Australia’s biggest financial institutions are in the spotlight, with ANZ Group Holdings (ASX:ANZ), Westpac (ASX:WBC), and National Australia Bank (ASX:NAB) posting half-year results, and Commonwealth Bank (ASX:CBA) delivering its third-quarter trading update. As lenders to all corners of the economy, the major banks give the best read-through on the nation’s financial health. Anyone investing in Australian equities can learn something from their results.

Here are our key takeaways.

Net interest margins stabilising at lower levels

Net interest margin, or NIM, is the difference between the rate at which banks borrow and lend. They borrow via deposits and wholesale markets and lend via home and business loans. By taking on this maturity mismatch, they earn a spread. The size of this spread is influenced by many factors, but the big ones are how aggressively banks need to set rates to win deposits and write loans.

Margins have been under pressure in recent years, squeezed by rising funding costs and intensifying competition. The expiry of the low-cost term funding facility, combined with customers switching from transaction to higher-rate savings and term deposits, has lifted interest expenses. Meanwhile, the growing dominance of mortgage brokers, who now account for 76% of new home loans, up from 52% in early 2020, is also eroding profitability, as banks pay commissions on broker-originated loans.

But margins appear to be finding a floor. ANZ’s NIM slipped just 2 basis points from the second half of fiscal 2024 to 1.56%, while Westpac’s fell 4 basis points to 1.92%. Meanwhile, NAB and CBA held margins steady. For CBA, we think its strategy of writing more loans directly is protecting margins, with just 32% of loans written via brokers in the March quarter, down from 46% in the December half. By comparison, 67% of Westpac’s loans were broker-originated in the six months to March.

Looking ahead, we expect NIMs to lift gradually as deposit competition eases, with non-major banks likely to scale back aggressive pricing to improve returns on equity. Falling cash rates should also see banks reduce discounts on new loans and offer less generous rates on deposits. But we expect only modest improvement from here, not a return to the margins the banks enjoyed before COVID.

Heavy, but necessary, investment cycle underway

The major banks are deep into multi-year investment programs focused on lifting efficiency, enhancing customer experience, and strengthening risk management. Execution risks look highest at Westpac and ANZ, given the scale and complexity of their respective transformation programs—Unite at Westpac and ANZ Plus at ANZ. But over time, we expect these investments to deliver meaningful benefits. Across the sector, we see scope for a material reduction in cost-to-income ratios as the banks simplify operations, digitise processes, and remove duplication.

Bad debts contained

Bad debts remain low across the banks, reflecting resilient household and business balance sheets. Home loan arrears are generally drifting higher across the majors—unsurprising given the higher interest rate environment—though this is off very low levels and remains well contained.

We think the banks are well provisioned for any near-term deterioration. Stress doesn’t appear widespread, though we expect arrears to edge up as tight monetary policy and lingering cost-of-living pressures weigh on the household sector.

CBA’s ROE remains well ahead of peers

If there’s one financial metric that matters most to bank investors, it’s probably return on equity. At the end of the day, the attractiveness of a bank—like any business—comes down to the returns it can generate on your capital.

We estimate the cost of equity for the major banks at 9%, reflecting a 4.5% risk-free rate and a 4.5% equity risk premium. A good bank should be earning at least this on new investments. Above it, the bank is creating value, and below it, destroying value.

So, how are the banks faring? In the first half, NAB generated an ROE of about 11%, and ANZ and Westpac about 10%. In each case, a fairly thin margin over our estimated cost of equity. CBA’s ROE for the six months ended December 2024 was almost 14%, comfortably above peers [Exhibit 1].

Exhibit 1: CBA better than peers, but the gap isn’t widening

Trailing twelve-month return on equity (smoothed)

Source: Company filings, Morningstar. Note: the big decline in major bank ROE in 2016 is a result of NAB’s Clydesdale impairment, while the 2021 reduction is mostly due to Westpac’s AUSTRAC fine.

CBA’s valuation looks stretched—better value elsewhere

All else equal, you’d pay more for a business generating higher returns on equity. But even the best business has a fair price, and CBA has long since surpassed it.

We know this might sound like a broken record. We’ve questioned CBA’s valuation before, and yet the shares keep climbing. Its latest result met expectations, offering little to unsettle investors. But what the market is willing to pay for those earnings remains, in our view, unjustifiable.

CBA now trades on 26 times forward earnings, with a dividend yield of just 3%. You can throw almost any bull case into a valuation model and still struggle to get to $170 per share.

It’s a high-quality franchise, no question. Its lower-cost deposit base, operating efficiency, and conservative underwriting relative to peers support superior returns, and justify a premium. But if we step back and look at ROE trends over the past decade, CBA hasn’t meaningfully pulled away from the other majors, yet investors are paying a much higher multiple than ever before, even as returns have drifted lower in line with the rest of the sector [Exhibit 2].

Exhibit 2: CBA’s huge valuation gap to peers

Trailing twelve-month price/book

Source: Company filings, Morningstar.

We understand investors might see it as the first port of call for Australian bank exposure, or even for general Australian market exposure. But let’s keep some perspective. The earnings outlook is not particularly inspiring. On our forecasts, CBA’s earnings are forecast to grow at an average of 5.5% during the next five years, a touch above nominal GDP. And if the banks are effectively a proxy for the Australian market, which has historically traded on about 15 times earnings, you have to ask why CBA—Australia’s biggest company—should trade on 26 times.

CBA has looked expensive for some time—and it has continued to defy gravity. We see more compelling opportunities elsewhere. Amongst the major banks, ANZ looks the most attractive, and we make the case in our earnings note on page 7.