The structure of regulatory revolutions

Austen Saunders and Rajan Patel

What can the history and philosophy of science teach us about regulatory reform? In this post, we borrow Thomas Kuhn’s idea of ‘scientific revolutions’ to argue that radical overhauls of regulation often occur after crises but that, once major reforms have been completed, it’s normal to have periods when rules do not change so much. For instance, major reforms made to banking regulations after the Global Financial Crisis of 2007–08 are now coming to an end with future change likely to be more incremental. This post is about why different circumstances call for these different approaches to regulatory change.

Dialectic and revolution

Some amount of regulatory change is always necessary because rules and risks are constantly reshaping each other. This is what Edward Kane described as the ‘regulatory dialectic’ – a two-way process by which risk-taking changes in response to regulation while, at the same time, regulation changes in response to new forms of risk-taking. Most of the time, targeted changes to rules are enough for them to keep up with the risks they address. These incremental changes are the normal business of ‘dynamic regulation’ which adjusts existing frameworks to keep them fit for purpose. But after a crisis, that’s not always enough. Then, a more radical overhaul might be needed to replace one framework with another.

We can understand the circumstances in which these different types of change are appropriate by borrowing from Thomas Kuhn’s theory of scientific paradigms and revolutions.

In ‘The structure of scientific revolutions’, Kuhn argues that scientific research takes place within what he calls ‘paradigms’. A paradigm provides a scientist with a body of accepted knowledge, assumptions about how to frame research questions, and standards for applying experimental procedures. Most work done by scientists takes place within an established paradigm and is what Kuhn calls ‘normal science’.

Regulators have their paradigms too – what we call ‘frameworks’ or ‘regimes’. Most of the time, supervisors of banks subject to the Basel standards or of insurers subject to Solvency II (to give a couple of examples), can take the regime they work with as a given. Their day job (what we might call ‘normal regulation’) consists of applying the current regime to specific risks as they arise. That’s not to say adjustments aren’t sometimes needed, but they’re part of this everyday work of regulating within an accepted paradigm. Rules might be tweaked, but the basic framework stays the same.

Until it cannot.

For Kuhn, the history of science is the history of its paradigms and of what he calls ‘crises’ that force scientists to replace old paradigms with new ones (this is a ‘paradigm shift’). He argues that crises arise after repeated failures to align experimental data with what an established paradigm teaches scientists to expect. In Kuhn’s words, ‘nature has somehow violated the paradigm-induced expectations that govern normal science’ and, it being impossible to change nature, the paradigm must change instead. These are moments of scientific ‘revolution’. A classic example is the ‘Copernican Revolution’ which occurred when a geocentric paradigm for astronomy (with the earth at the centre of the universe) was replaced by a heliocentric paradigm (with the sun at the centre) in response to new observations.

And just as regulators have their paradigms, so too have they their revolutions. For example, the Basel regime has been rewritten twice. In each case, new ideas were adopted in response to the perceived limitations of the old paradigm. Thus Basel II instituted a paradigm shift in regulation by allowing banks to make more use of their own models to estimate their capital requirements. This was seen as an important advance on Basel I’s use of standardised risk weights because that approach could not be reconciled with evidence thought to show that banks had the information and incentives needed to measure risk more accurately than regulators. But that assumption itself soon came under severe pressure when the Global Financial Crisis of 2007–08 showed that many banks had in fact failed to understand the risks they had taken on. Another round of regulatory reform was launched to address the very visible failings of pre-crisis regulations. Basel III was the result.

So regulators manage change in two ways that fit Kuhn’s model of ‘normal science’ and ‘revolutions’. In normal times rule makers make incremental changes to keep their regulatory frameworks fit for purpose. But when they think their frameworks are seriously inadequate, they replace them with new and (supposedly) improved versions.

Of the limits and end of knowledge

Does this mean that financial regulation might one day arrive at a final and perfect paradigm?


As already noted, people subject to regulation change their behaviours in response to it. Because profit-maximising banks invent new ways of making money when regulations make the old ways harder, the nature of banking changes over time in response to regulations. Regulation must change again in turn. This is a challenge Copernicus did not face. The sun did not change its place in the sky because of his theories.

In principle, the regulatory dialectic might eventually arrive at an equilibrium when neither behaviours nor regulation changed any more. But there is another factor to consider, which is what we might call ‘history’. This is everything which happens outside of banking and regulation but which effects both (in its broadest sense, it’s the constantly unfolding process of economic and social change within which all human agency is exercised). Economic shocks like Covid-19 (Covid) are one example of ‘history’. Politics is another. So too are changes in people’s preferences which make them demand more of some financial services and less of others. ‘History’ can therefore be thought of as a stream of exogenous shocks which randomly disturb the entwined evolution of finance and financial regulation and which prevents it from ever reaching a static resting point.

Aspects of the episode of market volatility in March 2020 known as the ‘dash for cash’ illustrate how regulation, risk-taking, and external factors interact. Since the Global Financial Crisis of 2007–08, non-banks’ share of total financial intermediation has increased significantly. Post-crisis regulation of banks encouraged this important structural change because standards designed to limit banks’ leverage (such as the leverage ratio) created incentives for highly leveraged activities to shift into less tightly constrained entities. Thus the first half of the regulatory dialectic (new rules) drove the second half (changes to risk-taking). But ‘history’ (developments not directly related to regulation) was also important. The volume of tradeable sovereign debt has increased immensely over the past 10 years as a result of governments’ choices about spending and taxation. Banks balance sheets have not grown so quickly, giving non-bank intermediaries an opportunity to step in to fill the gap. This factor was not directly related to the regulatory framework but it may have played a role in the extreme volatility seen in sovereign debt markets at the start of the Covid pandemic. Regulators need to respond to this new reality.

All that is solid won’t melt overnight

After regulators conclude that change is needed, it takes time to implement a new framework.

Researchers in the pure sciences do not have to worry about this in quite the same way. Kuhn argues that scientific revolutions can happen quickly because, in a crisis, scientists dissatisfied with the old paradigm can shift immediately to a new one. The exciting answers it offers to previously intractable problems makes scientists optimistic that a few years’ work governed by this new paradigm will be enough to fill in the (often large) gaps that need filling in. This happened at the start of the 20th century when quantum physics and special relativity rapidly superseded an old paradigm based on Newton’s science.

Regulation does not work like that. Even when there’s widespread agreement that a new regulatory paradigm is needed, regulators have to carry on doing their day job with their old tools while they build a new framework. This situation can last for a long time if international co-ordination is needed.

The implication for policymakers is that they need to carry out both continuous ‘dynamic regulation’ to maintain frameworks and periodic ‘revolutions’ to keep up with fundamental shifts in regulated markets – sometimes doing both at the same time. But this means that they need to develop the ability to distinguish between occasions when a dynamic response is needed to keep existing rules fit for purpose and times when a big change makes a fundamental rewrite of the rules necessary. To do this, they need to be alert to ‘history’ (the changes taking place in society around them) as well as to what’s happening in financial markets. If they do not, they might miss those big shifts which originate outside the regulatory system but which make paradigm shifts necessary. These capabilities need to be exercised both domestically and internationally. Individual regulatory authorities are often best-placed to spot how the risks they face are evolving and to design local solutions but, because many important regulatory frameworks are international standards, change often requires international co-ordination.

Austen Saunders and Rajan Patel work in the Bank’s Strategy and Policy Approach Division.

If you want to get in touch, please email us at or leave a comment below.

Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.

2 thoughts on “The structure of regulatory revolutions

  1. “Rules stay the same, until reality is realised”. That this be 1 lesson to learn from Copernican revolution is evident to us, even if not apparent to Copernicus. The requirement of and for “dynamic assessment” is established as RAF science for pilots and navigators.
    So, why is there 1 law for Navigation and another set of rules for others to go by? The requirement of and for a relevant Framework of Value(s) is 1 that calls not to go unheeded. The ‘heuristic principle’ of credit horizon is none the less required to Observe: – ‘Earth Curvature’.

  2. One grammatical error. “effects” should be “affects” in the following: But there is another factor to consider, which is what we might call ‘history’. This is everything which happens outside of banking and regulation but which effects both (in its broadest sense, it’s the constantly unfolding process of economic and social change within which all human agency is exercised).

Comments are closed.