CHENNAI: Opinions of credit-rating agencies influence the allocation and pricing of capital, yet much of the demand for ratings exists because regulators, investment mandates and financial institutions require them.
That raises an uncomfortable question: If mandatory credit ratings disappeared tomorrow, would anyone still pay for them?
The question becomes more urgent as AI makes data and analytical capability increasingly abundant. Financial statements, market data, payment behaviour, supply chains, news, litigation and thousands of other signals can increasingly be analysed continuously.
If everyone can access powerful models and vast amounts of data, what remains scarce? I believe the answer is independent judgement, accountability and trust.
The traditional rating asks how likely will an issuer meet its obligations over a defined period. AI changes the economics of an industry faster. Disappearing competitive advantages and disruption of business models could happen before financial statements reveal the damage.
The rating agency of the future will ask: Is this enterprise resilient enough to remain a going concern as its environment changes? That means looking beyond leverage, liquidity and cash flows to six dimensions of resilience:
■ Financial resilience: Can the balance sheet absorb shocks and fund transformation?
■ Business-model resilience: Will competitive advantage survive technological and industry disruption?
■ Technology resilience: Can the organisation adapt fast enough as technology changes?
■ Management resilience: Can leadership recognise and act on structural changes?
■ Governance resilience: Can the Board challenge management when yesterday’s strategy is no longer viable?
■ Trust resilience: Will customers, employees, lenders and investors continue to trust the institution when circumstances become difficult?
This is no longer simply a rating. It is an independent assessment of enterprise resilience.
The parallel is auditing, which relies substantially on sampling. Technology is increasingly making full-population analysis possible. Credit assessment can undergo a similar transformation. AI-enabled institutions could continuously examine the entire population of financial, operational, market and external signals, identifying anomalies and emerging deterioration.
The machine can detect the signal. The institution must interpret it. The moat will no longer be data.
As data becomes a commodity and Models become cheaper, the rating agency cannot build its future around owning better data or a better model.
Its moat will tend towards independence, judgement, transparency, speed and trust. The institution, that identifies deterioration early, explains it clearly, changes its assessment quickly when evidence warrants it, and stands behind its judgement, will retain value.
Perhaps the answer is not to preserve mandatory ratings at all. Large or systemically important enterprises could instead be required to obtain independent opinions on three distinct dimensions:
■ Audit of financial statements
■ Credit worthiness to meet financial obligations
■ Enterprise resilience to remain viable through structural change
Rating agencies may need to expand beyond rating while protecting the independence that gives them credibility. Their future could lie in continuous surveillance, early warning, transparent methodologies and broader assessments of enterprise resilience.
—(The writer is founder, Alive Consulting)