VoxEU Column Financial Markets Global crisis Institutions and economics International Finance

Why supervisors should continue measuring financial risks – the fallacy of simple rules

Measuring financial risk is difficult. But what lessons, if any, have we learnt from the current crisis? This column argues against a move to leverage ratios and instead proposes continuing to measure financial risk (despite the difficulties), but with higher capital charges for banks. Focusing on the basics can only bolster central banks’ ability to fulfil their duties – looking for imbalances, and promptly tackling them with informed decisions.

Measuring financial risk is notoriously difficult. But no more difficult than measuring any other risk. Take, for example, the risk of flooding. In order to calculate the optimal height of a seawall one must take into account the regular variation of the sea level caused by the moon’s gravity, the shape of the shoreline, the probability of extreme events such as hurricanes and tsunamis, alongside myriad other factors. Such deliberations – and experience - have resulted in very different structures of seawalls, from high-rising concrete walls to porous mounds. If you asked financial regulators, they would say that seawalls are ‘risk-adjusted'. Higher walls offer more protection but are costlier, so society faces a trade-off. In the world of financial risk, the calibration of capital requirements attempts to strike a similar trade-off.1

The leverage ratio

In the wake of the financial crisis, some regulators argue that risk-adjusted capital requirements should be abandoned in favour of a much simpler rule – the leverage ratio. Various forms of leverage ratios have been used in the US and Canada, but the idea has gained new impetus since the publication of the Turner Review (2009). Although most proponents of the leverage ratio seem to favour a form of ‘backstop’ for risk-weighted rules, others now seem to propose an outright paradigm shift (Haldane 2012).

Proponents of leverage ratios argue that:

  • Financial risk and uncertainty are simply too obscure to capture, so our risk weights are bound to be incorrect.
  • Complex decision rules will then impede flexibility but without the additional benefit of precision.
  • Even if we get risk weights right, there are – unlike the causes of natural risks such as flooding – very clever people who make a living from bending the rules.
  • The lack of consistency of risk weights between firms and countries creates an unlevel playing field (see Demirgüç-Kunt et al. 2010).

The conclusion? We should apparently break the infinite loop of ‘rules-arbitrage-more-rules’ by resorting to a flat capital charge for all kinds of financial exposures.

The flawed logic of leverage ratios

Though the simplicity of a leverage ratio may seem appealing, the logic behind it is flawed. If we applied it to flood protection, seawalls across the globe would all have the exact same height. Such a prescription would mean walls that are too small for one place (error of type 1) and too high for another (error of type 2). Both errors end up costing society at large.2 Furthermore, no financial wizardry will be required to ‘arbitrage’ a leverage ratio; a bank can simply shift the composition of its loan book towards more speculative projects without having to increase its capital. Although leverage ratios may seem to provide a harmonised global rulebook, we will merely end up pretending that all shores are equally prone to flooding,

The benefits of measuring financial risk

Supervisors should continue to measure financial risks. However, we should impose much higher capital charges than today’s. In the flooding metaphor, this means making all walls higher than before. But it does not necessarily mean making them the same height. It is true that we will never get risk-weightings ‘correct’, nor will we avoid regulatory arbitrage. But through up-to-date measurement of financial risk and higher capital charges, we can make it much more likely that we err on the appropriate side of caution in the future. It is of course a colossal political failure that banks were allowed to operate with as little as 2% core capital as a proportion of risk-adjusted assets before the current crisis.


The beauty of a non-risk-adjusted capital measure lies in its simplicity. It may enable both regulators and investors to quickly detect emerging risks, ‘to see the wood for the trees’. This also holds for many other statistics, such as the growth of a bank’s balance sheet, households’ indebtedness, or the banking system’s exposure to the property market. Thus, leverage ratios should be implemented as monitoring tools - so-called Pillar 2 measures as opposed to Pillar 1. The Basel III proposal has the equivocal formulation that a 3% leverage ratio shall be implemented over a transition period “with a view to migrating to a Pillar 1 treatment on 1 January 2018 based on appropriate review and calibration”.(Basel III 2011).

The duty of a central bank

The first duty of a central bank and its supervisors is to distill masses of economic information in order to detect underlying imbalances. The second is to promptly take informed, decisive action. Clearly, the financial crisis was exacerbated by the institutional failure to carry out these dutie, rather than the technicalities of the methods of calibrating capital requirements.


Basel III (2011), "A global regulatory framework for more resilient banks and banking systems", Bank for International Settlements, originally published December 2010, revised in June.

Demirgüç-Kunt, A, E Detragiache and O Merrouche (2010), "Bank Capital: Lessons from the Financial Crisis", World Bank Working Paper Series ,5473.

Haldane, A (2012), "The dog and the frisbee", speech, Federal Reserve Bank of Kansas City’s 36th Economic Policy Symposium, The Changing Policy Landscape, 31 Aug.

Turner Review (2009), "A regulatory response to the global banking crisis", FSA, March.

Modigliani, F and M Miller (1958), "The Cost of Capital, Corporation Finance and the Theory of Investment", American Economic Review, 48(3), 261–297.

1 Notwithstanding, of course, that the cost of equity rather than debt finance often is largely tax-induced, and thus small for society. But even in environments where the premises of the Modigliani-Miller theorem (1958) are close to hold there is little reason to abandon risk-weighted rules -- though indeed reason to increase capital levels.


2 As a minimum ‘backstop’, the two errors are likely to reversed, but still costly for society. The prescribed wall will be way too small for most places, rendering the rule meaningless, and too high for certain places.

2,730 Reads