Sovereign, corporate and consumer credit ratings are used to assess the creditworthiness of borrowers. Yet these ratings often fulfill other functions as well, serving as measures of more general qualities of countries, businesses and individuals. When ratings are used outside the context of lending, we call it ‘off-label use.’ This paper develops the argument in the context of consumer lending and discusses the use of credits scores in the U.S. by car insurance companies in calculating premiums, landlords in selecting tenants, and employers in hiring workers. We argue that off-label use can have harmful effects through two mechanisms: error propagation and enhanced performativity. Both amplify small initial differences, exacerbate inequalities, lock borrowers in upward or downward spirals and increase economic inequalities. Turbo performativity results when measures influenced by earlier credit scores become direct inputs for calculating new credit scores. Off-label use of consumer ratings, therefore, should be treated not just as a privacy issue but also as a factor in economic polarization.