H—lB

has met with any considerable degree of success to ascertain in the Dominion the cost of living at a given period. If similar investigations had been made in previous years your Commissioners' work would have been greatly facilitated. (3.) The Statistics of New Zealand, particularly those relating to changes in rates of wages, prices of commodities, and rents in the four cities. As already mentioned, these statistics require extension before they can be of any material assistance in an inquiry like ours. (4.) Special tables and other statistical matter kindly prepared for the Commission by Mr. M. Fraser, the Government Statistician. (5.) Evidence tendered to the Commission, which may be classified as — (a.) Material derived from developing and recombining the subject-matter of the official and private statistics enumerated above, supplemented by original research, as, for instance, the evidence of Dr. Mcllraith (see page 264), Professor Segar (see page 303), Mr. F. P. Wilson (see page 407), Mr. F. L. G. Jolly (see page 196). (b.) Authenticated price-lists and accounts described, explained, and handed in or read to the Commission by witnesses, (c.) Budgets of household expenditure for certain dates. (d.) Verbal statements, often from memory only, and of the vaguest description, (c.) Statements of medical practitioners (see Addendum, page 505). Of these the most valuable for the purpose of comparing prices were Dr. Mcllraith's tables and index numbers, portions of the official statistics (particularly the Labour Department's Inquiry), and the certified price-lists and household accounts. It is to be regretted that the work of the Commission has been made exceedingly difficult, and' less exact than should be desired, by the very serious defects in the statistics of the Dominion, a subject to which we return later in this report. 6. In dealing with the material accumulated regarding prices, the Commission was alive to most of the leading principles to be observed in combining and interpreting them, viz. : — (1.) That wherever possible the greatest care should be taken to select an accurate estimate of the price in question, but that since the inquiry is in the nature of a comparison of changes in price, or into price-ratios, approximate estimates might be admitted, provided the methods by which they were obtained were the same over the whole period. Thus, if the price of tea in a particular table of prices were found to be calculated from some particular kind of tea not generally used, and high in price, that might be accepted as the price of tea generally, provided, first, that the price had been taken from the same market on the same method throughout, and, secondly, that the price of the tea in question moved in sympathy with the changes in the cost of the general kinds of tea. (2.) That the prices compased should be the same prices— i.e., that allowance should be made for changes in discounts, and in other details in the methods of charging, such as ex ship, ex store, delivered over the counter or at the house ; and that retail prices should be carefully distinguished from wholesale prices. (3.) That the things priced should be the same not only in kind but also in quality. This is a matter impossible to allow for in the price statistics themselves, as they cannot be manipulated to allow for improvement in quality. But it is important to note that most of the articles with which the inquiry has been concerned have improved in quality during the last twenty years—some of them very considerably ; that is to say, not only has there been an improvement in quality in each grade of the commodity, but in many cases new and higher grades of the same article have appeared. If ordinary white bread, e.g., improves in quality and the money price remains the same there is a real fall in price, but there is no plain and convenient method of representing statistically such a fall. All that can be done is to call attention to the fact that the quality has changed. (4.) That the prices chosen should be thoroughly representative of the prices usually paid for the commodities specified. Actual prices paid were often quoted by witnesses, but it was difficult to discover whether they were representative prices,

XII