tending to follow the lower fringes and the rates outside following the upper fringes of the national rates. The rate inside SMSA's rose from 0.9 percent in 1956 to 1.4 in 1963 and remained at the higher level through 1965. Outside SMSA's, the rate rose from 0.9 in 1956 to 1.5 in 1961, dipped slightly in 1964, and regained the 1.5 level in 1965.

Regionally, there was considerable variation in homeowner vacancy rates during the 10-year period. The rates in the Northeast remained lowest and showed the greatest stability, continuing practically unchanged during the entire period. In the North Central Region, the rate increased rapidly between 1956 and 1958, dropped in 1959 and, except for a slight increase in 1963, remained fairly stable through 1965. In the South, the rates were relatively high and fast-growing, with the 1965 rate approximately double that reported for 1956. The West also had a high percentage of homeowner vacancies, ranking first among the re

gions in 1956 and 1957. The rates increased over the 10-year period, but the rise was considerably slower than in the South. The rates remained at the same level between 1956 and 1961, moved upward in both 1962 and 1963, then remained unchanged through 1965.

CHARACTERISTICS OF VACANT RENTAL AND HOMEOWNER UNITS

It should be kept in mind that the data on characteristics describe only the changes that have occurred in the vacant inventory; they do not necessarily reflect the changes in the total inventory. Comparable data on the characteristics of the occupied segment of the inventory are not available. The characteristics of the total inventory of units are affected by the addition

through new construction, by the removal of units through demolitions, and by changes in characteristics of existing units. It is expected, therefore, that such changes might be reflected in the characteristics of the vacant inventory. To the extent, for example, that the number of units constructed in 1960 or later rises with each year after 1960, the distribution of vacant units by year built may show an increase in the level of such units. Similarly, any changes in size, type of structure, or rents and property values occurring in the occupied portion of the inventory probably have some impact on the composition of the vacant inventory.

Vacant rental units.--Although the rental vacancy rates indicate an increase in the supply of vacant rental units between 1956 and 1965, the statistics on characteristics reflect only modest changes in the kinds of units offered for rent.

of

Size of units.--There is some evidence an increase in the size of the vacant rental units. The median number of rooms was 3.2 in 1956, 3.3 in 1960, and 3.4 in 1965. Also, the percentage for 2-bedroom units increased between 1960 and 1965-from 29 to 35. Conversely, 1-bedroom units declined; the percentage was 45 in 1960 and 37 in 1965.

Number of housing units in structure.-There has been an increase in rental vacancies located in the larger apartment buildings (structures with 10 housing units or more) and a decrease in vacancies in 1- and 2-unit structures. In 1956,

units in larger "apartment" buildings accounted for only 12 percent of the vacant rental inventory, and 1- and 2-family units accounted for 58 percent. In 1960, about 19 percent of the rental vacancies were "apartments," and 53 percent were in 1- and 2-unit structures. By 1965, the proportion representing "apartments" had increased to 26 percent; 1- and 2-unit structures had dropped to 43 percent. Practically all the increase in vacant "apartment" units occurred within metropolitan areas, whereas the decrease in vacant 1- and 2family housing units occurred inside and outside SMSA's.

Plumbing facilities. --The quality of rental vacancies, as measured by plumbing facilities, has improved. Three-tenths of the vacant units offered for rent in 1960 lacked a private toilet or bath or hot running water. In comparison, two-tenths of the rental vacancies lacked these facilities in 1965.

Duration of vacancy. --The figures on the length of time a unit remains vacant provide some information on turnover among rental vacancies. Although the supply of rental vacancies increased between 1956 and 1965, turnover among the units remained constant. In 1956, about 34 percent of

the rental vacancies had been vacant less than 1 month. An additional 34 percent had been on the market 4 months or longer. In 1960, the percentage was 31 for short-term vacancies (of less than 1 month duration) and 37 for long-term vacancies (units vacant 4 months or longer). The levels registered in 1960 continued through 1965.

As indicated in tables 4 and 9, rental vacancies that remain vacant the longest are more likely to be located outside SMSA's. This group of vacancies is also directly correlated with 1and 2-family houses, structures built before 1940, and units that lack plumbing facilities.

On the other hand, rental units vacant less than 1 month are more frequently found inside SMSA's and are more likely to be located in recently built structures containing 5 housing units or more. The majority have all plumbing facilities. Rent asked.--The monthly rent asked for vacant rental units has been shifting upward. The percent of vacant rental units offered at $100 or more has doubled since 1960 and quadrupled since 1956. Only 5 percent of the rental vacancies were in the $100 or more rent class in 1956, compared with 9 percent in 1960 and 23 percent in 1965 (table 5).

As expected, the cross-tabulations on data for rent show considerable differences between units at the lower end of the rent scale and those at the upper levels (table 7). The group of vacancies offered for rents under $50 monthly contain a relatively large proportion of 1- and 2room units and consist predominantly of 1- and 2-family houses and units without private plumbing facilities. These units also remain longest in the vacant inventory. Units offered at $75 and over are generally larger, more likely to be 10cated in structures with 5 housing units or more, have all plumbing facilities, and remain vacant for short periods of time.

This pattern of variation by rent levels is fairly consistent for the period 1956 to 1965. However, the differences are sharper between 1960 and 1965, reflecting the increase in units built 1960 or later. For example, approximately 63 percent of the units renting at $100 and over in 1965 were built in 1960 or later. The corresponding figures were 56 in 1964 and 43 in 1963; comparable data are not available for earlier years (table 13).

Year structure built.--Information on "year built" was first collected in this survey in 1962.

rental inventory

consists

more of

Although the vacant largely of aging stock, proportionately the newer units have become available in the past two years (table 8). As of 1962, about 5 percent of the rental vacancies had been built during the period 1960 to 1962, 15 percent between 1950 and 1960, 13 percent between 1940 and 1950, and 67 percent in 1939 or earlier. By the end of 1965, survey results show that 19 percent of the existing vacant rental stock had been built in 1960 or later and 58 percent had been built in 1939 or earlier.

The increase in new rental vacancies occurred almost entirely within metropolitan areas. Inside SMSA's, about 24 percent of the 1965 supply had been built in 1960 or later, compared with 5 percent in 1962. Outside SMSA's, the proportion for newly built vacancies remained the same--6 percent in 1965 and 5 percent in 1962.

As shown in table 8, the increase between 1962 and 1965 in new rental vacancies was almost entirely in structures containing 5 housing units or more. In 1962, about 54 percent of the new units were in 5-or-more-unit structures, 31 percent were in 2- to 4-unit structures, and 15 percent were single-family houses. In 1965, about

Size of unit.--Vacant units available for sale in 1965 were generally larger than those offered in 1956. The median number of rooms was 5.2 in 1956 as compared with 5.4 in 1960 and 1965. There was also an increase in 3-bedroom units; the percentage was 38 in 1956, 53 in 1960, and 45 in 1965.

Number of housing units in structure. --Between 1956 and 1965, vacant units offered for sale were predominantly single-family houses. Since 1960, however, there is some evidence of a slight decline in the proportion of such units. In 1965, one-housing unit structures accounted for 91 percent of the vacant homeowner stock, which is slightly lower than the 95 percent reported for 1960. This downward trend in single-family housing is more pronounced inside SMSA's than outside these areas.

Plumbing facilities.--With respect to plumbing facilities, the level of quality continued to be higher for vacant homeowner units than for vacant rental units. In contrast to the vacant rental supply, however, vacant homeowner housing showed little improvement in plumbing between 1960 and 1965. In 1960, approximately 12 percent of the for-sale vacancies lacked plumbing facilities--private flush toilet, private bath, or hot running water. In 1965, approximately 10 percent of the homeowner supply lacked these facilities. Practically all the units lacking plumbing were located in nonmetropolitan areas.

Duration of vacancy.--There was less turnover among vacant units offered for sale than among rental vacancies. In 1956, about 19 percent of the vacant homeowner units had been vacant less than 1 month and 46 percent had remained vacant at least 4 months. In 1960, the corresponding percentages were about the same as in 1956. In 1965, short-term vacancies were at approximately

the

outside SMSA's than inside SMSA's. Vacant units outside SMSA's were closely associated with houses built in 1939 or earlier and with houses lacking plumbing facilities. By contrast, long-term vacancies located in metropolitan areas included a high proportion of units not over 15 years old (built in 1950 or later) and only a few units without plumbing.

Sale price asked.--The median sale price asked for homeowner vacancies has increased. The median was $11,100 in 1956, $13,400 in 1960, and $13,200 in 1965. Vacant for-sale units located inside SMSA's were higher-priced than those 10cated outside SMSA's. Inside SMSA's, the median was $12,800 in 1956, $13,100 in 1960, and rose to $14,100 in 1965. Outside SMSA's, the median was $9,500 in 1956, increased to $11,300 in 1960, and dropped to $10,800 in 1965.

The increase in median sale price reflects the upward shift in sale prices of vacant units during the 10 years and is closely associated with the increase in the supply of recently built vacancies available for sale. Since 1956, relatively fewer vacancies are offered for sale at the lower price levels and more at the upper levels. As shown in tables 12 and 13, practically all the

changes in sale price levels occurred during the period 1956 to 1960. Sale prices $15,000 and over were reported for 27 percent of the homeowner vacancies in 1956, but for 41 percent in 1960 and 39 percent in 1965. Conversely, 43 percent of the homeowner vacancies were to be sold for less than $10,000 in 1956, but only 30 percent in 1960 and 34 percent in 1965.

Table 12 shows that the sale price asked for vacant units was closely related to the size of the unit and the presence of plumbing facilities. Most of the higher-priced units had 5 rooms or more and all had private plumbing facilities. A large proportion had not been previously occupied at the time of enumeration. Furthermore, the

figures in table 13 show that units built in 1960 or later comprised the largest percentage of the higher-priced vacancies.

Year structure built. --Although there was no change in the level of homeowner vacancies between 1962 and 1965, there was an increase in the proportion of newly built units. In 1962, about 24 percent of the vacant homeowner units were built in 1960 or later; in 1965, about 33 percent of the vacancies were built in 1960 or later (table 14).

80

0

0

1956

1957

1958

1959

1960

1961

1962

1963

1964

1965

V Under $10,000