Working papers results

We present a model of electoral accountability to compare the public finance outcomes under a presidential-congressional and a parliamentary system. In a presidential-congressional system, contrary to a parliamentary system, there are no endogenous incentives for legislative cohesion, but this allows for a clearer separation of powers. These features lead to clear differences in the public finance performance of the two systems. A Parliamentary system has redistribution towards a majority, less underprovision of public goods, more waste and a higher burden of taxation, whereas a presidential-congressional system has redistribution towards a minority, more underprovision of public goods, but less waste and a smaller size of government.

Now in:
European Economic Review, 1999 (forthcoming)

This paper analyses how to extract market expectations from asset prices, with a particular example: using the term structure of interest rates to estimate the probability the market attaches to the event that a country, Italy, joins the European Monetary Union at a given date. The extraction of such a probability from the term structure is based on the presumption that the term structure contains valuable information regarding the markets assessment of a countrys chances to join the EMU. The case of Italy is interesting because in the survey regularly conducted by Reuters the probability of joining EMU in 1999 fluctuated between 0.07 and 0.15, while, during the same period, the measures of computed by financial houses -- also based on the term structure of interest rates -- have ranged between 0.5 and 0.8. The paper proposes a new method for computing these probabilities, and shows that the discrepancies between survey and market-based measures are not the result of market ine fficiencies, but depend on an incorrect use of the term structure to compute probabilities. The technique proposed in the paper can also be used to distinguish between convergence of probabilities and convergence of fundamentals, that is to find out whether an observed reduction in interest rate spreads signals a higher probability of joining EMU at a given time, or simply reflects improved fundamentals.

This paper analyses the dynamics of wives labour force participation in Spain during the late 1980s from a non-parametric descriptive perspective. This research is motivated by two basic facts: One, there is evidence that female labour supply behaviour in Spain is changing since the late 1980s. Two, while the analysis of participation stocks is covered in the literature, there is no published research on mobility or flows. In the first part of this paper there is a description of the three-monts transition rates over two-waves. The underlying assumption is the First Order Markov Hypothesis. In the second part, the Markov assumption is questioned. This is done by carrying out an analysis of survival over 7 waves in which re-entries are ignored. Moreover, there is also an analysis of mobility contingent on past labour market state, which includes re-entries in the analysis. This allows me to study the likelihood of relapsing in a particular state, or the likelihood of surviving contingent on past labour market states. The results are interesting because they reveal features of female labour market behaviour unknown to date.

We show how to extend the construction of infinite hierachies of beliefs (Mertens and Zamir (1985), Brandenburger and Dekel (1993)) from the case of probability measures to the case of conditional probability systems (CPSs) defined with respect to a fixed collection of relevant hypotheses. The set of hierarchies of CPSs satisfying common certainty of coherency conditional on every relevant hypothesis corresponds to a universal type space. This construction provides a unified framework to analyze the epistemic foundations of solution concepts for dynamic games. As an illustration, we derive some results about conditional common certainty of rationality and rationalizability in multistage games with observed actions.

This work focuses on the political determination of a public education policy within the context of a general equilibrium macroeconomic model. The primary objective of this paper is to study whether publicly funded education can emerge and be sustained as a political and economic equilibrium in an economy where individual agents are selfish, rational and forward-looking. I construct an overlapping generations general equilibrium model that endogenizes the large involvement of the public sector in human capital investment. The agents work for the first two periods of their lives and then retire during the third period. The first generation agents may also allocate resources to the acquisition of human capital, but they cannot borrow against their future income. In a political equilibrium where the rational and forward-looking agents of the two oldest generations vote for a level of public funding of education, public financing of education is motivated by the complementarity between capital and labor in the production function and appears as an instrument to compensate for the absence of credit markets. Thus, public funding of education does not have to be chosen because of altruism or externalities. In an economy calibrated using U.S. data, I can match the high shares of GNP allocated to education observed in the U.S. economy. The share of publicly funded education is an increasing function of GNP which mirrors the observed disparities across countries with different levels of development. Furthermore, an increase in the social security tax rate reduces the share of GNP allocated to publicly funded education which might help explain differences across countries with similar levels of development. The constructed economy also supports the existence of poverty traps for relatively low levels of income per capita. An interesting feature is that, if we do not allow the young agents to work, the economy will get out of these poverty traps and will converge to a steady-state with high levels of physical and human capital.

I include the variables wives age and cohort and children in a participation equation to explore how the following two economic issues affect participation. First, a structural change in terms of participation over the life-cycle. Because a structural change does not affect all women of the young cohorts, I distinguish between long-run participating women (i.e. those whose participation behaviour resembles that found after the structural change) and a priori inactive women (i.e. those with a traditional behaviour). Second, it explores the impact of current social policies on mothers participation. Despite that a negative correlation between children and mothers participation (especially pre-scholars) is considered a stylized fact in the literature, long-run participating women may not withdraw from the labour market after maternity to avoid the likely experience loss (i.e. real wage decline) due to long absences. This analysis is carryed out by exploiting a longitudinal Spanish survey (the ECPF). Despite some lacking variables, the use of panel data methods yields to satisfactory results.

Much of the recent growth and development literature is based on the notion that economies may exhibit multiple equilibria, due to coordination failures. Surprisingly, little attention has been given to analyze which economic institutions may solve such failures. We examine the role of banks as 'catalysts for industrialization. When there are limits to contracting, and complementarities exist among investments of different firms, we derive coordination costs endogenously and show that banks can acts as catalysts provided that: (i) they are sufficiently large to mobilize a critical mass of firms, and (ii) they possess sufficient market power to make profits from coordination. We also show that the costs of coordination depend critically on the contracting instruments available to banks. In particular, allowing banks to hold equity reduces and sometimes eliminates the cost of coordination. We use our results to interpret the patterns of early industrialization of Belgium, Germany, and Italy in the late 19th century. These countries experienced quick industrialization with the active involvement of large and powerful universal banks, which engaged in both debt and equity finance.

Now in:
Quo Vadis Europe, Ed. by H. Sibert, Kiel

At the core of the ongoing political and academic debate on European integration lies a fundamental question: what is the appropriate assignment of policy tasks to different levels of government? This paper asks what economic theory has to say about this normative problem. Our starting point is traditional economic theory, which approaches the question of policy assignment from the perspective of social welfare maximization by a Pigovian benevolent planner. Then, we discuss the political economics approach to this same question. Two themes run through the paper. The first theme is that, when allowing for political economy considerations, straightforward normative conclusions on the appropriate degree of centralization are much more difficult to draw. The second theme relates to the existence of complementarities between policy dimensions. Complementarities imply that, in the absence of clear constitutional safeguards, the process of European integration is unstable and fragile. We conclude with a discussion of how to combine flexibility and commitment in the process of European integration.

Now in:
Quarterly Journal of Economics, November 1997

Political constitutions are incomplete contracts and therefore leave scope for abuse of power. In democracies, elections are the primary mechanism for disciplining public officials, but they are not sufficient. Separation of powers between executive and legislative bodies also helps preventing the abuse of power, but only with appropriate checks and balances. Checks and balances work by creating a conflict of interests between the executive and the legislature, yet requiring both bodies to agree on public policy. In this way, the two bodies discipline each other at the voters advantage. Under appropriate checks and balances, separation of powers also helps the voters elicit information.

This paper measures the relation between job flows and stablishment size applying econometric techniques best suited for analysing the dynamics of large cross-section. Using a balanced panel from the Mexican Manufacturing sector, it shows that, in line with cross-country evidence, initially small firms create proportionally more jobs than large firms. Since these results suffer from regression toward the mean, the paper applies an alternative technique and it does not find long-run tendency of small establishment to converge toward the mean. Furthermore, it shows how cross-sectional dynamics varies across industries and how it is linked to gross and net flows in each sector. We observe convergence to the mean in relatively stable sectors and asymmetric dynamic behaviour between expanding and declining industries.