Have banks boosted their return on assets, or has profitability risen purely due to increased leverage? More than 20 years of data from The Banker can help provide an insight. Writer Philip Alexander, Data research Guillaume Hingel

For almost half The Banker's 85-year history, its bank rankings have provided the most comprehensive data snapshot of the global banking industry. We are in the process of digitising back-data from the early years of the Top 1000 rankings, and we can already track and analyse continuous historic annual data for a number of key bank indicators going back to 1988. This was the period when financial policy-makers began to rethink bank regulation in a global context, following the damage to the US and European financial sectors inflicted by the 1982 Latin American debt crisis. The resulting discussions led to the original Basel accords governing bank capital adequacy.

More than 20 years later, in the wake of another financial crisis that has hit major banks in the US and Europe even harder, Basel is undergoing its third draft. The unresolved question is whether bank capital regulation has adequately captured and controlled the level of risks banks take. The events of the past two years have forced a reconsideration of whether bank profitability derived from successful lending decisions that allowed good return on assets without taking excessive risks, or simply involved boosting return on equity by increasing balance sheet leverage indiscriminately.

A stroll through The Banker's database can provide significant insight into how banks have made their money, whether those which showed high returns during the boom years relied too much on leverage, and if those that came through the crisis in better shape were taking fewer risks. It can also provide a taste of the changing shape of global banking, as returns on assets vary by region over time. To study these trends, we have looked at a sample consisting of banks in continuous existence over the period, based in major banking markets and drawn from among the top 100 in our rankings.

Contrasting Basel responses

In 1982, in the wake of the Latin debt crisis and the collapse of domestic savings and loan institutions, US banks were heavily undercapitalised, but regulators allowed a number of banks to continue operating even when technically insolvent. With the Brady Plan resolving the Latin debt defaults and the first Basel accord establishing international standards for capital adequacy, the average Tier 1 capital-to-assets ratio for US banks rose significantly in the early 1990s.

It remained higher than levels in Europe throughout the 1990s, as the US banks focused more on tangible common equity while the European market for hybrid capital expanded. The Banker only started tracking the break-down of capital between Tier 1 and Tier 2 in recent years, but we hope to be able to back-fill this data to provide a more detailed study in a later edition. The essential message seems to be that on-balance-sheet leverage was not necessarily a factor that weakened US banks - the off-balance-sheet securitisations and conduits that do not appear in the capital-to-assets figures were the prime culprits in many cases.

Throughout the period, the capitalisation of German banks remained low, dragged down primarily by the four landesbanken in the sample. Part of the reason why these banks failed to build their capital base is likely to be their low profitability, with the average return on assets for Germany consistently worse than any other major banking market in the past 20 years, never surpassing 1%. Heavy competition for funding and for corporate clients in a market where numerous landesbanken backed by municipal governments compete with the private sector appears to have a consistently negative effect on bank margins.

The unresolved question is whether bank capital regulation has adequately... controlled the level of risks banks take

Of the other countries most affected by the financial crisis, the UK on average retained relatively high levels of capitalisation until 2005, when leverage began to accelerate. But the picture is not uniform. Many large financial institutions, including Lloyds Banking Group and leading mutuals such as The Co-operative Bank, retained relatively high capitalisation throughout the period. But RBS, Northern Rock and Barclays all became much more aggressive from 2005 - the capital-to-assets ratio for Northern Rock was slashed from 4.7% in 2004 to 2.5% in 2005, while that for Barclays fell from 3.2% to 2% in the same year. The ratio for RBS was higher until the ABN Amro takeover in 2007, which pulled it down from 3.4% to 2.3%.

Risk-weighted assets

A further factor behind the low headline capital-to-assets ratios in Europe appears to be the more intensive approach to capital management on the asset side. Certain European banks seem to have made far more use of the process of risk-weighting their assets to calculate the overall capital adequacy ratio than their US counterparts. Basel II, which allows for much greater differentiation in risk-weighting by using internal modelling, was adopted sooner in Europe than in the US. As a result, it is possible to maintain a relatively high capital adequacy ratio under Bank for International Settlements definitions, even while the pure capital-to-assets ratio falls.

The major Swiss banks are the most striking example of this process: the average Tier 1 capital-to-assets ratio declined from one of the highest in our sample in 1989 to the lowest at the time the financial crisis arrived in 2007. Yet for the period since 2000, when The Banker began tracking risk-weighted assets, the proportion of risk-weighted to total assets for Credit Suisse and UBS has been dramatically lower than those for most other major banking groups. In recent years, this ratio dipped below 20%. In other countries, the ratio for Deutsche Bank is also markedly low.

This preponderance of assets with a low risk-weight allows a bank to leverage up without technically undermining its solvency as measured by Basel II. However, concerns about the ability to identify risk correctly in weighting the assets was one of the key factors that prompted a rethink of the Basel rules in the wake of the subprime crisis. Securitised subprime mortgages were often rated AAA but carried a yield pick-up over AAA government or corporate bonds. In reality, correlation between mortgage defaults in the securitised pool was much higher than the AAA rating and low risk-weighting had implied.

Bank of England governor Mervyn King noted in a speech in New York in October 2010: "Those risk-weights are computed from past experience. Yet the circumstances in which capital needs to be available to absorb potential losses are precisely those when earlier judgements about the risk of different assets and their correlation are shown to be wrong. One might well say that a financial crisis occurs when the Basel risk-weights turn out to be poor estimates of underlying risk, because the relevant risks are often impossible to assess in terms of fixed probabilities. Events can take place that we could not have envisaged, let alone to which we could attach probabilities."

Another factor may also allow banks to record low totals of risk-weighted assets while leveraging up on capital. For the banks where risk-weighted assets are lowest, the size of investment banking operations is significant, as demonstrated by relatively high proportions of non-interest income. Consequently, these banks have large derivatives trading operations, and assets in these divisions are more likely to be classified under the market risk category of Basel II, rather than credit risk. The bank will then net out all its derivative positions that effectively match each other - the process known as alternative risk transfer.

As long as the positions genuinely match, and the bank has managed to find clients to take the risks associated with both sides of the trade, then the market risk for the bank should be much lower than the asset size would imply. Once again, however, the Basel III proposals may change the mathematics of capital management for derivatives activity by introducing greater scrutiny and risk-weighting for the counterparty credit risk inherent in derivative transactions. All in all, those banks with very low ratios of risk-weighted to total assets may face greater pressure to deleverage in preparation for Basel III.

how banks make their money figure 1 and 2

Emerging market model

In emerging markets, capital adequacy levels were far higher in the late 1980s as a result of low lending penetration and poor credit quality. In China, the average ratio was more than 7%, and in default-racked Brazil, over 13%.

In both cases, those ratios descended as the 1990s progressed and economic transition began to take effect. In the case of China, that decline was interrupted by the Asia crisis in 1997, as lending opportunities grew scarce again and capitalisation was boosted to provide a cushion against the increase in risk.

What is also noticeable is that as risk diminished and returns became more predictable in emerging markets in the past decade, banks appear to have pursued genuinely counter-cyclical capital management, preparing during the good times for the inevitable turn in the cycle. So Brazilian and Chinese banks were in a stronger capital position than many developed market counterparts when the financial crisis began in 2007, because they had boosted capital during the boom years since 2003, presumably by retaining part of their growing profits.

And how those profits have grown. Our data provides a striking graphic of the recent rise of emerging markets. From the return on assets of less than 1% in China, and the wild swings between profit and loss in Brazil, both these markets have become top performers since 2007 - and even earlier in the decade in the case of Brazil.

how banks make their money figure 3 and 4

Fee income decline

As for developed markets, aside from Germany's consistently dismal performance, the consistent excellence of the US until 2007 is also remarkable. Was it a mirage? In some ways less so than the performance of European banks. While some banks in Switzerland, the UK or Germany offset low return on assets by ramping up leverage to increase their return on equity, the US banks generated their performance with lower leverage, but higher return on assets.

The greater available balance sheet also shows itself in the proportion of interest income - higher for the big US universal banks than for many peers on the other side of the Atlantic. Inevitably, interest income is also significant for banks in the high-growth, high-inflation context of emerging markets such as Brazil. In the present environment, it is hard to know whether the ability to generate greater revenues from fee-earning activities such as investment banking and private banking is an advantage or a disadvantage.

Credit risk is higher, and both capital and funding are scarce, making pure balance sheet growth a difficult strategy. But at the same time, investment bank revenues are under pressure from volatile markets, and leading private banks are still struggling to restore trust after failing to protect their clients' capital during the crisis. Moreover, the relative scarcity of funds has driven loan margins higher.

The consistent excellence of the US until 2007 is remarkable. Was it a mirage?

Almost across the board, banks are generating more of their income from lending today than they were in 2000. Bank of America is one striking exception - the benefits of the Merrill Lynch acquisition in terms of diversifying its revenue streams are already evident.

Perhaps, however, the banks with the most traditional models are in the best position, provided they are in the right geographic markets. Like the banks headquartered in emerging markets, European lenders such as UniCredit or Santander had small investment banking operations and fairly high capitalisation going into the crisis, with a more conservative approach to the risk-weighting of their assets. They generated higher returns by bringing best-practice plain vanilla banking to emerging markets - eastern Europe in the case of UniCredit and Latin America for Santander. Their limited exposure to capital markets was an advantage during the crisis. And as both continue to earn a high proportion of their revenues from interest on lending, they should have more room to increase the share of fee income in the mix. The Banker will be keeping a close eye on how that ratio develops in the coming years.

PLEASE ENTER YOUR DETAILS TO WATCH THIS VIDEO

All fields are mandatory

The Banker is a service from the Financial Times. The Financial Times Ltd takes your privacy seriously.

Choose how you want us to contact you.

Invites and Offers from The Banker

Receive exclusive personalised event invitations, carefully curated offers and promotions from The Banker



For more information about how we use your data, please refer to our privacy and cookie policies.

Terms and conditions

Join our community

The Banker on Twitter