Why you cannot regulate systemic financial risk

Governments of countries affected by the financial crisis are preoccupied with designing a set of regulations that would prevent the future banking crisis (called liquidity shortage or credit crunch) from happening. The actions are taken on the local level, introducing new regulations, and international level, e.g. Basel III. At the same time nothing is done to break up banks so they are not "too big to fail" or to separate investment banking from high street banking that would prevent using depositors cash for, often very risky, financial speculation.

But is it possible to regulate financial risks? To get to the bottom of it, it is necessary to understand what the major factor of liquidity shortage risk is. Ever since the full reserve banking was abandoned, i.e. the situation when banks were acting as safe storage business and depositors had to pay to keep their money there, when bankers realised some hundreds years ago that at any one time they only get a withdrawal call on a small proportion of the deposits held, banks started circulating money. Circulation is in fact money multiplication, i.e. for every £100 paid in as deposits banks lend out a portion of it. Typically it was between £80 - £90. This made the banking system work as a statistical machine whereby for every £100 liability, the bank held only £10 - £20 cash.

The risks of running a banking system in such a way is rather obvious. The first one is psychological: if depositors demanded a large proportion of their deposits, for whatever reason, a bank would not have money to satisfy them. This is called a bank run. Banks were insured for such a contingency in two ways.

The first one was banks lending to one another. In case one bank was short of liquidity another bank was lending it to satisfy the demand. As the interbank interest rate of lending was lower than what banks were offering their customers, all banks were still making money, it was a profit sharing scheme, and the whole system worked like an integrated statistical machine.

The second way of insuring against liquidity shortage was keeping the actual liquidity reserves at the appropriate level. For example, at every one time, for every £100 possible immediate withdrawal call, a bank has say £15. And this is a hard one. Banks create money out of nothing: they are statistical risk machines. Therefore the basic question is : where are the borders between reasonable commercial risk taken by the bank and reckless risk that leads to failure (or even much too high risk which amounts to a downright fraud)?

Sometimes it is possible to identify a straightforward case of fraudulent actions. For example in the run up to the current financial crisis that started in 2007/2008 many banks, including major well known banks, were lending with loan to deposit ratio greater than 100%. This was a classic fraud, known for hundreds of years as a pyramid scheme. (That is why people who have been responsible for the banking system in the run up to the financial crisis financiers, regulators and politicians were classic fraudsters or grossly incompetent/negligent. However their sheer number and collective massive influence guarantees their immunity from prosecution.)

However finding a line separating taking commercial risk and reckless risk by banks is impossible. It is changing on a day to day basis and it is impossible to cater by regulations or laws for every major economic eventuality. (All those who are old enough may remember that the communist system was an attempt at such a holistic approach to the economy that spectacularly failed.) For hundreds of years it was tough justice and self regulation, in a free market setting, that ensured that banking system worked. Later central banks came onto the scene like Bank of England. They ensured that problems were picked up early before they got out of hand and also acted as lenders of the last resort.

Firstly banks owners put their capital and their freedom or even life at stake. Dealing with some else's money is a serious business. If a bank failed, bankers were losing a lot: quite often all their and their family wealth, their freedom or even their lives (and even their families). For some centuries of banking such harsh penalties were watered down and concept of limited liability was introduced. However it is not unknown that only a few decades ago Bank of England was taking very tough action on banks and bankers that failed and were rescued. It was not widely publicised in order not to raise panic. Therefore, whilst not advocating returning to exactly ancient methods, financiers and all those who are responsible for handling someone else's money must face a very harsh personal justice in case of failure. This is the first level of deterrent. The current practice of rewarding financiers for recking the financial system is not only a travesty of natural justice but, on a practical level, also ensures more problems in the future.

Secondly banks were only lending to each other, to help in case of liquidity shortage, if they trusted each other. Therefore they were monitoring each other whether they were taking justified commercial risk and the liquidity shortage was a statistical event, or a bank in trouble took a reckless risk in which case it was denied a short term credit to cater for liquidity crunch. In such case the bank was allowed to fail with all the consequences for the bankers described above. In the UK, the Bank of England sometimes stepped in (not in the recent case of BCCI though) to rescue such a bank. Nevertheless this also involved harsh penalties for the failing bankers. Sometimes other banks stepped in and took over the failing bank.

The public must realise that depositing money in a bank carries a risk. However such risk can be insured. This means that it has to be backed by additional insurance capital, so the return on deposits would be lower than otherwise would have been. Whether it is a government backed and compulsory insurance or voluntary arrangement, is a policy decision.

In fact a bank going bankrupt every so often, like in any other business, is a very healthy part of the free market economy. It weeds out the weak businesses, reckless risk takers, unlucky and fraudster. It is a proper free market pruning process and banks must not be immune to it.

Thirdly, and above all, banks cannot be "too big to fail". For hundreds of years this was precisely the case. If banks are "too big to fail", then the free market is dead. Self regulation does not work. Instead it is a system akin to communist central planning that so spectacularly failed over two decades ago.

In a nutshell, in order to put the financial system back into shape governments must not regulate more but less as it is impossible to regulate a dynamic financial risk. It is the exactly same problem as a challenge of centrally planned economies: predicting people's micro-behaviour. Governments must also reintroduce a free market into financial industry. Banks must be small enough to fail. Banks must regulate themselves, apart from general law and basic, bottom line regulations (this is typically the role of central banks and general justice system). In the case of a banks' failure, all those carrying responsibility for managing such a bank must be held personally responsible with very harsh penalties.