Banks paying the cost of insuring against future bail outs

Returning to the other great issue of the day, namely "What to do with the banks?", I'd like to make the following, 'modest proposal'.
  1. Fractional-reserve banks require insurance to continue. (IMHO it's unlikely the banking system could survive de-growth or deflation for example.)
  2. That insurance is at present done by the state.
  3. The state should be paid a proper rate for that insurance.
  4. That insurance could, in part, be provided by ultra safe reinsurance companies such as Swiss Re and Munich Re, and/or the Lloyds names, who are mostly exposed to hurricane, and not credit risk. It's unlikely that they could shoulder all the risk of a bank. They would, one would think, negotiate a fair price for that risk.
So I propose the following.
  1. The insurance should be properly defined, and 'living wills' be created too (for the circumstances where the insurer is not prepared to insure the banks).
  2. Once the risks have been defined, some proportion (say 10%) of the whole insured amount should be insured by the reinsurance companies.
The government would then 'discover' the price needed to insure the banks if done in the private sector. The bank would then pay the same rate for the whole amount to the government as the rate paid to the reinsurance company for its part in the insurance.

11 comments:

Adrian Wrigley said...

What you are proposing is essentially a "tax" on fractional-reserve banks, based on the fair-price value of the risk.

The philosophy of this is that costs imposed by businesses on the government should be borne by those businesses - "The cost principle". An alternative would be to charge the businesses for the value they get under the protection of the government. This is the "benefits" principle. If the government provides services to businesses at cost, regardless of the value, this usually amounts to a significant net transfer.

In the case of the banks, the government is providing insurance, but is also a party in the banking cartel which creates the money. Government taxation is the root of fiat money value. Tax and banking law creates the privileges which are the base of the banks' profits. The value of these privileges to banking generally exceeds the insurance costs by a large margin.

Wouldn't it be more appropriate for the government to tax the banks for the value of the privileges granted?

TheClimatePhilosopher said...

Adrian,
Thank you; I've commented inline.

>>What you are proposing is essentially a "tax" on fractional-reserve banks, based on the fair-price value of the risk.

Yes

>>The philosophy of this is that costs imposed by businesses on the government should be borne by those businesses - "The cost principle".

Yes, in this instance - since the external costs I believe to be measurable - I am imposing 'the potential polluter pays for the insurance' principle.

>>An alternative would be to charge the businesses for the value they get under the protection of the government.

Presumably by value, you mean the value to banks - What banks would be willing to pay for a banking licence? Presumably that would equal the total value of 'supernormal' profits from banking?

Presumably if there are supernormal profits from banking this is to do with barriers to entry and so is a competition issue. Or perhaps you are arguing that the economies of scale associated with banking mean that there are natural economic barriers to entry?

>>This is the "benefits" principle.
OK. I'd be interested to see a calculation methodology. In general applying this principle might mean that total government tax revenues do not equal total government expenditure.

>>If the government provides services to businesses at cost, regardless of the value, this usually amounts to a significant net transfer.

Yes

>>In the case of the banks, the government is providing insurance, but is also a party in the banking cartel which creates the money.

In a sense, although the role of government in creating paper currency (pure seigniorage) and the role of banks in creating bank deposits (obviously an obligation from the bank to others) is obviously different. In general, the bank is purely a middleman.
But, yes.

>>Government taxation is the root of fiat money value.

I'm not sure what the root of paper currency fiat money value is. Habit I think.

The root of value of government debt is government taxation.

>>Tax and banking law creates the privileges which are the base of the banks' profits.
Presumably, the privilege in particular to swap bank promises to pay.

>>The value of these privileges to banking generally exceeds the insurance costs by a large margin.

Evidence? I'd imagine the cost of insurance could be quite high.

>>Wouldn't it be more appropriate for the government to tax the banks for the value of the privileges granted?

I'd be interested to see any proposals on these lines.


What my proposal does is tax the banks on an estimate of the free market value of the insurance cost, going forward into the future. It might prevent the most serious risk taking activity, by ensuring the market monitors the banks. It is therefore important to avoid the moral hazard associated with the bailouts.

TCP.

TheClimatePhilosopher said...

Comment from: Deepankar Panigrahi
Good proposal. The challenge is getting the operational part right - what is the fair price of insurance? The ideal would be that we had a proper systemic risk measure and based on that banks and other financial institutions pay an insurance premium. If they contribute more to systemic risk their premium would be higher. There is a proposal by ... See MoreTobias Adrian from NY Fed and Markus Brunnermeir but it is still not there.

This should not be limited to banks but extended to all financial institutions that pose systemic risk. Many hedge funds would fall in to this category, e.g., failure of LTCM was a systemic risk but not all, e.g., Amaranth failure had no impact. The premise of not regulating hedge funds are no longer valid (i.e., hedge funds get money from sophisticated and wealthy clients so they now the risk they are taking. Many retail investors are exposed to hedge funds indirectly and secondly failure of hedge funds can pose a risk similar to that of a financial institution). The shadow banking system is as important as the traditional banks and focusing on one creates opportunities for regulatory arbitrage.


----------


Thanks. On Hedge funds: good point; thanks.
Thanks for the link to Adrian and Brunnermeir's work. Obviously a difference with my proposal is that theirs seems that this is based on modelled-risk, whereas my proposal would try to find a market for the bank's risk.


----------


Regarding the difference in approaches, wouldn't we still have to model the risk to create a market? How will the market value the risk otherwise? I am not very clear on this point. Even insurance would require some form of a pricing model. ... See More
That said I think your idea of using re-insurance companies and insurance companies is very good since as you said the co-relations between natural disasters and financial crisis is very low.

There is a paper (it won the best paper award in the Journal of Finance last year), "Collective Risk Management in a Flight to Quality Episode" by Caballero (MIT) and Krishnamurhty (Kellog, Northwestern) on providing insurance but at pre-Knightian prices since once you have Knightian uncertainty (unknowns unknowns) the market will break down as happens during a financial crisis. Their argument is that insurance companies can insure against microeconomics risk factors or moderate shocks to the economy but not against systemic risk. They favour the idea of governments to provide this form of insurance. I think you will find the article interesting.

I have a question regarding climate change. I went briefly over the paper by Martin Weitzman, "On Modeling and Interpreting the Economics of Catastrophic Climate Change." Due to climate change if extreme events are more likely than before wouldn't that change the way insurance companies price risk. Also if the exisiting relationships breakdown, e.g., we have cyclones in many parts of the world at once, cyclone in a area not prone to cyclones,etc... wouldn't that make exisiting models irrelevant. How would insurance companies deal with this uncertainty? His paper was probably more aimed at government officicals to get them to do something about climate change but I think it does raise interesting questions for the insurance industry.

Peter said...

Insurance to cover a business risk is BAD!
The business should succeed or fail by its decisions. Insurance would penalise the well run by bailing out the bad!

The business risks are impossible to quantify, Banking is a meta level of this so it makes the risks worse not better. ( The cost of agency kicks in too)

The more direct the lender is to the borrower the better. Debt exchanges such as Zopa is as good as it gets!

There is an interesting issue about what the debt is to be used for

1) Consumption (bad)
2) Saving/Storing (Neutral)
3) Means of production (good)

TheClimatePhilosopher said...

Note that the insurance would match the maturity profile of the bank (so would have short term and long term items).
The insurers could demand disclosure to ask how much risk is on the balance sheet.

There was a suggestion (at the Union society debate) that we should be insuring the depositors of the bank, and not the bank itself (i.e. the shareholders). That seems reasonable.

We still of course have pro-cyclicality problems of banking however.

Peter said...

In the theory of the firm, the shareholders ARE the risk takers. However, the banks chose (Illegally in my view) to take risks that were beyond the size of the risk capable of being borne by the shareholders.

There may be a role for a mezzanine layer of debt equity that has a higher interest rate as it is also at risk, after the shareholders have lost all their equity, if the bank miscalculates the risk.

My point about the badness of insurance remains!

TheClimatePhilosopher said...

"My point about the badness of insurance remains!"
I acknowledge the basic principle of letting businesses fail, and not insuring specific business risks.

However
a) I am not here insuring all banks in general (at a constant rate) against their each specific risks, [that is almost what happens at present] but rather the opposite: I am proposing to insure each bank individually, (at a rate specific to the individual bank) against the risk that that bank would be unable to pay back their depositors in the event of a system-wide crash.

b) Finance has specific issues:
>>"In the theory of the firm, the shareholders ARE the risk takers. However, the banks chose (Illegally in my view) to take risks that were beyond the size of the risk capable of being borne by the shareholders."

*All* banks *always* take risks beyond the size of being borne by the shareholders (in so-called 'extreme' circumstances). That is the nature of banking. No bank would survive generalized degrowth and disinflation.



"There may be a role for a mezzanine layer of debt equity that has a higher interest rate as it is also at risk, after the shareholders have lost all their equity, if the bank miscalculates the risk."
Exactly. Mezzazine debt already exists and should have been wiped out.


My point is that governments *insure the bank's depositors* for better or for worse.

IF governments insure the banks' depositors (for better or for worse), then the banks should be charged for the privilege,. If the price is determine

There are good reasons WHY government's insure the banks' do this, GIVEN the system of fractional reserve banking, although that doesn't make fractional reserve baniking a good system.

Those good reasons are broadly, that bank defaults are not, in the short term, cathartic (like business defaults might be) (except perhaps, if one believes such things, in the system-wide sense of 'crashing capitalism'), but the opposite - all the banks are linked together and the net effect of 'letting a banks fail' would be that depositors lose their money and other banks fail, this would cause a rapid constriction in money supply, and deflation which would mean that individuals would find it impossible to pay off their debt (See Fisher's Debt Deflation Theory of Depressions).

Anyway, so the point is that this reform is a large improvement on the status quo, because it makes banks pay a proper price to insure their deposits (which would therefore incentivise less risky behaviour and more disclosure), rather than having this insurance be given to them by the goverment for free.

pjdqwe said...

In extremis, a bank that only lends to the government can be risk free. So banking does not have to include risk.

Any insurer of a bank would always have less information relating to the risks taken by the bank, than the bank being insured, this will always result in the insurer being "played". E.g. giving favours to rating agencies!

One should note that the US re-insurer ?AGI? failed.

I'll still maintain insurance makes things worse, not better!

TheClimatePhilosopher said...

Of course, not even governments are risk-free! Thanks for the comments. I'll think about them some more when I have time.

baludec5 said...
This comment has been removed by a blog administrator.
willson said...

If I were to wager a guess at why, I’d say that users don’t “browse” forms. The interaction style users engage in with forms is different, and requires its own study and design best practices.
latest technology