Liberalism and Growth Models

Liberalism and Growth Models

The 'liberal slant' on the comparative growth performance of countries is evident at the very beginning of the professional discussion of long-term economic growth. Indeed it could be argued that the father of economic liberalism, Adam Smith, first put forward a coherent growth model of a kind which is readily translatable into the aggregate production function form commonly found in modem growth theory. I use a stylized version of Smith's model in order to bring out the points where this distinct liberal slant may be detected. Of course, Smith did not adopt a mathematical formulation even of the simple kind presented below; the input of scholarship which would justify this formulation has been prodigious and may be found in the references cited in note 1. The Smithian aggregate production function can be specified as follows:

Y=flL,K,N,T), where fi,,fk,fn,ft > 0
Y = output;
K = capital stock;
L = labour force;
N = 'natural resources'; and T = level of technique.

Yis an increasing function of capital and labour input and natural resources and the state of technology with partial second derivatives of each factor negative. I shall follow other writers and regard N as a 'bit of a nuisance' (Higgins) and treat it as part of the capital stock. This makes no difference to the general argument though it would be important to consider N separately if one were closely concerned, as I am not, with highlighting all the main characteristics of Smith approach (for critical discussion, see Reid, 1989).

The particular contribution of Smith in the specification of the aggregate production function which attracts attention is his treatment of T. Smith rather dismisses philosophy and speculation as an important influence on T and concentrates on endogenous influences. Imagine that, for reasons to be gone into later, K increases relative to L. Y/L may increase initially but, cet. par., diminishing returns to capital must eventually set in, the capital/output ratio rises, and the process of growth is halted. However, T is itself a function of the extent of the market (Y/L) and the capital/labour ratio (K/L). The growth in the extent of the market facilitates specialization, offering greater opportunities for learning by doing which pays off in greater efficiency. The increase in capital per head is associated with an increase in the number of production processes and therefore with an extension of experience of specialization. It follows that:

T = T(Y/L,K/L), where t'(Y/K), t’(K/L) > 0

We can represent the Smithian process of growth which is technologically possible as being self-sustaining over some range of output. Curve A1 represents the relationship between an increase in capital per worker and output per worker for a given state of technology. For given L, Y/K increases but at a decreasing rate, i.e. the capital/output ratio rises, Curves A2 and A3 are similarly drawn for successively higher levels of T. Given equation, the potential growth path can be represented by a locus of points on successive A curves, as T increases in response to market expansion (Y/L) and the increase in capital per worker (K/L). Expressed in this way Smithian analysis recognizes the distinction between a growth effect produced by some permanent increase in K/L and a level effect produced by endogenous technical change (cf. Lucas, 1988).

It is hardly necessary for our argument to go into detail on the length and timing of the process of growth. Sufficient to say that Smith places primary emphasis on capital accumulation. The reaching of a maximum Y/L with a subsequent stationary state or even a decline in Y/L is closely related to the motives for capital accumulation. This is not to neglect the limitations that might be placed on the growth process from growing scarcity of that 'bit of a nuisance', land, or what was to become famous as the Malthusian effect, namely that population growth could be a function of growth in income per head. .The liberal slant in growth modelling first becomes evident in the insistence on understanding the micro-foundations of the growth process. What makes a society start moving along a growth path? The answer cannot be provided by specifying a technological relationship but depends on individual motivations. As we have already observed, Adam Smith put great emphasis on men's desire to 'better their condition'. Human action is both purposive and rational, a theme which is endlessly repeated in liberal political economy. With capital accumulation playing such a major part in Smith's model, how do we translate individual motivations into an increase in K/L? By an increase in resources set aside from personal consumption, in short by frugality. Indeed, Smith's prime example of bettering one's condition is that of saving, 'a desire which, though generally calm and dispassionate, comes with us from the womb, and never leaves us till we go into the grave'. The intended result is an increase in wealth and 'an augmentation of fortune is the means by which the greater part of men propose and wish to better their condition' (Smith, 1776/1976, p. 341).

In short:
S = s(Y),s'(Y)>0
S = &K

where S = Saving and &K = the increase in capital stock.
It follows that if saving increases with income, so will investment, so does the
capital stock and with it T increases, as previously explained.

The next element in the liberal slant is to define the conditions which will favour an increase in K without an offsetting rise in K/Y, the capital/output ratio. It is important that entrepreneurs envisage investment opportunities which offer profits over and above compensation for risk.2 Smith is the first of many liberal writers to emphasize the importance of good government and particularly institutional arrangements which favour security of property and the legal enforcement of contracts. Stable and wise government would be a necessary but not a sufficient condition for growth. A further condition would have to be some incentive to use capital efficiently so as to maximize profits. Competition which would allow freedom of entry into new markets at home and abroad and which would encourage process innovation would be an important supplement. Stern (1991, p. 128) actually observes that it goes 'beyond the standard theory' to add management and organization as an argument in the aggregate production function. Maybe; but this is an important theme in Smith who observes that 'monopoly ... is a great enemy of good management, which can never be universally established but in consequence of that free and universal competition which forces everybody to have recourse to it for the sake of self-defence' (Smith, 1776/1976, pp. 163-4).
The liberal slant is carried over into the Smithian recommendations regarding the role of the state in promoting growth. His is not an extreme laissez-faire position, as the following list clearly indicates:

(i) Remove barriers to competition, such as granting of monopoly privileges and - constantly emphasized - barriers to communication. Good roads, canals and navigable rivers are means for breaking down local monopolies. These require state action to remove barriers to private provision and even financial support by government - but not a state-run communications system.

(ii) Keep the growth of the public sector in check to prevent reduction in saving through increasing tax burdens;
(iii) Insofar as the state has to provide services which promote good government, try to relate effort to reward in public as well as in the private sector.

Again, it should be noted that Stem states that it goes beyond the standard theory to pay attention to infrastructure, in the widest sense, highlighting such impediments to growth as slow, costly and hazardous transport and corrupt and inefficient bureaucracy as impediments to growth, but credit for emphasizing such matters, which appear frequently in Smith, is only attributed to writings first published in the 1980s!

Smith's emphasis on good government as a precondition for growth and limited government as a cardinal element in maintaining it ran into trouble long before more sophisticated models appeared as rival products in the intellectual market­place. The doubts and reservations are well known and I only briefly mention those that emanate from liberal writers.
The first set of criticisms is associated with the Classical debate about whether or not economies would reach a stationary state. In this respect, the most that Smithian limited measures of intervention would achieve would be to postpone the day when the standard of living would gravitate to the subsistence minimum, even taking account of the benefits which could accrue from extending the market through encouragement of free trade. Smith's critics did not, however, envisage any solution which rested on further state intervention, and had different views about when the exact moment would come when, in Mill's words, 'the condition of the poorest class sinks, even in a progressive state, to the lowest point which they will consent to endure' (Mill, Toronto Edition, 1965, p. 753). It is worth parenthetic mention that Mill broke ranks amongst those who had an 'unaffected aversion' for the stationary state: 'I confess that I am not charmed with the idea of life held out by those who think that the normal state of human beings is that of struggling to get on; that the trampling, crushing, elbowing, and treading on each other's heels ... are the most desirable lot of human kind, or anything but the most disagreeable symptoms of the phases of industrial progress' (Mill, loc. cit.), adding later: 'the best state for human nature is that in which, while no one is poor, no one desires to be richer, nor has any reason to fear being thrust back by the efforts of others to push themselves forward'. Many liberals would say 'Amen'.

The second set of criticisms offers different reasons for the inadequacy of Smith's views on 'natural liberty' as the mainspring of economic growth. I need hardly dwell on Marx's attempt to use the stationary state of argument to demonstrate that the inevitable emergence of intolerable inequalities of income and employment opportunities would bring about the demise of capitalism. The influence of Marx on Schumpeter, a liberal convert, is manifest. Schumpeter accepted Marx's view that 'internal forces' would bring about a transformation of capitalism, though not as Marx predicted. The transformation to socialism would be gradual rather than violent. Increasing industrial concentration would separate ownership from control and remove the mainspring of capitalism - the proprietary interest in the use of capital. In its place would emerge state socialism fostered by the bureaucratic attitudes resulting from the organizational changes in large business enterprises. Schumpeter, unlike Marx, did not relish what he believed would come to pass.

Nevertheless, what is striking about Smith's approach is that important vestiges of it still remain in the modern growth literature, though this is not the result of the passing down of a Smithian torch from one generation of liberals to another.5 The intellectual battle between liberalist and socialist economists changed its form in the later nineteenth century and concerned such questions as the determination of factor shares according to the principle of marginal productivity and whether market socialism could be made to conform with principles of allocation based on consumer sovereignty. Indeed, I know of no specifically liberal model of economic growth to lay alongside the many variants of aggregate production models of the post-World War II era. That is not to say, as we shall soon observe, that there is not a liberal disposition towards such models, and particularly towards some of the policy conclusions that have been drawn from them. Thus, the Solow-type growth models adapted to include endogenous technological change have a clear affinity to the Smithian model, although the role of the entrepreneur is often not spelt out in the detail which Smith and later Schumpeter considered necessary.

The crucial question is whether the growth theories of today require liberals to modify their view about the Smithian conception of the role of the state in promoting economic progress. In the case of the Solow-type models with variable capital/output ratios and flexible factor and product prices, clearly the Smithian 'canons' are preconditions for the growth process to take place, but, in the primary version of the model, with no 'built-in' technological progress, no fiscal or monetary measures can influence the growth rate, in the long run anyway. Keith Shaw and I (Peacock and Shaw, 1976) have shown that if consumption rather than income per head is taken as the target variable, the fiscal policy could influence its growth rate; and it is well known that fiscal policy models have been produced to show how the natural rate of growth can be reached more swiftly by compensatory finance. But these are not much more than theoretical curiosities.

Once the model is developed along Smithian lines with technological improvements made a function of the growth in capital stock, then fiscal stimuli, so it is claimed, could raise the natural rate of growth and could do so within the time horizon which would make such a policy of interest to vole-maximizing governments (see Streissler, 1979; Peacock and Shaw, 1976). 1 suspect that liberals, assuming that they accepted that evidence supported this position, would be divided on the desirability of buttressing the Smithian 'canons' with fiscal stimulation. Those who considered growth as the most important argument in some widely agreed welfare function might consider that liberal principles would not be violated. Others would argue that the fiscal policy was in effect promoting the collectivization of saving for raising the rate of capital formation would imply a rise in the rate of saving above that sanctioned by a freely working capital market. The additional argument might be used that stimulation of investment by tax reliefs and grants would take the pressure off entrepreneurs to be efficient and innovative.

However, the recent emphasis on knowledge and therefore on research and education and training as a cardinal factor in growth (cf. Romer, 1986; Solow, 1991) presents a more interesting challenge to liberal thinking. Entrepreneurs may only have a limited interest in investment in knowledge and education because of the claimed difficulties of capturing the benefits. This leads to the familiar argument that the externalities created by research and knowledge require public support if not public production. The additional twist provided by Romer is that, even if externalities can be internalized, the introduction of increasing returns to scale throws doubts on the encouragement of competition, notably through trade liberalization, as a measure which would promote economic growth. It is beyond my competence to pass any judgment on the strength of this line of argument. What can be said is that a liberal who was convinced by the argument and accompanying evidence would wish to look very closely at how far the externalities created by knowledge required public support, and the efficiency of the delivery systems for education, training and research relying on such support; unless, of course, (s)he judged that such collective action would be contrary to liberal principles in the first place, whatever growth benefits might follow from it.

I have probably given the impression that the liberal position vis-a-vis growth economics is that of the critical bystander; and to some extent I believe that to be the right description. There is one area, however, in which liberal economists have been very much on the attack, namely with regard to growth models which assign a large role to collective action in raising growth rates in 'developing' countries. This attack is of more than passing interest because liberal thinkers, notably Bauer (1969) and Lachmann (1973), have questioned the very existence of a phenomenon labelled 'growth' and 'historicist prophecies' (Bauer) which claim to be able to map its progress and identify the forces which promote it. For example, Bauer argues:

(M)any of the concepts used ..., notably capital output ratios, are exceedingly vague. But even if they were more precisely defined they could not serve as the basis for scientific prediction as they are only simple descriptions of past occurrences and not variables between which functional and causal relationships have been established. And casual collection of a few statistics is not serious historical study, nor does simple extrapolation of arbitrarily chosen trends reflect scientific activity. (Bauer, 1969)

Latterly, liberal economists have been less 'Austrian' and more 'Smithian' in their attitude to macroeconomic modelling. It has become accepted that national income accounting and input-output analysis have considerably improved our knowledge of the workings of developing countries, though themselves originating in the need for development plans which have been, so it is claimed, conspicuously unsuccessful. Indeed, it is maintained by Deepak Lai, in a phrase which encapsulates the growing acceptance of the importance of efficient markets in development countries, that 'efficient growth which raises the demand for unskilled labour by "getting the prices right" is probably the most important means of alleviating poverty' (Lai, 1983, p. 102). Space prevents me adumbrating further the precise policy measures implicit in Lai's prescription of 'getting the prices right'; sufficient to note that his conclusion is that administrators in developing countries have much to learn still from 'that oft-neglected work, The Wealth of Nations, both so relevant and so modern' (Lai, 1983, p. 108). Adam Smith still lives and not only in Chicago!