Risk Committee Dormancy at a Fast-Growing Microfinance NBFC

The weaknesses are structural, not isolated. First, the board risk committee is not governing risk; by spending its time on credit approvals it has become a shadow credit committee, leaving enterprise risk, liquidity and conduct risk unowned. Second, the risk appetite statement is decorative: 'moderate' cannot be measured, delegated or breached, so no one can escalate. Third, the CRO lacks independence and resources because reporting to the CFO places risk under the function that is incentivised to grow the balance sheet. Fourth, the asset-liability committee has been dormant during a clear maturity mismatch: 78% short-term funding against a book where 62% matures in 12 to 24 months. Fifth, the branch incentive pays for disbursement volume without clawback, which rewards weak underwriting and explains the ₹21 crore of missing income proof. These interact. The growth target pressures branches to disburse, the incentive pays them to do so, internal audit finds the consequence, but the CRO cannot escalate independently and the board risk committee is not looking. Liquidity risk then compounds credit risk because early defaults and funding rollover pressure arrive together. The board should treat the combination as a risk-governance failure, not a series of operational lapses.

The board should separate risk oversight from credit sanction. The board risk committee should have a charter focused on risk appetite, top and emerging risks, liquidity, conduct, model risk and the effectiveness of the three lines of defence; credit approvals should sit with a management credit committee within delegated limits. The CRO should report functionally to the board risk committee, with administrative reporting to the CEO rather than the CFO, and should have an independent budget and unfettered access to the committee. The asset-liability committee should meet at least quarterly, with a standing liquidity dashboard covering coverage ratio, survival horizon, concentration of borrowings and rollover risk. The risk appetite statement should be rebuilt as a set of quantitative limits and triggers: minimum liquidity coverage, maximum short-term borrowing share, sector and borrower concentration caps, asset-quality triggers, and a conduct metric on incentive-driven lending. Each limit should have an owner, a reporting frequency and an escalation ladder. The branch incentive should be redesigned to pay on portfolio quality and repayment, with clawback for early delinquency. Internal audit should follow up on the ₹21 crore exception, and the board should receive a consolidated risk report that shows limit utilisation, breaches and remediation status rather than only credit approvals.

The immediate issue is that the company is already outside its own liquidity target while the CEO proposes faster growth. The board should not approve a 40% growth plan until the funding and liquidity position is stabilised. I would press for a board-approved liquidity contingency plan, a rollover schedule for short-term borrowings showing weekly gaps, and a freeze on new product or branch expansion that consumes liquidity. The asset-liability committee should be reconvened within days, not at the next quarterly cycle. Management should present a credible plan to restore the liquidity coverage ratio to 100%, including diversification into longer-term funding and a review of loan tenor mix. On credit, I would require an immediate sample review of the ₹21 crore flagged loans, provisioning adequacy, and a pause on incentive payments until the clawback policy is approved. The board should also test the growth plan against stress scenarios: a 20% funding withdrawal, a rise in early delinquency, and a ratings downgrade. Finally, the board should minute that growth is conditional on risk limits being met, so that management cannot later claim an implicit mandate for expansion. Escalation to the audit committee is appropriate because the internal audit findings and incentive design affect financial reporting and control.