This dissertation is composed of three independent empirical articles that use data on home and kitchen appliances and image equipment sales of one specific chain store and their clients characteristics. The first one investigates the effect of the client s characteristics, such as gender, age, education and income, on the probability of receiving a discount. It intends to investigate if the pricing authority that is delegated to the salespeople is being well used. If the effect of delegation were to implement price discrimination, clients with higher income would have a lower probability of getting a discount. On the other hand, discounts can be given for a different reason than the one desired by the firm. Once the customer recognizes that the salesperson has some discretion over the prices, she may wish to engage in bargaining to decrease the final price. The outcome of this process may also depend on the customer s characteristics if they affect the customer s skill in bargaining. In this case, the greater the client s ability, the bigger the discount. Since we find that the probability of receiving a discount is increasing in the client s income and education, and the value of the discount is also bigger for people with higher income and better education, we conclude that the discount giving is not a result of price discrimination, but a result of a bargaining process. The second article estimates the effect of asymmetric information on consumer loans, using the payment plan chosen by the client and his observable characteristics. We decompose the effect into an adverse selection and a causal component using a technique similar to Adams, Einav and Levin (2009). We find significant marginal effects for both components and that the causal effect of the payment plan choice is greater than the one caused by the adverse selection. A key aspect of the identification strategy is the choice of the exogenous variable. We use a dummy variable that is one when the month of the purchase is one of the saldão months, that is, when one special payment plan is also available: eleven monthly payments, with zero interest rate. The variable used clearly influences the payment plan choice but the repayment behavior does not depend directly on it, since it is not directly related to the value of the debt collateral. The third article studies the retail market effects of the first tax stimulus episode of 2009, when the government reduced the IPI (Industrialized Products Tax) of certain categories of kitchen and home appliances. Tax rates vary depending on the product. The reduction occurred to four different products: refrigerators, stoves, washing machines and non automatic washing machines. We estimate the effect of the IPI reduction on prices and the composition of demand. We find that the IPI reduction lead to a 5.97 percent reduction on price, when we use a simple linear model and a 2.98 percent reduction, when we control for product fixed effects. The characteristics of consumers who acquire the appliances do not change significantly with the reduction.