Neither Silicon Valley Bank (SVB) nor Credit Suisse should have failed, thanks to all the measures put in place after the global crisis in 2008.
In the words of Mark Carney, then governor of the Bank of England and the head of the Financial Stability Board, "[o]ver the past decade, G20 financial reforms have fixed the fault lines that caused the global financial crisis" (Carney 2017).
Those post-2000 G20 financial reforms are founded on the philosophy of modern regulations. The notion is that all important risk is identified and measured, to be used by banks and the financial authorities to determine the appropriate level of risk. Then it is easy for the banks and regulators to fine-tune risk. If we need more growth, reduce capital requirements, as we did in March 2020, or demand more capital if risk is too high, as we should have done before 2008. Risk plays a key role in that because the amount of capital is a direct function of the riskiness of a bank. (Perotti 2023, Dewatripont et al. 2023)
The case of Credit Suisse and Silicon Valley Bank challenges the modern philosophy of regulations. While there are many factors at work, here we want to focus on the trilemma of financial policy:
- The economy should grow, or at least recessions must be avoided.
- Inflation needs to be close to its 2% target.
- Financial stability is to be high.
Financial regulations – both of the micro variety, including Basel III, and the macroprudential variety – play a key role in achieving these objectives. The problem is they cannot be achieved simultaneously. So the authorities need to pick the ones they prefer, hence the trilemma.
This conflict was not apparent in the decade after the 2008 global crisis because all three objectives were in sync. Financial policy helped growth via quantitative easing and low interest rates, inflation stayed close to its target, and financial stability appeared high. But that was an illusion.
Lax monetary policy, designed to help the economy grow, made the financial system dependent on low interest rates. Banks adapting their operations to the low interest rate environment was not seen as a problem because they would only face difficulty if rates were to rise. Consequently, a necessary condition for that monetary policy to be sensible is that inflation would never rise. It was a bet on low inflation and low interest rates lasting forever, akin to writing a deep out-of-the-money and decades-long maturity put option on inflation.
The longer monetary policy stayed lax, the more systemic risk increased, along with the growing dependence on money creation and the low rates. While not exactly hidden, these problems were hard for the private sector to find out because of the lack of granularity in accounting disclosures. However, the authorities had all the data and could identify the problems.
The ultimate consequence of the lax monetary policy was to undermine objective three, namely, financial stability. That should not have been a problem, since all the authorities had to do was rein in risk by raising capital. The problem is that increasing capital when the economy is doing poorly, as it is now, is recessionary. High financial stability conflicts with the first objective, namely, economic growth.
The modern philosophy of financial regulations put the financial authorities in an impossible position with no good options.
If we want the economy to grow to meet the first objective, we must disregard the other two. First, by keeping funding costs low, limiting interest rate rises, and fuelling inflation. And second, by keeping the cost of lending easily affordable, which means capital must be low and bank leverage high, fuelling systemic risk.
Fight inflation, and we end up with a recession because lending becomes too dear and financial stability drops because banks' capital position is eroded, hurting growth and even causing recession.
Prop up financial stability by raising capital levels, and lending becomes too expensive, growth is curtailed, and the economy is pushed into recession.
All of this was foreseeable and avoidable, especially since housing regulations and monetary policy in the same institution should facilitate the cooperation of these two policy domains, helping the architects of financial policy to identify the weaknesses in the policy framework.
While there are many reasons why the authorities disregarded the possibility that the three policy objectives could be in conflict, and politics certainly plays a significant role in that, the failure of the modern philosophy of regulations is a major cause.
The financial authorities face two key problems.
The first is that the financial system is, in effect, infinitely complex, and even if the authorities successfully identify a lot of risk and areas where it is taken, there is an infinite scope for risk to emerge elsewhere. There is no way to identify and manage all of that risk effectively. Doing so would make financial regulations so onerous that the banks would cease functioning as institutions that intermediate between savers and investors.
The second problem is that, in general, financial risk cannot be properly measured. A few years ago, we proposed the notion of the riskometer(Danielsson 2009) , a mythical device that, once plunged deep into the bowels of Wall Street, gives us an accurate measurement of risk. The problem is that the riskometer does not exist, as it is not possible to directly measure most financial risk. We can only infer it by the impact it leaves on the world, such as price fluctuations. To translate those fluctuations into risk, we need a model. Since there are an infinite number of candidate models, there are an infinite number of alternative measures of the same risk, many of which are equally plausible ex ante.
These two problems mean that the modern philosophy of financial regulations is not sound. It is based on the notion that the financial authorities and banks maintain virtuous feedback between risk identification and measurement to the amount of risk being taken. Just like the thermometers in the risk managers' office allow them to keep the temperature steady at 21C°. The necessary conditions for this virtuous feedback loop to exist are that the financial system is not infinitely complex so that the authorities can identify risk wherever it may happen, and for the riskometer not only to exist but to be as accurate as a thermometer.
Neither condition is met, as the case of Credit Suisse and Silicon Valley Bank makes clear. That is why regulatory policy after 2008 is a failure.
That begs the question of what we should do about it.
The most obvious and likely option will be to simply ramp up the existing regulatory framework, tighten the rules, and increase bank capital. While that might provide an immediate calming of market distress, it will also make financial intermediation more costly, reduce lending in the all-important SME sector, and even be recessionary. It is, at best, only a very short-term solution and will increase systemic risk in the longer term.
We can leave finance to the market, treating the banks like any other firm in the economy. That is not politically feasible because when the next crisis happens, the government will be under such popular pressure to act that it will have to step in, as has been demonstrated many times in history. Consequently, it is better for the authorities to be prepared for the eventuality, which means we cannot leave banks to the market.
Or we can change the regulatory framework, perhaps requiring banks to hold 100% reserves for demand deposits and maturity match assets to liabilities. While that would prevent Credit Suisee and SVB-type scenarios, it also would make financial intermediation very costly, and hence be highly recessionary.
We can look to technology. The system might then be founded on central bank digital currencies (CBDCs), created much more ambitiously than they are currently conceived of. We could all hold central bank-issued tokens of fiat money. That would ensure perfect liquidity, with banks akin to tech companies overseeing the decentralised financial system – what is known as Web5 and DeFi. This might be a fine solution, but it is very ambitious and will take decades to implement, and a key unresolved issue is that we do not want the central bank to be involved in lending decisions.
We propose two alternatives. The first is not to think about the problem of regulating the financial system from a risk-based buffer perspective, which current regulations do, but instead to approach it from a shock absorption point of view. When shocks happen, how best should they be absorbed? Current regulations make banks act as shock amplifiers because they harmonise beliefs via standardised risk-measurement techniques and action through mandated buffers. This means, in practice, that when shocks arrive, the system's institutions react to them in the same way, all buying or selling simultaneously, which is the basic mechanism behind the shock amplification.
The solution is to make the institutions of the financial system more heterogeneous or diverse. Then, when a shock comes along, some banks buy while others sell, in aggregate creating random noise. Achieving this is straightforward since it is just a matter of tweaking regulations. Instead of emphasising risk and buffers, encourage different business models. The micro regulators should be actively encouraged to embrace new entrants with different business models, even more than now. This would quickly increase the shock absorption capacity of the system with the additional benefit of providing cheaper and better tailored services to the banks' clients, helping with economic growth (see Danielsson 2022 for more details).
Furthermore, it is easy to address the moral hazard created by banks being limited liability corporations managed by people who get bonuses when things go well, while being protected from the downside. We cannot return to the pre-Victorian approach of unlimited liability for all because it would mean that banks could never get equity capital from outsiders. But there is no reason why we could not require senior bank management to face multiple liability and, in the case of CEOs, possibly to have unlimited liability. If senior management faced a really serious loss when their bank failed, there would be far less need for masses of restrictive regulations.
The problems we now see in the financial system, including the downfall of Silicon Valley Bank and Credit Suisse, have arisen because the financial authorities have been trying to do the impossible: maintain growth while keeping inflation under control and financial stability high.
The question, then, is how financial regulations should respond to the current market turmoil and rising long-term systemic risk. While there are viable solutions, such as curtailing moral hazard, increasing shock absorption, and new technology, we suspect the lessons learned will be different. The financial authorities will double down on current approaches, with more stringent regulations and higher capital levels that ultimately will hurt the economy and increase systemic risk.
Authors’ note: The authors were on the panel for an online conference “Silicon Valley Bank Failure: Implications for Risks to the Global Financial System” on 23 March 2023. Watch the discussion here.
Perotti, E (2023), “The Swiss authorities enforced a legitimate going concern conversion”, VoxEU.org, 22 March.
Danielsson, J (2009), “The myth of the riskometer”, VoxEU.org, 5 January. r
Danielsson, J (2022), “The illusion of control”, VoxEU.org, 7 November.
Dewatripont, M, P Praet and A Sapir (2023), “The Silicon Valley Bank collapse: Prudential regulation lessons for Europe and the world”, VoxEU.org, 20 March.