Getting to grips with unknown unknowns

After agonising over this problem myself, I came to the personal conclusion that the key lay in the way we thought about global issues. Most economists, including myself, were taught the neoclassical “general equilibrium” model. This assumes perfect information, a friction-less world and basically an ideal model (like Humpty Dumpty) that will revert back to stable equilibrium when it is slightly out of kilter. 

We now know that this simple model is the foundation of most of the current econometric and asset-valuation models in use today for policy formulation, risk management and asset pricing. Unfortunately, we were lulled by this “ideal world” and could not think “out of the box” about a world that is full of unknown unknowns and “black swans”. The result was disaster myopia.  

Indeed, mainstream economics training led us to always go for “first best”, and thus most economic advice sought ideal policies that in practical terms were impractical to implement, or nigh impossible in political terms. In this world of specialist knowledge where we know more and more about less and less, common sense is not common.

Indeed, the fact that most of us realise that the world has changed profoundly since the 2007 crisis suggest that what we truly lack are moral and intellectual compasses to deal with multidimensional, complex, simultaneous transformations, such as rising population, disruptive technology, industrial transformation, climate change, terrorism and human conflicts and natural disasters. In this period, uncertainty, change and death are the only certainties.  

How do we deal with unknown unknowns? Human beings are by nature optimistic, which dulls our recognition that our environment has changed. Think about the Easter Islander who cut down his last tree. He thought building another statue to his gods would rescue him from oblivion. 

The older economists like Keynes, Hayek, Herbert Simon and Milton Friedman all understood that we must work with imperfections in information and knowledge. Hayek’s 1974 Nobel address, “The Pretence of Knowledge”, laid bare the perils of over-active policies assuming omniscience, when we don’t know enough. He was against Big Government because by his definition, no bureaucracy knew better than the self-order of markets. The physicist Richard Feynman argued that “it is not what we know, but what we do not know which we must always address, to avoid major failures, catastrophes and panics”.

Sociologists, anthropologists and political scientists have always known that human beings group together in order to deal with uncertainty. The limited company was an invention in Holland and Britain that reduced uncertainty for investors. Survival under complex, non-linear conditions requires social organisation, leadership and adaptation. This means hierarchy or structure and processes of searching for better information to make decisions, sharing that knowledge with everyone, experimenting with new things, learning, making mistakes, innovating, adapting, expanding and exploiting the new tools or knowledge and then evolving.

The massive confusion we are witnessing is the result of mega-trends – the increase in population to 9 billion in two decades, with more than half now living in cities; profound technological and industrial change; and also rising natural disasters, social conflict and effects of climate warming. 

Our minds and training cannot cope with simultaneous changes that are really systems interacting within systems in an ever-changing, multipolar, multidimensional global system. We have to have a major rethink of economic and political theory to address growing inequality, rising unemployment, financial disarray and climate change. But we tend to examine our markets in isolation.

Californian system thinker Frithoj Capra suggested that the “development process is not purely an economic process. It is also a social, ecological, and ethical process, a multidimensional and systemic process.”

The trouble is that bureaucracies tend to fight complexity with more complexity. The US Glass-Steagall Act that separated banks from capital markets worked for more than 60 years was only 37 pages long. Proposals for the Volcker Rule under Dodd-Frank are already at 325 pages, and the Dodd-Frank Act is 1,100 pages, growing probably to 30,000 pages after all the subsidiary rules are put in place. 

In the area of bank regulation, the first set of Basel 1 rules in 1987 was 30 pages, with Basel 2 in the 1990s rising to 347 pages, and the latest draft of Basel 3 being 616 pages. Much of the latest proposals are based on complex models to estimate risks. If these rules are so complex that only specialist risk managers and lawyers understand their meaning, how can company directors, many of whom are independent and part-time, fully grasp their significance? 

In simple terms, when we are dealing with unknown unknowns, is it better to have very complicated systems with complicated rules, or relatively simple systems that everyone understands, finds easy to use and can be personally accountable for?

There is increasing realisation that we simply do not have good enough theory to guide us in such complex times. 

In contrast, crossing the river by feeling the stones is not a random process. It is a practical skill that can be learnt, by companies, institutions and even countries. Basically, it is learning by doing. One begins by strategising within uncertainty, then prioritising within limited resources and identified risks. In the face of uncertainty, you experiment by “piloting” – searching for new options and alternatives. Your team must share the experience of discovery, for they will “seed” the new teams that are able to adapt and innovate through co-creation with other teams. If you succeed, you exploit and expand the pilot. If you fail, you review, regroup and try again.

Systems thinking and action means that we learn collectively to deal with the complexity, unpredictability, and reflexivity of the real world.

 

Andrew Sheng is president of the Fung Global Institute.