ArticlePDF Available

Missing Links in the Inclusive Growth Debate: Functional Income Distribution and Labor Market Institutions

Authors:
  • Autonomous University of Madrid (UAM)

Abstract and Figures

The aim of this article is to discuss the ‘inclusive growth’ proposals made by the IMF, the OECD and the EU, and to examine to what extent these proposals are consistent with the objective to be achieved. To do this, we examine the importance of two ‘missing links’ commonly overlooked by these institutions when promoting ‘inclusive growth’: functional income distribution and labor market institutions. Using a panel of 42 advanced countries for the period 1990–2018, we estimate an econometric model and we obtain two relevant results. First, functional income distribution has an important influence on income inequality. Second, collective bargaining coverage plays a significant role in explaining increases in the wage share and reductions in income inequality. Consequently, these two ‘missing links’ must be fully integrated into the policy debate of international organizations if ‘inclusive growth’ is to be successfully promoted.
Content may be subject to copyright.
Copyright © The authors 2021
Journal compilation © International Labour Organization 2021
*Autonomous University of Madrid (UAM) and Complutense Institute for International Stud-
ies (ICEI), email: nacho.alvarez@uam.es (corresponding author). **Amsterdam Institute for Ad-
vanced Labour Studies (AIAS), University of Amsterdam, email: m.j.keune@uva.nl. ***Fundación
1º de Mayo, email: jcruces@1mayo.ccoo.es. ****University of Castilla La Mancha (UCLM), email:
Jorge.Uxo@uclm.es.
Responsibility for opinions expressed in signed articles rests solely with their authors, and
publication does not constitute an endorsement by the ILO.
International Labour Review, Vol. 160 (2021), No. 3
Missing links in the inclusive
growth debate:
Functional income distribution
and labour market institutions
Ignacio ÁLVAREZ,* Maarten KEUNE,** Jesús CRUCES***
and Jorge UXÓ****
Abstract. This article examines the extent to which “inclusive growth” proposals
made by various international organizations are consistent with their own object-
ives. The authors identify two commonly overlooked “missing links”: functional
income distribution and collective bargaining coverage. Using a panel of 42 ad-
vanced countries for the period 1990–2018, they find that the first has an important
influence on income inequality and the second plays a significant role in explain-
ing increases in the wage share and reductions in income inequality. Consequently,
these two factors must be fully integrated into the policy debate of international
organizations if inclusive growth is to be successfully promoted.
Keywords: inclusive growth, collective bargaining, functional income distribution,
inequality, EU, IMF, OECD.
1. Introduction
Globalization and technological change have been a source of economic growth.
Yet the benefits of these processes have been shared unevenly, triggering un-
employment, low-paid jobs and high levels of inequality. In countries of the
Organisation for Economic Co-operation and Development (OECD), the Great Re-
cession of 2008 together with fiscal austerity and internal devaluation policies
have served to reinforce this trend over the last decade (Vaughan-Whitehead
2015; Ortiz et al. 2015). Today, work no longer guarantees a way out of poverty,
International Labour Review
338
and the living conditions of large sections of the population are under pressure
within advanced economies.
In this context, international institutions such as the European Union (EU),
the International Monetary Fund (IMF) and the OECD have begun to use “inclu-
sive growth” as a new approach to address current economic and social chal-
lenges in advanced countries (European Commission 2018; Eurostat 2019; IMF
2017a and 2020; OECD 2017a and 2017b). The concept of “inclusive growth”
was first developed in the 2000s when the World Bank realized that, contrary
to Kuznets’s (1955) logic, growth does not necessarily reduce inequality. Thus,
a concept initially applied to emerging economies has come into use in OECD
countries. In this article, we address certain weaknesses and missing links in
the debate.
The OECD defines “inclusive growth” as “economic growth that creates op-
portunity for all segments of the population and distributes the dividends of
increased prosperity, in both monetary and non-monetary terms, fairly across
society” (OECD 2015, 84–85). The concern among international institutions for
inclusive growth has helped reinforce a multidimensional view of well-being,
moving beyond per capita gross domestic product (GDP) to embrace other elem-
ents such as job quality, skills and education, basic services, infrastructure, en-
vironment and fiscal transfers (European Commission 2018; Eurostat 2019; IMF
2017a and 2020; OECD 2015). And, of course, inequality is a key element when
considering inclusive growth.
Inequality is a source of concern among these institutions not only because
of its effect on social cohesion but also because of its potential impact on eco-
nomic performance. They have come to realize that policies that increase income
inequality undermine social stability and trust among social groups and may
also curtail economic growth by constraining the ability of low-income groups
to contribute to production (European Commission 2018; Berg and Ostry 2011;
OECD 2015 and 2017b). For these reasons, inclusive growth has today become
a key objective of economic policy.
However, international institutions have mainly focused on factors such as
globalization, technological change, market failures and education gaps to ex-
plain the unequal distribution of economic growth and the increase in inequality
(IMF 2017a and 2020; OECD 2015 and 2017b). These institutions have paid insuf-
ficient attention to other dimensions that are essential to understanding the lack
of inclusive growth. In particular, they tend to disregard two fundamental de-
terminants in this debate: the role of functional income distribution, and the cap-
acity of labour market institutions to alter market “pre-distribution” of income.
These two dimensions have undergone important transformations during
recent decades in OECD economies. According to Bowley’s law, one would expect
the wage share to remain constant in the long run. However, this is not what
has happened. A drastic change in the distribution of income between wages
and profits appeared in OECD countries in the mid-1970s, when the wage share
began a steady downward trend (see figure 1).
Likewise, labour market institutions have experienced significant changes
over the same period, as we can see in figure 2. Over the past five decades, in
OECD countries, the average proportion of wage and salary earners who are
Missing links in the inclusive growth debate 339
trade union members has fallen from 46.5 to 25.8 per cent. Similarly, the average
proportion of employees who are covered by collective bargaining agreements
has decreased over this period from 59.4 to 48.1 per cent. What impact have
these changes had on personal income distribution? Is an inclusive growth strat-
egy feasible when the evolution of these two variables is not taken into account?
In recent years, with the surge in research on inequality in the wake of the
2008 crisis, new empirical and theoretical studies have begun to examine the
importance of functional income distribution in explaining increased inequal-
ity. Recent research (Atkinson 2015; Jacobson and Occhino 2012; ILO 2019) has
pointed out how changes in functional income distribution can shape changes
Note: EU-15 refers to the Member States of the EU before the accession of ten further countries in 2004.
Source: Annual macro-economic (AMECO) database of the European Commission.
Figure 1. Adjusted wage share in main OECD countries, 1970–2018
(percentages of GDP)
50
55
60
65
70
Japan
1970
USA EU-15
75
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
Source: OECD/AIAS ICTWSS database.
Figure 2. Employees covered by collective bargaining agreements and
trade union density in OECD countries, 1970–2018 (percentages)
50
55
60
45
25
30
35
40
20
65
1970–1979 1980–1989 1990–1999 2000–2009 2010–2017
Coverage (%) Union density (%)
International Labour Review
340
in income inequality in OECD countries. The economic literature has also in-
dicated the existence of a crucial link between labour market institutions and
functional income distribution (Moore et al. 2019; Checchi and García-Peñalosa
2010; Kristal 2010).
These same issues are increasingly being studied and confirmed by the re-
search departments of international institutions. However, as we will discuss in
the next section, the policies promoted by these institutions continue to ignore
them. As a result of failing to incorporate the role of functional income distribu-
tion and labour market institutions, the policy recommendations of the inter-
national institutions often evince important contradictions and biases in relation
to the objective of inclusive growth. Labour market institutions are frequently
presented as “rigidities” that can hurt growth and other macroeconomic out-
comes, disregarding the possibility of building a new income distribution model
that, using these labour institutions as a fulcrum, reconciles economic growth
with reduced inequality.
The aim of this article is to discuss the inclusive growth policies of the IMF,
the OECD and the EU – the most influential institutions when it comes to ex-
plaining economic policy changes in advanced countries – and to examine the
extent to which these proposals are consistent with the objectives they aim to
achieve. Besides reviewing the policies promoted by these organizations, and in
order to provide a solid basis for our discussion, we empirically analyse the two
aforementioned missing links: (1) the role of functional income distribution in
explaining inequality; and (2) the importance of labour market institutions
in explaining both the evolution of the wage share and income inequality.
Our starting hypothesis is that the erosion of labour market institutions helps
to explain the decline of the wage share in recent decades, and that both phe-
nomena have contributed to growing inequality in the OECD. Labour market
institutions – and particularly collective bargaining – play a key role in the ne-
gotiation of the distribution of productivity gains and are therefore crucial in
redirecting the functional income distribution and reducing inequality.
Our research contributes to the literature on inclusive growth in several
ways. First, it relates the evolution of personal income distribution to functional
income distribution, incorporating bargaining between capital and labour into
the debate on inclusive growth. Second, when analysing the determinants of
functional income distribution and inequality, we focus on the impact of labour
market institutions, and particularly on collective bargaining. Third, we are not
solely interested in analysing the determinants of income distribution; we also
compare our empirical analysis with the policy strategies promoted by the IMF,
the OECD and the EU. Lastly, we use a data panel that presents a sample of coun-
tries and years larger than those used in most previous research.
We estimate an econometric model using a panel of 42 advanced coun-
tries for the period 1990–2018. The countries that make up our sample are the
37 Member countries of the OECD (Australia, Austria, Belgium, Canada, Chile,
Colombia, Czechia, Denmark, Estonia, Finland, France, Germany, Greece, Hun-
gary, Iceland, Ireland, Israel, Italy, Japan, Republic of Korea, Latvia, Lithuania,
Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Poland, Portugal,
Missing links in the inclusive growth debate 341
Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey, United Kingdom of Great
Britain and Northern Ireland, United States of America) and those countries of
the European Union that are not part of the OECD (Bulgaria, Croatia, Cyprus,
Malta, Romania).
We obtain two relevant results from our estimates: first, that the functional
distribution of income has indeed a significant influence on the evolution of
income inequality; second, that collective bargaining coverage plays a crucial
role in explaining both declines in the wage share and increases in income in-
equality. These two results suggest the need to reformulate and broaden the
policy strategies of international institutions to consider the positive role that a
strengthening of collective bargaining can play in achieving inclusive growth.
As has been pointed out by other scholars, collective bargaining can play an
important part in reducing inequalities, by providing working people with the
institutional capabilities required to negotiate their share of productivity gains
(Hayter and Visser 2021).
The structure of this article is as follows. In the second section, we analyse
the inclusive growth strategies of the EU, the IMF and the OECD. In the third
section, we address the missing links and contradictions in such strategies and
we outline the importance of taking into consideration both functional income
distribution and labour market institutions. In the fourth section, we estimate
an econometric model to test the soundness of our main arguments. Lastly, in
the fifth section, we conclude and present the policy implications emerging from
our analysis.
2. Inclusive growth: A new goal for international
institutions
The EU, the IMF and the OECD have committed themselves to inclusive growth,
arguing that growth alone is not sufficient to ensure well-being: growth should
be to the benefit of all and not just a few, and should contribute to reducing in-
equality, which has been on the rise in recent decades.
In this section we analyse the inclusive growth strategies of these institu-
tions. We discuss the importance that these strategies attribute to functional
income distribution and labour market institutions (such as collective bar-
gaining coverage and union density) in reducing inequality and strengthening
the inclusiveness of growth.
2.1. The European Union
The EU made a commitment to inclusive growth in 2010 with its Europe 2020
strategy for smart, sustainable and inclusive growth (European Commission
2010). The Europe 2020 strategy has as its objective a “smart, sustainable and
inclusive economy delivering high levels of employment, productivity and social
cohesion” (European Commission, 5). Indeed, Europe 2020 focuses on increas-
ing employment, productivity and social cohesion. It does not, however, ad-
dress the need to reduce inequality or to share productivity more equitably; nor
are wages, collective bargaining and the wage share mentioned. The strategy’s
International Labour Review
342
rationale is that inclusive growth allows as many people as possible to share in
growth by having a job, irrespective of the wage levels of these jobs. And for
those who really cannot get a job, there should be social policy.
Over time, the EU’s approach to inclusive growth has changed somewhat.
Inequality has come to figure more prominently. For example, the 2017 Annual
Growth Survey states that “the Commission calls on Member States to redouble
their efforts on the three elements of the virtuous triangle of economic policy,
and in so doing, put the focus on social fairness to deliver more inclusive growth:
boosting investment; pursuing structural reforms; and ensuring responsible fis-
cal policies” (European Commission 2016, 5). Here, labour market issues fall
under “structural reforms”. The reforms continue, however, to be focused on
efficient labour markets and social policy instead of reducing income inequal-
ity. The survey does mention the need for wages to become more closely aligned
with productivity, but not out of a concern for inequality; rather, it refers to con-
cerns about competitiveness and aggregate demand. Moreover, in line with EU
tradition, the document underlines the importance of a role for the social part-
ners in national policymaking through social dialogue, but ignores the differ-
ences in power and interests between workers and employers and the possible
role of collective bargaining in reducing inequalities.
The 2019 Annual Growth Survey largely conveys the same message. How-
ever, wage growth and collective bargaining start to get some attention:
Wage growth, resulting from increased productivity, can reduce inequalities and
support upward convergence of living standards. Real wage developments con-
tinued to trail behind productivity in 2017 on average, following a longer-term
trend. In a context of declining collective bargaining coverage, policies enhancing
the institutional capacity of social partners could be beneficial in countries where
social dialogue is weak or has been negatively affected by the crisis. (European
Commission 2018, 11)
In rather vague terms, the European Commission, as the executive branch
of the EU, thus acknowledges the link between limited collective bargaining
coverage and the continued lag of wage growth behind productivity growth.
Nevertheless, it proposes no policies to curb this trend by strengthening
collective bargaining or union density.
2.2. The International Monetary Fund
The IMF has also been promoting inclusive growth in its flagship reports over the
past decade or so, discussing the need to address growing inequalities in income
and wealth (IMF 2007a, 2007b, 2017a and 2019). These reports have devoted
ample space to analysing the fall of the wage share in advanced economies and
causes of this. The April 2017 World Economic Outlook discusses a number of
factors that have caused this downward trend in the wage share, including tech
-
nological change and globalization, as well as declining unionization rates and
labour bargaining power (IMF 2017b, 121–172). Thus, the importance that labour
market institutions may have is confirmed in the analyses of the IMF’s research
departments. Moreover, the 2017 Fostering Inclusive Growth report considers the
wage share in its analysis, arguing that “The decline in the global labor share of
Missing links in the inclusive growth debate 343
income has generally implied higher income inequality” (IMF 2017a, 14). How-
ever, collective bargaining and power differences between workers and employ-
ers do not feature at all.
More generally, unionization and collective bargaining are entirely absent in
the policy recommendations of these reports concerning inclusive growth, and
the latter offer no consistent strategy to stop the downward trend in the wage
share (and thereby halt the rise in inequalities). The IMF proposes wage subsid-
ies for low-wage workers instead of a policy that would alter the functional
income distribution or labour’s bargaining power (IMF 2017a). Its 2020 G20 re-
port does include a policy to counter the decline in the wage share, proposing
to achieve this through reforms to strengthen competition and reduce corporate
market power (IMF 2020). But it again neglects any policy proposal on union
power or collective bargaining.
2.3. The Organisation for Economic Co-operation
and Development
In the case of the OECD, an evolution in analysis and policy content can be ob-
served over the years. Its 2008 report Growing Unequal? Income Distribution
and Poverty in OECD Countries includes an entire chapter analysing the links
between earnings and income inequality, but unions and collective bargaining
are not discussed. The report works towards two policy strategies aimed to re-
duce poverty and inequality, one concerning redistribution by governments and
the other “making the distribution of market income less unequal”, where “the
main instrument … is that of increasing the level of employment and spreading
work opportunities across a larger number of households” (OECD 2008, 302).
By 2019, a number of OECD reports and research papers had established
that there is a positive relationship between the coverage and coordination of
collective bargaining, on the one hand, and wage equality and employment, on
the other (OECD 2011, 2018 and 2019; Denk 2015). Moreover, the report Nego-
tiating Our Way Up: Collective Bargaining in a Changing World of Work (OECD
2019) explicitly analyses the relationship between collective bargaining and in-
clusive growth. It shows that unions and employers, through collective bargain-
ing, affect labour market outcomes (principally wage inequality, productivity
and employment) and therefore the inclusiveness of growth. One of its main
conclusions is that
The need for co-ordination and negotiation mechanisms between employers and
workers is heightened in the changing world of work. Whether considering key
issues such as wage inequality, job quality, workplace adaptation to the use of new
technologies, or support for workers displaced by shifts in industries, collective
bargaining and workers’ voice can complement public policies to produce tailored
and balanced solutions. (OECD 2019, 13)
In 2018, the OECD had published its report Opportunities for All: A Frame-
work for Policy Action on Inclusive Growth, which acknowledges the key role of
collective bargaining in determining functional income distribution and wage
inequality (OECD 2018). However, again this issue plays a very minor role in
the policy framework proposed by the report. The framework has three pillars:
investing in people and places that have been left behind; supporting business
International Labour Review
344
dynamism and inclusive labour markets; and building efficient and responsive
governments (OECD, 20). The role of wages is mentioned only as a reminder that,
for firms to be innovative, an increase in real wages similar to that of productiv-
ity may be necessary (21–24). Collective bargaining does not figure in the three
pillars; it is stated only that “the role of social partners and other stakeholders
could be strengthened to ensure the creation of quality jobs and non-discrimin-
ation in the workplace, as well as to facilitate a smooth transition towards the
future of work” (23).
We can therefore summarize this section as follows. The EU, the IMF and
the OECD have committed themselves to inclusive growth, arguing that growth
alone is not sufficient to ensure well-being: growth should be to the benefit of
all and not just of a few, and should contribute to reducing inequality. Initially,
they considered that the growth of inequality was fundamentally a consequence
of globalization, technological change, market failures, access to employment,
and educational differences. However, in recent years, fuelled by the academic
literature and their own research, the three institutions have gradually and to
varying degrees incorporated in their analyses labour market institutions – the
coverage and coordination of collective bargaining, and the strength of trade
unions – and the declining wage share trend. However, these issues have not
been integrated in their inclusive growth policies and no proposals have been
put forward to increase inclusiveness through strengthening the position of
trade unions or through a wider coverage and better coordination of collective
bargaining.
3. Missing links and contradictions in the inclusive
growth debate
In this section we address the missing links and contradictions in the inclusive
growth strategies promoted by the aforementioned international institutions,
underlining the need to include functional income distribution and labour
market institutions in the design of inclusive growth policies.
3.1. Functional income distribution
The tendency of the wage share to fall reflects the decoupling between product-
ivity growth and increases in real employee compensation (Pasimeni 2018; Biv-
ens and Mishel 2015; Stansbury and Summers 2017). As can be seen in figure 3,
whereas productivity and real wages increased at the same speed in the OECD
countries until the early 1980s, since then the gap between the two has been
widening steadily, indicating that productivity gains are no longer fully trans-
lated into wage growth. This decoupling occurs later in the case of the United
States, around the year 2000.
This evolution of functional income distribution is not neutral as far as in-
equality is concerned. There is an increasing body of literature explaining how
the increase in income inequality experienced by OECD countries over the last
few decades is connected to the downward trend in the wage share (Moore et
al. 2019; Jacobson and Occhino 2012; Piketty 2014). Empirical evidence shows
Missing links in the inclusive growth debate 345
that a higher capital share is associated with higher inequality in personal in-
come distribution, since capital is more concentrated than labour endowments
(Piketty 2014). In other words, since labour income is more uniformly distrib-
uted across households than capital income, the decline in the wage share con-
centrates total income at the top of the distribution.
Note: 1970 = 100 (1980 = 100 for Japan due to statistical availability). EU-15 refers to the Member States of the EU
before the accession of ten further countries in 2004.
Source: AMECO database.
Figure 3. Productivity–wage gap in main OECD countries, 1970–2018
100
120
140
160
180
1970
Productivity (GDP per person employed)
200
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
EU-15
220
100
120
140
160
180
1970
200
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
United States
220
100
120
140
160
180
1980
200
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
2018
Japan
Real compensation per employee
International Labour Review
346
According to Jacobson and Occhino (2012), the decline in the wage share
from 1979 to 2007 raised the Gini index by 2.3 percentage points in the case of
the US economy. As can be seen in figure 4, this inverse relationship can also be
demonstrated with our data sample, a panel of all OECD and EU economies for
the period 1990–2018. Thus, the wage share decline seems to go hand in hand
with increased income inequality.
Moreover, the fall in aggregate wage share masks a number of very different
effects on the various wage brackets. According to some estimates (OECD 2012;
Piketty 2014), the fall in the wage share is more pronounced when the richest
1 per cent of the population is not considered. In particular, the OECD (2012)
points to a fall in the wage share of 99 per cent of income earners in advanced
economies, whereas it has increased by 20 per cent for the top income popula-
tion (the top 1 per cent).
Any policies that seek to foster inclusive growth should therefore take into
account this aspect of functional income distribution. However, the strategies
designed by the EU, the IMF and the OECD not only fail to take into account the
importance of the functional income distribution but are often contradictory
and contribute to further accelerating the decline of the wage share. This was
true of the wage devaluation policies that were promoted by the IMF and the EU
during the Eurozone debt crisis with the aim of reducing real unit labour costs
to regain external competitiveness.
In 2012, the Directorate-General for Economic and Financial Affairs of the
European Commission explicitly argued that, in order to achieve “better employ
-
ment outcomes”, the coverage and extension of collective agreements should be
decreased and collective bargaining decentralized, while minimum wages and
the wage-setting power of trade unions should be reduced (European Commis-
sion 2012, 104). To this end, important labour market reforms have been carried
out in the intervening years in those EU countries that received bailout loans
Figure 4. Wage share and income inequality in OECD and EU countries, 1990–2018
0.45
0.50
0.55
0.60
0.40
0.20
0.25
0.30
0.35
Source: Authors’ calculations based on data from the AMECO database and the World Inequality Database.
Gini index
Wage share (% of GDP)
100
20 30 40 50 60 70 80 90
y = –0.0016x + 0.5148
R2 = 0.0506
Missing links in the inclusive growth debate 347
– Greece, Ireland, Portugal and Spain – triggering a strong erosion of collective
bargaining in these economies (Marginson 2015; Koukiadaki, Távora and Mar-
tínez Lucio 2016).
These reforms – in picturing collective labour relations as obstacles to mar-
ket functioning – failed to address the wage-related dimension of inclusive
growth (Keune 2015; Visser 2016). As can be seen in figure 5, the downward
trend in the wage share has accelerated precisely in those EU countries where
these drastic reforms were adopted between 2010 and 2012.
Hence, important contradictions emerge in the debate on inclusive growth.
Whereas, on the one hand, international institutions are formally committed
to achieving this goal, on the other, they promote policies that foster yet more
decline in the wage share, thereby contributing to increased income inequality.
3.2. Labour market institutions
The role that labour market institutions play in the evolution of both functional
and personal income distribution is the second major blind spot in the inclusive
growth strategies designed by international institutions. As we have already said,
labour market institutions are rarely incorporated into the analyses that explain
the determinants of inequality. And even when they are, they are not then taken
into account in policy design. However, both the academic literature and empir-
ical evidence suggest that these institutions must be considered.
The links between labour market institutions, functional distribution of in-
come, and economic inequality are well established.
Figure 5. Adjusted wage share in EU-27 countries
85
90
95
110
80
60
65
70
75
Note: 2008 = 100. EU-27 refers to the Member States of the EU before the accession of Croatia in 2013. Countries
where drastic reforms were adopted between 2010 and 2012 are indicated by way of comparison.
Source: AMECO database.
2008 2011 2012 2013 2014
100
105
20102009 2015 2016 2017 2018 2019
Greece EU-27 Portugal Spain
Ireland
International Labour Review
348
Several authors (Atkinson 2015; Wilkinson and Pickett 2009) point to the key
role of labour markets in creating inequality: “It is there that value is created and
divided between the various gradations of employees. It is there that the inequi-
ties which necessitate redistribution are set up” (Wilkinson and Pickett 2009,
249–250). They argue that these inequities have increased following a decline
in union membership and collective bargaining in recent decades.
Presumably, greater bargaining power for workers will lead to increases in
wages and in the wage share if the demand for labour is relatively inelastic. Em-
pirical research uses different proxy variables related to labour market institu-
tions to approximate the bargaining power of workers (employment protection
legislation, unemployment benefits, strike activity, bargaining coverage, union
density). Stockhammer (2009) finds that the decline in union density in recent
decades has had a negative effect on the evolution of the wage share, reflecting
the changes in bargaining power between capital and labour and the decreas-
ing capacity of workers to capture increases in productivity. Guschanski and
Onaran (2017), Fichtenbaum (2011) and Bengtsson (2014) also report a signifi-
cant positive effect of labour market institutions – particularly union density –
on the bargaining power of unions and on the wage share. Similarly, Pasimeni
(2018) finds that higher unemployment – used as a proxy variable for reduced
collective bargaining power among workers – is an important determinant of
the productivity–wage gap.
Trade union density is a variable that can, however, skew the results by under-
estimating the bargaining power of workers (union density is very low in some
countries where collective bargaining extension mechanisms reach a much higher
percentage of workers). If instead of considering union density, as most of these
studies do, we consider the coverage of collective bargaining, we can better cap-
ture the relationship between labour market institutions and the wage share.
In recent years, these findings have also been supported by a number of
studies from the IMF and the OECD. IMF researchers Jaumotte and Osorio Buit-
ron (2015) focus on the relationships between inequality and labour market
institutions, finding that de-unionization leads to greater inequality, since it is
strongly associated with an increase in top incomes. It weakens the bargaining
power of unions and their influence on corporate decisions, limits the impact
of unions on public policy and reduces the weight of equality-oriented social
values. According to the authors, the decline in union density explains 40 per
cent of the average increase in the top 10 per cent income share in their sample
countries. Interestingly, this effect can be partly offset when collective bargain-
ing coverage is guaranteed by extension agreements, which indicates the import-
ance of these mechanisms in controlling increases in inequality.1 Furthermore,
an IMF paper by Abdih and Danninger (2017) on the US economy finds that a
fall in unionization (highly statistically significant) helps to explain around
20 per cent of the fall in the wage share.
Some OECD publications point in the same direction, arguing that declining
union power, lower bargaining coverage and bargaining decentralization all lead
1 Yet Jaumotte and Osorio Buitron (2015) contend that the effects are not the same in all coun-
tries, and argue that strong unions in Southern Europe have resulted in more rather than less in-
equality as a result of union power causing higher unemployment and a loss of competitiveness.
Missing links in the inclusive growth debate 349
to greater wage inequality (OECD 2011), and finding that collective bargaining
systems with greater coverage and bargaining systems that are not fully decen-
tralized are more correlated with lower wage inequality (OECD 2019). Further-
more, Denk (2015) – of the OECD Economics Department – points to the negative
correlation between the labour income share of the top 1 per cent and collective
bargaining coverage. Where larger parts of the workforce are covered by collect-
ive agreements, inequality is lower, since top earners are not allowed to increase
their work-based income as much and low wages are allowed to increase more.
The main arguments of this literature appear to correspond with the em-
pirical evidence derived from our panel of 42 advanced countries for the period
1990–2018, as can be seen below. In figures 6 to 10 we present pooled data that
show how collective bargaining coverage is associated with functional income
distribution, wage inequality and income inequality.
Figure 6 indicates that the coverage of collective bargaining is positively as-
sociated with the wage share: in countries and periods where coverage was com-
paratively higher, during 1990–2018, the wage share also tended to be higher.
This suggests that widespread collective bargaining helps to capture a greater
share of total income for workers and that collective agreements can be an im-
portant tool of income redistribution.
Collective bargaining affects functional income distribution, but the import-
ance of this institution for inequality is wider reaching. The literature has high-
lighted the strong relationship that exists between collective bargaining and both
wage inequality and personal income inequality, which reinforces its possible
role in achieving inclusive growth. Sectoral and industry-level bargaining and
high coverage rates are associated with lower within-industry earning differ-
ences for workers doing similar jobs, and therefore with lower levels of earn-
ings inequality (Hayter and Visser 2021).
Figures 7 and 8 show that there is indeed a significant negative correlation
between collective agreement coverage and wage inequality. Where the coverage
of collective agreements is high, the 90/10 inequality ratio tends to be lower (figure 7).
Added to this, the percentage of workers with low wages is also significantly
Figure 6. Collective bargaining coverage and wage share in OECD
and EU countries, 1990–2018 (percentages)
80
90
70
30
40
50
60
Source: Authors’ calculations based on data from the AMECO and OECD/AIAS ICTWSS databases.
Wage share (% of GDP)
Collective bargaining coverage (%)
100
20 30 40 50 60 70 80 90
y = 0.046x + 52.018
R2 = 0.0611
010
International Labour Review
350
smaller in countries and years where collective bargaining coverage is higher
(figure 8). We therefore see how collective bargaining sets limits to wage disper-
sion by preventing both very low and very high wages (Visser and Checchi 2009).
Through its link with both functional income distribution and wage inequal-
ity, collective bargaining coverage also evinces a clear negative association with
income inequality. As we can see in figure 9, Gini coefficients tend to be lower
Figure 7. Collective bargaining coverage and wage inequality in OECD
and EU countries, 1990–2018
5
6
7
8
4
0
1
2
3
Note: The 90/10 inequality ratio is defined as the ratio of the salary earned by individuals in the ninth decile to the
earnings of workers in the first decile.
Source: Authors’ calculations based on data from the OECD (decile ratios of gross earnings) and the OECD/AIAS
ICTWSS database.
9th/1st income decile ratio
Collective bargaining coverage (%)
100
20 30 40 50 60 70 80 90
y = –0.0181x + 4.3639
R2 = 0.396
020
Figure 8. Collective bargaining coverage and low pay incidence in OECD
and EU countries, 1990–2018 (percentages)
25
30
35
20
0
5
10
15
Note: Low pay incidence is defined as the percentage of employees earning less than two-thirds of gross median
earnings of all full-time workers.
Source: Authors’ calculations based on data from the OECD (decile ratios of gross earnings) and the OECD/AIAS
ICTWSS database.
Low-paid workers (%)
Collective bargaining coverage (%)
100
20 30 40 50 60 70 80 90
y = –0.1086x + 22.4
R2 = 0.3069
010
Missing links in the inclusive growth debate 351
where collective bargaining coverage is higher. Since income inequality is deter-
mined by many factors other than the distribution of productivity gains and the
evolution of wages (for example, household composition and other types of in-
come), this association appears to be somewhat weaker than the ones stated above.
Nonetheless, as can be seen in figure 10, there is a rather strong relation-
ship between wage inequality and income inequality. In other words, labour
market outcomes are crucial to any understanding of income inequality and
Figure 9. Collective bargaining coverage and income inequality in OECD
and EU countries, 1990–2018
0.50
0.60
0.70
1
0.40
0
0.10
0.20
0.30
Source: Authors’ calculations based on data from the World Inequality Database and the OECD/AIAS ICTWSS
database.
Gini index
Collective bargaining coverage (%)
100
030 40 50 60 70 80 90
y = –0.001x + 0.5014
R2 = 0.1674
0.80
0.90
2010
Figure 10. Wage inequality and income inequality, (OECD and EU countries,
1990–2018)
0.70
0.80
0.60
0.20
0.30
0.40
0.50
Source: Authors’ calculations based on data from the OECD (decile ratios of gross earnings) and the World Inequal-
ity Database.
Gini index
9th/1st income decile ratio
7
345 6
y = 0.0417x + 0.2904
R2 = 0.3485
12
International Labour Review
352
the lack of inclusive growth. This also means that labour market institutions
– in particular, solid collective bargaining – may prove vital to promoting more
equitable growth.
4. A panel data analysis for OECD and EU countries
In this section we estimate a panel data model for 42 advanced economies to
evaluate our two hypotheses: first, that the evolution of the wage share influ-
ences income inequality and, second, that collective bargaining coverage is a
significant determinant of both functional income distribution and personal in-
come distribution. In other words, we are interested in the causal links indi-
cated in figure 11.
4.1.   Model specication: Determinants of functional 
and personal income distribution
We estimate two different specifications in our analysis. In the first one the func-
tional income distribution will be the dependent variable, and in the second one
we shall analyse the determinants of personal income distribution.
Most of the literature on inequality and inclusive growth has focused on
analysing changes in personal income distribution, rather than changes in func-
tional income distribution; few studies link the former to the latter. In contrast,
we analyse this relationship and the role that collective bargaining plays in both
cases.
In the research that has been conducted on the determinants of functional
income distribution, neoclassical studies (Bentolila and Saint-Paul 2003; Euro-
pean Commission 2007, 237–260; IMF 2007a and 2017b) have traditionally
pointed to technological change as the main determinant of the evolution of the
wage share. According to this approach, in recent years technological change has
become capital augmenting rather than labour augmenting and has also been
skill biased. This is believed to have caused a shift in income distribution, from
labour to capital, and falling wage shares for low-skilled workers.
Figure 11. The role of labour market institutions in driving inequalities
Source: Authors’ elaboration.
Erosion of labour
market institutions
Wage share
Income inequality
Missing links in the inclusive growth debate 353
Educational achievement has also been considered in neoclassical studies
(Grömling and Klös 2019; IMF 2007a and 2007b, 31–65) as a possible determin-
ant of income distribution, since it explains the skills of the workforce and
therefore their ability to share in the benefits of technological change.
Likewise, globalization has been highlighted by some studies – both neoclas-
sical and post-Keynesian (IMF 2007a; Dünhaupt 2017; Stockhammer 2017) – as
a key driver of changes in income distribution. The main effect of international
trade on functional income distribution is the impact it has on the bargaining
positions of capital and labour in advanced countries. Globalization and trade
openness reinforce the negotiating power of the most mobile factor – capital –
versus the most static factor – labour – and increase the strategic options of the
former by allowing the international relocation of production.
Recent post-Keynesian studies have incorporated an additional determinant
when studying the evolution of wage share: financialization (Pariboni and Trid-
ico 2019; Alvarez 2015; Dünhaupt 2017). The growing importance that financial
activity has gained in developed economies in recent decades has affected in-
come distribution in several ways (Stockhammer 2013 and 2017). First, strategies
to maximize shareholder value have led to an increase in the payments made
by firms to shareholders (in the form of dividends), raising overhead costs and
putting downward pressure on the wage share. In addition, the expansion of
international capital markets has established a market for corporate control,
shifting corporate priorities from growth to profitability. The evolution from a
model based on “retain and reinvest” to a model based on “downsize and dis-
tribute” (Lazonick and O’Sullivan 2000) brings additional pressures on labour,
weakening its bargaining position.
Other authors, particularly from the field of political economy (Kristal 2010),
have explained that the downsizing of government activity and the reduction in
welfare state generosity that, to a greater or lesser extent, OECD countries have
experienced since the 1980s constitute another determining factor in the evo-
lution of income distribution. Welfare state retrenchment has led to a progres-
sive shift towards private provision of public services and an erosion of some
crucial supports of workers’ bargaining power (reducing the generosity of
the so-called “indirect wage” – education, health, pensions – and the duration
of unemployment subsidies).
The literature that has analysed the evolution of functional income distri-
bution has also paid attention to the role played by labour market institutions
in explaining the wage share. As mentioned above, various authors (Stockham-
mer 2009; Fichtenbaum 2011; Bengtsson 2014; Guschanski and Onaran 2017)
have observed the negative impact that the erosion of these institutions has had
on the evolution of the wage share. However, most of these investigations have
used union density as the key variable to proxy the role of labour market insti-
tutions. We consider that collective bargaining coverage can better capture the
scope of institutional bargaining between employers’ and workers’ organiza-
tions, since it does not underestimate the negotiating power of labour in those
countries where collective agreements and extension mechanisms reach a much
higher percentage of workers than are members of unions.
International Labour Review
354
We first estimate functional income distribution as a function of the previ-
ously mentioned determinants: technological change (tech), globalization ( glob),
financialization ( fin), human capital (humank), welfare state retrenchment (wel-
fare) and labour market institutions (lmi). In addition, and like the IMF (2007a
and 2007b), we check the extent to which these factors also explain personal in-
come distribution ( gini), incorporating in this second specification of our model
the effect of functional income distribution itself (wageshare). We present
both specifications in equations 1 and 2:
Specification1:
Wage share = f (tech, glob, fin, humank, welfare, lmi) (1)
Specification2:
Gini index = f (wageshare, tech, glob, fin, humank, welfare, lmi) (2)
4.2. Estimation equations, variables and econometric
methodology
The two baseline specifications are defined in equations 3 and 4, where i repre-
sents the individual countries surveyed and t the years:
Specification1:
wageshare i, t = α +β1(∆K/L) i, t + β2(∆tradeopen) i, t + β3(∆marketcap) i, t
+ β4(∆humank) i, t + β5(∆governm) i, t + β6(∆coverage) i, t. (3)
Specification 2:
gini i, t = α + β1(∆wageshare) i, t + β2(∆K/L) i, t + β3(∆tradeopen) i, t
+ β4(∆marketcap) i, t + β5(∆humank) i, t + β6(∆governm) i, t + β7(∆coverage) i, t. (4)
Here, we incorporate the specific variables that we use to proxy the aforemen-
tioned analytical dimensions. Table A1 in the Appendix shows the variables and
definitions used in our estimates, as well as the data sources from which the
time series have been obtained.
Functional income distribution is measured with the adjusted wage share
(wageshare). We measure technological change (K/L) through the evolution of the
capital–labour ratio (capital stock at current purchasing power parities (PPPs)
divided by the number of persons in the labour market), and we use trade open-
ness (imports plus exports, as a share of GDP) to capture the impact of global-
ization on income distribution (tradeopen). We approximate financialization by
measuring the market capitalization of listed domestic companies as a percent-
age of GDP (marketcap), and we also incorporate a human capital index into
our equation, based on the average years of schooling and an assumed rate of
return to education (humank). Government activity and welfare state retrench-
ment are measured using government expenditure in relation to GDP ( governm).
Finally, collective bargaining coverage is defined as the proportion of all wage
and salary earners in employment covered by collective bargaining agreements
(coverage). In addition, in the second specification of the model (equation 4), we
measure personal income distribution through the Gini coefficient of market
pre-tax income ( gini).
Table A2 in the Appendix shows the basic descriptive statistics of our vari-
ables. Our baseline specifications are the result of a delicate equilibrium:
Missing links in the inclusive growth debate 355
maintaining a large sample of countries and years – together with Pariboni and
Tridico’s (2019), ours is one of the most extensive samples in the literature – and
including variables that have significance in explaining income distribution at
national level. Other similar studies have used somewhat more specific vari-
ables to approximate the dimensions presented here (for example, dividends
distributed by non-financial companies as a proxy for financialization, or social
transfers to measure welfare state retrenchment). However, the use of such vari
-
ables would entail giving up one of the main strengths of our research: using a
large sample of countries for a period of almost three decades.
Our panel data provide a strongly balanced panel, made up of 42 advanced
countries, with data for the period 1990–2018. As we said in the introduction,
the countries that make up our sample are the 37 OECD Member countries, to-
gether with those EU countries that are not part of the OECD. In our estimation
we use yearly observations.
Our initial hypotheses about the causal relationships in the model are the
following:
Specification1:
(∆wageshare)K/L < 0; (∆wageshare)tradeopen < 0; (∆wageshare)marketcap < 0;
(∆wageshare)humank > 0; (∆wageshare)governm > 0; (∆wageshare)coverage > 0 (5)
Specification2:
(∆gini )∆wageshare < 0; (∆gini )K/L < 0; (∆gini )tradeopen < 0; (∆gini )marketcap < 0;
(∆gini )humank > 0; (∆gini )governm > 0; (∆gini )coverage > 0 (6)
We expect the evolution of wage share to be negatively related to techno-
logical change, trade openness and financialization, as the literature indicates. In
contrast, we expect the wage share to be positively related to the human capital
index, government activity and collective bargaining coverage.
In relation to the second specification of our model, we expect the determin-
ants of personal income distribution to behave in a similar way and with the
same sign (see equation 6). However, we assume that the link between these de-
terminants and the Gini index is probably weaker and less significant than the
relationship in the first specification. Furthermore, and this is the key point of
the second specification, we expect the evolution of the wage share to be nega-
tively related to the Gini index.
When dealing with time series cross-sectional data for a broad sample of
42 countries over a period of almost 30 years, one has to consider using fixed
effects in order to capture unobserved effects. The use of fixed effects allows
one to control for time-invariant unobserved heterogeneity that is not captured
by the explanatory variables. That is, the fixed effects model assumes that there
is a different constant term for each individual and that individual effects are
independent of one another.
However, our sample shows the usual problems when working with time
series cross-sectional data. First, both the Levin–Lin–Chu and the Im–Pesaran–
Shin tests for panel data indicate the presence of unit roots, which cause non-
stationarity in the series and the corresponding risk of spurious correlations.
Second, when we use the modified Wald test for groupwise heteroskedasticity,
International Labour Review
356
we have to reject the null hypothesis of constant variance. Third, the Wooldridge
test shows the presence of autocorrelation in our sample. Furthermore, we test
for possible multicollinearity, but the variance inflation factor method reveals
the absence of a linear relationship between the model variables.
We correct for the presence of unit roots in our sample by estimating the
model in first differences, which eliminates the problem. It is also necessary to
use an estimation method that corrects the heteroskedasticity and autocorrel-
ation errors for panel data. The most efficient way is to model the functional
form of both the heteroskedasticity and autocorrelation, so that more efficient
and precise estimates can be obtained for the parameters (Cameron and Trivedi
2009). For this purpose, we can use feasible generalized least squares (FGLS)
and also panel-corrected standard errors (PCSE) estimations. Both methods
correct standard errors for contemporaneous correlated and heteroskedastic
errors. When we use FGLS we model the autocorrelation using a panel-specific
first-order autoregressive structure, and when we estimate using PCSE we apply
a Prais–Winsten transformation. We present both FGLS and PCSE estimates, al-
though the results of our estimates are not affected by the method chosen.
4.3. Econometric results
The results of our estimates are consistent with our hypotheses and also with
the previous literature, for both FGLS and PCSE. As can be seen in table 1, the
evolution of the wage share is negatively related to trade openness and to fi-
nancialization, as previously detected by Dünhaupt (2017) and by Pariboni and
Tridico (2019). Both variables are strongly significant.
On the other hand, and as Stockhammer (2013 and 2017) also confirms, the
impact of technological change on wage share is not very robust, being weakly
Table 1. Functional income distribution estimation results
Variables FGLS PCSE
K/L –0.00294 –0.00564*
(0.00206) (0.00318)
tradeopen –4.006**** –3.82****
(0.551) (0.762)
marketcap –0.00535**** –0.0072****
(0.00143) (0.00195)
humank 6.299*7.494*
(3.259) (4.133)
governm 0.128**** 0.139****
(0.0161) (0.0267)
coverage 0.0219*** 0.029**
(0.00779) (0.0125)
N714 714
R2 0.136
*, **, *** and **** indicate statistical signicance at the 10, 5, 1 and 0.1 per cent levels, respectively.
Notes: Dependent variable = wageshare. Standard errors appear in parentheses. For a description of the
variables, see table A1 in the Appendix.
Source: Authors’ calculations based on data from the sources listed in table A1 in the Appendix.
Missing links in the inclusive growth debate 357
negative and significant only at the 10 per cent significance level in the PCSE
estimate. Contrary to what some neoclassical analyses have suggested (Bentolila
and Saint-Paul 2003; IMF 2007a), technological change does not seem to be a
good determinant of the evolution of functional income distribution (or of per-
sonal income distribution, as we see below).
Government activity in the economy – a proxy for welfare state retrench-
ment – is positively related to the wage share, as is the human capital index,
although the first variable is much more significant than the second. These re-
sults also mirror the literature: Kristal (2010) and Dünhaupt (2017) show how
welfare state retrenchment has had a strong negative impact on the evolution
of the wage share, and Harrison (2005) also finds that wage shares are driven
by changes in government spending. Furthermore, the IMF (2007a) points to the
importance of education in preventing reductions in the wages of low-skilled
workers.
Finally, collective bargaining coverage is statistically significant and shows
the expected sign: greater coverage leads to a greater wage share. Previous re-
search has reported a significant positive effect of union density on the wage
share (Guschanski and Onaran 2017; Fichtenbaum 2011; and Bengtsson 2014).
We also observed this positive effect for collective bargaining coverage. The im-
portance of collective bargaining and collective agreements as determinants of
functional income distribution is therefore confirmed.
In our second estimation we show how the determinants chosen to explain
functional income distribution are also significant in most cases when one seeks
to explain personal income distribution (see table 2). However, and as we ex-
pected, some of them are not significant – for example, technological change and
Table 2. Personal income distribution estimation results
Variables FGLS PCSE
wageshare –0.00177**** –0.00154*****
(0.000278) (0.000362)
K/L –0.0000108 –0.0000381
(0.0000167) (0.0000238)
tradeopen 0.00399 0.00413
(0.00511) (0.00706)
marketcap 0.0000651**** 0.0000681***
(0.0000155) (0.0000223)
humank –0.0722** –0.0773*
(0.0323) (0.0462)
governm –0.000286*–0.000427**
(0.000148) (0.000183)
coverage –0.000185*** –0.000162**
(0.0000617) (0.0000736)
N558 558
R20.161
*, **, *** and **** indicate statistical signicance at the 10, 5, 1 and 0.1 per cent levels, respectively.
Notes: Dependent variable = ∆gini. Standard errors appear in parentheses. For a description of the variables,
see table A1 in the Appendix.
Source: Authors’ calculations based on data from the sources listed in table A1 in the Appendix.
International Labour Review
358
trade openness. Changes in financialization, human capital index and welfare
state retrenchment continue to be significant in this second specification of our
model and do maintain the expected signs.
Two aspects are particularly noteworthy in the second specification. First,
the wage share is a strongly significant determinant of the evolution of the Gini
index: this index increases when the wage share declines, as has previously been
reported in the literature (Moore et al. 2019; Jacobson and Occhino 2012). Sec-
ond, strong collective bargaining coverage not only positively affects the wage
share, as we have seen in table 1, but also reduces personal income inequality
(see table 2).
Therefore, functional income distribution matters when it comes to explain-
ing the evolution of personal income distribution. Likewise, labour market in-
stitutions – in this case, collective bargaining coverage – are not unrelated to
personal income distribution: greater bargaining power of workers seems to be
related to a less skewed distribution of income between capital and labour and,
therefore, to less personal income inequality.
5. Conclusions
In this article, we have sought to contribute to the inclusive growth debate pro-
moted by the IMF, the OECD and the EU in recent years. To this end, we have
examined the extent to which the policy proposals made by these institutions
are consistent with the objective of growth that reduces inequality. We have also
estimated an econometric model using a panel of 42 advanced countries for the
period 1990–2018 to analyse the determinants of both functional and personal
income distribution.
We obtain two important results from our estimates: functional income dis-
tribution has indeed a significant influence on the evolution of income inequal-
ity, and collective bargaining coverage plays a crucial role in explaining both
declines in the wage share and increases in income inequality. Taking into ac-
count these results of our econometric model, and after reviewing the main re-
ports, analyses and policy proposals of the three international institutions, we
can conclude that there are at least three major shortcomings in the strategies
of these institutions.
First, the analyses and reports of these institutions have long overlooked the
role of functional income distribution – the distribution of productivity gains be-
tween capital and labour – when explaining inequality. Although this omission
has been partially corrected in recent years as this dimension has been incorp-
orated into the research agenda of these institutions, this relationship has not
been integrated into their policy strategies, which makes the strategies of these
institutions less capable of truly achieving inclusive growth.
Second, the policy strategies of these institutions have also disregarded the
important role that labour market institutions – particularly collective bargain-
ing – can play in promoting growth in which gains are more fairly and evenly
spread throughout society. What is more, the strategies of these institutions
have been counterproductive, recommending labour market reforms that have
eroded collective bargaining coverage (as in the Southern European countries
Missing links in the inclusive growth debate 359
over the last decade). However, the results of our research are clear in this re-
gard: greater coverage of collective bargaining leads to a higher wage share and
lower levels of personal income inequality. Therefore, the promotion of labour
market reforms that reduce collective bargaining coverage will most likely
undermine inclusive growth.
This also means that the trend experienced by some labour market institu-
tions in recent decades should be worrying for all those who seek to promote
an inclusive growth strategy. As stated in the introduction, over the past five
decades the average proportion of employees covered by collective bargaining
agreements in OECD countries has decreased by more than 10 percentage points,
from 59.4 to 48.1 per cent.
Third, the policy strategies of the OECD, the IMF and the EU ignore their
own research departments when it comes to inclusive growth. In these research
departments, several studies in recent years have reached conclusions similar
to those raised here (Jaumotte and Osorio Buitron 2015; Abdih and Danninger
2017; Denk 2015). However, these conclusions have not been incorporated into
the economic policy of these institutions.
The results of our research, like some of the results of the studies just men-
tioned, have important policy implications and suggest the need to seriously
reconsider the strategies adopted by these international institutions. First, our
research clearly points to the need to halt and reverse the fall in the wage share
in order to reduce inequality. Second, these international institutions need to
take into account the positive effect that a strengthening of collective bargaining
can have in achieving inclusive growth. Since labour market institutions, and
particularly collective bargaining, are located where the distribution of prod-
uctivity gains takes place, they can be a key lever in achieving a new model of
income distribution. As long as international institutions continue to present
labour market institutions as mere “rigidities”, their policy strategies will con-
tinue to fail to achieve inclusive growth.
References
Abdih, Yasser, and Stephan Danninger. 2017. “What Explains the Decline of the U.S.
Labor Share of Income? An Analysis of State and Industry Level Data”, IMF Work-
ing Paper WP/17/167. Washington, DC: International Monetary Fund.
Alvarez, Ignacio. 2015. “Financialization, Non-financial Corporations and Income In-
equality: The Case of France”. Socio-economic Review 13 (3): 449–475.
Atkinson, Anthony B. 2015. Inequality: What Can Be Done? Cambridge, MA: Harvard
University Press.
Bengtsson, Erik. 2014. “Do Unions Redistribute Income from Capital to Labour? Union
Density and Wage Shares since 1960”. Industrial Relations Journal 45 (5): 389–408.
Bentolila, Samuel, and Gilles Saint-Paul. 2003. “Explaining Movements in the Labor
Share”. B.E. Journal of Macroeconomics 3 (1): 1–33.
Berg, Andrew G., and Jonathan D. Ostry. 2011. “Inequality and Unsustainable Growth:
Two Sides of the Same Coin?” IMF Staff Discussion Note SDN/11/08. Washington,
DC: International Monetary Fund.
Bivens, Josh, and Lawrence Mishel. 2015. “Understanding the Historic Divergence
between Productivity and a Typical Worker’s Pay: Why It Matters and Why It’s
Real”, EPI Briefing Paper No. 406. Washington, DC: Economic Policy Institute.
International Labour Review
360
Cameron, Colin A., and Pravin K. Trivedi. 2009. Microeconometrics Using Stata. College
Station, TX: Stata Press.
Checchi, Daniele, and Cecilia García-Peñalosa. 2010. “Labour Market Institutions and
the Personal Distribution of Income in the OECD”. Economica 77 (307): 413–450.
Denk, Oliver. 2015. “Who Are the Top 1% Earners in Europe?” OECD Economics Depart-
ment Working Papers, No. 1274. Paris: OECD.
Dünhaupt, Petra. 2017. “Determinants of Labour’s Income Share in the Era of Finan-
cialisation”. Cambridge Journal of Economics 41 (1): 283–306.
European Commission. 2007. Employment in Europe 2007. Luxembourg: Office for Offi-
cial Publications of the European Communities.
––––––. 2010. EUROPE 2020: A Strategy for Smart, Sustainable and Inclusive Growth.
COM(2010) 2020 final. Brussels.
––––––. 2012. “Labour Market Developments in Europe 2012”, European Economy Series,
No. 5/2012. Luxembourg: Publications Office of the European Union.
––––––. 2016. Annual Growth Survey 2017. COM(2016) 725 final. Brussels.
––––––. 2018. Annual Growth Survey 2019: For a Stronger Europe in the Face of Global
Uncertainty. COM(2018) 770 final. Brussels.
Eurostat. 2019. Smarter, Greener, More Inclusive? Indicators to Support the Europe 2020
Strategy. Luxembourg: Publications Office of the European Union.
Fichtenbaum, Rudy. 2011. “Do Unions Affect Labor’s Share of Income: Evidence Using
Panel Data”. American Journal of Economics and Sociology 70 (3): 784–810.
Grömling, Michael, and Hans-Peter Klös. 2019. “Inclusive Growth: Institutions Matter!”
Intereconomics 54 (3): 184–192.
Guschanski, Alexander, and Özlem Onaran. 2017. “The Political Economy of Income Dis-
tribution: Industry Level Evidence from 14 OECD Countries”, Greenwich Papers in
Political Economy, No. GPERC51. London: University of Greenwich.
Harrison, Ann. 2005. “Has Globalization Eroded Labor’s Share? Some Cross-Country Evi-
dence”, MPRA Paper No. 39649. Munich: University Library of Munich.
Hayter, Susan, and Jelle Visser. 2021. “Making Collective Bargaining More Inclusive: The
Role of Extension”. International Labour Review 160 (2): 169–195.
ILO. 2019. The Global Labour Income Share and Distribution: Key Findings. Geneva.
IMF (International Monetary Fund). 2007a. World Economic Outlook, April 2007: Spill-
overs and Cycles in the Global Economy. Washington, DC.
––––––. 2007b. World Economic Outlook, October 2007: Globalization and Inequality.
Washington, DC.
––––––. 2017a. Fostering Inclusive Growth. Washington, DC.
––––––. 2017b. World Economic Outlook April 2017: Gaining Momentum? Washington, DC.
––––––. 2019. World Economic Outlook, April 2019: Growth Slowdown, Precarious Recov-
ery. Washington, DC.
––––––. 2020. G-20 Report on Strong, Sustainable, Balanced and Inclusive Growth. Wash-
ington, DC.
Jacobson, Margaret, and Filippo Occhino. 2012. “Labor’s Declining Share of Income
and Rising Inequality”, Economic Commentary No. 2012-12. Cleveland, OH: Fed-
eral Reserve Bank of Cleveland.
Jaumotte, Florence, and Carolina Osorio Buitron. 2015. “Inequality and Labor Market
Institutions”, IMF Staff Discussion Note SDN/15/14. Washington, DC: International
Monetary Fund.
Keune, Maarten. 2015. “The Effects of the EU’s Assault on Collective Bargaining: Less
Governance Capacity and More Inequality”. Transfer: European Review of Labour
and Research 21 (4): 477–483.
Koukiadaki, Aristea, Isabel Távora, and Miguel Martínez Lucio, eds. 2016. Joint Regu-
lation and Labour Market Policy in Europe during the Crisis. Brussels: European
Trade Union Institute.
Missing links in the inclusive growth debate 3 61
Kristal, Tali. 2010. “Good Times, Bad Times: Postwar Labor’s Share of National Income
in Capitalist Democracies”. American Sociological Review 75 (5): 729–763.
Kuznets, Simon. 1955. “Economic Growth and Income Inequality”. American Economic
Review 45 (1): 1–28.
Lazonick, William, and Mary O’Sullivan. 2000. “Maximizing Shareholder Value: A New
Ideology for Corporate Governance”. Economy and Society 29 (1): 13–35.
Marginson, Paul. 2015. “Coordinated Bargaining in Europe: From Incremental Corro-
sion to Frontal Assault?” European Journal of Industrial Relations 21 (2): 97–114.
Moore, Sian, Ozlem Onaran, Alexander Guschanski, Bethania Antunes, and Graham
Symon. 2019. “The Resilience of Collective Bargaining: A Renewed Logic for Joint
Regulation?” Employee Relations 41 (2): 279–295.
OECD (Organisation for Co-operation and Development). 2008. Growing Unequal? Income
Distribution and Poverty in OECD Countries. Paris.
––––––. 2011. Divided We Stand: Why Inequality Keeps Rising. Paris.
––––––. 2012. OECD Employment Outlook 2012. Paris.
––––––. 2015. All on Board: Making Inclusive Growth Happen. Paris.
––––––. 2017a. Bridging the Gap: Inclusive Growth 2017 Update Report. Paris.
––––––. 2017b. “Time to Act: Making Inclusive Growth Happen”, Policy Brief. Paris.
––––––. 2018. Opportunities for All: A Framework for Policy Action on Inclusive Growth.
Paris.
––––––. 2019. Negotiating Our Way Up: Collective Bargaining in a Changing World of
Work. Paris.
Ortiz, Isabel, Matthew Cummins, Jeronim Capaldo, and Kalaivani Karunanethy. 2015.
“The Decade of Adjustment: A Review of Austerity Trends 2010–2020 in 187 Coun-
tries”, ESS Working Paper No. 53. Geneva: ILO.
Pariboni, Riccardo, and Pasquale Tridico. 2019. “Labour Share Decline, Financialisation
and Structural Change”. Cambridge Journal of Economics 43 (4): 1073–1102.
Pasimeni, Paolo. 2018. “The Relation between Productivity and Compensation in Europe”,
European Economy Discussion Paper No. 079. Luxembourg: Publications Office of
the European Union.
Piketty, Thomas. 2014. Capital in the Twenty-First Century, translated by Arthur Gold-
hammer. Cambridge, MA: Harvard University Press.
Stansbury, Anna M., and Lawrence H. Summers. 2017. “Productivity and Pay: Is the
Link Broken?” NBER Working Paper No. 24165. Cambridge, MA: National Bureau
of Economic Research.
Stockhammer, Engelbert. 2009. “Determinants of Functional Income Distribution in
OECD Countries”, IMK Study 5/2009. Düsseldorf: Macroeconomic Policy Institute.
––––––. 2013. “Why Have Wage Shares Fallen? A Panel Analysis of the Determinants
of Functional Income Distribution”, Conditions of Work and Employment Series
No. 35. Geneva: ILO.
––––––. 2017. “Determinants of the Wage Share: A Panel Analysis of Advanced and Devel-
oping Economies”. British Journal of Industrial Relations 55 (1): 3–33.
Vaughan-Whitehead, Daniel, ed. 2015. The European Social Model in Crisis: Is Europe
Losing Its Soul? Cheltenham: Edward Elgar.
Visser, Jelle. 2016. “What Happened to Collective Bargaining during the Great Reces-
sion?” IZA Journal of Labor Policy 5: 1–35.
Visser, Jelle, and Daniele Checchi. 2009. “Inequality and the Labour Market: Unions”.
In The Oxford Handbook of Economic Inequality, edited by Wiemer Salverda, Brian
Nolan and Timothy M. Smeeding, 230–256. Oxford: Oxford University Press.
Wilkinson, Richard, and Kate Pickett. 2009. The Spirit Level: Why Greater Equality Makes
Societies Stronger. New York, NY: Bloomsbury Press.
International Labour Review
362
Appendix
Table A1. Variables, definitions and sources
Analytical dimension Variable Denition Source
Functional income
distribution
Adjusted wage share
(wageshare)
Adjusted wage share,
total economy as a
percentage of GDP
at current prices.
AMECO database
Personal income
distribution
Gini coecient
(gini)
Gini coecient, pre-
tax national income.
World Inequality
Database
Technological
change
Capital/labour ratio
(K/L)
Capital stock at
current PPPs divided
by the number of
persons engaged in
the labour market.
Penn World Table,
version 9.1
Education Human capital index
(humank)
Human capital index
based on years
of schooling and
returns to education.
Penn World Table,
version 9.1
Globalization Trade openness
(tradeopen)
Imports plus exports
of goods and services,
as a percentage of
GDP.
OECD, National
Accounts at a
Glance
Financialization Market capitalization
(marketcap)
Market capitalization
of listed domestic
companies, as
a percentage of GDP.
World Development
Indicators, World
Bank
Welfare state Government
expenditure
(governm)
Total expenditure
of general government,
as a percentage of GDP.
OECD National
Accounts at a
Glance
Labour market
institutions
Collective bargaining
coverage
(coverage)
Percentage of all wage
and salary earners
in employment covered
by collective bargaining
agreements.
OECD/AIAS ICTWSS
database
Table A2. Summary statistics
Variable Observations Mean Standard
deviation
Minimum Maximum
wageshare 1 109 54.090 6.206 34.002 87.421
gini 925 0.4353 0.0670 0.2311 0.7124
K/L 1 148 241 794.5 131 484.1 8 899.9 690 601.9
humank 1 148 3.092 0.403 1.802 3.807
tradeopen 1 209 0.896 0.567 0.160 4.084
marketcap 888 60.818 48.449 0.046 326.359
governm 1 051 42.577 8.380 15.124 65.084
coverage 1 218 53.68 31.32 0.80 100.00
Source: Authors’ calculations based on data from the sources listed in table A1.
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Article
The aim of the concept of inclusive growth is to derive political recommendations on the basis of selected indicators so that a maximum number of socio-economic groups may benefit from the economic progress of a country. On the one hand, this initiative continues the long-running debate on alternative welfare measures beyond GDP-per-capita. On the other, the discussion on the relationship between growth and distribution is being revitalised. This article deals with the potentials and limitations of the concept of inclusive growth. Conventional inclusive growth concepts are less embedded within a growth context and not focused on political frameable institutions. The indicators focus on results or outcome rather than on growth drivers. In addition, outcome variables are mixed with supply-side indicators. Therefore, the conventional concepts should be further developed into an institution-oriented ‘inclusive growth accounting’ with a focus on the growth factors of labour, education and capital.
Article
Labor income has been declining as a share of total income earned in the United States for the past three decades. We look at the past effect of the labor share decline on income inequality, and we study the likely future path of the labor share and its implications for inequality.
Article
The purpose of this article is to explain the determinants behind the decline of labour share in the last three to four decades in OECD countries. In our view, this decline was determined by financialisation and was deepened by the structural changes that occurred almost simultaneously in those economies. Financialisation, or finance-dominated capitalism, from the 1980s onwards, was a key element in the strategic offensive of the advanced countries' dominant classes to appropriate higher shares of national income and to restore their control over the political process, a control that had been threatened by a generalised advancement of the labour movement in the 1970s. The development of a finance-dominated capitalism was helped by the process of global-isation, which affected not only OECD countries but also many others. A new, though unstable, macroeconomic model emerged, which we will call financial capitalism. In financial capitalism, trade unions lost power vis-à-vis capital, labour flexibility increased enormously, and a structural change from manufacturing to services was accelerated in rich countries. This resulted in negative consequences for labour share and income inequality. After having provided a theoretical discussion of the determinants of the compression of the wage share, making reference to the relevant literature, we submit our hypotheses to empirical scrutiny, performing a panel data analysis on 28 OECD Countries. The results of the estimations provide support to the theoretical argument.
Article
Purpose The purpose of this paper is twofold: first, to reassert the persistent association of the decline in collective bargaining with the increase in income inequality, the fall in the share of wages in national income and deterioration in macroeconomic performance in the UK; and second, to present case studies affirming concrete outcomes of organisational collective bargaining for workers, in terms of pay, job quality, working hours and work-life balance. Design/methodology/approach The paper is based upon two methodological approaches. First, econometric analyses using industry-level and firm-level data for advanced and emerging economies testing the relationship between declining union density, collective bargaining coverage and the fall in the share of wages in national income. Second, it reports on ten in-depth case studies of collective bargaining each based upon analysis of collective bargaining agreements plus in-depth interviews with the actors party to them: in total, 16 trade union officers, 16 members and 11 employer representatives. Findings There is robust evidence of the effects of different measures of bargaining power on the labour share including union density, welfare state retrenchment, minimum wages and female employment. The case studies appear to address a legacy of deregulated industrial relations. A number demonstrate the reinvigoration of collective bargaining at the organisational and sectoral level, addressing the two-tier workforce and contractual differentiation, alongside the consequences of government pay policies for equality. Research limitations/implications The case studies represent a purposive sample and therefore findings are not generalisable; researchers are encouraged to test the suggested propositions further. Practical implications The paper proposes that tackling income inequality requires a restructuring of the institutional framework in which bargaining takes place and a level playing field where the bargaining power of labour is more in balance with that of capital. Collective bargaining addresses a number of the issues raised by the Taylor Review of Modern Working Practices as essential for “good work”, yet is at odds with the review’s assumptions and remedies. The case studies reiterate the importance of the development of strong workplace representation and bargaining at workplace level, which advocates for non-members and provides a basis for union recruitment, organisation and wider employee engagement. Originality/value The paper indicates that there may be limits to employer commitment to deregulated employment relations. The emergence of new or reinvigorated collective agreements may represent a concession by employers that a “free”, individualised, deinstitutionalised, precarious approach to industrial relations, based on wage suppression and work intensification, is not in their interests in the long run.
Article
The U.S. labor share of income has been on a secular downward trajectory since the beginning of the new millennium. Using data that are disaggregated across both state and industry, we show the decline in the labor share is broad-based but the extent of the fall varies greatly. Exploiting a new data set on the task characteristics of occupations, the U.S. input-output tables, and the Current Population Survey, we find that in addition to changes in labor institutions, technological change and different forms of trade integration lowered the labor share. In particular, the fall was largest, on average, in industries that saw: a high initial intensity of “routinizable” occupations; steep declines in unionization; a high level of competition from imports; and a high intensity of foreign input usage. Quantitatively, we find that the bulk of the effect comes from changes in technology that are linked to the automation of routine tasks, followed by trade globalization.