In response to Robin Neill's query about how economists explain  
decreasing and increasing returns when all inputs are variable, may I  
suggest Bela Gold's article, "Changing Perspectives on Size, Scale and  
returns: An Interpretive Survey," in the _Journal of Economic  
Literature_, XIX (March 1981), 5-33.  In an article that I wrote with  
Sidney Carroll on "Alfred Chandler's Speed: Monetary Transformation,"  
(_Business and Economic History_, Vol. 22, no. 1, Fall 1993) we  
summarized Gold's argument in these words:   
  
"Gold provides extensive documentation in support of the argument also  
advanced by Armen Alchian, that what are usually called economies of  
scale are in reality economies associated with different techniques of  
production and with different factor proportion.  Gold finds little  
evidence that larger duplicates of smaller plants that are otherwise  
identical in all other important respects do, in fact, produce at lower  
cost.  He describes the long struggle in economic thought to preserve  
the idea that size is cause, rather than consequence, of cost reducing  
changes in production, and he concludes that at least two concepts of  
scale economies have coexisted.  The 'restrictive' theoretical concept  
attributes scale economies to size and holds all else constant,  
technology as well as factor proportions.  The 'observational' concept  
attributes scale-related cost reductions to changes in (among other  
things) technology, factor proportions, and organization of production."  
  
  
It is my impression that the distinction is not made clear in most  
introductory, or for that matter, intermediate or advanced courses.    
  
  
Anne Mayhew