Household craft production and consumption play a key role in modeling the degree of economic control at Moundville. If production was household or corporately centered, then both utilitarian and non-utilitarian artifact classes should have a dispersed distribution of consumption across the site. If artifact production was organized at the polity level by elites, then artifact classes associated with elites should have a restricted distribution of consumption in specific areas where elite-controlled production occurred. To understand the way that craft production and consumption were negotiated at Moundville, this study examines data from off-mound residential areas excavated as part of four seasons of the Early Moundville Archaeological Project (EMAP). There are three objectives to examining and analyzing these data. The first objective is a site wide consumption pattern gathered from previous investigations at Moundville. The second objective is subsurface sampling, which allows for a site-wide comparison of the abundance of artifact classes through an observation of density measurements. The third objective, the excavation units, provides distribution, abundance, and context data that are compared across different areas Moundville and different contexts. The data lend evidence to suggest that certain expectations of the political economy model are not adequately represented in off-mound areas. First, there is evidence for both non-utilitarian crafts and production debris in residential middens, including abundances that are comparable to mound-top data. Second, craft production is found in domestic areas, and does not seem to be concentrated in specific areas of the site. With regards to ritual economy models, the data did not follow the pattern suggested by Kelly's Osage model, which focused stages of production; rather, Knight's mode that sees differing corporate groups specializing in specific goods with complementary exchange is a better fit with certain aspects of my data. Utlimately, data from the three objectives indicate variation in the amounts of locally available goods, but with nonlocal goods, there is an overwhelming pattern of redundancy through time. To best account for this pattern, I propose an alternative ritual economy model, ritual replication, which I feel best accounts for the pattern of redundancy in artifact classes across Moundville's habitation areas.