386 SECTIONAL TRANSACTIONS.— F. 



compilation of reports for circulation among the collaborators is performed 

 by members of a University staff at the Department of Business Adminis- 

 tration in the London School of Economics. That ensures complete 

 detachment from the collaborating firms and from any trade association. 

 Secondly, the coding of returns which are sent through the office of the 

 statistical department of the Bank of England ensures complete anonymity, 

 while limitations of the group to firms receiving specific invitation to col- 

 laborate secures selectivity, and ensures a usefully comparable sample. 

 The handling of the returns inevitably involves statistical problems due, 

 for instance, to a changing sample from year to year and to the difficulty 

 of relating income and expenditure figures to valuation of capital. Arbitrary 

 decisions must be taken for dealing with such matters as rental of owned 

 premises, but these problems may be overcome by methods which do not 

 invalidate the results. 



For the most effective use to be made of the results it is necessary that 

 there shall be a wise exercise of discretion by the computors. Averaged 

 figures must frequently be accompanied by measurements of dispersion if 

 erroneous impressions are to be avoided, and new statistical methods must 

 continually be devised to give appropriate emphasis to changes in trading 

 conditions. There must also be a proper understanding of the reports in 

 each of the collaborating firms if full use is to be made of the results. This 

 involves in many cases the appointment of staflF with special training. 



The justification for comparing operating -costs in this way lies in the 

 revelation of significant differences. To be significant they must be 

 differences in the cost of undertaking broadly similar functions, so that 

 homogeneity of sample is all-important. Because differences in costs must 

 be expected between similar firms performing similar functions in different 

 situations, no special significance must be attached to average figures ; the 

 collaborating firms must regard them as landmarks, not as signposts. They 

 raise questions rather than propose solutions ; ' goal ' figures for particular 

 cost items are consequently to be deprecated. Further, there is no fixed 

 constellation of differences in costs likely to endure over time. Trading 

 conditions are not static and changes in costs with changes in sales are all- 

 important. Cost studies must therefore concentrate upon the rates of 

 change in costs which are associated with given rates of change in sales. 

 Operating cost studies become marginal studies. There is a steadily growing 

 need for associated annual studies in other branches of business, particularly 

 in the field of distribution, on the pioneer lines of British department stores. 



Prof. B. F. Shields. — -The Irish census of distribution (10.30). 



An examination of reports of the Tribunal on Prices, 1927, and Com- 

 mission on the Registration of Shops, 1934, with respect to retailers' costs, 

 excessive prices, redundancy, registration and licensing of shops. 



Distribution of population in the Saorstat and its relation to retail sales. 

 Types of shops. Methods of fixing retail prices and their influence on gross 

 profits. Gross profit margins classified. Joint supply. Retailers' expense 

 analysed. Spread of prices in various districts. 



Irish Census of Distribution : objects, scope, procedure, general analysis 

 and limitations of the inquiry. Comparison with other government 

 censuses. Expenditure of retail £. Investigation of human element in 

 retail distribution : family, proprietors, employees, males, females. Wages 

 and salaries, as a whole, in grouped firms, and areas in relation to businesses 

 of varying size, and as a percentage of sales — the larger the turnover, the 

 higher the total wages as a percentage of sales ? Single and multiple shops, 



