----------------- HES POSTING ----------------- 
 
Can anyone tell me where, if at all, economists have attempted to define  
the distinction between exogenous and endogenous phenomena? In  
particular, what is endogenous to the market, and what is exogenous to  
the market?   
 
I am well aware that this is discussed extensively in econometrics, but I  
think that the distinction one finds here is purely technical and does not  
really address the ontological questions involved. For example, if wage  
demand is specified as a function of employment levels then it is  
endogenous, if not it is exogenous; but this does not tell me the  
analytical grounds for the choice of equations. Moreover, it is perfectly  
possible to 'endogenise' phenomena that are on analytical grounds  
exogenous, eg if the size of the ozone layer is a function of CFC  
consumption, and agricultural output and mortality are functions of the  
size of the ozone layer, we may write a system of equations in which  
the climate is technically an endogenous variable of the economy; but in  
actual fact, this system does not tell us that the ozone layer is a  
constituent part of the market, merely that the market interacts with it.   
 
This arises because I am writing a paper on long waves in which I want  
to address the issue: is there an endogenous recovery from the  
exhaustion of a long wave of expansion? There is a wide discussion on  
this question in the literature, cast in terms of 'internal' and 'external'  
factors but although it is hotly disputed, I cannot find anywhere where  
the participants explicitly define what they mean by 'internal' and  
'external'.   
 
It seems to me that the question 'what is endogenous to the market' is  
co-terminous with the ontological question 'what is a market'? Since  
this is in a sense the definition of the subject matter of economics, it  
strikes me as a significant deficiency that the distinction is used so  
much, and seems to be discussed so little.   
 
Alan Freeman 
 
------------ FOOTER TO HES POSTING ------------ 
For information, send the message "info HES" to [log in to unmask]