ILO-The Case for Wage-led Policies

8/22/2019 ILO-The Case for Wage-led Policies

1/104

International Journal of Labour Research 2011

Vol. 3

Issue 2

owards a sustainablerecovery: Te case

for wage-led policies

INERNAIONAL LABOUR OFFICE, GENEVA


2/104

Photocomposed in Switzerland WEIPrinted in Switzerland SRO

Copyright International Labour Organization 2011First published 2011

Publications o the International Labour Office enjoy copyright under Protocol 2 o the Uni-versal Copyright Convention. Nevertheless, short excerpts rom them may be reproduced withoutauthorization, on condition that the source is indicated. For rights o reproduction or translation,

application should be made to ILO Publications (Rights and Permissions), International LabourOffice, CH-1211 Geneva 22, Switzerland, or by email: [email protected]. Te InternationalLabour Office welcomes such applications.

Libraries, institutions and other users registered with reproduction rights organizations may makecopies in accordance with the licences issued to them or this purpose. Visitww w.irro.orgto findthe reproduction rights organization in your country.

International Journal o Labour ResearchGeneva, International Labour Office, 2011

ISSN 2076-9806

labour relations / collective bargaining / wage determination / trade union role / collectiveagreement / labour flexibility / economic recession / Greece / employment / unemployment /

wages / monetary policy / economic recovery / developed countries / developing countries /income distribution / OECD countries / labour productivity / household income / consumerexpenditure / debt repayment / USA

13.06.3

Te designations employed in ILO publications, which are in conormity with United Nationspractice , and the presentation o material therein do not imply the expression o any opinionwhatsoever on the part o the International Labour Office concerning the legal status o anycountry, area or territory or o its authorities, or concerning the delimitation o its rontiers.

Te responsibility or opinions expressed in signed articles, studies and other contributions restssolely with their authors, and publication does not constitute an endorsement by the InternationalLabour Office o the opinions expressed in them.

Reerence to names o firms and commercial products and processes does not imply their endorse-ment by the International Labour Office, and any ailure to mention a particular firm, commercial

product or process is not a sign o disapproval.

ILO publications and electronic products can be obtained through major booksellers or ILOlocal offices in many countries, or direct rom ILO Publications, International Labour Office,CH-1211 Geneva 22, Switzerland. Catalogues or lists o new publications are available ree ocharge rom the above address, or by email:[email protected]

Visit our website:ww w.ilo.org/publns

ILO Cataloguing in Publication Data
mailto:[email protected]://www.ifrro.org/mailto:[email protected]://www.ilo.org/publnshttp://www.ilo.org/publnsmailto:[email protected]://www.ifrro.org/mailto:[email protected]


3/104

159

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

Contents

161ForewordDan Cunniah

163EditorialPierre Lalibert

167Wage-led growth: An introductionEngelbert Stockhammer

189Leveraging inequalityMichael Kumhof and Romain Rancire

197Te productivity and investment effects o wage-led growthServaas Storm and C.W.M. Naastepad

219Te economics o wage-led recovery:

Analysis and policy recommendationsThomas Palley

245Te impact o the crisis on labour relations

and collective agreements in GreeceYannis Kouzis


4/104


5/104

161

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

This issue o theInternational Journal o Labour Researchaddresses a cen-tral issue, i not the key issue or the labour movement, that o wages andwhat has happened to them over the past three decades.

It is clear that the combination o restrictive macroeconomic policies,trade liberalization and the financialization o corporate governance has dras-tically changed the landscape in which collective bargaining takes place. In

the North concession bargaining has become a amiliar concept, and in theSouth, the shif in the balance o power has meant that workers are not ableto capture the hard-earned ruits o economic growth.

Tis new context, by weakening labour market regulation both de jureand de acto has prooundly eroded trade unions ability to connect im-

provements in standard o living to productivity gains. Tis has resulted notonly in increased wage and income inequalities and higher incidence o low

pay, but also in an increasingly dysunctional macroeconomic picture. Aswages could not sustain aggregate demand as they once did, workers in sev-

eral countries relied more and more on credit to maintain their standard oliving, with the calamitous results that we all witnessed in . Short owage increases at home, growth strategies everywhere have become increas-ingly export-oriented.

Most o the contributions to this issue emanate rom papers presentedat an ACRAV workshop held in May under the title Wages, thecrisis, and economic recovery. Te workshop brought together academic andtrade union researchers, as well as ILO specialists, to take stock o wage de-

velopments and their consequences. But as the theme suggests, the workshop

was not just retrospective in its outlook, it also sought to identiy how wagescould play a role in creating a sustainable exit o the current situation.

One o the main findings to emerge rom the discussions is that wage-led growth economic strategies, ar rom undermining growth as is argued bymainstream economists, would on the contrary improve growth rates. Tis

ForewordDan CunniahDirector,Bureau or Workers ActivitiesInternational Labour Office


6/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

162

is an important argument as it directly contradicts the current competitive-ness policy orientation in much o the world an orientation based on per-manent wage moderation.

It is clear that such a recovery can only materialize i there is a global

rebalancing o wages and productivity. Tis will not only require that tradeunions intensiy their efforts at the bargaining table and in pushing or betterminimum wages, but that they fight to change the new global rules o thegame that are diametrically set against them. In this area, as or others, col-lective action is asine qua noncondition to achieving any success.

Finally a air warning to readers: this issue may not make or easyreading or most non-economists. But let me assure you that the effort is en-tirely worth it as the articles not only provide the analytical ground or re-sponding to the arguments o mainstream economists, but also put orward

an alternative based on sound evidence and credible analysis.


7/104

163

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

I a lot o attention has been deservedly given to the financial roots o thecurrent economic crisis, the role o wages on the other hand has yet to getthe consideration it merits both as a cause o the crisis as well as a solutionto the current economic predicament. o help fill this gap, this issue o the

International Journal o Labour Researchis wholly dedicated to this topic.Te stylized acts about wage and income in the past quarter o a cen-

tury are by now well documented: in a majority o countries around theworld, low pay has become more prevalent, leading to an increase in wage(and income) inequality. A key actor in this development has been the de-linking o wages rom productivity growth, a relationship which had been ahallmark o post-war era collective bargaining negotiations. At the macroeco-nomic level, these trends have maniested themselves in a secular decline inthe share o national income that goes to labour, the so-called wage share.Again, this is a widespread development that touches even the so-called win-ners o the international trade game, such as China or Germany.

Wage and income trends have been such that even the OECD and theIMF have over the past decade tried to come to grips with developments thatwent against conventional expectations. Not too surprisingly, the main cul-prits these organizations identified are related to technological developmentsand education differentials. In a nutshell, since the s, technological de-

velopments have increased the premium to education leading to greater di-erentials between the technology savvy and less educated workers. Since thecause is technological developments, something over which governments haveno control, the only policy prescription lef is to increase access to education

and hope or the best.Globalization is also seen as playing a role in the story by throwing

unskilled workers into competition with each other, thereby keeping wage

. See ILO: Global Wage Report(Geneva, ).

EditorialPierre LalibertEditor


8/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

164

increases or this group in check. However, since the OECD and the IMFtypically see globalization as a technology-driven phenomenon and a goodthing on the account o the efficiency it osters, the solution again is or gov-ernments to invest in education and training to upgrade workers skills and

ensure that the losers in international trade can get back on their eet.he contributors to this issue o the IJLRcontest not only the the-

oretical premises o mainstream labour market analysis, but offer alterna-tive explanations to the past wage trends. Using state-o-the-art econometricmodels, Stockhammer provides a different narrative or the decline in the

wage share which is linked first and oremost to financial and trade liberaliza-tion, as well as decline in union density. In his view, technology plays no stat-istically significant role in the story. Tis is to say that ar rom being an acto God, the decline in the wage share is to a great extent the result o policy

decisions, and thus amenable to urther policy developments.Storm and Naastepad underscore how the shif in the primary objective

o macroeconomic policy rom ull employment to inflation, nicely wrappedin the theory o the non-accelerating inlation rate o unemployment(NAIRU or short), was a central element in this development. However,by their account, i NAIRU-inspired policies were successul in introducingever more flexibility in the labour market and weakening labour unions, theylargely ailed to deliver on their stated objectives: improved economic and

labour market outcomes.For Storm and Naastepad, as well as Palley and Stockhammer, the ailureo the NAIRU model goes back to the theoretical premises o mainstream econ-omics which treats wages essentially as a cost, but not as a generator o demandand even less as a vector o technological change. In time, this one-sided obses-sion has led to dysunctional macroeconomic conditions in which private debtand exports had in some sense to make up or the growing earnings gap o

workers, which in turn sowed the seeds o the financial crisis o .Kumho and Rancire make an original contribution to the discussion

here as they present the first bona fide model that links income inequalityto financial crisis. Teir model, which allows or a struggle over wage andprofit (and different spending behaviours on the part o workers and capital-ists), shows how the outcome o this process has repercussions in the finan-cial sphere. Te research is all the more significant in that it emerges rom theresearch entrails o the IMF. One can only hope that it will spark renewedinterest within that institution in the issue o macroeconomic and financialeffects or wage and income inequality.

Storm, Naastepad, Stockhammer and Palley make parallel cases or pol-

icies avouring growth though wages. All our agree that the crisis has dem-onstrated that the neoliberal model is spent. With large-scale unemploymentin OECD countries, a more ragile financial sector and high level o house-hold debt, growth through private indebtedness is pretty much a oreclosedavenue. As a consequence, nations that depended on credit-led consumption


9/104

Editorial

165

in other countries to increase their exports also find themselves in a new pre-dicament. Growth through competitive austerity will not work: a new para-digm is needed to lead us to sustainable growth.

Stockhammer makes a distinction between pro-labour policies and

wage-led regimes. Te ormer reer to policies regarding their effects onwages (increase in minimum wages, strengthening trade union rights, im-proving social protection, macroeconomic policies that aim or ull-employ-ment, etc.), while the latter reer to the structure o the economy itsel.

An economic regime is a description o actual economic structures andinstitutions, including social security provisions, the financial system in placeand the degree o openness o the economy. While the economic regime isinfluenced by various orms o government policy, it should be clear that thenature o the economic regime is not a choice variable or economic policy

in any straightorward sense. It should not be understood as the outcome opolicy strategy.

Stockhammer (but also Storm and Naastepad) makes the point that wageincreases will have dierent economic outcomes depending on the kind oregime under which it takes place. Tus some countries are deemed wage-led and others profit-led, depending on this effect. While wage improvementstypically increase consumption and oten lead to increased investment and

productivity (notably by pushing up the utilization o productive capacity), thus

increasing aggregate demand, they also have a more negative effect i they reducethe profit rate and negatively influence net exports (exports minus imports). I acountry is wage-led, an increase in the labour share should translate into astergrowth while under a profit-led regime it would prove sel-deeating. Conversely,

pro-capital policies in a wage-led regime would be equally suboptimal.A review o the literature shows that most countries or regions typically

operate under wage-led regimes while a ew, particularly small export-oriented countries, come more closely to the description o profit-led. And,o course, the whole world as an economic space is wage-led, in large part

because it is a closed economy.As Stockhammer observes, contrary to neoliberal claims pro-capitalpolicies did not lead, over the past years, to increased investments (andeventually wages) and consequently did not set an economic virtuous circleinto motion. Ultimately, growth became dependent on finance-led consump-tion. In his view, the world now urgently needs to move to policies that willstrengthen wages and provide a more sustainable basis or development.

Storm and Naastepad make a parallel argument on the need or wage-ledgrowth policies, but put more emphasis on the productivity-enhancing effects

o wage changes. In their view, higher wages tend to oster higher product-ivity growth, in part because o the pressure to introduce labour-saving in-novation, but also because o improved social relations in the workplace.NAIRU-based models break down once you introduce these effects since,as they observe, more regulation has a bigger impact on labour productivity


10/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

166

growth than on real wage claims and, hence, is associated with lower struc-tural unemployment.

In his article, Palley brings all the pieces together to put orward anambitious plan or wage-led recovery. He shares the view that the current

policy orientation is a recipe or ailure as it will only urther depress a worldeconomy already short o aggregate demand. For this to happen, policy-makers need to get out o the current prisoners dilemma that has everyonegoing or austerity cum competitiveness policies and get on with policiesthat are optimal or all.

Re-linking wage and productivity increases through the acilitationo collective bargaining and the improvement o minimum wages is keyto a wage-led recovery. Palley proposes the establishment o a global min-imum wage (as a given percentage o each national median wage) to provide a

common and meaningul floor or the world economy.O course, pivotal to any such recovery plan are pro-employment fiscal

and monetary policies, as well as substantial reorm o the current financialand trade international architecture. Palley also makes a plea or renewedaction on the labour standards ront as they constitute a key dimension o anew globalization paradigm.

Tis issue ends with an important cautionary tale about Greece as theproverbial canary in the coal mine or the labour movement, particularly in

Europe. In his article, Kouzis shows how the fiscal crisis in Greece is beingused to unleash urther labour market deregulation and wage repressionwhen it should be patently clear that wage developments were not the causeo that countrys economic ailments. Among the key architects o this scan-dalous yet amiliar scenario, the European Commission stands out as beingeven more hawkish than the IMF

It is clear that the European labour movement is at an important cross-roads. he European Union project to which it lent credibility is ast indanger o becoming an albatross associated with austerity, economic stag-

nation and high rates o unemployment, particularly with young workers.Unless European unions are able to develop a coordinated response to thecrisis through better synchronization o wage bargaining strategies, the es-tablishment o some orm o minimum wage floor, and a pro-active politicalcampaign to reorm the institutions o the EU away rom their current aus-terity bias onto a genuine solidarity orientation, it is difficult to see how thenotion o social Europe will ever be more than an empty slogan.

When the president o the Socialist International himsel, along withother socialist governments, become the willing accomplices o liberal aus-

terity plans that may well sacrifice a generation o young workers, one cannothelp being overwhelmed by the proound vacuum o political leadership andthe urgent need to put orward an alternative plan o action.

As the contributions in this issue careully demonstrate, that alternativeexists.


11/104

167

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

Wage-led growth:An introduction *

Engelbert Stockhammer

Kingston University

* Te paper is part o the project New perspectives on wages and economicgrowth: Te potentials o wage-led growth. Te second section builds on joint

work with Marc Lavoie. An earlier version o this paper was presented at theworkshop Wages and Economic Recovery, held in May at the ILO. Teauthor is grateul to the participants and to Hubert Kohler and Marc Lavoie orcomments. Te usual disclaimers apply.


12/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

168

The past decades have witnessed alling wage shares and a polarization opersonal income distribution. Average wages and average labour com-pensation have not kept up with productivity growth. Functional incomedistribution has shited at the expense o labour. In many countries per-

sonal income distribution has also become more unequal. By many measuresincome inequality is worse than at any time in the twentieth century. At thesame time economic growth processes have become imbalanced. Financialcrises have become more requent; household debts have risen sharply; inter-national imbalances have increased, with some countries relying excessivelyon export growth. Tis paper argues that the polarization o income dis-tribution and the decline in the wage share play an important role in thegeneration o imbalanced and unequal growth, and that a pro-labour wage

policy will orm an important part o a policy package that generates a stable

growth regime. A wage-led growth strategy is thus advocated.Te advocacy o a wage-led growth strategy has a long history. It has

been articulated in reormist visions within the labour movement and wasdiscussed under the heading o underconsumption in nineteenth centuryeconomics. Te theory got a boost rom the theories o effective demand de-

veloped by Keynes and Kalecki. Te modern theoretical debates on wage-led demand were based on seminal papers by Rowthorn (), Dutt ()and Bhaduri and Marglin (). Te policy-oriented concept o a wage-led

growth strategy was prominently used by UNCAD ().Te second section o this paper will provide a policy-oriented rame-work or the analysis o the interaction between distribution and growth. Wewill distinguish between distributional policies and economic regimes. Pro-labour policies aim at increasing wages, whereas pro-capital distributional

policies aim at suppressing wage growth and increasing profit margins. Temacroeconomic regime o a country is determined by the structural eatureso its economy, such as its openness to international trade, its financial systemand the characteristics o its welare state. We will distinguish between wage-

led and profit-led economic regimes, or more precisely between wage-led andprofit-led demand and supply regimes. In a wage-led regime, an increase inthe wage share has positive effects that mean higher economic activity (in theshort run) and aster accumulation o capital (in the long run), both throughdemand-side effects, or aster productivity growth on the supply side. By con-trast, a profit-led economic regime would occur whenever a decrease in theshare o wages or an increase in the profit margins o firms generate positiveeffects on the economy.

Te third section investigates the causes o changes in income distribu-

tion, in particular the long-run reduction in the share o wages. Te ourthsection provides more details as to why an economy would exhibit a wage-ledeconomic regime, looking both at supply-side effects, that is the relationshipbetween the share o wages and labour productivity growth, and at demand-side effects. Tis section also has a summary o some recent empirical research,


13/104

Wage-led growth:An introduction

169

providing the approximate size o some key effects on the demand side. Tefifh section will classiy the actual experience o key economies within thisramework. In the era o neoliberalism, growth processes have become imbal-anced, either relying on growing debt ratios or on persistent export surpluses.

wo growth processes have emerged:finance-led growth(also called debt-ledgrowth), where growth was uelled by increasing household debt made pos-sible by asset and property price bubbles and financial engineering (examplesare Ireland, the United Kingdom, and the United States) and export-led

growth, where the main engines o growth have been net exports (examples areChina, Germany, and Japan). Both o these neoliberal growth processes havecome with wage suppression. Finally, the sixth section highlights a wage-ledgrowth strategy as a possible alternative. It combines pro-labour distributional

policies with structural policies that are avourable to wage-led growth. It has

the potential or an equitable and (economically) sustainable growth process.

Distribution and growth: A conceptual framework

Te relation between distribution and growth was at the centre o macroeco-nomic analysis in classical economics, but with the dominance o neoclas-sical economics in the twentieth century, issues o distribution have occupied

a secondary place, since income distribution was assumed to be regulatedby marginal productivity relations within a perect competition model. Inthe ollowing we offer a policy-oriented ramework to analyse the relationbetween distribution and growth. We will contrast pro-labour and pro-capitaldistributional policies and wage-led and profit-led demand and supply regimes.

Income distribution is the outcome o complex social and economicprocesses, but governments influence it by means o social policy and labourmarket policy. We define pro-capital distributional policies as policies thatlead to a decline in the wage share, and pro-labour distributional policies

as policies that result in an increase in the wage share. Pro-capital distribu-tional policies usually proclaim to promote labour market flexibility orwage flexibility, rather than increasing capital income. Tey include meas-ures that weaken collective bargaining institutions, weaken labour unions,lower minimum wages, and weaken employment protection legislation. Pro-labour policies are ofen reerred to as strengthening the welare state andlabour market institutions and include strengthening collective bargaining(e.g. by extending the reach o bargaining agreements to non-unionizedfirms), strengthening labour unions, increasing unemployment benefits, and

reducing wage and salary income inequalities.

. Here, and in the ollowing, we assume that (effective) labour demand is inelastic (orupward) sloping (or empirical evidence see e.g. Rowthorn ()). Tus an increase in real

wages will correspond to an increase in the wage share.


14/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

170

O course, there are also other actors influencing income distribution,such as technological changes, globalization, and financialization. Tese ac-

tors have recently played an important role, but we will not elaborate on themhere because this section ocuses on the interaction o distributional policiesand economic regimes. We will revisit the determinants o income distribu-tion in the ollowing section.

Next we consider the economic structure. An economic regime is a de-scription o actual economic structures and institutions, including social se-curity provisions, the financial system in place and the degree o openness o theeconomy. While the economic regime is influenced by various orms o govern-

ment policy, it should be clear that the nature o the economic regime is not achoice variable or economic policy in any straightorward sense. It should notbe understood as the outcome o policy strategy. We will distinguish between

wage-led and profit-led economic regimes. Furthermore, ollowing conven-tional practice we will distinguish between demand-side (both in the shortrun and in the long run) and supply-side (long-run) considerations. Te key de-mand-side variable is the level o aggregate demand, emphasized by Keynesianeconomists. he key variable or the supply side is productivity growth.

For our purpose, the question is, first, how aggregate demand reacts to

a change in income distribution. Tese effects will be quite complex and arediscussed in more depth in the ourth section. Here we will ocus on extremecases in order to illustrate our ramework. Demand may be wage-led or profit-led. A wage-led demand regimemeans that an increase in the wage share leadsto an increase in aggregate demand. Te wage-led scenario may arise whenhigher wages lead to higher consumption expenditures (higher consump-tion sales may then also induce higher investment expenditures). Conversely,aprofit-led demand regimemeans that an increase in the wage share leads toa decline in aggregate demand. Demand may be profit-led i investment is

highly sensitive to a reduction in profit margins. High profitability (at a givenrate o capacity utilization) may motivate firms to expand their productivecapacity and increase investment.

O course there are many actors other than income distribution that de-termine aggregate demand: monetary policy, fiscal policy, various shocks such

Table 1. Pro-labour and pro-capital distributional policies

Distributional policies Other factors

Pro-capital Pro-labour

Policies Labour market flexibility

Abolish minimum wages

Weaken collective

bargaining

Welfare state

Increase minimum wages

Strengthen collective

bargaining

Changes in

technology

Globalization

Financialization

Results Weak wage growth

Wage share

Increased wage

dispersion

Rising real wages

Stable (or ) wage share

Decreased wage

dispersion


15/104


171

as spikes in oil price, the bursting o a stock market bubble, changes in realexchange rates, changes in the growth rate o trade partners, etc. Indeed, or

most year-to-year changes, income distribution will only be a minor influenceon the determination o aggregate demand, with other developments playinga more prominent role. However, i there are long-lasting, deep changes inincome distribution as have occurred in the last quarter century, they willend up having a substantial role.

Finally, aggregate supply may also be wage-led or proit-led. he keysummary variable or the supply side is labour productivity. Productivity

will be profit-led, i an increase in wages discourages productivity-enhancingcapital investment and, as a result, the growth o labour productivity slows

down (most orms o technological progress require capital investment; thisis called embodied technological progress). Increases in wage growth mayhave a positive eect on productivity growth, either i irms react by in-creasing productivity-enhancing investments in order to maintain competi-tiveness, or i workers contribution to the production process improves. Tismay be the case either because o improved workers motivation or, in devel-oping countries, i their health and nutritional situation improves. Tis caseis ofen called the efficiency wage hypothesis, but we may also call it the

Webb effect, since a positive causal relationship going rom higher real wages

to higher productivity was already proposed by Sidney Webb (), one othe ounders o the London School o Economics.

A wage-led demand growth regimeis a stronger and more long-term con-cept than wage-led demand. While the latter simply implies that an increasein the wage share will lead to an increase in aggregate demand or in the rate

Table 2. Economic structure: wage-led and profit-led demand and supply regimes

Demand regime Supply regime

Economic

structure

Profit-led Investment very sensitive to

profit margins

A lower wage share leads to

higher investment

Wage restraint leads to

productivity-enhancing

investment

A higher wage share leads to

lower GDP and slower capital

accumulation

Higher real wage growth leads

to slower productivity growth

Wage-led The propensity to consume out

of wage income is higher than

that out of profit income

Wage growth has strong

positive effects on labour effort

and productivity-enhancing

investments

A higher wage share leads to

higher GDP and faster capital

accumulation

Real wage growth leads to

faster productivity growth

Other factors Other sources of demand:Government fiscal and monetary policies

Financial factors: financial asset and real estate price bubbles

Exchange rate evolution and changes in world demand

Changes in world commodity prices


16/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

172

o capacity utilization, the ormer additionally requires an increase in in-vestment expenditures and productivity growth. Over the long run it im-plies an increase in the rate o accumulation o the capital stock. In contrast,when an increase in the wage share implies a decrease in the rate o growth

o the capital stock and o productivity growth, we then speak o aprofit-leddemand growth regime.

able puts the analyses o distributional policies and o economic struc-tures together. For simplicity we do not distinguish between demand and

productivity regimes, but only discuss the economic regime, i.e., we assumethat demand and supply react in a similar direction to distributional changes.Tis allows us to gain insight in the likely growth dynamics o the differentregimes and strategies. Between the two sets o distributional policies and thetwo economic structures, our different combinations are possible. Tese do

have quite different properties. I pro-capital distributional policies are pur-sued in a profit-led economy, this will result in a profit-led growth process.Inversely, i pro-labour policies are pursued in a wage-led economy, this willresult in a wage-led growth process. Tese are the two cells in the main di-agonal in table . In both cases distributional policies and economic struc-tures are consistent. However, i pro-capital policies are pursued in a wage-ledeconomy or i pro-labour policies are pursued in a profit-led economy, this willresult in stagnation, or more likely in practice, will result in unstable growth

patterns as growth will have to rely on external stimulation.able is useul in categorizing different political ideologies associatedwith the our different combinations. ake the first cell (pro-capital policiesin a profit-led economy). Tis scenario corresponds to liberal ideology and

what is ofen called the trickle-down effect: higher profits are said to lead toimproved macroeconomic perormance. Workers will eventually benefit rom

wage cuts as higher profit margins will lead to investment and growth and re-wards will eventually trickle down to workers as well, in the orm o higheremployment rates and higher purchasing power. Tis scenario could be called

neoliberalism in theory.he cell that combines pro-labour policies with a wage-led economysummarizes what many economists (e.g. Marglin and Schor, ) regardas a key characteristic o the post-war era: the expansion o the welare state(in advanced economies) which led to a golden age o growth. Te next cell(pro-labour policies in a profit-led economy) could be called doomed socialreorms. It is the scenario that neoliberals claim would occur i progressive

Table 3. Viability of growth regimes

Distributional policies


Economic

structure

Profit-led Profit-led growth process Stagnation or unstable growth

Wage-led Stagnation or unstable growth Wage-led growth process


17/104


173

social reorms were implemented. Margaret Tatchers amous dictum thereis no alternative (INA) makes sense in this cell. Some Marxists use a similarscenario to illustrate the utility o attempts to establish a more humane

economy within the capitalist mode o production. Attempts to raise workersconsumption or the wage share inevitably lead to a slowdown o the economy.

Finally, there is the combination o pro-capital policies in a wage-ledeconomy. We will argue that this describes actually existing neoliberalism,

where two decades o pro-capital distribution have resulted in a mediocreeconomic perormance with a heavy reliance on a speculative financial sectoror on external demand to achieve growth (see the fifh section below). Teollowing sections will summarize some o the available evidence to evaluate

which scenario describes actual economies.

Decline in the wage share: What are the causes?

In the last quarter o a century dramatic changes in income distribution havetaken place. Tis reers to the personal distribution o income as well as tothe unctional distribution o income. Wage shares have allen in virtually allOECD countries, with decreases typically being more pronounced in conti-

nental European countries (and Japan) than in the Anglo-Saxon countries.In the euro area the (adjusted) wage share has allen rom . per cent in, to . per cent in (figure ). Personal income distribution hasbecome more unequal in almost all OECD countries (OECD, ), withthe very top income groups increasing their income shares substantially inthe Anglo-Saxon countries, in particular in the United States (Piketty andSaez, ; Atkinson, Piketty and Saez, ). In a multi-country study,Daudey and Garca-Pealosa () show that there is a positive correlation

. Although some researchers would argue instead that reliance on ree market mechanismsand more flexible labour markets have generated large increases in world real income overthe last three decades (Balcerowiz and Fisher, ). But these authors orget to compare thelast decades to the evolution o the s and s. For rich discussions o neoliberalism,see Harvey () and Glyn ().

Table 4. Actual growth strategies in the economic structure/distributional

policies framework

Distributional policies


Economic

structure

Profit-led Neoliberalism in theory: supply-side

policies will generate aggregate demand

(trickle-down theory)

Doomed social

reforms TINA

Wage-led Actually existing Neoliberalism unstable

and has to rely on exogenous growth drivers

(credit-led growth)

Post-war social

Keynesianism


18/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

174

between changes in personal and unctional income distribution. Overall,median real wage growth has clearly lagged behind productivity growth sincearound . Tis constitutes a major historical change as wage shares hadbeen stable or increasing in the post-war era.

Tis secular decline has led to a renewed interest in the determinantso the distribution o income in recent years, with major economic researchinstitutions such as the OECD and the IMF publishing prominent studies.Te OECD () documents changes in personal income distribution. TeIMF (a) and the European Commission () investigate changesin unctional income distribution, and the OECD () analyses the

wage elasticity o the labour demand unction. Te IMF (a) and theEC () make a strong case that technological change has been the maincause o changes in unctional income distribution, that globalization (o

trade and production) has also played an important role and, finally, thatchanges in labour market institutions have played a minor role. echnologicalchange is empirically measured as IC (Inormation and Communicationechnology) investment, or IC services. Te general thrust o the argumentis in line with the neoclassical theory o income distribution, which regardsdistribution as essentially technologically determined.

Globalization also eatures prominently in the debate. he standardtrade-theory argument is built on the Stolper and Samuelson () theorem,

which states that theabundantactor will gain rom trade liberalization. Fornorthern countries, supposedly, this is capital whereas labour is abundantin developing countries that have recently entered the global economy, suchas China and India. Globalization is thus supposed to benefit capital in theNorth, and labour in the South.

While the Stolper-Samuelson argument describes a competitive long-run equilibrium, the political economy o trade approach highlights dis-tributional eects o globalization in a bargaining setting. For example,Rodrik () argues that trade liberalization (even among similar countries)

will affect distribution and will benefit the more mobileactor, which willtypically be capital. Unlike the Stolper-Samuelson approach, Rodriks argu-ment is set in a bargaining ramework. Te change in distribution takes placebecause o a redistribution o rents, not because o the equalization o actor

. he Stolper-Samuelson theorem assumes that irms have no market power and thatneither capital nor labour are mobile; its effects take place through trade in competitive equi-librium. However, the recent period o globalization has been marked by an increase in cap-ital mobility. I capital can travel across borders, the implications o the theorem weakensubstantially (EC, , p. ). Moreover, classical international trade theory is unable to

explain the actual pattern o trade, which takes place mostly among developed countries.According to standard trade theory it is not obvious why North-North trade should affectincome distribution (assuming that relative actor prices are similar). Second, labour is nota homogenous input. While unskilled labour (in the north) may lose rom globalization,skilled labour may indeed gain. I so, it is a priori not clear how the total wage share in thenorth should be affected.


19/104


175

costs. Moreover, in the Stolper-Samuelson theorem one would expect distri-bution to change aerproduction has been relocated. In contrast, Epsteinand Burke () argue that due to threat effects, redistribution can take

place without changes in production locations.While there are differences in the theoretical arguments, the empirical

assessment is rather clear. All studies find substantial effects o globalization

on unctional income distribution. For example, the IMF concludes thatglobalization is one o several actors that have acted to reduce the share oincome accruing to labour in advanced economies (IMF, a, p. ).

A third set o actors that influence income distribution is financial de-regulation (or, more broadly speaking, financialization). Financial deregula-tion has had two important effects on the bargaining position o labour. First,firms have gained more options or investing: they can invest in financial assetsas well as in real assets and they can invest at home as well as abroad. Teyhave gained mobility in terms o the geographical location as well as in term o

the content o investment. Second, it has empowered shareholders relative toworkers. Te development o a market or corporate control has aligned man-agements interest to that o shareholders (Lazonick and OSullivan, ;Stockhammer, ). Rossmann () illustrates this with reerence to pri-

vate equity unds, which buy firms by way o debt that is transerred to thefirm. Te surplus is siphoned to the private equity und through dividend

payments or ees. Te restructured firms then are heavily burdened with ser-vicing their debt and have little alternative to pursuing an aggressive cost-cut-ting strategy. For countries where data is available, the increase in dividend

payout is well documented (Dumnil and Lvy, ). Power, Epstein andAbrena () document the increasing income share o rentiers.

. Financialization reers to the increased influence o financial institutions and financialmotives on non-financial activities.

Figure 1. Adjusted wage shares in the euro area, the United States and Japan, 19602007 (%)

Source: AMECO.

60

65

70

85

80

75

1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005 2007

Japan

Euro area(12 countries)

United States


20/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

176

So ar, ew econometric studies on changes in unctional income dis-tribution have included inancialization variables. he ILO argues thatfinancial globalization has led to a depression o the share o wages in GDP(ILO, , p. ), but does not provide evidence. Jayadev () analyses

the effect o financial openness and trade openness on the wage share in aneconometric analysis covering up to countries or the period to .Te openness variables are legal measures on openness. Capital account open-ness and trade openness are ound to have negative effects on the wage share.Remarkably, the IMF (b), in a study on personal income distribution

within countries has included oreign direct investment (FDI) stocks.In a detailed study attempting to replicate and extend IMF (a) and

EC () studies, Stockhammer () finds that the results or techno-logical change are not robust, whereas the effects o globalization are con-

firmed. He then extends the estimation specifications to include a measure orfinancial globalization, and allows or different effects o trade union densityin countries where trade union membership is a precondition or receivingunemployment beneits. He inds that inancial globalization has strongeffects and the organisational strength o labour unions has a robust effect.

Economic effects of a declining wage share

While the previous section discussed the causes o the decline in the wageshare, this section turns to its effects. It is standard in economic theory todistinguish between the demand-side and supply-side effects, where demandeffects reer to changes in expenditures or a given productive capacity andtechnology, while supply-side effects involve changes in machinery and tech-nology. he key summary variable or the supply side is (the growth o)labour productivity. We will ollow the same distinction here, being under-stood, as was pointed out in the second section, that demand effects can spill

over to the growth rate o capital accumulation.

Demand effects

What are the eects o change in the wage share on aggregate demand?Aggregate demand consists o private consumption expenditures, investmentexpenditures, net exports and government expenditures. In the ollowing

. Mainstream economics regards demand effects as purely short-run effects as it considersthe economy to be strongly anchored in a supply-determined equilibrium to which the econ-omy will return. Keynes, who pioneered the analysis o demand ormation, was rather scep-tical o long-run analysis. Post-Keynesian economics, built on the works o Keynes, Kaleckiand Steindl, highlights that aggregate demand plays a crucial role even in the long run.


21/104


177

section we ocus on the reaction o the private sector and treat governmentexpenditures as an exogenous policy variable.

A change in income distribution will have several effects on the compo-nents o demand that pull in different directions. First, consumption expendi-

tures are likely to be a positive unction o the wage share. Higher wages willtypically lead to higher consumption expenditures because wage earners nor-mally have a higher propensity to consume than recipients o capital income.Tis is because workers are typically poorer than capitalists (or other recipientso capital income). Furthermore, a large proportion o gross profits are savedby firms in the orm o retained earnings. Te size o this income distributioneffect will depend on the difference in income between capital and labour, onthe social security system, which influences savings rates, but also on other ea-tures such as house prices and capital gains on the stock market. Second, in-

vestment expenditures are likely to react negatively to an increase in the wageshare, i.e., to a decrease in the profit share (or a given level o national income).From an intuitive point o view, a reduction in the profit share or a given levelo national income implies that the profit margins o firms have gone down.Since expected uture profits ought to be an important stimulant or invest-ment, a reduction in profit margins, i.e., a reduction in the profit rate assessedat normal rates o utilization o capacity ought to have a negative effect on in-

vestment. Te precise effect will depend on the structure and liquidity o the

financial system and on what Keynes called the psychology o the investor, e.g.afer a financial crisis firms may be reluctant to invest because o increased un-certainty. Tird, net exports are likely to react negatively to increases in the

wage share because, or a given exchange rate, the increase in the wage sharewill decrease proit margins and/or make exports less competitive abroad.Te size o this effect will depend on the degree o openness o the economyand the types o products that the economy is importing and exporting.

Te effects on the three aggregates thus pull in different directions. Anincrease in the wage share is likely to increase consumption, but decrease in-

vestment and net exports. Te net effect is not clear a priori, but will dependon the relative size o these effects. I the consumption effect is stronger thanthe investment and net export effects then the overall effect is positive andthe economy is in a wage-led demand regime. Conversely, i investment andnet exports react more strongly, the overall effect o an increase in the wageshare on demand is negative and the demand regime is called profit led. Tisdistinction is based on the theoretical work o Bhaduri and Marglin ()and Blecker ().

Note that the model outlined above includes net exports. One countrys

exports are some other countrys imports. Tis raises the possibility o a al-lacy o composition: while each individual country can increase its demandby exporting more, not all countries can do so at the same time. Te worldeconomy as a whole is a closed economy. It is thus interesting to look at thedomestic effect and the total effects (i.e., including net exports) separately.


22/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

178

Te domestic effects only include the effects on consumption and investmentand should be interpreted as a scenario when the change in the wage share a-ects all trading partners simultaneously. It can be thought o as a change inthe world wage share.

Regarding the consumption behaviour, the saving differential betweenrich and poor is well established empirically. As an illustration, table re-

ports the savings rates or different income groups or Germany. In , thebottom quarter o the income distribution had a savings rate o . per cent,

whereas the richest quarter had a savings rate o . per cent. Savings ratesclearly increase with income level. Germany experienced a dramatic increasein inequality in the last decades. Tis also affects different saving propen-sities. In , the lowest quartile had a savings rate o . per cent whereasthe richest quartile had a saving propensity o . per cent.

Tese models have recently inspired a rich empirical literature trying toidentiy demand regimes by econometric means. able gives an overviewo the empirical results. Tese studies differ by the countries and time periodcovered, as well as by the method employed (or more extensive discussionssee Hein and Vogel, ; Stockhammer and Stehrer, ) and are thus di-ficult to compare. Overall, the majority o studies find that domestic demandregimes tend to be wage-led, whereas international trade turns many econ-omies into a profit-led regime.

o illustrate the orders o magnitude involved, table summarizes the re-sults or a large, relatively closed economy (the euro area) and or a small openeconomy (Austria), based on Stockhammer, Onaran and Ederer, , andStockhammer and Ederer, , respectively. A . percentage point increasein the wage share would lead to an increase in consumption by . (per-centage points o GDP) in the euro area and . in Austria. Investment woulddecline by . and . per cent respectively. Domestic demand is wage-led inboth cases (by . in the EU and . in Austria). Te net export effectis . in the EU, but . in Austria. Te total demand regime is thus

wage-led in the EU (a . percentage point increase in the wage share leads toa . percentage point increase in demand), but profit-led in Austria (.).

. While I consider these values plausible, other researchers disagree . Naastepad andStorm (/) tend to find much higher investment effects and much lower net exporteffects. Tese results are based on single-equation estimators. Systems estimators tend to findstronger profit effects (Barbosa-Filho and aylor, ; Flaschel and Proano, ). My ex-

perience is that the consumption effect is rather reliable, though ofen rather small in Anglo-Saxon countries. Investment effects are usually very sensitive to the exact specification o theestimation equation. Tis is probably because profits and demand are highly correlated and

investment is a highly pro-cyclical variable itsel. Te net export effect depends on assump-tions about the exchange rate. Several early studies did not allow or globalization to affectthe wage sensitivity o exports. wo concluding comments on the literature need to be made:first, the simultaneity issue between distribution and demand lurks unresolved in the back-ground. Second, the set o control variables controlling or other actors is rather limited inmost estimations.


23/104


179

Tese results have important policy implications or regional economicintegration. ake the euro area as an example. As elsewhere, wage shares haveallen drastically in the euro area. Tis has been encouraged by the European

Commission, which has advocated a strategy o improving competitivenessor a long time (EC, , and ). Indeed, many European coun-tries have implemented wage pacts that combine wage restraint with other

policy measures (Schulten, ). he results in table illustrate an im-portant economic difference between the euro area and its Member States.

Table 5. Savings rates by income group, Germany, 1995-2007

1995 2001 2007

1st quartile 7.3 5.4 4.1

2nd quartile 9.5 9.3 8.0

3rd quartile 11.3 10.1 9.04th quartile 13.8 13.1 15.8

Source: Stein (2011) based on SOEP (Socio-Economic Panel) data.

Table 6. Econometric studies on wage-led and profit-led demand regimes

Domestic demand Total demand

Wage-led Profit-led Wage-led Profit-led

Euro area SOE09 SOE09

Germany BB95, NS07, HV08,

SHG11, SS11

NS07, HV08, SHG11 BB95

France BB95, NS07, ES07,

HV08, SS11

(SO04), NS07, HV08 BB95, ES07

Netherlands NS07, SS11 HV08 NS07 HV08

Austria SE08, HV08, SS11 SE08, HV08

United Kingdom BB95, NS07, HV08 SS11 BB95, NS07, HV08

Japan BB95 NS07 BB95, NS07

United States BB95, HV08,

OSG12, (SS11)

NS07 BB95, HV08, OSG12 (SO04), NS07,

BFT06

Note: References in brackets denote statistically insignificant results.

BB95: Bowles and Boyer (1995); BFT06: Barbosa-Filho and Taylor (2006); ES07: Ederer and Stockhammer(2007); HV08: Hein and Vogel (2008); NS07: Naastepad and Storm (2006/7); OSG12: Onaran, Stockhammer

and Grafl (forthcoming); SO04: Stockhammer and Onaran (2004); SE08: Stockhammer and Ederer (2008);SHG11: Stockhammer, Hein and Grafl (2011); SOE09: Stockhammer, Onaran and Ederer (2009); SS11:

Stockhammer and Stehrer (2011).

Table 7. Effects on private excess demand (in percentage of GDP)

EU 12 (openness < 15 percent)

Austria (openness > 50 percent)

Consumption 0.37 0.36

Investment 0.07 0.15

Domestic sector 0.30 0.21

Net exports 0.09 0.39

Total effect 0.21 0.18

Source: EU12: Stockhammer, Onaran and Ederer (2009); Austria: Stockhammer and Ederer (2008).


24/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

180

While many Member States are small open economies, in which a wage re-straint may boost demand via exports, the euro area as a whole is a large, rel-atively closed economy. Most o the trade o the Member States takes place

within the euro area. A wage cut in the euro area will increase net exports,

but domestic demand will shrink by more. Wage policy is thus in a prisonersdilemma-type situation. For individual Member States wage restraint may bean attractive strategy, but i everyone pursues it, it will have negative effects.European wage coordination would, at least in principle, make it easier toovercome the prisoners dilemma and internalize the externalities o wageagreements (Stockhammer, ; Hein and ruger, ). However, the di-erences in wage bargaining systems make this difficult in practice.

Supply-side effects

On the supply side, the key question is how changes in the wage share or inreal wages affect productivity growth (or, more broadly speaking, techno-logical progress). Mainstream economists typically argue that competitivemarkets are most conducive to growth and, in the next step, argue or labourmarket (and product market) deregulation. Critical economists highlight theact that labour market institutions cannot only have positive social effects as

they help overcome market ailures, but they also may have positive effects oneconomic growth because good labour relations will improve the propensityo workers to contribute to the production process.

Recently, this has inspired several empirical studies. Storm andNaastepad () investigated labour market institutions in OECDcountries and ound that relatively regulated and coordinated (rigid) in-stitutions led to higher productivity growth. Hein and arassow () ana-lysed the link between income distribution and productivity growth or sixOECD economies by means o time series analysis and reported that higher

proit shares have a negative eect on productivity growth. Vergeer andKleinknecht () perormed a panel analysis or OECD countries rom to and ound that higher wage growth leads to higher product-ivity growth. Tey interpret this as implying that stronger labour marketinstitutions lead to aster long-run growth. Tese studies ace challengesin identiying the direction o causality and the distinction between short-run and long-run effects; and more research is certainly needed. However,it seems air to conclude that the available evidence does not suggest thatreal wage growth has any negative long-run effect on labour productivity

growth.Wages have a dual unction in capitalist economies. Tey are a cost o

production as well as a source o demand. An increase in the wage share hasseveral effects on demand and whether actual demand regimes are wage-ledor profit-led is subject to an ongoing academic debate. Our interpretation o


25/104


181

the available evidence is that domestic demand regimes are likely to be wageled in most economies. In open economies, the net export effects may over-

power the domestic effects and total demand in many individual countriesmay well be profit-led. However, countries trade among each other. Larger

geographical (or economic) areas are thereore likely to be wage-led. heworld economy as a whole is probably in a wage-led demand regime. Tereis comparatively less research on the supply-side effects o an increase in the

wage share. However, there are several studies that find positive effects owage increases on productivity growth, suggesting that the long-term effectso wage expansion are unlikely to be harmul.

Classifying recent growth regimes and strategies:

Credit-led growth, export-led growth or wage-led growth

Neoliberalism came with the promise that deregulation o goods markets,labour markets and financial markets would lead to higher growth and in-creased welare. Higher inequality was to be accepted because it yields eco-nomic benefits. In our terminology, neoliberalism posited a strongly profit-ledeconomic regime. But neoliberalism has ailed to deliver on its promise.Growth rates in the allegedly over-regulated post-war era were higher than in

the neoliberal phase. Deregulation did indeed generate increased inequality,but without much o the benefits that were supposed to come with it.But i the world economy is indeed wage-led, how did neoliberal econ-

omies grow at all? Neoliberalism, in practice, has operated in the south-eastcell o tables and , pursuing a strategy based on pro-capital distribu-tional policies, but within an essentially wage-led economic structure. Sucha strategy will lead either to stagnation or it has to rely on external actorsor stimulating growth. Indeed, the latter is what has characterized the per-ormance o what we might call actually existing neoliberalism. Instead

o generating a robust growth path based on the proit-investment link,growth has relied on either financial bubbles/rising indebtedness (in short,finance-led growth) or on a mercantilist strategy based on export surpluses(Stockhammer, ; Horn and van reeck, ). Boom-bust cycles drivenby stock markets, property markets or capital flows have been a key eature oactually existing neo-liberalism: the Latin American crises o the s ando the mid-s (the peso crisis), the EMS (European Monetary System)crisis ( to ), the South-East Asia crisis ( to ), the dot.combubble burst ( to ), and the Great Recession o to .

o understand this pattern one has to appreciate the central role o fi-nancial deregulation or the neoliberal growth model. Financial deregulationhas allowed financial innovation and has given rise to speculative boom-bustcycles and, over long periods, to increasing debt levels or financial institutionsand households. Booms on stock markets and property markets ofen attract


26/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

182

capital inflows that uel the bubbles urther (Reinhart and Reinhart, ).But the liberalization o capital flows also means that some countries will haveto have current account surpluses and net capital outflows. International fi-nancial deregulation has thereby given rise to two symbiotic growth models: a

credit-led growth model (with capital inflows) and an export-led model (withcapital outlows). While growth has been driven by consumption growthuelled by rising household debt in the Anglo-Saxon countries, and especiallyin the leading country the United States, other countries have subdued do-mestic demand, including that arising rom the government sector, and haveheavily relied on net exports as the key growth engine.

While admittedly not all countries neatly fit this dichotomy o credit-led and export-led growth models, it is useul as it captures an important parto the dynamics behind the growing international imbalances and highlights

that both models compensate or a lack o domestic demand. Both growthmodels have occurred in the centre as well as in the periphery. In particularin Europe the central countries (Germany and its smaller cousins) have ea-tures o export-led growth, whereas the peripheral countries within the eurozone have had credit-led growth. able gives a stylized classification o im-

portant countries.wo statistics will help substantiate the useulness o the distinction

in credit-led and export-led economies. First, table gives the increase in

household debt (as percentage o GDP) in major European economies andthe United States (comparable data or Japan and China were not readilyavailable). While household debt declined in Germany by . percentage

points o GDP rom to , it grew by a modest . percentage pointsin Austria, and by well above percentage points in the credit-led group. In

Table 8. Growth models of actually existing neo-liberalism

Credit-led Export-led

Centre United States, United Kingdom Germany, Austria, Japan

Periphery Greece, Ireland, Portugal, Spain China

Table 9. Increase in household debt 2000 to 2008 (as a percentage of GDP)

Country Percentage Country Percentage

Germany 11.34 United States 26.00

Netherlands 32.83 United Kingdom 28.13

Austria 7.91 Ireland 62.72

Greece 35.46

Spain 33.84

Italy 18.32

Portugal 27.38

Source: Eurostat: Financial Flows and Stocks by Sector; USA: Flows of Funds;

Ireland starts in 2001 instead of 2000.


27/104


183

the United States and the United Kingdom, it increased by and percent respectively. In Greece, Portugal and Spain, household debt shot up by, and per cent respectively. In Ireland, it even grew by a staggering per cent.

It turns out that those countries with rising household debt, with ewexceptions, have also been the countries that ran current account deficits,

whereas those with little change in household indebtedness have been thecountries with current account surpluses. In , beore the financial crisis,

Germany and Austria had current account surpluses o . per cent and. per cent (o GDP) respectively, while Japan and China had current ac-count surpluses o . per cent and . per cent. On the other hand, theUnited States and the United Kingdom had deicits o . per cent and. per cent, and Greece, Ireland, Portugal, and Spain had deficits o . percent, ., ., and . per cent respectively.

Actually existing neoliberalism has not led to a growth process via in-vestment. Rather it has relied on other actors or growth. Rising householddebt has temporarily made up or wage growth (Barba and Pivetti )

in the credit-led growth model; increasing trade surpluses have been thegrowth engine o a second group o countries, that have ollowed an export-led growth model. Both o these growth models are not sustainable. Financialbubbles eventually burst and debts have to be serviced and possibly repaid(or otherwise, bankruptcy occurs), while export-led growth relies on othercountries to import, and leads to the impoverishment o the importing coun-tries and growing international imbalances.

. In a sense, this is not unexpected, since by identity, as pointed out in particular by the

late Wynne Godley, domestic household net borrowing + corporate net borrowing + publicborrowing = current account deficit.. With the exception o Ireland, current account positions and net export positions aresimilar. Ireland, in past decades, has had current account deficits, but net export surpluses.Tis is because o the large amount o repatriated profits, thus leading to a large discrepancybetween GDP and GNP.

Table 10. International imbalances: current account

as a percentage of GDP (2007)

Country Percentage Country Percentage

Germany 7.9 United Kingdom 2.7

Austria 3.6 United States 5.2Netherlands 8.7

Greece 14.5

Japan 4.8 Ireland 5.3

China 5.2 Spain 10.0

Portugal 9.4

Italy 2.4

Source: OECD.


28/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

184

Wage-led growth: a viable economic strategy

But there is an alternative to neoliberalism. I, as we have argued, the worldeconomy (and, indeed, large countries and/or economic blocs) are indeed

wage-led, then a wage-led growth strateg y is a viable option. A wage-ledgrowth strategy would have to combine pro-labour distributional social andlabour market policies with a regulation o the financial sector.

Distributional policies that increase the wage share and reduce wage dis-persion include increasing or establishing minimum wages, strengtheningsocial security systems, improving union legislation and increasing the reacho collective bargaining agreements. All o these are against orthodox eco-nomic wisdom and, under the perceived pressure to reduce budget deficits,economic policy is recently moving in the opposite direction. However, in

times o crisis and a lack o effective demand, what economies need is morestate involvement, not less. A successul policy package to economic recovery

will also have sustained wage growth as one o its core building blocks. Onlywhen wages grow with productivity growth will consumption expendituresgrow without rising debt levels.

o be successul, a modern version o a wage-led growth strategy will re-quire a restructuring o the financial sector. Te deregulated financial sectorhas uelled speculative growth and resulted in the worst recession since the

s. I a repeat o the crisis is to be prevented, this will require managinginternational capital flows, a reocusing o the financial sector on narrowbanking, the elimination o destabilizing financial innovations, and a higherfiscal contribution o the financial sector (e.g., in the orm o a financial trans-actions tax).

References

Atkinson, A.; Piketty, .; Saez, E. . op incomes in the long run o history, inJournal o Economic Literature, Vol. , No. , pp. .

Balcerowicz, L.; Fisher, S. (eds). . Living standards and the wealth o nations:Successesand ailures in real convergence(Cambridge and London, MI Press).

Barba, A.; Pivetti, M. . Rising household debt: Its causes and macroeconomicimplications a long-period analysis, in Cambridge Journal o Economics,Vol. , pp. .

Barbosa-Filho, N.; aylor, L. . Distributive and demand cycles in the USeconomy a structuralist Goodwin model, inMetroeconomica, Vol. ,Issue , pp. .

Bhaduri, A.; Marglin, S. . Unemployment and the real wage: Teeconomic basis or contesting political ideologies, in Cambridge Journal oEconomics, Vol. , pp. .

Blecker, R. . International competition, income distribution and economicgrowth, in Cambridge Journal o Economics, Vol. , pp. .


29/104


185

Bowles, S.; Boyer, R. . Wages, aggregate demand, and employment in an openeconomy: An empirical investigation, in G. Epstein and H. Gintis (eds):Macroeconomic policy aer the conservative era. Studies in investment, savingand finance(Cambridge, UK, Cambridge University Press).

Daudey, E.; Garca-Pealosa, C. . Te personal and the actor distributionso income in a cross-section o countries, inJournal o DevelopmentStudies, Vol. , Issue , pp. .

Dumnil, G.; Lvy, D. . Costs and benefits o neoliberalism: A class analysis,inReview o International Political Economy, Vol. , No. , pp. .

Dutt, Amitava. . Stagnation, income distribution and monopoly power, inCambridge Journal o Economics, Vol. , pp. .

Ederer, S.; Stockhammer, E. . Wages and aggregate demand in France: Anempirical investigation, in E. Hein and A. ruger (eds): Money, distribution,and economic policy alternatives to orthodox macroeconomics (Cheltenham,UK, Edward Elgar).

Epstein, G.; Burke, S. . Treat effects and the internationalization o production,Political Economy Research Institute, Working Paper .

European Commission. . Broad economic policy guidelines, inEuropeanEconomy, No. .

. . Broad economic policy guidelines, inEuropean Economy, No. .

. . Broad economic policy guidelines, inEuropean Economy, No. .

. . Te labour income share in the European Union, in Employment in

Europe, Chapter , pp. .Flaschel, P.; Proano, C. . AS-AD Disequilibrium dynamics and the aylorinterest rate policy rule: Euro-area based estimation and simulation, inP. Arestis, E. Hein and E. Le Heron (eds): Aspects o modern monetary andmacroeconomic policies(Houndsmill, Palgrave MacMillan).

Glyn, A. . Capitalism unleashed: Finance, globalization and welare(Oxord,Oxord University Press).

Harvey, D. .A brie history o neoliberalism(Oxord, Oxord University Press).

Hein, E.; arassow, A. . Distribution, aggregate demand and productivity

growth: Teory and empirical results or six OECD countries based on apost-Kaleckian model, in Cambridge Journal o Economics, Vol. , No. ,pp. .

; ruger, A. . Macroeconomic coordination as an economic policyconcept opportunities and obstacles in the EMU, in E. Hein, . Niechoj,. Schulten and A. ruger (eds):Macroeconomic policy coordination in Europeand the role o the trade unions(Brussels, EUI).

; Vogel, L. . Distribution and growth reconsidered empirical resultsor Austria, France, Germany, the Netherlands, the UK and the USA, inCambridge Journal o Economics, Vol. , No. , pp. .

Horn, G.; van reeck, . . Ungleichheit und aussenwirtschaflicheUngleichgewichte. Eine Keynesiansiche Erklrung, Proceedings oKeynes-Gesellschaf .

ILO. . World o Work Report : Income inequalities in the age o financialglobalization(Geneva).


30/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

186

IMF. a. Te globalization o labor, in World Economic Outlook, Chapter ,Apr. (Washington, DC).

. b. Globalization and inequality, in World Economic Outlook, Chapter ,Oct. (Washington, DC).

Jayadev, A. . Capital account openness and the labour share o income, inCambridge Journal o Economics, Vol. , pp. .

Lazonick, W.; OSullivan, M. . Maximising shareholder value: A newideology or corporate governance, inEconomy and Society, Vol. , No. ,pp. .

Marglin, S. . Growth, distribution, and prices(Cambridge, MA, HarvardUniversity Press).

; Schor, J. (eds). . Te golden age o capitalism. Reinterpreting the postwarexperience(Oxord, Clarendon).

Naastepad, C.W. M.; Storm, S. /. OECD demand regimes (),inJournal o Post-Keynesian Economics, Vol. , No. , pp. .

OECD. . Corporate saving and investment: recent trends and prospects,in OECD Economic Outlook, Vol. , No. , pp. (Paris).

. . Growing unequal? Income distribution and poverty in OECD countries(Paris).

Onaran, .; Stockhammer, E.; Grafl, L. Forthcoming. Te finance-dominatedgrowth regime, distribution, and aggregate demand in the US, in CambridgeJournal o Economics.

Piketty, .; Saez, E. . Income inequality in the United States, , inQuarterly Journal o Economics, Vol. , No. , pp. .

Power, D.; Epstein, G.; Abrena, M. . rends in the rentier income share inOECD countries , PERI Working Paper a.

Reinhart, C.; Reinhart, V. . Capital flow bonanzas: An encompassing view o thepast and present, NBER Working Paper No. W.

Rodrik, D. .Has globalization gone too ar?(Washington, DC, Institute oInternational Economics).

Rossman, P. .Financialization and casualization o labour building a trade

union and regulatory response, Paper presented at the Global Labour UniversityConerence, Feb. , Mumbai.

Rowthorn, R. .Demand, real wages and economic growth. Tames Papers inPolitical Economy, Autumn, pp. , reprinted in Studi Economici,Vol. , pp. .

. . Unemployment, wage bargaining and capital-labour substitution, inCambridge Journal o Economics, Vol., pp. .

Schulten, . . A European solidaristic wage policy?, inEuropean Journal oIndustrial Relations, Vol. , pp. .

Stein, U. .Zur Entwicklung der Sparquoten der privaten Haushalte eineAuswertung von Haushaltsdaten des SOEP, IMK Working Paper .

Stockhammer E. . Financialization and the slowdown o accumulation, inCambridge Journal o Economics, Vol. , No. , pp. .

. . Wage coordination or wage flexibility?, inIntervention, No. , pp. .
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1267552http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1267552


31/104


187

. .Determinants o unctional income distribution in OECD countries,IMK Studies, No. (Dusseldor).

. . Neoliberalism, income distribution and the causes o the crisis, inP. Arestis, R.Sobreira and J. L . Oreiro (eds): Te financial crisis, financial

regulation and global impact. Volume . Te financial crisis: Origins andimplications(London, Palgrave Macmillan).

; Ederer, S. . Demand effects o a alling wage share in Austria, in Empirica,Vol. , No. , pp. .

; Hein, E.; Grafl, L. . Globalization and the effects o changes in unctionalincome distribution on aggregate demand in Germany, inInternationalReview o Applied Economics, Vol. , No. , pp. .

; Onaran, . . Accumulation, distribution and employment: A structuralVAR approach to a Kaleckian macro-model, in Structural Change andEconomic Dynamics, Vol. , No. , pp. .

; ; Ederer, S. . Functional income distribution and aggregate demand inthe Euro area, in Cambridge Journal o Economics, Vol. , No. , pp. .

; Stehrer, R. Forthcoming. Goodwin or Kalecki in demand? Functional incomedistribution and aggregate demand in the short run, inReview o RadicalPolitical Economics.

Stolper, W.; Samuelson, P. . Protection and real wages, inReview o EconomicStudies, Vol. , pp. .

Storm, S.; Naastepad, C.W. M. . Labor market regulation and productivitygrowth: evidence or twenty OECD countries (), inIndustrialRelations, Vol. , No. , pp. .

UNCAD. . rade and Development Report (Geneva).

Vergeer, R.; Kleinknecht, A. . Te impact o labor market deregulation onproductivity: A panel data analysis o OECD countries (), inJournal o Post Keynesian Economics, Vol. , pp. .

Webb, S. . Te economic theory o a legal minimum wage, inJournal oPolitical Economy, Vol. , Dec., pp. .


32/104


33/104

189

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

Leveraging inequality*

Michael Kumhof

Research Department, International Monetary Fund

Romain Rancire

Research Department, International Monetary Fund

* Tis article was first published in the IMF quarterly Finance & Development(December ).
http://www-bcc.imf.org/external/pubs/ft/fandd/2010/12/Kumhof.htm#authorhttp://www-bcc.imf.org/external/pubs/ft/fandd/2010/12/Kumhof.htm#authorhttp://www-bcc.imf.org/external/pubs/ft/fandd/2010/12/Kumhof.htm#authorhttp://www-bcc.imf.org/external/pubs/ft/fandd/2010/12/Kumhof.htm#authorhttp://www-bcc.imf.org/external/pubs/ft/fandd/2010/12/Kumhof.htm#authorhttp://www-bcc.imf.org/external/pubs/ft/fandd/2010/12/Kumhof.htm#author


34/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

190

The United States experienced two major economic crises over the past years the Great Depression o and the Great Recession o. Income inequality may have played a role in the origins o both. Wesay this because there are two remarkable similarities between the eras pre-

ceding these crises: a sharp increase in income inequality and a sharp increasein household debtto-income ratios.

Are these two acts connected? Empirical evidence and a consistent the-oretical model (Kumho and Rancire, ) suggest they are. When asappears to have happened in the long run-up to both crises the rich lenda large part o their added income to the poor and middle class, and whenincome inequality grows or several decades, debt-to-income ratios increasesufficiently to raise the risk o a major crisis.

Shifting wealth

We looked at the evolution o the share o total income controlled by thetop per cent o US households (ranked by income) compared with ratioso household debt to income in the periods preceding and (seefigure ). Te income share o the top per cent increased rom per centin to per cent in and rom per cent in to per cent

in (we used ewer years beore than beore because theearlier data were highly distorted by the First World War). During the sametwo periods, the ratio o household debt to income increased dramatically. Italmost doubled between and , and also between and ,reaching much higher levels ( per cent) in the second period.

In the more recent period (), the difference between the con-sumption o the rich and that o the poor and middle class did not widen asmuch as the differences in incomes o these two groups. Te only way to sus-tain high levels o consumption in the ace o stagnant incomes was or poor

and middle-class households to borrow (see figure ).In other words, the increase in the ratios o debt to income shown infigure was concentrated among poor and middle-class households. In ,the debt-to-income ratio o the top per cent o households was per cent;or the bottom per cent the ratio was per cent. wenty-five years later,in a striking reversal, the ratio was per cent or the top per cent and per cent or the bottom per cent.

Te poor and the middle class seem to have resisted the erosion o theirrelative income position by borrowing to maintain a higher standard o living;

meanwhile, the rich accumulated more and more assets and invested in assetsbacked by loans to the poor and the middle class. Consumption inequalitythat is lower than income inequality has led to much higher wealth inequality.

Te higher indebtedness o the bottom income group has implicationsboth or the size o the US financial industry and its vulnerability to financial
http://www-bcc.imf.org/external/pubs/ft/fandd/2010/12/Kumhof.htm#ref3http://www-bcc.imf.org/external/pubs/ft/fandd/2010/12/Kumhof.htm#ref3


35/104

Leveraginginequality

191

Figure 1. Lending disposable income

As income inequality increases, the rich lend to workers, whose leverage increases.

Note: Income excludes capital gains.

Source: US Department of Commerce, Statistical Abstract of the United States(top panel);

Picketty and Saez, 2003 (income shares, bottom panel); and Federal Reserve Board,

Flow of Funds database (debt to GDP).

% %

25

30

35

60

55

45

50

40

23

25

27

35

31

33

29

1920 1922 1924 1926 1928 1930

% %

70

90

110

150

130

20

24

28

36

32

1983 1986 1989 1992 20011998 20041995 2007

Private noncorporate + trade debt to GNP

(left scale)

Share of top 5 per cent in incomedistribution (right scale)

Household debt to GDP (left scale)

Share of top 5 per cent in incomedistribution (right scale)

Figure 2. Increasingly indebted

Workers have been borrowing more as capital owners lend from their risingdisposable income.

Source: Authors calculations based on model simulations.

Debt-to-income ratio

0.3

0.5

0.7

1.5

1.1

1.3

0.9

Bottom 95 per cent of the wealth distribution (workers)

Top 5 per cent of the wealth distribution (capital owners)

1983 1986 1989 1992 20011998 20041995 2007


36/104

International

Journal

of Labour

Research

2011

Vol. 3

Issue 2

192

crises. Te bottom groups greater reliance on debt and the top groups in-crease in wealth generated a higher demand or financial intermediation.

Between and , the US financial sector grew rapidly the ratioo private credit to gross domestic product (GDP) more than doubled, going

rom to per cent. Te financial industrys share in GDP doubled, rom to per cent. With increased debt, the economy became more vulnerableto financial crisis. When a crisis eventually hit in , it brought withit a generalized wave o deaults; per cent o mortgage loans became delin-quent, and output contracted sharply.

Tere are o course other possible explanations or the origins o the crisis, and many have stressed the roles o overly loose monetary policy,excessive financial liberalization, and asset price bubbles. ypically these ac-tors are ound to have been important in the years just preceding the crisis,

when debt-to-income ratios increased more steeply than beore. But it canalso be argued, as in Rajan(), that much o this was simply a maniesta-tion o an underlying and longer-term dynamic driven by income inequality.Rajans argument is that growing income inequality created political pres-sure not to reverse that inequality, but instead to encourage easy credit tokeep demand and job creation robust despite stagnating incomes.

Modelling the facts

An economic model can clearly illustrate these links among income in-equality, leverage, and crises. Our model has several novel eatures that re-flect the empirical acts described above. First, households are divided intoone income group at the top per cent o the income distribution (call themcapital owners) that derives all its income rom returns on the economyscapital stock and rom interest on loans, and a second group composed o theremaining per cent (workers), who earn income in the orm o wages.

Second, wages are determined by a bargaining process between capital ownersand workers. Tird, all households care how much they consume, but capitalowners also care about how much capital physical capital and inancialassets they own. Tis implies that when capital owners income increases atthe expense o workers, they will allocate it to a combination o higher con-sumption, higher physical investment, and higher financial investment. Telatter consists o increased loans to workers whose consumption originallyaccounts or a very high per cent o GDP giving them the means to con-sume enough to support the economys production.

Our model can be used to show what happens afer the economy ex-periences a lengthy shock to the distribution o incomes in avour o capitalowners. Workers adjust through a combination o lowering their consump-tion and borrowing to limit the drop in their consumption (see figure ). Tisgradually raises workers debt-to-income ratio, which ollows the pattern and
http://www-bcc.imf.org/external/pubs/ft/fandd/2010/12/Kumhof.htm#ref6http://www-bcc.imf.org/external/pubs/ft/fandd/2010/12/Kumhof.htm#ref6


37/104

Leveraginginequality

193

magnitude documented in figure . Workers higher debt is made possible bythe lending o capital owners increased disposable income.

More saving at the top and more borrowing at the bottom mean con-sumption inequality increases signiicantly less than income inequality.

Saving and borrowing patterns o both groups spur a need or financial ser-vices and intermediation. As a result, the size o the financial sector roughlydoubles. Te rise o poor and middle-class household indebtedness begetsfinancial ragility and a higher probability o financial crises. With workersbargaining power, and thereore their ability to service and repay loans, re-covering only very gradually, loans continue to increase and the risk o acrisis persists. When the crisis does occur assumed here to materialize afer years there are large-scale household debt deaults on per cent o theexisting loan stock, accompanied by an abrupt output contraction, as oc-

curred during the US financial crisis.Te model points to a number o ways the increase in debt-to-income

ratios in the pre-crisis period could be more pronounced than shown infigure . First, i capital owners allocate most o their additional income toconsumption and financial investment rather than to productive investment,

Figure 3. Borrowing from Peter to pay Paul

When workers wages drop, they borrow more to mainta

Documents

ILO-The Case for Wage-led Policies