Profit-Led Growth,Social Democracy,and the Left: An Accumulation of Discontent

1. Introduction

In the quarter-century since its initial publication, the Amit Bhaduri–Stephen Marglin (B-M; 1990) model, though subject to many extensions and amendments, has assumed a central role in the attempts of post-Keynesians to understand the interaction of distribution and accumulation. For its original authors, the model was motivated by the desire to formally identify the distinctive institutional features that yielded the postwar accumulation regime of the 1950s and 1960s wherein sustained accumulation coincided with marked increases in real wages and productivity. In their analysis, the viability of social democracy had depended on the persistence of a stagnationist accumulation regime wherein the growth of effective demand prompted by rising real wages would act as the overwhelming motive for continued investment, more than outweighing the fall in the profit share of national income. Such conditions were, nevertheless, far from guaranteed as varieties of capitalism might instead assume the form of “exhilarationist” regimes in which wage-led growth would be an impossibility. It was therefore a mistake to see wage-led cooperative capitalism as anything more than a transitory regime. The redistributive policies of the Left could only be persistently viable if the linkage between accumulation and the profit share of national income was severed.

The present paper seeks to better locate the B-M model as an historical outcome of the Left’s internal disputes over the prospects for social democracy. In better contextualizing the B-M model as a historical response to the perceived political economic failings of the social compromises upon which the growth of postwar advanced capitalist economies had rested, both the model’s popularity and its potential limitations can more easily be understood. Although the B-M framework has frequently come to be referred to as the neo- or post-Kaleckian growth model, such labels perhaps obscure the model’s diverse ancestry. The model constituted an attempt to reconcile seemingly incompatible theoretical perspectives, and to highlight those special conditions that made possible a “Golden Age” of social democracy.

2. Kalecki and Robinson

Reservations regarding the transformative potential of social democratic movements are deep-seated within the Left and, of course, well antedate the theoretical projects of J. M. Keynes and Michał Kalecki. Still, the contours of this debate were irreversibly altered by the “Keynesian Revolution,” and by the generation of authors that sought to extend the principle of effective demand beyond the short-period. This theoretical revolution highlighted the potential compatibility of sustained growth with a rising wage share in national income. In dethroning savings from its previously hallowed position, Keynesian theory seemed to offer a potential class compromise in which income redistribution and lowered rates of interest were the prices to be paid if capitalism was to retain its potential for dynamism and growth.

As is well known, Kalecki saw the maintenance of full-employment as a relatively simple technical problem so long as the State stood ready to meet the shortfall of private investment. The difficulty, with respect to maintaining full-employment, was political as business leaders and rentiers would rebel against lasting conditions that challenged their extra-economic power. ¹ For Kalecki, investment in the present was governed only superficially by the rate of profit, as this rate was, in turn, the result of past investment choices on the part of the capitalist class. A redistribution of real income toward workers, effected through nominal wages increase, while it might prove tenuous and fleeting in a world of oligopolistic pricing, would not necessarily depress the rate of accumulation, provided that the economy operated below full-capacity utilization, and was not heavily export-dependent. The principle impediment to the realization of full-employment accompanied by steady growth was not technical put political. The assumption that excess capacity would persist in the long-period was then a crucial feature of all of Kalecki’s work. Indeed his well-known comment that “the long-run trend is but a slowly changing component of a chain of short-period situations” can be understood in this way (Kalecki 1971: 165).

With this understanding in mind, the influence of Kalecki’s own approach on the B-M model appears more slight than is commonly appreciated. Rather than being colored by Kalecki, the modern post-Kaleckian approach arguably draws far more from Joan Robinson’s perspective. The decisive features of Robinson’s approach concern the treatment and formal modeling of the investment decision, and the related assumption that productive capacity would be fully utilized in the long-period. This model utilizes what is now known as the Cambridge Equation, the proposition that accumulation at any given moment in “logical time” serves to determine income distribution through the rate of profit. Moreover, Robinson (1962: 46–47) assumes that competition in the short-period leads firms to produce at “normal capacity” in the long-period, a point of production beyond which a “seller’s market prevails and capacity is being strained.” Investment in the present is then dependent upon the rate of profit firms expect to obtain in the future. Although in tranquil conditions the currently prevailing rate of profit might serve as the best proxy for the expected rate, investment is ultimately governed by animal spirits. Trend growth is thus determined by “the propensity to accumulate inherent in the system. It is steady or fluctuating according as it operates in tranquil conditions which generate inertia, or in a chancy world where uncertainty makes expectations volatile” (Robinson 1962: 87). For a given propensity to consume out of profits, and in the absence of savings by workers, the rate of growth serves to realize the profit rate expectations that originally motivated accumulation (Robinson 1962: 71). Growth is thus led by expectations as to the rate of profit. At full-employment, optimistic turns of expectations will raise the desired rate of accumulation, pushing against the capacity barrier at which the economy operates. Thus, through rising prices, and the consequent fall in real wages the rate of profit is enabled to rise in the next moment of logical time.

As was her habit, Robinson’s formal models are littered with reservations about the applicability of such analysis to actually existing economies. In her attempt to summarize the varying perspectives on growth, she concludes that “[t]hese models are all too much simplified and too highly integrated for it to be possible to confront them with evidence from reality” (Robinson 1962: 87). Thus, Robinson seems generally to have pursued these models as deductive thought experiments not susceptible to empirical testing. Certainly, any plain dismissals of wage-led growth as viable political strategy are absent.

3. The Stagnationist Tradition

The roots of the stagnationist perspective on capitalist development are not then directly found in Kalecki and Robinson’s own work. A comprehensive account of all those perspectives that have claimed for capitalism a tendency toward stagnation is beyond the scope of this paper, but in its modern guise, the most consistent and vocal champions of this perspective have been Paul Sweezy and his partisans of the Monthly Review School.

Perhaps no American author was better equipped to weigh the implications of Keynesian theory for neo-Marxist analysis than was Paul Sweezy. Judged by his Theory of Capitalist Development (1942), and by his role as a central contributor to, and editor of, the Monthly Review, Sweezy arguably did more than any other single author to shape the course of neo-Marxist political economy in the twentieth century. The general tenor of Sweezy’s commentary on Keynesian economics throughout his career deviated little. Keynes’ ideas represented the reawakening of bourgeois political economy from nearly a century of vulgarity and intellectual stagnation, while retaining inherent limitations. In Sweezy’s telling, the crucial advance of Keynesian theory was the willingness to see long-run stagnation as a normal outcome of the process of capitalist development that marked it as a clear advance. As a body of economic analysis, Sweezy’s objections to Keynesian theory were quite limited. As he notes, “[g]enerally speaking their logical consistency cannot be challenged, either on their own ground, or on the basis of the Marxian analysis of the reproduction process. The critique of Keynesian theories of liberal capitalist reform starts, therefore, not from their economic logic but rather from their faulty assumption about the relationship. . . between economics and political action” (Sweezy 1942: 348–49).

In developing his own account of stagnation, Sweezy drew liberally from Alvin Hansen’s (1938) concept of secular stagnation. ² Sweezy (1942) first details how the initial process of industrialization and the establishment of new industries absorbed enormous volumes of savings without immediately yielding chronic overcapacity. With industrialization having largely run its course, it had become “difficult even to imagine a series of new industries that would have the same relative importance comparable to. . . [those] of the eighteenth and nineteenth centuries” (Hansen 1938: 219). Thus, the failure of new industries to emerge at a sufficiently rapid rate and the slowing of population growth suggested a drift toward chronic depression in the absence of new and considerable countervailing forces. In sum, Keynesians had acknowledged that capitalism was subject to recurrent crises and that capitalism’s internal dynamics were capable of producing long-run stagnation. Keynesians were guilty of a certain political naiveté, but had not misapprehended the process of accumulation and growth.

The contributions of Josef Steindl (1952) were also of central importance to the evolution of the stagnationist tradition. Like Kalecki, Steindl saw the widespread and seemingly permanent emergence of oligopolistic firms as one of the decisive transformations of the postwar era. The long-term reduction in competitive pressures naturally brought with it comparative decline in investment. In treating the investment decision, however, Steindl emphasized the role of capacity utilization and treated the expected rate of profit as a variable of secondary importance. In his view, absent the intervention of activist fiscal and redistributive policies, rates of investment would inevitably decline, and growth would stagnate. For Steindl, the relative prosperity of advanced capitalism’s postwar Golden Age was the product of policy choices that recognized a substantial role for the state in ensuring full-employment and growth.

Crucially, Steindl saw the eclipse of the Golden Age that began in the late-1960s as, first and foremost, the product of policy choices. Following Kalecki’s predictions, full-employment policies had “led to a growing resentment of workers’ claims. . . [and] to complaints about work discipline” on the part of business. The consequent opposition of business leaders to a continuation of full-employment policies had produced instead a “policy of stagnation” (Steindl 1979: 8–9). For Steindl, no fundamental change in the “objective circumstances” governing growth and employment had occurred. The slowing of growth in the 1970s was a counterreaction, effected through a reversal of full-employment policies, against the political power of workers and trade unions. Either a long-run distributional shift leading to an increase in the wage share of national income or a full-fledged return to full-employment policies on the part of governments would be necessary to combat economic malaise (Steindl 1979: 13). In their absence, mature capitalism’s underlying tendency toward stagnation was sure to reappear.

4. The Rise of the Profit Squeeze Narrative

The new heralds of crisis saw within the postwar compromise the seeds of its own demise. The relative successes of labor, though they might have fueled a temporary consumption boom, could not be sustained in the increasingly competitive, globalized world capital had made for itself. Prominent in shaping this narrative on the Left were Andrew Glyn and Bob Sutcliffe. To the readers of the New Left Review (1971), and subsequently to a broader audience in British Capitalism, Workers and the Profit Squeeze (1972), Glyn and Sutcliffe announced that the British economy stood on the doorstep of crisis. Undoubtedly, this was a daring proclamation amid modest unemployment and the United Kingdom’s surging growth of the early 1970s. Concealed behind this façade, they held, was a marked fall in the share of profits in national income over the latter half of the 1960s.

The ebbing of the profit share and the accompanying fall in the rate of profit implied, first and foremost, a slower pace of accumulation. In their account, current profitability altered investment decisions “by influencing expectations about the profitability of [future] investment or through their role as the major normal source for the finance of investment” (Glyn and Sutcliffe 1971: 14). Thus, while rise in the wage share of national income was the “decisive advance” made by the working class in the twentieth century, for capitalism, “the continuation of [these] trends for a few more years would be catastrophic” (Glyn and Sutcliffe 1971: 27). A “Golden Age” of full-employment growth was an impossibility as “[t]here is no economic solution to the crisis in which the interests of capital and labour can be satisfied at once” (Glyn and Sutcliffe 1972: 200). British workers were thus faced with a choice to accede to the demands of capital accumulation lest it grind to a halt, or to pursue politically a socialist transformation.

Glyn and Sutcliffe’s early contributions were decidedly anglocentric, ³ but their contentions were soon to be generalized. In a lead article of the Monthly Review, Radford Boddy and James Crotty (1974) both restated and generalized the profit squeeze thesis articulated by Glyn and Sutcliffe. In their account, Keynesian theory had renewed faith in the possibility of full-employment growth under capitalism, a possibility that Marx had squarely dismissed. As the achievement of such growth had become Keynesians’ central aim, they had necessarily “glossed over the importance of the full-employment profit squeeze” (Radford Boddy and James Crotty 1974: 3). They rightly note that both Kalecki and Steindl had generally dismissed the possibility of a profit squeeze, suggesting instead that the costs entailed by rising wages would be passed through. Heightened international competition, manifest in the erosion of US trade surpluses, rendered capitalists’ inflationary accommodation of wage gains problematic. Mustering data from the United States to support their case, the apparent cyclicality of the wage share, along with the seeming rise in unit labor costs in the tail-end of business cycle expansions, lent considerable doubt to the Kalecki–Steindl proposition. Rising wages were not the only cause of the spike in unit labor costs during expansions, as it was also suggested that productivity growth would slow due in part to “an increasingly obstreperous labor force” (Radford Boddy and James Crotty 1974: 9). These dual mechanisms of the profit squeeze would play a central role in later models.

5. Rational Choice Marxism

By the close of the 1970s, profound skepticism with respect to the economic viability of social democracy was also manifest in the emergent tradition of Analytical Marxism. In a series of sweeping accounts of the evolution of social democratic movements, work that eventually coalesced into Capitalism and Social Democracy (1985), Adam Przeworski did much to shape the group’s perspective. In his telling, whatever its electoral successes, social democracy had failed to identify itself with any coherent body of economic theory prior to the “Keynesian Revolution” of the 1930s. Keynesian theory suggested to social democrats not only that they could mitigate the cycle through countercyclical policies, but also, crucially, that higher wages might drive growth in capitalist economies rather than impeding it. This theoretical alliance implied, for Przeworski, that social democracy lost its previous character as a reform movement based upon cumulative progress toward socialism, preferring instead to make the best of it under capitalism.

Przeworski’s further contention was that Keynesian social democratic compromises were inherently unsustainable. In collaboration with Michael Wallerstein, Przeworski (1982) developed a relatively simple formalization of this claim. Adopting a simplified Kaldor–Pasinetti model, capital accumulation depended upon capitalists’ savings out of profits, and thus, the rate of growth was determined solely by the rate of profit and capitalists’ saving rate, assuming that the productivity of capital was constant. Writing at the outset of the 1980s, the return to a model of accumulation in which full-employment and capacity utilization were presumed represented a curious choice, and one that implied predictable results with respect to the impact of redistribution on accumulation. Within this framework, however, Przeworski and Wallerstein argued that continued robust rates of accumulation hinged on a tenuous class compromise that was unlikely to be sustained in the presence of any meaningful measure of uncertainty.

The Left of the mid-1980s seemingly lacked the maturity and introspection to recognize that the old stalwart policies of income redistribution and social spending had become ineffective. As political strategy, it “represent[ed] a reaction of clinging to old ideas and policies that the Right claim[ed], with some justification, [had] been tried and found wanting” (Przeworski 1985: 206). Keynesian economics was, unfortunately, mistaken in dismissing savings as a meaningful constraint on growth. The difficulty in their minds was that Keynesian economics was merely the economics of the short-run. Stepping beyond the short-period, “when the economy is close to full employment the measures meant to increase aggregate demand and therefore decrease aggregate savings have the effect of limiting the rate of growth of potential output” (Przeworski 1985: 212).With the problem of effective demand discarded by assumption, long-run growth was inescapably profit-led. Thus, for Przeworski and Wallerstein, recognizing the flaws of the social democratic project was directly tied to an examination and rejection of the Keynesian–Kaleckian theoretical revolution.

6. The Left in Retreat

Such a wholesale rejection of the principle of effective demand in the long run was not so quickly forthcoming among Left economists. Nevertheless, as the decade of 1980s was ushered in by the electoral triumphs of conservative forces, assaults on organized labor, and extraordinarily tight monetary policy throughout much of the advanced capitalist world, the Left found itself on the defensive. Within this domain, the 1980s witnessed the development of two essential literatures that would form the basis of the B-M synthesis. One of these developments centered upon formal extensions of the stagnationist perspective. Led by Bob Rowthorn’s (1981) model, this literature provided a reconciliation between Kalecki and Steindl’s variants of the investment function, rendering investment a function of both of the current rate of profit and the rate of capacity utilization. The crucial point for our purposes is that within these models, only a wage-led growth regime is possible. Following Rowthorn, the mid-1980s saw the introduction of a number of explicitly stagnationist models that essentially retained the basic investment function proposed by Rowthorn, and extended it to consider the effects of fiscal and monetary policy, as well as foreign trade. Minor variations notwithstanding, these models were alike at the theoretical level in that they assumed that a redistribution toward wages would have a positive impact on the rate of growth, and appear far closer to the spirit of Kalecki and Steindl’s original contributions.

The 1980s also saw the maturation of the social structure of accumulation (SSA) paradigm, most prominently developed by Samuel Bowles, David M. Gordon, and Thomas E. Weisskopf. Through repeated collaborations, these authors presented what they termed a “Marxian ‘supply-side’ interpretation” of twentieth-century capitalism’s long-waves of accumulation and eventual crisis wherein the corporate rate of profit was the “fundamental underlying determinant of accumulation and growth” (Bowles, Gordon, and Weisskopf 1987: 43–44). Contributors to the SSA paradigm explicitly viewed their approach as an alternative to the stagnationist tradition and were eager to point out what they viewed as its empirical shortcomings. ⁴ In their view, the crisis of the 1930s was, in a broad sense, the result of “the capitalist class being too strong and the demand for goods and services being insufficient as a result.” Such conditions fortuitously allowed for the interests of capital and labor to align, and allowed for the success of “Keynesian and social democratic policies which. . . promised to redistribute income to labor, farmers, and other non-capitalist groups and [to] thereby stimulate demand” (Bowles, Gordon, and Weisskopf 1987: 55). Such an alliance could not, however, prove to be a durable model of growth. Taking the US economy as their central focus, the erosion of corporate profitability that began in the late-1960s stemmed from three basic institutional and political shifts: a heightened degree of international competition, a growing recalcitrance of labor to the demands of management fostered by low rates of unemployment, and the mounting cost burdens of environmental and safety regulations (Bowles, Gordon, and Weisskopf 1983: 79–97). In short, the crisis of the 1970s and early-1980s was “the type of supply-side crisis which results initially when the capitalist class is ‘too weak’” (Bowles, Gordon, and Weisskopf 1987: 55).

In its early incarnations, these arguments were presented as an historical narrative rather than as formal mathematical models of accumulation and growth. Subsequently, the research agenda evolved to include econometric estimations of the determinants of corporate profitability (Bowles, Gordon, and Weisskopf 1986). To their authors, these estimates offered meaningful confirmation of their historical narrative. Furthermore, in collaboration with Robert Boyer, Bowles (1988) offered a model of employment wherein investment was eventually constrained by a high employment profit squeeze. This first model, which made explicit use of the dichotomy between wage- and profit-led employment regimes, made only indirect policy suggestions. In a subsequently published elaboration of the model (Bowles and Boyer 1990), its larger intent was made far clearer. There, they noted that one important result of their model was that it “yield[ed] classical results—lower wages going along with higher levels of activity—even under rather extreme ‘Keynesian’ assumptions concerning savings, investment, and exports, and the effects of government borrowing. . . [and] suggesting the inherently contradictory nature of the social democratic full-employment program” (Bowles and Boyer 1990: 210).

In objecting to the Kalecki-Steindl investment function of the stagnationists, Bhaduri and Marglin (1990) were not then simply offering a technical amendment that allowed capacity utilization and profitability to exercise independent effects, but also formalizing nearly two decades of Left critiques of the long-run feasibility of Keynesian social democracy. Indeed in earlier work, Marglin (1984) had already mirrored the practical conclusions of the SSA authors discussed above. There he concluded that “[a] Left program. . . must respect the logic of the economic situation. . . As long as profitability remains the mainspring of investment, there are economic limits that constrain the wage share. Under capitalism, profits are indeed the geese that lay the golden eggs” (Marglin 1984: 142).

7. Conclusions

While Marglin and Bhaduri defined the terms of much future debate in post-Keynesian growth theory, the model was also the outcome of a generation of Left social scientists that had become deeply skeptical of the possibility of egalitarian redistribution under capitalism, and of the practical ambitions of Keynesian and social democratic parties. In their view, social democracy could only remain an economically viable form of class compromise under special conditions. In the absence of such conditions, the tenuous political compromises of cooperative capitalism would inevitably crumble. Subsequently, the model has given rise to countless empirical efforts to identify contemporary growth regimes as wage- or profit-led, increasingly removed from the political context in which the model arose. The B-M model’s ubiquity alone should not, however, assuage doubts about the validity of its synthesis of divergent traditions within radical political economy. Kalecki and the stagnationist authors who followed him saw the excess capacity and sluggish growth endemic to advanced capitalism as only remediable through egalitarian redistribution, or the continued socialization of investment. Attempting to reconciling this approach with that of authors who viewed social democracy as fatally flawed was the ambitious object of the B-M model, an object that in hindsight appears less that fully realized. With the considerable rise in the profit share of national income witnessed over the past thirty years in many advanced capitalist economies, and the modest growth that has accompanied it, some authors have strained to see the emergence of a profit-led regime. Apart from the reluctant acceptance of the policy prescriptions of the Right implied by this characterization of growth, authors working within this tradition have also tended, until recently, to neglect the role of household debt accumulation in sustaining effective demand. If, as seems increasingly plain, this rise in household debt was the necessary complement to the (illusory) realization of profit-led growth, the present prospects for revived rates of growth in the absence of social democratic reforms appear grim.