Learning Objective 2 Identify the questions management wants to ask and the relevant data
As the costs of gathering and storing data continue to decrease, organizations are saving large quantities of data. How these data should be used depends on the questions management would like to have answered. The management accountant plays an important role in helping management decide which questions and what data have the potential to create value for the organization. These are the first three steps in the decision-making process. We describe these steps and the management accountant’s contribution in the context of the investment decision facing Sierra Investments.
The first step in the decision-making process is gaining an understanding of the business problem. That is, what questions does the management accountant need to examine to advise the manager on a business decision. We illustrate the process by returning to Sierra Investments, which manages assets on behalf of clients. The firm is considering investing in loans offered by the peer-to-peer lending platform PeerLend Digital (PD), which connects individual borrowers to individual investor-lenders through an online platform. The platform allows all parties to bypass traditional banking infrastructure and their high operations costs. In exchange, borrowers get lower interest rates and investors get higher returns. The loans available on the PD platform range from $1,000 to $10,000. PD collects and keeps data on all loans that investors fund from the platform.
PD assigns a grade rating to each loan based on an estimate of the riskiness of a loan, that is, the probability of the borrower defaulting. To estimate risk, PD uses sophisticated models using historical loan data and features such as credit score, annual income, and amount of loan. PD assigns low risk loans a grade of A or B, moderate risk loans a grade of C or D, and high risk loans a grade of E or F.* PD assigns an interest rate to each grade that reflects the grade’s level of risk. Exhibit 11-2 shows the default rate (number of default loans divided by the total number of loans), interest rate, and the adjusted net annualized return for a certain selection of PD loans. Note that the interest rate increases with default risk. The adjusted net annualized return is the return after considering PD service fees and losses from defaults. For example, investing in C-grade loans yields a net annualized return of 7.6% on an interest rate of 13.0% with a default rate of 19.6%. Note that when a borrower defaults, an investor does not lose all of his or her investment. That is because a borrower often pays some interest and principal before the default.
| Grade | A | B | C | D | E | F |
|---|---|---|---|---|---|---|
| Default rate | 5.3% | 11.4% | 19.6% | 26.0% | 34.8% | 40.8% |
| Interest rate | 6.0% | 9.3% | 13.0% | 17.4% | 22.5% | 31.1% |
| Adjusted net annualized return | 5.2% | 7.0% | 7.6% | 7.9% | 7.4% | 7.7% |
How does PD earn money? PD charges the borrower a 3.5% origination fee on each loan and a 1% servicing fee to investors for collecting interest and principal. In the event of a default, PD charges investors collection fees to recover money from the borrower.*
Paige has two options:
The question is “Should Paige do further data analysis beyond what PD has already done to classify loans into different grades?” What might she learn from doing so?
Forming questions is a critical first step for the management accountant at the outset of any data science task. The widespread availability of data and access to computing resources allows the management accountant to run many different analyses. It is important, however, to pause and ask whether it is worthwhile to do so. In this case, Paige decides to run more analyses because the default rates and the cost of defaults are very high. If she can identify default loans more precisely, Sierra Investments can increase returns by avoiding those loans. Her next task is to identify data that would be most useful and relevant for the analysis.