Read Liberti and Mian (2010 JF) and in no more two pages provide answers to the following questions: 1) Explain the possible sources of endogeneity that the authors were facing if they were estimating a regression where the dependent variable is a measure of collateral requirements (e.g., collateral value to loan amount) and the explanatory variable is a measure of financial development using their cross-country dataset. 2) Explain how they side-stepped these problems using predictions about the collateral spread. Write down the relevant models and explain what each one solves and under what assumptions. 3) Discuss whether instrumental variable analysis could have been used to deal with endogeneity problems you discussed in 1) above. Explain the conditions that an external instrument must satisfy. Could an internal instrument have been used instead? For example, could the authors have used the Hausman-Taylor IV estimator instead? Explain your answer. 4) Assume you were given access to the dataset used in this study for writing one of the chapters in your thesis. What research question would you address? State your question in an intuitive way, explain why is an important question from an economic point of view, and how you would go about testing it. Try to use the data to your advantage.