----------------- HES POSTING ----------------- 
 
I am finding this thread (endogeneity/exogeneity) rather confusing. Can  
I suggest some distinctions that might help?   
 
First, Alan Freeman consistently refers to _the_ market rather than _a_  
market. I don't know how deliberate that is. _A_ market is clearly an  
analytical abstraction. We know that goods are never completely  
homogeneous so there is no definite boundary to a market. I bought a  
newspaper on my way to work this morning. What market is that  
transaction part of? The market for newspapers in Britain? In the street  
I stopped to buy it in? Are daily newspapers in the same market as  
weekly newspapers? There is surely no right answer. It depends on the  
facts of the case but also on what questions I want to pose (see  
below). 'The' market is harder to define, if it means anything at all. Is it  
the same as 'the economy'? It seems to be, from some of Alan's  
comments. Would it be clearer if we called it 'the economy'?   
 
Second, we need to distinguish between models and the real objects  
(whatever they are) which they relate to. 'Endogenous' and 'exogenous'  
are characteristics of variables, not things, relative to a particular model.  
Alan seems to want to link them to properties of objects. A key  
difference between a model and the things (whatever they are) that the  
model relates to is that a model is constructed to do a particular job -  
to explain or predict some particular phenomena. Endogeniety and  
exogeneity are the results of a particular modelling strategy which  
depends on the questions posed and need not have any ontological  
significance.   
 
My own view is that the market (or the economy) is an analytical  
construct whose boundaries are wherever we choose to put them, so  
the question is one of modelling strategy with no ontological  
significance but that may not be everyone's view.   
 
Tony Brewer ([log in to unmask]) 
 
------------ FOOTER TO HES POSTING ------------ 
For information, send the message "info HES" to [log in to unmask]