Putting a dent in entropy

It is the second week in January, and many of those New Year’s Resolutions will be already hitting their half-life. All those promises to ourselves that we will do less of X and more of Y in our personal and professional lives. But they were all made with good intent. And all expressing a desire (albeit an unconscious one), to put a dent in entropy.

?

Entropy?

Entropy is the inevitable process whereby chaos increases, and things decay and become less useful over time (the ‘over time’ bit is important). This gives us adages like: ‘rust never sleeps’ and Murphy’s Law – ‘whatever can go wrong will go wrong’.

The scientific principle of entropy evolved from the work of Carnot, Clausius and Boltzmann in the 19th Century. An era where men of science applied their minds to the great questions of the day – namely how to build a better steam engine. So, the explanations tend to be based on heat and pressure, and the average mileage you can get from a given bag of coal. It is also the foundation of the second law of thermodynamics. And the reason we can’t have nice things, like perpetual motion.

But, because the second law of thermodynamics is universal and inviolate, the concept of entropy applies to stars and steam engines as it does to fieldmice and financial services companies. It is just as relevant in a physics laboratory as it is in the boardroom of an investment firm.

As an example – you set up a department in your firm, you staff it up and give people tasks to do. You then close the door and leave them to their own devices for six months. When you return, your staff could have created the equivalent of the complete works of Shakespeare (and there’s no law of physics that says they couldn’t – on a timeline between now and the heat-death of the universe). But it is far more likely that chaos has ensued.

Because chaos has a gravitational pull; it is the black hole we are all falling towards. This truth was elegantly described by the physicist Boltzman when he said “systems tend to evolve from uncommon arrangements to common ones” - in any given scenario it is far more likely that the wrong thing will happen, rather than the right. So, if we consider success as it truly is: an ‘uncommon arrangement’, when ‘the stars align’ and we ‘capture lightning in a bottle’, then failure is always more likely.

So, given the inevitability of entropy, why doesn’t chaos surround us, why don’t planes fall from the sky and buildings collapse in front of us? Because we devote an enormous amount of energy keeping chaos at bay. At work, we don’t leave a department unsupervised for six months and expect a good outcome – we do ‘management’, we put in place policies and procedures, QA, oversight monitoring, and all the other controls. We are literally living, breathing entropy reduction machines.

?Entropy makes things fall, but life ingeniously rigs the game so that when they do they often fall into place.

John Tooby

Entropy and time’s arrow

As noted, entropy increases over time. And time is a one-way street; time is asymmetrical and forward-flowing - we were in 2023, we’re now in 2024. So, that New Year’s Resolution you made in 2023 to lose weight – you cannot go back and do it now. You can, of course, resurrect your ambitions in 2024 and successfully drop 10lbs, but it is not the same event, at the same time – and all those photographs of you, with your 2023 waistline, remain.

The concept of the inexorable forward flow of time is important to physicists, but it is equally important to those running financial services firms. Because:

• The future is uncertain
• You can’t change the past

?

The future is uncertain

The Jeff Bezos memo is currently doing the rounds on Linkedin, describing Amazon’s approach to decision-making. Amazon categorises decisions into:

• Type 1 decisions which are irreversible, and need to be considered carefully by senior management
• Type 2 decisions which are reversible and can be made quickly by people in operational areas

The concept is not unique to Amazon (it can be seen in the work of Gregory Bateson, Kahneman’s ‘Thinking Fast and Slow’, and even Descartes’ mind/body duality), but it is a clear effort to improve the speed and efficiency of decision making, particularly within organisations which are growing in size and complexity.

?

It is an attractive concept – it is easy to write up on a whiteboard, and it fits the ‘move fast and break things’ ethos of Silicon Valley. But firms in regulated sectors should treat it with caution, because – the future is uncertain.

For example, if you are a consumer credit firm you could choose not to review your Responsible Lending Policy, this month. That’s a decision which on the face of it looks reversible – it could easily be reviewed and amended next month. But, three years from now when you’re in a s166 and facing the prospect of paying redress to thousands of customers, the decision takes on a permanence that wasn’t anticipated.

Bateson talks about “the difference that makes a difference” – those small shifts or changes that bring about major (and potentially permanent) transitions. In Bateson’s view, outcomes are dependent on a wide range of interconnected elements, and effort is required to understand what the causal factors are. In essence, going from now to when is not easy:

• Now – recognising that your understanding of the present situation may be imperfect, incomplete, and biased
• When – understanding that the future is a cone, expanding outward over time with a range of possible outcomes. And in a complex financial services firm the range of possible outcomes expands exponentially.

?

So yes, empower operational areas to make decisions and take prompt action. But, put the necessary controls in place:

• A Risk Appetite statement which clearly demonstrates the guardrails
• Policies and procedures
• The training and culture to ensure people follow them
• QA and monitoring

?

Because, you can’t change the past.

?

You can’t change the past

If you knock some papers off the kitchen counter, their entropy increases significantly as they scatter across the floor. But the energy required to put them back to an ordered state is not significant. In contrast, if you knock some eggs off the counter they are smashed forever, and can never be returned to their former state – they have experienced an irreversible process. Time’s arrow is full of these indelible moments and irreversible processes – all those forks in the road that can’t be retraced.

For regulated firms, irreversible processes are an important consideration, because present actions bake in future risk. The actions (or inactions) today are evidence, and can become discoverable if the regulator investigates the firm in the future. And at that future point, it is not possible to go back and correct the record. What you did (or didn’t do) at the time is there in black and white, you just need to explain it.

Taking Consumer Duty as an example, the implementation was a series of moments in time:

• The October 2022 Board presentation
• The April 2023 deadline for providing information to distributors
• The July 2023 confirmation that the requirements had been met

If these milestones were missed or they were completed ineffectually then those events are fixed, as permanently as a mosquito in amber. And the same holds true for BAU activities in the firm– the Compliance monitoring, QA, annual product reviews, periodic oversight committees and policy reviews. If it didn’t happen when it was planned to happen, it didn’t happen.

This, of course, does not mean that errors are not correctable. Firms should spot and fix breaches and control failings on a continuous basis. But, there are no do-overs, you are writing the life story of your business in permanent ink.

?

Conclusion

So, carpe the diem in 2024. And when you’re developing the policies, Quality Assurance and controls to take your firm through this year and beyond, consider whether you doing ‘compliance’, or whether you are putting a dent in entropy?

?