Credit constraints in Colombia: evidence from the use of credit cards among low- and middle-income individuals

Purpose - The authors estimate the determinants of the value of purchases of semi-durable goods using permanent and transitory income, and the demographic characteristics of customers. The purpose is to identify whether individuals face remaining liquidity constraints, and how this friction affects their purchases. Design/methodology/approach - This study uses anonymized data of 516,525 credit card holders, with more than 7,501,065 records of purchases between 2010 and 2015. The authors decompose the income of individuals into permanent and transitory components to test the prevalence of the life cycle–permanent income hypothesis (LC–PIH). Determinants of the value of purchases for constrained and unconstrained consumers are estimated, considering the period in which individual characteristics are valid, the decisions not to make purchases in some months, and the potential endogeneity of the interest rate and the transitory component of income. Findings - The authors present evidence of liquidity constraints for individuals who have used a high percentage of the credit limit on their cards. For these restricted customers, the value of purchases is inelastic to the interest rate, whereas the response is sizable for customers who are less restricted. The restricted customers increase the value of purchases when faced with increases in their credit limit. The elasticity of the value of purchases of semi-durable goods to permanent income is less than that for transitory income; regardless of the constraints, this still supports the LC–PIH. Research limitations/implications - This credit card is targeted at low- and middle-income individuals in Bogotá. Although the results might be considered as indicative of the behavior of those with similar characteristics in Colombia, the authors regard this work as the study of a particular case. A limitation of this work is that the authors do not have alternative sources of credit at an individual level. Practical implications - The broad credit channel of monetary policy does not apply to the restricted customers. This should be considered not only by the monetary authority, to understand the true extent of this policy, but also by the financial institutions that use this business model. The monetary authority should be cautious not to overreact when intervening in the money market to try to prompt an adequate consumer response. Social implications - Financial institutions have the policy of modifying the credit limits of their customers' credit cards which affects the well-being of restricted customers. Given that the card is aimed at low and middle income individuals, the credit limits of customers who use a high percentage of their credit limit might be increased. Originality/value - This is the first paper to study liquidity restrictions with a retail credit card in Colombia and Latin America using information on customers' characteristics. The results are highly relevant for the implementation of monetary policy.