Table 6.1 — Average annual production of new housing units in tlie United States by type of unit, 



1920-1977, witfi projections to 2030 



(Thousand units) 





Total 





Conventional units 





Mobiles 





Period 











Used as 



Not used as 





demand 



Total 



One-family 



Multifamily 



Total 



primary 

 residences 



primary 

 residences 



1920-29 



803 



803 



527 



276 



— 



— 







1930-39 



365 



365 



301 



64 



— 



— 



— 



1940-49 



809 



780 



641 



139 



29 



— 



— 



1950-59 



1,522 



1,459 



1,208 



251 



63 



41 



22 



1960-69 



1.648 



1,443 



929 



514 



205 



164 



41 



1970-79 



2,145 



1,757 



1,102 



655 



388 



310 



78 









Low 



projections 









1980-89 



2,490 



2,160 



1,620 



540 



330 



260 



70 



1990-99 



2,080 



1,800 



1,460 



340 



280 



220 



60 



2000-09 



1,970 



1,680 



1,260 



420 



290 



230 



60 



2010-19 



2,060 



1,770 



1,300 



470 



290 



230 



60 



2020-29 



1,720 



1,470 



1,070 



400 



250 



200 



50 



Medium projections 



1980-89 



2,590 



2,250 



1,680 



570 



340 



270 



70 



1990-99 



2,240 



1,930 



1,540 



390 



310 



250 



60 



2000-09 



2,300 



1,960 



1,410 



550 



340 



270 



70 



2010-19 



2,270 



1,930 



1,390 



540 



340 



270 



70 



2020-29 



1,980 



1,680 



1,180 



500 



300 



240 



60 









High 



projections 









1980-89 



2,700 



2,340 



1.740 



600 



360 



290 



70 



1990-99 



2,410 



2,070 



1,610 



460 



340 



270 



70 



2000-09 



2,760 



2,350 



1,610 



740 



410 



330 



80 



2010-19 



2,550 



2,160 



1,510 



650 



390 



310 



80 



2020-29 



2,450 



2,070 



1,370 



700 



380 



300 



80 



Sources: Housing starts, 1920-49 and 1960-62 — Forest Service estimates 

 derived from data in the following sources: U.S. Department of Commerce. Bureau 

 of the Census. Housing construction statistics, 1889 to 1964 1966. 1950censusol 

 housing. Vol. I. Pts 2. 1953, U.S. Department of Labor. Bureau of Labor Statistics. 

 Nonfarm housing starts. 1889-1958. Bull. 1260. 1959; 1950-59 — U.S. Department 

 of Commerce. Bureau of the Census. United Stales census ot housing, 1960. Vol. 

 IV. Pt. 1-A, 1962; 1963-77 — U.S. Department of Commerce. Bureau of the Census. 

 Housing starts. Cons. Rep. Ser. C20-78-8. 1978. 



Total mobile homes. 1940-49 — Forest Service estimates derived from data in the 

 following sources: U.S. Department of Commerce. Bureau of the Census. 7950 



census of housing. Vol. i. Pt. 1. 1953; 1950-59— U.S. Department of Commerce, 

 Business and Defense Services Administration. Construction review. 7(3). 1961. 

 and Mobile Home/Recreational Dealer Magazine. Marl<et study, 1967-1968. 1969; 

 1960-63 — US Department of Commerce. Business and Defense Services 

 Administration. Construction review. 12(8) 1966; 1964-77 — U.S. Department of 

 Commerce. Bureau of the Census. Housmgs(ar(s. Cons. Rep. Ser. C20-78-8. 1978. 



Mobiles used as primary residences. Forest Service estimates derived from data 

 in the following source: US. Department of Commerce. Bureau of the Census. 

 United Stales census of housing, 1960. Vol. IV, R. 1-A. 1962 



Projections: U.S. Department of Agriculture. Forest Service. 



These shifts in housing production reflect changes 

 in demand associated with household formations, the 

 replacement of units lost from the housing stock, and 

 the maintenance of an inventory of vacant units. Anal- 

 yses of projections of these factors indicate another 

 rise in housing demand in the early 1980's — the 

 medium projection averages nearly 2.6 million hous- 

 ing units annually for the decade. Housing demand 

 drops in the 1990's, a reflection of the decline in birth 

 rates in the late 1960's and early 1970's. After the 

 1990's, demand will increasingly depend on popula- 

 tion growth. As a result, demand is likely to rise in the 

 decade after 2000, then decline slowly through the 

 remainder of projection period. 



The type of housing units demanded — single- 

 family, multifamily, mobile homes — is of major 

 importance in projecting demands for timber prod- 

 ucts because of large differences in average per unit 

 use. 



Over the last 50 years, about 70 percent of all hous- 

 ing units produced have been of the single-family 

 type. However, there has been wide variation in the 

 mix of housing types produced. Major booms in the 

 relative importance of multifamily housing occurred 

 in the 1920's and early 1970's. Production reached a 

 peak of over 1 million units in 1972 — about 35 per- 

 cent of total housing production — before falling to 

 less than 300,000 units in 1975. In the late 1950's, the 



198 



