Thursday, November 8, 2018

Europe's Banks

My visit to Europe resulted in many interesting conversations. There was a stark contrast between the complex regulatory vision of formal presentations and papers, and the lunch and coffee discussion reflecting experience of people involved in actually regulating banks. They seemed to be quite frustrated by the state of things. Disclaimer: this is all completely unverified gossip, and remembered through a fog of jet lag. If commenters have better facts, I'm hungry to hear them.

Risk weights are ungodly complex, and not many people actually understand them, or the layers of buffers and how they are applied.

Risk weights are suspiciously low. Big banks are allowed to use their own models, calibrated on 10 years of data. That means the data have, now, 10 years of stable growth and very low default. Look, say the banks, our investments are nearly risk free.

"Micro" regulators who look at the specifics of an individual bank are prone to offset the "systemic" and "macro-prudential" efforts. Look, say the banks, we have to fulfill all these macro-prudential rules, give us a break. Regulators do.

The financial regulatory community has been preoccupied with writing reports about one thing after another. Meanwhile, the elephant remains in the room:  Italy may default or leave the euro.

Italian banks remain stuffed with Italian government bonds. I learned some new words for this: a "doom loop." If the government defaults, the banks go with it.  Some smaller foreign banks still have large investments in Italian bonds. Another new word: "Moral suasion," governments encouraging banks to buy a lot of their bonds.  I imagine the Godfather had more colorful words for it. On the other hand, Italian banks are reportedly happy for the moment, since as long as Italy doesn't actually default, they make a bundle from high interest rates. Government debt is still treated with low or no risk weights.


In case it isn't obvious here's the problem. A sovereign default is bad enough. But if the banks are stuffed with government debt, then a sovereign default brings down the banking system. Depositors lose their shirts, and the banks, who know how to distinguish good from bad borrowers, are shut down. A calamity becomes a catastrophe. And an economy with failing banks will be bringing in a lot less tax revenue and more likely to default.  Government debt in a currency union without banking union is a singularly bad investment, because as currently construed governments give deposit insurance (explicit or implicit).

All this is obvious to anyone looking at it, and leads to a big sigh about "political pressure." 10 years on, and Europe can't quite bring itself to say that sovereign debts are risky. Understandably. This is a club of equals, and it's awfully hard to say that some debts are better than others.

But this elephant has been careening around the room for 10 years. There was a Greek crisis which should have gotten some attention!

Some want a full banking union, breaking local bank regulation and allowing large transnational  banks to operate fully, breaking the doom loop. Some want full fiscal union to go with monetary union. The current model, pressure from the rest of Europe for governments not to borrow so much, and thus to never face a potential sovereign default, has clearly failed.

In my view, monetary union without fiscal union works fine, so long as we all understand that governments can default, and their debt should be treated just as risky (and sometimes junk) corporate debt on bank balance sheets. And, of course, if capital requirements were doubled, tripled, or more, so that banks could sail through a sovereign default, the problem would solve itself.

It occurs to me that simply removing risky local government debt from banks would go a long way to solving the problem. Defaultable government debt should be held via floating-NAV mutual funds, not via bank accounts.

Additionally, there is a big kerfuffle going on that Italy's central bank owes Germany's a lot of money.   Italians see this coming, and there is a lot of capital flight out of Italy. When an Italian writes a check from an Italian bank to buy an apartment in Germany, the money flows from Italian bank to Italian central bank, to German central bank, to German bank. Except the Italian central bank essentially makes a promise to pay rather than actually paying. Italy is basically expanding government debt in this way. I don't totally understand it, nor did most people I talked to about it, and there is a wide disagreement whether this is another debt or just an accounting glitch. Still, that most people at a financial regulation conference know this is a big problem and nobody is quite sure what it means is telling.

With this background of lunchtime and coffee conversations, the written products of the financial regulation community have a surreal quality. The illusion of technocratic competence is always present in bank regulation discussions, but even more stark with this backdrop.

Look for example at the website of the Finanical Stability Board and the issues it thinks are important. As one example, the summary of FSB priorities for the Argentine G20 Presidency. It starts well enough, "Vigilant monitoring to identify, assess and address new and emerging risks." But what's the number one such risk? You would think, in honor of the Argentine presidency, with Italy the number one topic of conversation at lunch, and with who knows who owes who what in China, it would be "sovereign risk." Nope. Crypto-assets is number 1: "The FSB will identify metrics for enhanced monitoring of the financial stability risks posed by crypto-assets and update the G20 as appropriate." Then,
Disciplined completion of the G20’s outstanding financial reform priorities....During the course of the year the deliverables to the G20 will include the following areas: the correspondent banking Action Plan including improving the access of remittance providers to banking services; a toolkit for firms and supervisors on the use governance frameworks to reduce misconduct in the financial sector; leverage measures for investment funds to support resilient market-based finance; guidance on financial resources available to support central counterparty (CCP) resolution to deliver resilient and resolvable CCPs; a cyber security lexicon to support consistency in the work of the FSB, standard-setting bodies, authorities and private sector participants; and the private sector-led Task Force on Climate-related Financial Disclosures’ report on voluntary implementation of its recommendations to highlight good practice and foster wider adoption.  
Sovereign risk is not mentioned once in this document. And I did not find it anywhere on the FSB website.

If you read between the lines, there is a very worthy reaction to this tendency to produce complex hot air that changes with each presidency:
Pivoting to policy evaluation to ensure the reform programme is efficient, coherent and effective. The FSB is increasingly pivoting away from design of new policy initiatives towards dynamic implementation and rigorous evaluation of the effects of the agreed G20 reforms. 


16 comments:

  1. Valter Buffo, Recced, Milan, ItalyNovember 8, 2018 at 11:19 AM

    A (brief) comment from Italy about Italy.
    "It occurs to me that simply removing risky local government debt from banks would go a long way to solving the problem. Defaultable government debt should be held via floating-NAV mutual funds, not via bank accounts." And you are absolutely right in saying so: but, will never happen. Reason? The situation is intended to preserve the corporate balance of powers in the banking sector.
    "When an Italian writes a check from an Italian bank to buy an apartment in Germany, the money flows from Italian bank to Italian central bank, to German central bank, to German bank. Except the Italian central bank essentially makes a promise to pay rather than actually paying. Italy is basically expanding government debt in this way.". You are again right in saying so: notice that this process is "oversighted" by the ECB, and is an "alternative" way to recapitalise the banking sector (not just in Italy, notice) at the same time mantaining the above mentioned balance of powers, as well as most of the existing inefficiencies of the business practices.
    To sum up: here you have further indications that the whole season of QEs has been nothing more that a risky and inaccurate policy to freeze up the (existing) banking system as it is. This happened in the Eurozone, and in the US as well: and it happened precisely when the economy as a whole were in need of new players, new practices, new managements and new business strategies.
    As I wrote before here (and subsequently I have read from Nassim Taleb) QE has been nothing more than a massive (and indiscriminate) transfer of market/credit/default risk from the private sector to the public, and for free. A bonanza for all inefficiencies: no (established) player, amd their friendship, will pay for any mistake. The policy response to the run was: make life easier to those that generated the instability that originated the run.
    Newcomers have no access to the "easy financing": they must be kept at bay.
    And there are still those who go around asking why productivity remains stuck at the lowest level.

    ReplyDelete
  2. This isn't really a new problem. I am old enough to remember when the Reagan Administration had to bail out Chase Bank (via the IMF) to the tune of $8.4 Billion back in 1984 when that was still a lot of money.

    This was because Chase and others had invested in dodgy Latin American sovereign debt that...paid high interest rates. And when those countries defaulted, it somehow became a US Government responsibility to make them well!

    The only difference between that situation and the Italian and Greek ones is that no one wants to admit that Greece and Italy are effectively third-world countries from a fiscal point of view.

    Oh, and of course, the whole "Nice bank you got there, be a shame if anything were to happen to it!" business of <*cough*> "moral suasion."

    ReplyDelete
  3. Risk weighting is a total waste of time. The objective is to maximise the amount of maturity transformation banks can do for a given amount of risk. Put another way, the object of the exercise is to maximize the amount of liquidity / money creation that private banks can do for a given amount of risk.

    The monster flaw in that is that central banks can create limitless amounts of money at no cost and at no risk. So why have private banks create liquidity / money at all? Full reserve banking just abolishes private banks’ ability to create money.

    As the Nobel laureate economist Merton Miller put it in reference to Irving Fisher’s full reserve proposals, “Think how much national economic welfare could rise under Fisher’s banking scheme when thousands of no longer needed bank regulators (and hundreds of academic banking economists) find themselves forced at last to seek more socially productive lines of economic activity.”

    ReplyDelete
  4. Throwing out a question for anyone who knows.

    This point from G20 list
    guidance on financial resources available to support central counterparty (CCP) resolution to deliver resilient and resolvable CCPs
    sounds like it might be cryptic policy speak for the issue of
    Additionally, there is a big kerfuffle going on that Italy's central bank owes Germany's a lot of money.Italians see this coming, and there is a lot of capital flight out of Italy. When an Italian writes a check from an Italian bank to buy an apartment in Germany, the money flows from Italian bank to Italian central bank, to German central bank, to German bank. Except the Italian central bank essentially makes a promise to pay rather than actually paying. Italy is basically expanding government debt in this way. I don't totally understand it, nor did most people I talked to about it, and there is a wide disagreement whether this is another debt or just an accounting glitch. Still, that most people at a financial regulation conference know this is a big problem and nobody is quite sure what it means is telling.

    Is this correct? Or is 'CCP resolution' something else?

    ReplyDelete
  5. I feel like the outsider from Mars making this comment but here goes:

    The People's Bank of China has for decades purchased bad loans and defaulted bonds. The short story is the central bank printed money and bought bad debts. China has a 40-year track record of growth, and is now below a modest 3% inflation target.

    Yes, moral hazard. But a vital consideration is keeping a banking system liquid and solvent.

    The Greek GDP is still 25% below the level that it was 2008.

    There are times to take principled ideological positions. Then there are other times when discretion is the better part of valor.

    ReplyDelete
    Replies
    1. But under the Cochrane system (which comes to the same as full reserve banking I think), it's plain impossible for a bank to go insolvent. If a bank makes silly loans and the value of its stock of loans sinks to say 80% of book value, all that happens is that the value of its shares fall to about 80% of book value.

      It is wholly unjustified for a government or central bank to treat banks in any sort of favorable way compared to other industries. But under the existing bank system, that’s what they are forced to do periodically: e.g. witness the trillion or so dollars that the Fed loaned to sundry banks at the height of the crises. Don’t you wish you or your business could borrow at zero percent?

      Delete
    2. I am reminded of Mao's reply when asked about the impact of the French Revolution, "It's too early to tell." First, China started from zero (essentially) 40 years ago, so a lot of catch-up is already built in to their growth rates. But more importantly, I don't think you can make any determination about the efficacy of their "bad loan" policy just yet. These chicks have a habit of coming home to roost eventually.

      Delete
  6. Risk weights are not terribly complex as these things go. Yes there are alot of different things to keep track of depending on the asset class and whether you are using the standardized or advanced approach, but they are probably still simpler than many custom risk models developed by banks. Even under the advanced approach the calculation of input parameters like long-run probability of default often involve nothing much more complicated than simple averages. Maybe they are too complex from a regulatory point of view and could be replaced by higher capital across the board, but even if regulators got rid of them banks would still likely be building models to estimate the capital consumption of different products.

    The comment about banks using just 10 years of data strains does not ring true, or sounds like something that happened at one bank and is now the story everyone likes to tell. Not only would it get extreme pushback during an internal model review or a regulatory exam, but it assumes that management has no incentive whatsoever to maintain reasonable levels of capital. Moreover, no one has forgotten the 2008-09 experience, and data on that period is readily available, so investors would look askance at banks that are running the same business as before but have now suddenly lowered their capital levels relative to something like face value of assets.

    Speaking of Italy, a key problem is that it is hard to allocate resources within bank risk departments toward this type of big-picture thinking, which is largely concentrated among client-facing researchers. The vast majority of the risk department is involved in fairly mechanical tasks related to data management, routing reporting, and mathematical modeling that is often some variant of simple curve fitting. Less burdensome oversight coupled with higher across-the-board capital requirements might free up resources to study more high-level issues.

    ReplyDelete
  7. "That means the data have, now, 10 years of stable growth and very low default."

    This might be true for the US, but certainly not for Europe:

    2008 GFC
    2010/2011 European Debt Crisis
    2012 Spanish Real Estate Crisis
    2011 - 2013 Eurozone Recession
    2016 Italian Banking Crisis

    ReplyDelete
  8. Hi John,

    When should we expect your FTPL book to be published? I'm really enjoying the draft and I'm looking forward to reading the final version.

    Piotr

    ReplyDelete
  9. "Except the Italian central bank essentially makes a promise to pay rather than actually paying. Italy is basically expanding government debt in this way. I don't totally understand it, nor did most people I talked to about it, and there is a wide disagreement whether this is another debt or just an accounting glitch."

    Its not a glitch its a debt however , in the procedure, the Italian banking system has an asset to match the debt, and that asset could be an Italian Government Bond in the context of the Post.

    ReplyDelete
  10. Are these the TARGET2 balances? When an Italian writes a check from an Italian bank to buy an apartment in Germany, the money flows from Italian bank to Italian central bank, to German central bank, to German bank. Except the Italian central bank essentially makes a promise to pay rather than actually paying.

    ReplyDelete
  11. Are these TARGET2 balances? When an Italian writes a check from an Italian bank to buy an apartment in Germany, the money flows from Italian bank to Italian central bank, to German central bank, to German bank. Except the Italian central bank essentially makes a promise to pay rather than actually paying.

    ReplyDelete
  12. About the fact that Italy's central bank owes the German central bank a lot of money: the way I understood this, in this kind of operations (called Target2 system) both count as branches of ECB. The money is not due to be paid ever, at least as long as both countries stay in the Euro. In this sense I guess some say it is "an accounting glitch": it should be more about tracking where the money is flowing, and if there are financial flow imbalances created by having a "fixed exchange rate" between Euro countries. About what happens if Italy ever left the Euro, opinions diverge I think. Italy may have to actually pay this amount.

    Disclaimer: I'm not an economist, and not working in finance. Just an italian compulsive economic blog reader. So I am just writing here what I read elsewhere....

    ReplyDelete
  13. They are target 2 balances.
    It is a application of the system called contra accounting where parties to credits and debts keep a running tally of them.

    ReplyDelete
  14. thank you Professor Cochrane
    I believe your comments, especially the "coffee discussions", are right on target.
    You seem to understand much more about the Eurozone situation than most European financial pundits.
    Please keep on commenting.

    ReplyDelete

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.