The Relationship Between Sustainability and Financial Performance Using Panel and Multi-Equation Models
The Relationship Between Sustainability and Financial Performance Using Panel and Multi-Equation Models
The project´s purpose was to continue the in-depth progress of its predecessor on "The Relationship Between Environmental Performance and Shareholder Value" and settle some of the remaining methodical issues.The prior study concluded on the effects of European firms´ environmental and social performance on average monthly stock earnings.One major assumption implicit to this investigation proposed that the variables of environmental and social sustainability influence financial performance. Given the case that this effect may (as well) be reciprocal, the use of the models devised encounters several difficulties.As regards the question of cause and effect concerning the relationship of shareholder value and financial performance the methodical approach apllied to date may be improved in two respects:
- On the one hand, time lagged descriptive variables may be included. Here the influence of environmental or social sustainability on financial performance is investigated from the perspective gained one or several years later. This onset lends itself to suggest a fruitful combination with panel models. Such analyses allow to model potential intertemporal influences on financial performance. If such intertemporal patterns are unduly left without consideration, this may lead to distorted estimations of the parameters (and therefore of the sustainability parameters of interest as well).
- On the other hand, the question of causation in the relationship between sustainability and financial performance can be analysed econometrically by using multi-equation systems. With these, the influence of sustainability on economic performance as well as the influence of economic performance on sustainability may be modelled simultaneously.
The projects aim is to revise the research deficits described above by developing und empirically applying both multi-equation systems as well as panel models.