1 Data for both 1947 and 1940 are for nonfarm tenant-occupied dwelling units only.

HOUSING CHARACTERISTICS IN 108 SELECTED AREAS: 1946 VETERANS' HOUSING SURVEYS AND THE 1940 CENSUS OF HOUSING

(From July 1946 to January 1947, sample surveys of dwelling units to ascertain the general housing situation and the characteristics and housing plans of World War II veterans were conducted in 108 areas by the Bureau of the Census and the Bureau of Labor Statistics at the request of the National Housing Agency (predecessor of the Housing and Home Finance Agency). Detailed reports have been issued for each survey area and are available on request.1 This report presents a summary of 1940 and 1946 data on families, tenure, rent, and plumbing equipment, 1946 data on vacancies among privately financed dwelling units, and 1946 data on income of employed veterans for the 108 survey areas. The housing data include both veteran and nonveteran households.)

Summary of results

Substantial increases in home ownership and in the number of dwelling units with private plumbing equipment occurred between April 1940 and the latter part of 1946 in practically all of the 108 selected housing market areas in which veterans' housing surveys were made.

Approximately six-tenths of the survey areas in 1946, as compared with one-tenth of the same areas in 1940, had more owner-occupied units than tenant-occupied units. Of the 108 localities, the New York City area had the lowest proportion of owner-occupied units in both 1940 and 1946-16.3 percent in April 1940 and about 18.6 percent at the time of the 1946 survey. The areas with the highest owneroccupancy rates in 1940 were Tacoma (Wash.) with 57.8 percent, the Spokane area (Washington) with 56.8 percent, and the Kalamazoo area (Michigan) with 55.5 percent. In 1946, among the areas with the highest owner-occupancy rates were the Sharon-Farrell area (Pennsylvania), Fresno (Calif.), the Hackensack-Teaneck area (New Jersey), Toledo (Ohio), and the Spokane area, Yakima area, and Tacoma (Wash.), where about 7 out of 10 of the occupied dwelling units were owner-occupied. The median owner-occupancy rate for the areas surveyed increased from 38 percent in 1940 to approximately 53 percent in 1946. (The median is the rate for the area which would stand in the middle of the series if the areas were arranged in order according to owner-occupancy rates; in other words, it is the value which divides the series into two equal groups, one having rates higher than the median and the other having rates lower than the median.) (See table 34.)

According to the 1946 survey results, there was a noticeable increase in the proportion of dwelling units having standard plumbing equipment. Approximately four-tenths of the areas surveyed in 1946, as compared with only one-tenth of the same areas in 1940, had 90 percent or more of their units equipped with private bath, private flush

1In conjunction with the surveys included in this report, the Bureau of the Census conducted a national veterans' housing survey during the latter part of June 1946. Surveys in 9 northern cities were supplemented by the Bureau of Labor Statistics to provide separate figures for Negro veterans. Also, 19 other sample surveys of veterans' housing, in which a slightly different schedule was used, were conducted during March, April, and May, 1946. Census Population Report HVet-No. 114 contains a complete list of the veterans' housing surveys and a summarization of some of the items on veterans' characteristics, present living arrangements, and housing intentions.

In April 1947 the Bureau of the Census conducted sample population surveys in which housing data were collected for metropolitan districts. Twenty-eight of these surveys covered localities which are also contained in this report; however, the April 1947 surveys covered metropolitan districts, while the 1946 veterans' housing surveys generally covered the central city and nearby communities or, in some instances, were limited to the central city.

NOTE.-Data in tables 32, 33, 34, 35, 36, and 37 are for areas not comparable to the areas in tables 29, 30 and 31.

toilet, and running water. The median proportion of occupied units so equipped increased from 76 percent in 1940 to approximately 87 percent in 1946. (See table 33.)

The survey results indicated that very few vacant houses or apartments were being offered for rent or sale in any of the survey areas. In about four-fifths of the areas, the proportions of habitable nonseasonal vacancies for rent or sale were less than one-half of 1 percent. The St. Petersburg area (Florida) was the highest with 1.8 percent and Tucson (Ariz.), the Tampa area (Florida), and the Albuquerque area (New Mexico) were among the next highest with approximately 1 percent. The inclusion of habitable vacancies which were already rented or sold but not yet occupied and those held off the market for other reasons brought the habitable nonseasonal vacancy rates above 1 percent in only about one-fourth of the areas. In practically all of the areas, vacant dwelling units which were not habitable contributed little to the gross vacancy rates. With the exception of Tucson (Ariz.) and the St. Petersburg area (Florida) which had rates of 2.7 and 3.6 percent, respectively, gross nonseasonal vacancy rates did not exceed approximately 2 percent; in about one-half of the areas they were not higher than 1 percent. (See table 32.)

The results of the 1946 surveys showed that many married couples. had not established their own households but were sharing the living quarters of others either through necessity or choice. The proportions of such married couples (subfamilies) ranged from approximately 3 to 16 percent of the total primary families and subfamilies. In about one-half of the survey areas, at least 7 percent of the total primary and subfamilies were married couples who were sharing. Among the areas with the greatest proportions of subfamilies were Jackson (Miss.) with 16 percent and Greensboro (N. C.) with 14.8 percent. Among the areas with the least proportions of subfamilies were the Duluth area (Minnesota) with 2.8 percent and Pendleton (Oreg.) with 2.6 percent. Generally, more sharing occurred among nonwhite families than among white families. Also more sharing occurred among households containing a World War II veteran than among those not having a veteran. It should be pointed out, however, that veteran families consist largely of young married couples which constitute the group of persons most likely to share living quarters. (See tables 35 and 36.)

Another significant factor to be considered when examining housing requirements for veterans is their weekly income. Exclusive of southern cities, where separate figures were reported for white and nonwhite veterans, the Fall River-New Bedford area (Massachusetts) and the Providence area (Rhode Island) were among the areas reporting the lowest median weekly incomes for employed veterans-$37 and $39, respectively. Among the highest were Reno and Sparks, Nev., with $64 weekly and the Las Vegas 'area (Nevada) with a $66 weekly income. The surveys in the southern cities showed that, on the whole, nonwhite veterans were earning considerably less than the white veterans. The range of incomes for employed veterans in the southern areas was rather narrow; all the median incomes were between $29 and $35 for nonwhite employed veterans and between $44 and $59 for white employed veterans. (See table 37).

Results by individual survey areas are shown in the tables which follow. In several of the surveys, separate reports were issued for the

central city or other subarea so that the total number of places listed in the tables slightly exceeds 108. A complete list and description of the survey areas is given by States under the "Description of survey area" section.

Description of tables

Table 32 presents 1946 survey data on the estimated number of dwelling units in each area and percentage distributions of nonseasonal vacancies by availability for rent or sale. Table 33 shows plumbing equipment in occupied dwelling units in 1946 and 1940, while table 34 shows the proportions of owner-occupied and tenant-occupied dwelling units and the monthly contract rent of tenant-occupied dwelling units in both years.

Tables 35 and 36 show percentage distributions of primary families and subfamilies for veteran and nonveteran households from the 1946 surveys. The tables show the proportion of primary families which have no subfamilies living with them, the proportion which have one or more subfamilies living with them, and the proportion of subfamilies themselves. In addition, table 35 presents 1940 data on subfamilies for those areas for which such data are available.

In table 37 are shown the median weekly income and the percentage distribution of weekly incomes of employed veterans.

Separate figures for white and nonwhite households and veterans are presented for selected southern cities.

Dwelling units not reporting an item were distributed in the same proportion as the reporting units.

The 1940 data in tables 33, 34, and 35 include public housing while the 1946 figures represent only privately financed dwelling units. However, the number of public-housing units in any locality in 1940 was too small to have any significant effect on the comparability of the figures.

Excluded from the survey data in all the tables for both years. were the following: Hotels for transient guests, and similar places maintained by the YMCA and kindred organizations; missions and cheap one-night lodging houses; dormitories for students, and nurses' homes; educational, religious, military, and penal institutions; soldiers' homes; homes for orphans, the aged, blind, deaf, infirm, and incurables; military and labor camps in which workers live in a common barracks; and living quarters on military reservations. Definitions and explanations

Dwelling unit.-In 1940 a dwelling unit was defined as the living quarters occupied or intended for occupancy by one household. A household consisted of a family or any other group of persons living together, with common housekeeping arrangements, in the same living quarters.

While the same general concept of dwelling unit was retained in the 1946 surveys, the definition specified that, in general, a dwelling unit is a group of rooms or a single room occupied or intended for occupancy as separate living quarters by a person living alone or by a family or other group of persons living together. A group of rooms occupied or intended for occupancy as separate living quarters is a dwelling unit if (1) it has separate cooking facilities which must be more substantial than a two-burner hot plate or electrical appliance, or (2) it has a separate entrance. A group of rooms is considered as

having a separate entrance if no one has to pass through them to get to his own living quarters, and if the occupants of the group of rooms need not pass through any other living quarters to reach their rooms. A single room occupied or intended for occupancy as separate living quarters is a dwelling unit if (1) it has separate cooking facilities, or (2) it has a separate entrance and a private bath, or (3) it has a separate entrance and is rented unfurnished.

While there is no basic difference between the definitions used in 1940 and 1946, the definition of 1946 is more specific. Therefore, it is possible that living quarters may have been classified as separate dwelling units on one date and not on the other.

Habitable vacancies.-Included as habitable are vacancies in good condition or in need of minor repairs, plus only those in need of major repairs which are located in neighborhoods where units in similar condition are occupied.

Dwelling units in need of major repairs include units which require major repairs or replacement of floors, roof, plaster, walls, foundation, or other major structural replacement. A repair is major when its neglect seriously impairs the soundness of the structure and creates a hazard to its safety as a place of residence or if the repair has been neglected so long that the structure is already unsound.

The 1946 figure for habitable vacancies offered for rent or sale" represents habitable nonseasonal vacancies which were being offered for rent or sale at the time of the enumeration. Vacant units "already rented or sold" represent rental units which had already been rented and sale units which had been purchased, but the new tenants or owners had not yet moved in at the time or enumeration; these units constitute part of the active housing supply even though they are not actually being offered for rent or sale. "Other" vacancies include units undergoing extensive repairs, units in litigation, and units held off the rental or sale market for other reasons.

Not habitable vacancies.-Included as not habitable are all vacancies unfit for use plus those in need of major repairs which are located in neighborhoods where units in similar condition are not occupied. Vacancies unfit for use are those which are unfit for human habitation, i. e., so hazardous to the safety of occupants that they should be destroyed.

Seasonal. A vacant unit is considered seasonal if it is intended for occupancy during only a portion of the year. Although seasonal dwellings are found primarily in resort areas, certain dwelling units in rural areas used only for a portion of the year to house migratory workers are included as seasonal. However, a dwelling unit located in the closely built-up areas of most cities is not considered seasonal even though it is occupied only part of the year. Seasonal vacancies are not included in the vacancy rates in table 1.

Owner- and tenant-occupied dwelling unit.-A dwelling unit is classified as owner-occupied if it is owned either wholly or in part by the head of the household or some related member of his family living in the dwelling unit. All other occupied units are classified as tenantoccupied whether or not cash rent is actually paid for the living quarters. "Rent-free" quarters and living accommodations given in payment for services rendered are classified as tenant-occupied units. A unit owned by a person classified as a "lodger" and living in the unit is considered a tenant-occupied unit.