a farm. The dwelling unit may be a single-family detached dwelling, or it may be one of two such units in a duplex or one-over-one type structure. Or it may be one of many apartments in a multifamily structure. Space considerations do not apply in the use of the definition, provided the accommodations permit housekeeping use. Stopgap accommodations such as trailers are not included, nor are single persons or transient accommodations such as dormitories, commercial hotels and tourist cabins. Table 3.-New privately financed dwelling units started in urban areas: By type of structure, by year, 1936-46; by month, January 1945-March 1947 [Annual data rounded to thousands-monthly data rounded to hundreds] 239,000 Nonfarm dwelling units include those in urban places and all nonfarm dwelling units in rural areas. Urban places include all incorporated places whose populations, as counted in the most recent decennial census, were 2,500 persons or more, and a small number of densely populated unincorporated places classified as "urban" by the Bureau of the Census under a special rule. In order to be classified as urban under this rule an unincorporated place must have a population of 10,000 or more, and a population density of 1,000 or more per square mile. Classifications are made at the time of each decennial census and the place remains so classified until the time of the next census, at which time it is reclassified in accordance with its attained population and density. Rural nonfarm dwelling units include all units located outside of urban places but not on farms. The term "start" is intended to mean the beginning of active building and the statistics attempt to allocate starts to the period in which this initial activity occurred. This point in time has generally been taken to be the time at which Table 5.-New privately financed nonfarm dwelling units started: By geographic division, by type of structure, by year, 1940-46 [In thousands of family units] 1 Geographic data for 1945 and 1946 have not been adjusted for lapsed building permits and for lag between issuance of permit and actual start of construction. Consequently, totals for these years are not consistent with those in tables having national break-downs only, as shown in table 1. Less than 300 units. Source: U. S. Department of Labor, Bureau of Labor Statistics. Table 6.-New privately financed nonfarm dwelling units started: By geographic division, by urban and rural nonfarm areas, by year, 1940-46 1 "Urban" plus "Rural nonfarm" data will not necessarily add to "Total nonfarm" due to rounding. Geographic data for 1945 and 1946 have not been adjusted for lapsed building permits and for lag between issuance of permit and actual start of construction. Consequently totals for these years are not consistent with those in tables having national break-downs only, as shown in table 1. Source: U. S. Department of Labor, Bureau of Labor Statistics. Table 7.-New privately financed dwelling units started in urban areas: By geographic division, by type of structure, by year, 1940-46 Source: U. S. Department of Labor, Bureau of Labor Statistics. 1 Geographic data for 1945 and 1946 have not been adjusted for lapsed building permits and for lag between issuance of permit and actual start of construction. Consequently, totals for these years are not consistent with those in tables having national break-downs only, as shown in table 1. 2 Less than 51 units. Composition of the Estimates of Nonfarm Dwellings Started As indicated in the introduction, estimates for the years 1900 through 1929 were made by the Twentieth Century Fund and the National Bureau of Economic Research. They represent estimates for each 10-year period, in turn broken into annual totals for each year. For the years 1930 through 1939 BLS similarly distributed an aggregate estimate for the decade. The data for subsequent years represent estimates which have been prepared annually by the Bureau of Labor Statistics. These estimates recapitulate the monthly estimates now regularly prepared by the Bureau. To make these monthly estimates the Bureau uses separate methods for estimating the urban and rural nonfarm components. These are described below. Urban Area Estimates Estimates for urban areas are based largely on building permits issued by local government officials as reported to BLS. The number of cities for which building permits have been reported has been expanding constantly. Monthly data are now available in the BLS for almost 2,500 urban localities covering more than 85 percent of the urban population of the country. The estimate for urban places is prepared on the assumption that cities in the same state and of like population size and proximity to metropolitan centers will experience approximately the same volume of residential construction. For each city classification an estimating factor is computed, generally the ratio of total cities to reporting cities in each classification. This factor is multiplied by the number of dwelling units reported to get the estimated total for all cities in the classification (thus, if 5 out of 6 cities of 5,000 to 10,000 population in a given State report, the estimated total for the classification is 1.2 times the reported totals). The estimates obtained for the various city classifications are added to give the total estimate for all urban places. Such estimating, or "blow-up," factors are used to fill in gaps caused by approximately 1,000 nonreporting urban localities, but Table 8.-New privately financed dwelling units started in rural nonfarm areas: By geographic division, by type of structure, by year, 1940-46 1 Geographic data for 1945 and 1946 have not been adjusted for lapsed building permits and for lag between issuance of permit and actual start of construction. Thus "All Region" totals for these years are not consistent with those in tables having national break-downs only, as shown in table 1. 2 Less than 51 units. Source: U. S. Department of Labor, Bureau of Labor Statistics. |