resilience, not stability

The Public Deposit Option: An Alternative To “Regulate and Insure” Banking

with 31 comments

Many economists want to turn back the clock on the American economic system to that of the 50s and 60s. This is understandable – the ‘Golden Age’ of the 50s and 60s was characterised by healthy productivity growth, significant real wage growth and financial stability. Similarly, many commentators see the banking system during that time as the ideal state. In this vein, Amar Bhide offers his solution for the chronic fragility of the financial system:

governments should fully guarantee all bank deposits — and impose much tighter restrictions on risk-taking by banks. Banks should be forced to shed activities like derivatives trading that regulators cannot easily examine…..Banks must therefore be restricted to those activities, like making traditional loans and simple hedging operations, that a regulator of average education and intelligence can monitor.

There are a couple of problems with his idea – for one it may not be possible to effectively regulate bank risk-taking. On many previous occasions, I have asserted that regulations cannot restrain banks from extracting moral hazard rents from the guarantee provided by the state/central bank to bank creditors and depositors. The primary reason for this is the spread of financial innovation during the last fifty years that has given banks an almost infinite variety of ways in which it can construct an opaque and precisely tailored payoff that provides a steady stream of profits in good times in exchange for a catastrophic loss in bad times. As I have shown, the moral hazard trade is not a “riskier” trade but a combination of high leverage and a severely negatively skewed payoff with a catastrophic tail risk.

Minsky himself understood the essentially ephemeral nature of the financial system of the 50s from his work on the early stages of the process of financial innovation that allowed the financial system to unshackle itself from the effective control of the central bank and the regulator. As he observes:

The banking system came out of the war with a portfolio heavily weighted with government debt, and it was not until the 1960s that banks began to speculate actively with respect to their liabilities. It was a unique period in which finance mattered relatively little; at least, finance did not interpose its destabilizing ways……The apparent stability and robustness of the financial system of the 1950s and early 1960s can now be viewed as an accident of history, which was due to the financial residue of World War 2 following fast upon a great depression.

Amar Bhide’s idea essentially seeks to turn back the clock and forbid much of the innovation that has taken place in the last few decades. In particular, derivatives businesses will be forbidden for deposit-taking banks. This is a radical idea and one that is a significant improvement on the current status quo. But it is not enough to mitigate the moral hazard problem. To illustrate why this is the case, let me take an example of how as a banker, I would construct such a payoff within a “narrow banking”-like mandate. Let us assume that banks can only take deposits and make loans to corporations and households. They cannot hedge their loans or engage in any activities related to financial market positions even as market makers, and they cannot carry any off balance-sheet exposures, commitments etc. Although this would seem to be a sufficiently narrow mandate to prevent rent extraction, it is not. Banks can simply lend to other firms that take on negatively skewed bets. You may counter that banks should only be allowed to lend to real economy firms. But do we expect regulators to audit not only the banks under their watch but also the firms to whom they lend money? In the first post on this blog, I outlined how the synthetic super-senior CDO tranche was the quintessential rent-extraction product of the derivatives revolution. But at its core, the super-senior tranche is simply a severely negatively skewed bond – a product that pays a small positive spread in good times and loses you all your money in bad times. There is no shortage of ways in which such a negatively skewed payoff can be constructed by simple structured bank loans.

What the synthetic OTC derivatives revolution made possible was for the banking system to structure such payoffs in an essentially infinite amount without even going through the trouble of making new loans or mortgages – all that was needed was a derivatives counterparty. Without derivatives, banks would have to lend money to generate such a payoff – this only makes it a little harder to extract rents but it still does not change the essence of the problem. Even more crucially, the potential for such rent extraction is unlimited compared to other avenues for extracting rent. If the state pays a higher price for an agricultural crop compared to the market, at least the losses suffered by the taxpayer are limited by physical constraints such as arable land available. But when the rent extraction opportunity goes hand in hand with the very process that creates credit and broad money, the potential for rent extraction is virtually unlimited.

Even if we assume that rent extraction can be controlled by more stringent regulations, there remains one problem. There is simply no way that incumbent large banks, especially those with a large OTC derivatives franchise, can shed their derivatives business and still remain solvent. The best indication of how hard it is to unwind complex derivatives portfolios was the experience of Warren Buffett in unwinding the derivatives portfolio which he inherited from the General Re acquisition. As Buffett notes, unwinding the portfolio of a relatively minor player in the derivative market under benign market conditions and no internal financial pressure took years and cost him $404 million. If we asked any of the large banks, let alone all of them at once, to do the same in the current fragile market conditions the cost of doing so will comfortably bankrupt the entire banking sector. The modern TBTF bank with its huge OTC derivatives business is akin to a suicide bomber with his finger on the button that is holding us hostage – this is the reason why regulators handle them with kid gloves.

In other words, even if our dream of limited and safe banking is viable we have a ‘can’t get there from here’ problem. This does not mean that there are no viable solutions but we need to be more creative. Amar Bhide makes a valid point when he argues that “Why not also make all short-term deposits, which function much like currency, the explicit liability of the government?” But the solution is not to allow private banks to reap the rents from cheap deposit financing but to allow each citizen and corporation access to a public deposit account. The simplest implementation of this would be a system similar to the postal savings system where all deposits are necessarily backed by short-term treasury bills. If the current stock of T-bills is not sufficient to back the demand for such deposits, the Treasury should shift the maturity profile of its debt until the demand is met. In such a system, there would be no deposit insurance i.e. all investment/deposit alternatives except for the state system will be explicitly risky and unprotected.

One criticism of such a system would be that the benefits of maturity transformation would be lost to the economy i.e. unless short-term deposits are deployed to match long-term investment projects, such projects would not find adequate funding. But as I have argued and the data shows, household long-term savings (which includes pensions and life insurance) is more than sufficient to meet the long-term borrowing needs of the corporate and the household sector in both the United States and Europe.

The “regulate and insure” model ignores the ability of banks to arbitrage any regulatory framework. But the status quo is also unacceptable. However the system is sufficiently levered and fragile that allowing market forces to operate or simply forcing a drastic structural change upon incumbent banks by regulatory fiat implies an almost certain collapse of the incumbent banks. Creating a public deposit option is the first step in implementing a sustainable transition to a resilient financial system, one in which instead of shackling incumbent banks we separate them from the risk-free depository system.


Note: My views on this topic and some other related topics which I hope to explore soon have been significantly influenced by uber-commenter K. For a taste of his broader ideas which are similar to mine, try this comment which he made in response to a Nick Rowe post.

Bookmark and Share

Written by Ashwin Parameswaran

January 5th, 2012 at 12:18 pm

31 Responses to 'The Public Deposit Option: An Alternative To “Regulate and Insure” Banking'

Subscribe to comments with RSS or TrackBack to 'The Public Deposit Option: An Alternative To “Regulate and Insure” Banking'.

  1. i think i understood 50’s, 60’s banking .. but i don’t have a clue what the word even means in 2012.


    5 Jan 12 at 12:29 pm

  2. As you know, I obviously fully agree. Great stuff!


    5 Jan 12 at 2:03 pm

  3. K – Thanks! Your comments at Nick Rowe’s blog and here have been very illuminating on this topic.I’ve added a link to a comment of yours as a token of my appreciation.


    5 Jan 12 at 2:25 pm

  4. Gregory – Banking can be as complex and as obtuse as it wants, so long as it is not combined with insured public deposits.


    5 Jan 12 at 2:26 pm

  5. Ashwin –

    interesting thoughts.

    If I understand correctly, what you are saying is that an individual depositor can deposit the savings money into the T-bills market (short duration and highly liquid) and therefore prevent the bank from using his savings (which he prefers to keep it in liquid form) as reserves for the bank to lend long-term illiquid assets. Am I right?

    If the individual does want to earn more return, he would have to deposit money with the bank, which the bank can use to lend. Banks will lend at a rate of interest greater than the interest they pay on the deposits and make profit on this spread.

    But this still doesn’t solve the duration mismatch problem correct? Maturity transformation done by the bank may not necessarily align with the maturity requirement of the individual, therefore you end up with a duration mismatch problem.

    I see the public deposit option as a way to mitigate the damage banks can do by ‘borrowing short to lend long’, but it doesn’t solve the problem entirely.


    5 Jan 12 at 5:29 pm

  6. “what you are saying is that an individual depositor can deposit the savings money into the T-bills market (short duration and highly liquid) and therefore prevent the bank from using his savings (which he prefers to keep it in liquid form) as reserves for the bank to lend long-term illiquid assets. Am I right?” Exactly!

    I’ve explored the maturity transformation in a few posts. In the last post I made the case that the household sector’s demand for long-term savings is sufficient to meet long-term investment needs. The data in that post backs up this case.

    If banks (or anyone else) wants to maturity transform anyway, it is very difficult to stop this. But we can make it clear that this is a risky activity which will not be backstopped by the taxpayer. Right now we cannot do this because there is no risk-free deposit option available other than via the banking sector.


    5 Jan 12 at 5:45 pm

  7. Wow! Thank you, Ashwin!


    I think what you are saying is that it doesn’t solve *society’s* problem which is that we want liquid assets and long term liabilities. Nick Rowe said the same thing to me at one point on his blog. What I said to him is that people *want* to drive Ferraris and pay Skoda prices. Me too! But that’s not where supply meets demand. Banks can’t perform that magic trick. All they can do is to put lipstick on the pig and call it a Ferrari (to really mix the metaphors!).


    5 Jan 12 at 6:18 pm

  8. /* I think what you are saying is that it doesn’t solve *society’s* problem which is that we want liquid assets and long term liabilities*/

    K – that’s not what I’m saying. Even with a public deposit scheme (for liquid savings), an individual depositing his remaining portion (which he’s willing to lend) at the bank should be able to choose the time-preference element of his return.

    It is not the bank’s prerogative to choose the time-preference of the individual, because the banks will always make the choice that gives them greater rate of return (at a higher level of risk naturally). Such an action by the bank may not be acceptable to the individual.

    If each individual made this choice of time-preference, then it distributes the risk appetite over more entities as opposed to a single for-profit commercial bank.


    5 Jan 12 at 8:37 pm

  9. Ashwin,
    Liability insurance — implied or formal — begets rent-seeking. I can see how your proposal prevents that activity. However, it also leaves intact the problem that insurance addresses in the first place: deadweight losses from bank runs. The non-public portion of the financial system would still be vulnerable to such runs. This means that the public would end up insuring those assets after the fact, which is perhaps a worst outcome.

    I think the real question is, do we want to prevent the modern version of bank runs? If the answer is “yes”, then insurance has to cover the bulk of financial system liabilities, no matter how large the moral hazard. If we minimize financial risk, it is to the benefit of rent-seekers; if we allow it, it can produce large losses for middle class savers even if the wealthy experience yet larger losses. Glass Steagall and the FDIC, together, were designed to firewall middle class savers from such losses while allowing investment banks to fail. Can we go back to that regime? You seem to argue the answer is, “no”, because banks would just lend to hedge funds. I think eliminating “finance company” lending by banks might prevent this.

    David Pearson

    5 Jan 12 at 9:28 pm

  10. David – bank runs occur due to the presence of maturity transformation. If the need for short-term deposits is met by short-tenor T-bill issuance by the govt, then the data shows that no more maturity transformation is needed to meet the long-term investment needs for the economy.

    In such an environment, any maturity transformation attempted by a bank or any other firm is simply a risk taken on by that firm and its owners and creditors. If it fails, tough luck! I don’t see any reason to prevent a bank run in such a scenario. If middle class savers take on conscious risks in search of higher yields despite the presence of a risk-free deposit alternative, then why should the taxpayer have to bear the burden of bailing them out?


    5 Jan 12 at 10:06 pm

  11. ephemeral_reality:

    once we institute a “safe deposit” option, the rest of the lending market is just like any other market. At a certain price, every risk-maturity combination should clear depending upon the investment opportunities available and savers’ time preferences. I have nothing against maturity transformation in this domain so long as everyone understands the risk of failure.


    5 Jan 12 at 10:31 pm

  12. Just a brilliant, succinct, summary,

    “On many previous occasions, I have asserted that regulations cannot restrain banks from extracting moral hazard rents from the guarantee provided by the state/central bank to bank creditors and depositors. The primary reason for this is the spread of financial innovation during the last fifty years that has given banks an almost infinite variety of ways in which it can construct an opaque and precisely tailored payoff that provides a steady stream of profits in good times in exchange for a catastrophic loss in bad times. As I have shown, the moral hazard trade is not a “riskier” trade but a combination of high leverage and a severely negatively skewed payoff with a catastrophic tail risk.”

    Michael Strong

    6 Jan 12 at 3:46 am

  13. Ashwin –

    May be my point is not very clear. Do banks actually take savers’ time preference into account when they sweep their savings deposits into reserves that they can use for lending?

    Let’s run with an example. Say I want to deposit a CD at a commercial bank for 5-years. The bank can either lend the money for some private party for 5 years at interest (say an auto-loan for example) or buy a government bond that’s 5-year maturity.

    Now – the banks want to maximize their return with the reserves, so technically there’s nothing stopping them from buying a 30-year Treasury bond with the same money (that was given only for 5-year term). This creates a duration mismatch problem.

    Is my concern more clear now?


    6 Jan 12 at 6:57 am

  14. The above post suggests that “each citizen and corporation (should have) access to a public deposit account.” That idea is very similar to the system proposed here:

    See in particular, p.7 section entitled “Step1: Divorcing the payments system….”

    Ralph Musgrave

    6 Jan 12 at 7:24 am

  15. Michael – Thanks. I’m bored of repeating myself so the summary keeps getting shorter and shorter!


    6 Jan 12 at 9:57 am

  16. ephemeral_reality:

    OK I understand your point a lot more clearly now.

    Once we have a risk-free transactional deposit account alternative in place, every other account is clearly flagged as risky. In your example, the bank can certainly buy a 30-year bond with your 5-year deposit and if things turn sour when the deposit is up for payment, the bank and the depositor are at risk of losses due to early liquidation of the asset.

    Some will insist that we need regulations to ban maturity transformation in all sectors but this is an unrealistic aim. What you describe is conducted by many real economy firms as well who borrow short and invest long and gamble on being able to refinance when needed.

    In practise, the investor will likely insist on covenants that protect him against such blatant abuse – a practise that is common in the loan market. The key is that having being flagged as clearly a risky investment, losses can and should be incurred by the parties concerned rather than by the taxpayer.


    6 Jan 12 at 10:09 am

  17. Ralph – Thanks for the link. It sounds like what I have in mind. I’m fairly certain that many others have made similar arguments – I’m just puzzled that this option doesn’t get discussed more often.

    Just a small nit to pick – I’m not sure the account needs to be zero-interest. At the very least, it could certainly pay the overnight rate less charges/costs.


    6 Jan 12 at 10:11 am

  18. Ashwin,
    I sympathize with your desire to let bank speculators fail. However, I am addressing a different question: if bank runs can create depression-like conditions, do we have an interest in preventing them? I think your proposal, while it protects a subset of depositors, may not protect the system as a whole.

    Stepping back, we have a “system” that can coalesce into two equilibria:
    1) rent-seeking stability characterized by the build up of catastrophic sovereign risk; or
    2) instability characterized by periodic bank runs, bank failures and fast-and-deep depressions

    Was the great “banking moderation” of 1946-1986 an aberration, or a third equilibrium? I tend to think the former. In my mind deposit schemes have to consciously choose whether they want 1) or 2) above. Right now we are in denial — trying to choose the 3) of a return to “banking moderation”.

    David Pearson

    6 Jan 12 at 2:07 pm

  19. David – I agree that under these proposals, the system will be less stable but more resilient. IMO the instability will manifest itself as fast and relatively shallow recessions. When natural disturbances in a complex system occur regularly, they tend to be less severe in nature. This is where I differ strongly from Minsky. MInsky thought that every disturbance could trigger a debt deflation – I don’t.

    My poster child example of this principle is the LTCM bailout. If the Fed had not cut rates in late 1998 (at a time when apart from a section of the financial system, the rest of the economy was booming), then we would likely have had a mini-derivatives blowup. One of the major investment banks would likely have gone under then for a much smaller cost than was eventually the case in 2008.

    In a nutshell, regular moderate disturbances rather than the occasional catastrophic one.


    6 Jan 12 at 8:29 pm

  20. Ashwin, Re your “puzzlement” that “this option” is no more widely discussed, here are some possible reasons.

    1. At least 99% of the population are just not interested in, or have not got time to read up different ideas for bank regulation. Page 3 of the Sun and the Daily Mail are more “interesting”. And that goes for those supposedly intelligent university students. I know because along with some friends I’ve tried to get local university students interested. We’ve wasted our time.

    2. It is EXTREMELY difficult getting any new idea across, no matter how simple and beneficial it is. The NHS was strongly opposed by a large proportion of doctors in the UK when the NHS was being set up. And people got their heads chopped off for claiming the Earth went round the Sun a long time ago.

    Ralph Musgrave

    7 Jan 12 at 4:21 pm

  21. Ashwin, I argued for the two account system on the Naked Capitalism blog yesterday (see first comment at the URL below). Perhaps you could add to the debate there.

    Ralph Musgrave

    8 Jan 12 at 10:49 am

  22. Ashwin,
    I was thinking of the pre-War depression experience, where the cycles were deeper and faster. Granted, that was an economy dominated by tradables; however, today the discretionary services sector is taking on the cyclical nature of manufacturing. I guess I was really posing the question of whether the post-War experience was an aberration or a permanent change shift towards stability (i.e. shallower recessions in the presence of financial crises).

    BTW, one thing I’ve been musing over is whether the services sector is just, by its nature, unresponsive to monetary stimulus after a financial crisis. FDR targeted commodity prices and their effect on corporate profits when he decided to exit the gold standard. He, or his advisers, realized the effect that a better terms of trade towards manufacturing would have on the economy. Today, the situation is quite different, as only a small portion of the economy benefits from higher tradables prices.

    David Pearson

    8 Jan 12 at 5:04 pm

  23. Ralph – Thanks for the comments. WIll add a comment at Naked Capitalism.

    David – I obviously think the post-WW2 experience was an aberration. But I also see stability itself as a “bad” – small disturbances are what keep complex systems resilient, evolvable and able to innovate. The end-state of post-WW2 economic evolution is a system without idiosyncratic risk and loaded up on macro risk. The excess of macro tail risk is a problem but so is the absence of idiosyncratic competitive risk.

    On FDR, I’m with you. Moving off the gold standard reduced real rates from high positive levels to near-zero levels. Right now we’re at negative real rates which is a very different situation. Your point on mfg vs services is one I haven’t thought of but even within manufacturing, the adaptability of an economy not wedded to huge sunk cost investments of the Fordist era probably makes it harder to enact systematic stimulus. Any systematic macro policy becomes ineffective once economic agents have adapted to its existence and maybe this adaptation is a lot easier in all sectors of the economy now.


    9 Jan 12 at 12:41 pm

  24. It is essential that we stop the casino. The problem with stopping liquidity is that it can cause a depression. But stopping speculation will cause no such thing. This article is wrong, wrong, wrong. Speculation is a tax on everyone and as Will Rogers once said back in the 30’s, harms the prosperity of everyone.

    Gary Anderson

    9 Jan 12 at 1:33 pm

  25. […] Ashwin Parameswaran and I question whether this can be done. I quote from Parameswaran: […]

  26. When I say the article is wrong, wrong, wrong, I don’t mean that we can’t have state banks or public banks. I am saying only that speculation must be regulated and stopped in it’s tracks. Ellen Brown is pushing for public banks and I support that.

    Gary Anderson

    9 Jan 12 at 5:04 pm

  27. I am wondering if
    How to Regulate Bank Capital
    provides a useful approach. If the banks are financed by convertible debt, then the risk is shifted form the government to private parties willing to take the risk. This seems in principal to limit the governments role for insuring deposits.


    11 Jan 12 at 3:37 pm

  28. Jim – Within the ambit of regulatory options, I believe that CoCos are a good idea. But if we have to go down this route, I tend to prefer that all bank debt is essentially deemed to be convertible.

    What I prefer about the public option idea is that we’re simply adding an option to the menu of choices out there rather than depending upon a complex regulatory solution.


    11 Jan 12 at 5:04 pm

  29. “Convertible,” of course, is a euphemism in this case. A convertible bond is convertible by the holder, not by the issuer (or the regulator). Interest on such bonds really push the limit on what is fairly considered tax deductible. Are those “interest” payments really mandatory? It reeks of a tax scam. Why would I issue prefs if I can issue “converts.” If you want banks to have less default triggering liabilities make them issue stock.


    13 Jan 12 at 9:05 pm

  30. […] have already described the essence of the public deposit option in an earlier post as “a system similar to the postal savings system where all deposits are necessarily backed by […]

  31. […] — by creating a “public deposit” option and getting rid of deposit insurance. Ashwin Parameswaran (via Reihan Salam): But the solution is not to allow private banks to reap the rents from cheap […]

Leave a Reply