H—4B

for potatoes in North Island towns, averaging the price of new and main-crop potatoes m the ratio of: new, 25 per cent., old, 75 per cent. In the South Island, main-crop-potato prices only will be used in the November index. In December, new- and oldpotato prices will be averaged, using the following ratios—North Island : new, 75 per cent., old, 25 per cent.; South Island : new, 50 per cent., old, 50 per cent. In January and February prices for new potatoes only will be used in both Islands. Some variation in this procedure may be necessary in an abnormal year—e.g., a very late season or a poor main crop. In such circumstances, the Government Statistician should have discretion to vary the formula, but should publish a statement of any adjustments made at the time of publication of the indices. Potato prices will be collected at weekly intervals in the transitional months (October-December). Other Foods 37. This sub-group covers a wide range of items—namely, dairy products, cereals, sugar, honey, jam, sweets, aerated waters, condiments, &c. (for detailed list, see Appendix B). Prices are collected for 31 items in all twenty-one towns covered by the index. B. HOUSING 38. It has been the practice in New Zealand hitherto (and it still is the practice in many other countries) to regard, for price index purposes, movements in rents of unfurnished houses as sufficiently typical of changes in housing costs. The Committee, however, considered that the scope of the index should be widened to include all housingcosts within the proposed prices index. The following facts influenced us in making this recommendation : (а) The 1945 census revealed that only 35-38 per cent, of the private European dwellings in the Dominion were rented, while 7*73 per cent, were rentfree (mostly with job), 31-36 per cent, owner-occupied subject to mortgage, and 25-38 per cent, owner-occupied without mortgage. A small percentage were not specified. In the urban areas (to which the proposed index will have application) the proportions of rented and owner-occupied houses to total private dwellings are 43 per cent, rented, 57 per cent, owneroccupied. We considered, therefore, that costs of owner-occupied houses should be represented in the index. (б) Rented unfurnished houses comprise rather a special class of dwelling, being largely either (i) old houses in private ownership, or (ii) new houses erected by the State for renting. The rents charged for class (i) are, in general, subject to tenancies over long periods, while, as regards (ii), the State, as a matter of public policy, keeps these rents at a minimum. (c) For the reasons advanced in (6) above, the rentals of rented houses do not in any degree represent reletting values. We considered, therefore, that rentals of unfurnished fiats, as representing in some measure at least reletting values, should be included in the index. Acting on instructions from the Committee, the Government Statistician prepared a plan for the inclusion of rents of flats and costs of owner-occupied houses in the index as well as rented houses. The relative weights allotted to the different classes as ascertained from the estimate of housing-costs in the base period are — Weight—i.e., Proportion of Aggregate Housing Cost in Base Period. Kented houses and flats .. .. .. .. 40 Owner-occupied houses .. .. .. .. 60 100

18