Loan Zone: An Analytics-Based Approach to Risk-Based Auto Lending

This article is adapted from CU Rise Analytics’ whitepaper, “Rising Auto Loan Rates—What does this mean for your Credit Union?”  

Credit unions have made notable gains in the auto loan market in recent years, driven in part by the lower rates they have offered. Going forward, a new approach to setting rates—an analytics-based approach—should give credit unions new power to stay competitive when adjusting in response to changes in federal rates.

CUs’ Recent Market Gains

Credit union have expanded the number of auto loans they hold in the market. According to NCUA 5300 call report data compiled by CU Rise Analytics, of 108 million auto loans in the market as of the second quarter of 2017, 23 million (21.29 percent) were funded by credit unions. This increased to 24 million by Q2 2018. Additionally, in 2017, the number of CU auto loans grew 9 percent, while for the overall auto loan market, the number of loans grew only 4.8 percent.

In terms of auto loan balances, the overall market grew by 7.8 percent in Q1 2017 and 2.3 percent in Q1 of 2018. For credit unions, the auto loan balance growth rates were 15.2 percent and 11.1 percent, respectively. Credit unions held $319 billion of auto loan balances, almost 28.7 percent of the total in Q1 2018.

Importantly, credit unions’ average interest rate for new cars was found to be 3.65 percent in June 2018, less than the market average by more than 1 percent. According to data from Statista.com, the average interest rate for a new car 60-month loan term was 4.83 percent in July 2018.

CUs’ low rates give them a huge advantage over their competitors. Interest rates play a vital role in driving people’s search for financing. The 1 percent difference has likely made a big difference in growing CUs’ auto loan balances.

To continue to be successful in the auto loan marketplace, credit unions need to quickly read and adjust rates according to the market, portfolio and membership risk trends. Traditionally, credit unions have determined their rates by looking at such factors as their costs to obtain funds from the Federal Reserve and other sources; their operational costs, such as processing costs, salaries, utility expenses, office rent, marketing costs, overhead, etc.; and the risk premium they are collecting borrowers to offset any default.

In a competitive market, a financial institution might artificially reduce its interest rates to attract borrowers, driving other organizations to follow suit and lower their rates, diminishing their profits. Major lenders like big banks often define the base market rate to benefit their market position and capture prime and super prime loans. Smaller FIs with lower marketing budgets set their rates after these market leaders. Credit unions, which typically fall in that category, often play the role of a lower-cost option for consumers.

A New Approach to Risk-Based Pricing—The Analytics Way

Analytics are providing new-found power when determining interest rates. Today’s analytics technology allows credit unions to consider multiple risk variables in different permutations and combinations. Here is a list of a few important factors being used in analytics models for determining interest rates for auto loans.

Sophisticated data models can be built that scrutinize borrower profiles for various triggers. If a borrower had a bad loan in the past or suffered a financial crisis, the risk will be set as high. If that borrower showed a conscious effort to pay back the loan or rebuild his relationship with the credit union, the rate could be lowered. The attributes of risk vary from applicant to applicant based on current and past behaviors. Risk models allow credit unions to decide on the rate based on the individual borrower during the underwriting of loan.

These models generally run on two types of data: bureau data and performance data. Bureau data provides insight into past activity and behavior of the applicants. Performance data gives information for a particular loan taken or a broader relationship held with a financial institution. Modeling can help identify risky loans with accuracy, but to do so, the data needs to be well organized.

Analytics can also be used in other areas pertaining to loan products. It can help in calculating the risk of the overall portfolio by analyzing each and every loan individually across different categories. Credit unions can also leverage analytics to find members that have the most likelihood to leave the CU and refinance with other financial institutions.

To continue to grow their place in the lending market, CUs need to offer competitive rates to borrowers while minimizing risk, charge-offs and delinquencies. They can do so alone or by partnering with analytics companies like CU Rise.

Today’s higher-rate environment is going to make the auto loan market more competitive and impact the number of loans taken by borrowers—especially in the subprime group, which is generally charged higher interest rates. Adopting data analytics is the best bet for credit unions to adjust rates in accordance to federal funds rates and still stay competitive.
 

Suchit Shah is COO and  Priyanka Pandya is senior consultant at CU Rise Analytics, Vienna, Virginia. CU Rise Analytics is a global CUSO helping credit unions leverage the power of data to better understand their members. We are a value driven, one-stop shop for not only CU’s detailed data analytics initiatives but for best practices and advisory services. Learn more about us at www.cu-rise.com.