Econometrics of investment and business surveys,
by Pierre Poret.
The usual model of the accelerator (which relates capital to production) does not provide for the measurement of the role played by Keynesian demand constraints, which can limit production just as easily as insufficient returns.
At most, this near-accounting equation we find in most of the major macro-econometric forecasting models can tell us something about the production function selected. If we wish to conserve them, however, we can seek to improve them by reconsidering the supposed characteristics of the production function. In this way we show how business surveys shed an original and useful light on the possible deformations of the production function and on depreciation behaviours.
Models based on more than one school (Keynesian, classical, etc.) in which market constraint is unduly represented by production are theoretically biassed towards the acceptation of the Keynesian hypothesis. We therefore propose another formulation of the Keynesian accelerator, based on balances of opinion from surveys and on predictable demand trends, and allowing quite a rich description of behaviours (taking into account past anticipation errors and demand uncertainties). Finally, we present estimations made with more general models based on two schools (Keynesian and classical) and compare these estimations with the results of surveys on the number of investors with financial constraints. Both approaches showd the same trend: the effect of profitability has increased on average since the mid-seventies.
This study also brings out the threefold value of business survey results: they bring additional information to shed more light on the meaning of statistics or quantitative results, they provide direct knowledge of anticipations and give "leading indicators" which can be integrated into econometric modelling.
Econometrics of investment and business surveys, by Pierre Poret. The usual model of the accelerator (which relates capital to production) does not provide for the measurement of the role played by Keynesian demand constraints, which can limit production just as easily as insufficient returns. At most, this near-accounting equation we find in most of the major macro-econometric forecasting models can tell us something about the production function selected. If we wish to conserve them, however, we can seek to improve them by reconsidering the supposed characteristics of the production function. In this way we show how business surveys shed an original and useful light on the possible deformations of the production function and on depreciation behaviours. Models based on more than one school (Keynesian, classical, etc.) in which market constraint is unduly represented by production are theoretically biassed towards the acceptation of the Keynesian hypothesis. We therefore propose another formulation of the Keynesian accelerator, based on balances of opinion from surveys and on predictable demand trends, and allowing quite a rich description of behaviours (taking into account past anticipation errors and demand uncertainties). Finally, we present estimations made with more general models based on two schools (Keynesian and classical) and compare these estimations with the results of surveys on the number of investors with financial constraints. Both approaches showd the same trend: the effect of profitability has increased on average since the mid-seventies. This study also brings out the threefold value of business survey results: they bring additional information to shed more light on the meaning of statistics or quantitative results, they provide direct knowledge of anticipations and give "leading indicators" which can be integrated into econometric modelling.
