Extracts from an ancient Home Work assignment of Sanjeev Sabhlok, Fall, 1994

FRAMEWORK for the rest of the questions


Before going further, it would be useful to outline briefly the connotation or interpretation of the names of different "schools" of macro-economic thought which shall be referred to in the remaining questions:


i) The term Classicals is used herein to represent the broad canvas of ideas of Ricardo, J.S. Mill, Adam Smith, Walras and Say, who postulated a market economy with flexible price system in a general equilibrium context. Hicks would also include many other pre-Keynesian economists such as the Physiocrats in this list.


ii) The term Neo-Classicals is used to include Marshall, Edgeworth, Arrow, Debreu, and perhaps to some extent, Samuelson. These economists abide by the Classical assumptions and concepts: they differ from the classical in terms of their strict (often mathematical) derivations of aggregate demand and supply functions from micro-level optimisation. Their most famous conclusion was that the economy is always in equilibrium. The Arrow-Debreu theorem (Debreu, 1959: 83) proves that a private ownership economy has an equilibrium if certain "nice" conditions are met by the various involved functions.


iii) Keynesians are those who (after Keynes, 1936) show how the economy tends to overshoot and rarely if ever exactly reaches equilibrium. Therefore disequilibrium is the usual state. This happens due to price and wage rigidities. The major representatives of this school are Keynes and Hicks.


iv) The New Classicals are those who use Classical assumptions of price flexibility and competitive equilibrium with unbounded rationality but with rational expectations. Agents form expectations on the basis of all available information. Further, the relationships among the governing variables are subject to random disturbances. Stochastic behaviour is thus essential to its approach to equilibrium: this yields an equilibrium path for the economy. Lucas is the major founder/ representative of this school.


v) New Keynesians: The New Keynesian have attempted to reach a synthesis between the neo-classical and Keynesian schools. The consequences of price and wage rigidities are explored here through modification of classical perfectly competitive market assumptions, and Keynesian macroeconomic results are deduced modified neo-classical foundations.


It can be seen that we are following Stiglitz (1992) broadly in this interpretation. More on the interpretation of the above schools will be discussed in Q.4 below.



2. There is no doubt that macroeconomics - or the behaviour of the economy in the aggregate - derives from micro-foundations, of various component units interacting with each other. But there is a complaint, often-voiced, that Keynes (1936) caused a split in the traditional "unity" of economics by creating a separate subject of macroeconomics, which appeared to postulate aggregate behaviour not rigorously derived from micro-behaviour (Leijonhufvud:26). We examine the justification of this complaint here.


In many ways Keynes was a path-breaker; he "discovered" the multiplier and fiscal policy. Qualitatively remarkable results were derived by him based solely on his observations of the economy as well as certain "psychological laws". These results seemed robust enough to explain various aspects of macroeconomic behaviour observed till the 1960s. But in the 1970s, the Keynesian "halo" effect began to wear out after the western economies faced unexpected shocks, and "simple" Keynesian solutions began to fail, particularly in the context of inflation. About this time, some economists began to investigate the "micro-foundations" of Keynes with the view to determine whether there was something intrinsically wrong with them which gave unexpected results (such as inflation) when "supply" shocks hit the economy. The focus of study was therefore the behaviour of price.


Keynes firmly believed in the price system as the arbitrator between people and firms, in the economy. The New Keynesian economists - who concentrate on study of the microfoundations of Keynes - therefore interpret Keynesian disequilibria as caused by some kind of imperfections when introduced into the perfectly competitive model.


In the following paragraphs, the similarities and differences between the Neo-Classical and (old) Keynesian streams of economic thinking are examined in the context of the four major markets in the economy.


2.(a) Keynesian micro foundations: As indicated above, much of the economic analysis of Keynes arose from two sources:


a) Observations of real life (i.e., experience): To illustrate, almost the entire Chapter Two of Keynes’ book is based on observations about the behaviour of labour; in particular, his perceptive observation that the desire of labour is to maintain (or improve) its relative wage rate in the economy. Thus, if the price level rises, reducing the real wage, the already employed labour does not opt out of employment on the ground that its real wage has declined. At p.247 of his book he summarises two major variables of his model derived from experience/ deduction: i) the wage-unit as determined by the bargains reached between employers and employed, and ii) the quantity of money as determined by the action of the central bank (i.e., money supply is exogenously determined).


b) "Psychological laws" which he used as assumptions: In fact, he went so far as to classify some of these assumptions as rules (Keynes: p.97). To be specific, Keynes uses three psychological laws (1936:247):


i) The psychological propensity to consume,

ii) the psychological attitude to liquidity, and

iii) the psychological expectation of future yield from capital assets.


It appears that most of Keynesian analysis can be considered as an approach towards representation of "stylised facts" (cf. Kaldor) rather than a neat mathematical analysis of the economy.


i. Market for consumer goods:

Demand for consumer goods: Two things determine the demand side of this market according to Keynes: one is the price system with perfect competition (implicitly assumed). The amount of goods demanded can be therefore arrived at from optimising individual behaviour. The other is a "jump" into the aggregate behaviour (consumption function) without explaining whether the assumption can be optimally arrived at in a multi-period income-consumption framework. For the latter, he postulates the existence of the "psychological propensity to consume" as follows:


"The fundamental psychological law, upon which we are entitled to depend with great confidence both a priori from our knowledge of human nature and from the detailed facts of experience, is that men are disposed, as a rule and on the average, to increase their consumption as their income increases, but not by as much as the increase in their income" (Keynes: 96).


Thus apart from the price mechanism which operated in the background at the microeconomic level, the above "law" is the only so-called micro economic foundation of Keynes’ approach to the consumer goods market. Other factors like distribution of income of the consumer (i.e., his resource endowment) were also considered by Keynes, but not explicitly used in his model.


Therefore the aggregate demand for consumption goods is given by c = c0 + cy(Y) where c0 - the autonomous consumption, is exogenously determined; and cy(Y) is the endogenous consumption, with cy being the MPC, Y being the aggregate income.


Supply of consumer goods: On the supply side he seems to have assumed that firms supply at a price which clears the goods market, which is very akin to Walras’ law.




ii. Market for investment goods:

Demand for investment goods: According to Keynes the determinant of investment demand is the relative influence or strength of marginal efficiency of capital (MEC) and the rate of interest. The market rate of interest determines how much money is available for investment (cf. the liquidity demand discussed below). He (at p.135) defines MEC "as being equal to that rate of discount which would make the present value of the series of annuities given by the returns expected from the capital-asset during its life just equal to its supply price."


This is akin to the IRR (internal rate of return) concept of finance. The use of "marginal" here refers to the fact that new investment I is based on the IRR of the additional investment proposal, and not on the existing investments. This involves expectation about future yields from the project: Keynes calls it the psychological expectation of future yield from capital assets. Equilibrium is achieved when MEC = r (the market rate of interest).

Relationship between I and r: As I increases, MEC decreases. This happens because no one can hope to invest infinitely at a high IRR. But as MEC falls below r, investment comes to a halt. If r were low, MEC would be high enough to attract capital => low r induces high I; high r reduces I.


The MEC is thus the price of investment. Investment in this system is not determined by external agencies planning investment, but by millions of individuals taking their decisions on the basis of objective yardsticks. To that extent this approach can be considered to be a "micro-foundation".


Supply of investment goods: This derives from individual saving decisions, with the equality of I and S in the aggregate. To the extent that the individual considers the return on his saving, Keynesian microfoundations are similar to the neo-Classical; only here, the MEC is an explicit decision variable.


iii. Market for money: This is a very significant market in Keynesian economics. Money matters. In fact the key feature of the Keynesian system is that it integrates the market for money into the theory of income determination through the theory of the rate of interest. The equilibrating force is the interest rate. The interest rate was seen by him as money’s own price.


Demand for money: In his system the demand for speculative balances which is a decreasing function of the rate of interest, is added to the classical precautionary or transactions demand for money (Keynes:199). The micro-foundation for the money market is therefore the "psychological attitude to liquidity" mentioned earlier.


Supply of money: This was assumed by Keynes for most of his analysis to be exogenously determined by the central bank. However, his analysis can be adapted to the case of endogenous money supply also.


Inter-action of the demand for and supply of money: In essence, therefore an increase in money supply lowers the interest rate through the use of the liquidity preference function, which represents speculative demand. This (lower) interest rate induces an increase in investment, and through the multiplier mechanism causes a rise in employment and production. This is the LM curve of Hicks, in which r = g(Y) with g¢ >0, and shows that an expansionist monetary policy can keep depression and stagnation away. It is worth mentioning that Keynes found that the money market adjusts rapidly (compared to the labour market discussed below).

iv. Market for labour:

Demand for labour: In the demand for labour, Keynes can be said to have visualised (at least implcitly) microfoundations very similar to those of the neo-classical competitive equilibrium analysis, in the sense that a firm can be modelled as maximising its profit subject to the production function. The demand for labour is then determined based on the marginal product of labour (Keynes, p.5).


Supply of labour: As far as the supply of labour is concerned, Keynes differs radically from the Classical analysis. He goes to great length to point out how and why the Classical assumption that "the utility of the wage when a given volume of labour is employed is equal to the marginal disutility of that amount of employment" is false. The major observation made by Keynes was that this market does not "clear" instantaneously, and that there is usually an excess supply over the demand for labour at any given wage level. He made the perceptive observation that "a fall in real wages due to a rise in prices, with money-wages unaltered, does not, as a rule, cause the supply of available labour on offer at the current wage to fall below the amount actually employed prior to the rise of prices" (Keynes, p.13). The result is that (at least in the short run, and before full employment level is attained) the supply of labour is a function of money wage rate alone and the supply curve is perfectly elastic upto the point of full employment.


In fact this observation happens to be a momentous one, for it started the chain of thinking leading to his important conclusions, including the major one that full employment equilibrium is hardly, if ever, attainable.


2.(b) Competitive Equilibrium foundations: This is basically the neo-classical world-view, wherein optimisation deicisons with perfect competition and flexible prices lead to equilibrium in the market at all times. The work of Arrow and Debreu marked a turning . point in terms of mathematical sophistication used in this approach. The Competitive equlibrium therefore has very strong microfoundations based on choice theoretic principles. It represents a general equilibrium approach - as contrasted with partial equilibrium approach - to the study of the economic system.


Millions, even billions (if the global economy is considered to be a purely open system) of consumers and firms - the basic economic decision-making units, who are basically motivated by self-interest (which is of course based on aspects studied by psychology, sociology, etc.), appear to make the economy "run" together in a complicated web of inter-relationships.


Competition in this theory represents perfect arbitrage, or in other words, that prices of the same commodity cannot continue to remain different in two different markets (law of one price): these prices shall be moved towards equality by forces of physical purchase of the commodity from low price areas and sale to the high price areas.


Due to this inter-relation, prices in all markets are simultaneously determined by an "invisible hand". The neo-classical approach tries to model this invisible hand of Adam Smith.


Mathematically, two kinds of optimisation decisions are taking place zillions of times in the economy, all the time:


a) Consumers: The individuals are modelled by preference functions. The Consumer Problem is: maximisation of expected (lifetime) utility subject to an (inter temporal) budget constraint. This gives rise to various household decisions such as consumer spending, labour supply and portfolio behaviour. We note that this is a maximisation problem under uncertainty.


b) Firms: Maximisation of profit subject to the production function. This gives rise to the producer’s decisions such as investment, employment, production and pricing.


This (equilibrium) approach is generally accepted, in the sense that it is agreed that analysis should be based on models in which agents optimise, given their objectives and constraints, and all markets should be integrated as far as possible. However, this requires that various markets clear instantaneously, something on which major differences remain.


i. Market for consumer goods: Let there be k commodities and prices indexed by k; m households indexed by i and n firms indexed by j. Prices of the k commodities are p1, p2, ..., pk, represented by the price vector p.


Demand for consumer goods: Demand functions for individual households are derived from utility maximisation (which in turn can be derived from preferences). Let these be represented by the vector Xi = (X1i, X2i, ..., Xki), Xki ³ 0. Represent the demand for the kth commodity by the ith consumer by Dki(p), since demand is a function of price. The sum of all the demands for the kth commodity is å k Dki(p); write this as D(p). In recent years, various models have "fine-tuned" these foundations, e.g., Milton Friedman’s permanent income hypothesis and Modigliani’s life-cycle hypothesis (Bruno, 1990). The introduction of uncertainty and dynamic programming into the analysis has also made demand analysis rather complex in present day competitive equilibrium models.


Supply of consumer goods: Supply functions for individual firms are derived from profit maximisation. Let these be represented by the vector Yj = (Y1j ,Y2j , ..., Ykj), Ykj³ 0. Represent the supply for the kth commodity by the jth firm by Skj(p), since supply is a function of price. The sum of all the supply for the kth commodity is å k Skj(p); write this as S(p).


Excess demand (called e, or e(p)) is given by X - Y. For the kth commodity it is given by: ek(p) = Xk - Yk


When the total value of goods brought to the market equals the value of things sold in the market, then p.e(p) = 0. This can happen only when either e(p) = 0 (in which case p³ 0), or when e(p) <0 (in which case p=0). But we rule out the second case where e(p) <0, by assuming that Say’s and Walras’ laws hold. Equilibrium is obtained when excess demand is zero.


ii. Market for investment goods:

Demand for investment goods: This market derives from the decisions of firms which are optimising their profits given the production function over periods of time, given the market rate of interest as a first approximation for purposes of comparison with the internal return on the investment. In many ways this is similar to the Keynesian analysis shown above. It must be mentioned, however, that there do not appear to exist many sophisiticated (earlier) versions of this analysis.


Supply of consumer goods: Consumers are thought of as maximising their life-time utility by optimising current consumption and savings. This approach using microeconomics is however, fraught with complexity as decisions for the future involve uncertainty. The market rate of interest is used here as a first approximation of the future return on savings.


Equilibrium is obtained when investment and savings are equal.


iii. Market for money:

Demand for money: Money demand is viewed basically as the transactions demand and is a function of the individual’s income. Money demand is not responsive to interest rates in the absence of speculative demand for money. Various models have emerged to explain and to extend this money demand model, e.g., Baumol’s transactions demand model, Asset demand for money including mean-variance analysis by Hicks and Tobin, etc. The quantity theory of money summarises the classical view of money. We look at it briefly in Q.3 below.


Supply of money: The neo-classical analysis includes money as exogenous (determined by the central bank).


In the competitive equilibrium model of the economy, money is neutral since if money supply increases, prices increase in the same proportion, neutralising any effect of increase of money supply.


iv. Market for labour:

Demand for labour: If perfect competition prevails in the product and labour markets, a profit-maximising firm will employ labour upto a point where the MPL (marginal physical product of labour) multiplied by the price of output is equal to wage rate which is the same as the cost of employment. This can be written as (where W= nominal wage, P = price level, w= real wage): W = P. MPL or, W/P (i.e., w) = MPL.


The labour demand Nd in this model is given by: Nd = f(w) = f (MPL), which is a decreasing function of real wage rate.


Supply of labour: Supply of labour is an increasing function of the real wage rate, Ns = f(w). This is because wages are assumed to adjust instantaneously, so that if prices rise, and real wage falls, employment will reduce. We have noted above how Keynes found this a rather implausible situation.


The equilibrium level of employment is determined by the interaction of forces of the demand and supply of labour in the labour market.


2.(c) i) Comparison of the two sets of micro foundations:


Keynesian micro-foundations have been derived from an intuitive analysis of what goes on in real life - a very useful scientific device: e.g., Newton discovered his laws by keen observation of nature; justification for these laws as a special case of Einstein’s theory came in much later. Similarly, the New Keynesians accept that Keynes’ observations and results were on the right track, but they seek a basic, ab-initio analysis of the assumptions. In particular, the New Keynesians and also, to a limited extent, some New Classicists are presently trying to prove that the wage-price rigidity of Keynes can be derived from micro-foundations, through recourse to various assumptions such as imperfect competition. In this process, it is expected by some that the Keynesian stream of thought will become more rigorous and the New Classical stream perhaps more Keynesian in its results.

ii) Similarities/ differences in the two:


ii.a ) Similarities: From a reading of Keynes, one gathers that he accepted that the competitive optimising behaviour of the individual/ firm was going on in the background. In fact, he insisted that such behaviour was essential for his theory to stand; but that complete freedom to this process could only be "permitted" in certain special cases, unlike Adam Smith’s view on the subject. He wrote (1936:378),


"If we suppose the volume of output to be given, i.e., to be determined by forces outside the classical scheme of thought, then there is no objection to be raised against the classical analysis of the manner in which private self-interest will determine what in particular is produced, in what proportions the factors of production will be combined to produce it, and how the value of the final product will be distributed between them." Therefore, "apart from the necessity of central controls to bring about an adjustment between the propensity to consume and the inducement to invest, there is no more reason to socialise economic life than there was before."


ii.b ) Differences:


i) Keynes concluded that the competitive market economy rarely reached equilibrium, whereas the neo-Classical economists believe in the competitive economy reaching equilibrium always. New Keynesians see the perfectly competitive economy reaching states of disequilibrium.


ii) Rigour: One obvious difference between the two sets of micro-foundations is in terms of rigour. Keynes arrived at his formulations based on observation/ intuition. Of course, one should recognise that at the time of Keynes, many of the currently used mathematical tools were absent. Neo-Classicals do not believe in this approach and want a definite theoretical route, starting from utility and profit maximisation, to arrive at this result.


iii) Assumptions/ limitations:


a) Explanation for wage/price ridigities: Today a lot has been learnt about the micro foundations of aggregate equations of Keynes; however, major Keynesian assumptions, such as price and wage stickiness, have not yet become amenable to explanation, except by "contrived" theory (as cited in Bruno:95) such as Fischer (1977), Taylor (1980), Blanchard and Kiyotaki (1986), Ball and Romer (1987). These attempts are usually called New Keynesian, and much progress has been made in supplying the "microfoundations" which Keynes himself lacked, and which he substituted by excellent intuition and observation.


b) Aggregation problem: There are however, serious theoretical problems with the attempt to explain observed phenomenon using the competitive equilibrium method. Grandmont (1990: 46) points out the aggregation problem (as shown by the Debreu-Sonnenschein-Mantel theorem), wherein if the distribution of individual characteristics is random, the process of optimisation can lead to unlimited types of aggregate behaviour. This seems to makes the Competitive Equilibrium foundations suspect. Stiglitz (1992) deals with this problem at length in his paper. Thus, the problem with the use of a rational individual or firms as the starting point of research is that the jump to aggregate rationality is too sudden and steep, as well as theoretically improper.


iii) Consistency of Keynesian assumptions with Classical assumptions: There have been attempts to integrate the two views, with the equilibrium approach as the starting point. A major commonality is the price system. Keynes clearly everywhere believed that the price system underlies economic activity, whether it is the price of commodities or the price of money. There are other points of consistency of Keynes with the Classicals, particularly in the labour and investment goods market, as already discussed above.


The major consistency of Keynesian assumptions with Classical comes when the economy reaches full employment. At that point, Keynes accepts the price flexibility condition, and the economy becomes "Classical" in all its aspects.


iv) Difference of Keynesian assumptions from Classical assumptions: Inspite of many areas of consistency with the classical assumptions, Keynes formulated his edifice on the basis of basic differences with some Classical assumptions, as pointed out at various places above. He himself generalised that "its (the Classical System’s) tacit assumptions are seldom or never satisfied" (Keynes: 378). The main difference, as already pointed out, was in the assumption of price and wage rigidity in Keynesian theory below the full-employment level, as well as inability of the economy to reach a full-employment equilibrium.




Very briefly, and very broadly speaking, therefore, the argument that a serious weakness of Keynesian economics is its "lack of micro economic foundation" is partly justified, but equally serious efforts are being undertaken by New Keynesians to provide these foundations, and ultimately, these efforts might lead to a unification of economics again. "Perhaps micro-foundations for Keynesian economics will come about in the not-too-distant future" (Honkapohja, 1990:79), making the two streams merge.



We explain, compare and contrast below, the different views held of the role of money and prices, in the neo-classical and Keynesian models of the economy.


Role of money:


i) Neo-Classical theory:


The first observation in this theory is that the money market is always in competitive equilibrium. The Walras’ law states that the total demand for goods and money is equal to the total supply of goods and money. The Say’s law says that the total demand for goods equals the total supply of goods (since goods markets clear). From this we see that the market for money also clears.


As far as the demand for money is concerned, neo-classicals do not allow for speculative demand for money: thus money demand is not influenced by interest rates; and comprises essentially of the transactions demand, influenced by income.


In its role, money is neutral (in its effects on the real sector) in the neo-classical analysis. This derives from the homogeneity in prices (of degree zero) of the utility function; if money supply is increased uniformly then it would merely have the effect of raising the prices uniformly as well as the income/ endowment, which would leave the utility unchanged. The equation of exchange also illustrates this view: MV = PY. In this case, V is assumed constant as being determined by payments technology. Further, Y is independent of the variables above, since it is determined by factor supplies. Accordingly, increase in M leads to increase in P.


A notable (and often debated in the literature) assumption made here is that the real balance (Pigou) effect is not large. If the real balance effect were large, then a fall in money supply would enrich the person’s money holdings (real balance), and his utility would increase, causing a non-neutral effect on the real economy.


There are two major deviations from this simple neo-classical view, based on the quantity theory and rational expectations. Both the views lead to the same conclusion in the long run, but differ on the role of money in the short run.


a) Monetarists: Quantity theorists (or monetarists) (e.g., Friedman) find that money is not neutral in the short run, though their theory derives from the classical system. Their result derives from the classical accounting identity: the quantity equation: MV = PT. Monetarists believe that the velocity of money, V, is in a steady, long-term trend. If people have more quantity of money, they spend it and this leads to increase in consumption and investment (real effects), which raises incomes. Thus, according to the monetarists there is a real effect of increase in money supply in the short run. The point to note is that here the quantity of money impinges directly on spending (rather than through interest as in the case of Keynes). At the same time, Monetarists think that the effect of money on the real economy is rather short-lived, and soon the prices change (by rising in the case of increased money supply), thus causing money to return to classical neutrality quite soon. In order to prevent short-term disturbance to the economy, monetarists argue that monetary policy should be conducted according to a fixed rule (non-discretionary).

b) New Classical view: (e.g., Lucas) Another influential view deriving from the classical view of money is that of the New Classicals (rational expectations hypothesis) who believe that money is neutral both in the short and long run, since people know that when money supply increases, prices are going to rise, and therefore tend to increase the prices immediately, rather than going through the complicated process of increase in Y, etc. Essentially, since people know that prices are going to increase, they will do so at once. This view holds that government need not follow discretionary policy (as advocated by Keynesians), and resembles Monetarists to that extent. They claim that government would do best to follow a steady growth in M, rather than allow spurts, since spurts will merely get translated into inflation, and destabilise the economy.


ii) Keynesian theory:


In the Keynesian theory, money matters.


Keynes states at page 293 of his book: "So long as we limit ourselves to the study of the individual industry or firm on the assumption that the aggregate quantity of employed resources is constant, and, provisionally, that the conditions of other industries or firms are unchanged, it is true that we are not concerned with the significant characteristics of money. But as soon as we pass to the problem of what determines output and employment as a whole, we require a complete theory of a Monetary Economy."


This brings out instantly that Keynes himself saw a distinction between the competitive equilibrium which can take care of firms and individuals operating within a price system, and the role of money on the macro-scale. At the level of the individual consumer, firm, homogeneity of degree one in prices is valid; not so at the aggregate level. The assumption of neutrality of money of the Classical/ neo-Classical system is, therefore, according to Keynesian theory, a kind of fallacy of composition.


The difference lies basically in the understanding of the vital role played by interest rate in the aggregate. The rate of interest is determined by monetary forces, being that rate of interest which equates the total supply of money with the demand for money. We have already shown (in Q.2 above) that Keynesians use a money demand function incorporating transactions as well as speculative demand. The velocity of money, V, according to the Keynesians, changes with the interest rates, due to the existence of this speculative (liquidity) demand for money. That is why different results are arrived at compared with the neo-classicals.


Here, as money expands, interest rate is forced down, increasing investment and thus aggregate demand, finally increasing income and employment. The effect of all this however, does not last forever. In the long run, Keynesians also believe that the economy becomes classical, and money becomes neutral. Further, in the case of a full-employment economy, money is neutral (Smith, 1956). Thus, at full employment, further relaxation of money supply would lead to inflation.


While the above is the basic view of the Keynesians regarding the role of money, they do admit that there exist many other factors which affect the portfolio adjustment of individuals as well as the transmission mechanism of monetary effects, and it is quite possible that the effects of the increase in money supply may not be transmitted to the real sector at all. In fact, due to this latter analysis, Keynesians very often discount the influence of money as the primary policy variable in the economy and lay greater stress on fiscal policy. They are also called fiscalists. But they do see a role for the government in managing the interest rate.


Role of prices:


i) Neo-Classical theory:


Prices are flexible: Prices are assumed to be perfectly flexible in the neo-classical system and equilibrate the various markets. Any change in the price level cannot produce any disequilibrium in the money market. This is further illustrated in the "Homogeneity Postulate" of the classicals: according to which the demand functions for commodities are homogeneous of degree zero in (money) prices. It means that if all money prices are increased by the same proportion, the demand for and supply of commodities remains unchanged as relative prices remain unaltered. In a world where Say’s identity holds, the commodity market is always in equilibrium.


The quantity theory of money shows that the price level is determined by various parameters. If the production function is given by B(a T)b where a , b represent the psychology of the people. We have pq = MV, => p (price level) = MV / y = MV / B(a T)b .


Baumol (1977: 484). shows that it is impossible, using the General Equilibrium theory to have any determinate price level - any price level will do as well as any other because they will all be equally consistent with equilibrium. This is because of the homogeneity assumption mentioned above: any change in the price level cannot disturb the equilibrium of the commodity market and so, by Walras’ Law, the money market will always be in equilibrium. As any price level will be consistent with equilibrium, price level remains indeterminate. In general monetary theory is made impossible using the neo-classical General Equilibrium theory.


In this connection, it is worth mentioning that some economists (New Keynesians) have argued that the alternative hypothesis - that there exists a significant amount of price level stickiness - is in principle consistent with the equilibrium approach of neo-classical economics. Some of them contend that prices do not adjust promptly for a variety of reasons not amenable to optimisation analysis, such as institutional factors.




a) As Nagatani points out (1981: 3), this model does not concern itself with a knowledge of the actual price-formation process. He writes, "In macroeconomics, where it is crucial to understand the dynamic processes of price and quantity variations, this type of equilibrium-price relationship is manifestly inadequate. Actual prices often turn out to be too irresponsive to important changes in the data (and conversely, too sensitive to false signals) for them to maintain the system close to equilibrium." In this view, this theory overstates the importance of price in determining the general equilibrium. Some of these aspects are taken into account by considering a dynamic model.


b) Another major weakness of this approach is that it assumes perfect knowledge by the constituent units: in reality, therefore, no firm takes into account (or can ever hope to take into account) the prices of all other goods while making its production decisions. What happens is that a very partial view off events is taken. The same holds for a consumer: he does not compare the prices of all other goods while deciding on the purchase of a particular good: if he does that he would never be actually able to purchase anything at all, given that prices of almost all commodities are changing over time.


ii) Keynesian theory:


Keynesians do not have a theory of price, because according to them, prices are given at any point, and remain at that point (they are sticky). The only point when prices are influenced by the system is when actual demand exceeds aggregate demand and this, according to the Keynesians, can happen only at full employment level. At other times (which are more common, the economy being in disequilibrium), there should therefore be no increase in prices, since supply (of both goods and labour) will increase to meet the rising demand. Due to this, Keynes tended to ignore the role of prices, focusing instead on real output and employment.


Day mentions (Day, 1994) that Keynesian theory is not a "general" theory (cf. The General Theory of Employment, Interest and Income) since it does not explain prices. It does not tell us why there has been inflation inspite of recessions/ stagnation. As mentioned above, Keynesian framework focused on the determinants of real output, real income, and unemployment, and therefore did not devote much attention to inflation. Therefore Keynesian policy "failed" in the 1970s, when even after recession, prices did not drop, but accelerated. The Philips curve controversy is central to this discussion. The (old) Keynesians tried to explain the changing relationship between unemployment and inflation through a shift in the Philips curve.


A notable change comes into the Keynesian models when considered dynamically. Inflation and various random shifts in the Philips curve are explained coherently using dynamic Keynesian models: it can be said therefore, that if considered dynamically, Keynesian models do throw light on the price formation process.


Empirically, Keynesians are on a sound wicket. The Keynesian intuition, that nominal wages and prices are usually sticky, has been recently successfully tested for the U.S. by Poterba, Rotemberg and Summers (1986) (cited in Bruno,1990: 94).


It is worth mentioning here that there is a recognition among Keynesians that prices tend to be sticky downward and are somewhat more flexible in the upward direction.




4. Stiglitz (1992) sees that macroeconomics economic thinking over the years has led to two broad views of Capitalism: one, which shows how the "invisible hand" in capitalist economies brings about efficiency and promotes innovation and the other which shows how capitalist economies are necessarily characterised by distributive injustice as well as by periodic fluctuations. How do both aspects arise from the same economy? Attempted solutions to this puzzle have taken various forms over the years. Stiglitz then classifies various economists, a classification that we use as a frame for the discussion.


Usual understanding of the term "Classical"

and how different schools partake portions of it:


i) The general understanding of the term "Classical" is that based on Walras’ general equilibrium appraoch with perfectly competitive markets equilibrated by rapidly adjusting price system, thus co-ordinated by an invisible hand (Adam Smith). The term Classicals is therefore commonly used to represent the broad canvas of ideas of Ricardo, J.S. Mill, Adam Smith, Walras and Say. Keynes (1936:3) had himself used this interpretation. Hicks (1992) would also include many other pre-Keynesian economists such as the Physiocrats in this list.


The neo-Classical, the New Classical, the New Keynesian, and even to som extent, the Keynesian, views emerge from this basic foundation, as will be attempted to be shown below.


ii) The term Neo-Classicals is used to include Marshall, Edgeworth, Arrow, Debreu, and perhaps to some extent, Samuelson. These economists abide by the Classical assumptions and concepts: they differ from the classical in terms of their strict (often mathematical) derivations of aggregate demand and supply functions from micro-level optimisation. Their most famous conclusion was that the economy is always in equilibrium. The Arrow-Debreu theorem (Debreu, 1959: 83) proves that a private ownership economy has an equilibrium if certain "nice" conditions are met by the various involved functions.


Critique: Neo-classical approaches have failed (in the opinion of Stiglitz) to throw much light on reality. His major objection to this approach is that the Debreu-Sonnenschein-Mantel theorem showed in the 1960s that "any set of market excess demand functions satisfying Walras’s law can be derived from utility-maximising individuals" (p.44). This is a very serious drawback, since it imposes no restrictions on rational behaviour: therefore virtually any economy can be deduced from this. In fact, he repeatedly shows how the use of aggregation and "representative" agents (individuals firms) can lead to serious error.


Hicks points out to the neo-Classical "myth" that all economic problems can be solved by the price mechanism. There is considerable scope for intervention, according to him. Obviously, both Stiglitz and Hicks profess Keynesian ideas.


iii) Keynesians are those who (after Keynes, 1936) show how the economy tends to overshoot and rarely if ever exactly reaches equilibrium. Therefore disequilibrium is the usual state. This happens due to price and wage rigidities. The major representatives of this school are Keynes and Hicks.


There have been views that even Keynes professed classical ideas, at least to some extent. This is true in the sense (highlighted in the Qs.2 and 3 above), that Keynes believed that the price system equilibrates the various markets, at least in the long run, and that at full employment level, the economy attains all the "classical" features. He believed that the classical situation prevails once stabilisation (through intervention) is accomplished.


Origin of Keynes in Classical thought:


Hicks (1992), however, shows very lucidly that Keynes in fact had more classical features than this, though perhaps he was not aware that his roots went back so far into history. Hicks analyses the origin of classical and Keynesian macroeconomics in his article, "The Unification of Macroeconomics". He shows that macroeconomics emerged from the Norman conquest of England (in about 1100 A.D.), whereafter William Petty was given the task of estimating the wealth of Ireland. Half a century later, Cantillon attempted to work out a fundamental answer, and made considerable progress. Around 1750, this work was extended by Physiocrats, who created a model, which Hicks shows contained the seeds of the multiplier of Keynes. The essential model is as follows:


Suppose that landowners (in rural areas) increase the demand for town goods. The income of the town people therefore increases from the payment made for these exports by landowners. This payment is in the form of corn. Suppose the fraction of their incomes which the townpeople want to spend on imports (of corn) is k (k<1). They will also spend k part of their increased income on imports of food, i.e., k ´ increase in income. This must be the amount that they want to increase their imports by. Therefore in equilibrium,


Increase in exports = increase in imports

= k ´ increase in income


Therefore, increase in town income = (1/k) ´ increase in exports.

Here, k<1, therefore 1/k >1 . It implies that for each unit of increased exports, their income increases by a multiplier.


This is very similar in concept to the multipliers used by Keynes. Did Keynes know about the Physiocrats?


Difference between Keynes and the Classics: Inspite of the above similarities, Hicks does find an essential difference between Classical and Keynesian theories: He finds Keynes' to be a theory of employment and money, whereas Classical theory kept money out. Keynes did this by considering wages to be money wages, and assumed that these are determined exogenously.


According to Hicks, if in Keynesian theory, real wage is kept constant, rather than money wage, then it should be possible to unify the two schools of macroeconomics. Some such synthesis has been attempted by the neo-classical synthesis as well as by New Keynesians.


iv) The neoclassical "synthesis": This took the view that stabilisation of the economy is possible (Keynesian), and once that is accomplished, the classical view of the economy takes on a predominant role. Samuelson was the foremost "synthesiser". Stiglitz sees in this view a lack of intellectual foundation. We do not elaborate upon this view here, as the New Keynesians have taken over from this school.


v) The New Classicals discarded the attempt to synthesise Keynesian views with the neo classical approach. They attempted to work their way up directly from micro economic foundations. Basically therefore, the Walrasian approach to micro economic analysis has been rehabilitated in the form of the "New Classical" theory. The distinguishing characteristic of various sub-schools of New Classical Economics is faith in mathematical rigour and in competitive equilibrium.

Two situations are assumed to hold good by the New Classicals:


(i) Equilibrium has to exist given certain nice assumptions (Arrow-Debreu theorem). The condition for this to hold is perfect competition, and complete markets. But Stiglitz points out that this is perhaps the "singular" case where the economy is Pareto efficient.


(ii) Markets clear instantaneously (Walras’ and Says’ laws hold). Any deviation in the paradigm of instantaneous market clearing is considered to be tantamount to the introduction of ad hoc based free parameters.


There are two major sub-schools of New Classical economics:


a) Rational expectations: In this the individual is believed to have rational expectations. Agents form expectations on the basis of all available information. Further, the relationships among the governing variables are subject to random disturbances. Stochastic behaviour is thus essential to its approach to equilibrium: this yields an equilibrium path for the economy. Lucas is the major founder/ representative of this school.


b) Supply side economics: (Mundell) This focuses on the supply side of the economy (unlike Keynesians who concentrated on the demand side).


To a theoretical economist, the main attraction of this approach is that preferences (utility functions) and technological relations (production functions) are invariant to policy rules or regimes and thus the conclusions can be generalised to any regime where price works as the equilibrating mechanism. Further, this branch of economic analysis imposes a rigorous discipline on the user, and prevents internal inconsistencies in model construction.


One of the main findings of this model is that fiscal and monetary policy has very limited positive influence on employment. In fact, all interventions are considered to cause distortions in the Walrasian general equilibrium model, and are therefore discouraged - a conclusion which is radically different from the Keynesian conclusion..


Criticism of New Classical approach:


a) Stiglitz differs from this approach tremendously. He states that it is not necessary to begin every paper from first principles. For example, based on the large body of financial research, it is perfectly rational to begin with an assumption that equity markets do not function efficiently: but the New Classicals refuse to do so.


b) Bruno (1990:90) points out that a considerable effort has been made by New Classical economists to create a theory which cannot, under its usual assumptions, uphold commonly observed phenomenon. Stiglitz also points out that aspects of reality such as wage and price rigidities cannot be explained using this approach. Empirically, nominal price rigidities have been found to exist by in the U.S. by Poterba, Rotemberg and Summers (1986) (as cited by Bruno,1990) . Thus, the Keynesian intuition, that nominal wages and prices are usually sticky, has so far stood firm. The New Classicals, according to Stiglitz, simply reject that these rigidities occur. His anecdote of the attitude of New Classicals sums up his approach: He compares them with biologists who, not being able to understand how blood goes up to a giraffe’s brains, conclude that this is not possible, and therefore the giraffe has a short neck. The main test of a theory, according to him has to be that the predictions from the model have to be consistent with observations. This is a very powerful criticism.


c) Alex Leijonhufvud (1992) also tries to show that the unbounded rationality assumption of New Classical economics/ classical economics has tremendous limitations, and that a more reasonable assumption would be of "bounded rationality" - which is New Keynesian.


d) Grandmont (1990: 45) sees in the New Classical view a discrepancy in that "... it is surprising nowadays to see some "New Classical" macroeconomists claim that the distinctive and novel feature of their research strategy - the so-called "equilibrium approach" - is to portray economic units as entities which optimise an objective function under well-defined constraints describing the market or social institutions through which they interact. I would have thought that understanding the aggregate behaviour of economic or social systems from sound choice theoretic principles has been, well before 1972 or 1936, one of the goals shared by many microeconomic and macroeconomic theorists of different persuasions, including Friedman, Hicks, Kalecki, Keynes, Modigliani, Patinkin, Samuelson, Solow or Tobin."


Thus almost all economists who subscribe to the capitalist-based price system are, in a way claimants of the equilibrium approach. It is only that the New Classicals have thought it fit to deduce everything from this paradigm - which is proving to be a very ambitious and difficult project.


v) New Keynesians: The New Keynesian have attempted to reach a synthesis between the neo-classical/ classical and Keynesian schools. The consequences of price and wage rigidities are explored here through modification of classical perfectly competitive market assumptions, and Keynesian macroeconomic results are deduced through these modified neo-classical foundations. Things like imperfect information, imperfect competition, adjustment costs, etc., are introduced. According to Stiglitz, whereas New Keynesians agree with the New Classicals that good macroeconomics should be built on solid micro foundations, they differ on which set of micro foundations to consider.


He is of the view that economics being a social science, important insights at the micro-level should be considered, whether they fit into the standard optimisation model, or not. Therefore, aspects of reality such as adverse selection and moral hazard should also be taken into account by macroeconomics, since these are a part of the micro-reality. This approach goes along with the Keynesian approach of observing each phenomenon of real life and using it to explain reality. It is therefore to a large extent empirically consistent. For example, New Keynesian models are able to handle things like price and wage rigidity which New Classical models find difficult to handle within their restricted framework.


Stiglitz sees the need of a thorough understanding of how informational considerations affect the functioning of labour, capital, and product markets. He feels that even the most complicated dynamic programming simulations cannot take into account the complexity which real life has to offer. Therefore, aspects such as price and wage rigidities are worth explaining from whichever angle seems best. Other areas dealt with include credit rationing.




The view that Keynes was a Classical economist does bears a modicum of scrutiny. But, except from proving the fact that the seeds of his ideas can be attributed to earlier (Classical) economists, it is difficult to derive any firm conclusion that Keynes was a Classical. The primary reason is that his assumptions were often radically different from those of classical economists.


Among those whose thinking is similar to Classical thinkers, are the neo-classicals, the New Classicals and the New Keynesians. It has been shown above that these three different schools partake of differing amounts of the foundations of the Classical school. The neo-classical and the New Classical schools do not distort its assumptions, whereas the New Keynesian school modifies the assumptions to try to converge theoretically to empirically observed reality.


It can be said in the end that the Classical school of thought has a long history and a long ("invisible") hand: its influence can be found in almost all schools of economic thought. At the same time, its fundamental features are being used only by a limited group of economists today; since others have opted to modify the Classical assumptions to deduce results which are more Keynesian than Classical.




Baumol, W.J. (1977). Economic Theory and Operations Analysis. Prentice-Hall.

Bruno, Michael (1990). "Theoretical Developments in the Light of Macroeconomic Policy and Empirical Research" in The State of Macroeconomics, edited by Seppo Honkapohja,1990, Basil Blackwell.

Day, R.H. (Fall, 1994) Class discussions. The book Complex Dynamics (1994) and other notes were used for Q.1.

Debreu, Gerard (1959). Theory of Value. Cowles Foundation Monograph.

Grandmont, Jean-Michel(1990). Keynesian Issues and Economic Theory in The State of Macroeconomics, edited by Seppo Honkapohja,1990, Basil Blackwell.

Hicks, John (1992). The Unification of Macroeconomics, in Macroeconomics: A Survey of Research Strategies edited by Vercelli, A, and Dimitri, N.

Honkapohja, Seppo (1990). "Comment on J-M. Grandmont, ‘Keynesian Issues and Economic Theory’", in The State of Macroeconomics, edited by Seppo Honkapohja,1990, Basil Blackwell

Keynes, John Maynard (1936). The General Theory of Employment, Interest and Income. Harvest, 1964.

Leijonhufvud, Alex (1992). Keynesian Economics: Past Confusions, Future Prospects, in Macroeconomics: A Survey of Research Strategies edited by Vercelli, A, and Dimitri, N.

Nagatani, Keizo (1981). Macroeconomic Dynamics. Cambridge University Press.

Smith, Warren L. (1956) A Graphical Exposition of the Complete Keynesian System.

Stiglitz, Joseph E. Methodological issues and the New Keynesian Economics, in Macroeconomics: A Survey of Research Strategies edited by Vercelli, A, and Dimitri, N.