I. Introduction

The role of price liberalization—its scope and timing—in the overall process of transforming socialist into market-oriented economies is one of the main issues facing transitional socialist economies. Price liberalization raises fundamental issues, both at a macroeconomic and a microeconomic level. However, this paper does not discuss the issue of what appropriate macroeconomic policies have to accompany the process, nor the establishment of an efficient relative price structure. ^1/ Instead, this paper examines how liberalizing prices in economies with limited competition among suppliers can affect output and consumption.

Recent experience shows that the extent of price liberalization varies significantly across countries. One of the arguments in favor of the “gradual” approach to price liberalization which has been followed in many countries is based on the idea of monopoly pricing. Specifically, it is argued that given the highly monopolistic structure of an economy, price liberalization enables monopolists to appropriate rents, with no major improvements in the overall efficiency of the economy or the welfare of the population. ^2/ Widespread price controls conflict with a movement toward a market economy and, through their effects on the budget deficit, tend to destabilize the macroeconomic situation. ^3/ Moreover, price controls can hinder or considerably slow the structural changes in the economy, including the privatization process. ^4/

Therefore, some have proposed that the solution lies not in a piecemeal approach of gradual price reform but, rather, in a comprehensive approach that combines price liberalization with measures affecting the supply side of the economy, namely the problem of entry and ownership control of enterprises. Price liberalization in the presence of large monopolistic rents can increase monopolistic profits while decreasing consumption, and cause the economy to stall at a low consumption/production equilibrium. Nevertheless, eliminating monopolistic rents can be difficult, as shown by the actual experience of transitional socialist economies. Both eliminating existing barriers to entry and privatizing the economy’s productive structure requires a considerable amount of time. Under these circumstances, a role for fiscal policy may be envisaged. By redistributing monopolistic rents, a tax/subsidies policy may improve welfare. This paper will formally investigate this argument.

The plan of this paper is as follows. In the following section the basic argument underlying the policy conclusions is exposed. In Section III, the structure of the model is described. In Section IV, equilibrium is characterized formally for both a classical socialist economy and a transitional socialist economy; the role of price liberalization is then discussed. In section V, the equilibrium outcome of price liberalization in economies characterized by limited competition is contrasted to price liberalization in full market economies in which barriers to the entry are eliminated. In Section VI some policy implications of the model are discussed.

II. The Argument

The Polish experience suggests that in order to yield positive effects on the real economy, price liberalization has to be complemented by reforms affecting the market structure, and thus enterprise behavior. In connection with the recent Polish “big bang,” it has been argued that price liberalization may not be sufficient to give the “right” signals to economic actors, unless the state relinquishes the ownership of the supply side of the economy and the existing production trusts are dismembered. ^1/ The idea is that managers of state-owned firms enjoy participation rents, which preclude access to the market to potential competitors. Under these circumstances, price liberalization may lead to a low output/consumption equilibrium, by providing additional rent opportunities as new firms cannot enter the market. In this regard, the Polish experience may be particularly relevant. The failure of supply to respond to price liberalization is puzzling, as the dimension of the output drop is not fully explained by the need of adapting production to the new relative price structure. ^2/ Such an outcome could represent a low output/consumption equilibrium, which has nothing to do with an efficient restructuring of the economy, and may actually jeopardize growth prospects and the sustainability of reforms in these countries. In the simple formal model developed in the next section, it is shown how this low output/consumption equilibrium can arise as a result of the lack of appropriate structural reforms accompanying price liberalization. ^1/ However, the low equilibrium is not a warranted outcome, as the model may give rise to a multiplicity of equilibria. In the context of the model, policies are examined that could contribute to a better equilibrium outcome. Policies may affect the equilibrium in two ways, by encouraging market participation on the supply side of the economy, or by influencing expectations.

The model is constructed in the “search equilibrium” tradition. The choice of this analytical framework reflects the belief that traditional demand and supply models are inadequate to study socialist economies in transition, as they presuppose the existence of the basic market institutions. In these models, price liberalization is interpreted as the removal of price distortions in a context in which demand and supply schedules do not differ substantially from the ones in a market economy. Accordingly, rationing and inefficiencies arise only because the “social planner” has chosen the wrong vector of relative prices. In the actual experience of transitional socialist economies, price liberalization represents a step toward the introduction of markets, suggesting the inadequacy of analyses based on the assumption that markets are functioning, although distorted, by “wrong” prices. The “search equilibrium” approach does not have these drawbacks. By focusing on the basic mechanism of exchange, one does not need to assume the existence of all the institutions and behaviors typical of market economies. However, the model describes economic behavior from first principles and is in the equilibrium tradition in the sense that the agents behave rationally.

Despite its simplicity, the model accounts for some fundamental features of transitional socialist economies. The equilibrium outcome in centrally planned economies will be associated with both a low probability that a consumer will find a good on the market and with a large number of consumers searching for goods; in this sense, the model can explain shortages as the presence of a large number of consumers in the market despite their small chance of procuring the good. It is important to stress that along these lines shortages result from optimal agents’ behavior in a setting where rationing does not result from price rigidities. However, price liberalization will reduce shortages by reducing the number of buyers in the market. It is also possible that aggregate consumption will decline. Interestingly, this seems to be consistent with the actual experience of transitional socialist economies.

The model permits an equilibrium analysis of both consumers’ and firms’ behavior and identifies trade frictions, which reflect features of socialist economies. This is true not only for the consumers’ but also for the producers’ behavior. ^1/ In fact, delivering goods from the production sites to the market place is often difficult as markets are highly segmented and shipment costs are high, giving rise to rent opportunities. Since firms are state-owned, the model takes production decisions as a given, which is certainly the case under central planning, but it focuses on market participation decisions.

Focusing on the effects of price liberalization in a socialist economy in transition, this paper argues that if not accompanied by the elimination of existing economic rents, price reform may yield undesirable outcomes—and actually reinforce the structural problems of the economy. The model presented has important implications for the role of prices in the economy. Because of the presence of trade frictions, the price will not clear the market in the Walrasian sense, as there is no obvious demand and supply to equate. Prices will be determined by a bargaining process between the buyer and supplier, This aspect differentiates this model from “disequilibrium” models which have been extensively applied to socialist economies (see Portes (1986)).

III. The Model

The model is in the “search equilibrium” tradition of Diamond (1984) and Mortensen (1989). The economy comprises two types of agents—suppliers and buyers. For simplicity, no heterogeneity is allowed within each type. Suppliers produce a basic good that can be either durable or perishable; unit costs are fixed. Buyers consume it, deriving a constant flow of utility in each period; the model assumes that the utility derived from the consumption of the good is larger than the utility derived from the consumption of the other goods. ^2/ The model abstracts from relative price issues by focusing on the consumption good. In this sense, the analysis will be in a partial equilibrium framework, so that the price of the basic consumption good is expressed in terms of a basket that includes the other goods in economy, with the price normalized to unity.

Since there are exchange frictions and transaction lags, there is not a market in the Walrasian sense, whereby demand and supply are equated. Trades can occur only if a supplier meets a buyer, as agents undergo a costly search process. Agents have to locate agents of the opposite type before an exchange can take place. Consequently, there is a positive probability that a buyer will go to the market and not find any partner with whom to trade.

As for price determination, it is well known that in search equilibrium models prices are indeterminate, since there is no obvious demand and supply to equate. In other words, the presence of transaction costs and lags—which in the model are captured by a trade technology—gives rise to an exchange surplus that must be split between trade partners. As long as trade participation costs are covered, any sharing rule will satisfy voluntary participation in the exchange. Thus, in this class of models, prices are usually set assuming an arbitrary bargaining rule that specifies the relative bargaining power of the parties—the buyer and the supplier in the model. Two different price rules will be studied, one conferring the entire bargaining power on the consumer, and one allowing equal bargaining power to the buyer and the supplier. ^1/ The former can be thought of as describing price determination in a classical socialist economy, the latter as describing price determination in a transitional socialist economy. This description of pricing in a classical socialist economy intends to capture the fact that consumption prices are set by the state and cannot be influenced by the producer. In classical socialist economies, consumption prices are often kept artificially low by planners’ decisions (through the application of subsidies) whereas in reforming socialist economies producers are allowed to increase prices.

The model characterizes equilibrium in transitional socialist economies as an intermediate step between a “socialist” and a “market” equilibrium. The transitional socialist economy shares with the classical socialist economy the presence of economic rents on the production side of the economy, while it shares with the market economy the pricing rule. Indeed, it has been suggested (see Weitzman(1989) that the main difference between socialist and market economies is that the former can be characterized as a seller’s market, whereas the latter is a buyer’s market. Therefore, it will be assumed that in socialist economies there are substantial participation rents for the suppliers, whereas competitive pressures bring to zero the value of these rents in a full market economy.

IV. Equilibrium in Classical and Transitional Socialist Economies

Consider an economy producing a basic consumption good depreciating at the rate δ, 0 < δ ≤ 1. The case of δ ≤ 1 is studied, because in command economies the reuse value of consumption goods is usually considered high. ^2/ Let’s call b the number of buyers and s the number of suppliers of the good. Trades occur according to a trade technology m(b, s). The function m(.,.) is homogeneous—i.e. m(b, s) = s^km(b, 1) —, increasing in both its arguments; furthermore, m(0, s)=m(b,0)=0; finally m(b, s)/b and m(b, s)/s are decreasing in b and s, respectively. ^1/ One can think of m(.,.) as generating an output consisting of the number of trades as a function of buyers and suppliers as the inputs.

The number of potential suppliers, denoted by S, is a given. Each supplier produces a unit of the good before trying to sell it and is engaged in either producing the good or searching for customers, so that if x denotes the number of agents engaged in production, it must be the case that S = s + x; each supplier sells a unit of the good. It can be assumed that after a trade occurs both the supplier and the consumer abandon the market, the latter to consume the good and the former to produce in order to replace the sold quantity. At the same time, the flow of suppliers out of production activity and into search activity will occur at an exogenous rate τ, 0 < τ ≤ 1, so that this flow will be equal to τ [S-s], since the number of agents engaged in production will be equal to the difference between the total number of potential suppliers in the economy (S) and the number engaged in selling the good (s). Therefore, production, x, will follow the law of motion:

since the flow into production is given by the number of suppliers able to sell the good—m(b, s).

Similarly, consumption, y will follow the law of motion:

where a dot over a variable indicates a time derivative. Indeed, m(b, S-x) represents the flow into consumption activity and δy is the depreciation of consumption and hence represents the flow out of consumption.

The dynamic maximization problem for the typical agents can now be studied. The typical supplier will maximize the expected present value of his future income stream; if r is the exogenous interest rate, and the good is treated as an asset, optimality will require that the rate of return times the value of the asset equal the flow of income plus the expected capital gain. The only decision the agents have to take is whether to participate in the search process. Whenever a buyer and a seller meet, trade occurs. Therefore, the expected present value of the income deriving from bringing a unit of the good to the market for the representative seller, V, will be:

where r is the interest rate; λ is a constant term summarizing the presence of rents for the suppliers, alternatively λ is a government subsidy to the supplier; ^1/ c_s is a fixed search or fixed selling cost, also it is assumed that λ > c_s; m(b, s)/s is the rate at which the supplier finds a buyer. J=p-q, where q is a constant unit production cost and p is the price. In other words, the opportunity interest on holding the asset, the good for sale, must be equal to the rate at which the good is sold per period, multiplied by the capital gain associated with selling the good, plus the expected present value of future profits attributable to holding a unit of the good for sale.

Similarly, for the buyer, one can define the expected present value of the utility deriving from holding a unit of the good for consumption, W, as follows:

where z is a utility index for the services derived from the consumption of the good, p its price, and U the capital value of searching for goods to buy. Since the consumption good depreciates at an instantaneous rate δ, the asset pricing condition for the capital value of the good to the consumer will be given by its instantaneous utility minus the loss due to depreciation plus any future gain attributable to holding the good. A similar argument yields the condition for the capital value of a buyer looking for a supplier:

where c_b is a fixed search cost, m(b, s) /b is the rate at which a buyer finds a unit of the good, i.e., a supplier. In socialist economies, since all firms are state-owned, the market participation rents for the suppliers will be positive, and are summarized by the term λ, i.e V(t) > 0, for each t. Moreover, the production decisions are out of the suppliers’ control, since suppliers will engage in production when not engaged in search activity.

On the side of buyers, it is assumed that competition among them drives their participation rents to zero, i.e, U(t)=0 for each t. This reflects the observation that in transitional socialist economies the number of buyers overwhelms the number of suppliers. The number of buyers is instantaneously determined, whereas the number of sellers is predetermined according to equation (1). U=0 implies that c_b = [m(b, s) /b] W, thus it defines b as an implicit function of W. From the assumption of homogeneity of m(.,.)i it follows that:

Using the fact that s = S-x:

where b/s solves the free entry condition (5) for the buyers. The function π (.,.) gives the sale frequency per seller; we have that ∂π (.) /∂w > 0, and ∂π(.) /∂x <0 if k>1, ∂π(.) /∂x > 0 if k < 1, ∂π(.) /∂x = 0 if k=1.

Equation (1a) and (1b) can thus be rewritten as

The other dynamic equation of the system will be:

where the price p has been written as a function of x.

This economy will be studied under three different pricing rules to be characterized as classical socialist, transitional socialist, and market economy. In a classical socialist economy, it is assumed that the planners enforce a rule which guarantees that consumers fully appropriate the exchange surplus stemming from transaction costs and lags, i.e., that suppliers do not play a role in determining the price and that prices are subsidized in favor of the consumer; price liberalization in the transitional phase is identified with a price rule determined by a bargaining process in which suppliers and consumers split equally the exchange surplus, reflecting both suppliers affecting the price at which they sell the good, ^1/ and the end of subsidized pricing. However, rents are still large so that λ > c_s and V is positive as in the classical socialist economy. In a full market economy, the symmetric Nash bargaining rule is accompanied by the condition that participation rents are driven to zero for the suppliers i.e λ = 0 and V = 0. The idea is that flexibility of prices is not the only important element characterizing the transition from planned to market economy, but that to implement an efficient result it has to be accompanied by the elimination of rents on the supply side of the economy.

Denote by T the total exchange or trade surplus, which represents the total gain from a transaction for both the buyer and the seller: T = W+J-U-V. The exchange surplus T represents the difference between the joint returns from exchange and the joint expected returns from re-engaging in the search process.

Define

Θ = [W-U]/[W+J-U-V],

as the share of the exchange surplus going to the buyer—and thus (1-Θ) is the share going to the supplier.

1. Classical socialist economy

In a classical socialist economy, it is assumed that Θ is set equal to 1, reflecting an institutional setting such that the bargaining process is solved completely in favor of the consumer as a result of the fact that consumption prices are subsidized. This assumption needs some justification. In the actual experience of a command economy suppliers accept the price given by the central planner, hence play no role in price determination. In fact, it has been suggested that prices provide no production incentives, and that supply decisions are mainly the result of rent seeking decisions. ^1/ In this sense, assuming that the exchange surplus is allocated entirely to the consumer, captures an important feature of price determination in classical socialist economies. Furthermore, this price rule can be interpreted as optimal from the planners’ point of view, since it will maximize aggregate consumption for a given level of rents λ. ^2/

The participation rents are created by state control of productive activity and barriers to entry. These rents together with the pricing rule characterize a classical socialist economy in the model. In other words, the number of suppliers is small because there is no free entry to the market and this, in turn, generates participation rents; at the same time, free entry on the demand side together with a low price for the good brings about a large number of buyers and, therefore, a low rate at which buyers find suppliers. In this way, the model also describes and motivates the presence of shortages in command economies that may be interpreted as a low buyers probability of finding a unit of the good for purchase. Formally, shortages, σ, are defined as the difference between the number of buyers on the market and the number of exchanges taking place:

σ = b - m(b, s).

A conclusion of the model is that shortages will be larger in a classical socialist economy than in either a transitional socialist or a market economy. ^1/

Assuming that Θ=1 implies that J−V=0, then p = q + (λ−c_S)/r. Consequently, the price is fixed and is not a function of consumption. Equilibrium is thus perfectly determinate. This case is illustrated in Figure 1. Equilibrium will be characterized by the following differential equations:

View Full Size

• Download Figure
• Download figure as PowerPoint slide

$\begin{array}{cc}\dot{\mathrm{W}}=(\mathrm{r}+\delta )\mathrm{W}+\mathrm{q}+[(\lambda -{\mathrm{c}}_{\mathrm{S}})/\mathrm{r}]-\mathrm{z}& \left(10\right)\end{array}$

$\begin{array}{cc}\dot{\mathrm{y}}=\pi (\mathrm{W},\mathrm{x})(\mathrm{S}-\mathrm{x})-\delta \mathrm{y}& \left(11\mathrm{a}\right)\end{array}$

$\begin{array}{cc}\dot{\mathrm{x}}=\pi (\mathrm{W},\mathrm{x})(\mathrm{S}-\mathrm{x})-\tau \mathrm{x}.& \left(11\mathrm{b}\right)\end{array}$

To study the dynamics of this system, one can focus on the differential equation describing the change in consumption, as production and consumption will differ to the extent δ < τ. If one sets κ=(δ/τ), the steady state equilibrium will be given by the cross of the two singular curves:

$\begin{array}{cc}\mathrm{z}-\mathrm{q}-\left[(\lambda -{\mathrm{c}}_{\mathrm{S}})/\mathrm{r}\right]=(\mathrm{r}+\delta )\mathrm{W}*& \left(12\right)\end{array}$

$\begin{array}{cc}\delta \mathrm{y}*=\pi (\mathrm{W};\kappa \mathrm{y}*)(\mathrm{S}-\kappa \mathrm{y}*)& \left(13\right)\end{array}$

as in steady state x*=κy*.

It is easy to establish that the consumption singular curve—i.e., the curve describing the steady state points for consumption—will be increasing in the y, W space. Its positive slope reflects the fact that higher expectation about future returns from consumption is needed to sustain a higher level of steady state consumption. Formally, this is immediate as the right-hand side of the equation will be decreasing in y and increasing in W. ^2/ As W* is not a function of consumption, y, under the classical socialist price rule, it will be the case that a unique equilibrium will exist. This is illustrated by Figure 1. Note that the pricing rule chosen to describe pricing behavior in a classical socialist economy not only describes well the fact that consumption prices in Eastern European countries are generally low in terms of the other goods, but also can be characterized as rational from the social planner’s point of view as it will maximize y, unless there are increasing returns to scale in the trade technology m(b, s).

2. Transitional socialist economy

A transitional socialist economy can be characterized as a change in the institutional set up such that both parties have some bargaining power, reflecting both the fact that suppliers can freely set their price and also that buyers play a role. Formally, this assumption can be stated as θ=1/2, and this solution is known in the literature as “Nash symmetric bargaining”; it should be noted that the results of the model would not be substantially modified for any θ < 1. However, it is assumed that suppliers still enjoy positive participation rents, i.e., λ > 0 and hence V > 0. In other words, we consider an economy that is different from the classical socialist case only because the planner does not fix prices any longer but leaves the suppliers and the buyers free to bargain. From θ=½ it follows that W-U = J-V, that in turn implies:

$\begin{array}{cc}\mathrm{p}(\mathrm{W},\mathrm{x})=[1/(1+\mathrm{r})]\{\mathrm{q}\mathrm{r}+\mathrm{z}+(\lambda -{\mathrm{c}}_{\mathrm{s}})+\left[\pi (\mathrm{W},\mathrm{x})-\delta \right]\mathrm{W}\}.& \left(14\right)\end{array}$

We first consider the case in which the trade technology exhibits increasing returns to scale, then the case in which it exhibits decreasing or constant returns. This technological assumptions on the function m(.,.), summarize the possible presence of trade externalities in our search economy. If there are positive trade externalities, there will be increasing returns to scale; if there are congestion externalities there will be decreasing returns to scale, whereas If neither of two effects are present, there will be constant returns to scale.

a. Increasing returns to scale in exchange

The idea of increasing returns in the exchange technology, intends to capture the presence of a positive trade externality, making easier trading the larger is the market size. This externality is likely to be present at low levels of economic activity, whereas the existing empirical evidence seems to deny its relevance for higher levels of economic activity. ^1/ Therefore, this assumption cannot be ruled out for transitional socialist economies, which are usually characterized by low levels of economic activity.

If k > 1, i.e., m(.,.) exhibits Increasing returns to scale, it will be the case that ∂p/∂x < 0, and ∂p/∂W > 0. ^2/

Let us call z(W, y) the buyer’s pay-off function, i.e., z(W, x) = z - p(W, x). One can now study the system of differential equations:

$\begin{array}{cc}\dot{\mathrm{x}}=\pi (\mathrm{W},\mathrm{x})[\mathrm{S}-\mathrm{x}]-\tau \mathrm{x}& \left(15\mathrm{a}\right)\end{array}$

$\begin{array}{cc}\dot{\mathrm{y}}=\pi (\mathrm{W},\mathrm{x})[\mathrm{S}-\mathrm{x}]-\delta \mathrm{y}& \left(15\mathrm{b}\right)\end{array}$

$\begin{array}{cc}\dot{\mathrm{W}}=(\mathrm{r}+\delta )\mathrm{W}-\mathrm{z}(\mathrm{W},\mathrm{x})& \left(16\right)\end{array}$

The singular curves will be:

$\begin{array}{cc}\pi (\mathrm{w},\kappa {\mathrm{y}}^{*})[\mathrm{S}-\kappa {\mathrm{y}}^{*}]=\delta \kappa {\mathrm{y}}^{*}& \left(17\right)\end{array}$

$\begin{array}{cc}{\mathrm{W}}^{*}(\mathrm{r}+\delta )=\mathrm{z}({\mathrm{W}}^{*},\kappa \mathrm{y}).& \left(18\right)\end{array}$

The singular curve for consumption will be increasing in the y, W space, whereas the locus of steady state points for the expected value of consumption, W, will now be a function of consumption through its effect on the price. Expressly, W will increase with y if k > 1, since the positive trade externality will produce a negative effect of aggregate consumption on the price.

Under these circumstances, the dynamics may be much more complicated, as the two singular curves may cross more than once, and give rise to a multiplicity of steady state equilibria and, therefore, also to an indeterminacy of the equilibrium. For instance, in the case illustrated in Figure 2, there are three crosses. Of the three equilibria consistent with price liberalization, two (A and B in Figure 2) are associated with two different saddle paths (S₁ and S₂), while the third (C) is either a source, or for certain values of r and δ, generates stable cycles around itself. The main result is that price liberalization does not necessarily guarantee a superior equilibrium—here defined in terms of the output/consumption level. Indeed, the economy can get stuck in a low consumption equilibrium, as steady states of the liberalized economy can imply lower output than in the classical socialist economy. Moreover, depending on the steady state that arises, interesting dynamic paths, eventually associated with stagflationary outcomes, with increasing prices and declining output, may arise. The reason for the multiplicity of equilibria has to do with the reinforcing effect that the presence of increasing returns has on the agents’ participation decisions. In particular, the high equilibrium path may result from the positive trade externality more than offsetting the negative effect of an increase in buyers’ participation on the expected returns from consumption. Note that this mechanism operates also for a decrease in buyers’ participation, hence the possibility of the two saddle paths. This example illustrates the importance of consumers’ expectations and suggests a role for policy in affecting this expectations. However, its relevance for policy analysis depends on the realism of the increasing returns to scale assumption. Indeed, if market size externalities are not important and, therefore, one thinks the trade technology exhibits constant returns to scale, the equilibrium will be unique; this will also be the case if there are congestion externalities that are described by assuming decreasing return to scale.

View Full Size

• Download Figure
• Download figure as PowerPoint slide

b. Decreasing or constant returns to scale in exchange

Decreasing returns to scale in the trade technology intend to capture a negative congestion externality, affecting trade negatively as the size of the market increases. If k < 1, i.e., the trade technology—m(b, s)—exhibits decreasing returns to scale, the price will be an increasing function of consumption, y. Therefore, the consumer pay-off function z(W, ky) will decrease with y, when the price is set according to the Nash rule, i.e., Θ = 1/2. Under these circumstances, the singular curve (17) for the expected value of consumption, W*, will be decreasing with y. Thus, a unique equilibrium will exist, since the singular curve for consumption, y*, will increase with W as it does in the increasing returns to scale case. However, the equilibrium will be associated with a lower level of consumption than in the classical socialist economy, when Θ=1. ^1/ This is illustrated in Figure 3, where the singular curve for W has been drawn, showing that the intercept with the vertical axis is lower when Θ=1/2 than when Θ=1.

View Full Size

• Download Figure
• Download figure as PowerPoint slide

Therefore, under decreasing returns to scale, price liberalization produces a low consumption equilibrium. The reason why a low equilibrium may arise following the change in price setting is that a price increase affects participation on the buyers’ side negatively as the returns to search activity decrease, since the right hand side of (2) goes down, whereas the presence of participation rents will prevent an offsetting increase in the number of suppliers.

Under constant returns, neither of these two externalities are at work. The constant return to scale case is different from the decreasing returns case only because then the singular curve will be constant since there are no trade externalities and, therefore, no feedback effects of aggregate consumption on prices; however, the equilibrium consumption level will still be lower than under the socialist pricing rule. This is illustrated in Figure 4.

View Full Size

• Download Figure
• Download figure as PowerPoint slide

V. Equilibrium in a Market-Type Economy

The purpose of this section is to show that the possibility of inferior equilibria associated with price liberalization is linked to the presence of rents on the supply side of the economy. This section analyzes a market-type system in which the pricing rule is identical to the one characterizing the transitional socialist economy, i.e., ϴ-½ but the rents are eliminated, i.e., V=0 as λ = 0. After characterizing equilibrium in this new environment, the equilibrium level of consumption in the full market economy will be compared with the one prevailing in the transitional socialist economy.

Setting Θ=1/2 when participation rents are driven to zero—i.e., when λ=0 and V(t)=0 for each t—will imply that J = W, solving for the price:

Substituting the above expression for the price in the steady state condition for the expected value of consumption will give

The equilibrium will then be unique both under decreasing and increasing returns to scale, and there will be no possibility of complicated dynamics of the type described for the transitional socialist economy with increasing returns to scale. This case is illustrated in Figure 5.

View Full Size

• Download Figure
• Download figure as PowerPoint slide

Since the singular curve for consumption is increasing in W, one can compare the equilibrium in a classical socialist economy with the one prevailing in a full market economy. Let’s call W_s and W_m respectively the equilibrium level of W prevailing in the classical socialist economy and in the fully market economy. If W_m > W_s, then equilibrium in a full market economy will be associated with a larger level of consumption than in the classical socialist economy. We will have that W_s (>) (<) W_m if

From the above inequality, it is evident that a large rent, summarized by λ, will be associated with a smaller consumption level in the classical socialist economy than in the full market economy. Suppose λ = 0, then the pricing rule Θ = 1 would implement an equilibrium with a higher consumption level than in the case Θ=½. This illustrates the fact that it is the presence of economic rents that make the classical socialist equilibrium outcome inferior to the market one.

VI. The Role of Government Policies in the Transition Process

It is not surprising that the presence of externalities in the search process creates a role for Pigouvian taxes and subsidies. However, here we are interested in describing policies which might minimize the output loss consequential to price liberalization in a context in which participation rents cannot be eliminated quickly, rather than in describing the ways in which policy may internalize the search externalities. The idea is that structural reforms—such as privatization or the definition of property rights—may require time to be implemented whereas price liberalization may be quicker, as shown by the Polish experience. Under these circumstances, there could be a role for fiscal policy—to produce incentives to influence the agents’ participation decisions so that the negative effects of the participation rents are partially offset. In the model, although price liberalization gives bargaining power to the supplier, it does not produce a supply response because of restrictions to entry. Instead, there is a drop in market participation on the buyer side, causing the economy to stall at a low equilibrium level of activity. This mechanism creates a role for policy, as a lower consumption/output outcome can be avoided by redistributing rents from the suppliers to the buyers. A consumption subsidy—financed by taxing suppliers’ rents—could enhance buyers’ participation, avoiding a drop in production that could result from too many buyers abandoning the market. At the same time, this would result in a better bargaining position for the buyers and therefore limit the extent of the price increase. In a sense, tax policy could be used to affect the result of price bargaining by affecting the “threat-points” ^1/ of the Nash bargaining process. It is important to stress that this tax subsidy policy should not have a negative effect on the fiscal budget, as it should be self-financing.

Policies will defer depending on the features of the trade technology. If the function m(.,.) exhibits decreasing or constant returns to scale, so that the equilibrium is unique, lump sum subsidies to search activity would be welfare improving. This subsidy could be financed through a lump sum tax on the suppliers equal to the difference between λ and c_s. The subsidy to search activity would shift the output/consumption singular curve to the right, as a lower level of expected utility from consumption activity would be needed to support the same steady state level of output/consumption. The lump sum tax on supplier rents would move up the intercept of the W singular curve with the vertical axes, as a lower level of output/consumption would be needed to support the same steady state level of expected utility from consumption activity. While it remains to verify if lump sum transfers are feasible. Nevertheless, the model suggests that tax policy may have an important role in minimizing transition costs and that subsidies to consumption could allow the economy to escape the low activity equilibrium.

If, instead, trade is affected by positive externalities—making for easier exchange the larger the market—i.e., if the trade technology m(.,.) exhibits increasing returns to scale, there are likely to be multiple equilibria. There is scope for policy to affect expectations, and hence the steady state where the economy will end up eventually. In this case the government could play an important role in promoting confidence or optimism within the private sector. Tax policy could be exploited to eliminate the possibility of the low steady state by shifting the intercept of the singular curve for the expected returns from consumption upward. However, a lump sum subsidy to consumption might eliminate both the steady state A and steady state C, leaving S₂ as the only possible equilibrium path. Unlike the case of a unique equilibrium, here a subsidy to consumption could be financed by future taxes if the objective of policy is to move the economy from the lower to the higher equilibrium path. Indeed, if the economy is at a lower equilibrium steady state—point A in Figure 2—a temporary tax on the suppliers and a subsidy to the buyers could rule out the low steady state leaving the high equilibrium path as unique. This policy could be reversed when the economy reach’s the higher steady state, without causing a movement back to the lower equilibrium path.

VII. Conclusions

The main conclusion of this paper is that the process of price liberalization in transitional socialist economies could result in substantial output losses if the competitive structure of the economy is not quickly modified to eliminate supply rents. In this sense, price liberalization has to be seen as only one, albeit important, step in the transition from a centrally planned to a market economy. Accordingly, this paper stresses the urgent need to eliminate the existing barriers to the entry on the supply side of the market. However, since this task may require time, it is suggested that a tax subsidy policy play a role in limiting the output loss. This paper shows that policies that encourage buyers’ participation—such as transferring rents from the suppliers to the buyers—could succeed in avoiding a low activity equilibrium outcome.