Housing units in China are sometimes sold with fittings (including internal finishes, interior decorations, fixtures, electrical appliances, etc.) and sometimes as bare units (without fittings). This study formulates hypotheses that explain these phenomena and test these hypotheses with empirical data from a sample of 1,704 development projects in China that were sold in the primary market between 2003 and 2007. Based on developers’ risk-aversive behavior, we hypothesize that they are less likely to sell new units with fittings in areas with low land prices. The value of fittings is the highest if the developers can accurately predict the tastes of potential buyers. The risk facing the former is that the fittings may not suit the latter. The risks are higher when the cost of installing the fittings constitutes a relatively high proportion of the market value of a housing unit, which is the case when land prices are relatively low. Since housing construction costs are similar across the different regions in China, land price is a major determinate of housing prices. Due to the problem of information asymmetry between developers and buyers, the former’s reputable or goodwill is important. The market would give a higher valuation for fittings provided by more reputable developers. In addition, potential buyers may worry that the fittings had been used to cover up defective or sub-standard work. Therefore, we hypothesize that reputable developers are more likely to sell housing units with fittings, especially for developments with a high construction cost to market value ratio. Demand and supply conditions play a role in determining a developer’s decision to sell units with fittings. When supply is relatively abundant, competition amongst developers is keen. Developers will tend to compete by providing fittings as sweeteners to attract buyers. The provision of fittings may also be a non-price competition strategy used by developers to differentiate their products from that of their competitors. Therefore, we hypothesize that developers are more likely to sell housing units with fittings when demand is weak or when supply is abundant. Finally, we conjectured that as people’s incomes increase, their time becomes more valuable. Therefore higher income households are less likely to buy bare units, since the costs of shopping for interior decoration contractors, finishes and fittings, etc. are higher for higher income households. We conjecture that developers tend to provide units with fittings in areas with higher average household income. With the exception of the last hypothesis, our empirical results supported all hypotheses. The lack of empirical evidence to support the last hypothesis could be due to a possible opposing force. This force arises from the possibility that higher income households have more sophisticated tastes that are more difficult to predict. Developers, being risk-averse, would tend to supply bare housing units in high income areas. The results in this study contribute to our understanding of developers’ behaviors in the real estate market, particularly in China. The results are also of practical relevance to home buyers, developers, investors, and policymakers. / published_or_final_version / Real Estate and Construction / Doctoral / Doctor of Philosophy