On the importance of failure in free-market banking systems

An article from Gillian Tett in the FT earlier this week reminded me of the importance of experimenting short-term failures for free markets to succeed in the long-term. This is a key principle of a self-sufficient and self-correcting system that has unfortunately been forgotten with the rise of the paternalist state.

This is what Larry White referred to with the expression ‘antifragile banking and monetary system’, based on Nassim Taleb’s antifragile concept. In her article, Tett mentions how actors in financial markets, as well as central bankers, have changed their views following the crisis. The vocabulary she used is misleading. Facts have not changed. Facts have always remained the same. Economic agents simply didn’t know how to interpret those facts. We call that learning.

As Larry White says, “the banking system is not inherently fragile” and “a more thorough look at theory and empirical evidence indicates clearly that banking is not naturally fragile.” (his emphasis)

He then adds:

The view of banking institutions as naturally fragile is implausibly anti-Darwinian. It defies the Darwinian principle of natural selection (“the survival of the fittest”). Given a few centuries, financial institutions that are inherently prone to collapse should be expected to collapse and thereby to disappear over time, while sturdier structures should be expected to survive. The inherent-fragility view of banking cannot explain how modern banking survived, much less how it flourished and spread across the world, as it did for the seven-plus centuries between its emergence around 1200 (Lopez 1979: 12) and the arrival of official safety nets after 1900 in the form of government-sponsored lenders of last resort and national deposit insurance.

While single institutions are indeed prone to failure, the whole financial system is not inherently weak but is made weaker by restrictive/intrusive/limiting institutional frameworks. Paternalist frameworks attempt to forbid banks to engage is what regulators view as ‘risky’ activities. The consequence of this being to prevent the system from learning from its mistakes and reinforcing itself.

Crashes following financial liberalisations are often used as evidence of the impossibility of the banking system to self-regulate by proponents of a tight control of the financial sector by the state. Those critics are misplaced. Crashes following financial liberalisation are normal. Like children discovering some risky new games, financial actors haven’t yet found out the inherent risks associated to their new activities. The system crashes and it learns.

However, crashes need not be that painful. State intervention (or ‘protection’) postpones the learning process. Meanwhile, innovations accumulate in order to bypass some of the state’s introduced artificial limitations. When the system is finally ‘released’, multiple live experiments, some really spontaneously emerging from the natural evolution of the system, some others pure products of regulation avoidance, are launched simultaneously, with potentially disastrous consequences when many of them crash at the same time.

But free markets are not to blame in such circumstances. The system followed its natural Darwinian selection rules and should the state not have restricted financial activities in the first place, the process would have been smoother and spread out over a longer period of time. Historically, the freest banking systems were also the most robust ones (comparing the US and Canadian banking systems at the end of the 19^th and early 20^th century quickly confirms that).

This mechanism inevitably played a role in some of the previous financial crises, from the S&L crisis of the 1980s to the recent 2008 crash. China seems to have taken a cautious view and aims only at liberalising its banking system step by step. This sounds sensible.

Learning from experience is crucial in all aspects of life. By preventing the system from learning through various regulations, protection schemes, deposit insurances and other paternalist interventions, the state and regulators prevent whole generations of bankers (as well as the overall population) from discovering what they should and should not do. This is dangerous. Short-term mild pain is often a necessary evil that promotes long-term robustness. We call that wisdom.