“If we learn something from this, it is that the higher you go, the more incompetent institutions are; The opposite applies too… the most competent people are in small villages. They know exactly what measures to take. We must move to localism.”

So proclaimed Nassim Nicholas Taleb, Distinguished Professor of Risk Engineering at New York University, and the author of The Black Swan: The Impact of the Highly Improbable, writing about lesson’s to be learned from Covid-19 in last week’s Sunday Times.

Is Taleb right? Or should there be tight central coordination to maximise efficiency and ensure equality?  What serves us best dealing with Covid-19 crisis, or more broadly with the societal challenges we face?

The creation of the Nightingale Hospitals and rapid decisions on what functions hospitals should stop and how wards should be reallocated was a great success for central government. PPE provision less so. The aim to ensure that the initial PPE distribution was to some extent equitable across the country was logical. But opening up to local solutions has demonstrated the value of this. Similarly with efforts to scale up testing and as may happen again, the pace of opening up local services.

Considering the question more broadly, and thinking about precisely what should be centralised, the obvious but not particularly informative answer is that coordination and decision making should be made at the right level.  Sometimes that is centrally, but I agree with Taleb that the default should be hyper-local. In his words, the village. In UK terms this means the Parish or Ward.

Why should this be the case?

If the challenge is complex and unpredictable, requires integration of different pieces of information and discussions made at speed, then those involved can be better coordinated when they have relationship-based interactions, rather than transactional ones. Humans max out at 150 people, the manageable number to interact with in a group, sometimes referred to as The Dunbar Number after the Oxford anthropologist who highlighted the link between primate brain size and optimal social group size. This group works best if there is a shared geography, but there doesn’t have to be, with the key element being the group having a shared interest.

The next advantage of small groups is it is easier to see trade-offs. Everything in life has a trade-off, but when decision-making is remote, these are hidden from people, so things appear free from cost and compromise. This then leads to the arms race between the political parties at election time, each offering more and more goodies.

The most important reason though for devolution to small groups is that people embrace solutions if they are actively involved in formulating them. That’s the way our brains work. If there had been any doubt beforehand, Covid-19 has shown us the latent potential in local groups, whether that be the many neighbourhood support schemes, or local academics and businesses working together at speed to create new ventilator kit or PPE, delivering these often in weeks or even days.

With the war effort motivational incentive and associated freedom of action, people have done amazing things and of a complexity and pace most would have doubted was possible before the crisis.

Still, I do actually support some aspects of centralisation and the need for a tight coordinating group at the centre of government that can link across departments, like the Number 10 Policy Unit or equivalent. But it needs to also embrace grassroots local activity, with rings of coordination in between, such as the mayors. Polycentric governance with good links between the rings and a principle of devolution to the most local level possible. That is true participatory democracy.

What goes in each ring is open for debate. Placing some things is easy, such as central control of the army or intellectual property law. Regional control has coordinating advantages, but must not replace one distant bureaucracy with another closer to home and with current voting systems comes a risk that some regions may be lumbered with the same political incumbents forever.

This is why I favour some form of proportional representation at and below regional level, and much stronger protection for public representatives from the vexatious, troubled or plain nasty, who create misery, impede progress and inhibit too many people from entering public service.

How and who we elect should vary according to the domain we are talking about, because our attitude to type of governance also changes according to the polycentric ring and the subject of the control.

“I am, at the Fed level, libertarian; at the state level, Republican; at the local level, Democrat; and at the family and friends’ level, a socialist.”, as American experts on neighbourhood planning, brothers Vince and Geoff Graham, so brilliantly put it.

The role of central government should be to provide the enabling tools and at least in certain domains, parameters of the destination. Localities are best placed to work out how to get there. Rather like with an orienteering race, where there are several choices which must be made to get to the target destination, do you take the safe but longer route along an easy path, or a shorter but more uncomfortable route crossing a marsh? Or is your preference for the fast but high risk route which has no markings.

The point is you assess the trade-offs and make a decision appropriate for your situation and risk tolerance.

Central government can help with data, informing us as to whether progress is being made. If completely lost, then higher levels of government should step in, orientate and share best practise. County can support Parish, Regions the Counties, and national government the Regions.

With this, a more nurturing and less paternalistic state can emerge after the shock of the Covid-19 crisis.