Post Keynesian Pricing Theory:

Alternative Foundations and Prospects for Future Research.[1]

Dr Paul Downward

Reader in Economics

StaffordshireUniversity

Leek Road

Stoke on Trent

ST4 2DF

Tel:+44 1782 294101 (direct line)

+44 1782 294074 (message)

Email:

JEL classification code: B51, B52, D11, D21, D43, D57, D83, L11

PsycINFO classification code: 3660, 3920

Abstract

Using references to Post Keynesian pricing theory, this paper shows how decision making under uncertainty, habits and institutions and consumer theory can be integrated to provide a coherent alternative perspective on microeconomics. Possibilities of clear research synergies between behavioural, psychological and Post Keynesian economists are noted.

1. Introduction

This short paper has two aims. The first is to provide readers of this Journal with an overview of the ideas that underpin Post Keynesian pricing theory. This apparently arbitrary objective in fact provides an opportunity to illustrate how subjects from previous papers in this symposium, decision making under uncertainty, habits and institutions, and consumer theory can be integrated. The second aim is to illustrate further potential links with economic psychology and pricing, being suggestive of future research agendas.

In the next section the three main origins of Post Keynesian pricing theory are outlined. Section 3 addresses the issue of the coherence of these contributions. Section 4 suggests themes that, in the author’s opinion, offer potential for research synergies with readers of this journal.

2.Alternative Foundations of Post Keynesian Pricing Theory[2]

As Fontana and Gerrard indicate in their paper in this symposium, Post Keynesian Economics has a long and diverse origin. The same is true of pricing wherein there is a rejection of the core mainstream economic principle that market prices are an index that measures, and aggregates, subjective individual valuations on alternative uses of scarce resources. Of course, these principles were first developed out of the ideas put forward in the ‘marginal revolution’ of Jevons, Menger and Walras.

These ideas were a reaction to a different perspective of prices put forward by the ‘Classical economists’. Despite at times exhibiting stark and fundamental differences in key areas of their analysis and ideology, the classical economists, Smith, Ricardo and Marx, united in distinguishing market prices from natural prices. Market prices governed day to day transactions. However, these were regulated by natural prices, which reflected long-run values deriving from the use of labour in production. Natural prices ensured the reproducibility and/or growth of the economy. This distinction facilitated an examination of the economy in terms of social relations, classes and institutions.

2.1 Neo-Ricardian Economics

Recognising analytical problems with the labour theory of value (summarised and extended in Steedman, 1977), one branch of Post Keynesian pricing analysis draws upon the classical approach without the labour theory of value by revisiting the work of Sraffa (1960). The aim is to challenge neoclassical economics. Typically, referred to as Neo-Ricardians, theorists analyse the determination of prices in input-output systems of equations. Demand thus has no central role to play in determining prices. In contrast the value of (input) commodities in production are marked up according to profit rates to yield (output) prices of production. Marked-up prices and income distribution are thus closely linked in this setting. Other key themes addressed by Neo-Ricardians are capital theory and attempts to forge links with Keynes’ theory of effective demand (see for example Harcourt, 1972; Roncaglia, 1978; Eatwell and Millgate, 1983; Steedman, 1992)[3]. It should be noted that the historical focus has been upon the ‘production of commodities by commodities’ – that is, manufacturing.

More recently, Eichner (1991)[4] offered a ‘long-period’ analysis of prices in an input-output system of production that has its roots in Sraffa (1960) but which is linked to a ‘short-period’ analysis of pricing that is discussed below. This model determines the system-wide set of equilibrium prices that allows the economy to experience steady-state expansion. Finally, drawing upon Eichner’s approach, Lee (1998) provides an analysis of long-run prices that does not rely upon a duality between production and prices. In this regard, Lee attempts to introduce history into the analysis that is otherwise based in equilibrium terms.

The other branches of Post Keynesian pricing theory have different origins. They are rooted in the marginalist controversy, prompted by the Oxford Economists Research Group in the 1930s and the institutionalist economics of the US of the same era. At the core of the marginalist controversy lie tensions of methodology that persist in Post Keynesian economics today. This concerns the respective merits of formal theory and a historical/behavioural approach to theory.

2.2 Industrial Organisation Theories

In the former case key exemplars are Kalecki (1940), Eichner (1991) and Cowling and Waterson (1976). Sharing an emphasis with both neoclassical economics and Neo-Ricardian analysis upon equilibrium, these contributions essentially draw upon and refine the Structure-Conduct-Performance industrial organisation paradigm. This paradigm applies optimising behaviour to more realistic industrial scenarios than perfect competition, wherein large firms, supplying differentiated products can earn persistent ‘abnormal’ profits. A key historical result, addressed in the marginalist controversy, was that mark ups on marginal costs reflected the reciprocal of the price elasticity of demand as a measure of monopoly power.

Kalecki (1940) thus presents a profit maximising, model of imperfect competition in which marginal revenues and costs are equated. Cowling and Waterson (1976) extended this analysis to argue that mark ups would also depend upon the industrial concentration in an industry and the degree of ‘Cournot’ competition. Likewise, Eichner, (1991) offered a determinate model of pricing that he intended to provide a microfoundation for macroeconomics. He developed a microeconomic model of the core oligopolistic sector of the US, wherein price-leader ‘megacorps’ seek to maximise their growth subject to institutional constraints, such as the degree of entry to the market and government scrutiny.

2.3 Historical/Behavioural Theories

The final branch of Post Keynesian pricing theory, tends to reject the formalism and equilibrium approach of the above strands of thought and embrace the original analyses of the Oxford Economists research Group and US institutional economists. In the former case, the seminal contributions were Hall and Hitch (1939) and Andrews (1949, 1964). Based on interviews with businessmen, Hall and Hitch argued that the, then emerging, general approach to pricing that emphasised marginalist rules was misplaced. Uncertainty facing businessmen meant that firms did not make use of either marginal revenue or costs to set prices, nor made (even implicit) calculations of the price-elasticity of demand in setting mark ups and prices. In contrast firms marked up their prices on full average costs to the extent that the ‘group of competing producers’ (p. 19) would allow. The mark up was set to cover both overheads and a conventionalallowance for profit per unit. In order to allocate overheads and hence determine their margin for profit and the level of prices, businessmen in practice used a hypothesisedor normallevel of output to govern their pricing decisions, rather than a consideration of actual levels of output, which may change or deviate from this hypothesised level as demand changes.

On the whole, therefore, nominal prices would stabilise at a conventional level corresponding to the normal level of output for firms, reflecting normal competitive conditions. A kinked demand curve was presented by Hall and Hitch to capture these conjectures. Andrews’ (1949, 1964) in many respects echoed their work but described the pricing procedures as normal-cost pricing. Andrews was, moreover, keen to emphasise that competition was fierce and that ‘known rivalry’ embracing both actual and potential competitors helped to structure prices. In this respect full-cost pricing should not be thought of as bureaucratic behaviour. In contrast price stability reflected uncertain expectations about potential competition.

In the US, the institutional economist Gardiner Means’ (1935, 1936, 1962, 1972) central concern was with articulating how prices are set in large ‘modern’ corporations. In this respect he was less concerned with the nature of external market constraints on prices than, say, Andrews or Hall and Hitch. His thesis was that one can distinguish price-setting as reflecting two extremes. In large modern corporations prices are ‘administered’ rather than set in the ‘auction’ conditions implied in neoclassical price theory. Administered prices are prices that are fixed by administrative fiat before transactions occur and are held constant for periods of time and for sequential sets of transactions. Nevertheless, as Means (1962) illustrates, pricing is carried out in terms that are analogous to full/normal-cost pricing. The implication thus is that once again, prices are set with reference to mark-up procedures and prices will change more closely with cost changes than demand changes.

Some of the later work of Kalecki (1954, 1971) also presents a theory of pricing in which ‘exact’ profit maximising behaviour is rejected because of the uncertainties faced by firms. The emphasis is on firms marking up average direct costs with mark ups being constrained by the degree of monopoly of the firm and the price of similar products. The degree of monopoly however is not measured by the reciprocal of the price elasticity of demand but reflects institutional features of the market such as sales promotion activities, trade-union power, the degree of industrial concentration and the level of overheads relative to direct costs. Although Kalecki treats overheads differently than Hall and Hitch and Andrews, one can argue that the similarities are stark. For debate, particularly over the behaviour of costs, see Kriesler, (1987), Downward and Reynolds (1996), Lavoie, (1996) and Lee, (1996). It remains that Kalecki’s analysis also emphasises that firms follow mark-up procedures to determine prices and that as a result prices vary directly with costs but only indirectly with demand.

3. A Coherent perspective?

Though drawing upon a wide set of ideas there are two ways in which the coherence of Post Keynesian pricing can be understood and out of which research agendas with behavioural and psychological economics established.

The first of the approaches is the traditional perspective that finds congruence in structural aspects of pricing in inter-industry, particular industry and firm-level contexts. The emphasised link between the theories is that prices are set according to mark-up procedures and that prices will vary with costs but not directly with demand. Of course, this proposition contrasts greatly with the general equilibrium, perfectly competitive core of neoclassical economics.

Attempts have been made to directly synthesise the Neo-Ricardian and industrial organisation approaches to pricing. As noted earlier, Eichner (1991) considers his megacorp approach to pricing to be a ‘short-period’ analysis that determines ‘actual prices’, which are then linked to the long-run analysis of prices consistent with steady-state growth in the Neo-Ricardian framework as discussed above. Reynolds (1987) also argues that the ‘short-period’ aspect of Eichner’s work is closely linked to the Structure-Conduct-Performance approach while, in general, Sawyer (1990) argues that Post Keynesian economics, as defined by these contributions, strongly overlaps with the industrial organisation literature. This possibility of synthesis is shared in Arestis (1992) and Lavoie (1992) in their textbook analyses of Post Keynesian economics.

The second approach is more radical, selective and draws upon methodological criteria. As indicated above textbook perspectives of Post Keynesian economics argue that the fundamental difference between neoclassical and Post Keynesian accounts of pricing lie in the emphasis upon mark-up procedures and asymmetrical effects of costs and demand upon prices. Yet, there is a problem. The ergodic/non-ergodic, closed-system/open system ontological distinctions referred to by Fontana and Gerrard as distinguishing between mainstream and Post Keynesian analysis can clearly apply to the above pricing theories. To put it bluntly, the Neo-Ricardian and industrial organisation perspectives both emphasise equilibrium solutions. The latter also explicitly draw upon an optimising explanation of behaviour. These are properties of ergodic or closed systems, which are at odds with uncertainty and historically contingent decision making. The rationale for mark-up procedures is thus different, emphasising a logical counterpart to marginalist rules. To this end, though particularly drawing upon a Critical Realist methodological perspective, Downward (1999, 2000) argues that only the historical/behavioural contributions constitute a constructive alternative to mainstream analysis.

4. Potential Links with Behavioural and Psychological Economics

On the basis of the above perspectives of coherence, potential links between Post Keynesian and Behavioural and Psychological economics can be identified. Let us take the Neo-Ricardian approach first. Despite the historical emphasis on the inter-industry linkages associated with the physical production of output, it is clear that there is a need to understand how complex products are supplied and the effects that this has on price – i.e., value – in modern service and consumer-oriented industries. The ‘value-chain’ perspective popularised as generic by the Business Strategist Michael Porter, for example, is potentially consistent with the Neo-Ricardian perspective.

One could use the input-output structure of Neo-Ricardian economics to analyse the physical linkages between activities in firms or implied by supply-chains between firms. An extension of this analysis would be to recognise that often technical factors such as the quality of inputs can act as an important constraint here. Another possibility is to recognise the conception of products as ‘brands’ with sets of distinct qualitative characteristics that are combined to produce economic value as implied in conceptions of brand equity in marketing and accounting. The aggregation of these characteristics into an expression of value is clearly possible in Neo-Ricardian terms. In both of these cases, the reasons for the weights by which components of activities or brands are aggregated – i.e., the structure of interlinked demands – need investigating.

Obvious lines of enquiry here are examining how firms signal their quality to customers and how quality symbols affect the consumption and supply of products and services. Such symbols will act as thresholds on behaviour and, as such, limit and refine search. In general the behaviour and negotiations of industrial buyers and suppliers as they balance the need to be competitive with the need to promote longer-term links needs investigation. In general, the message is that clearly psychological and behavioural analysis as much as technical concern is important in fleshing out how these forms of economic activity develop and evolve. To put it bluntly, there are social and power relationships involved not simply technical coefficients of production. Consistent with psychological economics, as Hodgson’s contribution to this symposium recognises, rules and habits underpin behaviour in which agents grapple to make ‘rational choices’ under conditions of uncertainty. Clearly in the multiactivity and multiagency contexts noted above such behaviour will be manifested and needs investigation. Of, course, one should recognise that the Neo-Ricardian concept of ‘value’ does have it roots in the classical intellectual tradition. This emphasises the material conditions of production. Movements towards psychological and more subjective notions do imply some compromise here.

Similar arguments apply in consideration of the industrial organisation theories of pricing. While essentially drawing upon structural features of markets to explain prices, such as concentration and the price elasticity of demand, it is clear that the agency that shapes these structural features needs investigation. The concept of concentration thus could be unpicked in terms of the type of Neo-Ricardian analysis discussed above. The sensitivity of consumers to price can also be investigated. The sense in which this can take place also applies to the historical/behavioural approaches to pricing. Some suggestions now follow.

Further links with behavioural and psychological economics could easily be made in examining the focus upon rule-following behaviour. Mark-up pricing procedures conform to this behaviour. One can argue that this is the case because information on costs – because of cost accounting procedures – provides reliable information for firms but stable margins are set and defended to produce predictability in the external sphere of markets in which information is less complete. Change thus tends to be incremental.

This is a feature emphasised, for example, in Hall and Hitch (1939) in which unstable prices arise when falls in demand are sufficiently high and persistent for firms to revise their expectations of normal output. Here, they seek to promote an increase in output by cutting prices, which will also reduce average total costs. In the short run, however, firms avoid this behaviour for fear of promoting a price war. ‘Habits’ in this context become manifest as pricing procedures and rules.