Abstract:The effectiveness and the efficiency of a credit or an insurance program critically depend on the design of the program.In designing a program, the government should carefully consider various factors, including incentives of private parties, possible information advantages of private parties, and competition and liquidity in the target market. Due to failures to reflect these factors in the program design fully, many credit and insurance programs may not be serving their target populations effectively and efficiently.Some partial loan guarantee programs allow private lenders to set the lending rate within a range. In these programs, lenders can increase the profit by lending more to high-risk borrowers who pay a high interest rate. Resulting inefficiencies include underserving of low-risk borrowers, excessive profits for lenders, and excessive costs to the government. Possible solutions include tying the guarantee fee to the lending rate and taking a partial ownership of loans. A main problem with direct loans and full loan guarantees is prepayments by low-risk borrowers. Two main ways to recapture the value of the prepayment option are a prepayment penalty and a high lending rate. The preferable way is a prepayment penalty because a high lending rate may further prompt low-risk borrower to prepay. To finance highly risky activities, such as technology start-ups, the government should take an equity position. Without sharing a few "jackpots," the government cannot recoup its investment. Nevertheless, the government takes a debt position in venture capital investments. The consequence is either that the government suffers large losses or that the government fails to finance truly high-risk, high-return projects. In some insurance programs, the government shares risk with private parties in a way that favors private parties. Private insurers have opportunities to take advantage of mispricing of insurance policies. Policyholders underpay during good times, but they don't necessarily make up the underpayment during bad times. An efficient way of sharing risk is that the government bears all of catastrophic risk and let private insurers bear policy-specific risk. The government implicitly guarantees GSE debts. Since GSEs have incentive to take excessive risk, the government is providing a very expensive guarantee free of charge. An explicit guarantee made available to both GSEs and private entities at a fee should lower the cost to the government, make the market more competitive, and contain financial crises effectively.For proper budget discipline and efficient management, it is important to estimate the costs of credit programs accurately. The cost estimation for credit programs has two main steps: estimating future cashflows and discounting future cashflows to arrive at the net present value of cashflows. Some models estimating future cashflows do not fully consider the effects of economic fluctuations. The consequence can be a systematic underestimation of the cost of a credit program. Some also argue that discounting uncertain cashflows with Treasury rates leads to an underestimation of the cost to taxpayers by ignoring the risk premium. This argument is off the mark. Discounting with Treasury rates is appropriate, considering that budgeting is more about accounting accuracy than economic optimization and that the costs of other government programs do not include the risk premium. The focus should be on estimating cashflows more accurately, fully taking into account the possibilities of extraordinary losses cause by economic downturns.

Managing Consumer Credit Delinquency in the US Economy: A Multi-Billion Dollar Management Science Application

Big Data Credit Report in Credit Risk Management of Consumer Finance

Loss rate forecasting framework based on macroeconomic changes: Application to US credit card industry

Credit Risk Management of Consumer Finance Based on Big Data

Modelling repayment patterns in the collections process for unsecured consumer debt: A case study

Deciphering Big Data in Consumer Credit Evaluation

How Do Machine Learning and Non-Traditional Data Affect Credit Scoring? New Evidence from a Chinese Fintech Firm

Risk and return prediction for pricing portfolios of non-performing consumer credit

A Decision Support Approach for Accounts Receivable Risk Management

Machine Learning-Driven Lending Decisions in Bank Consumer Finance

Checking account activity and credit default risk of enterprises: An application of statistical learning methods

A machine learning workflow to address credit default prediction

Data Analysis on Credit Card Debt: Rate of Consumption and Impact on Individuals and the US Economy

Data Driven Credit Risk Management Process: a Machine Learning Approach

Pragmatic segmentation-based credit risk management using Machine Learning

Could Coaching Improve Consumer Credit Use Behavior? Evidence From a State Program

Design and Implementation of Federal Credit and Insurance Programs

Data Analytics for Credit Risk Models in Retail Banking: a new era for the banking system

Towards a data-driven debt collection strategy based on an advanced machine learning framework