Mid-Tier Bank: Record Profit Built on Bond Gains While NIM Compresses
A mid-tier Pakistani commercial bank has total assets of PKR 1.2trn: gross loans PKR 480bn (ADR 42%), investments PKR 620bn, of which PKR 300bn is AFS with an average duration of 4.5 years and PKR 320bn is held to maturity. Sixty per cent of its PKR 950bn deposit base is current and savings accounts. The SBP has cut the policy rate by 400bp over nine months. Reported profit after tax rose 22% year on year to PKR 34bn, helped by PKR 28bn of pre-tax realised gains from selling AFS securities and a lower effective tax charge; shareholders' equity also rose on AFS revaluation. Net interest margin fell from 5.6% to 3.9%. ADR is below the 50% threshold above which the SBP requires incremental deposits to be parked with the central bank at below-market return. The board, citing 'record profit and record equity', proposes raising the payout ratio from 30% to 55% while simultaneously issuing fresh Tier-2 capital at KIBOR+450bp.
1. Normalise sustainable earnings and evaluate the proposed 55% payout.
Read the worked answer
Strip out the non-recurring item. The PKR 28bn realised gain is the disposal of an appreciating asset, not earnings power. At a 29% tax rate it contributed about PKR 19.9bn after tax. Normalised PAT ≈ 34.0 - 19.9 = PKR 14.1bn.
Size the recurring hit from the rate cuts. Earning assets of about PKR 1.1trn repricing 170bp lower costs roughly PKR 18.7bn of annual pre-tax net interest income. That is the run-rate the bank must replace, and it is larger than the one-off gain that flattered this year's profit.
Test the payout against cash earnings. A 55% payout on reported PAT of 34bn is PKR 18.7bn, which is more than the entire normalised profit of 14.1bn - a payout ratio of 133% of sustainable earnings. Even at 30% (PKR 10.2bn) headroom is thin once the NII run-off is reflected next year.
Second inconsistency. Raising Tier-2 at KIBOR+450bp to fund a dividend that exceeds recurring profit borrows expensive subordinated money to pay out a one-off gain. Economically the bank is distributing capital and simultaneously paying a spread to replace it.
Recommendation. Base the dividend on normalised earnings with a 25–35% payout, disclose the normalised figure prominently, and use retained capital plus the Tier-2 issue to fund loan growth into a lower-rate environment rather than to plug a payout gap.
2. Is the AFS revaluation a genuine strengthening of capital? What will happen to it?
Read the worked answer
Arithmetic of the revaluation. Using duration as an approximation: 300bn x 4.5 x 4.0% ≈ PKR 54bn pre-tax, about PKR 38bn after tax, credited to OCI and hence to equity. That is why 'record equity' appears alongside falling margins.
Why it is not free capital. Three reasons. First, it is unrealised and reverses if rates rise; a 100bp back-up removes roughly PKR 13.5bn of it. Second, it disappears entirely if the bonds are held to maturity - the gain is only crystallised by selling, which is exactly what generated this year's PKR 28bn realised gain, so part of the equity uplift has already been converted into a one-off P&L item. Third, and most important, the gain is the present value of a stream of coupons that is about to fall: reinvesting PKR 300bn at 400bp lower costs roughly PKR 12bn of pre-tax annual interest income for the life of the portfolio.
Analytical treatment. Treat the AFS revaluation as a timing transfer, not as capital generation for dividend or growth purposes. Show equity both as reported and excluding the AFS reserve, and compute the sustainable return on that adjusted equity.
The costly error to avoid. Selling AFS to 'lock in' the gain crystallises it in P&L, destroys the highest-yielding assets on the book, and leaves the bank reinvesting at the bottom of the rate cycle. Realising gains to protect a dividend is a classic value-destroying trade.
3. What operational levers should management pull on ADR and NIM, and what are the risks?
Read the worked answer
The ADR lever. At 42%, the bank sits just below the 50% threshold at which incremental deposits must be placed with the SBP at a penal below-market return. Every rupee of deposits above that threshold earns a sub-market yield, so the effective marginal cost of deposits jumps sharply. Two options: (a) grow loans by about PKR 60–70bn to cross 50% (roughly 12–14% loan growth), or (b) shrink the deposit base and let the ADR rise arithmetically.
Why (a) is not automatic. Lending into a slowing economy at the bottom of a rate cycle at KIBOR+300–400bp may not cover through-the-cycle credit cost if infection ratios rise. Growth must be evaluated on risk-adjusted spread, not on the arithmetic of crossing a regulatory line.
Why (b) is cleaner but painful. Repricing or shedding expensive term deposits raises ADR without credit risk, but 40% of the book is term funding which is sticky and reputationally sensitive; a deposit run is a worse outcome than a penal placement.
NIM levers. Push CASA from 60% toward 65% through transaction and digital propositions; reprice term deposits faster than the policy rate falls; and shift part of the AFS book out of long duration into floating-rate or short PIBs to reduce reinvestment risk.
Risk to flag. A rate cut cycle compresses deposit spreads with a lag, so the reported 3.9% margin likely understates the eventual trough. Model NIM assuming further compression before stabilisation, and stress the loan book at 200bp higher credit cost before committing to loan-led ADR repair.