SECTIONAL TRANSACTIONS.— F. 387 



owned and not owned in the Saorstat. Rate of stock turnover per annum 

 in various businesses in different areas and according to the size of business. 

 Analysis of returns for large size businesses. Duration of ' ownership ' of 

 retail shops. Predominant position of Dublin in retail trade. 



Mr. H. T. Weeks. — The trial census of distribution (ii.o). 



The Trial Census of Distribution was designed to show how valuable 

 a complete census on wider lines would be. Field investigators covered 

 six towns of equal size and by inspection recorded the trade description of 

 each selling outlet, the products sold, whether the shop was a branch of 

 a multiple or co-operative, whether a lock-up or house shop. In an 

 official census information on total turnover could be obtained, and 

 possibly turnover of commodity groups, but since such information could 

 not be obtained in this unofficial inquiry rateable value was added to serve 

 as an illustration of (but not a substitute for) turnover figures. 



The results of the census showed surprising variety between the six 

 towns in most trades. There were extreme differences in the numbers of 

 outlets for many important products, and evidence of a different growth 

 in the distributive system which could only be guessed at from other 

 statistics. 



The Trial Census was frankly propaganda to prove to businesses in 

 distribution that a census would be valuable. Specifically it would help 

 in fixing sales quotas, guiding local sales drives, and determining develop- 

 ment policy, and the four sponsoring firms contributed notes on how they 

 would use the material from the Trial Census. 



But the Trial Census has more than a purely commercial interest. It 

 suggests that if control is necessary in distribution, existing statistics are 

 insufficient and previous measurement by census is essential. It shows that 

 a Census of Distribution would be essential to distributors, to administrators 

 and to theorists on distribution. 



Mr. H. G. Selfridge, Jun. — Consumers' control and the expenses of 

 distribution ( 1 1 . 1 5 ) . 



It is a pity that an item of merchandise in a customer's home is called by 

 the same name as is the same item in the manufacturer's warehouse, because 

 that tends to hide the exceedingly important economic fact that the inter- 

 vening processes through which it has to pass are somewhat similar in 

 quality', and generally at least as expensive, as the processes that make the 

 finished item out of its constituent raw materials. 



Prof. Plant has described the methods by which we have obtained very 

 satisfactorily accurate figures as to what those processes of distribution 

 cost. Allowing for a reasonable and necessary net profit, they cost just 

 over 30 per cent, of what the customer pays for all she buys in department 

 stores and similar shops. 



An analysis of the processes shows how complicated is the function of 

 the distributor. The most important difference between what he and what 

 the manufacturer does lies in the fact that the distributor's customer is an 

 individual with very individualised wants, an individual, furthermore, who 

 is entitled to and does ignore those considerations that business training and 

 experience might tell her should decide how those wants are to be satisfied. 

 Whatever the distributor does, is fundamentally at the behest of his cus- 

 tomer, whose money, indeed, pays for the whole of his activity. That the 



