Thursday, February 13, 2014

Calomiris and Meltzer on Financial Reform

Charles Calomiris and Alan Meltzer have a very nice Op-Ed on financial reform in the Feb 13 Wall Street Journal
At a Senate hearing in January, Elizabeth Warren asked a bipartisan panel of four economists (including Allan Meltzer ) whether the Dodd-Frank Act would end the problem of too-big-to-fail banks. Every one answered no.
See, economists can agree on something!

An important point: The financial crisis was not a chain of dominoes, but a run. A popular theory holds that A (Lehman say) failing causes B to fail if A owed B a lot of money, and so on. This is simply not what happened. Lehman's failure made investors worry that other banks who had made similar bets might also be on the edge, so they ran to get their money out.
The principal danger to the banking system arises when fear and uncertainty about the value of bank assets induces the widespread refusal by banks to accept each other's short-term debts....
The collapse of interbank credit in September 2008 was not the automatic consequence of Lehman's failure. 
Rather, it resulted from a widespread market perception that many large banks were at significant risk of failing. 
What to do?
To ensure systemwide resiliency, most of Dodd-Frank's regulations should be replaced by measures requiring large, systemically important banks to increase their capacity to deal with losses. The first step would be to substantially raise the minimum ratio of the book value of their equity relative to the book value of their assets. 
It has been interesting to watch a consensus develop among people who think about financial stability. In 2008, ideas were all over the map, and there was a lot more support for many parts of Dodd-Frank, including the idea that regulators could keep banks from taking too much risk. It all seems to be boiling down to much more simple idea: Banks need to get money by issuing equity, a lot more equity, instead of borrowing it.

How much?
The Brown-Vitter bill now before Congress (the Terminating Bailouts for Taxpayer Fairness Act) would raise that minimum ratio to 15%, roughly a threefold increase from current levels. 
Here, Calomiris and Meltzer veer off into the land of the current debate. Sure, more equity, but how much more equity? 20%? 30%? And what's the denominator? All assets? No, surely, as then banks choose riskier assets. Risk weighted assets?  No, surely as then they game the risk weights.

These thoughts drive me to think the answer is to set a price, not a quantity. For every dollar of short-term debt, pay the government (say) 10 cents. I don't know the exact number either, but a wrong tax rate does a lot less damage than a wrong quantiative restriction.

I think they recognize that's where the discussion is -- but this is a lot simpler than the 10,000 pages of Dodd-Frank regulation!
 There is plenty of room to debate the details, but the essential reform is to place responsibility for absorbing a bank's losses on banks and their owners
Yes!  Rather than keep a fragile, short-term-debt-laden run-prone financial system, and hope that the wise guidance of regulators will keep any bank from losing money, or from being in doubt of losing money, let us construct a resilient financial system, in which investors transparently bear losses commensurate with their rewards.


  1. John,

    Just to be clear:

    "The financial crisis was not a chain of dominoes, but a run."

    "Banks need to get money by issuing equity, a lot more equity, instead of borrowing it."

    More equity will not prevent a run. It will lessen the legal protections that investors in banks have and limit the federal government's intervention into the financial system (bailouts).

    To limit runs requires either:
    1. Securities be non-marketable (for either debt or equity)
    2. Investors be predominately long term

    The first solution is a property of the securities themselves, the second solution is a property of the investors.

    And you know something, this will never happen unless the federal government itself takes the first step - meaning the federal government begins selling equity. Time to lead by example instead of passing the buck and pointing fingers.

  2. John,

    I agree with your conclusion that finding a price mechanism instead of a quantity mechanism is a better approach. I thought a good speech and analysis on this was given by Fed Governor Jeremy Stein (link attached) where he comes to a similar conclusion using a surtax on size of institution. Obviously, you could cut it that way or you could also cut it based on riskiness (more tax for short-term debt), which is essentially the risk-based capital model.

    Stein has a responsibility to defend the current state of international negotiations but the conclusion he draws isn't that far from yours.

  3. Dear Professor Cochrane,
    Are you familiar with Jeremy Bulow and Paul Klemperer's proposal, which is designed to rely on market prices and removes the possibility of regulatory forbearance?

  4. John,

    Does a tax interfere with monetary policy? The central bank may want to induce a positively sloped yield curve, while a tax may present a negatively sloped lending curve.

    Traditionally, banks make money on the spread between short term borrowing and long term lending. That spread is partially controlled by monetary policy and partially by the market (supply / demand for long term loans and bonds).

    Introduce a tax on the short end borrowing, and suddenly fiscal policy and monetary policy are working against each other (not always a bad situation).

    Imagine this situation:

    Primary dealer bank earns a spread of 3% on money that it lends to the federal government. Federal government decides to tax primary dealer bank's short term borrowing by charging 4% on the amount of short term borrowing the primary dealer bank engages in.

    Suddenly, it becomes unprofitable (on an after tax basis) for said primary dealer bank to lend to the federal government. Meaning that federal government gets less tax revenue AND it has trouble selling it's bonds.

    That is not to say that the profitability of lending to the federal government should be a protected. I just don't think you have looked at the issue very deeply.

  5. John -

    You seem to feel that government intervention is needed to increase the use of equity capital in the financial system. Even your phrasing in this article...e.g., "Let us construct a financial system..." smacks of central planning, as opposed to your usual market oriented demeanor. What are the reasons for this departure? Is it a true departure, or is there a reason that government caused this problem in the first place that requires intervention to fix?

    1. Mike,

      "You seem to feel that government intervention is needed to increase the use of equity capital in the financial system."

      There is a difference between government intervention in the financial system (government tells private citizens how they should pay for things), and government participation in the financial system (government tells private citizens how it is going to pay for itself).

      A government can increase the use of equity capital in the financial system by using equity to fund itself instead of telling banks to use more equity.

  6. A simple proposal like a tax on short term debt sure beats Dodd-Frank's 10k pages of insanity, but I have to wonder how effective it would be. The line between debt and equity can get a bit muddled, and a tax on one may simply blur the line further. The Byzantine structures of some the CDO's that were being traded during the bubble years raises the suspicion that securities lawyers would probably break out in howls of laughter at the idea that you could force them to seriously alter their capital structure with some little tax on issuing this or that security. They figured out how to game the risk weightings, this one might prove to be even easier. Or not, I don't know. Just putting it out there.

  7. Interesting--and I wholeheartedly concur, let's have big and simple solutions as opposed to 10,000 pages of regulation. (Same for the tax code, btw).

    I like the idea that banks pay fees based on their fraction of short-term liabilities. That seems clear, and hard to game what is a "short-term" liability (they will find a way, but).

    So, to avoid fees, banks will seek long-term liabilities. They can't run either.

    Yes, equity owners should bear full bring of all losses.

  8. "The financial crisis was not a chain of dominoes"

    It probably was a chain of dominos but the mechanism was different. It was not that A owed money to B so when A failed, B failed. More like A owed money to B who owed money to C. C got nervous and refused to roll over loans so now B had to refuse to roll over loans with the result that A (Lehman) failed. The direction of transmission is different but the underlying cause is the same: (1) large inter bank loans of hot money; and (2) the poor quality of A's assets and equity.

    1. Curiosity:

      Is the U. S. legal system set up to resolve circular credit claims? For instance bank A borrows from bank B who borrows from bank C who borrows from bank A. Money flows in a circle in the lending process and then flows in a counter circle in the repayment process - until it stops flowing.

      Bankruptcy proceedings are normally handled as creditor versus lender, but how does a legal system handle circular defaults - bank C can't repay bank B who then can't repay bank A who then can't repay bank C?


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.