144 SECTIONAL ADDRESSES 



not support the thesis of a limit to efficient size. McNair, in a review of 

 the Harvard Bureau's inquiries up to 1928, presents figures ^^ showing 

 definitely a higher profit per sales for the larger department stores, lumber 

 dealers, jewellers, grocers and shcie retailers. Since 1928 similar results 

 have been repeated for department stores and have been newly established 

 for specialty stores.^ This greater profit of the larger firms was due to 

 their more economical purchase of merchandise rather than to their 

 lower operating expense. Where chain stores are concerned it is often 

 found that the larger the branch the lower is the operating cost, while the 

 larger the whole firm or chain the lower the merchandise cost. A clear 

 example is that of the chain groceries studied by the Harvard Bureau. 

 As the whole chain gets larger the cost of merchandise falls. In chains 

 of under nine branches the costs of purchasing the merchandise forms 

 81-9 per cent, of retail sales ; in chains of ten to fifty branches it is 80 -8 

 per cent. ; in chains of over fifty branches it is down to 79 -4 per cent. As 

 the branch shop — the link of the chain — gets larger the operating costs 

 fall. In firms with shops selling less than $50,000 a year on the average, ex- 

 penses formed 19 per cent, of retail sales ; for firms with shops selling on 

 the average more than this expenses were down to 16-7 per cent. It is 

 true that the purchasing cost of merchandise rises slightly for firms with 

 larger shops and that operating costs rise slightly with larger chains, but 

 this is not sufficient to off^set the economies just mentioned, so that on 

 balance the chains selling the greatest value of goods (whether by reason 

 of large shops or a large number of shops) have the lower total of costs 

 and higher profit per given amount of sales. 



The U.S. Federal Trade Commission have recently investigated chain 

 stores on a wider scale. Here the several trades covered show widely 

 different results. Some trades, such as groceries, confectionery, men's 

 furnishings, men and women's ready-to-wear, follow the Harvard grocery 

 chain-store pattern. As the number of branches increases, merchandise 

 cost falls considerably, operating costs rise slightly and, on balance, profits 

 rise. On the other hand, variety stores either with a limit of $10 or no 

 limit, department stores and dry goods, show a falling profit as the branches 

 increase in number. It is noticeable that the nine trades where (according 

 to the Trade Commission's own table ^*) profits per sales rise upward with 

 increasing size of the chain, are trades with, on the whole, a higher rate of 

 turnover or ' stock-turn ' than the seven trades where profits fall downward 

 with increasing size of chain. Only one of the nine upward profit trades 

 has a turnover rate lower than 3 • 40 ; only one of the seven downward 

 profit trades (department stores) has a turnover rate above 3 '42. The 

 rate of stock-turn is with certain qualifications a possible index, as we 

 shall see, of the degree of standardisation. 



It is possible that profitability depends not upon number of branches 

 but upon standardisation and the reduction in the number of lines sold. 

 Thus where the trade lends itself to a multiplication of lines it is rash for 

 a firm to branch out, since not enough sales will be effected in each line to 



*' Economic Journal, December 1930. 



^^ Harvard Bureau of Business Research Bulletins, 83 and 84. 



'* Chain Stores : Sales, Costs and Profits of Retail Chains, Table 34, p. ^&. 



