Contenu du sommaire : Développements récents de l'analyse économique (XLIVè congrès annuel de l'Association française de science économique, 1995))

The result found by Diamond [1971] shows that imperfect information on thé consumer side drastically affects pricing on imperfectly competitive markets. This paper aims at summing up the literature that follows this route in a differentiated products context and to show that such a context is very appropriate to study the impact of consumer search on competition. Furthermore, taking into account imperfect consumer information sheds new light on the theory of markets for hete­rogeneous products.

The paper studies the optimal contract offered by a buyer to a producer, under asymétrie information on both the marginal cost and the fixed cost. The model illustrates the results obtained when generalizing the adverse selection Principal-Agent model by relaxing the assumptions on outside options. Asymétrie informa­tion can reduce or increase the volume of trade. The informational rent is not monotonie. Quantity distortions relate to the shape and the curvature of the func­tion linking the fixed cost to the marginal cost.

We survey the theory of common agency under complete information in which an Agent chooses an action (which may be pluridimensional ) affecting the well-being of several other individuals (the Principals) which offer monetary payments contingent on the action choosen.

This paper is an application to "marchés à commandes" of McAfee-Reny's result showing that rent extraction from informed agent is possible with correlated information. We determine sufficient conditions to lead efficient solution in this public market when we use temporal correlation as ex-post information.

This paper focuses on the choice of the access charge by an operator through a decentralized mechanism. We consider an incumbent firm which operates as an upstream monopolist in the infrastructure market but faces competition as service provider. We examine the impact of a verticaly differentiated service market and analyse the conditions under which the operator may be induced to let other firms compete in the downstream market.

Studying differentiated product markets has become one of the main topics in New Empirical Industrial Organization. This paper presents different methodological approaches that have been used. It stresses on their pros and cons according to the type of available data and the features of the market under study. The fra­mework of a structural static equilibrium model is provided when dealing with micro-level data relative to demand side and market-level data relative to the sup­ply side. An application to the Household Telephone Equipment Market in France is then proposed.

This paper investigates the dynamical properties of optimal paths in an exam­ple of one-sector overlapping generations model in which the utility function of the representative agent is non-separable. We show that the optimal growth path may be oscillating and optimal two-period cycles may exist. Applying these results to the model with altruism, we show that the condition of operative bequest is fully compatible with endogenous fluctuations.

In this article, we analyze the dynamics of the debt-income ratio when the inte­rest rate is an increasing function of this ratio. Within such a framework, we show that the equilibrium debt-income ratios are multiple and that their dynamics can exhbit hysteresis effect. An empirical illustration permits to show how these theo­retical results can be used to show that the 60 % debt-income ratio imposed by the Maastricht Treaty may be inadequate.

We examine several estimation methods of one of the most useful instruments in interest rate risk management : the term structure of interest rates. We present mainly simulation-based methods allowing for parametric estimation of continuous time models.

The purpose of this paper is to study the quantitative implications of the keyne-sian and R.B.C. models. The structural parameters of the R.B.C. model are obtai­ned through estimates of the observed and simulated keynesian model. This application of indirect inference indicates that the data are not sufficently informa­tive to distinguish between these two representations of the aggregate economy. Nevertheless, this result comes from the introduction of an "error term" into the R.B.C. model.

The purpose of this paper is to test the dependence structure of financial series of stock returns and foreign exchange rates. To do this, R/S analysis and various ARFIMA methods have been implemented in order to detect the presence of long-term memory. The main result concerns Canadian markets for which we find persistence and anti-persistence phenomena in the foreign exchange market (dollar-Canadian dollar) and in the stock market (TSE 300) respectively.

Holistic methodology develops an analysis of scientific dynamics which is more complex than those produced by traditional methodologies. In holistic methodology the advance of economic knowledge is the result of a three-dimen­sional arbitration by the profession between scientific characteristics, social rele­vance and production costs. The consequence of this permanent arbitration has been - and still is-a renunciation to explore certain scientific paths, which in fact appears paradoxical as far as progress in knowledge is concerned. In macroeco­nomics, a striking example of such a renunciation is the general omission of the opportunity costs of production in the standard modélisations of short-term choi­ces of the firms. This conventional treatment dominates economic literature and has major consequences on the representations of the working of the economy. Explanation of this choice can be found in the satisfaction of social demands to the economic profession.

This article deals with an important question of epistemological analysis of social sciences. Based on the analysis developped by K.R. Popper, it studies the difficulties that self-fulfilling theories and predictions - and more generally effects that predictions or theories can have on anticipated events - represents for economics.

Our paper presents some results of an experimental study comparing the des­criptive power of the Expected Utility (EU) model and of a Rank Dependent Utility (RDU) Model allowing the decision maker to subjectively transform objective pro­babilities. The descriptive power of the EU model and of a RDU model is evaluated by mean of two indexes constructed from their respective preference functionals. The main conclusion of this experimental study is a slight descriptive superiority of the RDU models over the EU model.

We show that the context of choice, defined as the choice set, influences the preferences expressed by experimental subjects. Making available an additional option in the choice set, can induce reversals in preference judgements. We observe such changes for preferences among lotteries, which contradict the axiom of irrelevant alternatives. But the asymétrie dominance effect, observed in similar studies is also rejected.

Recently, evolutionary games has analyzed the way evolutionary dynamics is modified when the system is subjected to random perturbations. This approach has led to a new concept of stability, the one of stochastic stability proposed by Foster and Young. For 2x2 symmetric games with two symmetric strict Nash equilibria, there is only one stochastically stable equilibrium, the one which satis­fies Harsanyi and Selten's criterion of risk dominance. This paper reviews the recent literature on this domain and highlights some limits of the stochastic stabi­lity. In particular, if the process is non-ergodic, the later criterion cannot be applied. We finish by studying some non-ergodic stochastic evolutionary game dynamics.

In this paper which deals with global environmental problems, we use an evo­lutionary approach in order to determine the conditions under which O.E.C.D countries are led to reduce unilaterally their pollution emissions. Thus we remark that a country's choice of a unilateral reduction depends on the information at its disposal concerning the other nations' characteristics. Where such information is complete, the commitment emerges as the result of a strategical calculation. The conclusion seems less obvious in a case of asymmetrical information where eve­rything leads us to believe that the commitment would be bound to fail.

We study a merged representative sample of French firms and workers. The difference between the workers' internal and external wage rates depends upon the compensation policy of the firm and the expectation of the compensation policies of all other firms. We estimate this difference at the level of the firm and model its determinants. The results are as follows : (1) 2/3 of the firm size effect in wage differences is explained by the tendency of high paying firms to employ workers with high external wage rates ; (2) 1/2 of the sectoral wage differences disappear ; (3) even a biased measure of the internal/external wage difference, ignoring measurable individual characteristics, can explain a large part of the observed wage differentials.

In this study, a rich data set is used to show that there is a size effect on the level of wages as well as on the returns to seniority. Our results indicate that for blue collar workers, the larger the establishment size is, the higher are the returns to seniority in the beginning of workers' careers and the more rapid is the decrease of these returns to seniority. The opposite is observed for the other cate­gories of employees. We then discuss possible links between four alternative explanations of the seniority-wage profile and the observed size effect. We finally present evidence in support of two of these explanations.

Flows of workers into and out of plants are each year quite huge and can only be slightly explained by job creations and destructions. The excess of wor­kers turnover looks like a pure replacement process, which allows plants to select the most adequate labor force, which will be promoted later on. This Rota­tion/Selection/Promotion process intensifies during upturns. These results can be justified by explicitely modelling hiring as an integration process with no need of the traditional hypothesis of adjustment costs.

We propose a model that explains differences in wages and unemployment rates between two regions. First, we show that when the two regions are in autarky, differences in firms, productive capacity lead to inter-regional differences in wages and unemployment rates. Second, whith free labor mobility, the migra­tion equilibrium is characterized by the equality of expected utilities and there is no reason for wages and unemployment rates to be identical between the two regions.

The generally accepted explanation of persistent unemployment in our econo­mies is based on the theory of suboptimal equilibria. These equilibria are the out­come of general equilibrium theory where some market imperfections are introduced (real rigidities, asymmetry of information, etc.). In this paper, another approach is adopted where persistent unemployment is interpreted as an involun­tary unemployment equilibrium, in the spirit of Keynes. This cannot be done but by rejecting Walras's law. When budgetary constraints of wage-earners differ from those of entrepreneurs, it is possible to have a partial Walras's law. In such a model, there exists a continuum of Keynesian equilibria. Effective demand theory allows one to select an equilibrium in the continuum. These equilibria are stable if the nominal wage is rigid.

In this paper, we develop two simulation models -an optimal growth model and an overlapping generations model- to investigate the long-run effect of ageing on per capital consumption, national saving and intergenerational welfare inequali­ties. Using trench demographic data, it is first shown that the optimal intertemporal path asks for an increase in the saving rate up to 2050. Nevertheless, the opposite result should be observed under an unchanged fiscal context. Two alternative saving-promoting policies are thus examined : a decrease in retirement benefits and the accumulation of a public fund.

We study a unique data set giving the French annual income distribution from 1977 to 1994. We show that households from recent generations tend to have lower income and standard of living at the begining of the life cycle than older generations. Concerning infra-generational inequalities, we show that the variance of French income distribution tends to be contracyclical.

Five social groups including the poor are defined by a synthetic indicator mixing three dimensions of poverty. Their consumption and income elasticities are computed on a pseudo-panel of the INSEE Family Budget Surveys of 1979, 1984 and 1989.

The CREDOC Research Institute surveyed 754 poor people in 1994, appro-ching them directly in areas concentrating a low-income population. This méthode proved successful to assess the efficiency of public policies designed to fight extreme poverty. It was also an opportunity to examine poverty on a broad perspective. Whereas a method based on poverty levels would have restricted the scope of people interviewed, the selection made by the CREDOC allowed surveyors to cover a whole range of financially or socially unstable people who did not fall below the poverty line.

Income poverty lines may fail to distinguish households experiencing depriva tion/exclusion Analysing large sample of Irish households we find that current income is an important influence on deprivation levels but so are many other aspects of current situation and how it arrived there It is also impor tant to distinguish different dimensions in using deprivation indicators with income to measure poverty

French banks have being facing significant changes in their financial environment : fall in credit demand, shifts in the structure of their liabilities. We describe those changes and analyse, using several theoretical models, their possible effects on the behavior of banks (risk exposure, degree of competition...).

We try and test the existence of a credit channel in France since the begining of financial deregulation. We analyse the influence of monetary policy on the spread between bank lending interest rates and money market rates ; moreover we study the predictive power of these spreads on the evolution of GDP. We show that banks smooth rather than amplify the variations of money market rates. Fur­thermore, the spreads between bank lending rates and market rates contain infor­mation about future activity. Finally, our tests do not provide a formal proof of the influence of monetary policy on activity by the credit channel.

The paper presents estimates of X-Efficiencies and Scale-Efficiencies in the French banking industry for the 1988-1992 period. By using the « distribution free » method of efficiency estimation, we show that average X-efficiencies are in the range of 70 % to 80 %. These results show also the existence of scale econo­mies in French banking industry. These results are also in conformity with the hypothesis that some excess capacity could exist in French banking industry. Finally, a positive relationship between cost-inefficiency and risk-taking is verified.

Long-term credit relationships and the structure of local banking markets This paper provides a spatial competition model of the banking sector, in which banks and firms can endure long-term relationships. We show that the fraction of borrowers with long-term contracts depends negatively on the number of banks. Considering sunk costs, we then show that the existence of long-term contracts leads to a decrease in the equilibrium number of banks.

This paper presents a model of endogenous growth with an imperfectly com­petitive financial intermediation sector. A concentrated banking sector implies a higher level of intermediation margin, which exerts a depressive influence on savings and investment. This prediction alone would suggest that the high growth performance of Germany and Japan are not related to the particular features of their financial sectors. However, if large financial institutions exert a positive influence on the productivity of industrial investment, thanks to their expertise and ability to advise entrepreneurs, the disadvantages of imperfect competition in terms of growth may be more than offset by the increase in productivity. The paper shows that this can be the case in a simple model of endogenous growth.

We analyse the effects of trade liberalisation in a Cournot model under three regimes : autarcy, asymmetric and symmetric trade. Welfare comparisons lead to some known results of reciprocal dumping. We determine the cut-off point of transport costs which leads to Pareto efficiency of one or the other of the regimes under consideration. Free entry allows to obtain positive welfare effects of trade liberalisation. We show by means of simulations that the same is true if asymme­tric variable costs are introduced into the model.

This article analyses the process of geographical adjustment in the sectors of textiles and garments, in the perspective of the removal of the Multi-Fibre Agree­ment. It uses a gravity-type model based on a sample accounting for 85 per cent of total world exports in those sectors. Particular attention is paid to potential pro­ductive re-location among OECD countries and a group of peripheral Asian coun­tries. Implications in terms of reduction of bilateral exchange costs in the location model are discussed. The introduction of such costs will limit the role of compara­tive advantages in the deepening of domestic specialisation.

Using a general three-country model which encompasses some existing models in the literature on regional agreements, dealing with monetary, fiscal or commercial policy, this paper determines the conditions under which regional agreements appear as solutions. The solution concept which is used corresponds to the core in cooperative game theory. It is underlined that the free-riding argument leads to results which depend on the negotiation context, while in the absence of a free rider problem the results are very sensitive to the existence of transfers or of differences between countries.

In contrast to the traditional analysis, this paper analyses the costs implied by the exclusion rather than by the inclusion of a country to the European Monetary Union (EMU). In a three-country model, we show that the monetary policy of the excluded country will be more expansionist and more counter-cyclical than the monetary policy of the EMU countries. This resembles a strategy of competitive depreciation. The country whose shocks are most symmetric to those of the exclu­ded country suffers most from this exclusion, through an increase in the variance of its production. We also examine the proposals to deal with the relations between EMU and non EMU countries of the EU.

A simple analytical model is presented where the rebellion activity of the politi­cal opponent is an increasing function of the prédation and repression activity of the government, and a decreasing function of its redistribution activity. These pre­dictions are tested on national African data, using a probit model, and are not rejected. Lastly it is shown that political risk, measured by the probability of politi­cal violence erupting, has a negative impact on economic growth.

This paper examines how different kinds of authoritarian regimes may affect long-run economic growth. On the theoretical side, non-democratic governments are differentiated according to the objectives they are likely to follow when deci­ding public spending and taxation policies. It is found that growth should be higher under « benevolent dictator » regimes, comparatively lower under « bureaucratic-authoritarian » regimes, and much slower under « predatory » regimes. A classifi­cation of non democratic countries into these three categories is attempted accor­ding to indicators of economic liberty and of the nature of their economic system. Conditional convergence tests carried out on the basis of this classification sup­port the predictions of the model.

This study concerns the impact of China's « opening and reform » policies on short-term urban industrial growth. The econometric analysis uses data on 434 Chinese cities throughout the 1988-91 period and tests two types of contributions : a direct one, measured by a simple relation between industrial growth and expla­natory variables and an indirect one, measured by spillover effects at a provincial level. The results highlight the direct role of foreign investment and the indirect role of exports. In terms of balance in regional development, it means that while foreign investment favours a geographically concentrated development, exports tend to reduce this unequal development through diffusion.

This paper shows how the modalities of exchange rate policy (exchange arrangements as well as rate variations) contribute to explain the strong real depreciation of the Chinese currency. It relies on an explicative model of the real exchange rate which incorporates, besides devaluation, monetary policy and price controls. It then tests the response of manufactured exports to the change of real exchange rate.