FINANCIAL MARKETS AND ECONOMIC DEVELOPMENT: MYTHS AND INSTITUTIONAL REALITY

It is a commonplace that economics is about markets. Among markets the stock exchange has a mythical role as representing the epitome of the operation of perfect competition. Recently, it has also taken on a mythical role as a necessary condition of economic development. Here I attempt to inject some realism into these myths by starting with the idea that Leon Walras, normally vilified as having completely ignored institutions and the evolution of economic systems, should in fact be classified as an institutional economist.

The argument starts by challenging the idea that Walrasian general equilibrium theory is an abstract aberration of a mad French intellectual with no application to the real world. This is just not so, with respect to Walras's original theory, and also with respect to its extensions by Arrow and Debreu. Despite criticisms from post-Keynesians and institutional economists that the theory is devoid of real-world content, and the claims of applied general equilibrists that it can be applied independently of any real-world institutions, the theory of price formation of the Elements (1954) is a rather accurate rendition of operations of a particular institution, the Paris Bourse. If it can be criticized it is because it has little applicability outside of the historical context and particular conditions it describes. It is 'institution-bounded'.

But if the theory is institution-bounded this also calls into question the generally accepted proposition that competitive Walrasian markets allocate resources efficiently and thus promote growth by channelling resources to their most productive uses. Most economists, when pressed for an example of efficient market allocation, refer to the stock market. But the idea of efficient allocation, or a Pareto allocation, is just a description of the distribution of shares after the determination of the equilibrium prices on the Paris Bourse. If Walras's theory of price determination is institutionally bounded, then the concept of an optimal allocation is also bounded to the assets that can be traded in such markets.

What financial markets deal in are legal claims to the income deriving from the ownership of resources; they exist because of the requirement that in a capitalist economy everything must be owned by someone, but it need not be perpetual ownership. Stock markets thus provide for the distribution and redistribution of those claims. But it is not the resources that they distribute amongst users; rather, they distribute the losses and gains that arise from the failure of the market to distribute resources efficiently to the areas of highest productivity. They do this by adjusting the prices of claims so that their returns are equalized, taking into account realized rates of returns and expected future profits. The market allocates capital gains and losses that arise from investment errors, not the resources themselves. If there were no errors, then there would be no gains or losses or price adjustments, since all resources would be allocated so as to produce a uniform return. This is not an unimportant task, simply a rather different one than is usually claimed for financial markets.

Having challenged two of the myths surrounding the operation of stock markets, we can consider the myth that Walrasian financial markets are more efficient at channelling new resources to investment, and thus contribute positively to growth and development. This myth has gone so far as to be used to justify the creation of securities markets in developing countries and the Central and Eastern European transforming economies in order to increase their efficiency in the use of their scarce available resources. Even sub-Saharan Africa has become a target, and countries such as Ghana, Nigeria and Botswana have recently created stock markets in the hopes of stimulating their development efforts.

The experience of the now developed countries suggests that it is not financial markets, but banks that determine the allocation of resources through the creation and allocation of credit. Banks not only provide the positive impetus to growth by overcoming liquidity constraints, but are also crucial to the creation of the financial markets in general and stock markets in particular.

In this context we can deal with an associated myth: that 'markets' might some day produce enough disintermediation to replace banks. Since most organized markets could not exist without bank credit (the Walrasian market providing an important exception to this rule), banks and markets should be considered as complements, not substitutes. Remember that J. P. Morgan managed to control US financial affairs without ever setting foot on the floor of the New York Stock Exchange. Indeed, some say he professed not to know where it was, despite his office being just across the street.

The rest of this chapter will start from the provocation that Walras should in fact be considered an honorary, if posthumous, member of EAEPE, and that the recognition of his contributions in this area may be used as a means of assessing the exaggerated claims for generality that have been made for the concept of the efficient allocation of resources via equity markets in the process of development.

Walras and the Real World of Financial Markets – Circa 1860

The modern habit of referring to stock markets as examples of efficient markets is based on purely circular reasoning. It is but a reflection of Walras's clear intention to produce a theory that mirrors and reflects the operation of the stock market! Walras (1954, pp. 83–84) in the Elements declares that:

we must go to the market to study value in exchange. [...] The markets which are best organized from the competitive standpoint are those in which purchases and sales are made by auction, through the instrumentality of stockbrokers [...] This is the way business is done on the stock exchange [...] Let us go into the stock exchange of a large investment centre [...] Let us take for example trading in 3 per cent French Rentes on the Paris Stock Exchange.

Walras promises a theory that will render 'perfectly comprehensible' 'the confused uproar and chaotic movement', the 'clamour and bustle', that impresses any visitor to a stock exchange. In the type of market organization that prevailed on the Paris Bourse when Walras was writing, brokers met once a day at the same time in the same place to trade an officially approved list of stocks. Trading was organized by an employee of the exchange who started the trading session by quoting an opening price for the first stock on the list. Brokers signalled the size of the buy or sell orders given by their clients for that price; if the orders did not match, a new price was called, lower if offers predominated, and vice versa. This auction process of 'groping' or tâtonnement only stopped when the market clearing price was discovered. This was the 'equilibrium' price, also known as the price 'fix'. All orders to buy or sell at this price are executed, as are orders that can be matched to buy at a higher price or sell at a lower price. There is a 'single price' for all bargains executed.

As Walras points out, what the agents de change carry with them in their little leather carnet of orders is the equivalent of their clients' demand and supply curves. The process of tâtonnement carried out by the 'auctioneer' simply serves to make this information public and allows him/her to determine the equilibrium price as if it were the intersection of the market supply and demand functions that set the equilibrium price. Once the price of the last stock is fixed, the market is over, and no further trading takes place until the following day. It is thus a market that trades at 'discrete' intervals, with periodic trading 'suspensions', during which new orders are collected on the basis of new information, that reaches clients. Price formation could thus be considered to take place in conditions of perfect information since all orders that exist at any price are presumed to be in the possession of the agents de change when they arrive at the market and they are made public during the auction. The trading suspensions mean that trade takes place on the basis of given and known quantity and allocation of stocks in individual portfolios.

Note that these are not ad hoc assumptions that Walras imposes on his model in order to reach a particular desired result (as so often happens in current theorizing). They result from the official regulations governing the operation of the market. The French Commercial Code restricted trading to officially designated bourses via a given number of officially appointed agents de change acting as officials of the Crown. In exchange for their monopoly on trading, the agents were forbidden from participating in the auction by entering their own orders, preventing them from intervening to influence market supply and demand. At the close of each day's market session the auctioneer had thus produced a Pareto-optimal allocation of stocks across individual portfolios, for if this were not the case, individuals would have chosen to place their orders differently. Since all trades were voluntary, by definition any trade undertaken makes one side better off without making the other side worse off.

The idea that perfectly competitive markets produce an efficient allocation of resources is thus grounded in a particular institutional and regulatory framework, linked to a particular historical period. It is the market regulations that provide the conditions of perfect knowledge, the given quantities of stocks to be traded and the exogenous initial distribution of stocks required to specify the Pareto-optimal allocation.

There is little difference between the Paris Bourse and the theory in Walras's Elements, except that in a real call market prices are fixed sequentially, while Walras proposed a simultaneous system of price determination in which 'the whole world may be looked upon as a vast general market made up of diverse special markets' (Walras, 1954, p. 84); 'at the same time [...] trading is going on in [...] French Rentes, similar trading is taking place in [...] English, Italian, Spanish, Turkish and Egyptian [...] stocks and bonds; besides cash transactions there are future transactions, some firm and others optional' (original emphasis) (p. 86).

If demand is also determined by the prices of other stocks, this may require reopening trading on stocks that had already been called earlier and had their prices fixed. There were provisions in markets to 'call back' a stock (see Schwartz, 1991), and Edgeworth (1881) had suggested a similar process, which he called 'recontracting'. Both are similar to Walras's account given above since they are types of contingency contract, similar to futures and options, which were already prevalent in Walras's day. Thus, although simultaneous determination of prices could not take place in a temporal sequential call, the use of 'call backs' and futures and options contracts could provide a close equivalent. All modern extensions of the theory, from Debreu's (1959) Theory of Value onward, simply provide proof that such close equivalents in the form of contingent contracting could indeed replicate the results of simultaneous trading.

The basic justification for the organization of economic activity by means of such Walrasian competitive markets is their contribution to the 'efficient' allocation of resources. The definition of 'efficient' is that no other allocation can be achieved, which, starting from any given initial distribution of those resources, would improve the position of any individual without making another worse off. The elimination of Pareto improvements implies that all opportunities for profitable arbitrage have been exhausted. This is not to say that some other initial allocation could not produce greater overall wellbeing, but that the operation of competition in the market is presumed to be independent of the initial allocation. Our discussion of Walras's modelling of the Paris Bourse shows that this description of allocative efficiency is simply another way of defining the competitive price fix.

Criticisms that tâtonnement and recontracting are unrealistic because they abolish time by ignoring the sequential order of actual trading appear to be misplaced. The more relevant criticism is that Walrasian general equilibrium theory is based on a peculiar set of regulations and institutions that produced perfect information, given initial allocations and price determination via an auction procedure. The theory thus has no claim to generality since it only has application to a particular market, the stock market. The results of the theory, and in particular the propositions of allocational efficiency or the ability to produce Pareto allocations, do not generalize either across markets or across institutions or across historical periods.

That allocative efficiency is just as 'institution bounded' as competitive equilibrium can be quite easily seen by reference to Walras's contemporary, Alfred Marshall, who also tried to emulate the stock market in his theory of price. But as geography and history would have it, his real-world example was the older and differently evolved London Stock Exchange.

Marshall and Continuous Trading Markets

Marshall's point of reference, the London Stock Exchange, did not at all resemble the Paris Bourse. British clients were also represented by brokers who had exclusive access to the Stock Exchange, although not through government regulation, but as a member in what was essentially a private club. A client order received at any time during the trading day could be taken to the Exchange to seek out a counter party. This trading was usually done by consulting a number of brokers, known as 'jobbers', who bought and sold for their own investment purposes, in order to find the best available price. Other orders for the same stock might be negotiated at the same time by other brokers, or at other times during the day, and they could be executed at different prices, so that there was no guarantee of a 'single' price for all buyers and sellers, nor was there any possibility of complete knowledge of either the trades or prices occurring during the market day. Neither could the allocation of stock be taken as given and the quantities of stock outstanding brought to the market were a variable proportion of the total. There was no public information on transactions and reporting trades was voluntary.

In his presentation of price formation, rather than referring directly to the stock market, Marshall (1920, p. 332) discusses 'the corn market in a country town'. Actual prices vary over the day as individual buyers and sellers engage in a 'continuous' open outcry auction market in which repeated bilateral bargaining takes place throughout the day. The same calling out of prices takes place, but it is now the buyers and sellers themselves who do the calling, there is no auctioneer. In the London Stock Exchange this same process takes place, but it is the individual brokers, as agents for their clients (or the broker and the jobbers), who bargain until a price is agreed at which a specific exchange can take place.