What bank capital regulation works best for the economy?
02 November 2018
By Andrea Gamba
Worries about the ability of firms to roll over debt, a rising demand for cash, accompanied by the difficulties of non-banking finance companies in raising funds as banks hoard money. In short, an escalating liquidity crisis threatening the stability of the banking system. But this is not the 2007-08 financial crisis, this is the Indian economy in October 2018.
It is a stark reminder of the fragility of global financial markets, an echo of the last great financial crisis, and a situation that policymakers in Europe and North America have sought to avoid through the introduction of a raft of banking regulations. The 2007-08 financial crisis prompted policymakers and regulators to revisit the rule book and determine, given the apparent inadequacy of the existing Basel Accords regulatory framework, what action might be taken to better regulate banks and the banking system.
The policymakers and regulators were assisted by the academic community and the publication of numerous papers focusing on the causes of the financial crisis and possible remedies. The result was an addition to the Basel Accords regulatory framework developed under the auspices of the Bank for International Settlements (BIS). Moving on from inadequacies exposed in Basel II, Basel III is due to be fully implemented by 2019.
Among those academic interventions in the aftermath of the financial crisis was a paper I produced in 2012 and published in 2014, together with colleagues from the IMF and the Ca' Foscari University of Venice. That paper adopted a microprudential, firm level, view of banking regulatory issues. In particular we considered the impact of three regulatory provisions on two measures of bank efficiency and welfare.
The regulatory measures were: capital requirements - the amount of capital a bank has to hold, usually expressed as a capital adequacy ratio (being mainly common stock and retained earnings as a percentage of risk-weighted assets); liquidity requirements - a requirement for banks to hold sufficient high-quality liquid assets to cover total net cash outflows over 30 days; and prompt corrective action (PCA) - regulatory provisions that force banks to sell assets, restrict payouts or even close, subject to levels of capital.
The two metrics were: enterprise value - the efficiency with which the bank is able to fulfil its maturity transformation role using its debt, in the form of customers' short-term deposits to provide long-term loans offered to companies in the productive sector; and social welfare - the contribution to the overall value to society of banking activities.
After all, it is one thing to give in to the widespread calls for tough regulations in the wake of the crisis in order to protect banks and investors in anticipation of an economic shock. But what if, in imposing the regulatory measures, you prevent the banking system from functioning effectively, providing credit to the productive sector, and helping to drive economic growth post-shock?
Our research showed, that when seen in a dynamic context, prompt corrective action is the best approach because it is a regulatory response that depends on the state of the bank. You only intervene when the bank is in trouble, rather than in advance, in anticipation of the bank getting into difficulty, and certainly not when the bank is doing well. Whereas measures that are not contingent on the state of the bank, and that combine capital and liquidity requirements, may be detrimental to the bank.
Our findings were well received by the academic community at the time, including regulators, however there were limitations to our work. We were taking a microprudential approach and looking at each bank in isolation. So with colleagues from New York University and the Ca' Foscari University of Venice we decided to expand our perspective and talk about the banking sector's risk rather than an individual bank’s risk. In order to examine the regulatory challenge from the perspective of the banking sector we devised a general equilibrium dynamic model with aggregate shocks that represents the whole economy in its upturns and downturns.
One of the primary elements of Basel III (and the Basel Accords generally) has been the need to increase capital requirements reducing bank debt. In our new analysis, we consider capital requirements in a world in which banks have another important role besides making credit: they provide liquidity to consumers.
As mentioned previously, banks have a maturity transformation role. They intermediate between households and the productive sector of the economy, taking money from consumers, through short-term borrowing, and using that money to create long-term loans.
Hence, banks create liquidity for households and consumers by selling deposits to them, and in doing so, they make sure that the money is available to consumers at any moment in the future to be withdrawn and spent on consumption. This is a liquidity service and banks create money in the economic system by allowing consumers to carry forward wealth into the future.
In an upturn, when times are good, banks have an incentive to make more loans and therefore take on more debt in the form of deposits. However, the risk is that banks over extend and the economic situation changes, leaving them exposed to large liabilities to households and to increased delinquencies in their loan investment. Individual banks are, understandably, focused on delivering returns for their shareholders. They have the incentive to act to further the interests of those shareholders by pursuing high returns at significant risk, knowing that any losses due to failure of the bank and disruption of the banking system are likely to be distributed much more widely.
Because a banking crisis has large economic and social costs, rather than allowing a completely laissez-faire approach to banking, governments seek to prevent banks from running into trouble and harming themselves and the economy through regulation. One popular measure, as with the Basel Accords, is to apply capital requirements. The challenge is knowing what the optimum approach is for setting capital restrictions given the special role of bank deposits. And invariably there is pressure to increase the levels of capital restriction.
This is usually when the banking lobby protests, arguing that capital requirements should not be raised too high, because otherwise it would be impossible to run the banks. Their usual argument is that when regulators impose higher capital requirements it increases the cost of capital for banks, because they must rely more on equity capital, which requires a higher return than debt. That increased cost is then passed on to borrowers, who will have to pay more for their loans, and so it is bad for the economy. Interestingly, our research shows that there is some merit in the protest of the banking lobby but not for the reasons they suggest.
In terms of how capital requirements are applied, they could be applied using a constant capital ratio rule; the same for all banks and for all states of the economy. Alternately they could be state contingent, that is depending on the state of the economy.
Basel III introduces the notion of a "discretionary counter-cyclical buffer" as part of the capital requirements. This provides for national regulators to demand the bank to hold additional capital during periods of high credit growth - rapid credit growth is often followed by banking crises.
However, our research suggests that this approach may not be optimal for the broader economy, because of the role of banks in creating liquidity. Imposing conditions that make leverage counter-cyclical reduces deposits, thus increasing their price and therefore reducing their return. In this sense, bank capital becomes more expensive relative to debt. The overall result is a reduction of consumption and productive investment, reducing wealth and welfare in general.
In detail, during an expansion consumers want to consume more. Therefore, they tend to use deposits more in order to fund their short-term consumption. If capital requirements become more stringent in this phase, the banks are constrained regarding the extent to which they can use deposits in order to take on more debts. Thus, there is a high demand from consumers for deposits but a restricted supply. Consequently consumers are willing to accept very little return, if any, on their deposits. To the extent that effectively there may well be a negative return - consumers pay in real terms to have their deposit accounts because they need the liquidity.
To satisfy their funding needs, banks have two sources - equity and debt. As the banks do not have the option of exceeding the regulatory constraints placed upon them on debt, they issue equity - raise capital from the markets - and this capital is very expensive relative to debt. Thus, the regulatory framework increases the cost of capital for the bank. This is the direct result of what, it might easily be argued, is over restrictive regulation.
It seems only natural that there should be an expansion of leverage allowed in good times. We argue that it is beneficial for the economy to allow the leverage to be pro-cyclical. In our model, the best regulatory approach would be one that allows the bank to expand leverage during periods of economic prosperity, just ensuring that there is a sufficient degree of resistance to restrain the banks' inclination to further increase leverage substantially. The capital requirement would be reduced steadily, allowing leverage to increase. Although never to the extent that leverage might expand if it were a completely unfettered market, without regulation.
And there is no need for the regulator to resist the expansion of leverage at all when the economy enters a downturn. That is because the economic slowdown imposes a natural brake on a bank's desire to increase its leverage. Overall, optimal bank leverage is pro-cyclical and regulation is counter-cyclical, in the sense of being more restrictive in economic upturns.
In the alternative regulatory approach of a constant capital ratio, regardless of upturns and downturns, there would be less debt, fewer deposits, and less consumption in upturns. So eventually the economy will be adversely affected due to a reduction in investment. This situation will also create a bigger differential between the cost of debt and cost of equity for the banks. So in a way our research confirms the banking lobby's argument that bank capital becomes expensive. Yet it is not, as the banks suggest, because of the level of capital requirement – but rather when the regulations imposed are non-cyclical.
The role of the regulator
One of the implications of our analysis concerns the role of the bank regulator and more specifically what can be reasonably expected from a regulator. This is an issue that has not been fully discussed in the debate on bank regulation, but which is central to it. Because only once we are clear about what can be expected of a regulator can we actually decide what the regulator should do.
Bank regulators are not like the central planner figure from economic theory. All they have are a relatively small set of regulatory levers that essentially affect one aspect of the economy - how banks operate. It seems unrealistic to expect to maximise the overall welfare to the economy by just regulating bank leverage or bank liquidity, for example. That is a very narrow perspective, after all. If we are thinking of the benefit to the broader economy, perhaps there should be full integration between monetary policy and bank regulation, as the two are integrated ways of creating liquidity in the economy, as our analysis shows.
If a government is to successfully relaunch the economy in the aftermath of a crisis, it makes sense for the arbiters of monetary policy - often the central banks - to work closely alongside bank regulators. It should not be the role of the bank regulators alone as they possess insufficient levers to drive economic growth at a macroeconomic level. It requires additional measures, such as the quantitative easing deployed after the recent crisis, beyond the scope and tools of what a bank regulator can do. In other words, bank regulation should be integrated into the broader monetary policy that central bankers control.
Although such an approach may be some way off, it merits serious discussion. In the meantime, our research shows that regulators can achieve something that goes a long way towards optimal - if the degree of leverage allowed is naturally pro-cyclical - exactly because this allows a monetary expansion, through bank leverage, in upturns. Capital restrictions may be counter-cyclical, in the sense of ‘leaning against the wind’ and making the bank leverage less pro-cyclical than in the corresponding unregulated approach, as the economy expands. But an entirely counter-cyclical approach to leverage - as currently suggested in Basel III - that would be mistake.
Gamba A., G. De Nicolò, and M. Lucchetta. (2014) "Microprudential Regulation in a Dynamic Model of Banking", Review of Financial Studies, Vol 27(7), July 2014, 2097–2138.
Gamba A., D. Gale, and M. Lucchetta, (2017) "Dynamic Bank Capital Regulation in Equilibrium", SSRN/FEN.
For more articles like this download Core magazine here.