In response to a congressional request, GAO analyzed the loan activity of 40 State and local bond issuers that borrowed in the tax-exempt market between December 1981 and July 1982 and provided preliminary information regarding: (1) the extent to which lower income homebuyers are benefiting from the mortgage revenue bonds; (2) the effectiveness of Federal purchase price ceilings and State and local income limits in targeting loans to the intended households; and (3) the efficiency of mortgage revenue bonds in general. Preliminary analysis indicated that mortgage revenue bonds are costly when compared with both the benefits they provide and the costs of similar alternatives. This is because tax-exempt housing bonds also provide large tax savings to bond purchasers. GAO calculations indicate that the long-term revenue loss to the Treasury could be roughly four times the benefit provided to homebuyers in the form of reduced monthly mortgage payments. Direct grants to lenders would substantially reduce Federal costs while still providing equivalent mortgage interest savings to homebuyers. A carefully structured tax-credit for homebuyers could also have the same effect. Mortgage revenue bond proponents argue that the positive economic effects of additional home purchases outweigh the cost. GAO found that a high percentage of tax-exempt subsidized homebuyers would have bought without subsidy. GAO also found that most subsidized home loans were not made to low- and moderate-income households, but rather to those who probably could have purchased homes without assistance. In the absence of Federal income guidelines, State and local jurisdictions have set their own income ceilings, and most jurisdictions set ceilings that allow the participation of relatively affluent households.