A non-bank lender in used-vehicle finance has grown its loan book 40% a year for three years to Rs 18,000 crore. Gross NPAs are 1.9%, down from 3.4% three years ago. Net interest margin is 7.8%. Cost of funds has risen 90 basis points over the year and the company has passed on 40 basis points. Average loan tenure is 42 months. Provision coverage is 52%. The stock trades at 3.4x book against a sector median of 2.1x. Eighty per cent of incremental disbursement in the last year went to first-time borrowers with no formal credit history.
Write a buy or sell recommendation. Lead with the call, support it with the figures given, and name what would make you wrong.
100 points, 60% to pass.
The central question is whether the falling NPA ratio is credit quality or arithmetic. A book growing 40% a year has a denominator that outruns the numerator: loans written this year have not had time to go bad, so a rapidly growing book mechanically reports a falling NPA percentage. Against 42-month tenure and 80% first-time borrowers, the vintage is untested. Margin is also compressing — 90 basis points of cost absorbed against 40 passed on. A strong pitch asks what NPAs look like on a static-pool basis and what coverage of 52% implies if they normalise. The bull case is genuine underwriting edge in a segment banks cannot serve.