LCR headroom built on Level 2 assets and five wholesale investors

Start with the arithmetic. Net cash outflows = A$1.9bn / 1.18 = A$1.61bn, so headline surplus HQLA is about A$290 million, or 18 percentage points.

Then test eligibility. Level 1 = 54% x 1.9 = A$1.03bn. Level 2A = A$646m and Level 2B = A$228m, so Level 2 totals A$874m. Level 2 assets may not exceed two-thirds of Level 1 (the 40% cap): the cap is A$684m, so A$190m of Level 2 is ineligible. Adjusted HQLA = A$1.03bn + A$0.68bn = A$1.71bn, and the adjusted LCR is about 106%, not 118%. The Level 2B cap at 15% of adjusted HQLA is not breached, so the binding constraint is the aggregate Level 2 cap.

Then test the outflow assumptions.

• Investor concentration. A$1.4bn maturing inside 90 days, rolled from five institutions, is a name concentration the 30-day metric barely reflects. If those five reprice or withdraw together, the real outflow is larger and faster than modelled.
• Broker deposits. A$600m treated as stable retail is generous. Such deposits are typically less stable, rate-sensitive and capable of leaving quickly; a conservative treatment materially raises the outflow denominator.
• Specials. A$800m repricing next quarter is a repricing risk rather than a 30-day outflow, but it makes the deposit base fragile and raises the cost of any NSFR fix.
• Subsidiary lines. A$250m of committed undrawn lines is a contingent outflow that must be modelled, not netted away.

Answer to the board: the headroom is real but thin and overstated at the headline. Before approving 8% loan growth, present the LCR net of the Level 2 cap, stress the five-investor concentration, and reclassify broker deposits conservatively.

NSFR mechanics. At 103% the bank has three percentage points of room. Growth of 8% in a loan book of this size adds a substantial block of required stable funding - long-dated loans attract a high RSF factor - while short-dated wholesale funding contributes little or nothing to available stable funding. Three-month NCDs, and wholesale funding with less than six months to run, effectively receive no ASF credit, and funding between six and twelve months receives only partial credit. Stable retail deposits receive close to full ASF credit; less stable and broker-sourced deposits less.

So growing assets while rolling three-month NCDs mechanically drives the NSFR down, and an 8% book increase funded that way would likely push it below 100%.

Funding actions, in order of preference

1. Terminate the short-dated wholesale dependence. Issue term debt beyond twelve months, which earns full ASF credit and simultaneously lengthens the 90-day maturity wall. This is the highest-value action.
2. Shift the deposit mix. Grow transaction and savings balances; be selective on broker-sourced and specials-driven deposits, which are the most expensive and least stable form of ASF.
3. Consider Tier 2 issuance, which provides 100% ASF credit and addresses the capital constraint at the same time - relevant given the board's framing of the question.
4. Match asset and liability tenor. Price longer fixed-rate lending off term funding rather than the short curve, so asset growth carries its own stable funding.
5. Securitisation can be used, but check how retained versus sold tranches are treated for both LCR and NSFR before treating it as a solution.

Recommendation: growth is fundable, but not on the current funding mix. Gate the 8% plan on term issuance and a deposit-mix shift, and report the projected LCR and NSFR under the plan, not just the current position.

Principles. Triggers must be leading, observable daily or weekly, unambiguous, and tied to a pre-agreed action ladder with named owners. A plan that only activates once the LCR is near 100% is a post-mortem, not a plan.

Market and counterparty triggers

• NCD or term spread to BBSW widening beyond a set threshold (say 40bp) or a failed rollover with any of the five panel investors.
• Rating outlook change or downgrade, which also activates collateral and counterparty concerns elsewhere.
• Reduced availability of repo against the bank's Level 2A RMBS, or an increase in haircuts.

Firm-specific triggers

• Projected 30-day LCR below 110%, or NSFR below 100% on any forecast horizon.
• Deposit outflow above 2% in a week or 5% in a month, measured excluding normal seasonality.
• Drawdown of more than 50% of the A$250m subsidiary back-up lines.
• Any single wholesale investor declining to roll, or concentration in the top five rising further.
• Adverse movement in the mark-to-market of any derivative collateral obligations.

Action ladder: each stage should pre-specify actions (draw committed lines, activate the central bank repo facility, accelerate deposit campaigns, sell liquid securities, securitise, slow lending growth), the decision-maker, and the communication plan for investors and the supervisor.

Testing: run a reverse stress test to identify the combination of deposit run, wholesale non-renewal and collateral call that breaches the minimum, then run a quarterly walk-through against actual market data from a historical stress period. A trigger that has never been tested is a hypothesis, not a control.