Whatever happened to the "Goodwin pattern"?

Setterfield, Mark Working Paper Whatever happened to the "Goodwin pattern"? FMM Working Paper, No. 64 Provided in Cooperation with: Macroeconomic Policy Institute (IMK) at the Hans Boeckler Foundation Suggested Citation: Setterfield, Mark (2021) : Whatever happened to the "Goodwin pattern"?, FMM Working Paper, No. 64, Hans-Böckler-Stiftung, Macroeconomic Policy Institute (IMK), Forum for Macroeconomics and Macroeconomic Policies (FMM), Düsseldorf This Version is available at: https://hdl.handle.net/10419/246829 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/legalcode FMM WORKING PAPER No. 64 • April 2021 • Hans-Böckler-Stiftung WHATEVER HAPPENED TO THE 'GOODWIN PATTERN'? Mark Setterfield 1 ABSTRACT The 'Goodwin pattern' – an anti-clockwise rotation in real activity x wage share space recurring at intervals that correspond roughly to the duration of business cycles – is an enduring feature of high-frequency dynamics in capitalist economies. It is well known that the centre or focus of this rotation shifts over time. More recently, however, the Goodwin pattern seems to have broken down, the wage share no longer increasing as the real economy improves over the course of short-term booms. In this paper, the breakdown of the Goodwin pattern is associated with the consolidation of an `incomes policy based on fear' that is partand-parcel of neoliberalism. As a result of this incomes policy based on fear, the institutional structure of the labour market disciplines labour at any rate of unemployment. This decouples wage-share dynamics from the state of the real economy, with the result that as recently witnessed, the wage share is rendered invariant to tightening of the labour market in the course of short-term cyclical booms. ————————— 1 Professor of Economics, New School for Social Research, New York, email [email protected]; FMM fellow. Whatever happened to the ‘Goodwin pattern’? Mark Setterfield∗ January 14, 2021 Abstract The ‘Goodwin pattern’ – an anti-clockwise rotation in real activity ×wage share space recurring at intervals that correspond roughly to the duration of business cycles – is an enduring feature of high-frequency dynamics in capitalist economies. It is well known that the centre or focus of this rotation shifts over time. More recently, however, the Goodwin pattern seems to have broken down, the wage share no longer increasing as the real economy improves over the course of short-term booms. In this paper, the breakdown of the Goodwin pattern is associated with the consolidation of an ‘incomes policy based on fear’ that is part-and-parcel of neoliberalism. As a result of this incomes policy based on fear, the institutional structure of the labour market disciplines labour at any rate of unemployment. This decouples wage-share dynamics from the state of the real economy, with the result that as recently witnessed, the wage share is rendered invariant to tightening of the labour market in the course of short-term cyclical booms. JEL codes: E11, E12, E25, E64 Keywords: Goodwin pattern, distributional conflict, worker insecurity, incomes policy based on fear 1 Introduction According to a substantial empirical literature, capitalist economies are characterized by short-cycles that involve a common empirical signature: the ‘Goodwin pattern’. The Goodwin pattern describesan anti-clockwiserotation in real activity×wageshare space, recurring at intervals that roughly correspond to business cycle frequency. Numerous explanations have ∗Professor of Economics, New School for Social Research; email [email protected]. 1 been forwarded for this key stylized fact of high-frequency capitalist macrodynamics.1More recently, however, attention has focused on the possibility that the Goodwin pattern has broken down. While the centre or focus of the Goodwin pattern has long been understood to shift between discrete historical periods, the apparent breakdown of the Goodwin pattern is an altogether more contemporary phenomenon. The purpose of this paper is to document and furnish a theory of the recent breakdown of the Goodwin pattern. Central to the theory advanced is the notion that contemporary neoliberal capitalism is based on an ‘incomes policy based on fear’ (Cornwall, 1990) that substitutes institutionalized worker insecurity for the ‘threat of the sack’ and, in so doing, loosens the connection between variations in the rate of employment, wage bargaining, and the behaviour of the wage share. The remainder of the paper is devoted to the empirical basis for and development of this theory, and is organized as follows. Section 2 explores the evidence that has established the Goodwin pattern as a prominent stylized fact in advanced capitalist economies. Section 3 examines the evidence that this pattern has recently broken down. Section 4 provides an explanation for the breakdown of the Goodwin pattern in terms of the consolidation of an incomes policy based on fear arising from the institutional structure of the labour market in neoliberal capitalism. Finally, section 5 concludes. 2 An enduring stylized fact 2.1 What is the ‘Goodwin pattern’? The ‘Goodwin pattern’ is named for the famous Marxian model of cyclical growth developed by Goodwin (1967). It can be regarded as a basic stylized fact of advanced capitalist economies, describing a cyclical pattern of high-frequency co-movements between measures of 1The progenitor of these analyses is Goodwin (1967), but see also the qualitatively different models of Skott (1989); Barbosa-Filho and Taylor (2006); and Stockhammer and Michell (2016). For recent surveys of this literature, see Blecker and Setterfield (2019) and Flaschel et al. (2020). 2 real economic activity on one hand, and measures of the distribution of income on the other. Specifically, the Goodwin pattern is associated with either clockwise rotations in wage share ×real performance space, or alternatively (and equivalently) counter-clockwise rotations in real performance ×wage share space.2In other words – and without imputing any causeeffect relationships between the variables involved – as real economic performance improves, the wage share is observed to first fall then rise, and when real performance subsequently deteriorates, the wage share first continues to rise before subsequently falling back. It is generally accepted that the centre or focus of the rotation characteristic of the Goodwin pattern can shift, and that such shifts are evident in the historical record. In what follows, we treat this observation as being indicative of structural modifications to the basic workings of capitalism – in this case, the nature and outcomes associated with distributional conflict – that are brought about by the transition between different long-run phases of growth. Two particular phases of growth will be of paramount importance in what follows: the post-war Golden Age (1948-73) and the more recent Neoliberal Boom (19902007). By conceiving the Goodwin pattern as a high-frequency phenomenon occurring within low frequency phases of growth, our interpretation of this statistical regularity is similar to that of Mohun and Veneziani (2008), who see it as characteristic of shorter cycles taking place in the midst of longer-term trends associated with structural change in capitalist economies. 2.2 Empirical evidence Figure 1 provides evidence of the Goodwin pattern in the US economy from 1948–2019. Figure 1 is drawn in wage share ×real performance space (with the wage share on the abscissa and the employment rate on the ordinate), depicting clockwise rotations over time. 2As will become clear below, observation of the Goodwin pattern is robust to changes in the precise measure of real economic performance. 3 Three different complete business cycles are drawn to attention,3one of which (1960-69) is representative of Golden Age experience, the other two (1990-2000, 2000-2007) making up the Neoliberal Boom.4Two observations are merited based on the data presented in figure 1. First, there are extremes of unemployment (i.e., markedly lower than average employment rates) in 1975, 1982 and 2010, resulting from crises (1973-74, 1979-81, and 2007-09) that can be associated with the termination of phases of growth in the US economy (the Golden Age and the Neoliberal Boom, respectively). Second, apart from the fact that the patterns traced out over the course of each cycle are generally consistent with the clockwise rotation (in wage share ×real performance space) typical of the Goodwin pattern, it is notable that the rotation from 1960-67 is further to the right in figure 1 than that during either 1990-2000 or 2001-2007. This is consistent with the observation made previously, that the centre or focus of the Goodwin pattern shifts over time, being centred on different point values of the wage share and employment rate during different historical periods. These periods, in turn, correspond to longer-term structural changes associated with the transition between different phases of growth – in this case, the Golden Age and the Neoliberal Boom. Specifically, the leftward shift of the centre or focus of the Goodwin pattern is consistent with the deterioration of the wage share under neoliberalism as compared with the mid-twentieth century Golden Age. Moreover, the fluctuations in the employment rate associated with the Goodwin pattern after the 1960s are (as can be seen from figure 1) orders of magnitude larger than those observed in the data prior to the 1970s. This development is suggestive of a greater willingness to sacrifice employment in order to discipline labour and so control the wage share – what Balogh (1982), in his assessment of similar developments in the UK, termed the ‘incomes policy of Karl Marx’ (see also Setterfield (2006)). 3The dating of these business cycles is based on National Bureau of Economic Research (NBER) business cycle dating – see www2.nber.org/cycles. 4Notice that figure 1 also draws attention to US experience since the 2007-2009 financial crisis and Great Recession. We will return to discuss this experience in section 3. 4 89.0 90.0 91.0 92.0 93.0 94.0 95.0 96.0 97.0 98.0 54.0 56.0 58.0 60.0 62.0 64.0 66.0 Employment rate US Wage share US 1990 2000 1992 1997 2003 2007 2010 2019 1967 1960 1963 1965 1975 1982 Figure 1: The distributive cycle: US 1948–2019 Descriptive analyses of the data similar to that above, and providing evidence confirmatory of the existence of the Goodwin pattern in advanced capitalist economies, are widespread. For example, Flaschel and Groh (1995) and Harvie (2000) observe the Goodwin pattern over periods of 10-15 years in the data for samples of 8 and 10 OECD countries, respectively. Patterns in US macroeconomic data similar to those observed in figure 1 have been brought to attention previously by various authors. Using a data set similar to that on which figure 1 is based for the period 1948-2008, Zipperer and Skott (2011, pp.35-6) remark on the shift in the centre or focus of the Goodwin pattern in the US economy, first towards lower rates of employment in the 1970s and 1980s, and then towards higher rates of employment and lower values of the wage share thereafter. Barbosa-Filho and Taylor (2006, pp.389-91), meanwhile, focus on the relationship between the rate of capacity utilization (rather than the employment rate) and the wage share of income in the US economy from 1929-2002.5Despite changing the measure of real economic performance and examin5Note that for the purpose of demonstrating the relationship between real activity and income distribution characteristic of the Goodwin pattern, the capacity utilization rate can be regarded as a plausible substitute for the employment rate. While the capacity utilization rate and employment rate are correlated, however, 5 ing a longer period of time, Barbosa-Filho and Taylor (2006) once again observe the same clockwise rotations in the data that are observed in figure 1, and also draw attention to similar shifts in the centre or focus of this pattern.6The only exception they note is during the period 1944-50, when the pattern of rotation runs contrary to that associated with the Goodwin pattern, but this they explain in terms of simultaneous declines in the wage share and the rate of capacity utilization from their wartime peaks. Aside from the descriptive statistics illustrated and discussed above, there are various other sources of evidence suggesting that the Goodwin pattern is representative of higherfrequency cycles in advanced capitalist economies.7For example, Barrales-Ruiz and von Arnim (2017) use wavelet decomposition methodology to study cycles of different periods in US data since 1948. The authors find a consistent Goodwin pattern for cycles of 4 to 8 and 8 to 16 years in length, regardless of the measure of real economic activity used (the output gap, income-capital ratio, or employment rate). Building on this approach, and using a variety of different measures of both real activity and the wage share, Barrales-Ruiz et al. (2020) first show that when considered individually, all activity and wage share measures display similar high-frequency dynamics. Their wavelet analysis then confirms this dynamical coherence, and an analysis of leads and lags reveals that real activity leads the wage share at high frequencies, consistent with the Goodwin pattern. Meanwhile, McAdam et al. (2019) study data spanning more than a century for three advanced capitalist economies: the US (1898-2010), the UK (1856-2010) and France (18962010). They find that at high frequencies (2–4 and 4–8 years), the wage share has a negative impact on output growth. Elsewhere, using US wage share and employment rate data 1948-2016, Araujo et al. (2019) study the impulse-response functions of an estimated VAR the relationship between the two variables need not be straightforward (Skott, 2017). 6Barbosa-Filho and Taylor (2006) in fact report counterclockwise rotations in the data, but this is simply because it is presented in real performance ×wage share space. 7For a more extensive review of this literature, see BRM20. 6 model. They find that the wage share responds to shocks to the employment rate and the employment rate responds to shocks to the wage share in a manner consistent with the Goodwin pattern. The duration of these responses depends on the way in which the employment rate and wage share data is filtered, but in all cases the effects of the shock disappear within 3 to 5 years.8Similar results using VAR methodology are reported in Barrales-Ruiz et al. (2020). Based on the stylized facts presented in their own earlier work (Zipperer and Skott, 2011), Skott and Zipperer (2012, pp.293-4) conclude that there is strong evidence of counterclockwise movements in real activity ×wage share space for the US economy (and somewhat weaker evidence of the same pattern in other advanced capitalist economies) at roughly business cycle frequency, that is robust to changes in the precise definition and measurement of variables.9These same conclusions appear consistent with the variety of evidence reviewed above. That having been said – and perhaps inevitably, given the nature of empirical macroeconomics – the seeming ubiquity of the Goodwin pattern is not without qualifications and caveats. First, the Goodwin pattern is not evident in lower frequency cycles of the sort associated with the rise, decline and rise of the consecutive phases of growth. Hence McAdam et al. (2019) find that the wage share leads growth positively at frequencies of 32 years and above in the US, UK, and France, which observation is inconsistent with the countercyclical rotation in real activity ×wage share space characteristic of the Goodwin pattern. Set against this, those who consider the Goodwin pattern to be characteristic of real-world capitalism seldom (if ever) claim that the pattern can be found in low-frequency data. 8Araujo et al. (2019) discover that their results are sensitive to the choice of real performance variable, the effects of the wage share on real performance disappearing when the capacity utilization rate is substituted for the employment rate. They suggest that the goods market and labour market should not be considered equivalent, which leads them to a reformulation of Goodwin-type dynamics based on a three-way interaction between the goods market, labour market, and wage share. A similar reformulation can be found in Skott (1989, 2010), which contribution is discussed briefly below. 9Skott and Zipperer (2012) actually refer to clockwise movements in real activity ×profit share space, but this amounts to the same thing. 7 employment intensity over the course of the post-war period (Paternesi Meloni and Stirati, 2020). Secular increases in labour market slack would therefore appear to help explain the long-term decline in the wage share evident in figure 1. Our concern is with high-frequency variations in the wage share, however – and inspection of figures 3 and 4 reveals that by any measure, labour market slack varies over the course of the business cycle and fell markedly during the 2009-2019 recovery. On the face of it, then, there is still more or less slack in the US labour market over the course of the cycle, as a result of which there should be (in principle) correspondingly less or more pressure on the profit share. But as is evident from the contrast between the cyclical variation in unemployment and unemployment intensity in figures 3 and 4, and the now-acyclical behaviour of the wage share in figure 2, measures of labour market slack continue to vary at high frequency in a manner that the wage share no longer does. The fact that the wage share has now essentially ceased to respond to even the substantial rise in the employment rate (6 percentage points) registered during the 2009-2019 recovery suggests that some other structural explanation of the events depicted in figure 2 is required. 4 The disappearance of the Goodwin pattern: An explanation 4.1 The transition to neoliberalism According to its proponents, neoliberalism works by freeing the private sector from the fetters of the state and revitalizing the supply-side of the economy. In particular, the labor market is made ‘flexible’: disincentives to work (such as taxes and unemployment benefits) are lowered, and impediments to wage adjustment (such as trade unions and minimum wage legislation) attenuated or removed. The alleged result is a labour market that generates 14 more jobs, resulting in higher output and so a healthier economy overall. Even the increased inequality associated with the initial changes to the structure of taxes and social benefits are allegedly offset by aggregate output gains that eventually ‘trickle down’ to the less affluent. Macroeconomic performance did, in fact, improve in the US after 1990, as compared to the previous two decades of inter regnum following the demise of the Golden Age. As the first two rows of table 1 make clear, the first full business cycle of the neoliberal era (1990-2000) re-established the low average rates of unemployment and inflation last seen towards the end of the post-war ‘Golden Age’ (represented by the period 1960-73 in the first column of table 1). This improvement in macroeconomic performance did not occur because of the effects of neoliberalism as envisaged by its progenitors, however. Instead, changes to the supply-side of the economy during the 1980s constituted a process of ‘zapping labor’ (Harrison and Bluestone, 1988). Post-1980 changes in public policy that claimed to make labor markets more flexible, together with various contemporaneous changes in corporate behavior, all succeeded in markedly increasing worker insecurity. These changes included labor law ‘reforms’ that, among other things, made unionization by workers more difficult and de-unionization by firms easier (Block et al., 1996), resulting in a precipitous drop in the rate of unionization in the US, from a post-war peak of about 35%, to 20% by the early 1980s, to just 10% by 2019.18 Other changes included increases in ‘non-standard’ (i.e, part time and/or temporary) employment, that eroded the post-war norm of year-round, full-time work. This gave rise to under-employment in the form of ‘involuntary’ part-time/temporary employment, a counterpart to involuntary unemployment. The introduction of periodic ‘downsizing’ exercises by firms, meanwhile, created the credible threat of job loss independently of general economic conditions (Osterman, 1999). Worker insecurity was further enhanced by the emergence of a credible threat (on the part of firms) to relocate production between political jurisdic18The contemporary rate of unionization in the private sector is just 6%, union representation being much greater in the public sector. 15 tions. This threat of plant relocation began as a domestic phenomenon that witnessed the migration of industry within the US from traditional industrial areas in the north-east and mid-west to southern and south-western states. It then took on an international dimension thanks to the pro-corporate form taken by ‘globalization’, as encoded in various international trade agreements. The latter helped create an international institutional environment that makes little or no reference to labor standards, thus exposing US workers to competition from low-wage foreign workers – not only as a result of the greater internationalization of product markets, but also by enhancing the credibility of the threat of plant relocation. Together, these developments have transformed ‘trade’ from its mid-twentieth century form – a process of competition between spatially-distinct and geographically immobile firms based on product and process innovations – into a competition between political jurisdictions to attract footloose corporations, on the basis of tax reductions and denuded environmental and labour standards.19 The combined effects of these various facets of the process of zapping labor are evident in the final row of table 1. This reports an index of worker insecurity based on numerical measures of the various phenomena described above.20 The index of worker insecurity rises precipitously during the 1980s and 1990s, as the institutional architecture of the neoliberal economy emerged. The process of zapping labor described above made the neoliberal economy work by instituting an ‘incomes policy based on fear’ (Cornwall, 1990). In general, incomes policies are formal and/or informal institutions that frame and mediate aggregate wage and price setting behaviour in a manner that reduces conflict over income shares. The archetypal incomes policy is based on cooperation and conciliation, ameliorating distributional conflict 19This is what Palley (2019) calls ‘barge economics’, referring to former General Electric CEO Jack Welch’s quip that, ideally, all physical plant would be located on barges that would continuously float between political jurisdictions to immediately take advantage of any cost advantage, regardless of its source. 20Details of the calculation of this insecurity index can be found in Setterfield (2005) and Setterfield and Lovejoy (2006). 16 Table 1: US Macroeconomic Performance, 1960-2000 1960-73 1974-79 1980-89 1990-2000 Unemployment 4.9 6.8 7.3 5.6 Inflation 3.1 9.6 5.6 3.1 Wage share 57.4 59.1 58.4 57.7 Interest rate 4.77 7.67 9.97 5.24 Worker insecurity N/A 0.21 0.57 0.82 Source: Setterfield (2006) through consensus building. The neoliberal ‘incomes policy based on fear’ is, instead, a model of domination, in which conflict is ameliorated essentially by means of coercion – specifically, the imposition of capitalists’ preferred distributional settlement on workers (Setterfield, 2006, 2007). First, workers were disempowered by the institutional changes outlined above. This reduced their ability to bargain for nominal wage increases, lessening their capacity to either seek improvements in their standard of living or even increase wages in the face of rising prices in order to protect their existing standard of living – an outcome that, ceteris paribus, lowers the real wage and hence the wage share of income.21 Since any increase in nominal wages results in an increase in the costs of production that may be passed on in the form of higher prices, the disempowering of workers under the incomes policy based on fear also reduced underlying inflationary pressures in the US economy. Not only could workers not initiate an inflationary process (by bargaining for higher nominal wages in the face of corporate resistance to increases in the real wage), neither could they propagate an inflationary process by merely seeking to defend their standards of living against price increases resulting from, 21In the presence of labour-saving technological change, any failure on the part of workers to bargain for real wage increases that keep pace with the rate of increase of productivity will have similar effect, lowering the wage share of income. 17 for example, commodity price ‘shocks’.22 These developments are clearly seen in the second and third rows of table 1 where, following their rise during the 1970s, inflation and the wage share fell during the 1980s and 1990s. By relieving pressure on inflation, the incomes policy based on fear simultaneously reduced the need for the Federal Reserve to perform this function themselves, by resort to restrictive monetary policy interventions designed to depress aggregate demand, raise unemployment, and discipline workers into moderating wage claims. The incomes policy based on fear made this policy response un-necessary: the very structure of the labor market created by the process of zapping labor now disciplined workers, by creating job and/or income insecurity at any rate of unemployment, eliminating the need for high unemployment to perform this disciplinary task.23 This, in turn, freed the Federal Reserve to lower interest rates and allow unemployment to fall (and with it, incomes and hence profits to rise). Again, these developments are evident in table 1 (rows one and four), where marked declines in both interest rates and the rate of unemployment can be seen after 1990. In sum, the neoliberal restitution of Golden Age standards of macroeconomic performance evident in column five of table 1 (rows one, two, and four), where we see the return of a low inflation, low unemployment, and low interest rate economy in the US for the first time since the early 1970s, can be explained by the process of zapping labor evident in rows three and five of the same table, and its creation of an incomes policy based on fear. Taken together, these developments describe a process of ‘balancing the macroeconomic books on the backs of workers’ (Setterfield, 2006), wherein the costs of achieving improved macroeconomic performance are borne uniquely by working households. 22The increases in the price of oil resulting from the US invasion of Iraq in 2003 and the impact of hurricane Katrina on oil refining capacity located on the Gulf of Mexico in 2005 are salient examples of this latter process during the period of the Neoliberal Boom. 23Even Alan Greenspan, the former Chair of the Federal Reserve, was explicit about this process, referring in Congressional testimony to the ‘labor market fear factor’ gripping American workers as a reason for diminished inflation concerns at the Federal Reserve. 18 4.2 The transition to neoliberalism and recent shifts in the centre or focus of the Goodwin pattern Central to the account provided above is the effective substitution of institutional conditions for unemployment as the key ‘worker discipline device’ moderating distributional conflict and hence the wage share of income (and the accompanying rate of price inflation). To see this in more detail, consider the following system of equations: ˆw=µ(vW−v) (1) ˆp=φ(v−vF) (2) µ=h(e, I), he>0, hI<0 (3) where ˆwis the rate of growth of nominal wages, vWis the target wage share of workers, vis (as previously defined) the actual wage share, ˆpis the rate of price inflation, vFis the target wage share of firms, Irepresents institutional features of the labour market that create employment and/or income insecurity amongst workers, and the parameters µand φ denote the relative power of workers in the wage bargain and the relative power of firms in product markets, respectively.24 Equations (1) and (3) constitute a simplified form of the conflicting-claims model of inflation due originally to Rowthorn (1977). The simplifications arise from the omission of productivity growth and expectations as independent variables. Hence note that in equation (1), for example (and ceteris paribus), workers would need to increase nominal wages at a rate equivalent to the growth of labour productivity simply in order to maintain the current value of the wage share (much less advance it towards the 24Note that, for the sake of simplicity, vWis treated as being independent of eand Ieven as µis modelled as being endogenous to these variables. See, however, Setterfield and Lovejoy (2006) for discussion of the endogeneity of workers’ aspirations. Note also that, in what follows, we abstract entirely from changes in φ, vFand vWin order to highlight the particular importance of changes in the values of µfor the structure and performance of US capitalism over the past thirty years. It is, however, quite possible that changes in φ,vFand/or vWhave contributed to the macroeconomic outcomes analyzed in this chapter. 19 target value, vW), given that (as previously noted) v=ω q=w/p q. This concern is eliminated if, for the sake of simplicity, we assume that ˆq= 0.25 It is also possible that workers reference an expected rate of inflation when seeking wage increases in accordance with equation (1) given that ceteris paribus, and once again recalling that v=w/p q, they need to raise nominal wages in tandem with any increase in prices in order to maintain the current value of the wage share. Abstracting from concerns with expectations and productivity growth is worthwhile in the current context, however, because our objective here is to highlight the role of conflict in the determination of outcomes such as vand ˆpand, in particular, the changing influences on this conflict as a result of the transition of US capitalism towards neoliberalism. The underlying theory in equations (1) and (2) is that workers and firms care about achieving ‘fair shares’ of total income (reflected in the targets vWand vF), and strive to inflate nominal wages and prices (respectively) in the pursuit of these fair shares.26 Hence ˆwis increasing in vWand decreasing in vin equation (1), while ˆpis increasing in vand decreasing in vFin equation (2). Equations (1) and (2) combine to yield equilibrium values for the rate of inflation (ˆp∗) and the wage share (v∗) under the equilibrium condition ˆw= ˆp, which is necessary to ensure a constant wage share ( ˙v= 0) under the hypothesized condition ˆq= 0 (given that v=w/p q). These equilibria can be written as: ˆp∗=µ∗φ(vW−vF) µ+φ(4) v∗=µvW+φvF µ+φ(5) Note that in equation (5), the equilibrium wage share is a weighted average of the distribu25Note that this simplifying assumption also means that statements about the wage share are equivalent to statements about the value of the real wage and vice versa, since with ˆq= 0 any variation in ωis reflected in equal proportional variation in v, and vice versa. 26The wage share pursued by firms, νF, can be associated with a target value of the mark up (τF) in a simple mark-up pricing equation of the form p= (1 + τ)w/q, given that it follows from this equation that τF=1−νF νF. 20 tional targets vWand vF: as long as µ, φ 6= 0, neither workers nor capitalists are able to set v equal to their preferred, target values. Instead, equilibrium price inflation ˆp∗emerges as the residual outcome of the ‘balance of conflicting forces’ that keeps the functional distribution of income constant in equilibrium. Equation (3) relates the bargaining power of workers directly to the employment rate e=N/L, where Nis the level of employment (as previously defined) and Lthe size of the labour force, and indirectly to the degree of institutionalized worker insecurity, I. This equation captures a trade off between the rate of unemployment (1−e) and the degree of institutionalized worker insecurity necessary to maintain a constant level of worker bargaining power, µ. As will become clear below, this allows us to model the substitution of institutionalized worker insecurity for unemployment as a worker discipline device under neoliberal capitalism. Referring to equations (3), (4) and (5), consider first the derivatives: dˆp∗ de =dˆp∗ dµ dµ de =φ2(vW−vF) (µ∗+φ)2he>0 (6) and: dv∗ de =dv∗ dµ dµ de =φ(vW−vF) (µ∗+φ)2he>0 (7) These results establish the existence of ‘standard’ (price inflation) and wage-share Phillips curves, respectively: as the employment rate rises (unemployment rate falls) and the labour market tightens, the rate of inflation and the wage share of income both rise. Now consider the derivatives: dˆp∗ de =dˆp∗ dµ dµ dI =φ2(vW−vF) (µ∗+φ)2hI<0 (8) and: 21 dv∗ dI =dv∗ dµ dµ dI =φ(vW−vF) (µ∗+φ)2hI<0 (9) These results demonstrate the capacity for variations in institutionalized worker insecurity to ‘shift’ the standard and wage-share Phillips curves, so that as Irises, both the rate of inflation and the wage share of income fall at any given rate of employment (unemployment). Finally, consider the total derivatives: dˆp∗=∂ˆp∗ ∂e de +∂ˆp∗ ∂I dI (10) and: dv∗=∂v∗ ∂e de +∂v∗ ∂I dI (11) It is clear from inspection that because the partial derivatives in both (10) and (11) are of opposing signs (see equations (6)-(9)), we can find de, dI > 0 such that dˆp∗, dv∗= 0. This is the essence of the neoliberal supply side as described earlier, and its apparent restitution of Golden Age macroeconomic performance (seen in table 1) through the institution of an incomes policy based on fear (dI > 0). The result is illustrated in Figure 5. Suppose we begin at the equilibrium v1,ˆp1in the north-east quadrant of Figure 5. This puts the economy at points Aand A0(respectively) on the wage-share and standard Phillips curves WSPC1and SPC1in the south-east and north-west quadrants of the Figure. Ceteris paribus, a rise in institutionalized worker insecurity Iwill lower worker bargaining power (to µ2) and so lower the equilibrium wage share and inflation rate (to v2and ˆp2, respectively. But if accompanied by a simultaneous increase in e(and hence decrease in Ufrom U1to U2) sufficient to exactly offset the impact of the change in Ion µ, the equilibrium in the north-east quadrant of Figure 5 will be restored to its initial configuration. This will bring the economy 22 Figure 5: Restoring Macroeconomic Performance with an Incomes Policy Based on Fear 45o v U = 1 – e U = 1 – e ˆˆ ,wp U 1 U 1 U 2 U 2 v 1 v 2 v F v W μ 1 μ 2 1 ˆ p 2 ˆ p WSPC 1 WSPC 2 SPC 1 SPC 2 A B A' B' 23 have institutionalized worker insecurity in the US economy over the past thirty years. This institutionalized worker insecurity has effectively replaced unemployment as the key worker discipline device that moderates worker bargaining power, the resultant ability of workers to bargain for wage increases, and hence the behaviour of the wage share of income. The upshot of this mechanism is that the wage share has effectively become ‘decoupled’ from labour market outcomes: it is no longer necessary to raise unemployment in order to discipline labour and reduce the wage share, while tightening the labour market no longer reduces labour market discipline and increases worker bargaining power sufficiently to directly affect the wage share. This account of the modern dynamics of the US labour market is consistent with the observed breakdown of the Goodwin pattern after 2000. It also suggests that the breakdown in the Goodwin pattern can be explained in terms of a transformation in the structure of class conflict during a particular phase of growth (neoliberalism) – conflict itself remaining central to our account of the patterns evident in the data. One important lesson that emerges from this observation is that distributional conflict is an enduring feature of capitalism. Whether conflict is explicit (as, for example, during the strike waves of the early 1970s) or suppressed by institutional arrangements that codify a specific (either more or less equitable) balance of power among conflicting claimants, it remains central to capitalist dynamics. Awareness of this fact, coupled with an ability to account for the precise manifestation of conflict, are therefore vital for explaining outcomes during any particular phase of capitalist growth. References Araujo, R. A., M. J. D´avila-Fern´andez, and H. N. Moreira (2019). Some new insights on the empirics of Goodwin’s growth-cycle model. Structural Change and Economic Dynamics 51, 42 – 54. Balogh, T. (1982). The Irrelevance of Conventional Economics. New York: W.W. Norton. 30 Barbosa-Filho, N. and L. Taylor (2006). Distributive and demand cycles in the US economy: a structuralist Goodwin model. Metroeconomica 57(3), 389–411. Barrales-Ruiz, J., I. Mendieta-Mu˜noz, C. Rada, D. Tavani, and R. von Arnim (2020). The distributive cycle: Evidence and current debates. Working Paper 2020-07, Department of Economics, University of Utah. Barrales-Ruiz, J. and R. von Arnim (2017). Longer-run distributive cycles: wavelet decompositions for the US, 1948-2011. Review of Keynesian Economics 5(2), 196–217. Blecker, R. and M. Setterfield (2019). Heterodox Macroeconomics: Models of Demand, Distribution and Growth. Cheltenham, UK: Edward Elgar. Block, R. N., J. Beck, and D. H. Kruger (1996). Labor Law, Industrial Relations, and Employee Choice: The State of the Workplace in the 1990s. Kalamazoo, MI: W.E. Upjohn Institute for Employment Research. Cauvel, M. (2018). Three essays on the empirical estimation of wage-led and profit-led demand regimes. Ph. D. thesis, American University, Washington DC. Cornwall, J. (1990). The Theory of Economic Breakdown: An Institutional-Analytical Approach,. Oxford: Basil Blackwell. Flaschel, P. and G. Groh (1995). The classical growth cycle: Reformulation, simulation and some facts. Economic Notes 24, 293–326. Flaschel, P., H. Kr¨amer, S. Luchtenberg, C. P. no, and M. Setterfield (2020). Capitalism, Inclusive Growth, and Social Protection: Inherent Contradiction or Achievable Vision? Cheltenham, UK: Edward Elgar. Goodwin, R. (1967). A growth cycle. In C. Feinstein (Ed.), Socialism, Capitalism and Economic Growth, pp. 54–58. Cambridge: Cambridge University Press. Harrison, B. and B. Bluestone (1988). The great u-turn: corporate restructuring and the polarizing of America. New York: Basic Books. Hartley, B. (2019). Accounting for endogeneity and secular trends in aggregative studies of growth and distribution, new school for social research, mimeo. Harvie, D. (2000). Testing Goodwin: growth cycles in ten OECD countries. Cambridge Journal of Economics 24(3), 349–376. Komlos, J. (2019). Trends and cycles in U.S. labor-market slack, 1994-2019. Applied Economics Quarterly (formerly: Konjunkturpolitik) 65(3), 209–235. Lavoie, M. (2017). The origins and evolution of the debate on wage-led and profit-led regimes. European Journal of Economics and Economic Policies: Intervention 14(2), 200–221. 31 McAdam, P., S. Bridji, and M. Charpe (2019). Labor share and growth in the long run. Working Paper Series 2251, European Central Bank. Mohun, S. and R. Veneziani (2008). Goodwin cycles and the US economy 1948-2004. In P. Flaschel and M. Landesmann (Eds.), Mathematical Economics and the Dynamics of Capitalism: Goodwin’s Legacy Continued, pp. 107–130. London: Routledge. Nikiforos, M. (2017). Uncertainty and contradiction: An essay on the business cycle. Review of Radical Political Economics 49(2), 247–264. Osterman, P. (1999). Securing Prosperity: The American Labour Market: How It Has Changed and What to Do About It. Princeton, NJ: Princeton University Press. Pacitti, A. (2020). The cost of job loss, long-term unemployment, and wage growth. Journal of Post Keynesian Economics, 1–28. Palley, T. I. (2019). The fracturing of globalization: Implications of economic resentments and geopolitical contradictions. Challenge 62(1), 49–66. Paternesi Meloni, W. and A. Stirati (2020). Unemployment and income distribution: Some extensions of shaikh’s analysis, institute for new economic thinking working paper no. 137. Technical report. Rowthorn, R. E. (1977). Conflict, inflation and money. Cambridge Journal of Economics 1(3), 215–39. Setterfield, M. (2005). Worker insecurity and u.s. macroeconomic performance during the 1990s. Review of Radical Political Economics 37(2), 155–177. Setterfield, M. (2006). Balancing the macroeconomic books on the backs of workers: A simple analytical political economy model of contemporary U.S. capitalism. International Journal of Political Economy 35 (3), 46–63. Setterfield, M. (2007). The rise, decline and rise of incomes policies in the US during the postwar era: an institutional-analytical explanation of inflation and the functional distribution of income. Journal of Institutional Economics 3(02), 127–146. Setterfield, M. and T. Lovejoy (2006). Aspirations, bargaining power, and macroeconomic performance. Journal of Post Keynesian Economics 29(1), 117–148. Shaikh, A. (2016). Capitalism: Competition, Conflict, Crises. Oxford: Oxford University Press. Skott, P. (1989). Conflict and effective demand in economic growth. Cambridge, UK: Cambridge University Press. 32 Skott, P. (2010). Growth, instability and cycles: Harrodian and Kaleckian models of accumulation and income distribution. In M. Setterfield (Ed.), Handbook of Alternative Theories of Economic Growth. Cheltenham: Edward Elgar. Skott, P. (2017). Weaknesses of ’wage-led growth’. Review of Keynesian Economics 5(3), 336–359. Skott, P. and B. Zipperer (2012). An empirical evaluation of three post-Keynesian models. European Journal of Economics and Economic Policies: Intervention 9(2), 277–307. Stockhammer, E. and J. Michell (2016). Pseudo-Goodwin cycles in a Minsky model. Cambridge Journal of Economics 41(1), 105–125. Zipperer, B. and P. Skott (2011). Cyclical patterns of employment, utilization, and profitability. Journal of Post Keynesian Economics 34(1), 25–58. 33 Imprint Publisher Macroeconomic Policy Institute (IMK) of Hans-Böckler-Foundation, Georg-Glock-Str. 18, 40474 Düsseldorf, Contact: [email protected], https://www.fmm-macro.net FMM Working Paper is an irregular online publication series available at: https://www.boeckler.de/de/fmm-working-paper-22457.htm The views expressed in this paper do not necessarily reflect those of the IMK or the Hans-Böckler-Foundation. ISSN 2512-8655 This publication is licensed under the Creative commons license: Attribution 4.0 International (CC BY). Provided that the author's name is acknowledged, this license permits the editing, reproduction and distribution of the material in any format or medium for any purpose, including commercial use. The complete license text can be found here: https://creativecommons.org/licenses/by/4.0/legalcode The terms of the Creative Commons License apply to original material only. The re-use of material from other sources (marked with source) such as graphs, tables, photos and texts may require further permission from the copyright holder.