“The best laid schemes o’ Mice an’ Men
Gang aft agley,
An’ lea’e us nought but grief an’ pain,
For promis’d joy!”
– Robert Burns
On Turning her up in her Nest, with the Plough, 1785.
Customarily credit rating in India depends on multiple factors. Such factors include financial performance (profitability, cash flow, debt level, and liquidity); industry risk (creditworthiness risk); macroeconomic conditions, management quality (management’s competence and stability); repayment history (credit repayment behaviour); type of debt and lending and borrowing history. Thus rating is a function of marketing strategies, competitive edge, level of technological development, operational efficiency, competent management, risk factors, cash-flow trends and potential, financial flexibility, etc.
Credit ratings, which are current opinions regarding creditworthiness, are usually assigned for specific instruments or for a defined purpose and often continuously. However such ratings are limited by their respective definitions and do not constitute investment or financial advice. Ratings and substantiating documentation, viz., audited financial statements, annual reports, programme appraisals etc., however, provide an insight in the operations of the rated entities. Ratings definitions, criteria and methodologies are periodically reviewed for transparency, credibility, independence and objectivity, without being compromised by conflicts of interest, abuse of confidential information or other undue influences.
Of late, the issues of Artificial Intelligence (AI), geopolitical dynamics and climate change have become important because of their extensive impact on the performance of the firm and industry with concomitant ratings implications. The transforming world heightens apprehensions, discordance and incertitude. But “worry never fixes anything” (Ernest Hemingway) in impermanent life. In conformity with the winds of change sweeping the world, it has been increasingly realized that the credit rating process in India, as indeed elsewhere, needs to go beyond the basics to be expanded to incorporate contextually significant factors such as AI, geo-political risks and climate change-related risks because of their wide-ranging ramifications on real-world outcomes.
With a growing and diversifying Indian economy in general and the banking and financial sector in particular, the credit rating industry in India is expected to go places. There is a manifest need for greater care and vigil in crafting the emerging India and carrying out mid-course corrections, as and when needed. Consequently, it is necessary to set benchmarks and assess the emerging elements for a substantive change in credit rating analytics.
AI- Creating value beyond the hype
Rating agencies are information providers that evaluate the quality of an issue or issuer aimed at diminishing information asymmetry and enhancing transparency in financial markets. The game-changing impact of AI on the global economy, transforming businesses and employment has been increasingly recognized and leveraged by inextricably intermeshing it into their companies’ workflows across sectors and geographies.
AI has pervaded multiple fields of innovation and development, e.g., driverless cars, online product recommendations, personalized healthcare, and natural language conversations. Against this overarching canvas of proactive engagement with technology and the tipping point of widespread AI integration, rating agencies and researchers need to dovetail their methodologies to examine governance with accountable and transparent systems. Accordingly, AI could deftly handle repetitive, time-consuming tasks, thereby not just reversing the downward productivity trend, but also triggering a major sustained surge in productivity. Such changes could lead to a sharper focus on unleashing more creative endeavors because there is magic “when human creativity meets AI ingenuity” (Dave Birss).
In today’s digital age, the issue of the current capabilities, future potential, and inner workings of AI and generative AI (GenAI) assumes greater immediacy because of the compelling need to process voluminous data and surging computational power, together with advancing technology, quick decisions and forecasts to drive cost savings, revenue streams, productivity, and efficiency. In a fluid and dynamic world, traditional algorithms need to be modified to facilitate speedier organizational transformation, innovate smarter, and unlock the power of AI to go to scale. GenAI can generalize this information and eliminate the need for new algorithms in the limitless future.
Initially, standard statistical techniques, such as discriminant analysis, principal component analysis, logit, probit, and other regression methods were used by researchers for long, forming the basis of traditional credit scoring methodologies. With rapid technological advancements, data management, and AI- the great breakthrough technology becoming integral to everyday life in professional and private contexts, new models based on machine learning and AI emerged to offer new solutions and possibilities for a better future. Voilà, AI is on a roll! Some well-known AI techniques used for assigning ratings include artificial neural networks (ANN), Support Vector Machines (SVM), k-Nearest Neighbors (k-NN) and Classification and Regression Trees (CART).
As breakthrough technologies, e.g., fire, steam power, electricity, and computing evolved, AI is similarly poised to evolve and become more capable in identifying technological solutions that can discern ingenious indications. McKinsey & Co. aver that the speed improvement “can be translated into an increase in productivity that outperforms past advances in engineering productivity, driven by both new tooling and processes”. With AI metamorphosing technology, many researchers are exploring artificial neural network models because of their greater accuracy, adaptability, and robustness. Support Vector Machine techniques are, however, recognized for their classification ability. Furthermore, several researchers highlight the decision tree-based approach’s superior degree of interpretability and flexibility in feature extraction. A one-size doesn’t fit all and there cannot be a solitary solution to complex macro issues spanning distinctive regional and sectoral diversities. Such peculiarities necessitate more disaggregated risk models and tweaked standard approaches. There is no panacea in this “new normal”. For, the World Bank recently stressed “From India to Senegal: driving local solutions” is the key.
Additionally, financial markets’ inherent uncertainty and unpredictability complicate the creation of unimpaired models. Inaccuracies can lead to spill-over effects and contagion, as changes in ratings impact the pricing of various assets. Enhanced credit rating methodologies and robust AI systems to mitigate these challenges and reduce associated risks is imperative. Consequently, AI is remodelling credit rating agencies by streamlining data analysis, risk assessment, and fraud detection. Strategic digital transformation and value creation automate routine tasks, enhance predictive modelling, and potentially reduce bias. AI enhances speed, accuracy, and data integration aligning with the long-term strategy and client needs optimising credit rating processes.
While progress will accelerate in unforeseen ways enhancing sluggish productivity growth since the global financial crisis, persisting challenges include ensuring data quality, model explainability, cyber defence manoeuvres and techniques, robust security, and building trust in AI. Such challenges for the future of work hamper “the revolution of rising expectations” in India and raise concerns about AI being the new “snake oil”. At its root, these and other challenges necessitate practical insights through real-world case studies that can be applied in multiple security scenarios to address issues of accessibility, affordability, and quality. Security scenarios must secure all potential vulnerabilities, and ensure full protection of clients’ networks and sensitive information. Such challenges were starkly highlighted in Cybersecurity protection company CrowdStrike’s faulty software update, which triggered a global crisis in technology systems in July 2024 with massive disruptive effects across the development spectrum, particularly banks, brokerage firms, and trading infrastructure. AI also has concerns about control and competition, existential risk, public image, and regulatory challenges.
There are larger issues of AI’s ethical implications, its potential downsides, the blurring of the dividing line between truth and fabrication, and the risk-reward matrix in the AI black box. Consequently, Mr. Shaktikanta Das, the RBI Governor stressed the compelling need for strengthening cyber-security in an increasingly digital world at the Global Fintech Festival (GFF), Mumbai 2024. Since securing AI systems facilitates an indomitable advantage, there must be a steadfast commitment to a cyber-secure culture and adopt and emulate best cybersecurity practices.
Geo-political Risks
Interest rates remained benign to aid capital market issues and borrowings by corporates from the banking system. This helped the credit rating industry. Further, post-COVID pent-up demand improved the risk profile of corporates in India contributing to more than a unity credit ratio (upgrades/downgrades). However, the domestic improvement was partially offset by geo-political risks, including the ongoing Ukraine war and the Middle East/Red Sea scenario.
Going forward, in the near term, the government’s focus on enhancing spending in the infrastructure sector augurs well for the economy and the industry. This will be somewhat counterbalanced by geo-political transitions marked by the end of unipolarity (theatres of conflict in West Asia, Russia-Ukraine, and Indo-Pacific, geo-economic fragmentation, oil price volatility, the rising significance of the global South, etc.) and the northward interest rates. Such myriad concerns significantly impact the credit rating business in India by heightening economic instability, amplifying sector-specific risks, and raising regulatory uncertainties. These paradigm shifts lead to intensified global tensions, disrupting trade and causing fluctuations in the global markets. They damage investor confidence and shrink capital flows into emerging market economies (EMEs) like India.
With the evolving global scenario, informed decisions assume greater significance in industries, e.g., energy, infrastructure, and manufacturing, which are sensitive to international trade dynamics and supply chain disruptions. The volatility in global oil prices and energy supplies, stemming from the Russia-Ukraine conflict, has further marred the financial outlook for Indian companies reliant on imports or foreign debt, leading to potential downgrades. These conflicts have influenced sovereign credit ratings globally, with a distinct possibility of raising borrowing costs and perceived business risks in strife-torn regions. The ongoing regulatory changes and uncertainties, exacerbated by geopolitical conflicts, require credit rating agencies to be vigilant and adaptive in their assessments of evolving corporate balance sheets and their effect on the overall credit ratio as the broader market sentiment and liquidity are incessantly influenced by events of considerable contemporary significance.
Climate Change-related Risks
While climate change does not represent a “do or die” moment for the credit rating business, its profound impact is expected to significantly reshape the credit rating business in India by introducing new risks that affect firms, industries, and financial institutions. Manifesting deeper climate risk and transition management risks can be classified into physical risks, natural disasters, resource scarcity, rising sea levels, and transition risks, e.g., policy changes, technological disruption, and reputational risks.
The increasing frequency and intensity of extreme weather events damage assets, disrupt operations and hike insurance premiums. Resource scarcity and rising sea levels imperil agricultural yields, manufacturing processes, and coastal infrastructure. With the government enforcing stricter environmental regulations and carbon taxes, and industries transitioning to low-carbon technologies, traditional business models face disruptions affecting creditworthiness. Such macroeconomic disruptions necessitate adroitly navigating the rapidly altering clean transition investment landscape.
Accordingly, credit rating agencies need to incorporate these pre-investment and post-investment climate capabilities into their overall assessments to meet the complex challenges of today and the unforeseen events of tomorrow. This requires evaluating a company’s exposure to physical and transition risks, its climate strategy, and its financial performance under various climate scenarios. “But man is not made for defeat… A man can be destroyed but not defeated” (Ernest Hemingway, The Old Man and The Sea, 1952).
With the stakes rising higher, the rising emphasis on green finance will create new opportunities for rating agencies to assess the creditworthiness of green bonds and climate-friendly projects in a more comprehensive and holistic framework. This will, however, make the credit risk assessment process more complex. By adapting to these challenges in a timely and effective manner, credit rating agencies can provide investors with more accurate and thorough evaluations, reflecting the financial impact of climate change in an evolving environment- an environment, where all elements are critically in ferment.
Note: This is a modified version of the Op-ed first published in Financial Express on September 17, 2024.
ABOUT THE AUTHOR
Dr. Manoranjan Sharma is Chief Economist, Infomerics, India. With a brilliant academic record, he has over 250 publications and six books. His views have been cited in the Associated Press, New York; Dow Jones, New York; International Herald Tribune, New York; Wall Street Journal, New York.
