ANZ's Andrew Cornell puts the case for bank leverage, says banks without risk don’t serve the purpose as 'handmaidens' of the economy, no matter how safe they are

By Andrew Cornell*

It’s indisputable the level of core capital banks must carry is rising higher and higher; nor is it in dispute capital was too low when the financial crisis hit.

Indeed, some prominent authorities, including former Bank of England governor Mervyn King and Financial Times columnist Martin Wolf, believe in even higher levels of capital.

The implicit argument is more capital – real money such as equity – reduces the leverage in a bank. The explicit counter argument is this may cost more than its worth in terms of financing economic growth and profitability of banks.

Banks with more capital have more buffers against losses and are safer, albeit less profitable. However, in being safer they should actually be more profitable in the long run and hence those investors funding banks, with either debt or equity, should be prepared to accept lower returns (for the lower risk).

There’s a core economic theory behind this known as the Modigliani-Miller hypothesis (“The Cost of Capital, Corporation Finance and the Theory of Investment”, 1958) which specifically argues the cost of equity capital will be lower if banks reduce leverage and risk. And indeed, since the crisis the M-M hypothesis has underpinned much debate.

The banking industry doesn’t necessarily agree with the theory – or at least its applicability in the real world. Speaking on behalf of the industry, the Institute of International Finance (IIF) has long argued the benefits don’t warrant the costs.

The IIF notes in a new paper the wave of regulation, known as Basel III, requires more capital and is morphing into Basel IV with the addition of minimum liquidity coverage ratios (to insure against bank runs), net stable funding ratios, leverage ratios and TLAC (Total Loss Absorbing Capacity).

“The industry - and many leading regulators - consider that this new push beyond the broadly justified increases already achieved could go too far, leading to a shrinkage of bank balance sheets -including by reducing lending to businesses and consumers,” the paper says.

The paper, “Bank Capital: Does Modigliani-Miller Have It Right?” by Hung Tran, Sonja Gibbs and Khadija Mahmood concludes “increasing capital and related requirements beyond some point - which many believe has been reached - will do more to reduce the volume and increase the cost of lending than to increase stability in the system”.

Straw men

From the regulatory perspective however – and far be it from me to speak for regulators – M&M are straw men. Or at least they are for the purposes of the banking industry’s argument against higher levels of capital (according to regulators).

Indeed, they would argue the global regulatory response has specifically not relied upon M-M. For example in the Basel Committee on Banking Supervision’s core paper “An Assessment of the long-term economic impact of stronger capital and liquidity requirements”.

That analysis actually assumes M-M was wrong, that the cost of debt and equity was not affected by bank riskiness (see page 22 of the above document). The study did assume the full cost of regulatory imposts would be passed on to borrowers.

Despite those higher costs being passed on in full, the BCBS found the net benefits of more capital were clearly positive – “a conclusion that was similarly reached in subsequent work by the International Monetary Fund, Organisation for Economic Cooperation and Development and a number of academics. That makes the extent to which there may be some (partial) M-M benefit a moot point”, as one Basel insider notes.

Others are happy to be more emphatic: “there are reasons to think the cost of capital might decline as a bank becomes less risky. I would agree with that view - to suggest that debt holders are indifferent to the level of bank leverage when setting risk premiums seems somewhat unrealistic”.

But you’ll find plenty of bankers in treasury departments who believe investors are indifferent and the IIF does an empirical analysis in its latest paper.

“A fact check of the Modigliani-Miller hypothesis against developments from 2000 to the present, looking at 2000-2007 and 2010-2016 (skipping the crisis years 2008-2009) shows that Modigliani-Miller cannot on its own support arguments for new capital requirements,” the authors write.

“Our analysis suggests that for banks in the US and Europe, Core Tier 1 (CT1) capital ratios increased from less than 8.5 per cent in 2000-07 to 12.5 per cent-13.5 per cent in 2010-2016 - more than meeting Basel III minimum requirements. Japanese banks have also increased their CT1 capital ratios from below 7 per cent to over 12.5 per cent.

“However, despite the significant increase in CT1 capital, the cost of equity capital has risen noticeably in all three regions, from 9 per cent-10 per cent to around 14 per cent. This is directly contrary to the Modigliani-Miller prediction.”

All things equal

A regulator however would counter thus:  if the outlook for bank earnings deteriorates (margin compression, weak macroeconomic conditions, subdued growth, poor outlook, whatever), then the underlying risk profile rises and may well offset/exceed the impact of reduced leverage.

But the BCBS analysis never argued the cost of equity would decline - simply that actual returns would probably decline due to lower levels of leverage (a point on which banks and regulators agree).

And we have not been in what economists like to call a ceteris paribus world – all things being equal.

Actually, the IIF paper argument is not in contradiction: “Particularly important in explaining the rise in bank cost of equity has been the sharp decline in bank profitability, as reflected in the drop in return on equity from around 15 per cent to 8 per cent for US banks and around 4 per cent for European banks; Japanese banks also saw a decline in ROE, from over 11 per cent to around 8.5 per cent.

“This poor profitability, coupled with earnings uncertainty, has greatly depressed bank valuations: price to book ratios have fallen to well below 1x across the board - pushing up the cost of equity.  There is also evidence that investors will not accept utility-like returns from banks given their central role in the economy, which implies an inherent cyclicality and volatility in earnings.”

Nevertheless, overall bank funding costs have fallen materially in the US, Europe and Japan – because interest rates have plunged. Have the Basel II and IV reforms amplified this reduction (because banks are safer) or reduced them (because bank profitability is lower)?

Ultimately, regulators don’t care about bank shareholders as long as banks don’t fail and play their central role in financing economies. The BCBS continues to argue global banking systems are now better placed to do this – and it is up to monetary policy to stimulate macro-economically.

Nevertheless, the debate is crucial. It is one of costs and benefits. How safe is too safe? Banks without risk don’t serve the purpose as 'handmaidens' of the economy, no matter how safe they are.

*Andrew Cornell is managing editor of ANZ's BlueNotes. This article first appeared on the BlueNotes website here, and is used with permission.