indicator, does not give
the full picture of the impact of the agreements on people’s livelihoods. With proper data collection
some of this type of problems could be overcome. For example, the evaluation of the TUC-DFID
Partnership Agreement (2011) states that the indicators used do not capture the impact and that
what “improvement” would look like is not always clearly defined. Similarly, in the evaluation of
Sida’s support to the LO-TCO secretariat (Runchel et al., 2008) concerns about the quality and
quantity of proper evaluations are raised.
Though there is relatively little solid evidence on the channel of aid, specific projects carried out by
NGOs (as well as international organisations or governments) have been evaluated by using
experimental or quasi-experimental methods. The issue of interest in these is on the effectiveness of
a particular intervention. Current research favours randomised controlled trials (RCT), whereby an
intervention is applied to a randomly selected group of individuals or units, and there is a similar
control group. Alternatively, if pure randomisation is not possible, the control group can be
constructed by using statistical methods to match the treated group as closely as possible. To our
knowledge, this approach has not been applied to specific projects carried out by trade unions or
projects related to developing industrial relations or working conditions. In the case of labour
standards, there are examples from, for example, the use of conditional cash transfers to reduce
child labour.
The main advantage of this approach is the ability to solve the problem of causality. In
observational studies, or in typical qualitative analysis, it is always left open as to whether the
detected developments are caused by a particular intervention or whether they have taken place for
another reason. In careful econometric impact evaluation, one measures the interesting outcome
variable before and after the intervention, not only within those units of analysis that are subject to
the intervention but also from a comparison group. The comparison group should consist of units
that are otherwise similar so that the only reason why outcomes vary between treated and control
units is the intervention. The best way to ensure that the groups are alike is via randomization: then
subjects cannot choose themselves whether they become subject to the intervention or not. For a
thorough discussion of the approach, see Banerjee and Duflo (2009).
It is clear that such an approach is not suitable for every possible research question: macro policies
can hardly be randomized across countries. It is also always open to what extent the results can be

29
generalised to other settings. (This is the question about external validity.) And while it can be
argued that the experimental approach can reveal what works and what does not work, the question
why something happens is often left unanswered. However, the last problem can be alleviated by a
careful design: the experiment can involve different types of interventions that are being tested at
the same time against the no-intervention case and, in the best possible situation, the experiment
deals directly with a particular behavioural mechanism (Mullainathan et al. 2011).
To our knowledge there has not been a study whereby an intervention carried out by using a trade
union is tested by using this approach. This is quite surprising, since these interventions are often
meant to affect labour standards, something which is high on the agenda of global development
objectives. Moreover, the interventions are of the type that could be subject to rigorous impact
evaluation, at least when they deal with firm-level or worker-level outcome variables. While
potential outcomes can be somewhat abstract, they are quite similar topics (such as job satisfaction)
that are routinely investigated using surveys and interviews in the developed countries. In our view,
there are no conceptual-level reasons why these issues could not be examined using impact
evaluations in poor countries as well, although practical problems in managing surveys need to be
taken into account.
6 CONCLUSIONS
In this paper we have surveyed a number of studies concerning the economic impact of trade unions
in developing countries and we have discussed the potential avenues for future research.
Overall, there seems to be a relative scarcity of evidence on basic issues such as the impact of
unionisation on various outcomes. This partly reflects the lack of sophisticated data that are
available for the industrialised countries and that allow the study of issues such as the impact of
unions on issues such as training intensity or productivity. On the other hand, given that the
coverage of collective bargaining is lower in developing countries and the high incidence of
informality sets certain restrictions on the scale of the possible impact, it is possible that the current
observable impact of unions on the total economy scale would be, in many cases, small. The
evidence that does exist suggests that labour standards are not detrimental for economic
performance and that unions may have distributional effects. Hypothetically, using the “voice” face
of unions to improve labour standards and, ultimately, productivity could be highly beneficial.
However, there does not yet seem to be clear quantitative empirical evidence on how this could be

30
best attained. This is, in fact, one of the interesting areas for further research: Can unions be used to
implement sufficient labour standards in conditions where the funds and the administrative capacity
of the government are limited and, if so, under what conditions?
Other interesting areas for future research that have emerged from the discussion in this paper
include the following:
• While rigorous impact evaluations have proliferated in almost all other areas in development
economics research, they are almost non-existent in the field of aid channelled via unions.
• Even if the evidence suggests that unions can have at least some negative impact on
employment, they also raise minimum wages, and therefore it would be of key priority to
examine empirically what the joint effect of these changes on poverty is, which is, after all,
the first of the Millennium Development Goals. Kanbur and Fields (2007) examine the
impacts of minimum wages on poverty theoretically, and it would be useful to set their
analysis explicitly in a bargaining framework. One of their points is that the distributional
impacts of minimum wages are dependent on the extent of income-sharing between the
employed and unemployed. Interestingly, such income-sharing is common among union
members in developed countries, and the suitability of such a framework for poor countries
is worth exploring.
• Intuition suggests that monopsony power may be quite prevalent in developing countries.
This raises the question of how the impacts of unions depend on the extent of market power
on the employers’ side.
• While we have some evidence on the impacts of unions on their members’ wages and well-
being, extremely little is known about the spillover effects of union power on outsiders, in
particular, on poorer workers in the informal sector; are there instances where an increase in
unionised sector wages gradually trickles down to the benefit of others in the same
area/industry, or is it more likely that the outsiders’ prospects worsen, as they are shut away
from access to better jobs?
• On a more normative note, there is a need to examine the proper balance between job
protection that unions tend to seek versus the development of the official social safety net.
One possible route would be a gradual transition from job protection to social protection if
and when social assistance systems are being scaled up.
• The broader institutional implications of unions are also worth exploring. Much modern
development research underscores the importance of proper institutions for success in

31
development. (A good recent example of this is Pritchett 2012.) When are unions conducive
to social stability and good governance and when are they more likely to extract rents for
protected elite workers (perhaps civil servants whose position is over-protected)?
Indeed, the research that would be most useful in the future may not be similar to those studies that
have been conducted in developed countries. The central issues that need to be studied as well as
policy goals are clearly different for developing countries. Also, unions in less developed countries
represent very different points on the spectrum of unionism and thus any policy measures and goals
need to be in line with the prevailing conditions in each country. In particular, from the donor
countries’ perspective it would also be important to study how different modes of support for
unionism and forms of technical assistance can best contribute to the goals. Addressing many (if not
all) of these topics requires new, better data and sometimes also cleverly designed experimental
settings.
REFERENCES
Aidt, Toke and Zafiris Tzannatos, 2002. Unions and collective bargaining - Economic effects in a
global environment, The World Bank.
Addison, J. and C. Schnabel (eds.), 2003. International Handbook of Trade Unions, Edward Elgar.
Almeida, Rita and Pedro Carneiro, 2012. “Enforcement of Labor Regulation and Informality,”
American Economic Journal: Applied Economics, American Economic Association, July, Vol.
4(3), 64-89.
Azam, Jean-Paul Sandrine Rospabé, 2007. “Trade unions vs. statistical discrimination: Theory
and application to post-apartheid South Africa,” Journal of Development Economics, September,
Vol. 84, Issue 1, 417–444.
Baccaro, Lucio, 2011. Labor, Globalization and Inequality: Are Trade Unions Still Redistributive?
In David Brady (ed.), Comparing European Workers Part B: Policies and Institutions (Research in
the Sociology of Work, Vol. 22), Emerald Group Publishing Limited, 213-285.
Banerjee, A. and E. Duflo, 2009. “The Experimental Approach to Development Economics,”
Annual Review of Economics, 1:151–78.
Barrientos, Stephanie and Sally Smith, 2007. “Do workers benefit from ethical trade? Assessing
codes of labour practice in global production systems.” Third World Quarterly, Vol. 28, Issue 4.

32
Batt, Rosemary and Hiro Nohara, 2009. “How Institutions and Business Strategies Affect Wages: A
Cross National Study of Call Centers.” Industrial and Labor Relations Review, 62(4), 533-52.
Bazillier, Remi, 2008. “Core Labor Standards and Development: Impact on Long-Term Income,”
World Development, Elsevier, January, Vol. 36(1), 17-38.
Bennett, James T. and Bruce Kaufman (eds.), 2007. “What Do Unions Do? A Twenty-Year
Perspective.” New Brunswick, N.J., Transaction.
Besley, Timothy and Robin Burgess, 2004. “Can Labor Regulation Hinder Economic Performance?
Evidence from India,” The Quarterly Journal of Economics, MIT Press, February, Vol. 119(1), 91-
134.
Betcherman, Gordon, 2012. “Labor Market Institutions: A Review of the Literature.” Background
paper for the World Development Report 2013. The World Bank.
Bhattacharya, A., 2006. “Labour market Regulation and Industrial Performance in India”, Indian
Journal of Labour Economics, Vol. 49(2), 211-231.
Boeri, Tito and Jan van Ours, 2008. “The Economics of Imperfect Labor Markets”, Princeton
University Press.
Bonnal M., 2010a. “Economic Growth and Labor Standards: Evidence from a Dynamic Panel Data
Model,” Review of Development Economics, Vol. 14, No. 1, 20-33.
Bonnal M., 2010b. “Export Performance, Labor Standards and Institutions: Evidence from a
Dynamic Panel Data Model,” Journal of Labor Research, Vol. 31, Issue 1, 53-66.
Botero, Juan, Simeon Djankov, Rafael Porta and Florencio C. Lopez-De-Silanes, 2004. “The
Regulation of Labor,” The Quarterly Journal of Economics, MIT Press, November, Vol. 119(4),
1339-1382.
Brummond, P., 2012. Variation in Monopsonistic Behavior across Establishments: Evidence from
the Indonesian Labor Market, Cornell University, mimeo.
Bulut, T. and C. Lane, 2011. “The private regulation of labour standards and rights in the global
clothing industry: An evaluation of its effectiveness in two developing countries”, New Political
Economy, 16(1), 41-71.
Busse, Matthias, 2004. “On the determinants of core labour standards: the case of developing
countries,” Economics Letters, Elsevier, May, Vol. 83(2), 211-217.

33
Busse, Matthias, Peter Nunnenkamp and Mariana Spatareanu, 2011. “Foreign direct investment and
labour rights: a panel analysis of bilateral FDI flows”, Applied Economics Letters, Taylor and
Francis Journals, Vol. 18(2), 149-152.
Calderón, César and Alberto Chong, 2009. “Labor market institutions and income inequality:
an empirical exploration”, Public Choice, Vol. 138(1), 65-81.
Calvo, Guillermo A, 1978. “Urban Employment and Wage Determination in LDC's: Trade Unions
in the Harris-Todaro Model,” International Economic Review, February, Vol. 19(1), 65-81.
Casale, D. and D. Posel, 2011. Unions and the gender wage gap in South Africa. Journal of African
Economies, 20(1), 27-59.
Cassoni, Adriana, Gaston J. Labadie and Gabriela Fachola, 2005. “The Economic Effects of Unions
in Latin America: Their Impact on Wages and the Economic Performance of Firms in Uruguay”,
Peter and Gustavo Márquez (eds.), 2005. “What difference do unions make? Their impact on
productivity and wages in Latin America”, Inter-American Development Bank.
Compa, Lance, 2004. Trade unions, NGOs, and corporate codes of conduct. Development in
Practice, Vol. 14, Issue 1-2, 2004.
Deaton, A., 2010. Instruments, randomization and learning about development. Journal of
Economic Literature, Vol. 48, 424-455.
Donado, Alejandro and Klaus Wälde, 2012. “How trade unions increase welfare,” Economic
Journal, Royal Economic Society, Vol. 122(563), 990-1009, 09.
Feldmann, Horst, 2009. “The Quality of Industrial Relations and Unemployment in Developing
Countries,” Review of Development Economics, Wiley Blackwell, Vol. 13(1), 56-69, 02.
Fields, Gary S., 2007. “The Harris-Todaro Model” Working Papers. ILR Collection. 8-1-2007.
Cornell University.
Fields, G. and R. Kanbur, 2007. Minimum wages and poverty with income sharing. Journal of
Economic Inequality 4, 367-374.
Freedom of association in practice: Lessons learned. Global Report under the follow-up to the ILO
Declaration on Fundamental Principles and Rights at Work. Report of the Director-General, 2008.
Freeman, Richard B. and James Medoff, 1984. “What Do Unions Do?” Basic Books, New York.
Freeman, Richard B., 2007. “Labor Market Institutions around the World,” NBER Working Papers
13242, National Bureau of Economic Research, Inc.

34
Freeman, Richard B., 2010. Labor Regulations, Unions, and Social Protection in Developing
Countries: Market Distortions or Efficient Institutions? In Dani Rodrik and Mark Rosenzweig,
(eds), Handbook of Development Economics, Elsevier, Vol. 5, 4657-4702.
Galli, Rossana and David Kucera, 2004. “Labor Standards and Informal Employment in Latin
America,” World Development, Vol. 32, No. 5.
Graham, D. and N. Woods, 2006. “Making corporate self-regulation effective in developing
countries”, World Development 34(5), 868-83.
Greenhill, Brian, Layna Mosley and Aseem Prakash, 2009. “Trade-based Diffusion of Labor
Rights: A Panel Study, 1986–2002 / November / Vol. 103 / Issue 04, 669-690.
Hall, J.C. and P.T. Leeson, 2007. “Good for the goose, bad for the gander: international labor
standards and comparative development”, Journal of Labor Research 28 (4), 658-676.
Hasan, Rana Devashish Mitra Beyza P. Ural, 2006. “Trade Liberalization, Labor-Market
Institutions, and Poverty Reduction: Evidence from Indian States,” India Policy Forum, Global
Economy and Development Program, The Brookings Institution, Vol. 3(1), 71-122.
Hayter, Susan (ed.), 2011. “The role of collective bargaining in the global economy: negotiating for
social justice”, International Labour Office.
Hayter, S. and V. Stoevska, 2010. “Social Dialogue Indicators: Trade union density and collective
bargaining coverage”, International Statistical Inquiry 2008-09. Technical Brief.
Hayter, Susan and Bradley Weinberg, 2011. Mind the gap: collective bargaining and wage
inequality. In Susan Hayter (ed.), “The role of collective bargaining in the global economy:
negotiating for social justice”, International Labour Office.
Hirsch, Barry T., 2008. “Sluggish Institutions in a Dynamic World: Can Unions and Industrial
Competition Coexist?,” Journal of Economic Perspectives, American Economic Association, Vol.
22(1), 153-176, Winter.
Hoff, K. and J. Stiglitz, 2001. Modern Economic Theory and Development. In G. Meier and J.
Stiglitz (eds.), Frontiers of Development Economics, Oxford University Press.
Johnson, Susan, 2005. An Empirical Examination of Union Density in Six Countries: Canada,
Ecuador, Mexico, Nicaragua, the United States, and Venezuela. In Kuhn, Peter and Gustavo
Márquez (eds.), “What difference do unions make? Their impact on productivity and wages in Latin
America”, Inter-American Development Bank.

35
Keen, Mark, Hatty Dinsmore and Sadie Watson, 2010. Evaluation of the TUC-DFID Programme
Partnership Agreement, Final Report. IODPARC, prepared for TUC.
Kerr, Andrew Francis J. Teal, 2012. “The Determinants of Earnings Inequalities: Panel Data
Evidence from South Africa,” IZA Discussion Papers 6534, Institute for the Study of Labor (IZA).
Kanbur, Ravi, 2009. “Conceptualizing Informality: Regulation and Enforcement”, February, V.V.
Giri Memorial Lecture. Revised version published in Indian Journal of Labour Economics.
Koçer, R.G. and S. Hayter, 2011. Comparative study of labour relations in African countries.
University of Amsterdam, AIAS Working Paper 116.
Kucera, David and Ritash Sarna, 2006. “Trade Union Rights, Democracy and Exports: A Gravity
Model Approach,” Review of International Economics, Vol. 14, No. 5.
Kuhn, Peter and Gustavo Márquez (eds.), 2005. “What Difference Do Unions Make?” Washington
DC: Inter-American Development Bank.
Kuruvilla, Sarosh and Anil Verma, 2006. “International Labor Standards, Soft Regulation, and
National Government Roles”, Journal of Industrial Relations, February 48: 41-58,
La Porta, Rafael and Andrei Shleifer, 2008. “The Unofficial Economy and Economic
Development,” Brookings Papers on Economic Activity, Economic Studies Program, The
Brookings Institution, Vol. 39(2(Fall)), 275-363.
Lee, Chang and Mingwei Liu, 2011. Collective bargaining in transition: Measuring the effects of
collective voice in China. In Hayter, Susan (ed.), “The role of collective bargaining in the global
economy: negotiating for social justice”, International Labour Office.
Locke, Richard M, Fei Qin and Alberto Brause, 2007. “Does Monitoring Improve Labor Standards
- Lessons from Nike”, Industrial and Labor Relations Review, 61, 3. (2007-2008)
Locke, Richard, Thomas Kochan, Monica Romis and Fei Qin, 2007. “Beyond corporate codes of
conduct: Work organization and labour standards at Nike's suppliers”, International Labour Review,
Vol. 146, Issue 1-2, 21–40, March–June.
Manning, Alan, 2003. Monopsony in Motion. Princeton University Press.
Masud, Nadia and Boriana Yontcheva, 2005. “Does Foreign Aid Reduce Poverty? Empirical
Evidence from Nongovernmental and Bilateral Aid”, IMF Working Paper 05/100, IMF Institute.
Menezes-Filho, Naercio, Helio Zylberstajn, Jose Paulo Chahad and Elaine Pazell, 2005. Unions and
the Economic Performance of Brazilian Establishments. In Kuhn, Peter and Gustavo Márquez

36
(eds.), “What difference do unions make? Their impact on productivity and wages in Latin
America” Inter-American Development Bank.
Mosley, Layna and Saika Uno, 2007. “Racing to the Bottom or Climbing to the Top? Economic
Globalization and Labor Rights”. Comparative Political Studies 40:8, 923-948, August.
Mullainathan, S., J. Ludwig, J.R. Kling, 2011. Mechanism Experiments and Policy Evaluations.
Journal of Economic Perspectives, 25(3), 17-38.
Murillo, Victoria, Mariano Tommasi, Lucas Ronconi and Juan Sanguinetti, 2005. The Economic
Effects of Unions in Latin America: Teachers’ Unions and Education in Argentina. In Kuhn, Peter
and Gustavo Márquez (eds.), “What Difference Do Unions Make?” Washington DC, Inter-
American Development Bank.
Nancy, Gilles and Yontcheva, Boriana, 2006. “Does NGO Aid Go to the Poor? Empirical Evidence
from Europe”, IMF Working Paper 06/39.
Naylor, Robin A., 2003. “Economic models of union behaviour,” chapter 2 in Addison, J. and C.
Schnabel (eds.), International Handbook of Trade Unions, Edward Elgar.
Neak, Samsen, Raymond Robertson, 2009. Globalization and Working Conditions: Evidence from
Cambodia. In Robertson, Raymond, Drusilla Brown, Pierre Gaëlle, Laura Sanchez-Puerta (eds.),
Globalization, Wages, and the Quality of Jobs Five Country Studies, The World Bank, Washington
D.C., 97-130.
Neumayer, Eric Indra de Soysa, 2006. “Globalization and the Right to Free Association and
Collective Bargaining: An Empirical Analysis”, World Development, Vol. 34, Issue 1, 31-49,
January.
Palley, Thomas I., 2004. “The economic case for international labour standards,” Cambridge
Journal of Economics, Oxford University Press, Vol. 28(1), 21-36, January.
Polaski, Sandra, 2006. “Combining global and local forces: The case of labor rights in Cambodia”,
World Development, Vol. 34, Issue 5, 919-932, May.
Pouliakas, Konstantinos and Ioannis Theodossiou, 2010. “The economics of health and safety: An
interdisciplinary review of the theory and policy,” MPRA Paper 20336, University Library of
Munich, Germany, Revised 07, May 2011.
Pritchett, Lant, 2012. Folk and the Formula. Pathways to Capable States. UNU-WIDER Annual
Lecture.

37
Rama, MartÌn, 2003. “Globalization and the Labor Market,” World Bank Research Observer,
World Bank Group, Vol. 18(2), 159-186.
Reinecke, Gerhard and Maria Elena Valenzuela, 2011. Illustrating the gap: Collective bargaining
and income distribution in Chile. In Hayter, Susan (ed.), “The role of collective bargaining in the
global economy: negotiating for social justice”, International Labour Office.
Rodrik, Dani, Arvind Subramanian and Francesco Trebbi, 2004. “Institutions Rule: The Primacy of
Institutions over Geography and Integration in Economic Development,” Journal of Economic
Growth, Springer, Vol. 9(2), 131-165, 06.
Rossi, Arianna and Raymond E. Robertson, 2011. Better Factories Cambodia: An Instrument for
Improving Industrial Relations in a Transnational Context (June 9, 2011). Center for Global
Development Working Paper No. 256.
Runchel, Frank, Agneta Gunnarsson and Jocke Nyberg, 2008. Global Trade Union Building in
Defence of Workers’ Rights. Sida evaluation 2008:57.
Saavedra, Jaime and Máximo Torero, 2005. Union Density Changes and Union Effects on Firm
Performance in Peru. In Kuhn, Peter and Gustavo Márquez (eds.), “What difference do unions
make? Their impact on productivity and wages in Latin America”, Inter-American Development
Bank.
Schurman, Susan J. and Adrienne E. Eaton, 2012. “Trade Union Organizing in the Informal
Economy: A Review of the Literature on Organizing in Africa, Asia, Latin America, North America
and Western, Central and Eastern Europe”, Report to the Solidarity Center. Rutgers University,
January.
Singh, Ajit and Ann Zammit, 2004. “Labour Standards and the ‘Race to the Bottom’: Rethinking
Globalization and Workers' Rights from Developmental and Solidaristic Perspectives”, Oxford
Review of Economic Policy, 20(1), 85-104.
Trade unions’ views on working with donor governments in the development sector - A review of
18 donor governments’ support mechanisms. TUCDN Development papers 2012/4.
Traxler, Franz and Bernd Brandl, 2011. The economic impact of collective bargaining. In Hayter,
Susan (ed.), ”The Role of Collective Bargaining in the Global Economy: Negotiating for Social
Justice”. Edward Elgar, Cheltenham, Northampton, 227-253.
Urízar H., Carmen and Sigfrido Lée , 2005. The Effects of Unions on Productivity: Evidence from
Large Coffee Producers in Guatemala. In Kuhn, Peter and Gustavo Márquez (eds.), “What

38
difference do unions make? Their impact on productivity and wages in Latin America” Inter-
American Development Bank.
Vandenberg, Paul, 2010. “Impact of Labor Market Institutions on Unemployment: Results from a
Global Panel”, No 219, ADB Economics Working Paper Series, Asian Development Bank.
Visser, J., 2003. Unions and Unionism around the World. In Addison, J. and Schnabel, J. (eds.),
International Handbook of Trade Unions. Werker, E., and F.Z. Ahmed (2008), What Do
Nongovernmental Organizations Do? Journal of Economic Perspectives 22 (2), 73-92.
World Development Report 2013. Jobs. 2012. The World Bank.
Zegarra, Eduardo and Renato Ravina, 2005. Teacher Unionization and the Quality of Education in
Peru: An Empirical Evaluation Using Survey Data. In Kuhn, Peter and Gustavo Márquez (eds.),
“What difference do unions make? Their impact on productivity and wages in Latin America”,
Inter-American Development Bank.
Yoon, Youngmo, 2009. “A comparative study on industrial relations and collective bargaining in
East Asian countries”, November, International Labour Office, Geneva.

Trade unions in the south and co-operation between unions in the South and in the North: A survey of the economics literature

More Related Content

What's hot (9)

Viewers also liked (6)

Similar to Trade unions in the south and co-operation between unions in the South and in the North: A survey of the economics literature (20)

More from Palkansaajien tutkimuslaitos (20)

Recently uploaded (20)

Trade unions in the south and co-operation between unions in the South and in the North: A survey of the economics literature