Un Essaie d'Application de la Théorie Quantitative de la Monnaie à l'Economie Portugaise, 1854-1998

Abstract

L'objectif de ce travail nous conduira à tester la pertinence d'accepter la théorie quantitative du revenu nominal dans l'économie portugaise. C'est-à-dire, on ira voir si un accroissement de l'offre de monnaie de x% entraîne, ou non, un accroissement de x% du revenu nominal. Les testes qu'on ira faire cherchent à démontrer cette théorie pour la longue période et aussi pour le processus d'ajustement qu'on appellera de courte période. Si on peut retenir cette hypothèse de comportement, on rencontre l'affirmation de Friedman, «There is an extraordinary empirical stability and regularity to such magnitudes as income velocity that cannot but impress anyone who works extensively with monetary data». Mais précisons, comme disait déjà Robertson, cette stabilité empirique ne nous oblige pas à accepter l'encaisse monétaire comme déterminé seulement par le revenu monétaire.

Neste trabalho testaremos a pertinência de aceitar a teoria quantitativa do rendimento nominal para a economia portuguesa. Isto é, tentaremos verificar se um acréscimo da oferta de moeda de x% conduz, ou não, a um acréscimo de x% do rendimento nominal. Os testes que faremos procurarão demonstrar a validade desta teoria no longo prazo e também para o processo de ajustamento que designaremos por curto prazo. Se se puder reter esta hipótese de comportamento, reencontraremos a afirmação de Friedman: «There is an extraordinary empirical stability and regularity to such magnitudes as income velocity that cannot but impress anyone who works extensively with monetary data». No entanto, como dizia já Robertson, esta estabilidade não nos obriga a aceitar o encaixe monetário como sendo determinado apenas pelo rendimento nominal.

Following Friedman we can say that from the macro-economic point of view we need a theory to explain a) the short-run division of a change in nominal income between prices and output; b) the short-run adjustment of nominal income to a change in money supply; and finally c) the correction mechanism that brings together that adjustment and the long-run equilibrium.

Our objective is to study the last two points in the context of the Portuguese economy in the period 1859-1998. We have studied several models involving money to explain the behaviour of nominal income and we conclude that we cannot exclude a money supply (M1) elasticity of one. This result is obtained for the long as well as for the short-run. We can take the quantity theory as a money income theory for our data. Robustness of the results and good forecasting ability are the main characteristics of our models.

We have applied several econometric models. Our first model is a single ecm equation with unrestricted and restricted coefficients. We also have found a co-integration vector by applying the Johansen method. With that vector we have constructed a VECM and obtained forecasts and impulse responses to shocks. We have done the same with a parsimonious model derived from the last one and estimated by FIML.