This paper presents the results of an empirical study on the relationship between internal structural change and performance, in a sample of 138 large U.K. companies, 1959-1976. The paper considers the introduction of the multidivisional form (M-form) within the sample and addresses itself to three principal questions:

Firstly, whether re-organization to an M-form structure followed a significant decline in the performance of the firm?

Secondly, whether M-form adoption was followed by a significant improvement in performance?

Finally, was the performance impact, if any, of re-organization related in a systematic way to firm characteristics — e.g. size, diversification etc. — in a manner consistent with the M-form hypothesis?

During the period from 1949 to 1974 exchange rates were almost fixed in the OECD area. Since then some major rates have been moving, but even now most rates remain fixed relatively to each other and it is possible that we are returning to more fixed rates. Therefore it is of some relevance to try to summarize the experience of the few exchange rate changes that took place during the fixed rate period. Section II goes through the currency history of 20 main OECD-countries and identifies the 12 devaluations and 3 revaluations which occured in that period.

We have for each of these 15 cases tried to make a comparable story following the time-path of the same 16 main economic indicators - listed in Table 1 - in the last 3 years before the exchange rate change and in the first 3 years after the change. The changes have different sizes (and signs), happened in 9 different countries and are parts of rather different economic policies, but we shall show below that there are, nevertheless, a few main features that appear to generalize.

Some 20 years ago, J. Drèze (1962) opened a new prospect for the statistical treatment of the « simultaneous equation model » in econometrics (SEM). Indeed, he advocated the use of Bayesian methods for the analysis of such models. In particular, he pointed out the flexibility of this approach as it allows to take account not only of the a priori information provided by economic theory or observation, but more fundamentally, of the imprecise nature of this kind of information.

Ever since, it has been clear that the Bayesian approach of the SEM would be rather difficult to implement. The main problems one can identify are

1. the representation of a priori information by a prior density function which should be easily interpretable and at the same time flexible enough;

2. the numerical treatment (i.e. integration) of the posterior density so as to obtain moments and marginal densities of parameters, whether they be parameters of the structural form, of the reduced form, or functions thereof.

Le développement de la théorie de la dualité a donné naissance à de nouvelles formes fonctionnelles de système de demande. Simultanément, les hausses importantes des prix du pétrole et du gaz ont suscité un regain d’intérêt pour les applications empiriques de ces systèmes aux facteurs énergétiques. Le modèle translog de Christensen, Jorgenson et Lau (1973) est de loin la spécification la plus couramment utilisée. Pourtant ce modèle ne constitue qu’un cas particulier de la fonction quadratique généralisée d’xordre β introduite par Denny (1974). D’autre part, la présence de facteurs tels les coûts d’ajustement, les attentes adaptatives et l’inertie à court terme des producteurs aux variations des prix relatifs, suggère qu’un modèle dynamique devrait fournir a priori, un meilleur outil d’analyse que l’habituelle formulation statique de la théorie classique de la demande.

1. INTRODUCTION

This paper introduces the idea of a consumer who has a time varying Lancasterian matrix of characteristics imbedded in an intertemporal Stone-Geary utility function. The thrust of the paper is theoretical, its main aim being the generalisation of the theoretical contributions of Dixon and Lluch (1977). The main contribution in Dixon and Lluch (1977) is that DELES, a system of equations is proposed which describe the consumer’s planned and actual expenditure paths for durables. In order to derive these equations they have a consumer maximising an inter-temporal Klein-Rubin utility function defined on the stocks of durables, subject to a budget integral constraint and a vector of durable goods stocks’ adjustment equations.

This paper differs from Dixon and Lluch's contribution in that we assume that utility is not derived from the stocks of durables directly. Utility is derived from the characteristics that the durables yield. For example a car yields satisfaction to a consumer indirectly through characteristics like speed, comfort, safety, reliability, economy in petrol consumption, prestige and similar characteristics associated with the car. The idea of characteristics is best known from the work of Lancaster (1966, 1971), who assumes a constant objectively known characteristics matrix. This paper differs further from Lancaster (1966, 1971) in that we assume that characteristics of a durable change over time. Take the example of a car again. Over time, though, characteristics like capacity, associated with a particular car may be time invariant, most other characteristics like car reliability, petrol consumption, comfort, speed, prestige may change with the age of the car.

Recent research on the behaviour of the firm under uncertainty is so formulated that the issue of choice of technique is excluded from serious consideration. The reasons are clear. With random demand or factor prices and no input variable ex-post it follows trivially that if the firm is risk neutral its choice of technique must be the same as under mean preserving certainty. On the other hand, if, as is usual, some but not all inputs are variable ex-post the standard assumption of a putty-putty technology implies that input levels are chosen sequentially and only when the uncertainty is resolved are factor proportions determined. The question of choice of technique under uncertainty is consequently assumed away; the analysis determines only the effect of uncertainty on the quantity of the fixed factor installed. A recent analysis in this vein by Perrakis (1980) concludes that if output is not fixed ex-ante, uncertainty over the price of variable inputs will normally raise the capital stock if the firm is risk neutral although risk aversion will offset this tendency; Nickell (1977 or 1978 ch. 5) obtains essentially similar results.