Corporate debt in India is waiting for its moneyball moment

Big-ticket corporate lending blowups are not limited to public-banks. They are common occurrences in private banks, mutual funds and private-equity sponsored non-bank finance institutions. It would be a miracle if there were long-term systemic flaws in the risk processes of corporate debt. Only a fraction of such defaults can be attributed to borrower fraud and corrupt bankers. However, an unnecessary generalization of this article may have prevented an introspection of existing risk methods. There have been growing changes such as high data usage, but only a few top banks have basically re-ined the lending process. For most banks, the corporate lending process is very subjective, sometimes driven by the toe of selective terms and selective readings of data. Sometimes used bespoke ‘expert models’ also have limited risk assessment capability.

Moneyball in Corporate Loans: Billy Beans of corporate debt should come forward. Billy Oakland is the legendary manager of the athletics baseball team. His exploits in using analytics to select players for his team and its subsequent success Moneyball: The Art of Winning an Unfire Game and a film of the same name. Like our Indian Premier League, baseball teams have a budget to buy players for this season. Prior to the introduction of Billy Analytics, the selection and pricing of players was determined by baseball experts with decades of experience recognizing talent.

When selecting players, apart from the basic performance statistics and their immediate success, the personality traits of the players perceived by these professionals influenced their selection. Underwriters can draw parallels between financial statements and how they view perceived management quality in selecting clients for credit. In the film, Talent Spotter, who does not like analytical approach, says, “It’s not about baseball numbers, it’s not science” and “baseball is unlikely to be understood only by people.” If one replaces baseball with corporate debt, it reflects the ideas of an undercover underwriter with doubts about analytical intervention in big-ticket loans. The film manages to entertain as well as inform. The data analyst has a degree in Economics but has a prior knowledge of baseball. In fact, analyst Paul Depodesta himself is a baseball scout. Ultimately, sports analytics has improved the decisive skills of talent sportsmen like Billy Bean, but has not replaced them.

Human Element vs. Skill: Without any pressure to meet the quarterly credit targets, it may be worth noting that the models present are unlikely to beat the best underwriter who cares for himself. But consistency issues grow. In marginal cases, without a black-and-white answer, research suggests that the mental state of the underwriter is important. On high-sentiment days, more loans are approved and higher defaults are observed for those loans. (The Mood for a Loan: The Kajal Effect of Sentiment on Credit Origin, Agarwal et al., 2012)

Challenges of ‘Expert Models’: Some lenders are aware of the problem of decision volatility, but continue to rely heavily on qualitative factors. In some cases, this leads to the use of partial-scale estimation methods such as expert judgment models. Such models use tough information (financial reports) and soft information (management quality). Soft elements, being qualitative, difficult to verify and their interpretation depends on the individual. Research suggests that they can lead to worse loan decisions, especially if the person collecting it is under time pressure. One study found behavioral nuggets: when loan executives have previous sales experience, the softer the information gathered the more optimistic the bias will be explained. (Making Sense of Soft Information: Interpretation Bias and Loan Quality, Compable et al., 2018)

As such, hybrid models that use qualitative and quantitative variables may not always be statistically consistent. Qualitative inputs often do not stand up to statistical consideration of their estimated power. It is not surprising that one would have less predictive power than models that use hard information, if one adds soft information that is collected erratically.

Start by measuring decision quality: One may first like to identify and measure the shortcomings of the traditional underwriting system and determine what is working and what is not. There are several parts to the assessment and judgment before the final warranty decision of ‘Go’ and ‘No Go’. Some of these components are industry perspective, financial statement forecasts, future loan requirements and maintenance quality. The value of these components improves as the credit call goes from the initial underwriter recommendation to the credit committee and is high. While most lenders are able to track the final decision quality by the debt delinquency rate, some may not be able to assess its quality at each stage. For example, if financial projections deviate from the actual results, even if the account is guilty, it indicates a weakness in the process.

A high delinquency rate can result from bad liability or a bad economy. Similarly, a low delinquency rate is not always a high-quality surety decision. Knowing the flaws helps to create a definitive analytical intervention. As in Moneyball, the goal is not to replace the underwriter, but to make good decisions.

Deep Mukherjee visits Faculty of Finance and Risk Management Consultants at IIM Calcutta

Subscribe to it Mint Newsletters

* Enter a valid email

* Thank you for subscribing to our newsletter.

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *