How economists explain changes in income shares
Wage income shares of Gross Domestic Product have been falling, often dramatically, in all advanced countries since the mid-1970s/early 1980s. This represents an element of novelty with respect to the constancy postulated by one of the Nicholas Kaldor’s famous stylized facts about economic growth, and a worrying phenomenon, among other things, for its connection with increasing inequality as well as for its negative consequences on consumption and aggregate demand (Stockhammer, 2013; Cynamon and Fazzari, 2016; Pariboni et al., 2020). Many analyses and interpretations have been advanced, which can be very broadly and with a certain degree of simplification divided into two main groups.
The first comprises a variety of explorations that are embedded in neoclassical theory of distribution, and hence, according to this theory, must consider the substitutability between production factors. The latter would normally imply a certain stability over time of income shares, since any reduction in the return to a factor of production would cause a greater intensity of its use. That is, in the case of labor, lower wages (other thing given) would cause higher labor to capital and labor to GDP ratios. Hence, in order to explain the observed large changes in income shares, the various contributions that belong to this broad group tend to emphasize biased (i.e. unskilled-labor saving) technical progress (Hutchinson and Persyn, 2012; Hogrefe and Kappler, 2013; Bassanini and Manfredi, 2014; IMF, 2017) and/or the major increase in world-wide (unskilled) labor supply resulting from the expansion of international trade and the large newcomers – such as China and India – that have been integrating into global markets (Acemoglu; 2003; Guscina 2006). It should be emphasized that both channels (technical change and international trade) according to this approach would not affect unemployment, which would always tend to be at its equilibrium level, but would negatively affect the equilibrium wage, particularly for unskilled workers in mature industrial economies.
The other broad group of interpretations of the changes in income shares can be identified as not subscribing to traditional neoclassical substitutability between production factors, and hence embracing a ‘conflict’ and/or institutional theory of income distribution. Such a theoretical framework can more freely account for the changes in income distribution as caused by institutional and economic changes that have adversely affected labor bargaining positions, since in this framework, unlike in the neoclassical one, a fall in wages would not necessarily lead to a change in techniques or consumption patterns such as to bring about a greater intensity in the use of labor vis-à-vis other production factors. In other words, according to these explanations, there are no off-setting effects on income shares deriving from the shift towards more labor-intensive forms of production, hence the income shares are quite responsive to the evolution of the bargaining position of labor. The latter, in turn, may be affected by a number of social, economic and institutional factors. Rather than reflecting technological factors (marginal productivity), and relative scarcity of resources (unskilled versus skilled labour supply), income shares can be seen as the most immediate indicator of the balance of forces between labour and capital’ (Franzini and Pianta, 2015, p. 71).
Contributions in this strand of analysis have generally adopted, both analytically and in empirical investigations, a political economy approach, focusing on the role of changes in labor market structure, (de)regulation and institutions along a variety of dimensions, such as: i) the process of structural change intended as a shift toward the low-pay and low-productivity segment of the service sector (Stockhammer, 2017; Storm, 2017; Beqiraj et al., 2019; Pariboni and Tridico, 2019a); ii) the retrenchment of welfare state and trade unions (Kristal, 2010; Bengtsson, 2014; Stockhammer, 2017; Tridico and Paternesi Meloni, 2018; Hein et al., 2020); iii) increasing labor market flexibility and the weakening of worker-protective labor laws (Brancaccio et al., 2018; Deakin et al., 2014); iv) the increasing process of globalization, which fostered trade openness (Rodrik, 1997; Onaran, 2011; Stockhammer, 2017) and financial flows related to offshoring practices (Elsby et al., 2013) and/or to portfolio investments (Jayadev, 2007); and v) the effects of increasing financialization (Stockhammer, 2009; Hein, 2015; Dünhaupt, 2017; Özdemir, 2019; Pariboni and Tridico, 2019b; Pariboni et al., 2020), including the influence of the returns on financial assets on the profit rate (Pivetti, 1991; Hein, 2014, albeit within different analytical frameworks). All or several of those factors are likely to have been very important, as often confirmed by empirical evidence. However, it is somewhat surprising that the role of persistent unemployment and more generally of labor market slack has remained relatively unexplored in this body of literature. Yet in the analyses of the Classical economists and Marx, which can be regarded as the historical and analytical roots of the conflict theory of income distribution, labor market conditions were regarded as very relevant factors in affecting wages, along with, of course, other institutional, historical and political elements.
In our new INET Working Paper, entitled ‘Unemployment and income distribution: some extensions of Shaikh’s analysis’, we try to begin filling this gap by uniquely focusing on labor market slack influence on the private sector wage share. Our enquiry can also contribute to clarifying whether unemployment could be one channel through which some of the factors highlighted by the just reviewed literature act upon income distribution. For example, according to the literature on financialization and corporate governance, one of the reasons why the latter would have adversely affected the wage share is the reduced incentive to enhance long-term growth of firms through real investment, with firms more oriented to short-term strategies and ‘downsize and distribute’ behaviour. On the aggregate level, this may involve higher job insecurity, an increase in bad jobs compared to relatively good ones, as well as higher unemployment, which thus would be one of the channels through which those changes might have affected distribution. Similarly, international capital mobility may not only have a threat effect of potential off-shoring, but it may also have involved actual de-industrialization and thus higher unemployment. Finally, a variety of contributions have regarded interest rates as directly affecting the normal profit rate or the mark-up charged by firms, and hence income distribution. But it is an open question whether this can be regarded as a mechanical transmission, that is, one that would always take place, or one that also depends on ‘wage resistance’ and workers’ bargaining strength, which influence nominal and real wage dynamics. In this respect, if labor market conditions are found to have a significant role, the implication would be in favor of the latter view rather than the former.
The Classical approach and ‘conflict’ wage determination
Following the classical tradition, given the limit set by output per worker and the necessarily positive rate of profit in an economy featuring private ownership of the means of production, the real wage will be comprised between a maximum level, corresponding to a minimum rate of profit, and a minimum level, corresponding to some historically determined consumption floor, that is, the ‘subsistence wage’ of the classical tradition (Stirati, 1992). Between these two limits, there lies a set of abstractly feasible income distributions: which will prevail will depends on the relative bargaining strength of the parties. In turn, such bargaining strength may be affected by several circumstances, among which two broad sets used to be regarded by the classical economists as the most important: on the one hand, the institutional setting, particularly concerning labor market and wage regulations and the degree of organization of workers and employers, as well as their political representation; on the other, labor market conditions (that is, labor market slack or tightness).
In contemporary economic literature, labor market conditions and their impact on wage dynamics are usually described using the unemployment rate. There are reasons, however, to regard unemployment as a useful but incomplete indicator of labor market conditions: in this section, for simplicity, we refer to ‘unemployment’ as a description of labor market conditions, but it should be understood that it could be substituted by other variables or sets of variables describing them.
The just described broad line of analysis might involve different ways of conceiving and representing the relation between unemployment rate and income distribution. That, is, it could be conceived as affecting the growth rate of the wage share, hence causing a decoupling between productivity growth and real wage growth; or affecting the wage share level (after a period of adjustment), which would involve a relatively tight connection between wage and productivity growth, except during the phases of adjustment; or productivity and unemployment could be thought to affect real wage growth rather independently of each other, in which case unemployment would tend to affect real wages rather independently of the wage share dynamics. In fact, none of these approaches appears fully satisfactory, but empirical analysis can contribute to a better understanding of these dynamics, which of course may differ over time and across countries.
In our paper, we explore the empirical relation between the wage share and the unemployment rate (or an additional index of labor market slack) in the medium to long run – that is, averaging across the cycle. This is a necessary qualification for two reasons. The first is that we are interested in exploring the underlying persistent factors affecting ‘normal’ income distribution and not its short run fluctuations. In addition to this, we want to clear our analysis from short-run disturbing factors. In particular, it is well known that productivity varies in a pro-cyclical manner, owing to labor hoarding and short-term changes in the intensity of its use; accordingly, this causes changes of the wage share in the opposite direction in the early phases of expansions and contractions, in a way that obviously has no connection with lasting changes in income distribution.
Against the background outlined in the previous sections, we have empirically explored the role of labor market slack in affecting the adjusted wage share in the private sector of the economy. While the pro-cyclical behavior of real wages in the short-run has been well documented in empirical analyses (Stirati, 2016), our purpose is to enquire whether persistent changes in unemployment are found to have lasting consequences on income distribution. To do so, we take as a source of inspiration the recent work by Shaikh (2016, chapter 14), who finds that an index of unemployment intensity constructed combining the unemployment rate with a measure of unemployment duration performs well in explaining the changes in the wage share in the United States over the post-war period. We extend Shaikh’s analysis in two main directions. On the one hand, we apply a similar empirical investigation to other large mature economies besides the US, namely, Canada, France, Germany, Italy, Japan, Spain, Sweden and the United Kingdom. On the other hand, we extend the box of tools by using different econometric techniques capable of detecting the long-term relation between changes in the unemployment rate – or the index of unemployment intensity – and income distribution to the same set of countries. At the same time, however, we assess Shaikh’s claim that there is a ‘normal’ unemployment rate (or unemployment intensity index) acting as a long-term attractor for the economic system.
Despite the fact that in the period examined, going from 1960 to 2017, most of the changes in the wage share are in the negative area, our explorations show that a ‘structural’ Phillips-type relationship between unemployment rate or unemployment intensity and the pace of change of the wage share in the private sector is remarkable for the US, Italy and Germany; more moderate in France and Japan and exists for sub-periods in the United Kingdom and Spain; while it cannot be detected in Canada and Sweden.
The descriptive statistical analysis of filtered data along Shaikh’s example indicates that the pace of change of the wage share often reflects the size of unemployment or unemployment intensity, at least within a certain range of the unemployment level, while the relation tends to weaken for persistently very high levels of unemployment. This might reflect the fact that while increasing unemployment may slow down real wage growth, it can hardly (and at any rate only very slowly) cause a decline in real wages after that their dynamics has been already reduced to very low or close to zero by high unemployment. In such circumstances the decline in the wage share would only reflect the pace of productivity growth and hence become rather independent of further increases in the unemployment rate.
We then explore whether a long-run relationship between the two unemployment measures and the wage share taken in levels or prime differences can be detected using econometric tools. To do so, we make use of three different empirical strategies, each one capable of identifying the long-term correlation from a different angle. Overall, our findings confirm the depressing impact of increasing labor market slack on the wage share for almost all countries (i.e., except Canada and mixed results concerning France), albeit with varying statistical significance and economic magnitude. Accordingly, our empirical explorations suggest that the Classical view that lasting changes in labor market conditions tend to have persistent effects on income distribution is confirmed although, as predictable, with exceptions and varieties – given the role of other influences (institutions, politics) on the bargaining position of the workers. This appears to have some interesting theoretical implications for the analysis of income distribution, namely, that bargaining power – and hence the actions undertaken by all the parties involved – does matter. This appears to be in contrast not only with mainstream models, but also with several non-mainstream contributions and models assuming that firms can fix their mark-up according to their own strategic aims or product-market conditions regardless of workers claims and bargaining power (with the latter affecting inflation, but unable to alter income distribution); it also conflicts with the notion that interest rate fixing on the part of central banks is the sole or main determinant of profit rates.
Our empirical results may also provide some insights about the shape of the relationship between unemployment and the wage share. Time series autoregressive distributed lag-based estimates (ARDL estimates) suggest that changes in unemployment tend to determine long-period changes in the wage share (with coefficients over the period ranging between -0.2 and -0.8 across countries). The most reliable panel-ARDL analysis (carried out by means of a pooled mean group estimator) provides a coefficient of -1 concerning the impact of the unemployment rate on the private sector wage share level in the long run. When the panel analysis is run over sub-periods (in order to take into account changes in policy regimes), the coefficient on unemployment turns out to be somewhat higher for the 1960-1980 period than for the more recent decades. This is not too surprising, if we consider that the impact of changes in the unemployment rate tend to be smaller when unemployment rates are already very high and have been so for some time, and real wage growth already much subdued. Besides, it could also be the case that reduced trade union power tames the positive reaction of wages to reductions in unemployment, particularly when the latter is still relatively high.
The descriptive statistical analysis of filtered data along Shaikh’s example indicates that the pace of growth or decline of the wage share often reflects the size of unemployment or unemployment intensity, at least within a certain range of the unemployment level, while the relation tends to weaken for persistently very high levels of unemployment. This might reflect the fact that while increasing unemployment may slow down real wage growth, it can hardly (and at any rate only very slowly) cause a decline in real wages after that their dynamics is already close to zero by high unemployment. In such circumstances the decline in the wage share would only reflect the pace of productivity growth and hence become rather independent of further increases in the unemployment rate.
Finally, impulse response functions estimated by means of SVAR-based techniques indicate a range of the impact of a one point change in the unemployment rate, after five years, between -0.3 and -0.8, with a subsequent stabilization of the impact, despite the fact that results suggest that changes in unemployment persist over the ten year time-span considered. Thus, the empirical analysis implies that after a certain number of years, the wage share stabilizes at the new (lower) level.
Concerning the comparative impact of unemployment rate and the index of unemployment intensity, the latter appears to perform remarkably better in explaining the pace of changes in the wage shares only in the United States and Italy, and exhibits very similar coefficients between the two countries. In all other countries, the two measures of unemployment have similar coefficients.
The United States, Italy, the United Kingdom and, to a more limited extent, Germany turn out to be the countries where the estimated relationships between unemployment and the wage share are economically and statistically the most significant. This again is an interesting result, which contrasts with the often-claimed greater rigidity of the Italian labor market and wage-setting process with respect to the Anglo-Saxon model. An alleged rigidity that in the case of Italy is often claimed to be a major cause of high unemployment.
Although they are significant both from the economical and statistical point of view, the estimated impacts of unemployment on the wage share do not appear to be of a size capable of explaining the entire extension of the actual changes in income shares. This on the one hand tends to confirm an independent role for other institutional and structural factors already discussed in the literature and, on the other hand, suggests further explorations concerning the possibility that other dimensions of labor market conditions may have an impact (for example, participation rates and the size of the underemployed).
Is there an ‘equilibrium’ or ‘normal’ unemployment rate?
Similar to mainstream macro-models, albeit based on different analytical grounds, several non-mainstream models of Marxian or neo-Kaleckian orientation predict the existence of some ‘equilibrium’ or ‘normal’ unemployment rate representing an attractor for the economic system (see Stockhammer, 2008; Shaikh, 2016). In Shaikh (2016) contribution, the normal unemployment is that required to keep the wage share and the corresponding normal profit rate at the levels that ensure that the actual growth of the economy matches (on average) the Harrodian natural rate (the sum of labor force growth and productivity growth). In turn, the latter condition is required to keep unemployment constant. Other contributions (see Stockhammer, 2008) see this particular unemployment rate as the one consistent with stable (conflict) inflation. This normal unemployment rate can change over time owing to institutional changes. In particular, it would become lower as a consequence of reduced labor protection, since a reduced labor strength requires lower ‘disciplining’ unemployment in order to preserve profitability and/or price stability.
Empirical results in our paper do not provide support to the view that there is some endogenous spontaneous economic mechanism tending to cause a reversion of the economy toward a normal unemployment rate (except for the United States when considering the unemployment rate). We find that the unemployment rate and the unemployment intensity are not stationary nor trend-stationary (which would be consistent with a moving attractor) and in addition that their average values have much increased after the 1980s, while according to the theory the ‘normal’ unemployment should have become lower, owing to the weakening of pro-labor institutions. Hence, although possibly not conclusive, the evidence is not supportive of the notion.
According to Shaikh, the long-term tendency towards the normal’ unemployment rate would be based on a strong connection between profitability and accumulation. Our results thus contribute to casting some doubts concerning the existence of such a strong and systematic connection, along with increasing empirical evidence that output growth is consistently a major determinant of aggregate business investments, while mixed results and relatively small impacts, if any, are reported concerning profitability and the cost variables. It is also partly on the basis of this evidence that in recent years the view of demand-led growth has prevailed, albeit in diverse analytical frameworks within a broadly defined post-Keynesian approach (for a survey, see Cesaratto, 2015; Lavoie, 2016). This is not to deny that persistently low or high unemployment rates can then set in motion reactions in the economic, social, and political spheres that will affect the labor market conditions and income distribution. However, the routes may be more complex than suggested by the profit-investment nexus, and may involve strategic and political actions, with no clear-cut and mechanical implications concerning the final outcomes and the time-spans involved.
Acemoglu, D. (2003). Labor‐and capital‐augmenting technical change. Journal of the European Economic Association, 1(1), 1–37.
Bassanini, A., & Manfredi, T. (2014). Capital’s Grabbing Hand? A Cross-industry Analysis of the Decline of the Labor Share in OECD Countries. Eurasian Business Review, 4, 3–30.
Bengtsson, E. (2014). Do unions redistribute income from capital to labour? Union density and labour’s share since 1960. Industrial Relations Journal, 45(5), 389–408.
Beqiraj, E., Fanti, L., & Zamparelli, L. (2019). Sectoral composition of output and the wage share: The role of the service sector. Structural Change and Economic Dynamics, 51, 1-10.
Blecker, R. A. (2016). The US economy since the crisis: slow recovery and Secular Stagnation. European Journal of Economics and Economic Policies: Intervention, 13(2), 203–214.
Brancaccio, E., Garbellini, N., & Giammetti, R. (2018). Structural labour market reforms, GDP growth and the functional distribution of income. Structural Change and Economic Dynamics, 44(C), 34–45.
Cesaratto, S. (2015). Neo-Kaleckian and Sraffian controversies on the theory of accumulation. Review of Political Economy, 27(2), 154–182.
Cynamon, B. Z., & Fazzari, S. M. (2016). Inequality, the Great Recession and slow recovery. Cambridge Journal of Economics, 40(2), 373–399.
Deakin, S., Malmberg, J., & Sarkar, P. (2014). How do labour laws affect unemployment and the labour share of national income? The experience of six OECD countries, 1970–2010. International Labour Review, 153(1), 1-27.
Deleidi, M. (2018). Post Keynesian endogenous money theory: A theoretical and empirical investigation of the credit demand schedule. Journal of Post Keynesian Economics, 41(2), 185–209.
Dünhaupt, P. (2017). Determinants of labour’s income share in the era of financialisation. Cambridge Journal of Economics, 41(1), 283-306.
Elsby, M. W. L., Hobijn, B., & Şahin, A. (2013). The Decline of US Labor Share. Paper prepared for the Brookings Panel on Economic Activity, September 2013.
Franzini, M., & Pianta, M. (2015). Explaining inequality. Routledge, Taylor and Francis Group. London and New York.
Girardi, D., & Pariboni, R. (2020). Autonomous demand and the investment share. Review of Keynesian Economics, 8(3), 428–453.
Girardi, D., Paternesi Meloni, W., & Stirati, A. (2020). Reverse hysteresis? Persistent effects of autonomous demand expansions. Cambridge Journal of Economics, 44(4), 835–869.
Guscina, A. (2006). Effects of globalization on labor’s share in national income. International Monetary Fund, Working Paper No. 06/294.
Hein, E. (2014). Distribution and growth after Keynes: A Post-Keynesian guide. Edward Elgar Publishing.
Hein, E. (2017). Financialisation and tendencies towards stagnation: The role of macroeconomic regime changes in the course of and after the financial and economic crisis 2007-9 (No. 90/2017). Working Paper, Institute for International Political Economy Berlin.
Hein, E. (2015). Finance-dominated capitalism and re-distribution of income: a Kaleckian perspective. Cambridge Journal of Economics, 39(3), 907–934.
Hein, E., & Vogel, L. (2008). Distribution and growth reconsidered: empirical results for six OECD countries. Cambridge Journal of Economics, 32(3), 479–511.
Hein, E., Paternesi Meloni, W., & Tridico, P. (2020). Welfare models and demand-led growth regimes before and after the financial and economic crisis. Review of International Political Economy, online first article, DOI: 10.1080/09692290.2020.1744178.
Hogrefe, J., & Kappler, M. (2013). The labour share of income: heterogeneous causes for parallel movements?. The Journal of Economic Inequality, 11(3), 303-319.
Hutchinson, J., & Persyn, D. (2012). Globalisation, concentration and footloose firms: in search of the main cause of the declining labour share. Review of World Economics, 148(1), 17-43.
IMF (2017). Understanding the downward trend in labor income shares. In: World Economic Outlook: Gaining Momentum?
Jayadev, A. (2007). Capital account openness and the labour share of income. Cambridge Journal of Economics, 31(3), 423–443.
Khotari, S. P., Lewellen, J., & Warner, J. B. (2014). The Behaviour of Aggregate Corporate Investment. MIT Sloan School, Research Paper No. 5112–14.
Kristal, T. (2010). Good times, bad times: postwar labor’s share of national income in capitalist democracies. American Sociological Review, 75(5), 729–63.
Lavoie, M. (2016). Convergence towards the Normal rate of capacity utilization in neo-Kaleckian models: the role of non-capacity creating autonomous expenditures. Metroeconomica, 67(1), 172–201.
Lazonick, W. (2014). Profits without prosperity. Harvard Business Review, 92(9), 46–55.
Naastepad, C. W. M., & Storm, S. (2006). OECD demand regimes (1960-2000). Journal of Post Keynesian Economics, 29(2), 211-246.
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. FMM Working Paper No. 10.
OECD (2012). Labour Losing to Capital: What Explains the Declining Labour Share?. In: OECD Employment Outlook 2012, 109–161.
Onaran, O. (2011). Globalisation, macroeconomic performance and distribution. In: Hein, E., & Stockhammer, E. (eds.). A Modern Guide To Keynesian Macroeconomics and Economic Policies. Cheltenham: Edward Elgar, pp. 240–266.
Özdemir, O. (2019). Financialization and the Labor Share of Income. Review of Economic Perspectives, 19(4), 265-306.
Palley, T. I. (2016). Inequality, the financial crisis and stagnation: competing stories and why they matter. Real World Economics Review, 74, 1–18.
Pariboni, R., & Tridico, P. (2019a). Structural change, institutions and the dynamics of labor productivity in Europe. Journal of Evolutionary Economics, online first article, DOI: 10.1007/s00191-019-00641-y.
Pariboni, R., & Tridico, P. (2019b). Labour share decline, financialisation and structural change. Cambridge Journal of Economics, 43(4), 1073-1102.
Pariboni, R., Paternesi Meloni, W., & Tridico, P. (2020). When Melius Abundare Is No Longer True: Excessive Financialization and Inequality as Drivers of Stagnation. Review of Political Economy, online first article, DOI: 10.1080/09538259.2020.1769282.
Paternesi Meloni, W., & Stirati, A. (2018). A short story of the Phillips curve: from Phillips to Friedman…and back?. Review of Keynesian Economics, 6(4), 493–516.
Pivetti, M. (1991). An essay on money and distribution. London, Macmillan.
Rodrik, D. (1997). Has Globalization Gone Too Far?. Washington, DC: Institute of International Economics.
Schoder, C. (2014). Effective Demand, Exogenous Normal Utilization and Endogenous Capacity in the Long Run: Evidence from a Cointegrated Vector Autoregression Analysis for the USA. Metroeconomica, 65(2), 298–320.
Lazonick, W. and O’Sullivan, M. (2000). Maximizing shareholder value: a new ideology for corporate governance. Economy and Society, 29(1), 13–35.
Shaikh, A. (2016). Capitalism: Competition, conflict, crises. Oxford University Press.
Sharpe, S. A., & Suarez, G. (2014). The Insensitivity of Investment to Interest Rates: Evidence from a Survey of CFOs. FEDS Working Paper No. 2014–2. Available at SSRN: https://ssrn.com/abstract=2976891.
Stirati, A. (1992). Unemployment, institutions and the living standard in the classical theory of wages. Contributions to Political Economy, 11(1), 41–66.
Stirati, A. (2001). Inflation, unemployment and hysteresis: an alternative view. Review of Political Economy, 13(4), 427–451.
Stirati, A. (2016). Real wages in the business cycle and the theory of income distribution: an unresolved conflict between theory and facts in mainstream macroeconomics. Cambridge Journal of Economics, 40(2), 639–661.
Stockhammer, E. (2008). Is the NAIRU theory a monetarist, new Keynesian, post Keynesian or a Marxist theory?. Metroeconomica, 59(3), 479–510.
Stockhammer, E. (2009). Determinants of functional income distribution in OECD countries. IMK Studies 05–2009.
Stockhammer, E. (2013). Wage-led Growth: An equitable strategy for economic recovery. Springer.
Stockhammer, E. (2017). Determinants of the wage share: a panel analysis of advanced and developing economies. British Journal of Industrial Relations, 55(1), 3–33.
Storm, S. (2017). The new normal: Demand, secular stagnation, and the vanishing middle class. International Journal of Political Economy, 46(4), 169-210.
Tori, D., & Onaran, O. (2017). The effects of financialisation and financial development on investment: Evidence from firm-level data in Europe. Greenwich Papers in Political Economy, No. 44.
Tridico, P., & Paternesi Meloni, W. (2018). Economic growth, welfare models and inequality in the context of globalisation. The Economic and Labour Relations Review, 29(1), 118–139.
van Treeck, T. (2009). The political economy debate on ‘financialization’ – a macroeconomic perspective. Review of International Political Economy, 16(5), 907–944.
van Treeck, T. (2015). Inequality, the crisis, and stagnation. European Journal of Economics and Economic Policies: Intervention, 12(2), 158–169.
 There are also other related factors, acting in a similar manner, which this body of literature has considered among the drivers of the decline in the labor share. Among them: industry concentration and the consequent rising market power; the emergence of the innovative ‘superstar firms’; the transition toward a more capital-intensive economy focused on intellectual property; the falling relative price of investment goods and technological progress; the process of automation.
 See among others Van Treeck (2009) and Tori and Onaran (2017).
 See Lazonick and O’Sullivan (2000); OECD (2012); Lazonick (2014); van Treeck (2015); Blecker (2016); Palley (2016); Hein (2017).
 See Onaran (2011); Stockhammer (2017).
 See Pivetti (1991), and for a different formulation Hein (2014; 2015).
 See Stirati (2001) and Paternesi Meloni and Stirati (2018).
 The variability of the unemployment rate can be limited since a persistent lack of employment opportunities may induce adjustments on the supply side, while sustained labor demand can stimulate participation and reduce underemployment. Enlarged indicators, like the index of unemployment intensity proposed by Shaikh (2016) or other multidimensional measures, can be used to represent the labor market slack.
 Note that although institutions appear to have a role in many mainstream models of wage and employment determination, in fact they only determine equilibrium employment and unemployment, while wages are determined by labor marginal product at equilibrium employment and the mark-up determined in consumption goods market (e.g., with constant marginal product, changes in institutions would not alter income distribution but only the equilibrium non-inflationary unemployment rate).
 We also test differences according to institutional models by grouping Anglo-Saxon countries on the one hand and European on the other. The results confirm the expected higher impact of unemployment on the wage share in the former group; though as we shall see there is also wide within-group heterogeneity as well as similarities across institutional models.
 Very speculative considerations might be advanced in this regard. The first is that both countries differ from other European mature economies for not having much ‘active labor market policies’: these most often involve training activities that (somewhat fictitiously) interrupt the measured unemployment spells of individuals; it may thus be the case that the duration measure is altered in other countries by these activities and does not fully capture the degree of labor market slack. The second one is that perhaps in both countries there are sections of the labor force insisting in very difficult labor markets where unemployment duration can become very high (the southern regions, in the case of Italy).
 An increasing body of empirical literature on the determinants of aggregate investment finds large accelerator effects together with a low sensitivity of business investment to profitability and interest rate changes. Khotari et al. (2014), Schoder (2014), Girardi and Pariboni (2020) and Girardi et al. (2020) assessed the positive role of aggregate demand, and particularly its autonomous components, in shaping the process of capital formation. Moreover, several works belonging to the wage/profit-led literature that estimates an investment function found poor or not significant profitability effects and larger GDP effects (Naastepad and Storm, 2006; Hein and Vogel, 2008; Obst et al., 2017). The insensitivity of investment to interest rates has been documented by Sharp and Suarez (2014), while Deleidi (2018) finds no significant relationship between loans granted to enterprises and the corresponding interest rates.