Content uploaded by Engelbert Stockhammer
Author content
All content in this area was uploaded by Engelbert Stockhammer on Feb 03, 2014
Content may be subject to copyright.
ePubWU Institutional Repository
Engelbert Stockhammer and Stefan Ederer
Demand effects of the falling wage share in Austria
Working Paper
Original Citation:
Stockhammer, Engelbert and Ederer, Stefan (2007) Demand effects of the falling wage share in
Austria. Department of Economics Working Paper Series, 106. Inst. für Volkswirtschaftstheorie
und -politik, WU Vienna University of Economics and Business, Vienna.
This version is available at:
Available in ePubWU: August 2007
ePubWU, the institutional repository of the WU Vienna University of Economics and Business, is
provided by the University Library and the IT-Services. The aim is to enable open access to the
scholarly output of the WU.
Vienna University of Economics & B.A.
Department of Economics Working Paper Series
Demand effects of the falling wage share
in Austria*
Engelbert Stockhammer†, Stefan Ederer‡
Working Paper No. 106, August 2007
Abstract ― This paper aims at empirically estimating the demand effects of
changes in functional income distribution for Austria. Based on a Post-Kaleckian
macro model, this paper estimates the effects of a change in the wage share on the
main demand aggregates. The results for the behavioral functions for consumption,
investment, prices, exports and imports are compared with the specifications of the
WIFO macro model and the IHS macro model. A reduction in the wage share has a
restrictive effect on domestic demand as consumption decreases more strongly than
investment increases. Because of the strong effects on net exports the overall effects
of a decrease in the wage share are expansionary. However the latter effect
operates only as far as the fall in the wage share increases competitiveness. As
wage shares were also falling in Austria’s main trading partners, the effect seems
to have been neutralized.
Keywords: distribution, demand, investment, consumption, foreign trade,
macroeconomics, Keynesian economics
JEL-Classification: E12, E20, E22, E25, E61
* The authors are grateful to Amit Bhaduri, Kazimierz Laski, Özlem Onaran and Martin Riese for comments.
Support from FWF Project Nr. P18419-G05 is acknowledged. The usual disclaimers apply.
† Corresponding author: Dr. Engelbert Stockhammer, Institut für Geld- und Fiskalpolitik, Vienna University of
Economics and Business Administration, Augasse 2-6, A-1090 Vienna, Austria, e-mail:
engelbert.stockhammer@wu-wien.ac.at, tel: +43 1 31336 4509, fax: +43 1 31336 728
‡ Vienna University of Economics and Business Administration, e-mail: stefan.ederer@wu-wien.ac.at
1 Introduction
The wage share has fallen substantially in Austria over the past 25 years (Figure 1). While the
causes for this fall have recently been subject to debate,1 there has been comparatively little
research its effects. In particular there is no analysis of its effects on aggregate demand. A
priori one would expect a rise in the wage share to have a positive effect on consumption, a
negative effect on investment expenditures and a negative effect on net exports (because
given the level of productivity an increase in the wage share implies a loss in international
competitiveness). However, most of today’s macro economic theory does not attribute much
significance to functional income distribution in the formation of aggregate demand, as a look
into intermediate or advanced macroeconomic textbooks (Blanchard 2006, Mankiw 2006,
Romer 2006) will readily confirm. This is also reflected in the most widely used
macroeconomic models for Austria, the WIFO model (Baumgartner, Breuss and Kaniovski
2004) and the IHS model (Hofer and Kunst 2005). Both do not include the wage share
explicitly. However, both include the unit labor costs as influencing prices and net exports.
1 Marterbauer and Walterskirchen (2003) investigate the Austrian case and highlight the role of unemployment.
IMF (2007), and OECD (2007), investigate the effects of globalization, Breuss (2007) explicitly includes the
effects of Eastern Enlargement.
Figure 1: Adjusted wage share in Austria and the Euro area
Adjusted wage share (GDP at factor costs)
55
60
65
70
75
80
85
90
1960
1963
1966
1969
1972
1975
1978
1981
1984
1987
1990
1993
1996
1999
2002
2005
Euro area (12 countries)
EU12 (inc l. W-Ger)
Austria
Source: AMECO
The neglect of income distribution in today’s macroeconomics is in contrast to a long-
standing tradition in economics that has given income distribution a prominent role in demand
formation. Most of classical economics was centered around the distribution of income
between landowners, capitalists and workers (Ricardo 1951, Marx 1976). In the postwar era
Post Keynesian economists (Robinson 1956, 1962, Kalecki 1954, 1971, Kaldor 1956, 1957)
have highlighted the role of income distribution. In these models functional income
distribution would affect demand even in a closed economy setting as wage incomes typically
are associated with higher marginal propensities to consume than capital incomes and
investment is, to some part, financed out of retained earnings.
2
In this paper the question what the effects of a change in the wage share on aggregate
demand is, will be investigated empirically for Austria based on a Post-Kaleckian macro
3
model. The model estimated is a version of the Bhaduri and Marglin (1990) model which
allows for wage-led as well as for profit-led demand regimes according to the relative size of
the consumption differential, the sensitivity of investment to profits and the sensitivity of net
exports to unit labor costs.
While the theoretical model is Post-Keynesian, our estimation strategy is pragmatic.
Each behavioural function is estimated in a form that ensures comparability with the
respective specifications in the WIFO macro model and the IHS model; the key difference
being that measures of functional income distribution are included. In doing so, we hope to
demonstrate the plausibility of our results, but also to prepare the ground for a fruitful
discussion between Post Keynesian and ‘mainstream’ economics.2 Thus our specifications
can be seen as extensions rather than criticisms of these models.
The paper is structured as follows. Section 2 presents the theoretical background and the
Post-Kaleckian model, on which the empirical estimations are based. Section 3 summarizes
the empirical literature on these models. Section 4 presents the econometric results for the
effect of changes in functional income distribution on private consumption, private
investment and net exports. Section 5 summarizes the key findings and discusses policy
conclusions.
2 Theoretical background: wage-led und profit-led demand regimes
This section will present the macroeconomic model that forms the basis for the empirical
analysis of the effects of changes in functional income distribution on aggregate demand. The
2 A discussion of what ‘mainstream’ economics is beyond the scope of this paper. As the most widely used
macro models for Austria, the WIFO model and the IHS model by definition qualify as mainstream. Both are
short run models with a Keynesian flavour which, on a theoretical level, is why the behavioural equations
estimated are comparable, if different in detail, with our Post-Kaleckian model. New Keynesian macro models
that use long run restrictions extensively, like the OeNB model (Fenz and Spitzer 2005), would be much more
difficult to compare to our model.
model allows for wage-led as well as profit-led demand regimes and is similar in spirit to
Bhaduri and Marglin (1990). While in the classical Kaleckian model (for a closed economy)
an increase in the wage share will always lead to an increase in demand (Kalecki 1954,
Rowthorn 1989, Blecker 1999), this is not necessarily the case here because there is a positive
effect of profits on investment that has been highlighted by classical economics (Ricardo
1951, Marx 1976) and the recent literature on the role of internal finance because of
imperfections on the credit market (Fazzari and Mott 1986, Hubbard 1998). The question
whether the positive effect of wages on consumption or the negative effect of profits on
investment is larger, becomes an empirical one. In an open economy additional negative
effects will operate through net exports (Blecker 1989, 1999, 2002).
Aggregate demand (Y) is the sum of consumption (C), investment (I), net exports (NX)
and government expenditure (G). All variables are in real terms. In a general formulation,
consumption, investment and net exports are written as functions of income (Y), the wage
share (), and some other control variables (summarized as z). These latter are assumed to be
independent of output and distribution. Government expenditures are considered a function of
output only. Aggregate demand then is:
),(),,(),,(),( GNXI zYGzPYNXzYIYCY
+
+
Ω
+Ω= (1)
and (2)
),( P
zfP Ω=
This model is rather general in that it can be reduced to a standard Keynesian short-run
model (e.g Blanchard 2006) if C/ and I/ are assumed to be zero. For example the
WIFO and the IHS models do not include effects of income distribution on consumption and
4
5
investment.3 Only in the net exports function does income distribution usually play a role,
albeit in an indirect way. Typically export and import functions include a price term and
prices are thought to depend (among other things) on unit labor costs. Unit labor costs are by
definition related to the wage share.
The inclusion of income distribution shall briefly be motivated. In the consumption
function the basic assertion is that wage incomes and profit incomes are associated with
different propensities to consume. The Kaleckian assumption is that the marginal propensity
to save is higher for capital incomes than for wage income; consumption is therefore expected
to increase when the wage share rises. Standard investment functions depend on output and
the real interest rate or some other measure of the cost of capital. In addition to that,
investment in our model is expected to decrease when the wage share rises because future
profits may be expected to fall. In classical economics it was a straightforward assumption
that the capital accumulation was a positive function of the rate of profit. Consequently
investment ought to be a function of profits. Today it is often argued that retained earnings are
a privileged source of finance and may thus influence investment expenditures. This had
already been highlighted by Kalecki (1954) and been rediscovered in the 1980s by
mainstream economists (Stiglitz and Weiss 1981).
Net exports are a negative function of domestic demand, a positive function of foreign
demand, and will depend negatively on domestic prices. Domestic prices in turn depend on
unit labour costs (ULC) and import prices. This is the structure in our model as well as the
WIFO model and the IHS model. It is important to note that this structure implies that net
exports depend (among other things) on (changes in) the wage share. As the price equation
indicates the marginal effect of a change in ULC on prices, expressing the relation between
3 In the case of consumption one may argue that there is an indirect effect because a redistribution from, say,
profits to wages will affect disposable income as part of profits are retained and thus do not enter disposable
income.
real unit labor costs and prices is only a matter of re-parameterization.4 As real unit labor
costs are by definition related to the wage share, domestic prices are thus a function of the
wage share (as in equation 2).5
Government expenditures can react to income distribution; however this is ignored in our
analysis, which focuses on the private sector. A serious treatment of the public sector is
beyond the scope of this paper.
The model only covers the goods market. Typically the goods market is complemented by
a distribution function (Marglin and Bhaduri 1990) that describes the effects of changes in
economic activity on income distribution. However, the focus of this paper is on the demand
effects and the wage share ()6 is taken as exogenous. Thus feedbacks, for example, from
growth on income distribution via lower unemployment and a better bargaining position of
labour are ignored at this stage. It is therefore a partial model of a basic private open economy
type. Because of our focus on the effect of changes in the functional income distribution, the
effects of fiscal policy are excluded from the analysis.
Differentiating Y with respect to and collecting terms gives
1
2
1
*
h
h
d
dY
−
=
Ω (3)
where ⎟
⎠
⎞
⎜
⎝
⎛
∂
∂
+
∂
∂
+
∂
∂
+
∂
∂
=Y
G
Y
NX
Y
I
Y
C
h1 and ⎟
⎠
⎞
⎜
⎝
⎛
Ω∂
∂
+
Ω∂
∂
+
Ω∂
∂
=NXIC
h2.
4 Assume a price equation of the form lnP=f(lnULC, zP). Real unit labor costs (RULC) defined as ULC/P and are
identical to the wage share; thus ln = lnULC – lnP. Then ePRULC = ePULC/(1-ePULC), where ePRULC is the
elasticity of prices with respect to RULC and ePULC is the elasticity of prices with respect to RULC.
5 In the OECD data set that is used for the empirical analysis, ULC are defined as nominal output divided by real
productivity. Real unit labor costs are definded as nominal labor unit costs divided by the GDP deflator.
Therefore, in this definition, real unit labor costs are identical to the wage share.
6 Functional income distribution and its measure, the wage share, are used synonymously throughout this paper.
6
7
The term 1/(1-h1) in equation 3 is a standard multiplier and has to be positive for stability.
The sign of the total derivative therefore depends on the sign of the numerator. h2 is the sum
of the partial derivatives of the components of demand with respect to income distribution.
This sum is private excess demand, that is, the change in demand caused by a change in
income distribution given a certain level of income. It is impossible to sign h2 a priori, since
we hypothesize that C/>0, I/<0, and NX/<0. The sum of these effects can
therefore only be determined empirically. Determining the sign of private excess demand is
therefore the focus of the empirical estimations in this study.
The total effect of the increase in the wage share on aggregate depends on the relative size
of the reactions of the components of GDP, namely consumption, investment and net exports
to changes in income distribution. If it is positive (Y*/>0), the demand regime is called
wage-led. If the effect is negative (Y*/<0), it is called profit-led.
Given that Austria is a small open economy, one would expect net exports to play a major
role in empirically determining the overall outcome. However, it is important to distinguish
between the domestic sector of the economy and the open economy for policy reasons as well
as for theoretical ones. For economic policy it is important to realize that while individual
countries can increase demand by increasing exports, the world as a whole of course cannot.
Therefore with respect to the empirical investigation, what is at stake is whether private
excess demand in the domestic sector is wage-led or profit-led. The domestic sector is defined
with respect to consumption and investment only, assuming that the net export position does
not change (as would be the result if wages were to change simultaneously in all countries). If
consumption reacts more sensitively to an increase in the wage share than investment,
domestic demand will be wage-led.
8
3 Related literature
The Bhaduri and Marglin (1990) model is a flexible Post-Kaleckian macro model that is
widely used in modern Post Keynesian economics.7 It differs from the classical Kaleckian
model as it allows for wage-led as well as profit-led demand regimes. The question whether
the positive effect of an increase in the wage share on consumption outweighs the negative
effect on investment and on net exports becomes an empirical one. It has thus inspired
empirical literature, which will briefly be surveyed here. As none of this empirical literature
(with one exception) refers Austria, some papers on Austria that have a different theoretical
background are also covered.
The empirical tests of the Bharduri-Marglin models can be grouped into two estimation
strategies. The first group of papers tries to estimate the full model, that is, a goods market
equilibrium relation and a distribution function. Stockhammer and Onaran (2004) estimate a
structural VAR model consisting of the variables capital accumulation, capacity utilization,
profit share, unemployment rate and labour productivity growth for the USA, UK and France.
The goods market is modelled by a model based on Marglin and Bhaduri (1990). From the
empirical investigation it is concluded that unemployment is determined by the goods market,
and that the impact of income distribution on demand and employment is very weak. Onaran
and Stockhammer (2005) employ a similar model for Turkey and Korea. The advantage of the
systems approach is that the interaction between the variables can be incorporated. The
disadvantage of the VAR is that it is difficult to identify effects of individual variables.
The second group of papers focuses on the goods market and estimates consumption,
investment and net export equations. This is also the approach pursued in this paper. The first
paper along these lines was Bowles and Boyer (1995). While the paper has become a seminal
reference point for later research, the econometric methods employed are not up-to-date. In
7 Kaleckian economics is considered one of the main streams within Post Keynesian economics.
9
particular, they fail to discuss the time series properties of the economic variables and ignore
the issue of unit roots. As a consequence, they do not apply difference or error correction
models that form the core of modern time series econometrics. Hein and Ochsen (2003) focus
on the interest rate as exogenous variable and estimate savings and investment
econometrically and try to characterise the accumulation regimes of France, Germany, the
USA and the UK. Naastepad and Storm (2006/2007) estimate a similar model for eight OECD
countries. The estimated model is strictly derived from the theoretical one. Consequently the
estimated equations are typically in ratio form, which are not the ones favoured by modern
time series econometrics. Compared to our findings, their effects on consumption and
investment are high, but those on net exports are small.
Our approach is similar in spirit to Bowles and Boyer (1995) but uses modern econometric
techniques and seeks comparability with standard behavioral equations. A similar approach
has been taken by Ederer and Stockhammer (2007) for France and by Stockhammer, Onaran
and Ederer (2007) for the Euro area. Hein and Vogel (2007), building on Ederer and
Stockhammer (2007) estimate a similar model for Austria, France, Germany, the Netherlands,
UK and the USA, but offer a much simpler treatment of international trade.
Other than Hein and Vogel (2007) no estimations of the Bhaduri-Marglin model are
available for Austria. Hein and Vogel offer no detailed discussion of the Austrian case. The
differences and similarities will be discussed below. Marterbauer et al. (2006) is related to the
behavioural assumption of the Post-Keynesian consumption function without sharing its
theoretical basis. They calculate different consumption propensities according to income
groups for Austria. The lowest income third has a long-run marginal propensity to consume of
0.8, whereas for the highest third they get 1.2.
Throughout the empirical section comparisons will be made to the WIFO model of the
(Baumgartner, Breuss and Kaniovski 2004) and the IHS model (Hofer and Kunst 2005).
These are the two best established macro-econometric models for the Austrian economy and
10
are used for short to medium run simulations. Both are similar in structure and complexity
(consisting of some 30 behavioural and roughly the same number of identities). While we use
these models for comparisons of individual behavioural equations, they have been built as full
macroeconometric models without necessarily optimizing the fit of an individual equation.
However, the equations can still serve as a reference point in checking the plausibility of our
results.
4 Empirical results
The model is estimated by means of separate single equations for consumption, investment,
exports and imports. As far as possible, the specifications where chosen as to be consistent
with the macroeconomic models for Austria of WIFO and IHS and augmented for a
distributional variable. The econometric specifications are following the standard practice in
modern econometric modelling. Thus, error correction models (ECM) are estimated whenever
feasible. When the results were unsatisfactory and/or there was no indication of cointegration,
an unrestricted autoregressive distributed lag model (ADL) was estimated. ADL models are
general in that various specifications can be written as restrictions on an ADL model. In all
cases where ECM specifications did not work, the ADL suggested that a difference model
should be applied.
All data is taken from the OECD Economic Outlook database (downloaded in 2006). The
sample is 1960-2005. C, I, X, M, Y, W and R are real consumption expenditures, investment
expenditures, exports, imports, GDP, wages and profits respectively. Wages and profits were
deflated with the GDP deflator. In the OECD database, the wage share and real unit labor
costs are identical by definition. Variable definitions can be found in the Appendix (Table
A.1). Unit root tests8 suggest that all these variables are integrated of order one (I (1)). Thus,
8 The results of the unit root tests are reported in the Appendix (Table A.2).
11
ECM, ADL or difference specifications are applicable. WIFO and IHS are using annual data
of the national accounts published by Statistik Austria. The results of our estimations
therefore may therefore differ from the WIFO and IHS estimations because of the database.
However this difference should be minor.
There is a major qualification of the results to be reported. Functional income distribution
is assumed to be exogenous which obviously is not the case. Demand will affect functional
income distribution in at least two ways. First, mark-ups typically vary pro-cyclically (e.g. if
mark-ups are set on normal unit labour costs). Second, unemployment will typically (though
usually with a time lag) have a negative affect on the wage share. Endogenizing income
distribution nevertheless would require a different modelling strategy such as the structural
VAR approach by Stockhammer and Onaran (2004).
4.1 Consumption
Consumption usually is estimated as a function of income. In order to include a distributional
variable in the consumption function, we apply two different strategies. First, income is
separated into wage income and profit income. When estimated separately, the difference
between the marginal propensities to consume out of wages and profits gives the effect of a
change in functional income distribution. Second, the wage share () is directly included as
explanatory variable.
In our estimations the GDP of the private sector stands for income. In standard
macroeconomic models like the WIFO and the IHS model, disposable income is used as
explanatory variable. As tax cannot be assigened to wage and profit incomes, functional
income distribution cannot be calculated on the basis of disposable income. In our estimations
therefore we use GDP of the private sector as income.
12
Table 1 reports the results of the estimations. The ADF test rejected cointegration at the
10% level, which is rather surprising because consumption functions usually are modelled as
ECM.9 The test results however are close to the critical values, thus we also estimate the two
functions as ECMs. For econometric reasons all variables in the equations are in logarithmic
form.10 The relevant coefficients of both estimations are meaningful and statistically
significant.
Table 1: Regression results for the consumption function
Dependent variable: Δln C Dependent variable: Δln C
Variable Coeff.
Prob. Variable Coeff.
Prob.
c 1.97
0.01 c 0.23
0.04
Δln W 0.47 0.00 Δln Ω 0.06
0.67
Δln R 0.25 0.00 Δln Y 0.67 0.00
ln C (-1) -0.36 0.01 ln C (-1) -0.33 0.01
ln W (-1) 0.23 0.01 ln Ω (-1) 0.10 0.08
ln R (-1) 0.10 0.02 ln Y (-1) 0.30 0.01
Adj. R² 0.64 Adj. R² 0.66
DW stat. 2.04 DW stat. 1.97
Elasticities:
eCW 0.64 eCΩ0.29
eCR 0.27
eCY 0.91 eCY 0.91
Marginal effects (at sample means):
∂C/∂W 0.74
∂C/∂R 0.44
∂(C/Y)/∂Ω 0.30 ∂(C/Y)/∂Ω 0.33
Notes: eCW, eCR, eCY, eC are the elasticities of consumption with regard to wages, profits, income and the wage
share, respectively. C/W, C/R, (C/Y)/ are marginal effects of consumption in response to changes of
wages, profits and the wage share, respectively. The effect of a change in the wage share is calculated as the
percentage increase of consumption relative to total GDP. All marginal effects are calculated at the mean of the
sample.
In the first equation, the long-run elasticities for wage and profit income are 0.64 and 0.27,
respectively. Due to the use of logarithms, these are elasticities and have to be converted into
9 Both the WIFO and the IHS are applying an ECM for the consumption function.
10 Economic variables that are supposed to grow exponentially have to be used in logarithmic form in order to
get stationary. See e.g. Stewart (2005: 742ff) for further discussion.
13
marginal effects by multiplying with the share of consumption in wages and profits,
respectively. This yields marginal propensities to consume of 0.74 (for wage income) and
0.44 for profit income. The difference of these two values (0.30) is the effect of a change in
functional income distribution on consumption. A redistribution of income of 1 percentage
point (of GDP) from profits to wages would in the average induce additional consumption
expenditures of 0.30% of GDP.
As the function is estimated in logarithmic form, the sum of the wage and profit income
elasticity is the total income elasticity of consumption. Its value is 0.91 which is close to 1,
the value assumed by the permanent income hypothesis. For the second equation, the long-run
marginal effect of a change in income distribution is 0.33 and the long-run income elasticity
of consumption expenditures is 0.91, which are almost exactly the same values as obtained
through the first estimation strategy. Both the first estimation, where income distribution was
included by separating income into wage and profit income, and the second, where we
directly applied a distributional variable yield (almost) the same results.
In the WIFO macroeconomic model, the only explanatory variable is the disposable
income of households. The IHS model is estimating consumption of durables, non-durables
and services separately, all of them with disposable income as explanatory variable.11 The
income elasticity estimated by the WIFO is 1.12, which is slightly above our value. The IHS
reports income elasticities between 0.70 and 1.54, depending on the type of consumption.
Thus, the comparison of our income elasticities with the WIFO and the IHS models indicate
that our estimation results are very plausible. Hein und Vogel (2007) get an effect of a rise of
the wage share on consumption of -0.24% of GDP. This value is also very close to our results.
However, they apply an estimation equation in differences. The results therefore are not
directly comparable.
11 The equation for durables additionally contains the long-term real interest rate as explaining variable.
14
Our estimation results imply that a change in functional income distribution is clearly
affecting consumption expenditures. Adding a distributional variable to standard consumption
functions is therefore providing additional information about macroeconomic consumption
behaviour.
4.2 Investment
Investment usually is modelled as a function of output (Y) and the long-term real interest rate
or some other measure of capital cost. In our model, profits are included as explanatory
variable in the equation (see section 2). As the coefficient on profits gives the effect of profits
holding total income constant, this coefficient represents the effect of a redistribution of
income on investment. The explanatory variables thus are GDP, profits and the long-term real
interest rate. First, the investment function was estimated as an ECM without a restriction. For
the second equation, the long-run coefficient of output on investment was restricted to one. A
similar restriction in the investment function is used in the WIFO model. It implies that in the
long-run the investment share in output is stable.
Table 2 reports the results of the two estimations. For both equations the estimated long-
run coefficients are very similar. The coefficients on the interest rate are statistically
insignificant both in the short and the long run. In the first equation, the long-run coefficient
on profits is also statistically insignificant. Applying the restriction in the second equation
turns the coefficient significant.12
12In the estimations of the IHS the interest rate is also not statistically significant. The WIFO is using a variable
for capital cost, which is calculated from the inflation rate of capital goods, the nominal interest rate and a
depreciation rate. Additionally, a factor for the Austrian tax system was included. The capital cost however only
have a short-run effect. Unfortunately no information is given about the statistically significance of this variable.
Similarly, Wesche (2000: 18) concludes from a study with firm-level data that “the traditional interest-rate
channel is negligible for Austrian enterprises ”.
15
The long-run elasticity of investment with respect to the wage share is 0.15 for the first
estimation and 0.30 for the second. These results correspond to a marginal effect of the wage
share on investment of -0.08 and -0.15 (at sample means). A redistribution of income of 1
percentage point (of GDP) from profits to wages would therefore reduce investment
expenditures by 0.08% or 0.15% of GDP. Due to the application of logarithms in the
equations, we have to sum up the long-run coefficients on GDP and profits in order to get the
total income elasticity. Both equations yield approximately the same income elasticity of 1.30.
Table 2: Regression results for investment function
Dependent Variable: Δln I Dependent Variable: Δln I
Variable Coeff.
Prob. Variable Coeff.
Prob.
c -1.97
0.01 c -1.59
0.00
Δln Y 3.46 0.00 Δln Y 3.34 0.00
Δln R -0.68 0.02 Δln R -0.63 0.02
Δln i 0.00 0.90 Δln i 0.00 0.80
ln I (-1) -0.43 0.00 ln I (-1)/Y(-1) -0.44 0.00
ln Y (-1) 0.50 0.00 ln R (-1) 0.13 0.00
ln R (-1) 0.06 0.55 ln i (-1) 0.00 0.82
ln i (-1) 0.00 0.87
Adj. R² 0.63 Adj. R² 0.63
DW stat. 2.11 DW stat. 2.07
Elasticities:
eIR 0.15 eIR 0.30
eIY 1.31 eIY 1.30
Marginal effects (at sample means):
∂(I/Y)/∂Ω -0.08 ∂(I/Y)/∂Ω -0.15
Notes: eIR, eIY are the elasticities of investment with regard to profits and income, respectively. I/ is the
marginal effects on investment in response to changes of the wage share. All marginal effects are calculated at
the mean of the sample.
The WIFO model differentiates between investment in machinery and equipment and in
construction investment (excluding housing investment). Both equations are estimated with
the GDP of the private sector and capital costs as explaining variables. The IHS model also
distinguishes investment in machinery and equipment and construction investment (including
housing investment). The explaining variables are GDP and the long-term real interest rate.
16
The income elasticities estimated by the WIFO are 1 (for machinery and equipment) and 2
(for construction), where the former one is imposed by a restriction. The long-term income
elasticities in the IHS model are 1.39 for machinery and equipment and 0.79 for construction
investment. All these values are similar to our results. Our estimations therefore seem to give
plausible results.
Hein und Vogel (2007) estimate investment as a function of the profit share and GDP in
first differences. Yet, the coefficient on the profit share is not statistically significant. For the
income elasticity they obtain a value of 1.79. Their results therefore are similar to our first
equation. However, imposing a restriction following the WIFO model, we get a statistically
significant coefficient on profits.
As for consumption, a distributional variable seems to affect investment. Though, the
effects of a change of functional income distribution on investment are considerably smaller
than on consumption. Furthermore, estimating the equation without restriction gives
statistically insignificant coefficients for profits. The results for the distributional effect on
investment therefore should be treated with caution. Although they will be included in the
calculation of the overall effect (see section 4.4), the values reported above have to be
considered as upper bound estimates.
4.3 Prices
For the effects on consumption expenditures and investment, it is not important whether a
change in income distribution is associated with a change in the price level. However, exports
and imports depend on relative prices. In order to identify effects of a shift in functional
income distribution on net exports, the price effects of this shift have to be identified in a first
step. Typically prices are explained by unit labor costs and import prices. This price equation
permits the calculation of a rise in real unit labor costs (and thus the wage share) on prices.
17
The resultant rise in prices entails a deterioration of international competitiveness and
therefore affects net exports. This second step of the calculation is subject of the next section.
Here the price effects will be analyzed.
Two regressions for prices are estimated. The first one estimates export prices (Px) as a
function of domestic prices (P) and import prices (Pm). In the second one domestic prices are
explained by nominal unit labor costs (ULC), import prices and domestic GDP (Table 3).
Both our estimations were performed in difference form after ECM specifications proved
unsuccessful. The coefficients of both estimations are statistically significant (with the
exception of GDP in the price equation) and have the expected signs.
Table 3: Regression results for price functions
Dependent Variable: Δln Px Dependent Variable: Δln P
Variable Coeff.
Prob. Variable Coeff.
Prob.
C 0.00
0.40 c 0.00
0.08
Δln Pm 0.57 0.00 Δln Y 0.07 0.16
Δln P 0.33 0.01 Δln Pm 0.15 0.00
AR(1) 0.15
0.36 Δln ULC 0.34 0.00
Δln P(-1) 0.35 0.00
Adj. R² 0.81 Adj. R² 0.92
DW stat. 1.79 DW stat. 2.13
The price elasticity (with respect to domestic prices) of export prices is 0.33, the import
price elasticity is 0.57 (Table 3, left side). The equation for domestic prices is estimated as a
partial-adjustment-model with the lagged prices as additional explaining variable. The
resultant long-run unit labour costs of domestic prices is 0.52 [0.34/(1-0.35)]. To obtain the
unit labor costs elasticity of export prices, the price elasticity of export prices is multiplied by
this value. We get a value of 0.18.
In the WIFO model export prices are estimated directly as a function of unit labor costs.
Domestic prices depend on unit labor costs, import prices and the output gap. Both equations
are estimated in differences. The IHS estimated various price equations for consumption and
investment prices as functions of unit labor costs, import prices and the output gap. All
18
equations are estimated in differences. Export prices are taken as exogenous. The WIFO gets
a unit labor costs elasticity of exports of 0.23, which is rather close to our result (0.18). The
import price elasticity of export prices of the WIFO is 0.45. This value also is not far from the
one obtained by our estimations (0.57).
The elasticities estimated by the domestic price equation are (partly) also very similar. For
the unit labor costs elasticity we get a value of 0.52, where the WIFO is estimating a
coefficient of 0.57. For the import price elasticity of domestic prices however we get different
values. The result of our estimations is 0.22 [0.15/(1-0.35)], whereas the WIFO gets a value of
0.54. Because the IHS is taking export prices as exogenous, we cannot compare results for
this equation. For domestic prices the IHS gets a unit labor costs elasticity of 0.47 and an
import price elasticity of 0.31. Both values are in the same range as our estimations. The
results of the estimations of the price functions therefore seem plausible.
The elasticity of domestic prices with respect to nominal unit labor costs calculated above
is 0.52. Thus, an increase in nominal unit labor costs by 1% increases domestic prices by
0.52%. The difference between the rise in nominal unit labor costs and prices by definition is
the resultant change in real unit labor costs. In order to get the increase in domestic prices
arising from a rise of real unit labor costs of 1%, which is of interest here, nominal unit labor
costs have to rise by 2.10% [1/(1-0.52)]. Prices at the same time rise by 1.10% [0.52*2.10].
The elasticity of domestic prices with respect to a rise in real unit labor costs therefore is 1.10.
This calculation is shown in columns 1 and 2 of Table 4. In order to obtain the real unit labor
costs elasticity of export prices (which will enter the calculation of the net exports effect in
the following section) this value is multiplied by the price elasticity of export prices (column
3). The result is presented in column 4. The real unit labor costs elasticity of export prices is
0.36.
19
Table 4: Price elasticities
1 2 3 4
ePULC ePRULC ePxP ePxRULC
0.52 1.1 0.33 0.36
Note: Column 1 and 3 estimates from Table 3. Column 2: ePRULC is calculated as ePULC (column 2) divided by 1-
ePULC. Column 4: ePxRUlC is the real unit labor costs elasticity of export prices.
4.4 Net exports
Exports are estimated as a function of export prices relative to an index of competitor prices
on export markets (Pxc). As these countries represent the main trading partners of Austria, the
GDP of the Euro 12-countries was include in the equation in order to represent foreign
demand. Foreign prices are represented by the price deflator for competitors’ prices already in
Euro, thus the (nominal) exchange rate is not included in the estimation. As there is no
indication of cointegration, exports are estimated in first differences.
Imports are also estimated in difference form. The explanatory variables are GDP and
export prices relative to import prices. The use of export prices is motivated by the fact that in
the GDP deflator prices for non-tradable goods and services are included, which are not
relevant for imports. The export price deflator therefore is a better representative for
competitors’ prices on import markets.
Table 5 presents the results of the estimations for exports and imports. All relevant
coefficients are statistically significant and have the expected signs. The price elasticity of
exports is -0.28, the elasticity with respect to foreign GDP is 1.33. For imports the price
elasticity is 0.79, which is remarkably higher. The demand elasticity of imports is 2.00.
20
Table 5: Regression results for export and import functions
Dependent Variable: Δln X Dependent Variable: Δln M
Variable Coeff.
Prob. Variable Coeff.
Prob.
c 0.02
0.02 c 0.00
0.97
Δln Y_EU12 1.33 0.00 Δln Y 2.00 0.00
Δln Px/Pxc -0.28 0.00 Δln Px(-1)/Pm(-1) 0.79 0.03
AR(1) 0.09
0.67 AR (1) -0.15 0.35
Adj. R² 0.46 Adj. R² 0.63
DW stat. 1.92 DW stat. 2.10
Export and import functions of the WIFO and IHS models also depend (among other
variables) on relative prices. The WIFO export equation uses export prices relative to a world
price deflator and weighted GDP of the Austrian trade partners. The equation is estimated as
an ECM, though the price variables are not included in the error correction term. In the long
run, therefore prices would not influence export performance. The IHS is estimating a
difference model with export prices relative to import prices and demand of Austrian export
markets as explaining variables. The coefficients of the WIFO model are -0.28 for the price
elasticity and 2.4 for the elasticity with respect to foreign demand. Thus, the price elasticity
estimated by the WIFO is close to our results. The higher value for the coefficient on foreign
demand is probably a result of the use of a different variable. The IHS gets a price elasticity of
-0.24 and an elasticity with respect to foreign demand of 1.12. Both are similar to our results.
Both WIFO and IHS use the ratio of imports and the sum of demand components
weighted by a constant import share. The explanatory variables in the WIFO model are the
import prices relative to domestic prices. The IHS uses import prices relative to export prices.
The explanatory variable for relative prices in our equation thus is the same as in the IHS
model. The price elasticity of imports is 0.2813 in the WIFO model it and 0.42 for the IHS
model.
13 This is only the short run value of the WIFO equation. Since a partial adjustment model is applied, the short
run effect has to be corrected in order to determine the long run effect. However, this would increase the value to
1.91, which we consider implausibly high.
21
The calculation of the effect of a shift in income distribution on exports and imports
involves several steps (which are presented in table 6). The elasticity of exports and imports
with respect to a rise of 1%-point in real unit labor costs includes the price elasticities and the
effect of real unit labour costs on prices. As explained in section 4.3, we derive from the
equation for prices that an increase of real unit labor costs by 1% increases export prices by
0.36%. This value shows up in column 1. By multiplying with the estimated export price
elasticity of exports and imports (column 2) we get the real unit labor costs elasticity of
exports and imports, respectively (column 3). This value has to be transformed by dividing
through the average of real unit labor costs (column 4) in order to get the semi-elasticity of
exports and imports with respect to real unit labor costs, which by definition is the same as the
effects of changes in the wage share. As a result, we get the elasticity of exports and imports
with respect to the wage share, which is -0.20 (column 5).
Table 6: Export and import effects
1 2 3 4 5 6 7
Exports ePxRULC eXPx eXRULC 1/RULC eXΩX/Y d(X/Y)/dΩ
X/Y 2005 0.36 -0.28 -0.10 1.97 -0.20 0.53 -0.11
X/Y avg. 0.36 -0.28 -0.10 1.97 -0.20 0.29 -0.06
X/Y 1960 0.36 -0.28 -0.10 1.97 -0.20 0.13 -0.03
Imports ePxRULC eMPx eMRULC 1/RULC eMΩM/Y d(M/Y)/dΩ
X/Y 2005 0.36 0.79 0.28 1.97 0.57 0.49 0.28
X/Y avg. 0.36 0.79 0.33 1.97 0.57 0.29 0.16
X/Y avg. 0.36 0.79 0.33 1.97 0.57 0.14 0.08
Note: Column 1 from Table 4. Column 2 estimates from Table 5. Column 3: eXRULC is calculated as ePxRULC
(column 1) multiplied by eXPx (column 2). Column 5: eX and eM are the wage share (semi-)elasticities of
exports and imports, respectively, obtained by transforming column 3 through the reciprocal value of real unit
labor costs (column 4). Column 7: Partial effect of a rise of the wage share on exports and imports.
This elasticity has to be transformed into marginal effects. As export and import shares
display a clear trend due to globalization, the transformation of elasticities into marginal
effects yields different results according to the point in time where these effects are evaluated.
We therefore report values calculated at the beginning (1960), at the mean and at the end
22
(2005) of the sample. An increase in the wage share of 1%-point thus leads to a decrease in
exports of 0.11% (2005), 0.06% (mean) and 0.03% (1960) of GDP. The corresponding values
for imports are 0.28%, 0.16% and 0.08% of GDP. The effect of an increase in the wage share
of 1%-point on net exports is therefore -0.39% (2005), -0.22% (mean) and 0.11% (1960) of
GDP. The difference between the two values at the beginning and the end of the sample
reflects the increasing importance of international trade and globalization. The sum of the
export and import effect gives the total effect of an increase in the wage share by 1%-point on
net exports (export effect minus import effect). The results for net exports are displayed in
table 7 in the following section.
Hein and Vogel (2007) directly estimate net exports explained by the profit share,
domestic GDP and GDP of the Euro12-area. They get a net exports effect of a rise in the wage
share by 1%-point of -0.34% of GDP which lies between our results at the mean (-0.22) and
the end (-0.39) of the sample.
4.5 Total effects
Table 7 puts together the partial effects presented above. Adding these up gives the private
excess demand resulting from a one percentage point increase in the wage share. The results
of the net exports effect calculated at different points in time are reported in different
columns. For consistency, consumption and investment effects are also calculated with the
values at the end, the mean and the beginning of the sample. The positive effect of an increase
in the wage share on consumption is reduced from 0.48% of GDP in 1960 to 0.38% in 2005.
This is due to both a fall in the consumption share and the wage share.14 The investment
effect is -0.11% (1960) and -0.15% of GDP in 2005, however it is not very robust. The
positive consumption effect is substantially larger than the negative investment effect. Thus,
14 The average effect is smaller than for 2005 because the wage share rose in the 1970s and fell afterwards.
23
the domestic sector of the economy is clearly wage-led. The results suggest that an increase of
1%-point of the wage share leads to an increase of domestic demand of 0.37% of GDP in
1960 and 0.21% of GDP in 2005. At the mean of the sample, the effect on domestic (private
excess) demand is 0.16% of GDP.
Table 7: Total effects on private excess demand
shares
2005 shares
mean
shares
1960
Consumption 0.36
0.31 0.48
Investment -0.15
-0.15 -0.11
Domestic effect 0.21 0.16 0.37
Net exports -0.39 -0.22 -0.11
Total effect -0.18 -0.06 0.26
The effect of an increase of the wage share on net exports is 0.11% of GDP in 1960, which
is much lower than the effect on domestic demand. At the beginning of the sample thus the
overall effect is clearly positive (0.26%) and the demand regime is wage-led. In 2005
however, the net exports effect is higher than the domestic effect. The total private excess
demand effect is -0.18% of GDP and the demand regime turns profit-led.
To get the total effects of a shift in income distribution on equilibrium demand, the effects
on excess demand have to be adjusted by the multiplier (see equation 3). The focus of our
analysis has been on the effects of shifts in functional income distribution. Consequently the
econometric specifications have been chosen such that the distributional effects are plausible.
The income elasticities have not received similar attention. In particular the income elasticity
of imports is rather high and probably picks up effects of an (exogenous) increase in the
international division of labor. As these elasticities critically enter the calculation of the
multiplier, the results have to be interpreted with great caution. The presentation of the
24
multiplier here thus is for completeness rather than for realism. The multiplier and the
corresponding results for total output are presented in table 8.
Table 8: Total effects on (equilibrium) output
Shares
2005 Shares
mean
Shares
1960
Demand effect -0.18 -0.06 0.26
Multiplier 0.83
1.25 1.87
Total effect -0.15 -0.09 0.50
The multiplier consists of partial effects of changes in income on consumption, investment
and imports. Tables 1, 2 and 4 contain the corresponding coefficient estimates. Both the
income elasticity of consumption and investment are close to one. The income elasticity of
imports is 2, which probably also picks up the effects of an (exogenous) increase in the
international division of labor. Again, the coefficients represent elasticities and have to be
converted into partial effects. The points in time at which the elasticities are converted
therefore make a big difference. At the beginning of the sample we get a multiplier of 1.87.
The positive demand effect therefore is enlarged up to 0.50. The total effect of a rise in the
wage share of 1%-point on output is 0.50% of GDP. At the mean of the sample, the slightly
negative demand effect is multiplied by 1.25. The resulting total effect is -0.09. In the year
2005, due to the increasing import share, the multiplier is below one. Thus, the (negative)
demand effect is slightly diminished. The resulting total effect is -0.15% of GDP. However,
these values have to be considered as less reliable than the private excess demands calculated
in Table 7.
Summing up, the demand regime in Austria is clearly wage-led in the 1960s and turns into
profit-led in 2005 with a private excess demand effect of around -0.18.
25
5 Conclusion
The paper has investigated how changes in the wage share affect aggregate demand in
Austria. Behavioural equations for consumption, investment and international trade, which
included a term for functional income distribution, but are otherwise comparable with
standard macroeconometric models for Austria have been estimated. Three key findings have
emerged. First, domestic component of the demand regime is wage led. The effect of a change
in wage share on consumption is much greater than its effect of investment. Second, including
international trade, the demand regime turns profit-led. This is mostly due to trade, which in a
small open economy plays an important role. Thirdly, the role of the international demand
component has increased over time, reflecting increasing international division of labor and
globalization.
As the overall demand regime in Austria is profit led, it is tempting to conclude that the
fall in the wage share has positively contributed to growth. However, the implications of our
analysis are in fact more complex. A decline in the wage share will increase demand ceteris
paribus, that is with world market prices being constant. It will only increase demand as far as
it leads to an increase in competitiveness. In our context the crucial questions is how unit
labor costs, and ultimately wage shares in the main trading partners develop. Indeed wage
shares have sharply fallen all over Europe. Austria’s wage moderation thus has not fully
translated in an increase in competitiveness. Figure 2 shows the development of the wage
share and a price measure of competitiveness.15 While there clearly is a correlation of the
wage share and competitiveness, the latter also crucially depends on wage (and productivity)
developments abroad and on nominal exchange rates.
15 Nothing hinges on the particular measure of competitiveness that has been taken from the AMECO database
for convenience. Other measures would broadly show a similar pattern.
Figure 2: Wage share and competitiveness in Austria
Wage share and competitiveness in Austria
0.7
0.75
0.8
0.85
0.9
0.95
1
1.05
1.1
1960
1963
1966
1969
1972
1975
1978
1981
1984
1987
1990
1993
1996
1999
2002
2005
wage share
competitiveness
Note. The wage share is the adjusted wage share at factor costs; competitiveness is the nominal unit labor costs,
performance relative to the EU15 with double export weights. Both are standarized 1979 = 100. Source:
AMECO database
As exchange rates have been irrevocably fixed in the Euro system, wage moderation has
become a more effective means of increasing competitiveness. However, our analysis
highlights that wage moderation is a potentially dangerous policy instrument. While any small
country can stimulate its economy by means of wage moderation, this will only work if its
trade partners do not follow a similar strategy. In the period under investigation, this latter has
been the case. Thus Austria’s wage moderation thus has not fully translated in an increase in
competitiveness.
EU member states predominantly trade among each other. The EU as a whole is a
relatively closed economy. A (real) wage moderation at the EU level would therefore have
only a small expansionary effects via net exports, but the full negative effects of its wage-led
domestic sector. Indeed Stockhammer, Onaran and Ederer (2007) find that a 1%-point
26
27
decrease in the wage share has negative demand effect of around 0.2%-points of GDP in the
Euro area.
It therefore seems that wage policy in the Euro area can be characterized as a prisoners’
dilemma situation. While many member states will individually exhibit profit-led demand
regimes, collectively they form a wage-led demand regime. While for each country it may be
expansionary to exercise wage moderation (assuming constant wages abroad), wage
moderation in all countries will have a contractionary effect. This is likely to generate a
downward bias in wage settlements if wages are negotiated nationally. A coordination of
wage bargaining across the Euro area (or the EU in general) therefore seems desirable.
The approach taken in this paper has several limitations that should be addressed in future
research. While these limitations are rather obvious, their fixes are not. Firstly, functional
income distribution has been taken as exogenously given. A full model would endogenize the
wage shares. Secondly, the paper has focused on the private sector. A full model would
include the state sector, which is difficult because it requires assigning various taxes to wage
and profit incomes. Thirdly, the paper has only analyzed the demand side of the economy. A
full model would explicitly model the supply side. While standard production functions fit
uneasily with Post Keynesian models, because the economy will only be at full capacity
utilization by coincidence, there is no reason to assume that productivity growth is exogenous
to wage growth.
28
6 References
Baumgartner, J, Breuss F, Kaniovski, S, 2005. WIFO-Macromod – An econometric model of the
Austrian economy. In: OeNB (ed): Macroeconomic Models and Forcasts for Austria.
Proceedings of OeNB Workshops No. 5
Bhaduri, Amit, Marglin, Stephen, 1990. Unemployment and the Real Wage: The Economic Basis for
Contesting Political Ideologies. Cambridge Journal of Economics, 14: 375-93
Blanchard, Olivier (2006). Macroeconomics, 4th Edition. Upper Saddle River, NJ: Prentice Hall
Blecker, R., 1989. International Competition, Income Distribution and Economic Growth. CJE 13: 395-
412
Blecker, R., 1999. Kaleckian macromodels for open economies, in Deprez, J. and Harvey, J.T (eds):
Foundations of international economics: post Keynesian perspectives. Routledge, London,
New York.
Blecker, R., 1999. Kaleckian macromodels for open economies, in Deprez, J. and Harvey, J.T (eds):
Foundations of international economics: post Keynesian perspectives. Routledge, London,
New York.
Blecker, Robert, 2002. Distribution, demand and growth in neo-Kaleckian macro-models. in:
Setterfield (ed): The economics of demand-led growth. Cheltenham, UK: Edward Elgar
Bowles, S, Boyer, R, 1995. Wages, aggregate demand, and employment in an open economy: an
empirical investigation. In: G Epstein and H Gintis (eds): Macroeconomic policy after the
conservative era. Studies in investment, saving and finance. Cambridge: University Press
Breuss, Fritz, 2007. Globalization, EU enlargement and income distribution. Wifo working paper
296/2007
Charemza, W, Deadman, D, 1997. New Directions in Econometric Practice. Edward Elgar,
Cheltenham.
Ederer, S, Stockhammer, E, 2007. Wages and aggregate demand in France: An empirical investigation.
Hein, E, Truger, A (eds): Money, distribution, and economic policy – alternatives to orthodox
macroeconomics. Cheltenham: Edward Elgar
Fazzari, S, Mott, T, 1986. The investment theories of Kalecki and Keynes: an empirical study of firm
data, 1970-1982. Journal of Post Keynesian Economics 9, 2: 171-187
Fenz, G, Spitzer, M. 2005. AQM The Austria Quarterly Model of the Oesterreichische Nationalbank.
In: OeNB (ed): Macroeconomic Models and Forcasts for Austria. Proceedings of OeNB
Workshops No. 5
Flassbeck, H, Spiecker, F, 2005. Die deutsche Lohnpolitik sprengt die Europäische Währungsunion.
WSI Mitteilungen 12/2005, 707-13
Gordon, David, 1995. Growth distribution, and the rules of the game: social structuralist macro
foundations for a democratic economic policy. In: G Epstein and H Gintis (eds):
29
Macroeconomic policy after the conservative era. Studies in investment, saving and finance.
Cambridge: University Press
Gordon, David, 1995. Putting the horse (back) before the cart: disentangeling the macro relationship
between investment and saving. In: G Epstein and H Gintis (eds): Macroeconomic policy after
the conservative era. Studies in investment, saving and finance. Cambridge: University Press
Hatzius, Jan, 2000. Foreign direct investment and factor demand elasticities. European Economic
Review 44 (1), 77-143
Hein, E, and Ochsen, C., 2003. Regimes of interest rates, income shares, savings, and investment: a
Kaleckian model and empirical estimations for some advanced OECD-economies, in:
Metroeconomica, Vol. 54, 404-433
Hein, E, Krämer, H, 1997. Income shares and capital formation: patterns of recent developments.
Journal of Income Distribution 7, 1: 5-28
Hein, E, Vogel, L, 2007. Distribution and growth reconsidered – empirical results for Austria, France,
Germany, the Netherlands, the UK and the USA. IMK working paper 3/2007
Hubbard, R, 1998. Capital market imperfections and investment. Journal of Economic Literature 36:
193-225
IMF, 2007. The globalization of labor. Chapter 5 of World Economics Outlook April 2007. Washington:
IMF
Kaldor, Nicholas, 1956. Alternative Theories of Distribution. Review of Economic Studies 23, 2: 83-100
Kaldor, Nicholas, 1957. A model of economic growth. Economic Journal LXVII: 591-624
Kalecki, Michal, 1954. Theory of Economic Dynamics. Reprinted in J. Osiatynski (ed): Collected Works
of Michal Kalecki, Vol. 1, Oxford: Clarendon Press
Kalecki, Michal, 1954. Theory of Economic Dynamics. Reprinted in J. Osiatynski (ed): Collected Works
of Michal Kalecki, Vol. 1, Oxford: Clarendon Press
Kalecki, Michal, 1971. Selected Essays on the Dynamics of the Capitalist Economy. Cambridge
Mankiw, Nicholas Gregory (2006). Macroeconomics. New York, 6th edition. NY: Worth
Marginson , P, Sisson, K (2004). European integration and industrial relations. Multi-level governance
in the making. Houndsmill: Palgrave MacMillan
Marglin, S, Bhaduri, A, 1990. Profit Squeeze and Keynesian Theory. In: S. Marglin and J. Schor (eds):
The Golden Age of Capitalism. Reinterpreting the Postwar Experience. Oxford: Clarendon
Marterbauer, M; Kaniovski, S; Kratena, K; Wüger, M (2006): WIFO-Weißbuch: Mehr Beschäftigung
durch Wachstum auf Basis von Innovation und Qualifikation. Teilstudie 11: Maßnahmen zur
Belebung der privaten Inlandsnachfrage.
Marterbauer, M; Walterskirchen, E., (2003). Bestimmungsgründe der Lohnquote und realen
Lohnstückkosten. Wifo-Monatsberichte 2/03, 151-59
Marx, Karl, 1976. Capital. A Critique of Political Economy Volume One. Penguin Books
Naastepad, R, Storm, S (2006/7), OECD demand regimes (1960-2000), Journal of Post-Keynesian
Economics 29 (2), 213-248
30
Naastepad, Ro (2006). Technology, demand and distribution: a cumulative growth model with an
application to the Dutch productivity slowdown. Cambridge Journal of Economics 30 (3) 403-
434
OECD, 2007 OECD workers in the global economy: increasingly vulnerable? Chapter 3 of OECD
Employment Outlook 2007
Palley, Thomas, 2006. Currency Unions, the Phillips Curve, and Stabilization Policy: Some Suggestions
for Europe. Intervention 3 (2),351-370
Ricardo, David, 1951. On the principles of political economy and taxation. Edited by Piero Sraffa.
Cambridge University Press
Robinson, Joan, 1956. The accumulation of capital. London: Macmillan
Robinson, Joan, 1962. Essays in the theory of economic growth. London: Macmillian
Rodrik, Dani, 1997. Has globalization gone too far? Washington: Institute for International Economics
Romer, David, 2006. Advanced macroeconomics. 3rd edition. Boston: McGraw-Hill
Rowthorn, R, 1981. Demand, real wages and economic growth. Thames Papers in Political Economy.
Autumn 1-39, reprinted in Studi Economici 1982, vol. 18 3-54 and also in Sawyer (ed): Post-
Keynesian Economics
Schulten, Thorsten (2004). Solidarische Lohnpolitik in Europa. Zur Politischen Ökonomie der
Gewerkschaften. Hamburg: VSA-Verlag
Stewart, Kenneth (2005): Introduction to Applied Econometrics. Belmont: Brooks/Cole Thomson
Learning.
Stiglitz, J, Weiss, A, 1981. Credit rationing in markets with imperfect competition. American Economic
Review 71: 393-410
Stockhammer, E, Onaran, Ö, Ederer, S. 2007. Functional income distribution and aggregate demand in
the Euro area. Vienna University of Economics & B.A. Working Paper No. 102, Feb 2007
Stockhammer, E. and O. Onaran (2004), 'Accumulation, Distribution and Employment: A Structural
VAR Approach to a Kaleckian Macro-model' Structural Change and Economic Dynamics 15
(4), 421-47
Visser, Jelle, 2004. Patterns and variations in European industrial relations. Chapter 1 in: European
Commission: Industrial relations in Europe 2004. Luxembourg: European Communities
31
Appendix
Table A.1: Variable definitions
Notation OECD Notation Description
Ω - Wage share
C CPV Private consumption, real
I IPV Private investment, real
i IRLR Long-term real interest rate, deflated by GDP deflator
M MGSV Imports, real
P PGDP GDP deflator
PMPMGS Import price deflator
PXPXGS Export price deflator
PXC PXC Competitors prices on export market
R - Gross operating surplus, real, deflated by GDP deflator
RULC - Real unit labor costs
ULC ULC Nominal unit labor costs
W - Compensation of employees real, deflated by GDP deflator
X XGSV Exports, real
Y GDPV GDP, real
Y_EU GDPV GDP, Euro 12-countries, real
Table A.2: Unit root tests
Levels First Differences
Variable
Constant/
Trend Lags
Test
stat. Signifikance
Constant/
Trend Lags
Test
stat. Signifikance
ln C c, t 0 -1.302 - c 0 -4.774 ***
ln W c, t 1 -1.090 - c 0 -2.501 **
ln R c, t 0 -3.328 ** c 1 -7.138 ***
ln I c, t 0 -2.612 - c 0 -6.271 ***
ln Y c, t 0 -1.643 - c 0 -5.086 ***
ln Ω c 0 -2.239 ** c 0 -6.253 ***
i c 0 -2.395 ** c 0 -7.393 ***
ln X c, t 0 -1.577 - c 0 -4.998 ***
ln Y_EU12 c, t 1 -2.350 - c 0 -3.866 ***
ln Px/Pxc c 0 -1.621 * c 1 -4.426 ***
ln M c, t 0 -1.731 - c 0 -6.136 ***
ln P/Pm c 0 -1.360 - c 1 -5.916 ***
ln ULC c, t 0 0.980 - c 0 -3.225 ***
ln Px c, t 1 -0.649 - c 0 -3.430 ***
ln P c, t 1 -1.006 - c 0 -1.858 *
ln Pm c, t 0 -0.169 - c 0 -4.181 ***
Notes: * denotes statistical significance at the 10% level, ** at the 5% level, *** at the 1% level. Critical values
according to Charemza and Deadman (1997).
32
Table A.3: Cointegration tests
Consumption function
Dep. variable Expl. variable Test Stat. Signifikance
C W, R -3.333 -
C Y, Ω -2.962 -
Investment function
Dep. variable Expl. variable Test Stat. Signifikance
I Y, R -3.128 -
Notes: * denotes statistical significance at the 10% level, ** at
the 5% level, *** at the 1% level. Critical values according to
Charemza and Deadman (1997).
Bisher sind in dieser Reihe erschienen:
Eigl R., Experimentielle Methoden in der Mikroökonomik, No. 1, Mai 1991.
Dockner E., Long N.V., International Pollution Control: Cooperative versus Non-Cooperative Strategies, No.
2, September 1991.
Andraea C.A., Eigl R., Der öffentliche Sektor aus ordnungspolitischer Sicht, No. 3, Oktober 1991.
Dockner E., A Dynamic Theory of Conjectural Variations, No. 4, Oktober 1991.
Feichtinger G., Dockner E., Cyclical Consumption Pattern and Rational Addictions, No. 5, Oktober 1991.
Marterbauer M., Die Rolle der Fiskalpolitik im Schwedischen Wohlfahrtsstaat, No. 6, Dezember 1991.
Pichler E., Cost-Sharing of General and Specific Training with Depreciation of Human Capital, No. 7,
Dezember 1991.
Pichler E., Union Wage Bargaining and Status, No. 8, Dezember 1991.
Pichler E., Costs of Negotiations and the Structure of Bargaining - a Note, No. 9, Dezember 1991.
Nowotny E., The Austrian Social Partnership and Democracy, No. 10, Dezember 1991.
Pichler E., Walther H., The Economics of Sabbath, No. 11, April 1992.
Klatzer E., Unger B., Will Internationalization Lead to a Convergence of National Economic Policies?,No. 12,
June 1992.
Bellak C., Towards a Flexible Concept of Competitiveness, No. 13, May 1992.
Koren St., Stiassny A., The Temporal Causality between Government Taxes and Spending, No. 14, August
1992.
Altzinger W., Ost-West-Migration ohne Steuerungsmöglichkeiten?, No. 15, September 1992.
Bellack Ch., Outsiders' Response to Europe 1992, Case of Austria, No. 16, December 1992.
Guger A., Marterbauer M., Europäische Währungsunion und Konsequenzen für die Kollektiv-vertragspolitik,
No. 17, January 1993.
Unger B., van Waarden F., Characteristics, Governance, Performance and Future Perspectives, No. 18,
January 1993.
Scharmer F., The Validity Issue in Applied General Equilibrium Tax Models, No. 19, May 1993.
Ragacs Ch., Minimum Wages in Austria: Estimation of Employment Functions, No. 20, June 1993.
Ragacs Ch., Employment, Productivity, Output and Minimum Wages in Austria: A Time Series Analysis, No.
21, September 1993.
Stiassny A., TVP - Ein Programm zur Schätzung von Modellen mit zeitvariierenden Parametern, No. 22,
December 1993.
Gstach D., Scale Efficiency: Where Data Envelopment Analysis Outperforms Stochastic Production Function
Estimation, No. 23, December 1993.
Gstach D., Comparing Structural Efficiency of Unbalanced Subsamples: A Resampeling Adaptation of Data
Envelopment Analysis, No. 24, December 1993.
Klausinger H., Die Klassische Ökonomie und die Keynesianische Alternative. Revision ein Mythos?, No. 25,
December 1993.
Grandner T., Gewerkschaften in einem Cournot-Duopol. Sequentielle versus simultane Lohnverhandlungen,
No. 26, April 1994.
Stiasssny A., A Note on Frequency Domain Properties of Estimated VARs, No. 27, June 1994.
Koren St., Stiassny A., Tax and Spend or Spend and Tax ? An International Study, No. 28, August 1994.
Gstach D., Data Envelopment Analysis in a Stochastic Setting: The right answer form the wrong model?, No.
29, August 1994.
Cantwell J., Bellak Ch., Measuring the Importance of International Production: The Re-Estimation of Foreign
Direct Investment at Current Values, No. 30, January 1995.
Klausinger H., Pigou’s Macroeconomics of Unemployment (1933). A Simple Model, No. 31, February 1995.
Häfke Ch., Helmenstein Ch., Neural Networks in Capital Markets: An Application to Index Forecasting, No.
32, January 1995.
Hamberger K., Katzmair H., Arithmetische Politik und ökonomische Moral, Zur Genologie der
Sozialwissenschaften in England, No. 33, May 1995.
Altzinger W., Beschäftigungseffekte des österreichischen Osthandels, No. 34, July 1995.
Bellak Ch., Austrian Manufacturing Firms Abroad - The last 100 Years, No. 35, November 1995.
Stiassny A., Wage Setting, Unemployment and the Phillips Curve, No. 36, January 1996.
Zagler M., Long-Run Monetary Non-Neutrality in a Model of Endogenous Growth, No. 37, June 1996.
Traxler F., Bohmann G., Ragacs C., Schreckeneder B., Labour Market Regulation in Austria, No. 38,
January, 1996.
Gstach D., A new approach to stochastic frontier estimation: DEA+, No. 39, August 1996.
Bellak Ch., Clement W., Hofer R., Wettbewerbs- und Strukturpolitik: Theoretische Begründung und neuere
Entwicklungen in Österreich, No. 40, June 1996.
Nowotny E., Dritter Sektor, Öffentliche Hand und Gemeinwirtschaft, No. 41, August 1996.
Grandner T., Is Wage-Leadership an Instrument to Coordinate Union’s Wage-Policy? The Case of Imperfect
Product Markets, No. 42, November 1996.
Pirker R., The Constitution of Working Time, No. 43, Januar 1997.
Nowotny E., Konsequenzen einer Globalisierung der Weltwirtschaft für unsere Gesellschaft, No. 44, Januar
1997.
Grandner T., Territoriale Evolution von Kooperation in einem Gefangenendilemma, No. 45, February 1997.
Häfke Ch., Sögner L., Asset Pricing under Asymmetric Information, No. 46, February 1997.
Stiassny A., Die Relevanz von Effizienzlöhnen im Rahmen von Gewerkschaftsverhandlungsmodellen, No.
47, May 1997.
Stiassny A., Unsicherheit bezüglich der Preiselastizität der Güternachfrage als reale Rigidität, No. 48, May
1997.
Klausinger H., Die Alternativen zur Deflationspolitik Brünings im Lichte zeitgenössischer Kritik, No. 49, June
1997.
Wehinger G.D., Exchange Rate-Based Stabilization: Pleasant Monetary Dynamics?. No. 50, August 1997.
Wehninger G.D., Are Exchange Rate-Based Stabilizations Expansionary? Theoretical Considerations and
the Brazilian Case, No. 51, August 1997.
Huber C., Sögner L., Stern A., Selbstselektierendes Strompreisregulierungsmodell, No. 52, August 1997.
Ragacs Ch., Zagler M., Economic Policy in a Model of Endogenous Growth, No. 53, October 1997.
Mahlberg B., Url T., Effects of the Single Market on the Austrian Insurance Industry, No. 54, February 1998.
Gstach D., Grander T., Restricted Immigration In as Two-Sector Economy, No. 55, March 1998.
Sögner L., Regulation of a Complementary Imputed Good in a Competitive Environment, No. 56, March
1998.
Altzinger W., Austria's Foreign Direct Investment in Central and Eastern Europe: 'Supply Based' or Marked
Driven?, No. 57, April 1998.
Gstach D., Small Sample Performance of Two Approaches to Technical Efficiency Estimation in Noisy
Multiple Output Environments, No. 58, June 1998.
Gstach D., Technical Efficiency in Noisy Multi-Output Settings, No. 59, June 1998.
Ragacs Ch., Zagler M., Growth Theories and the Persistence of Output Fluctuations: The Case of Austria,
No. 60, October 1998.
Grandner T., Market Shares of Price Setting Firms and Trade Unions, No. 61, October 1998.
Bellak Ch., Explaining Foreign Ownership by Comparative and Competitive Advantage: Empirical Evidence,
No. 62, March 1999.
Klausinger H., The Stability of Full Employment. A Reconstruction of Chapter 19-Keynesianism, No. 63, April
1999.
Katzmair H., Der Modellbegriff in den Sozialwissenschaften. Zum Programm einer kritischen Sozio-Logik,
No. 64, June 1999.
Rumler F., Computable General Equilibrium Modeling, Numerical Simulations in a 2-Country Monetary
General Equilibrium Model, No. 65, June 1999.
Zagler M., Endogenous Growth, Efficiency Wages and Persistent Unemployment, No. 66, September 1999.
Stockhammer E., Robinsonian and Kaleckian Growth. An Update on Post-Keynesian Growth Theories, No.
67, October 1999.
Stockhammer E., Explaining European Unemployment: Testing the NAIRU Theory and a Keynesian
Approach, No. 68, February 2000.
Klausinger H., Walras’s Law and the IS-LM Model. A Tale of Progress and Regress, No. 69, May 2000.
Grandner T., A Note on Unionized Firms’ Incentive to Integrate Vertically, No. 70, May 2000.
Grandner T., Optimal Contracts for Vertically Connected, Unionized Duopolies, No. 71, July 2000.
Heise, A., Postkeynesianische Beschäftigungstheorie, Einige prinzipielle Überlegungen, No. 72, August
2000.
Heise, A., Theorie optimaler Lohnräume, Zur Lohnpolitik in der Europäischen Währungsunion, No. 73,
August 2000.
Unger B., Zagler M., Institutional and Organizational Determinants of Product Innovations. No. 74, August
2000.
Bellak, Ch., The Investment Development Path of Austria, No. 75, November 2000.
Heise, A., Das Konzept einer nachhaltige Finanzpolitik aus heterodoxer Sicht – ein Diskussionsbeitrag, No.
76, April 2001.
Kocher M., Luptacik M., Sutter M., Measuring Productivity of Research in Economics. A Cross-Country
Study Using DEA, No. 77, August 2001.
Munduch, G., Pfister A., Sögner L., Stiassny A., Estimating Marginal Costs fort he Austrian Railway System,
No. 78, Februray 2002.
Stückler M., Überprüfung von Gültigkeit und Annahmen der Friedman-These für Rohstoffmärkte, No. 79,
July 2002.
Stückler M., Handel auf Terminkontraktmärkten, No. 80, July 2002.
Ragacs Ch., Minimum Wages, Human Capital, Employment and Growth, No. 81, August 2002.
Klausinger H., Walras’ Law in Stochastic Macro Models: The Example of the Optimal Monetary Instrument,
No. 82, November 2002.
Gstach D., A Statistical Framework for Estimating Output-Specific Efficiencies, No. 83, February 2003.
Gstach D., Somers A., Warning S., Output specific efficiencies: The case of UK private secondary schools,
No. 84, February 2003.
Kubin I., The dynamics of wages and employment in a model of monopolistic competition and efficient
bargaining, No. 85. May 2003.
Bellak Ch., The Impact of Enlargement on the Race For FDI. No. 86 Jan. 2004
Bellak Ch., How Domestic and Foreign Firms Differ and Why Does it Matter?. No. 87 Jan. 2004
Grandner T., Gstach D., Joint Adjustment of house prices, stock prices and output towards short run
equilibrium, No. 88. January 2004
Currie M., Kubin I., Fixed Price Dynamics versus Flexible Price Dynamics, No. 89, January 2005
Schönfeld S., Reinstaller A., The effects of gallery and artist reputation on prices in the primary market for
art: A note, No. 90, May 2005
Böheim, R. and Muehlberger, U., Dependent Forms of Self-employment in the UK: Identifying Workers on
the Border between Employment and Self-employment. No. 91, Feb. 2006
Hammerschmidt, A., A strategic investment game with endogenous absorptive capacity. No. 92, April 2006
Onaran, Ö., Speculation-led growth and fragility in Turkey: Does EU make a difference or “can it happen
again”? No. 93, May 2006
Onaran, Ö., Stockhammer, E., The effect of FDI and foreign trade on wages in the Central and Eastern
European Countries in the post-transition era: A sectoral analysis. No. 94, June 2006
Burger, A., Reasons for the U.S. growth period in the nineties: non-keynesian effects, asset wealth and
productivity. No. 95, July 2006
Stockhammer, E., Is the NAIRU theory a Monetarist, New Keynesian, Post Keynesian or a Marxist theory?
No. 96, March 2006
Onaran, Ö., Aydiner-Avsar, N., The controversy over employment policy: Low labor costs and openness, or
demand policy? A sectoral analysis for Turkey. No. 97, August 2006
Klausinger, H., Oskar Morgenstern als wirtschaftspolitischer Berater in den 1930er-Jahren. No. 98, July
2006
Rocha-Akis, S., Labour tax policies and strategic offshoring under unionised oligopoly. No. 99, November
2006
Stockhammer, E., Onaran, Ö., National and sectoral factors in wage formation in Central and Eastern
Europe. No. 100, December 2006
Badinger, H., Kubin, I., Vom kurzfristigen zum mittelfristigen Gleichgewicht in einer offenen Volkswirtschaft
unter fixen und flexiblen Wechselkursen. No. 101, January 2007
Stockhammer, E., Onaran, Ö., Ederer, S., Functional income distribution and aggregate demand in the Euro-
area. No. 102, February 2007
Onaran, Ö., Jobless growth in the Central and Eastern European Countries: A country specific panel data
analysis for the manufacturing industry. No. 103, March 2007
Stockhammer, E., Ramskogler, P. Uncertainty and exploitation in history. No. 104, April 2007
Ramskogler, P., Uncertainty, market power and credit rationing. No. 105, August 2007
Stockhammer, E., Ederer, S. Demand effects of the falling wage share in Austria. No. 106, August 2007
