This paper aims to decompose the banking spread using a structural macroeconomic model. We embedded a general equilibrium framework with loans to households and entrepreneurs that may be in default, a banking sector in monopolistic competition and subject to administrative costs, and we also added a tax structure related to bank intermediation. These characteristics for the composition of the spread are in line with the empirical literature on banking spread determinants in Brazil and with the accounting decomposition of the spread made by the Banco Central do Brasil (BCB). Our quantitative findings reveal that household spread reduction is greater when we increase competition in the banking sector (3.77 p.p. quarterly or 54 percent decrease compared to baseline calibration). Furthermore, reducing administrative cost is the most effective way of diminishing entrepreneur spread (1.35 p.p. quarterly or 46 percent decrease compared to baseline) and it is also capable of diminishing household spread (2.5 p.p. quarterly or 36 percent decrease compared to baseline). Results also suggest some careful actions by policy makers only supported by BCB accounting decomposition without an economic model underpinning the analysis. This dissertation also reveals some challenges regarding the inclusion of credit default with a banking sector in imperfect competition and its contribution to the spread formation.