« Boffo, Marco (2013) Interrogating the knowledge‐based economy: from ...»
Indeed, while old growth theory crucially assumed ‘perfectly working markets and constant returns to scale’, ‘in order to endogenize growth and productivity’, EGT models breach both (Fine, 2006a, p.78). Thus, and firstly, EGT models are dependent on increasing returns to scale, from which follow that bigger economies will have higher productivity, and, consequently, the advantage of developed over developing countries (Fine, 2003b, p.207; Snowdon, Vane, 2005, pp.624-625). Secondly, and to substantiate the existence of increasing returns to scale, most EGT models rely ‘upon the presence of positive externalities’, with ‘constant returns to scale for the individual producer but positive spillover effects for the economy as a whole’, ‘as education, invention, learning, networks such as industrial districts’ or whatever other positive externality of choice ‘spread individual gains more widely’ (Fine, 2003b, p.207, emphasis added). Thirdly, and since both increasing returns to scale and positive externalities have long been acknowledged within the discipline as central examples of market failure, EGT models depend critically on the latter. Therefore, any ‘market imperfection can be used to generate’ an EGT model ‘as long as it generates increasing returns’, and ‘almost any sector of the economy can be perceived to experience market imperfections’ (Fine, 2003b, p.207).
Ultimately, translated in the mathematical apparatus of neoclassical economics, these are all sources of non-convexity. Therefore, this leads EGT models to drop the assumptions of technological convexity and overall concavity of the production function (which assures decreasing marginal returns to the factors of production and globally constant returns to scale, i.e. homogeneity), to allow for increasing returns to scale (Herrera, 2006; Romer, 1990b, 2005).70 Further, it is important to stress that, while these are far from being new elements in economic theory, they had, until the emergence of EGT, been relegated to the area of microeconomic (welfare) theory, where they have been traditionally identified as causes of static deadweight losses.71 Thus, ‘[w]hat is remarkable is that’ EGT ‘has taken such static, microeconomic deadweight losses … and transformed them into a macroeconomic influence on the growth rate’ (Fine, 2003b, p.208). This is extremely significant in an age of expansion of scope and applicability of IPRs. Indeed, (as discussed in sub-section 1.3.2) patents have been traditionally perceived within economics as implying a trade-off between static and dynamic inefficiency, since the monopoly power which they institute is a cause of deadweight losses measured by consumer surplus triangles. Thus, EGT, by identifying sources of (microeconomic) static deadweight losses as positive influences on (macroeconomic) growth, and, therefore, elevating the former to pride of place, has proved extremely important in shifting the conception of economists away from the characterisation of knowledge as a public good and towards favouring its commercialisation. Thus, EGT has complemented Coase’s redefinition of the “attributes” of the commodity, hitting the final nail on the coffin of (the non-excludability component of) the characterisation of knowledge as a public good.72 Indeed, in essence, Romer, 1990a transported within growth theory Coase’s attack to the non-excludability component of the public good, for it consecrated ‘[r]ivalry’ as ‘a purely technological attribute’ Incidentally, it should be noted here that, if we take the task of a theory of economic growth to explain the hierarchical relations and causal mechanisms linking factors of production and socioeconomic structures, dynamics and processes to each other, and these to economic growth, to call EGT a theory could be seen as inappropriate, since it involves no new theoretical content or reflection, but merely tinkering with the mathematical apparatus of neoclassical economics. Thus, whether there is, and what kind of, growth depends, in EGT models, on the mathematical properties of the model; but this is nothing like a theoretical explanation of actual economic growth and processes, rather the collapsing of the latter into mathematics (similarly, Herrera, 2006).
Indeed, as is well known, increasing returns to scale, externalities and market failure unsettle the ‘equivalence between Pareto optimum and competitive equilibrium’ (Herrera, 2006, p.246).
This has been acknowledged by Romer himself, for whom: ‘What endogenous growth theory is all about is that it took technology and reclassified it, not as a public good, but as a good which is subject to private control. It has at least some degree of appropriability or excludability associated with it, so that incentives matter for its production and use. But endogenous growth theory also retains the notion of non-rivalry that Solow captured. As he suggested, technology is a very different kind of good from capital and labour because it can be used over and over again, at zero marginal cost. The Solow theory was a very important first step. The natural next step beyond was to break down the public-good characterization of technology into this richer characterization – a partially excludable nonrival good. To do that you have to move away from perfect competition and that is what the recent round of growth theory has done’ (Romer, 2005, p.681).
(p.S73), and ‘[e]xcludability’ as ‘a function of both the technology and the legal system’ (p.S74). By ‘addressing the public good notion, and yet subordinating it to the privatization of knowledge through strengthened intellectual property rights, Romer took what had previously been the canonical justification for state subsidy of science and inverted it into a brief for the privatisation of science as a solution to the problems of flagging productivity and growth’ (Mirowski, 2011, p.74; similarly, also with respect to education and human capital as a productive factor in EGT, see Herrera, 2006, p.247, p.252). Significantly, in Romer’s own words, ‘[w]hat matters for the results’ of his model ‘is that the knowledge is a nonrival good that is partially excludable and privately provided’ (Romer, 1990a, p.S85), and this amounts to having the best of both worlds: on the one hand, non-rivalry allows for positive externalities and spillovers on the economy as a whole, while, on the other hand, the negation of the nonexcludable character of knowledge allows to tout patents (a source of static inefficiency) as good for growth.
However, that EGT cannot constitute a genuine engagement with the issues related to the production, reproduction and distribution of knowledge and technology is a clear consequence of its reliance on the neoclassical production function. Indeed, and firstly, as discussed (in footnotes 66 and 68) above, the neoclassical production function has been proven conceptually wrong and refuted in the heat of the Cambridge capital controversies. These saw the opposition of neoclassical economists and their radical political economy critics hailing from (or otherwise associated or aligned with), respectively, the Massachusetts Institute of Technology in Cambridge, Massachusetts, and Cambridge, England. Despite intellectual defeat having been inflicted on the neoclassical camp, as even acknowledged by Samuelson (1966), the revival of neoclassical growth theory represented by the rise of EGT has simply taken the neoclassical production function for granted as the starting point for empirical and theoretical work, as if the Cambridge controversies had never taken place (Fine, 2000, 2003b, 2012a;
Mirowski, 2011; Herrera, 2006). Therefore, as such, EGT is premised on an obsolete and discredited, albeit not therefore discarded, element of (disciplinary) knowledge. Secondly, the neoclassical production function is meant to represent a technological relation linking factors of production to (physical) output and, therefore ‘the physical/technological boundaries of what can be accomplished in the production process’ (Mirowski, 2011, p.76; similarly, Mirowski, 2007). However, the history of the production function within economics shows that, whatever its mathematical form, the latter was not dictated by engineering principles or physical laws but, rather, conceived in analogy with the utility function, and in deference to mathematical tractability and compatibility with the overall technical apparatus of neoclassical economics (Mirowski, 2007, 2011; for more on the history of the production function see Mirowski, 1989 and Ingrao, Israel, 1990). Indeed, the neoclassical production function was originally ‘modeled upon the original utility function … to permit an equal freedom of “choice” between supposedly coexistent yet distinct means of producing the good in question, the same way the consumer could “choose” between different baskets of commodities’ (Mirowski, 2007, p.491).
Thus, in the standard neoclassical economics framework, firms are posited as choosing techniques of production by solving constrained optimisation problems of cost minimisation, selecting the combination of inputs for which the isoquant of production is tangent to the lowest possible isocost line. This requires that marginal rates of technical substitution between inputs (i.e. the ratio between their marginal productivities, indicating the proportions in which one input can be exchanged for the other leaving the maximum level of output unchanged) equal the ratio of costs of production. But this is a fallacious and unrealistic representation of production on two important accounts. On the one hand, it is not possible to choose and change techniques of production freely and effortlessly, as ‘it takes time, new knowledge and a period of breaking the new process in’ before techniques of production can be developed, put to use and mastered effectively within any concrete production process (Mirowski, 2007, p.491), especially since productive activity involves transformative processes over the physical, spatial and temporal dimensions (Metcalfe, 2010). On the other hand, and more subtly, the neoclassical representation of the choice of production technique requires, conceptually and for marginal rates of technical substitution to be calculable in the first place, that inputs are freely and infinitely substitutable for one another. However, this breaches the second law of thermodynamics. Indeed, the latter affirms that the level of entropy (or disorder) of a closed system increases continuously and, therefore, that all closed systems transition from a state of order to one of disorder through degradation and decay through time (Boulding 1966;
Georgescu-Roegen, 1976); this implies the irreversibility of physical processes and that the representation of inputs in a production function as infinitely substitutable for one another is illegitimate (similarly, Mirowski, 2007, 2011 drawing on Georgescu-Roegen, 1976). Thus, to claim that EGT has endogenised technical change and knowledge, i.e. that it has made their production, reproduction and accumulation explicable from within the neoclassical model, other than in the shallow sense understood by mainstream economists, is, at best, a misunderstanding, for EGT is based on an obsolete and unrealistic element of knowledge (with respect to both the socio-economic processes it is meant to represent and the laws of physics).
Thirdly, many contributions to EGT (though not necessarily all) rest on the assumption that there exists a “stock” of knowledge, ideas or human capital, which can be introduced as an argument in the production function and which occupies a central place in the analysis, for it is its accumulation and properties which allow for increasing returns to scale (see, for key and seminal examples, Romer, 1986, 1990a; Lucas, 1988). Those upholding the distinction between codified and tacit knowledge have criticised EGT models for being ‘constructed around (the formalized representation of) a universal stock of technological knowledge to which all agents might contribute and from which all agents can draw costlessly’, thus neglecting the natural excludability deriving from the part of knowledge which is tacit (Cowan et al., 2000, p.226).
Following the role of EGT in attacking the non-excludability component of the conception of knowledge as a public good, this criticism is clearly beside the point. More debilitating, though, is the heterogeneous nature of different elements of knowledge. Indeed, the definition of a stock of knowledge implies the ‘need for a metric in which the constituent parts can be rendered commensurable’ (Cowan et al., 2000, p.227), but competitive markets are incapable of performing such a task precisely because of the heterogeneity of knowledge (Boulding, 1966; Cowan et al., 2000; Metcalfe, 2001, 2010). Indeed, as perceptively put by Metcalfe (2001, p.580), ‘[a]re ideas to be added, multiplied together, or aggregated in combinatorial fashion, in which case the stock grows faster than exponentially? Whatever the process of aggregation, we still need the weights (prices) with which an idea in carbon chemistry, say, is to be combined with an idea in the production of insurance services. It is not obvious what the weights are, and they certainly are not to be found in market prices’. Further, this implies that, assuming that a stock of knowledge is measurable at all, this can be done in ordinal (as opposed to cardinal) terms only. Yet, although EGT models do not ‘explicitly suppose the stock of knowledge to be cardinally measurable’, ‘they often assert this by implication’ (Steedman, 2003, p.128). Indeed, when a stock of knowledge is inserted as an argument in a production function exhibiting constant returns to scale, or when exhibits decreasing or increasing returns, or when ‘ is set equal to some power of multiplied by other variables’, it is implicit that the stock of knowledge is thought of as cardinally measurable, otherwise all of these assertions would be entirely meaningless. Yet, no contribution to EGT has demonstrated that a stock of knowledge can be measured cardinally, nor is any such demonstration possible, given the heterogeneous character of different elements of knowledge (Steedman, 2003).