Recent concerns over housing affordability for low-income households appear to be difficult to resolve by developing policy options that focus only on traditional single-family owner-occupied dwellings and/or rental apartments. In terms of developing a housing policy that would improve the quality of housing for lower income families, it seems appropriate to explore the merits of an often-ignored alternative, namely manufactured housing.

In this respect, this paper employs the American Housing Survey (AHS) between 1993 and 2001 to compare owned manufactured housing to rental housing and traditional owned housing as a tenure alternative for low-income households. This comparison for the three tenure types is made along several dimensions. Initially, a general comparison is made regarding the quality ranking of the structures and neighborhoods, housing cost, and housing affordability. Subsequently, regression models are used to determine the factors that affect the households’ neighborhood and structural quality rankings and changes in those rankings over time. Separate equations are estimated for each tenure type. In addition, a model is estimated to consider the factors affecting household mobility and the extent to which these effects differ for the three tenure types. Finally, the appreciation of conventional owned housing is compared to the appreciation for owned manufactured housing in two cases. These two cases are, first, where the structure is owned but the land is leased, and second, when both the land and structure are owned.

Our results contradict several preconceived notions regarding manufactured housing. Specifically, there are four important observations that are implied by the results. First, manufactured housing is found to be a low-cost housing alternative. Importantly, it is observed to have higher average quality rankings across both the neighborhood and structural dimensions of housing services than rental units. These results hold even when the sample is stratified by metropolitan and non-metropolitan location. As such, on average, manufactured housing appears to be a “good value” for low income households.

Second, those factors that contribute to lower structural quality or lower neighborhood quality, as well as changes in those quality measures over time are similar between manufactured housing and owned housing. These finding suggest that a properly planned manufactured housing development will not automatically deteriorate over time and communities do not have to develop uniquely different policies to include manufactured housing in the mix of units that make up the housing stock.

Third, the factors affecting household mobility across the three tenure types are quite similar. Of particular importance is the fact that like traditional owned units, and in direct contrast to rental units, the longer a household resides in manufactured housing at a specific location the less likely they are to move while holding constant other factors that influence household mobility. This finding suggests that having owned manufactured housing in a neighborhood will not inherently increase mobility among households living in manufactured units and, therefore, lead to neighborhood instability as associated with rental units.

Finally, while manufactured housing without land ownership does not appear to be a particularly good investment, ownership of land in conjunction with an owned manufactured unit generally provides a positive return. These returns do appear to be associated with relatively high variance. However, with manufactured housing as a generally lower cost alternative to renting, low income households might be expected to accumulate more wealth (through savings and land value appreciation) while in manufactured housing than in a rental unit. In sum, owned manufactured housing appears to be a relatively attractive option for housing low-income families in a manner that would be beneficial to them and to the communities in which they live.