Preventing NPAs with AI powered Customer Due Diligence

In March 2024, MadeCo, a respected kitchen manufacturing company in Melbourne Australia, entered voluntary administration (liquidation), burdened by $2.5 million in debts. The collapse was largely driven by two insolvent clients who left $520,000 in unpaid invoices, triggering a cash flow crisis that MadeCo could not overcome. Despite efforts to recover, the financial strain proved insurmountable, forcing the company to close its doors. The shutdown leaves 30 employees jobless and 70 creditors unpaid, highlighting the precarious nature of supply chains and the devastating impact of delayed payments. 

 There is enough evidence that delayed customer payments can wreak havoc on a company's cash flow, could lead to closure. Consider Carillion, the UK construction and facilities management behemoth that collapsed in 2018. Similarly, Toys "R" Us in the USA faced bankruptcy in 2017, with delayed payments from customers and suppliers exacerbating its financial troubles. Debenhams, the British department store chain, went into administration in 2020, partly due to delayed customer payments. Thomas Cook, the UK travel company, collapsed in 2019, with delayed payments from customers and partners contributing to its financial instability. Abengoa, the Spanish multinational energy and engineering company, filed for bankruptcy in 2016, citing delayed customer payments as a factor disrupting its debt repayment capabilities. These are the stories of companies that are known to public, yet the counterparty risk largely remains unaddressed across all sectors.

While late payments are a widespread issue in the business world, they particularly affect small and medium-sized enterprises (SMEs). In the United States, more than half (55%) of B2B invoiced sales are not paid on time. Similarly, 54% of SMEs face late payments from their customers, with payments arriving an average of 6 days after the due date. The problem persists across fiscal years, as evidenced by 55% of small businesses carrying outstanding invoices from the previous tax year. The sectors experiencing high Non-Performing Asset (NPA) rates in India include Food Processing, Textiles, and Infrastructure, indicating these industries face significant challenges in loan repayment and financial stability.

In India also the Small and Medium Enterprise (SME) sector appears most vulnerable, with an 8.2% NPA rate in the State Bank of India, followed by the corporate sector at 7.1%, agriculture at 5.9%, and retail at 4.8%. These NPAs significantly affect the banking sector, leading to reduced profitability and increased provisioning requirements, while also impacting the industrial sector through reduced credit access and higher borrowing costs. The broader economy suffers from decreased employment opportunities and potential financial stability issues. However, recent trends as of June 2024 show improvement, with gross NPAs of scheduled commercial banks falling by 15.2% year-on-year, and the overall GNPA ratio improving to 2.8% from 3.8% a year ago.

The NPA story has a long tail, and its shadow extends over entire company operations. NPAs directly lead to financial losses as the outstanding loan amounts may become unrecoverable, while the opportunity cost of lost interest income further strains profitability. Additionally, regulatory requirements mandate provisioning for NPAs, tying up capital that could otherwise be used for growth or investment. Indirect costs, such as legal expenses for recovery and administrative overhead for managing bad loans, also add to the financial burden. Over time, a high NPA level can erode investor confidence, increase borrowing costs, and weaken the company's financial position, hindering its ability to operate and grow sustainably.

As general formula can be represented as  Overall NPA Cost=(Direct Loss)+(Opportunity Cost)+(Provisioning Cost)+(Recovery Cost)+(Administrative Cost)

Why are we in this situation ? Arcane credit models

Financial data often resembles a carefully assembled family photograph: a snapshot meticulously arranged to present an idealized image. Just as family photos rarely capture the full complexity of relationships and daily realities, financial statements can obscure the true dynamics of a company's financial health while it is in motion. Also, credit models often rely on outdated financial data, potentially compromising their effectiveness in today's rapidly changing economic landscape. This time lag in data can lead to inaccurate risk assessments and suboptimal lending decisions.  

 Good news for these stakeholders is that in recent years new data sources have revolutionized risk assessment and credit modeling. Financial institutions now leverage diverse alternative data types to gain deeper insights into borrowers' creditworthiness. These innovative sources include social media activity, mobile phone usage patterns, utility and rent payment histories, geolocation information, email and phone data, IP analysis, behavioral metrics, and trust scores. By harnessing these varied data points, risk assessors can paint a more comprehensive picture of an individual's financial health. This multifaceted approach allows for more accurate predictions of creditworthiness, especially for those with limited traditional credit histories, potentially opening doors to financial services for previously underserved populations.

Credit models, while adept at assessing financial capacity, often overlook crucial intangible factors that define a company's true value and potential. By focusing primarily on quantitative metrics and historical data, these models risk missing key indicators of a firm's resilience and growth prospects. A more holistic approach is needed, one that balances traditional financial analysis with an evaluation of qualitative factors such as management integrity, innovation capacity, and strategic vision. This comprehensive assessment can uncover hidden strengths and provide a more accurate picture of creditworthiness.

By incorporating measures of intent, integrity, and potential alongside traditional financial metrics, credit evaluation can more effectively identify companies primed for sustainable growth. This holistic approach enables creditors to assess not just current financial status, but also a company's underlying values, future aspirations, and capacity for innovation. The result is a more nuanced understanding of risk, leading to lending relationships that are both more precise in their risk assessment and more aligned with businesses demonstrating genuine potential for long-term success and positive impact.

While we may not realize, late payments leading to NPAs have a far-reaching impact on the overall economy, extending well beyond individual businesses. They contribute to a slowdown in economic growth by reducing cash flow and spending, particularly affecting small and medium-sized enterprises (SMEs) which are vital employers and economic drivers. The ripple effects include job market instability, with potential salary delays and reduced hiring, as well as decreased business investment and innovation. Late payments can lead to increased business failures, higher borrowing costs as companies seek short-term loans, and significant reductions in economic output. Moreover, they create supply chain disruptions that reduce overall economic liquidity. In countries like the UAE, where consumer spending is a major component of GDP, these effects can be particularly severe.

DigiAlly's Trust Score is an innovative risk assessment tool that uses advanced AI and machine learning to provide a comprehensive evaluation of companies. It goes beyond traditional credit risk analysis by examining operational efficiency, market resilience, and industry standing. The system also assesses intangible factors like a company's nature, intentions, integrity, and growth potential. By analyzing diverse data sources, including financial reports, market indicators, social media, and alternative data, Our Score offers a holistic view of a company's risk profile and prospects. This multifaceted approach enables more informed decision-making for lenders, investors, and other stakeholders, providing a deeper understanding of a company's overall health and potential than conventional risk assessment methods.

Savvy investors and analysts must look beyond staged portraits to uncover the underlying financial realities that drive business performance and risk. To maintain relevance and accuracy, financial institutions must evolve their credit modeling approaches to incorporate more timely and diverse data sources, ensuring a more dynamic and responsive risk evaluation process.

#credit #creditmodels #AI #machinelearning #Creditrisk #manufacturing #finance #banking #SME #lending #supplychain #digially