## Abstract

The aim of this paper is to analyze the effects of interest rates on rates of capacity utilization, capital accumulation and profit in Italy within a post-Kaleckian theoretical framework. The model employed in the analysis, which was developed by Hein/Schoder (2011), is based on monetary- authority-controlled real long-term interest rates that affect the functional distribution of income. Interest rates directly and indirectly affect the equilibrium rates of capacity utilization, accumulation and profit at a given debt–capital ratio. Our findings based on two econometric methods revealed that a higher real long-term interest rate has an adverse effect on these three endogenous variables in the Italian economy.

## 1 INTRODUCTION

The theoretical background of the global conduct of monetary policy, after the replacement of the monetarist hegemony, has emerged as a New Keynesian approach,^{1} or more broadly, the New Consensus Models (hereafter NCMs) (Arestis/Sawyer 2002) in the 1990s. These models put forward that interest rates set by central banks only have short-term effects on output and employment, but they are ineffective in the long term. Despite the fact that this paradigm dominated the monetary policy sphere, post-Keynesian (hereafter PK) theorists have a different take on the role of interest rates in the economy. In a nutshell, PK authors criticized these models for the assumption of long-run neutrality of money and monetary policies (Hein/Schoder 2011). Within this framework, adopting the post-Kaleckian model developed by Hein/Schoder (hereafter HS) (2011), the aim of this article is to examine the long-run effects of interest rates on the equilibrium rates of capacity utilization, capital accumulation and profits in the Italian economy.^{2} The signs of these effects characterize different monetary regimes of a given economy.^{3}

Italy underwent an important institutional and structural transformation within the last three decades. Among the important elements of this transformation were several instances of privatization with a total bill of almost 120 billion euros within the two decades starting from the early 1990s, the liberalization of capital movements in the mid-1990s (Gabbi et al. 2016: 234–235), the financialization of the Italian state over the 1993–1999 period (Lagna 2016), neoliberal reforms in the labor market and welfare system. These developments resulted in sharp increases in income and wealth inequalities and along with the austerity measures applied led to suppression of aggregate demand in the Italian economy.^{4}

Italy temporarily withdrew from the European Monetary System in 1992 and rejoined it in 1996. In the second half of the 1990s, decreasing interest rates hand in hand with restrictive fiscal policies were helpful for keeping up with the fiscal criteria of the Maastricht agreement in order to join the euro area. The public debt/GDP ratio fell from 125 per cent in the mid-1990s to 100 per cent in 2007 (Cesaratto 2018). However, this set of policies came with the cost of stagnation of aggregate demand and productivity growth (Cesaratto/Zezza 2019). Meanwhile the debt of non-financial firms in Italy increased and reached its peak with 44 per cent in 2006 just before the 2007–2009 crisis and remained above 40 per cent until 2008 and started to decline afterwards (see Figure 1). This period was very restrictive and played a role in the emergence of a stagnating domestic demand-led regime (Hein/Martschin 2021). In this period, the long-term real interest rate remained significantly higher than the GDP growth rate (see Figure 2 for the interest rates). Following the 2007–2009 crisis, both the differential between the two and the long-term real interest rate itself increased further despite the fact that short-term real interest rates turned negative during the 2010–2019 period. Within this period, which was characterized by a stagnating *export-led mercantilist regime*, fiscal and monetary policies became less restrictive from 2014 onward and the monetary policy of the ECB became expansionary (Hein/Martschin 2021).^{5} Cesaratto/Zezza (2019), based on Bugamelli et al. (2017), confirm that the recovery that started to take place around this period was mostly due to exports. Although monetary policy was not the sole factor that affected the economic developments in Italy in the last three decades, its role was as important as those of fiscal and wage policies.

### Short and long-run interest rates in Italy

Citation: European Journal of Economics and Economic Policies Intervention 2024; 10.4337/ejeep.2024.0126

*Notes:*

*ISN*,

*ISRV*,

*ILN*and

*ILRV*stand for annual nominal short-term interest rates, annual real short-term interest rates deflated by GDP deflator, annual nominal long-term interest rates and annual real long-term interest rates deflated by GDP deflator, respectively. See the Appendix for detailed information.

*Source:*AMECO.

### Short and long-run interest rates in Italy

Citation: European Journal of Economics and Economic Policies Intervention 2024; 10.4337/ejeep.2024.0126

*Notes:*

*ISN*,

*ISRV*,

*ILN*and

*ILRV*stand for annual nominal short-term interest rates, annual real short-term interest rates deflated by GDP deflator, annual nominal long-term interest rates and annual real long-term interest rates deflated by GDP deflator, respectively. See the Appendix for detailed information.

*Source:*AMECO.

### Short and long-run interest rates in Italy

Citation: European Journal of Economics and Economic Policies Intervention 2024; 10.4337/ejeep.2024.0126

*Notes:*

*ISN*,

*ISRV*,

*ILN*and

*ILRV*stand for annual nominal short-term interest rates, annual real short-term interest rates deflated by GDP deflator, annual nominal long-term interest rates and annual real long-term interest rates deflated by GDP deflator, respectively. See the Appendix for detailed information.

*Source:*AMECO.

In this study, adopting the theoretical and empirical model by HS (2011), we examine the Italian economy for the period from 1995 to 2022. Econometric research based on PK distribution and growth models augmented with monetary variables is limited in number and, to the best of our knowledge, there is no such empirical study on Italy. The empirical findings of this study are comparable to those of HS (2011) on the US and German economies.

The remaining part of this article is organized as follows. Section 2 presents the theoretical background of the post-Kaleckian model by HS (2011), the model itself and some related theoretical frameworks. Section 3 surveys the empirical studies related to the field of research to date. Section 4 presents the data, discusses the economic trends in the Italian economy, explains the econometric approach employed, portrays the econometric findings and synthesizes them. Section 5 concludes the paper.

## 2 THEORETICAL BACKGROUND AND THE MODEL

We start this section with a discussion on the NCMs and then present the theoretical background leading to the emergence of the HS (2011) model. We then proceed with the assumptions and equations of the latter and explain the monetary regimes that might arise within this theoretical framework. Meanwhile we consider some related literature and conclude this section with a discussion of the behavior of the long-term capacity utilization rate.

PK models on money and monetary policy differ significantly from NCMs in several respects. Post-Keynesian authors reject the ‘Wicksellian theory of loanable funds’, in which the rate of interest is treated as the variable that equates loanable funds and real investment. This is the *natural* rate of interest, above which the inflation rate decreases. Post-Keynesians assert that either a natural rate does not exist or has multiple values. Another important difference is that tight monetary policies negatively influence output not only in the short but also in the long term. However, New Keynesians and the NCMs mainly do not accept these views (Lavoie 2006: 55–57). Despite these differences, both PK models and NCMs treat money supply as an endogenous variable; however, the latter do not recognize the role of the endogenous credit generation process as a pre-condition for investments and increase in expenditures (Sawyer 2008). NCMs are characterized by the following three equations: first, an aggregate demand function based on households’ and firms’ optimization behavior, in which there is a negative relation between the real interest rate and the output gap. Second, a Phillips curve augmented by expectations, in which the rate of inflation is a function of the output gap in the short run. Third, a central bank that sets the nominal interest rate as a function of the real equilibrium interest rate, the output gap and the deviation of the realized rate of inflation from its target (HS 2011). HS (2011) reject the NCM framework and propose a post-Kaleckian alternative.

The integration of PK monetary theory into the distribution and growth models in this tradition was put on the agenda in the late 1980s/early 1990s, with the exception of an earlier study by Pasinetti (1974).^{6} The research agenda on the ‘monetary theory of production’ became more active starting from the 1980s. While Keynes had advocated a monetary theory of production, the followers of his tradition developed the monetary theory and the models of distribution and growth in separate channels until the 1980s (Hein 2008: 37). In order to combine these two lines of research, Hein (1999, 2006, 2007) developed a series of models drawing on the contributions to the PK monetary theory of different generations of economists such as Kaldor (1970, 1982, 1985), Moore (1988, 1989), Lavoie (1984, 1992, 1995) and Dutt (1989, 1995). The models proposed by Hein are based on different PK/post-Kaleckian growth-distribution models developed by Rowthorn (1981), Dutt (1984, 1987), Amadeo (1986a, 1986b) and Bhaduri/Marglin (1990) (Hein 2014: 337). Among those models, Hein (1999) only focused on the short run, when the debt–capital ratio is constant. Another related theoretical and empirical work in the same vein, by Hein/Ochsen (2003), however, did not take into account the role of the debt–capital ratio. The other two models by Hein (2006, 2007) have similar features except the specification of the accumulation function. These two models endogenized the debt–capital ratio and extended the analysis to the long run. The model by HS (2011) was adapted from these models for econometric investigation. Hein (2014: ch. 9) made a thorough analysis of these models by drawing from all his previous work including Hein (2012b). Nishi (2013), in a different but comparable theoretical framework with exogenous income distribution and an accumulation function without normal rate of capacity utilization, analyzed the short-run, which is characterized by a constant capacity utilization rate defined as output-to-capital ratio, relation between growth regimes and debt-growth regimes (debt-burdened or debt-led) for a closed economy.

The HS (2011) model is based on a closed economy without government expenditures or taxes. Private sector produces a single product which can be used both for consumption by households and investment by firms. Productivity is constant, as are the coefficients of production. The identities and the behavioral equations of the model are as follows, respectively:

The nominal capital stock, *pK,* is equal to the sum of credits received by firms from rentier households, *B*, accumulated retained earnings of firms, *Z*, and dividends paid to rentier households, *D*. The latter two are defined in equations (5) and (6), respectively, where *i* is the (real) interest rate and *d* the dividend rate.^{7} Profit rate, *r*, is reformulated in terms of the product of the output–capital ratio, *u*, and the profit (capital income including interests and dividends) share *h* in equation 7 (HS 2011). The output–capital ratio ^{8} where ^{9}

Equation (8) endogenizes profit share *h* as a function of the rate of interest payments made by firms to rentier households (*Z/pK*), which is theoretically equal to

In equation (9), the capital accumulation rate *u* from its ‘normal’ rate *h.* The accumulation is a negative function of interest payments and dividends received by rentier households. The employment of the deviation of the capacity utilization rate from its normal rate as a proxy for demand allows the elimination of the variations in the capital–potential output ratio. The latter variable influences the output–capital ratio, which is treated as the capacity utilization rate in the model. The model assumes that the normal rate of capacity utilization is not constant but adjusts towards the realized rate and becomes endogenous. The employment of the two measures of the distributed capital income (or rentier income, up to a certain extent) in the accumulation function is grounded on Kalecki’s (1937) ‘principle of increasing risk’. Distributed profits decrease both internal and external means of finance for real investment. The use of a separate parameter for dividends captures the shareholder power on firms, which may engender extra negative effects on the real investment. If the debt–capital ratio is given, an increase in the interest rate has a negative effect on accumulation. However, this effect might be counterbalanced or even be overcome if share of total profits increases along with interest rates (HS 2011). Hein (2006) employed a different specification of this function in which the accumulation rate depends on retained earnings, not separately on shares of total capital income and rentier income.

In equation (10), the saving rate function is specified. Total savings, as a flow variable measured for a certain time interval – annually in our study – are equal to the sum of undistributed profits of firms, savings from rentiers’ income, and those out of labor income. By assumption, the propensity to save out of rentiers’ income,

Equation (11) expresses the goods market equilibrium condition, which closes the model. For the stability of the equilibrium, the condition, which is also called ‘Keynesian stability condition’, in equation (12) must hold. The latter states that the saving rate is more responsive to changes in the rate of capacity utilization than the accumulation rate (HS 2011). Combining the equations above gives the equilibrium rates of capacity utilization,

The model is based on the PK horizontalist theory of money and interest rates. Interest rates are exogenous to the economy and affected by central bank policies; however, loans and money supply are endogenous and determined by the demand of firms.^{10} Commercial banks provide loans to firms that are deemed to be creditworthy; however, the endogenous credit generation process is not treated in this model.^{11} The monetary authority accommodates the required reserves to banks when loans are transformed into deposits. In this process, the central bank’s (base) interest rate is marked up by commercial banks. This mark-up depends on the degree of competition in the banking sector, banks’ liquidity preferences and risk perceptions. In this framework, the central bank’s role in the determination of commercial banks’ interest rates might not be symmetrical. While increasing the central bank interest rate always leads to a higher commercial bank rate, lowering it might not end up with a decrease in the latter if commercial banks’ liquidity and risk premia increase due to higher uncertainty or profit desires of the banking sector (HS 2011).

The model acknowledges that competition, expectations and liquidity and risk perceptions are subject to change and these factors may constrain the ability of central banks to control interest rates. However, it is also argued that monetary policy does not only affect the short-run interest rates but also the pace of the long-run interest rates.^{12} The monetary policy sets both nominal and real rates of interest since the prices are assumed to be constant in the model. In the real world, central banks take into account inflation or expected inflation when deciding on the nominal interest rate. They change the latter in order to influence the real rate of interest. In the model, it is the rate of interest payments (*Z/pK,* where *Z* = *iB*) that affects investment, saving and profit share. Since this variable is influenced by the price level, assuming that firms’ debts are not indexed to inflation, it is the real rate of interest that influences the equilibrium rates of capacity utilization, capital accumulation and profit (HS 2011).

After replacing *h* with the right side of equation (8) in equations (13) through (15), the effects of a change in the interest rate on the equilibrium rates of capacity utilization, capital accumulation and profit are obtained as below, respectively:

The signs of the derivatives above depend on the values of a set of parameters and variables. If the profit share is insensitive to variations in the rate of interest at a given debt–capital ratio and the stability condition in equation (12) holds, the signs of the derivatives depend on the rentiers’ saving propensity and the responsiveness of investment to interest payments. The effect on the equilibrium rate of capital accumulation also depends on the workers’ saving propensity, the sensitivity of investment to economic activity and the profit share. The size of the debt–capital ratio amplifies or diminishes these effects (HS 2011).

Three different regimes might arise in this model. Under a ‘normal regime’, an increase in the rate of interest leads to a decrease in the equilibrium rates of capacity utilization, capital accumulation and profit. However, under a ‘puzzling regime’ the results are the opposite. An ‘intermediate regime’ emerges in which higher interest rates have a positive effect on the equilibrium rates of capacity utilization and profit, but a negative effect on that of capital accumulation (HS 2011).

The effects considered so far are the ‘primary effects’. If the profit share is elastic to interest payments, indirect effects that emerge through the income distribution channel must also be taken into account. These ‘secondary effects’ might change the signs of the derivatives or their magnitudes. They depend on whether demand, accumulation and profits are ‘wage-led’ or ‘profit-led’. If investment is not very sensitive to the profit share and the workers’ propensity to save is low, an increase in the profit share leads to a decline in capacity utilization, which is the wage-led demand case. Under a wage-led demand regime, if accumulation is very sensitive to capacity utilization, the capital accumulation also becomes wage-led; otherwise, it is profit-led. If the demand is profit-led, the accumulation is necessarily profit-led (HS 2011; Hein 2014).^{13}

Under the normal regime, if both demand and accumulation are wage-led, the negative primary effects of higher interest rates on the rates of capacity utilization, accumulation and profit increase through the secondary effects. However, under the puzzling regime, the positive effects on those might diminish or even turn negative. If both demand and accumulation are profit-led, the positive primary effects of rising interest rates on capacity utilization, accumulation and profit rates increase through the secondary effects under the puzzling regime. However, under the normal regime, these positive effects might dampen or even turn negative. Under the intermediate regime, when the profit share is elastic to interest payments, if demand is wage-led, the positive effects of a rise in interest rates on the rates of capacity utilization and profit will diminish or reverse. However, if demand is profit-led, these effects will be strengthened. If capital accumulation is wage-led, the negative effects of increasing interest rates on capital accumulation will be even higher in absolute value. However, if capital accumulation is profit-led, these effects will diminish or turn positive under this regime (HS 2011).

## 3 REVIEW OF EMPIRICAL LITERATURE

In this section we provide a review of the literature related to our empirical investigation on Italy. We first review a group of empirical studies that analyze the Italian economy from a Keynesian perspective. We proceed with a second group that determines the demand and growth regime in Italy along with other economies. Finally, we review a third group of studies at the cross-section of monetary policy, income distribution and growth, which we classify under three subgroups. We also address some studies that do not fall into any of these groups.

In the first group of studies, which analyzed the Italian economy from a Keynesian perspective, Tridico (2015) focused on the labor market and flexibility in Italy, Bagnai (2016) employed a balance-of-payments constrained growth model for Italy and Goes (2020) examined the relation between income taxes, personal income distribution and growth in Italy.

In the second group of studies, which determine the demand and growth regimes of economies, Italy is analyzed along with other countries.^{14} Naastepad/Storm (2006) analyzed the nexus between real wage growth, productivity growth and aggregate demand, Onaran/Galanis (2014) and Onaran/Obst (2016) examined the effects of changes in functional income distribution on demand within a framework that distinguishes between single-country and global effects, and Obst et al. (2017) focused on the role of fiscal policy.

We classify the third group of studies into three subgroups. Among those, the first subgroup only focused on the relation between monetary policy and income distribution. Among those, Argitis/Pitelis (2001) found that interest rates had small but negative effects on the share of industrial profits in total income in the US and UK. Rochon/Rossi (2006) studied the impact of inflation-targeting policy on income distribution, measured in terms of wage share, in 10 countries that adopted this regime and found that there seemed to be a general decline in this share in these countries. Kappes (2021) provided a review of the empirical literature that analyzed the impact of monetary policy on personal income distribution. This work, as the author mentions, briefly touched upon the PK studies and did not address the ones in the post-Kaleckian tradition such as that of HS (2011). Kappes concluded that the empirical studies reviewed by the author had shown that conventional and, with some exceptions, unconventional expansionary monetary policies had decreased income inequality and contractionary policies had had opposite effects.

The second subgroup of studies within the third group also incorporated growth into empirical analysis. Among those, Argitis (2008) applied a panel regression on 13 OECD countries in order to analyze the impact of inflation-targeting policies on distribution and growth. He found that the increase in the share of interests in total income had negative effects on output and employment. Argitis/Michopoulou (2010) built a PK model for a closed economy without government and applied it to a group of OECD countries in a panel study. The estimation of the model consisted of three equations in which the growth rates of consumption, investment and GDP were regressed on the rentier income, defined as the share of interest income received by the banking sector as percentage of GDP, and the wage share. The authors confirmed the negative impact of the rentier share on economic growth within their theoretical and empirical framework. This second group of studies did not address monetary regimes.

The third subgroup within the third group of studies identified potential monetary regimes that might arise due to variations in interest rates and determined them in countries studied. Among those, Hein/Ochsen (2003) developed a Kaleckian model and applied it to four countries. In this model, capital accumulation depends on the interest rate, capacity utilization rate and profit share. They found that over the full sample, increases in interest rates had negative effects on the equilibrium rates of output, accumulation, and profit in France and Germany. However, the corresponding effects for the UK and US point that the regimes are undetermined in these countries. Their analyses of subperiods gave mixed results. HS (2011) examined Germany and the US over the period 1960–2007. In the next section, we address the findings of this study in comparison to ours on Italy. In addition to these studies, Onaran et al. (2011) employed a post-Kaleckian model that did not identify monetary regimes but integrated the role of rentier income and housing and financial wealth. Among their other results, they found that rentier income had a negative effect on investment but an increase in non-rentier profits had a positive effect on it, while the overall effect of a pro-capital income on investment turned out to be limited.

## 4 DATA, ESTIMATION OF THE MODEL AND FINDINGS

In this section we present the data and trends in the Italian economy, the econometric methodology employed in our analysis and the findings of our estimations. We end this section with the synthesis of our econometric findings for the characterization of the monetary regime in Italy.

### 4.1 Data and trends

In order to estimate the model, we compiled and processed data from AMECO, EUROSTAT and ISTAT. When available, use of quarterly or monthly data in macroeconomic studies significantly increases the number of observations. However, the institutional sector accounts provided by ISTAT, which consists of the data for non-financial firms’ interest and dividend payments and households’ interest and dividend income and savings, are only available annually for the 1995–2022 period. The data for the debt stock of non-financial corporations are available for the 1995–2021 period. However, this did not decrease the number of observations in the econometric analysis since this variable is not employed in the regressions but in the calculations of effects of interest rates on the endogenous variables. The detailed description of the data and their sources are presented in the Appendix. In Figure 3, we present the evolution of the endogenous variables of the model. In order to complement it, Table 1 shows the average values of all the variables over the full sample and the subperiods determined with respect to the business cycles in the Italian economy.

#### Capital income share and rates of profit, capacity utilization and capital accumulation in Italy

*Source:*See the Appendix.

#### Capital income share and rates of profit, capacity utilization and capital accumulation in Italy

*Source:*See the Appendix.

#### Capital income share and rates of profit, capacity utilization and capital accumulation in Italy

*Source:*See the Appendix.

#### Average values of the variables in percentage terms over total period and business cycles

*Notes:* *u*, *h*, *g*, *r*, *Z/pK*, *D/pK*, *ZD/pK*, *W/pK*, *S _{h}/pK*,

*B/pK*,

*ISN*,

*ISRV*,

*ILN*and

*ILRV*denote the capacity utilization rate, the capital income share, the capital accumulation rate, the profit rate, rate of interest payments of non-financial corporations, rate of net dividend payments of non-financial corporations, rate of households’ net interest and dividend income, rate of adjusted labor income, savings rate of households, debt–capital ratio, annual nominal short-term interest rates, annual real short-term interest rates deflated by GDP deflator, annual nominal long-term interest rates and annual real long-term interest rates deflated by GDP deflator, respectively.

*Source:* See the Appendix. ^{a}The average value for the 1995–2021 period. ^{b}The average value for the 2020–2021 period.

The trends in Figure 3 and the average values of the variables over the business cycles in Table 1 point at the following observations. The capital accumulation rate reached its peak in the early 2000s; however, after the 2007–2009 crisis it rapidly declined, and after the onset of the eurozone crisis negative rates were observed until 2017. The partial recovery period before the COVID-19 crisis in 2020 was not long-lived. It seems to have recovered over the 2020–2022 period; however, in real terms the capital stock in Italy in 2022 and that a decade ago in 2012 are virtually the same.

A similar pattern is observed for the growth rate of the GDP. The average annual GDP growth rate gradually declined at each new business cycle and the average value over the sample period was less than 1 per cent. In the last one and a half decades, the Italian economy was severely affected by the Great Recession in 2007–2009, the eurozone crisis and the COVID-19 crisis in 2020. The austerity measures applied by the Prime Minister Mario Monti after the eurozone crisis were no remedy to the Italian economy. The annual average growth rate over the 2013–2019 period was less than 1 per cent per annum. After hitting a bottom in 2020 with a 9 per cent contraction rate, the annual GDP grew by 5 per cent per year over the 2020–2022 period. However, the GDPs of 2019 and 2022 were very close to each other in real terms; more strikingly, the annual Italian GDP has still not surpassed that of 2007.

In Italy, we also observed a long-term decline in the rate of capacity utilization, as measured by the output–capital ratio. In each business cycle, we detected a lower average of this variable. The average profit rate also declined over the sample period; however, we observed a partial recovery over the 2013–2019 period before the COVID-19 crisis. The profit rate recorded the all-time lowest average values after the two crises. The capital income share fluctuated around 40 per cent over the 1995–2022 period in Italy while its 1995–2007 average was relatively higher than the sample average with a slight downward trend. However, since the Great Recession, the business-cycle averages of the capital income share have been slightly increasing.

The business-cycle averages of non-financial corporations’ rate of net interest payments,^{15} household’s rate of net interest and dividend income and adjusted labor income declined over the sample period. Households’ gross saving rate followed a similar path, but after the COVID-19 crisis we observed a jump in this rate. The business-cycle averages of non-financial corporations’ net dividend payments rate,^{16} however, have been increasing after the 2007–2009 crisis. This observation confirms with the financialization process of the Italian economy as discussed by Gabbi et al. (2016). The average rate of the debt stock of non-financial corporations was around 34 per cent and the business-cycle average of this rate decreased around 1 per cent in the last decade.

Business-cycle averages of short-term interest rates both in nominal and real terms declined over the sample period. The short-term real rate turned negative over the 2009–2011 period and the nominal rate in the 2013–2019 period. They remain negative after the COVID-19 crisis. While the long-term nominal rate follows a similar path, average real long-term interest rates increased after the 2007–2009 crisis but significantly declined during the 2013–2019 period. However, we observe a negative rate in 2020 and a very low rate in 2021.

### 4.2 Econometric methodology and findings

At the first step of our empirical analysis, we applied Augmented Dickey–Fuller (ADF), Kwiatkowski–Phillips–Schmidt–Shin (KPSS) and Phillips–Perron (PP) unit root tests in order to determine the stationarity properties of the series.^{17} All tests confirmed that *(u −* *u _{n})* is stationary, i.e. I(0). While ADF and PP tests indicated that

*h*,

*g*and

*D*/

*pK*are integrated of order 1, i.e. I(1), they are I(0) with respect to the KPSS test. ADF and PP tests imply that

*S*/

_{h}*pK,*

*W*/

*pK*and

*Z*/

*pK*series are I(0) while they are I(1) at 5 per cent with respect to the KPSS test. Taking into account the controversial results of the unit root tests and in order to avoid potential econometric problems, we estimated the model using the OLS method by differencing the variables.

^{18}We also employed the Seemingly Unrelated Regression (SUR) method and it proved to be convenient for the estimation of the model according to the Breusch–Pagan and Hansen–Sargan tests. We checked the robustness of the regressions using the following diagnostic tests in the estimations using the OLS method: the null hypothesis of the F-test is that all coefficients are jointly equal to zero, that of the LM test is that there is no autocorrelation, that of the White’s test is that there is no heteroskedasticity, that of the ARCH test is that there is no autoregressive conditional heteroskedasticity, that of the normality test is that the error terms are normally distributed.

^{19}Regression Equation Specification Error Test (RESET) checks the adequacy of specifications and Cumulative Sum of Recursive Residuals Test (CUSUM) the parameter stability. In the SUR estimation, the null hypothesis of the Breusch–Pagan test is that there is no variance among model equations and that of the Hansen–Sargan test is that instruments used are valid. The model equations are presented in Table 2.

#### Equations used in the estimations of parameters of the model

*Notes:* *u*, *u _{n},*

*h*,

*g*,

*r*,

*Z*/

*pK*,

*D*/

*pK*,

*ZD*/

*pK*,

*W*/

*pK*and

*S*/

_{h}*pK*denote the capacity utilization rate, the normal rate of capacity utilization, the capital income share, the capital accumulation rate, the profit rate, rate of interest payments of non-financial corporations, rate of net dividend payments of non-financial corporations, rate of households’ net interest and dividend income, rate of adjusted labor income and savings rate of households, respectively.

We tried alternative OLS specifications for the model equations in order to solve the problems with the error terms. Profit share and capital accumulation rate equations of the OLS specification includes the dummy variable *DCOVID*, which takes into account the effects of the COVID-19 lockdowns during the 2020–2021 period in Italy. We also employed this variable in the capital accumulation rate equation in the SUR estimation. The OLS estimations of the model equations are presented in Tables 3 through 5 and the SUR estimation is represented in Table 6.

#### Estimation of the profit share equation via OLS method

*Notes:* *h*, *Z/pK* and *DCOVID* denote the capital income share, rate of interest payments of non-financial corporations and the dummy variable for the years 2020–2021, respectively. *, ** and *** represent 10 per cent, 5 per cent and 1 per cent significance levels, respectively. The numbers corresponding to the diagnostic tests are P-values.

#### Estimation of the capital accumulation rate equation via OLS method

*Notes:* *u*, *u _{n},*

*h*,

*Z*/

*pK*,

*D*/

*pK*and

*DCOVID*denote the capacity utilization rate, the normal rate of capacity utilization, the capital income share, rate of interest payments of non-financial corporations, rate of net dividend payments of non-financial corporations and the dummy variable for the years 2020–2021, respectively. *, ** and *** represent 10 per cent, 5 per cent and 1 per cent significance levels, respectively. The numbers corresponding to the diagnostic tests are P-values.

#### Estimation of the household saving rate equation via OLS method

*Notes:* *ZD/pK*, *W/pK* and *S _{h}/pK* denote the rate of households’ net interest and dividend income, rate of adjusted labor income and savings rate of households, respectively. *, ** and *** represent 10 per cent, 5 per cent and 1 per cent significance levels, respectively. The numbers corresponding to the diagnostic tests are P-values.

#### Estimation of the model equations via SUR method

*Notes:* *u*, *u _{n},*

*h*,

*g*,

*Z*/

*pK*,

*D*/

*pK*,

*ZD*/

*pK*,

*W*/

*pK*,

*S*/

_{h}*pK*and

*DCOVID*denote the capacity utilization rate, the normal rate of capacity utilization, the capital income share, the capital accumulation rate, rate of interest payments of non-financial corporations, rate of net dividend payments of non-financial corporations, rate of households’ net interest and dividend income, rate of adjusted labor income, savings rate of households and the dummy variable for the years 2020–2021, respectively. *, ** and *** represent 10 per cent, 5 per cent and 1 per cent significance levels, respectively. The numbers corresponding to the diagnostic tests are P-values.

In the profit share equation in Table 3, the validity of the OLS specification is rejected by the F-test and the impact of net interest payments on the profit share is insignificant. Besides, it suffers from heteroscedasticity. The SUR estimation in Table 6 gives an identical result. These findings imply that the parameter

In the OLS estimation of the capital accumulation equation in Table 4 and the SUR estimation in Table 6, the animal spirits turned out to be significant with an absolute value close to zero. While this coefficient values range between 0 and 0.06 in Hein/Ochsen’s (2003) study on France, Germany, the UK and US, the findings of HS (2011) on Germany and the US are similar to ours. We found the impact of the deviation of the rate of capacity utilization from its trend in Italy to be between 0.21 in the OLS estimation and 0.32 in the SUR estimation.^{20} Hein/Ochsen (2003) used the GDP growth rate in the accumulation rate function and the estimates of this coefficient over the full and sub-periods lie within the interval of 0.10–0.40. In HS’s study the values of the corresponding coefficients were 0.15 for Germany and 0.14 for the US. The impact of the profit share on the accumulation is also significant and positive in the three estimations and this coefficient ranges between 0.12 and 0.23. While in Hein/Ochsen’s study this coefficient even turned out to be negative for the UK in a subperiod, HS had estimated these coefficients to be 0.33 for Germany and 0.14 for the US. The findings on the estimated impact of net interest payments by non-financial corporations on the accumulation are in line with the hypotheses of the model. In Italy, a 1 per cent increase in these payments lead to 0.49 (OLS estimation) to 0.53 (SUR estimation) per cent decline in the rate of accumulation. In HS’s study, stronger effects for Germany (1.03) and the US (0.72) were observed. As for the effect of dividend payments by non-financial corporations on accumulation, the corresponding coefficient turned out to have the opposite sign in the Italian economy. This finding is in contradiction with the hypothesis of the model and requires further explanation. HS also failed to find negative effects of dividend payments on accumulation in Germany and the US, their estimated coefficients being insignificant.

The saving rate equation was estimated using the lags of the explanatory variable in order to solve the problems with the error terms in the OLS estimation in Table 5. The propensity to save out of wages turned out to be 0 and that out of dividend and interest income to be 0.72, respectively. In the SUR estimation we obtained a similar result, the corresponding coefficients are 0 and 0.70, respectively. Hein/Ochsen (2003) estimated these parameters using an equation that employed income shares instead of income rates. The propensity to save out of wages ranged between −0.19 and 0.08 and that out of rentiers’ income between 0.76 and 1.5.^{21} HS (2011) found these two propensities to be 0.13 and 0.60 for Germany and 0.09 and 0.76 for the US, respectively.

### 4.3 Synthesis and interpretation of findings

We synthetize our econometric findings, calculate the effects of the variations in interest rates on the equilibrium rates of the endogenous variables and interpret them in this sub-section. We compute the elasticities in equations 16, 17 and 18 and present them in Table 7 by combining the three sets of parameters from the estimated specifications of each equation.

#### Synthesis of econometric findings and calculation of effects of interest rates

*Notes:* *s _{W}* and

*s*are the estimated propensities to save out of labor income and rentiers’ income, respectively.

_{Z}*h*,

*u*and

^{a}Following the argumentation of Hein/Schoder (2011: 718), average instead of equilibrium values are used for calculations.

^{b}Average value for the period 1995–2021.

The calculation of the elasticities using two constellations of parameters qualitatively gave the same results in terms of the effects of the interest rate variations on the equilibrium rates capacity utilization, capital accumulation and profit. The calculated effects were decomposed into ‘primary’ and ‘secondary’ categories as defined in Section 2. The latter are zero since the profit share is inelastic with respect to net interest payments and therefore primary and total effects are the same.

According to the calculations, the stability condition was met in the two estimations. The synthesis of the results shows that according to the OLS and the SUR estimations, the Italian economy is characterized by a *normal* regime, as defined in the framework of HS (2011), since an increase in the interest rates has a negative effect on the equilibrium rates of capacity utilization, capital accumulation and profits. According to these two estimations, a 1 per cent increase in the real long-term interest rate results in 0.39 per cent to 0.95 per cent decrease in the equilibrium rate of capacity utilization, 0.25 to 0.48 per cent decrease in the equilibrium rate of capital accumulation and 0.15 to 0.38 per cent decrease in the equilibrium rate of profit. The findings of HS (2011) on Germany and the US are qualitatively identical to ours but the magnitudes of the elasticities are significantly different. An increase in the interest rates by 1 per cent led to almost 2 per cent decline in equilibrium rate of capacity utilization in Germany and 1.82 per cent decline in the US. The effect on the equilibrium rate of accumulation was also negative: 0.37 per cent for Germany and 0.32 per cent for the US. The effect on the equilibrium profit rate turned out to be almost zero in Germany and negative (−0.13 per cent) in the US. The signs remained the same for both countries in the sub-period calculations, but their magnitudes varied. For Germany they found a very weak negative effect on the equilibrium rate of profit for the 1960–1982 sub-period and a very weak positive effect for the 1983–2007 sub-period, while for the full sample period this effect turned out to be zero.

## 5 CONCLUSION

In this empirical study, we examined the effects of exogenous variations in interest rates on the equilibrium rates of capacity utilization, capital accumulation and profit in Italy. We based our analysis on a post-Kaleckian distribution and growth model augmented by monetary variables. This model by HS (2011) generates different monetary regimes as a function of the parameters in the saving, capital accumulation and profit share functions.

Our findings suggest that the monetary regime in Italy is characterized by the ‘normal regime’, as defined within the theoretical model, over the 1995–2022 period. The findings are robust to the choice of econometric specification and imply that, at a given debt–capital ratio, increasing the real long-term interest rates has a negative effect on the rates of capacity utilization, accumulation, and profits in Italy. The magnitude of the calculated effect on the equilibrium rate of capacity utilization is very close to unity, while the other effects are limited but not negligible. As the model assumes that short- and long-term real interest rates can be affected by the central bank, the results point out that tight monetary policies have ended up with lower rates of capacity utilization, accumulation, and profits in Italy. Eventually, these outcomes underline the need for central banks targeting low long-term interest rates. Even though monetary policy cannot be solely held responsible for these outcomes, in the case of Italy it can be argued that the inflation-targeting policies based on NCMs have not gone hand in hand with high economic growth and accumulation rates.

It should be noted that throughout the period that we analyzed, the Italian economy witnessed an important structural transformation whose main aspects are privatization, liberalization of capital movements, financialization, and neoliberal reforms in the labor market and welfare system. These developments and subsequent rise in inequalities, along with austerity measures, played an important role in the suppression of aggregate demand, which cannot be solely explained by the monetary policy stances or variations in the interest rates. Although it might be argued that the variations in the interest rates are indirectly affected by these developments, they were not captured by the framework of our analysis. Another point to be mentioned is the important role of government expenditures and exports in the growth in the Italian economy, which is also outside the framework of the model employed for the analysis.

We expect our findings to contribute to the debate on the application of inflation-targeting policies implied by NCMs and the discussions for its alternatives. Further research in this field can analyze the long-term dynamics of an endogenous debt–capital ratio along with the relation between short- and long-term interest rates.

## ACKNOWLEDGEMENTS

I would like to thank two anonymous referees and Eckhard Hein for their valuable comments that helped improve this work.

- 1↑
Sawyer (2008: 62) argues that these models are ‘neither new nor Keynesian’, as the title of his work suggests, and states that his views on the concept of Keynesian carry a ‘strong Kaleckian tinge’.

- 3↑
We opt for using the term

*monetary regime*instead of*accumulation regime*to distinguish the former from the regimes defined in other post-Kaleckian models that examine the effects of functional income distribution on the rates of capacity utilization, capital accumulation, and profit. - 4↑
According to WID (2023), the top 1 per cent income share increased from around 5.5 per cent in early 1980s to above 12 per cent in 2021 in Italy. More strikingly, the top 1 per cent wealth share increased from 16.4 per cent in 1995 to almost 26 per cent in 2014 and remains around 22 per cent as of 2021.

- 5↑
According to the classification of Hein/Martschin (2021), the Italian

*demand and growth regime*was*domestic demand-led*over the 2001–2009 period and turned export-led mercantilist in the 2010–2019 period. This was supported by the*macroeconomic policy regime*, which is identified on the basis of the way monetary, fiscal and wage policies are applied. - 6↑
The development of the literature on the PK monetary theory has preceded that on financialization. The latter incorporates topics such as shareholder value orientation, top management salaries (Hein 2012a: 26), debt-financed consumption and accumulation. See Hein (2012a, 2014: ch. 10) for extensive theoretical research in and review of this field. For a comprehensive study on financialization in Italy, see Gabbi et al. (2016).

- 7↑
The model implicitly assumes constant prices; therefore, the interest rate

*i*can be considered both nominal and real. - 8↑
This assumption implies that productive activity is not constrained by labor supply (Blecker/Setterfield 2019: 355).

- 9↑
In the literature, output gap as a percentage of potential output (Stockhammer/Onaran 2004) and growth rate of GDP (Onaran/Stockhammer 2005) are used as a proxy for demand as a substitute for the rate of capacity utilization.

- 10↑
The discussion between the horizontalist and the structuralist post-Keynesians on endogenous money supply goes back to the 1990s. For the contributions to this debate, see Lavoie (1996) and his earlier studies cited in the latter and for a recent one, see the chapters in Part III in Rochon/Rossi (2017).

- 11↑
Deleidi (2018), within a PK endogenous money framework, conducted an empirical analysis on the relationship between different types of interest rates and quantity of loans granted by commercial banks in the eurozone. The author detected a negative relation between the interest rates and the loans received for house purchases, while no significant relation was observed between loans granted to firms and loans for consumption and the corresponding interest rates.

- 12↑
The post-Keynesian theoretical and empirical literature on the relation between short- and long-run interest rates is well developed. In a recent study Akram (2022) summarizes Keynes’ views on this relation, surveys the empirical literature and proposes a model of the long-term interest rate dynamics.

- 13↑
See Hein (2014: ch. 7) for under which conditions demand and accumulation regimes become wage-led or profit-led in an extension of the Bhaduri/Marglin’s (1990) model in both closed and open economies with savings out of labor income.

- 14↑
Several PK empirical studies have been conducted in order to determine demand and growth regimes in emerging and advanced capitalist economies, without addressing monetary variables. For an up-to-date summary of the findings of the cumulative empirical research in the field of wage- vs. profit-led demand regimes, see Oyvat et al. (2020).

- 15↑
As also observed by HS (2011) for the US and Germany, due to data problems net interest payments are not equal to the product of the net debt–capital ratio and the long-term interest rate in the Italian economy. We found the correlation between the former and the latter to be 0.58 and the average of the ratio of the former to the latter is equal to 0.23 over the sample period.

- 16↑
The sum of net interest payments and net dividend payments of non-financial corporations are not equal to households’ net interest and dividend income for at least two reasons. First, dividends are also paid to foreign households or companies or received from foreign companies. Second, only a fraction of net interest payments is received directly or indirectly by domestic households, the rest being paid to foreign or domestic financial institutions.

- 17↑
The unit root tests are not reported due to limited space but available on request. Due to the limited sample size, we did not test for structural breaks. However, it should be mentioned that the Italian economy witnessed important structural changes during the sample period of 1995–2022.

- 18↑
For the accumulation rate, we also tested cointegration among variables using the Bounds test by Pesaran et al. (2001). The critical values provided by the latter for the test are asymptotic. We employed those calculated by Narayan (2005), who provided the values for sample sizes between 30 and 80. No evidence was found for cointegration. The cointegration tests are available on request.

- 20↑
The former value is obtained by summing the coefficients of the contemporaneous and the lagged variables in the second OLS specification.

- 21↑
The series for the rentier income share was calculated as the product of capital stock and interest rate in this study. In their calculations of the elasticities, the propensities exceeding unity were taken as 1.

## REFERENCES

Akram T. , 'A simple model of the long-term interest rate ' (2022 ) 45 (1 )

*Journal of Post Keynesian Economics*: 130 -144.Amadeo E.J. , 'Notes on capacity utilisation, distribution and accumulation ' (1986a ) 5 (1 )

*Contributions to Political Economy*: 83 -94.Amadeo E.J. , 'The role of capacity utilisation in long period analysis ' (1986b ) 2 (2 )

*Political Economy*: 147 -160.AMECO Database, Annual Macro-economic Database of the European Commission’s Directorate General for Economic and Financial Affairs, URL: https://ec.europa.eu/info/business-economy-euro/indicators-statistics/economic-databases/macro-economic-database-ameco_en (accessed June 2023).

Arestis P., Sawyer M. (2002): ‘New Consensus’, New Keynesianism, and the economics of the ‘Third Way’, Levy Institute Working Paper, No 364.

Argitis G. , 'Inflation targeting and Keynes’s political economy ' (2008 ) 31 (2 )

*Journal of Post Keynesian Economics*: 249 -270.Argitis G. & Michopoulou S. , 'Monetary policy, interest payments, income distribution and the macroeconomy ' (2010 ) 6 (1–2 )

*Review of Applied Economics*: 29 -39.Argitis G. & Pitelis C. , 'Monetary policy and the distribution of income: evidence for the United States and the United Kingdom ' (2001 ) 23 (4 )

*Journal of Post Keynesian Economics*: 617 -638.Bagnai A. , 'Italy’s decline and the balance-of-payments constraint: a multicountry analysis ' (2016 ) 30 (1 )

*International Review of Applied Economics*: 1 -26.Bhaduri A. & Marglin S. , 'Unemployment and the real wage: the economic basis for contesting political ideologies ' (1990 ) 14 (4 )

*Cambridge Journal of Economics*: 375 -393.Blecker R.A. & Setterfield M. ,

*Heterodox Macroeconomics: Models of Demand, Distribution and Growth*, (Edward Elgar Publishing, Cheltenham, UK and Northampton, MA, USA 2019 ).Bugamelli M., Fabiani S., Federico S., Felettigh A., Giordano C., Linarello A. (2017): Back on track? A macro-micro narrative of Italian exports, Bank of Italy Occasional Paper, No 399.

Cesaratto S. , 'Italy: A question of interest rates and trust ' (2018 ) 53 (6 )

*Intereconomics*: 294 -295.Cesaratto S. & Zezza G. , 'What went wrong with Italy and what the country should fight for in Europe ', in H. Herr, J. Priewe & A. Watt (eds),

*Still Time to Save the Euro: A New Agenda for Growth and Jobs with a Focus on the Euro Area’s Four Largest Countries*, (Social Europe Publishing , Berlin 2019 ) 47 -61.Deleidi M. , 'Post Keynesian endogenous money theory: a theoretical and empirical investigation of the credit demand schedule ' (2018 ) 41 (2 )

*Journal of Post Keynesian Economics*: 185 -209.Dutt A.K. , 'Stagnation, income distribution and monopoly power ' (1984 ) 8 (1 )

*Cambridge Journal of Economics*: 25 -40.Dutt A.K. , 'Alternative closures again: a comment on ‘Growth, Distribution and Inflation’ ' (1987 ) 11 (1 )

*Cambridge Journal of Economics*: 75 -82.Dutt A.K. , 'Accumulation, distribution and inflation in a Marxian/post-Keynesian model with a rentier class ' (1989 ) 21 (3 )

*Review of Radical Political Economics*: 18 -26.Dutt A.K. , 'Internal finance and monopoly power in capitalist economies: a reformulation of Steindl’s growth model ' (1995 ) 46 (1 )

*Metroeconomica*: 16 -34.EUROSTAT Database, European Statistical Office Database, URL: https://ec.europa.eu/eurostat (accessed June 2023).

Gabbi G., Ticci E. & Vozzela P. , 'The transmission channels between the financial and the real sectors in Italy and the crisis ', in E. Hein, D. Detzer & N. Dodig (eds),

*Financialisation and the Financial and Economic Crises*, (Edward Elgar Publishing , Cheltenham, UK and Northampton, MA, USA 2016 ) 234 -254.Gelman A., Hill J. & Vehtari A. ,

*Regression and Other Stories*, (Cambridge University Press, Cambridge, UK 2021 ).Goes M.C.B. , 'Personal income distribution and progressive taxation in a neo-Kaleckian model: insights from the Italian case ' (2020 ) 32 (4 )

*Review of Political Economy*: 615 -639.Hein E. , 'Interest rates, income shares and investment in a Kaleckian model ' (1999 ) 5

*Political Economy: Review of Political Economy and Social Sciences*: 5 -22.Hein E. , 'Interest, debt and capital accumulation — a Kaleckian approach ' (2006 ) 20 (3 )

*International Review of Applied Economics*: 337 -352.Hein E. , 'Interest rate, debt, distribution and capital accumulation in a post-Kaleckian model ' (2007 ) 58 (2 )

*Metroeconomica*: 310 -339.Hein E. ,

*Money, Distribution Conflict and Capital Accumulation*, (Palgrave Macmillan, Basingstoke 2008 ).Hein E. ,

*The Macroeconomics of Finance-dominated Capitalism and its Crisis*, (Edward Elgar Publishing, Cheltenham, UK and Northampton, MA, USA 2012a ).Hein E. , 'The rate of interest as a macroeconomic distribution parameter: horizontalism and post-Keynesian models of distribution and growth ' (2012b ) 6 (2 )

*Bulletin of Political Economy*: 107 -132.Hein E. ,

*Distribution and Growth After Keynes*, (Edward Elgar Publishing, Cheltenham, UK and Northampton, MA, USA 2014 ).Hein E. & Martschin J. , 'Demand and growth regimes in finance-dominated capitalism and the role of the macroeconomic policy regime: a post-Keynesian comparative study on France, Germany, Italy and Spain before and after the Great Financial Crisis and the Great Recession ' (2021 ) 2 (3 )

*Review of Evolutionary Political Economy*: 493 -527.Hein E. & Ochsen C. , 'Regimes of interest rates, income shares, savings, and investment: a Kaleckian model and empirical estimations for some advanced OECD economies ' (2003 ) 54 (4 )

*Metroeconomica*: 404 -433.Hein E. & Schoder C. , 'Interest rates, distribution and capital accumulation – a post-Kaleckian perspective on the US and Germany ' (2011 ) 25 (6 )

*International Review of Applied Economics*: 693 -723.ISTAT Database, Italian National Institute of Statistics Database, URL: https://www.istat.it/en (accessed June 2023).

Kaldor N. , 'The new monetarism ' (1970 ) 97

*Lloyds Bank Review*, July : 1 -18.Kaldor N. ,

*The Scourge of Monetarism*, (Oxford University Press, Oxford, UK 1982 ).Kaldor N , 'How monetarism failed ' (1985 ) 28 (2 )

*Challenge*: 4 -13.Kalecki M. , 'The principle of increasing risk ' (1937 ) 4 (16 )

*Economica*: 440 -447.Kappes S.A. , 'Monetary policy and personal income distribution: a survey of the empirical literature ' (2021 ) 35 (1 )

*Review of Political Economy*: 211 -230.Lagna A , 'Derivatives and the financialisation of the Italian state ' ((2016 ) 21 (2 )

*New Political Economy*: 167 -186.Lavoie M. , 'The endogenous flow of credit and the post-Keynesian theory of money ' (1984 ) 18 (3 )

*Journal of Economic Issues*: 771 -797.Lavoie M. ,

*Foundations of Post Keynesian Economic Analysis*, (Edward Elgar Publishing, Aldershot 1992 ).Lavoie M. , 'Interest rates in post‐Keynesian models of growth and distribution ' (1995 ) 46 (2 )

*Metroeconomica*: 146 -177.Lavoie M. , 'Horizontalism, structuralism, liquidity preference and the principle of increasing risk ' (1996 ) 43 (4 )

*Scottish Journal of Political Economy*: 275 -300.Lavoie M. ,

*Introduction to Post-Keynesian Economics*, (Palgrave Macmillan, Basingstoke 2006 ).Moore B.J. ,

*Horizontalists and Verticalists: The Macroeconomics of Credit Money*, (Cambridge University Press, Cambridge, UK 1988 ).Moore B.J. , 'The endogeneity of credit money ' (1989 ) 1 (1 )

*Review of Political Economy*: 65 -93.Naastepad C.W.M. & Storm S. , 'OECD demand regimes (1960–2000) ' (2006 ) 29 (2 )

*Journal of Post Keynesian Economics*: 211 -246.Narayan P.K. , 'The saving and investment nexus for China: evidence from cointegration tests ' (2005 ) 37 (17 )

*Applied Economics*: 1979 -1990.Nishi H. , 'On the short-run relationship between the income distribution-growth and debt-growth regimes ' (2013 ) 27 (6 )

*International Review of Applied Economics*: 729 -749.Obst, T.,, Onaran, Ö.,, Nikolaidi, M. (2017): The effect of income distribution and fiscal policy on growth, investment, and budget balance: the case of Europe, Macro-Economic Policy Institute (IMK) FMM Working Paper, No 10.

Onaran Ö. & Galanis G. , 'Income distribution and growth: a global model ' (2014 ) 46 (10 )

*Environment and Planning A*: 2489 -2513.Onaran Ö. & Obst T. , 'Wage-led growth in the EU15 member-states: the effects of income distribution on growth, investment, trade balance and inflation ' (2016 ) 40 (6 )

*Cambridge Journal of Economics*: 1517 -1551.Onaran Ö. & Stockhammer E. , 'Two different export-oriented growth strategies: accumulation and distribution in Turkey and South Korea ' (2005 ) 41 (1 )

*Emerging Markets Finance and Trade*: 65 -89.Onaran Ö., Stockhammer E. & Grafl L. , 'Financialisation, income distribution and aggregate demand in the USA ' (2011 ) 35 (4 )

*Cambridge Journal of Economics*: 637 -661.Oyvat C., Öztunali O. & Elgin C. , 'Wage-led versus profit-led demand: a comprehensive empirical analysis ' (2020 ) 71 (3 )

*Metroeconomica*: 458 -486.Pasinetti L.L. ,

*Growth and Income Distribution*, (Cambridge University Press, Cambridge, UK 1974 ).Pesaran M.H., Shin Y. & Smith R.J. , 'Bounds testing approaches to the analysis of level relationships ' (2001 ) 16 (3 )

*Journal of Applied Econometrics*: 289 -326.Rochon L.P. & Rossi S. , 'Inflation targeting, economic performance, and income distribution: a monetary macroeconomics analysis ' (2006 ) 28 (4 )

*Journal of Post Keynesian Economics*: 615 -638.Rochon L.P. & Rossi S. ,

*Advances in Endogenous Money Analysis*, (Edward Elgar Publishing, Cheltenham, UK and Northampton, MA, USA 2017 ).Rowthorn R.E. , 'Demand, real wages and economic growth ' (1981 )

*Thames Papers in Political Economy*: 1 -39.Sawyer M. , 'Neither new nor Keynesian: a critique of the New Keynesian programme ' (2008 ) 35

*METU Studies in Development*, June : 61 -80.Stockhammer E. & Onaran Ö. , 'Accumulation, distribution and employment: a structural VAR approach to a Kaleckian macro model ' (2004 ) 15 (4 )

*Structural Change and Economic Dynamics*: 421 -447.Tridico P. , 'From economic decline to the current crisis in Italy ' (2015 ) 29 (2 )

*International Review of Applied Economics*: 164 -193.WID Database, World Inequality Database, URL: https://wid.world (accessed June 2023).

## APPENDIX

#### Data: definition and sources

All the data are annual and cover the period from 1995 to 2022 except the debt–capital ratio and interest rate, which is available for the period from 1995 to 2021. The data were retrieved from AMECO (Annual Macro-economic Database of the European Commission’s Directorate General for Economic and Financial Affairs), EUROSTAT (European Statistical Office) and ISTAT (Italian National Institute of Statistics) in June 2023. Due to problems with the data, the rate of net interest payments is not generally equal to the product of the real interest rate and the debt-capital ratio, as also observed by HS (2011).

** K:** total net fixed capital, year-end outstanding amount in volume at linked prices. Source: ISTAT.

* g:* Annual growth rate of

*K*.

** pK:** Total net fixed capital, year-end outstanding amount at current prices. Source: ISTAT.

* GDP growth rate:* Growth rate of the real GDP at market prices at constant prices. Source: ISTAT.

* u:* Rate of capacity utilization. Net domestic product at constant prices, which is equal to GDP at market prices at constant prices minus consumption of fixed capital at constant prices, over

*K*. Source: ISTAT.

* u_{n}:* Trend of

*u*extracted using the Hodrick–Prescot filter with a smoothing parameter of 100.

* h:* Capital income share as percentage of GDP at current factor cost. Source: AMECO.

* r:* Profit rate. Capital income, calculated by multiplying

*h*by real GDP, over

*K*. Sources: AMECO and ISTAT.

* i = Z/pK:* Rate of net interest payments of non-financial corporations. Net interest payments of non-financial corporations (

*Z*) over

*pK*. Source: ISTAT.

* d = D/pKd:* Rate of net dividend payments of non-financial corporations. Net dividend payments of non-financial corporations (

*D*) over

*pK.*Source: ISTAT.

* S_{h}/pK:* Saving rate of households. Gross savings (

*S*) over

_{h}*pK.*Source: ISTAT.

* W/pK:* Rate of adjusted labor income. Adjusted labor income calculated by multiplying adjusted wage share in total economy as percentage of GDP by GDP at market prices at current prices over

*pK.*Sources: AMECO and ISTAT.

* ZD/pK:* Rate of households' net interest and dividend income. Households’ net interest and dividend income (

*ZD*) over

*pK.*Source: ISTAT.

** B:** Total financial liabilities net of total financial assets. Source: EUROSTAT.

* = B/pK:* Debt–capital ratio.

*B*over

*pK*. Source: EUROSTAT and ISTAT.

** ISN:** Nominal annual short-term interest rate. Source: AMECO.

** ISRV:** Real annual short-term interest rate. Nominal annual short-term interest rate deflated by the GDP deflator. Source: AMECO.

** ILN:** Nominal annual long-term interest rate. Source: AMECO.

** ILRV:** Real annual long-term interest rate. Nominal annual long-term interest rate deflated by the GDP deflator. Source: AMECO.