If banks don't have 100% capital ratios, they are subsidised.

My latest magnum opus – 6,000 words. It's available here. The basic argument, is very simple, and is briefly as follows.

Bank subsidies are only necessary because of an inherently risky element in the existing bank system: the lending on of depositors’ (supposedly safe) money, and any activity that needs subsidising reduces GDP, unless there is a very good social justification for the subsidy.

Or as Prof Adam Levitin put it, “Banking is based on two fundamentally irreconcilable functions: safekeeping of deposits and relending of deposits.”

Ergo the bank system that maximises GDP is one where the lending on of supposedly safe money is banned. That is, money which savers want to be totally safe should be lodged in a genuinely safe manner, e.g. with the state. As to loans, they should be funded by equity.

That “GDP maximising” system is called “full reserve” banking.


What about FDIC type insurance?

It might seem that the lending on of depositors’ money can be done in unsubsidised form with the help of some sort of self-funding FDIC insurance system and/or a lender of last resort facility. However, where a money lender is funded just by equity, shareholders actually insure themselves. So which is better: FDIC type insurance or shareholder “self insurance”?

Well in theory if they both gauge the premium right, there won’t be any difference. However there is a big problem with FDIC type insurance (and indeed support for banks via lender of last resort (LLR)).

It’s that FDIC/LLR type insurance is inherently expensive because of the moral hazard involved: the temptation to take excessive risks, keep the profits when that works, and send the bill to the insurer when it doesn’t. That moral hazard largely explained the crunch, and the costs of that were horrific.

Conclusion: the existing bank system is in check mate – full reserve maximises GDP, while the existing / conventional bank system does not.