Inside Story

Profits and prices

Banks’ returns are almost back to the levels of the late boom years, writes Ian Rogers. But it’s the costs for customers, not profits, that we should be focusing on

Ian Rogers 27 October 2011 1559 words

Ever upwards: The Commonwealth Bank’s Sydney headquarters (above).
Photo: Neerav Bhatt/ Flickr



CONTROVERSIES over bank profits are never far away in Australia, and the profit reporting season for banks that kicks off today, with National Australia Bank the first to report, will no doubt refresh the arguments.

The lines will be pretty familiar: reports of profits for individual banks somewhere between $4 billion and $6 billion, reminders of the tactical repricing of variable rate home loans to increase margins eleven months ago, and more criticism of pressure on bank staff to sell more products (often credit cards and credit insurance) to maintain their take-home pay.

These are valid topics, but there is a broader issue that deserves more attention: the overall cost of banking to customers, of which bank profits are only one part. The information is patchy, but any sophisticated debate about banking will need to make some effort to engage with this dimension.

Since so many debates over banks start with profits, though, let’s start there. Australia’s big banks are not just big, they are massive, with combined annual revenue of $70 billion and aggregate annual profits currently around $17 billion. Based on their present value on the stock exchange, ANZ, Commonwealth, National Australia Bank and Westpac are four of the five largest companies in Australia. Together, they account for 90 per cent of profits of the banking industry as a whole, though they hold only 78 per cent of the industry’s $3 trillion in assets.

To come to grips with these dazzlingly large numbers we need to find a way of comparing them. We can do this by calculating the banks’ return on equity (in other words, their net profit divided by their net assets) or their return on capital (net profit divided by total assets).

Take the profit of just one bank as an example. In August, the Commonwealth Bank reported a net profit for the year to June 2011 of $6.4 billion. The return on equity was 18.4 per cent (if you use statutory net profit over the full year) or as high as 20 per cent (if you use the “cash” profit that management likes to talk about, and if you look only at the second half of the year). The bank’s return on assets in 2011 was 1 per cent.

How does that last figure compare with banks overseas? Readers might recall recent coverage of a report on bank profitability by the Bank for International Settlements. A BIS survey found that the profits of major banks in Australia were the highest in the world during 2010. The BIS uses pre-tax profits in working out return on asset measures, and put the pre-tax return on assets of major banks in Australia at 1.14 per cent.

The United States, where the banks’ return on assets in 2010 was 1.02 per cent, and Canada, where it was 1.01 per cent, were the only two other jurisdictions where profit on this measure exceeded 1 per cent. Among the thirteen countries surveyed, the median pre-tax return on assets for major banks was 0.66 per cent. (It’s worth remembering that for most banks in most places profits have not recovered to pre-crisis levels, though in Australia and North America they are not too far away.)

Back in Australia, the Australian Prudential Regulation Authority publishes data each quarter on bank profitability. Its figures suggest that the big banks, and the Commonwealth Bank in particular, have made profits at the high end of the scale over the past year. The data shows that returns on equity for major banks in Australia peaked at a shade over 20 per cent in late 2007 and early 2008, and were typically between 16 per cent and 19 per cent in the later years of the long boom. (The authority uses statutory net profit in its calculation, so its figures will be slightly lower than those cited by banks themselves for cash profits, and lower again than the pre-tax measures used by the BIS.)

In other words, while bank executives continue to talk about “challenging” conditions, the data suggests the opposite: bank returns are rising, and are nearing levels experienced in the later years of the boom.

The Australian Bankers Association acknowledges that the industry must continue to engage with the debate over bank profits. For instance, a “fact sheet” from the ABA circulated last week makes the point that “only two banks feature in the fifty most profitable listed companies in Australia.” Steven Münchenberg, chief executive of the ABA, asserts that “critics, who claim banks are too profitable or who call for a reduction in profits, don’t think through the implications of a weakened banking system, nor are they willing to outline what a safe level of profit should be.”

In fact, the industry’s most credible critics – financial regulators – have addressed this point. In a speech to a banking industry conference in December last year, Ken Henry, who was secretary of the Treasury at the time, asked: “Is a 15 per cent post-tax rate of return on equity too high or too low?” He went on to conclude that 15 per cent might be about right, but only so long as the industry was competitive and contestable. (Henry will soon join the board of National Australia Bank.)

The Reserve Bank of New Zealand has also taken an interest in the topic, writing in its Bulletin in June that “the return on equity in the New Zealand banking system appears to have been very high by OECD standards.” (The Bank for International Settlements did not include New Zealand banks in its recent survey.)

So: bank profits are high, rising, and reaching levels almost as high as they have ever been. But so what?

Debates over bank profits (and margins on particular products such as home loans and transaction fees) are certainly worth having. But it is the overall cost of banking to consumers and business – including interest margins and fees – that matters most.

Part of that story has to do with what it costs the banks to deliver their services. Tracking these costs is a thorny task, as is the related task of monitoring productivity levels, and the rate of productivity improvements, in banking. What follows is a survey of the higher-level data that critics of banks might like to engage with and perhaps drag to the forefront of the debate.

Banks themselves like to talk about the “expense ratio” or the “cost to income ratio,” a measure they seem to consider a useful way of thinking about their efficiency – though in reality it is just another measure of a bank’s profit margin rather than the achievements of a bank’s managers and staff in devising more effective ways of working.

Detailed productivity measures in banking are virtually impossible to find. A few consultants know the comparable industry data, and can tell you, for example, that the cost of originating a home loan varies from around $800 at the most efficient bank to more than $2000 at the least efficient. But which banks are more efficient and how this data is trending over time are hidden from public view. (Though, for what it’s worth, Westpac is a known high-cost originator of home loans, in spite of centralising processing in Adelaide more than fifteen years ago.)

To be fair, overall productivity levels in the broad sector have bounded along. Australian Bureau of Statistics data published in 2010 shows that over five years the financial services sector improved its productivity by 8 per cent, over ten years by 15 per cent and over 20 years by 56 per cent. In the wider economy, meanwhile, productivity declined by more than 3 per cent over five years, improved by only 2 per cent over ten years and improved by 19 per cent over twenty years.

An analysis of this data by the Grattan Institute suggests that the industry is the second most productive in Australia. The institute puts the gross value added per hour worked in financial and insurance services at around $170, or roughly three times the average for all industries in Australia. Only mining, with a figure of around $290, was more productive, according to the analysis.

So much for productivity. What about prices paid by users of banking services?

These are falling, at least as a proportion of household spending. The Australian Bankers’ Association (which these days is producing sober and accessible surveys on industry trends on its website) said that in 2010 service-fee revenue from transaction accounts made up 28 per cent of total bank service-fee revenue from households, down from 37 per cent in 2009 and from more than 40 per cent in the late 1990s.

The association’s analysis also drew on Australian Bureau of Statistics data to show that the proportion of household expenditure taken by bank fees in 2010 was 0.61 per cent – down from a ratio of around 0.75 per cent in the four years prior.

It would be nice to know a lot more about unit costs in many product lines in banking, and a lot more about margins in key product lines. (Though some are known: in home loans, for example, the return on capital exceeds 40 per cent.) Banks won’t readily disclose this data, but they might open up a little if the public debate on this industry took a turn for the better. •