HSBC Holdings (HSBC) · Deep analysis

Report previewHSBC.USReport date Last updated
← Back to reports
Report contents

At a glance

  • Business model: HSBC gathers deposits in Hong Kong, the UK, corporate and institutional banking, and international wealth, then lends or invests them and earns transaction-banking and wealth fees. Its $1.8 trillion deposit base at December 31, 2025 and loan-to-deposit ratio near 55% support pricing and let it avoid marginal credit, but surplus liquidity earns wholesale rates, tying revenue to short-term rates.
  • Financial quality: FY2025 net income was $22.3 billion, an 11.6% ROE on a 0.71% ROA; nine months of 2026 produced $22.8 billion and roughly a 15% annualized ROE. Revenue rose 5.3%, so lower credit charges, notable items, or tax explain part of the gain. CET1 of 13.9% at 3Q26 leaves capital room, though buybacks and OCI losses compete for it.
01

Business model

Business model: deposits first, fees second

HSBC sells its balance sheet and the network attached to it. It gathers deposits from retail, corporate and institutional customers, then puts that money to work as loans, trade finance and securities, earning the spread; fees come on top from transaction banking, cards, wealth management and insurance. The payer is the borrower or the fee-paying client, and no single customer group dominates. The company organizes itself into four businesses: two home markets (Hong Kong and the UK) and two network businesses (Corporate and Institutional Banking, and International Wealth and Premier Banking). Together they held $1.8 trillion of customer deposits at December 31, 2025, per the company's investment case.

This is a financial intermediary, and a mature one. It targets a 50% dividend payout ratio for 2026 through 2028, which is a capital-return policy, not a growth one. The stage is better described as restructuring inside maturity: a 2025 reorganization, a cost program and the privatization of Hang Seng Bank are all attempts to push capital toward the network franchises and away from activities management calls non-strategic or low-returning.

The network carries half the funding

The four deposit figures in the same disclosure add up to nearly the group total: more than $540 billion in Hong Kong, more than $360 billion in the UK, more than $595 billion in Corporate and Institutional Banking and more than $280 billion in International Wealth and Premier Banking. If those segments partition the base, the corporate and institutional arm is the single largest funding pool, ahead of Hong Kong. That reframes what HSBC is: a transaction bank for companies and institutions that also owns two large retail franchises. Management said more than 90% of FY2025 revenue came from Banking NII, Wholesale Transaction Banking and Wealth. Trading and capital markets generate the headlines and the volatility, but not most of the revenue. That 90% figure is management's characterization of a year partly under the old structure; a full year of reporting under the four-business split is the test of whether it holds.

A 55% loan-to-deposit ratio is a returns tradeoff

Almost half the deposit base is not lent. The surplus sits in liquid assets earning market rates, which means group revenue tracks short-term dollar, Hong Kong dollar and pound rates more closely than it tracks loan demand. That is the clearest fragility in the model: a cutting cycle squeezes income on the surplus pool even when the franchise wins every customer it pitches. It also explains the strategy. Wealth balances of $2.1 trillion and transaction banking fees are the fee engines meant to grow faster than spread income shrinks. Management targets revenue growth rising to 5% year over year by 2028 and a return on tangible equity of 17% or better for 2026 through 2028, both excluding notable items.

The same ratio is where bargaining power shows. Leadership in Hong Kong deposits supports pricing on the funding side; on the lending side HSBC is a price-taker against local banks and bond markets. A deposit-rich bank that barely lends half its base can walk away from marginal credit, which is a real advantage in a downturn. The cost is that excess liquidity earns wholesale rates instead of loan spreads. Management is buying resilience and paying for it with return on equity, and the 17% target is the price it has set for that choice.

Cost cuts are small next to the revenue question

The simplification program is meant to deliver about $1.5 billion of annualized savings by the end of 2026; $1.2 billion was achieved by the end of 2025, with $0.3 billion left to action in the first half of 2026. HSBC also plans to redeploy roughly $1.5 billion of costs out of non-strategic activities and expects $0.3 billion of synergies from privatizing Hang Seng Bank. Those numbers are modest for a bank this size. Cost-out helps, but a return target in the high teens will be settled by revenue and credit costs, not by $1.5 billion of savings. The Hang Seng transaction matters for a different reason: it folds two overlapping Hong Kong franchises into one, and the operational risk is customer and staff attrition in a market where HSBC is already the leader.

The mix figures above are the year-end 2025 position. HSBC published its third-quarter 2026 earnings release on October 27, 2026, but the segment deposit split is disclosed in this form only periodically, so the year-end numbers remain the cleanest read on funding.

What would overturn the framing is fee income from wealth and transaction banking growing fast enough to lift group revenue while rates fall and deposits hold. HSBC would then be a fee-compounding wealth and payments franchise that happens to carry a spread book, and the rate sensitivity described above would overstate its fragility.

Questions worth following

  • Can HSBC keep net new wealth inflows positive as rates fall?

    Wealth balances of $2.1 trillion supply the fee income meant to offset spread compression. If inflows turn negative, the fee engine stops covering a lower rate environment and the 2028 revenue target leans harder on rate moves the bank cannot control.

  • Can HSBC hold Hong Kong deposit share as rivals pay up for savings?

    Hong Kong is the group's strongest franchise and cheapest funding source. Losing share would either push funding costs higher or force HSBC to give up spread to keep balances, and both outcomes hit net interest income on the largest deposit pool the group controls.

  • Will folding Hang Seng into HSBC cost more Hong Kong customers than it saves?

    The deal is justified by about $0.3 billion of synergies and by combining overlapping Hong Kong franchises. If integration pushes away Hang Seng's distinct customer base or experienced staff, the savings shrink while the group's core deposit franchise weakens at the same time.

02

Financial quality

Financial quality: better 2026 earnings still hinge on rates

HSBC earns a thin spread on a very large balance sheet and multiplies it with leverage. FY2025 net income was $22.3bn, an 11.6% ROE on a 0.71% ROA and roughly $16 of assets per dollar of equity. Nine months into 2026 the same machine is running near a 15% annualized ROE, with $22.8bn of net income already above the whole of FY2025. Those earnings recur; they are also unusually dependent on the level of rates. Cash flow adds little to the judgment: operating cash flow went from $104.3bn in FY2021 to $26.4bn in FY2022 and $29.8bn in FY2025, which for a bank tracks deposits and the trading book more than earning power.

USD bn unless notedFY2021FY2022FY2023FY2024FY20253Q26Assessment
Revenue73.9576.17129.51143.29134.1038.39Adequate
Revenue less interest expense64.2553.7264.4467.4071.0221.01Strong
Revenue less interest expense / revenue (%)86.970.549.847.053.054.7Adequate
Net income13.9215.5623.5323.9822.297.55Adequate
ROE (%)7.058.2712.9612.9511.633.85Adequate
ROA (%)0.470.530.780.790.710.22Adequate
Net margin (%)18.820.418.216.716.619.7Adequate
Equity / assets (%)6.705.996.106.136.135.69Adequate, bank-normal
Asset turnover (x)0.0250.0260.0430.0470.0430.011Adequate, bank-normal
Operating cash flow104.3126.4339.1165.3029.77Not disclosedAdequate
Operating cash flow / net income (x)7.51.71.72.71.3Not disclosedAdequate
Capex3.564.413.693.884.66Not disclosedStrong
Free cash flow100.7522.0235.4261.4225.11Not disclosedAdequate

*3Q26 is a single quarter. Flow ratios are not annualized and are not comparable with the full-year columns.

Returns come from spread, not volume

Two parts of the history need care. Revenue is not comparable across 2022 and 2023: interest expense jumped from $22.5bn to $65.1bn while reported revenue rose 70%, far more than HSBC's actual growth, so the presentation changed. Revenue less interest expense, the steadier line, went from $53.7bn to $64.4bn. The current ratio is also uninformative, at 0.16 to 0.33 in 2023–25 against 28 to 40 in 2021–22, because current assets consist of cash in some periods and a much broader set of balances in others. Liquidity has to be read from deposit mix and the LCR, not that ratio.

Since 2023 the gain in returns has come from the price of money, not from volume. Assets grew 6.4% from FY2023 to FY2025 while asset turnover held near 0.04x. In FY2025 revenue fell 6.4%, but interest expense fell 16.9%, so revenue less interest expense rose 5.4% and its share of revenue climbed from 47.0% to 53.0%. The pattern repeats in 2026: over nine months, revenue less interest expense is up 11% year over year against 5.3% revenue growth. Book equity at 3Q26 ($195.9bn) is slightly below the FY2021 figure while assets are 16% larger, so buybacks have kept the ROE denominator flat. That supports the ratio and thins the cushion at the same time.

The 2026 step-up outruns revenue

The improvement is larger than the top line explains. Over nine months, revenue rose 5.3% and net income 31.6%, lifting the net margin from 17.1% to 21.4%. That $5.5bn gap has to come from lower expected credit loss charges, smaller notable items, or a lower tax charge, and the reported series does not separate them. The newest quarter sharpens the question: revenue jumped 14.7% year over year to $38.4bn, but net income fell 4.6% from 2Q26. Costs or provisions rose in the quarter, or 2Q26 contained something that did not repeat.

QuarterRevenueRev. less int. exp.Net incomeTotal assetsTotal liabilitiesEquityOperating cash flowROE, qtr (%)AssessmentTrend
1Q2533.8517.747.323,054.42,856.2190.8Not disclosed3.90AdequatePeak of 2025
2Q2533.7317.664.733,214.43,014.5192.6Not disclosed2.47WeakWeakening
3Q2533.4817.905.283,234.23,035.5191.4Not disclosed2.75AdequateStabilizing
4Q2533.0317.724.943,233.03,027.4198.216.582.54AdequateStable
1Q2633.8419.127.353,306.03,108.7196.8Not disclosed3.72StrongImproving
2Q2634.1619.047.913,438.23,241.1196.781.764.02StrongImproving
3Q2638.3921.017.553,444.83,248.6195.9Not disclosed3.85StrongFlattening

USD bn. Premises and equipment are well under 1% of assets (FY2021: $13.2bn against $2,958bn) and are immaterial to a bank's capital position, so the fixed-asset line is omitted. DBS Group Research, in a November 6 note, put 9M26 return on tangible equity excluding notable items at 18.8%, above the roughly 15% annualized reported ROE. At least part of that gap is basis (tangible equity, adjusted for notable items), so the underlying run-rate is strong but flattered by the adjustment.

Capital is the real constraint

Capex was $4.7bn in FY2025, or 3.5% of revenue, so free cash flow of $25.1bn looks ample. For a bank it says almost nothing about the risk-weighted assets that must be capitalized, and it is CET1 that constrains buybacks. DBS put CET1 at 13.9% at 3Q26, down 0.2 percentage points after $2.6bn of pre-tax losses through other comprehensive income from higher yields, and expects $7–8bn a year of repurchases in FY27–28. A 13.9% ratio leaves room. A repeated OCI drag of that size against buybacks of roughly $8bn would not, which is why capital headroom and the pace of repurchases are the same question.

Three things need checking before a 15% ROE is treated as the new base.

  • Composition of the 2026 step-up. The expected credit loss charge as a share of average loans, notable items and the tax rate each explain part of the $5.5bn. If provisioning is running below through-cycle levels, the earnings base is borrowed from later years, and so is the buyback.
  • Deposit mix and capital. DBS reported the Hong Kong CASA ratio slipping as customers moved into time deposits, and put the sensitivity of net interest income at about $2bn for a 100bp parallel shift. Deposit competition cuts both ways: it lifts asset yields and raises funding costs, and the capital effect arrives through OCI before it reaches the income statement.
  • Cash flow quality. Operating cash flow of $81.8bn in 2Q26 alone against $29.8bn for all of FY2025 is a balance-sheet event, not a measure of earnings quality. The LCR and the loan-to-deposit ratio belong beside it; neither appears in the ratio history above.

What would settle the assessment: credit charges normalizing while revenue still grows, and CET1 holding at or above current levels through the buyback program. A quarter where revenue rises and net income does not, as in 3Q26, is the pattern to watch.

Questions worth following

  • Can HSBC hold its 2026 earnings run-rate as credit costs normalize?

    Nine-month net income rose 32% on 5% revenue growth, so the higher ROE depends on credit charges, notable items and tax staying favorable. If they normalize, earnings and buyback capacity fall back toward the 2024–25 level.

  • Will rate-driven OCI losses slow HSBC's buyback pace?

    Buybacks of $7–8bn a year depend on CET1 generation. Higher yields push losses through other comprehensive income and reduce that capital, so the pace of repurchases is the clearest test of whether distributions are outrunning the capital the bank is earning.

  • Are Hong Kong savers shifting out of CASA deposits?

    CASA deposits are HSBC's cheapest funding. A persistent shift into time deposits would raise funding costs and dilute the benefit of higher asset yields, making DBS's mid-single-digit banking NII growth forecast for FY2027–28 too optimistic. The damage would land in earnings and buyback capacity, not in solvency.

Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.

Go beyond the preview

Continue reading the full report

  1. Key risksWhat could break the investment case
  2. Valuation and margin of safetyWhat the current price assumes, and how much room for error it leaves
  3. SummaryOur overall judgement and what to watch next

Free to sign up · 20 research points · unlocking this report uses 10

20 points cover 2 full reports. No card needed.

Research reflects its generation date. Preview the first two modules.

More on HSBC Holdings PLC ADR