Roger Miles Researcher, Speaker, Author and Moderator;

Former Speaker, Moderator & Thought Leader at previous RAO conferences, Roger

The principles underpinning behavioural economics have been attracting legislators seeking to rein in the excesses of the financial sector. Are they motivated by saving face, political gain or doing right by the public? Roger Miles conducts some research.

With half of the world’s population going to the polls this year – at least in democracies – we should expect 2024 to bring fresh shocks to government and governance. It’s 15 years since perhaps the last big governance shock, when global financial markets crashed.

That event set off an earthquake in the landscape of #riskgovernance, as regulators moved abruptly towards behavioural economics. As so many countries’ citizens ponder whether to eject or reinstate their governing parties, now seems a good time to ask: What has been the real impact of behavioural science in curbing anti-social excesses? In the financial sector, or elsewhere?
 
Shamelessly today, I’m raiding the archive for a thought-stirring article first published on the tenth anniversary of the crash. In it I gently dig into the complexities of the new ‘behaviour-based’ regulatory agencies who now merge economics and law with anthropology.

Have we, come 2024, successfully completed a transition from contract- and product-focused regulation to a real culture of holding individuals accountable? (Recent events at the #PostOffice would suggest: not everywhere. Possibly nowhere, even?)

Has the insertion of behavioural science into rule-making changed anyone’s ideology, at a governance level? Or has it been, like much of ‘Nudge’, become a sour tale of political expediency?

As you know, I love to truffle around and to join the dots between behavioural research and the not-always-honourable motives of players in the governance space – whether government ministers, financial market players, or others. It’s only fair to question how far the new behaviour-based regulation has truly prevented any fresh outbreaks of public harms, or become simply a new set of MI: an ‘input indicator’ signalling virtuous engagement but little more.

Jump into this shark pool with me as I re share my thoughts with you in an article celebrating it’s seventh birthday. 7 year itch?

As the current crisis of trust in public housing safety shows, it can take a catastrophe to force us all to overhaul the ways that we perceive and govern risk.  After the financial market crash of 2007-8, many market watchers – including me – saw just such an opportunity: we could at last bring in better, people-centric ways to regulate financial risk-taking.  Before the crash, years of patient work by“alternative” economists had made steady but mostly unspectacular progress in getting the attention of legislators. Now, ten years on, has the crash made government  agencies take on board the new economics? Are new laws reflecting a fresh understanding, among legislators, of how people really behave?

On the surface, one change at least is clear. Behavioural economic principles now direct regulatory enforcers’ first attention towards the people doing the selling, rather than to products sold as in the past. Specifically in the case of behavioural regulation of financial services (the so-called Conduct regime), senior bankers are held responsible. Bank principals are now held to account in person for their decisions, under threat of prosecution through the Senior Managers Regime (whether UK-FCA, or analogous conduct controls in other countries). Around the world, financial firms’ managers face direct, personal prosecution, including the threat of prison. A new regulatory species, the behavioural enforcer, is at work, from Australia to the USA, from Hong Kong to London, calling out badly behaved people in financial high places.

So-called conduct controls have proved a hit with governments everywhere. Two less-than-ethical motives for this are sometimes apparent: saving face, and realising massive windfall revenues from fines. We are told that the impulse for all of this is a happy new conjunction of behavioural science and the law. Critics would counter that what’s really happening is not a fundamental change – new economic principles are not, in fact, embedding into new laws and the whole conduct project has less to do with new ideology, or scientific method, and everything to do with good old-fashioned political expediency.

Certainly the new behavioural control regime takes some account of the real science – such as around the impact of bias effects on the sales process – but many bankers see it as an excuse for government to initiate friendly-headline-grabbing prosecutions against “fat cats”, a lazily assumed token target group.

There are at least some grounds for optimism. Behavioural research is laying bare many of the flaws in conventional law-making and regulatory practice, with their fatal attachment to econometrics.

Take the case of market benchmarks such as LIBOR: looking back, it now seems weirdly optimistic, or just plain wrong, that anyone would have designed a system of market restraint around the idea of asking sellers to self-report their own probity.  As an approach to control design this seemed almost to invite abuse. It fails completely to allow for human response effects that any behavioural economist would warn against – for instance, the strong incentive for traders to collude privately, and to suppress their own recognition of any public detriment, when very large personal gains are in view.

 “Always keep in mind the  long  view of why regulation is created. Every new regulation is, in a sense, a product of failure.”

 As new cases of market and customer abuse continue to appear, bad behaviour seems still to be with us, suggesting that perhaps conduct control hasn’t quite gained the traction that was promised. Financial services regulators have explicitly tied their use of behavioural economic interventions to expected outcomes of culture improvement.   However,  the regulators have not fully co-opted behavioural economic measures into their new control designs, preferring instead to leave it to practitioners to respond with proposals for “Conduct and Culture dashboards” – a blue-sky request which is making plenty of work for the finance industry (and, in fairness, this author) during the current business year.

The new regulator’s lack of specific guidance on “behavioural economics-based auditing” also reflects another,  more familiar explanation.  Before the crash, the reporting of risk was widely “gamed”; traders routinely manipulated indicators; salespeople oversold; trading line losses were buried deep inside aggregated reports; and proscribed trading partners were given anonymous accounts. There was no transparent account of how salespeople were behaving day-to-day; if there had  been, perhaps alarm bells might have rung sooner.

Immediately after the crash, the many governments hit by the event faced a triple challenge: saving face and credibility; devising a new system of regulation to replace the failed one; and doing both of these things without spending a lot of public money at a time when national treasuries were, let’s say, fragile.

On that last point alone: to everyone’s surprise, Treasury ministries included, the public-money dividend of borrowing behavioural economics to use in financial regulation has been unimaginably huge. Revenues from fines against financial firms have risen ten-fold and more, in barely four years since the start of the conduct project. For any cash-strapped government, $300bn+ of windfall revenues is beyond a blessing. Ironically then, it seems that the new behaviour-based regulatory regime is itself in the grip of seriously perverse incentives including the fact regulatory failure delivers substantial revenue. To see how this happened, let’s consider where regulation, generically, comes from.

Always keep in mind the long view of why any form of regulation is created.

Every new regulation is, in a sense, the product of failure. New rules respond to the events of old rules not working, and of governments being wrong-footed by this. No national leader has ever woken up one sunny morning with the thought that “I fancy a spot of regulatory drafting”.

Back in 2008, every government hit by the financial crash – that is, most in the developed world – faced not just a financial crisis, but also an existential one. Voters were suddenly asking: How effective is my government, these people who let this happen on their watch? Some governments fell at the ballot box, as a more or less direct consequence of losing public trust.

Why old-school econometric regulation failed, and the new behavioural regulation should succeed.

Before world financial markets crashed in 2007-8, the regulation of financial firms tended to consist of regulators asking firms to report theirown assessment of risk-taking. By general assent, firms used econometric analysis to do this – that is, they audited their own risk-taking by looking back over their own financial trading records, sometimes comparing these against other firms’ levels of financial activity. The old style of regulation thus saw risk mainly in terms of recorded movements of money (that is, proxy and retrospective indicators), and did not trouble itself with real-time observation of actual bankers trading. For example, a standard paper risk indicator was the “heat map”: the predicted cost if a hazard occurred versus the judged likelihood of the event occurring.

The focus of all this risk assessment was cost. Classical econometric modellers (Econs) were not much interested in a risk’s impact on human beings, or in how people make decisions in real life. Worse, they leaned on some distinctly unsafe assumptions about what motivates human beings. For example, Econs tended to see humans in general as rational maximisers – cynically implying that we’re all only motivated by a desire for money.

More troublingly, Econs ignored many known patterns of how people actually behave in the real world – the so-called Econometric Problem of the Normal People. Econs preferred to dismiss as aberrations or noise, the facts that in real life:

■  humans use emotions when making decisions (the affect heuristic);

■  as social animals, humans are frequently altruistic;

■  humans tend to rush blindly to over-consume scarce resources, leaving a broken market (or ecosphere) in the aftermath (the tragedy of the commons).

Modern behavioural regulation will work better – at least in theory – because regulators say they’ll watch how financial practitioners and customers actually behave, in real-time. These aspects include our human tendencies to: biases (our brains mis-seeing and misinterpreting rational information); and alter our behaviour to fit in better with the people around us (even when those people are misbehaving).

By better reflecting a real understanding of how humans make decisions, including how emotions and other biases bring different outcomesfrom rational analysis, the new behavioural regulators hope to save us from a repeat of the market crash. Are the new regulators making correct use of behavioural research findings? And if so, will this work to stamp out “bad banker behaviour”? Time will tell.

As with any crisis of public confidence, government looks for ways to reassert its authority and legitimacy, ideally presenting its choices in a voter-friendly narrative. Looking at the wreckage of financial markets – just as, ten years on, we observe the charred remains of a tower block – public officials realised that it is unhelpful to use administrative jargon to explain to people how they were responding to the crisis. Much as citizens caught up in a tsunami don’t much care about hydrodynamics – they’re too busy trying to stay alive – so voters bankrupted by a financial crash had little interest in the mechanics of collateralised debt  instruments; they just wanted to get back  to normal life. Worse – to voters who have just lost their homes and jobs, a taxpayer- funded bank bailout looks suspiciously like an imposed act of social injustice. Voters were (and many remain) deeply troubled by the thought that their elected representatives seemed to have lost control over financial firms, or worse, to have caved in to“big finance” interests. Protest voting, and populism more widely, are lingering products of voter anomie since the bailouts.

Facing a cliff-edge of public trust at the time of the financial crash, politicians needed urgently to find a new regulatory concept that could help deflect public anger elsewhere but that also could be overlaid easily on existing control agencies, for a persuasive new regulatory “narrative”.

Against that somewhat desperate brief, behavioural regulation offered a wonderfully expedient option, focusing as it does on prosecuting senior individuals and legitimising the use of expressive, “salient and vivid” (eye-wateringly huge) fines against errant brands.

To the committed behavioural analyst (confessing to a personal interest here), the recent boom in behaviour-based regulation looks less like a revolution in public agencies’ application of science, more like simply realpolitik business as usual. The co-opting – or more plainly,  cherry-picking – of behavioural principles into regulation has mainly succeeded in handing governments around the world a fairly cheap way to attack business leaders whom the public were angry at, in an unfocused way, as having “got away with it”. For example, behavioural regulation makes great play of its focus on bias effects but has rather less to say about other important behavioural economic themes such as reciprocity and how the informal organisation subverts formally designed controls.

 “It plays well that conduct fines reduce public deficits, and government can use the new regime to recast its own role as fearless champion of consumer rights.”

No matter: it plays well that conduct fines help reducing public deficits, and that government can use the new controlregime to recast its own role as a fearless champion of consumer rights. Meanwhile our community of “new economists” laments how their good work has been hijacked by state agencies whose main motives are in addressing “lack of funds and political helplessness”1. A golden moment of opportunity may have already passed as recent public audit reports have begun to question enforcers’ lack of evidence that “behavioural enforcement” acts to reduce mis-selling.

The historic pattern is clear. Large-loss and high-jeopardy risk events, or the public prospect of them, tend to focus the public dialogue and to create more noise. Our research community might best respond by exploiting more ruthlessly the opportunities presented by other crises of public trust, to help refocus the debate.

There’s no lack of choice of these, right now. Government is facing tougher than ever tests of its authority and public legitimacy, not just over financial market fragility; take your pick of the issues, from major infrastructure projects, to public housing safety flaws, to police efficacy, or even to national identity. Now maybe just the time to welcome a more honest inclusion of human risk factors into any of these legislative agendas, such as a wider inclusion of significant behavioural economic themes of risk cognition, the impacts of informal organisation, and the limits of personal agency in behaviour modification. Truly behaviour-inclusive drafting of new regulations could save the public purse from having to pay legislators to re-learn, yet again, the same old lessons after future failures from (now largely predictable) patterns of human error.

Roger Miles

Roger researches human-factor risks in organisations and advises professional standard-setting bodies. His books on improved reporting frameworks for Conduct and Culture have been endorsed by financial regulators worldwide. He is …