Eckhard Hein DISTRIBUTION AND GROWTH AFTER KEYNES A Post Keynesian Guide (Edward Elgar 2014)

Transcription

1 Eckhard Hein DISTRIBUTION AND GROWTH AFTER KEYNES A Post Keynesian Guide (Edward Elgar 2014) CHAPTER 3 NEOCLASSICAL DISTRIBUTION AND GROWTH THEORY: OLD AND NEW AND A CRITIQUE

2 Content 3.1 Introduction 3.2 Neoclassical microeconomics and the unity of price and distribution theory 3.3 The aggregate marginal productivity theory of distribution 3.4 Old neoclassical growth theory: the stability of full employment growth with exogenous technological progress 3.5 The new (neoclassical) growth theory: endogenising productivity growth 3.6 Fundamental critique of the neoclassical distribution and growth theory 1

3 3.2 NEOCLASSICAL MICROECONOMICS AND THE UNITY OF PRICE AND DISTRIBUTION THEORY 2

4 Walrasian General Equilibrium Theory Given preferences (households) Given conditions of production (firms) Given initial endowments Perfect competition (households and firms as price takers) profit and utility maximisation under constraints Market theory: equilibrium prices and quantities on all goods and factor markets (symmetric treatment) 3

5 Profit maximising demand for factor of production i (3.1) Π: profits, Y: output, x: factors of production, p: price With prices given and factor j fixed: (3.2) First order condition for profit maximum: (3.3) 4 j j i i y x p x p Y p j j i i i y i x p x p x Y p x. 0 i i y i i y i p x Y p p x Y p x

6 Factor supply is derived from utility maximization: (3.4) p p i j U x i U x U: utility j dx dx j i Rising price of good i will induce households to consume less of the initial endowment with good i and supply it on the factor market. 5

7 Figure 3.1 Neoclassical partial factor market equilibrium 6

8 Income of production factor i: (3.5) (3.6) Y p * i * i p * : equilibrium price, q * : equilibrium quantity Same procedure for all factors of production and dividing by total income yields income share of each factor of production: z i i x Yi Y z: income share i Price theory is also distribution theory (for factors of production and for households), because endowments with factors of production is given 7

9 3.3 THE AGGREGATE MARGINAL PRODUCTIVITY THEORY OF DISTRIBUTION 8

10 Aggregation of factors of production and output (but how?) Cobb/Douglas production function: (3.7) Y L K, 1 Y: output, L: labour, K: capital, α, β: partial elasticities of production constant returns to scale 9

11 Falling marginal products: (3.8) (3.9) L Y 1 L Y L K K K 1 Falling average products: (3.10) Y L K 1 L L L K (3.11) Y K L K K L K 1 10

12 Solving (3.7) and (3.8) for α, resp. β, and substituting (3.9), resp. (3.10): partial elasticities of production: (3.12) Y L Y L Y Y L L (3.13) Y K Y K Y Y K K 11

13 Income shares are determined by partial elasticities of production: (3.14) w r Y L w r : real wage rate (3.15) r Y K r: real interest rate 12

14 Substituting (3.14) resp. (3.15) into (3.12) resp. (3.13) (3.16) wr wr L Y Y L W: wages W Y (3.17) r Y K rk Y Y Π: profits

15 Since 1, marginal productivity theory of factor prices explains distribution completely: (3.18) W ( ) Y Y Bad news for labour unions: increasing real wages has no effect on wage shares. Marginal productivity theory of factor prices implies: (3.19) r w r Y K Y L L K 1 L 1 K LK1 L K 14

16 Change in factor price relation has adverse effects on capital intensity (3.20) Elasticity of substitution of Cobb Douglas function (ε): (3.21) 15 r w L K r ) / ( ) / ( ) / ( ) / ( r r w r w r L K L K

17 (3.22) K L w r r r w r 1 (3.23) (K / L) (r / w r ) r w r 2 (3.24) (K / L) (r / w r ) r w r K L r w r 2 r w r r w r

18 3.4 OLD NEOCLASSICAL GROWTH THEORY: THE STABILITY OF FULL EMPLOYMENT GROWTH WITH EXOGENOUS TECHNOLOGICAL PROGRESS 17

19 3.4.1 Foundations Solow (1956) as a response to Harrod s (1939) instability hypothesis Harrod Actual rate of growth Warranted rate of growth (I=S) Natural rate of growth (labour force growth + productivity growth) Instability is related to warranted rate of growth Solow Locates the instability problem to the natural rate of growth Replaces fixed coefficient production function by CD function Attempts to show that natural rate of growth is stable 18

20 In Harrod s terms the critical question of balance boils down to a comparison between the natural rate of growth which depends, in the absence of technological change, on the increase of the labor force, and the warranted rate of growth which depends on the saving and investment habits of households and firms. But this fundamental opposition of warranted and natural rates turns out in the end to flow from the crucial assumption that production takes place under the conditions of fixed proportions. There is no possibility of substituting labor for capital in production. If this assumption is abandoned, the knife edge notion of unstable balance seems to go with it. (Solow 1956, p. 56, italics in the original) 19

21 Solow (1956) assumes: Labour force and productivity growth are exogenous Propensity to save is given Saving determines Investment (i.e. Say s Law is valid) Flexible prices on labour and goods markets Substitutional production function One good economy 20

22 3.4.2 The long term growth path: actual, warranted and natural rate of growth (3.7) Y L K, 1 CD function without productivity growth Actual rate of growth (3.25) Yˆ Lˆ Kˆ 21

23 Saving (S) determines investment (I) and capital stock growth (3.26) dk I S sy (3.27) Kˆ dk K I K S K sy K s v g w g w : warranted rate of growth, s: propensity to save, v: capital output ratio In equilibrium: (3.28) Yˆ Kˆ 22

24 Inserting (3.28) into (3.25): (3.29) Yˆ (1 ) Yˆ Yˆ Yˆ Lˆ 1 Lˆ. Yˆ, Lˆ, Lˆ, Growth rate of labour force determines equilibrium growth rate of all variables (3.30) ˆL ˆK Ŷ s v g n g w. Natural rate of growth (g n )towhichwarrantedrate(g w ) adjusts via the flexible capital output ratio (v) 23

25 With exogenous equilibrium growth (natural rate), the real rate of interest (the rate of profit) is determined endogenously: (3.31) Y g n (3.32) K s Real rate of interest is equal to the marginal productivity of capital: (3.15) It follows from (3.32) and (3.15) (3.33) I S Y K r g n r s g n K sy Exogenous long run equilibrium growth rate determines real interest rate/profit rate Y g s n K 24

26 3.4.3 The stability of the equilibrium Assume actual growth rate exceeds natural rate (3.34) Yˆ g 0 n Inserting (3.34) into (3.25): (3.35) Yˆ Yˆ Yˆ Yˆ g g g g n n n n Lˆ Kˆ Lˆ, ( 1) Lˆ Kˆ, Lˆ Kˆ, ( Kˆ g n ). Since β < 1 (falling marginal product of capital): (3.36) Kˆ Yˆ g n 25

27 Figure 3.2 Adjustment paths of the growth rates of the capital stock and of real income towards the natural rate of growth Capital stock growth determines output growth via production function. Output growth determines capital stock growth via saving function. 26

28 Divide the Cobb Douglas production function in equation (3.7) by the capital stock: (3.37) Y K L K 1 v 1 k Constant propensity to save (3.38) sy K s L K s v s 1 k v: capital output ratio, k: capital labour ratio 27

29 Figure 3.3 Existence and stability of the neoclassical full employment equilibrium growth rate 28

30 3.4.4 An increase in the propensity to save (3.39) g w ˆK s 1 v 0 g n ˆL s 0 v 0 In disequilibrium capital stock grows at faster rate than output capital output ratio rises until: (3.40) g w s 1 v 1 g n Saving and capital stock growth have no impact on equilibrium growth rate, but only affect growth path. Without labour force growth or technical progress the economy tends towards a stationary state, irrespective of capital accumulation. 29

31 Figure 3.4 An increase in the propensity to save and its effects on the neoclassical growth equilibrium 30

32 Figure 3.5 Shift of the equilibrium growth path by an increase in the propensity to save 31

33 3.4.5 Technological progress as an unexplained residual value Technological progress as an exponential trend (3.41) Y L K e t, 1 τ: exogenous growth rate of technological progress (3.42) Ŷ ˆL ˆK (3.43) ˆL ˆK Ŷ s v g n g w natural rate of growth is determined by exogenous labour force and productivity growth 32

34 Figure 3.6 An increase in the rate of technological progress and/or in the growth rate of the labour force 33

35 3.4.6 Implications for Convergence Figure 3.7 Unconditional convergence 34

36 Figure 3.8 Conditional convergence: low productivity country with a higher productivity growth rate during the convergence process 35

37 Figure 3.9 Conditional convergence: Low productivity country with a lower productivity growth rate during the convergence process 36

38 3.5 THE NEW (NEOCLASSICAL) GROWTH THEORY: ENDOGENIZING PRODUCTIVITY GROWTH 37

39 3.5.1 Introduction Purpose of new growth models: Productivity growth as an endogenous variable, determined within the economic system (present vs. future consumption) Implications of Solow model are not generally observed empirically: No general convergence of productivity (growth rates) only convergence clubs Productivity growth is positively related to investment share in GDP Human capital seems to have a positive impact on productivity growth R&D expenditures are positively related to productivity growth 38

40 3.5.2 The AK model Basic idea: capital accumulation at firm level has knowledge spillovers (positive externalities) for economy as a whole exactly compensating for decreasing marginal productivity of capital (Frankel 1962, Romer 1986, Rebelo 1991) (3.44) Y AK B, A 0 Y: real output, K B : broad capital stock (including human capital), A: constant broad capital productivity (3.45) dk B I B sy, s 1 I B : broad net investment, s: propensity to save 39

41 Since A (equation (3.44)) is a constant: (3.46) dy Y dk K B B I K B B g Inserting (3.44) and (3.45) into (3.46) (3.47) g n dy Y I K B B sy K B sa Saving/capital accumulation has permanent effect on natural rate of growth! 40

42 3.5.3 Human capital accumulation Figure 3.10 Structure of the human capital accumulation model Source: Based on Arnold (1995, p. 416) 41

43 Production function for output (3.48) Y Y(K,zH) Y: output, K: capital, H: human capital, z: share of human capital used in production Human capital production (3.49) dh (1 z)h, 0,1 : efficiency in the production of human capital, ψ: elasticity of production of human capital 42

44 Growth rate of human capital (3.50) dh H ˆ 1 H (1 z) H (1 z) (3.51) Hˆ Kˆ Yˆ (1 z) g n Natural rate of growth is determined by preferences (z) and technology in education (, ψ) Since human capital accumulation has external effects there are further potential positive effects on growth (Lucas 1988), which however require government intervention 43

45 3.5.4 Research and development Figure 3.11 Structure of the R&D model Source: Based on Arnold (1995, p. 425) 44

46 The state of generally available technological knowledge (A) can be conceived of as the sum of intermediate products (x i ): (3.52) Production function for output (consumption good only) (5.53) L P : labour in production Production of intermediate products (technological knowledge) (5.54) A x i Y C YA,L P da L R A, 0, 1 L R :labourinr&d,θ: efficiency of labour in R&D, ϑ: elasticity of production of technological knowledge 45

47 Growth rate of technological knowledge: (3.55) Aˆ da A L A 1 R L R Long run equilibrium growth rate: (3.56) Yˆ Cˆ Aˆ L R g n Natural rate of growth determined by preferences (share of labour in R&D) and technology in R&D (θ, ϑ) 46

48 Problems of R&D model (Romer 1990, 1994) Technological knowledge has public good properties (non rivalry) Perfect competition has to be given up, and temporary monopoly has to be allowed for in order to induce firms to invest in R&D External effects of R&D have to be taken into account Positive external effects require government regulation and intervention 47

49 3.5.5 The role of income distribution in the New Growth Theory New Growth Theory: Determination of the profit rate/real rate of interest through the production technology and through preferences, and then endogenous determination of growth by the savinginvestment mechanism. AK Model: From the production function in equation (3.44): (3.57) Y K B A r Inserting equation (3.57) into (3.47) for the long run equilibrium rate of growth, the natural rate, in the AK model we obtain: (3.58) g n sa sr 48

50 Income distribution in new neoclassical growth theory: Personal distribution Imperfect capital and credit markets: inequality restricts ability to accumulate human capital and to invest in R&D Political economy argument: distorting effects of taxation introduced to reduce inequality Political instability: unequal distribution has adverse political economy effects on accumulation and growth: illegal activities, social instability, left wing governments taxing the rich at the expense of private saving and investment 49

51 3.5.6 Progress and limitations of New Growth Theory In conventional theory, whenever increasing returns that are (dominantly) internal to the firm, externalities, public goods (or bads), incomplete and asymmetric information and so on are involved, there is a problem of market failure. Since the literature on new or endogenous growth revolves around precisely these phenomena, the question of public, institutional arrangements and mechanism design are close at hand. While capital accumulation is still at the centre of the analysis, these wider issues, which figured prominently in the classical authors, have been brought back into the picture. (Kurz 2006, p. 163) 50

52 But many problems apart from those relating to aggregate demand and aggregation of output, capital,... 51

53 1. Specific assumption regarding the elasticity of production of the stock of human capital (or of technological knowledge) with respect to additional human capital (or technological knowledge) Constant employment in human capital production or in R&D generates constant rates of growth of the economy; slight deviations either generate continuously increasing or falling productivity growth rates 52

54 2. Theory of technological progress? Preferences and technology determine productivity growth (institutions, class, history do not matter) 3. R&D models: perfect competition as the basis of neoclassical price, distribution and welfare theory has to be given up 4. Productivity growth as an endogenous result of capital accumulation is not really new: Smith, Ricardo, Marx, Kaldor, 53

55 5. Very little has been said in this survey about income distribution (in other words, about the determination of factor prices). That is because there is no special connection between the neoclassical model of growth and the determination of factor prices. The usual practice is to appeal to the same view of factor pricing that characterizes static neoclassical equilibrium theory. (Solow 2000, p. 378) 54

56 3.6 FUNDAMENTAL CRITIQUE OF THE NEOCLASSICAL DISTRIBUTION AND GROWTH THEORY 55

57 3.6.1 Two main areas of fundamental critique Relevance for a monetary production economy? The model economy traces just the path that it would follow if it were planned by the single, immortal household, solving infinite time utility maximization constrained only by given technological possibilities and inevitable identities. Under these favorable assumptions, the decentralization to competitive firms does not matter; in effect the industrial sector faithfully carries out the wishes of the household. (Solow 2000, p. 353) Logical consistency in an economy with more than one good 56

58 3.6.2 The capital controversy and its implications: the general issue Joan Robinson (1953/54): What is (the meaning of) capital? Cambridge controversies in the theory of capital (Harcourt 1972) Aggregation and inter dependence of prices and income distribution (known since Ricardo (1817), acknowledged by Wicksell (1934), and rigorously demonstrated by Sraffa (1960)) Optimal adjustment via substitution in an economy moving through historical time? 57

59 Probably the best method of exposition is to think of the neoclassical growth model as being a story about an imaginary economy that has only one produced good that can be consumed directly or stockpiled for use as a capital good. It is then an exact theory of that economy; and it becomes a difficult practical matter whether it provides a useful analogue of a multi commodity model. (Solow 2000, p. 351) 58

60 Neoclassical parables (Samuelson 1962) 1. Payment of production factors depends on their physical marginal productivity. The ratio of factor payments is determined by the ratio of marginal productivities. 2. There exists a unique inverse relationship between the factor price ratio and factor intensity, i.e. between capital intensity (K/N) and real interest rate real wage ratio [r/(w/p)]. 3. There is a unique inverse relationship between the interest rate and the capital output ratio (K/Y). 4. There is a unique inverse relationship between the interest rate and per capita consumption. 59

61 3.6.3 A simple model Prices and distribution in a two goods sector model (Harris 1978) Capital good 1, consumption good 2 (3.59) a r a 01w 1 11p1 1 p (3.60) a r a 02w 2 12p1 1 p a ij : input output ratios, a 0j :labour output ratios Given production technology by a ij known: a 0j unknown: price of consumption or capital good (p i ), nominal wage rate (w), real interest rate/profit rate (r) Four unknown variables and two equations! 60

62 Closure of the system Determination of prices: (3.61) Wages are determined by subsistence wage (w r s ) and class struggles (3.62) or Real rate of interest/rate of profit is determined by monetary factors (central bank or financial markets) (3.63) p 2 1 w r w r r sp 2 F i r F : rate of profit of enterprise, i: monetary rate of interest 61

63 Equations (3.59), (3.60), (3.61) and (3.63): Prices and distribution are interdependent: p i (r), w(r) or p i (w), r(w) Price of the capital good depends on distribution and therefore the value of the capital stock depends on distribution, as does the value of output Price system has one degree of openness 62

64 Relative capital intensities of the two industries: a 12 (3.64) a 02 a 11, 0 a 01 μ: relative capital intensity Value of μ determines the shape of the wage profit curve, as shown by Harris (1978, pp ) 63

65 Figure 3.12 Linear wage profit curve (μ=1) 64

66 Figure 3.13 Convex wage profit curve (μ>1) 65

67 Figure 3.14 Concave wage profit curve (μ<1) 66

68 The value of net output/income depends on distribution: (3.65) Y r wl rkr labour productivity: capital labour ratio: y = Y/L, k = K/L (3.66) y w rk k y r w capital intensity (k) depends on distribution: k(r) 67

69 Figure 3.15 Capital labour ratio with a convex wage profit curve 68

70 Figure 3.16 Capital labour ratio with a concave wage profit curve: reverse capital deepening 69

71 Figure 3.17 Capital labour ratio with a linear wage profit curve 70

72 Figure 3.18 Choice of techniques Technique a : a A 11,a A 01 ;a A A 12,a 02 Technique b : a A 11,a A 01 ;a B B 12,a 02 Technique c : a B 11,a B 01 ;a A A 12,a 02 Technique d : a B 11,a B 01 ;a B B 12,a 02 71

73 Figure 3.19 Re switching of techniques 72

74 Figure 3.20 Re switching of techniques and the capital labour ratio 73

75 Figure 3.21 Labour demand function with choice of techniques 74

76 3.6.4 The conclusion from the capital controversy No way to aggregate heterogenous capital goods to a homogenous value which is invariant with respect to income distribution. The choice of technique, capital intensity and hence employment cannot be derived from an aggregate production function. The rate of interest cannot be interpreted as the marginal product of capital, the wage rate is not the marginal product of labour. Factor income shares cannot be determined from their elasticities of production taken from an aggregate production function. 75

77 An increase in the rate of profit (in the real rate of interest) is compatible with reverse capital deepening and with reswitching of techniques. Measured capital intensity is not uniquely related to the real rate of interest. Generally, factor demand is not uniquely related to factor prices. Unemployment can hence not be cured by means of lowering the real wage rate. The adjustment processes towards the neoclassical growth equilibrium are not warranted in a more than one good (nonmonetary) economy 76

78 ( ) reswitching is a logical possibility in any technology ( ). Reswitching, whatever its empirical likelihood, does alert us to several vital possibilities: Lower interest rates may bring lower steady state consumption and lower capital/output ratios, and the transition to such lower interest rate can involve denial of diminishing returns and entail reverse capital deepening in which current consumption is augmented rather than sacrificed. There often turns out to be no unambiguous way of characterizing different processes as more "capital intensive," more "mechanized, more "roundabout," except in the ex post tautological sense of being adopted at alowerinterestrateandinvolvingahigherrealwage.suchatautological labeling is shown, in the case of reswitching, to lead to inconsistent ranking between pairs of unchanged technologies, depending upon which interest rate happens to prevail in the market. If all this causes headaches for those nostalgic for the old time parables of neoclassical writing, we must remind ourselves that scholars are not born to live an easy existence. We must respect, and appraise, the facts of life. (Samuelson 1966, pp , emphasis in the original) 77

79 (...) it is now generally recognized that there is no rigorous method of aggregating a heterogeneous collection of capital goods (Bliss, C., in: New Palgrave. A Dictonary of Economics, 1987, p. 885) 78

80 Why are aggregate neoclassical production functions still widely used? Attempt at ignoring and repressing uncomfortable arguments and conclusions by the macroeconomic mainstream (Pasinetti 2000) Theoretical irregularities like reverse capital deepening and re switching of techniques are empirically irrelevant and can therefore safely be ignored when doing theoretical and in particular empirical work. 79

81 3.6.5 Empirical relevance of the neoclassical aggregate production function A way out? (f)rom a practical point of view, economists are always dealing with aggregates ( ). The condition under which these aggregates can be formed, that is, under which the aggregates act as if they were homogeneous factors of production, are very restrictive; nonetheless, I believe that, under most circumstances and for most problems, the errors introduced as a consequence of aggregation of the kind involved in standard macro analysis are not too important ( ) (Stiglitz 1974, p. 899) 80

82 So why is the aggregate production function so widely and uncritically used? The answer seems to involve a form of Friedman s (1951) methodological instrumentalism. All theories, so the argument goes, involve heroic abstraction und unrealistic assumption, but what matters is their predictive ability. The aggregate production function, it is argued, passes this test with flying colours. (Felipe/McCombie 2010, p. 189) 81

83 The problem with this defence is, as we shall show, that the estimation of the a putative aggregate production function using constant price monetary (value) data cannot provide any inferences about the values of the putative parameters of the production function (output elasticities, aggregate elasticity of substitution) or the rate of technical progress. The reason is that there is an underlying accounting identity that relates these variables. This identity can be easily rewritten in a form that resembles a production function. This precludes any meaningful estimation of the production function and interpretation of the coefficients as estimates of the underlying technology. This critique is arguably the most damaging for the aggregate production function, because it applies even if there were no aggregation problems. (Felipe/McCombie 2010, pp ) 82

84 Aggregate neoclassical production function: (3.41) Y L K e t, 1 If marginal productivity determines factor prices, partial elasticities (α, β) determine factor income shares Dividing equation (3.41) by labour (L) yields: (3.67) y k e t y: labour productivity; k: capital intensity; τ: rate of technical progress; β:profitshare; α=1 β:wageshare Equation (3.67) in growth rate: (3.68) yˆ kˆ 83

85 National income accounts identity: (3.69) Dividing by L yield: (3.70) Derivative with respect to time: (3.71) Extending: (3.72) 84 rk L w Y r rk w y r t k r t r k t w t y r k k t k r r r t r k w w t w t y r r r

86 Dividing by y and taking into account that generally (3.73) yˆ wˆ r w y r ˆr rk y kˆ rk y x ˆ x t/ x w r /y Since the profit share is h rk / yand the wage share 1 h : (3.74) yˆ (1 h) wˆ r hrˆ hkˆ Taking (1 h)ŵ r hˆr (3.75) yˆ hkˆ Comparing equations (3.68) derived from the Cobb/Douglas production function and equation (3.75) derived from the national income accounts identity reveals that these are structurally similar. 85

87 It is thus no surprise when income shares are more or less constant through time (time series data) or through sectors (cross section data) that the Cobb Douglas would give a good fit: it can be derived from income identities. (Lavoie 1992, pp ) From this perspective, it also follows that Solow s residual (total factor productivity growth), the Holy Grail of the neoclassical growth model, is nothing but a weighted average of the growth rates of the wage and profit rates ( ), where the weights are the factor shares ( ). (Felipe/Fisher 2003, p. 255) The implications of the critique are far reaching. It implies that those areas of neoclassical macroeconomics that use the aggregate production function (with, or without the assumption that factors are paid their marginal products) have no theoretical or empirical basis. (Felipe/McCombie 2010, p. 190) 86

88 3.6.6 Abandon aggregate neoclassical modelling but retain the Walrasian general equilibrium approach for the analysis of distribution and growth? Hahn (1982): Capital critique is valid, but Walrasian general equilibrium theory remains intact Sraffa model is a special case of Walrasian model Problem: Walrasian model does not allow for a uniform or general rate of profit any more ( ) there is no presumption that the neoclassical theories of intertemporal or temporary equilibrium are general, whereas the classical theory is special. The truth is that these theories are different; while the former are short period, the latter is long period. (Kurz/Salvadori 1997, p. 467, emphasis in the original) 87

89 To sum up, the neoclassical scheme, in the Arrow Debreu version, needs neither rates of profit nor wage rates as such. It determines prices of given resources, and only prices. To find a place in this framework of analysis, any economic phenomenonmust bere moulded andre interpreted in these terms. With this conceptual framework, the shift of dominant economic theory in the direction of the neoclassical version descending from the Arrow Debreu scheme has practically entailed a general escape of economic analysis from the investigation and explanation of the problems of the distribution of income (and of wealth). At the dawn of classical economic theory, David Ricardo in 1817 opened his Principles with the famous proposition quoted at the beginning that To determine the laws which regulate this distribution (of income between rent, profit and wages) is the principal problem in Political Economy. At the end of the 20 th century, dominant economic theory has ended up with a theoretical framework (the neoclassical one, in the Arrow Debreu version), where the process and problems of income distribution have become secondary and, essentially, irrelevant. (Pasinetti, 2000, p.414) 88

90 While neoclassical economics envisions the lifetime utility maximizing consumption decisions of individuals as the driving force of economic activity, with the allocation of given, scarce resources as the fundamental economic problem, the English Cantabrigians [the Neo Ricardians and Post Keynesians, E.H.] argue for a return to a classical political economy vision. There, profit making decisions of capitalist firms are the driving force, with the fundamental economic problem being the allocation of surplus output to ensure reproduction and growth ( ). Because individuals depend on the market for their livelihoods, social class (their position within the division of labor) becomes the fundamental unit of analysis. (Cohen/Harcourt 2003, p. 208) 89