Skip to content

LIBOR and Lilliput

23 April 2013

Both Gary Gensler and Martin Wheatley, who as today’s Financial Times reports take opposite views on LIBOR (“Libor must be scrapped” and Lombard), are right. Gensler, head of the US Commodity Futures Trading Commission, thinks the index is completely discredited, while Wheatley, head of the Financial Conduct Authority, doesn’t want to scrap it before a proper alternative has emerged. Lombard likens this to the feud between the Big-Enders and Little-Enders, noting that in the end everyone managed to eat their egg. But, as those of you who have been reading this blog (or my evidence to the Wheatley review) will know, there is a way to reconcile these conflicting views. It will also make a major contribution to stabilising banks and giving them a chance of survival without the implicit taxpayer support on which, five years after the crash, they remain wholly dependent.

Everyone seems to have lost sight of the inevitable consequences of the increasing dependence on interbank security, where a bank’s most saleable assets are pledged (with a haircut) to secure funding. As more of this happens, the risk to the unsecured depositors increases since the bank’s equity cushion becomes insufficient to protect them. Banks may think their cost of capital is reduced overall but this is because the taxpayers are not paid the risk premium their support for depositors requires. When Government finally explores ways to stop this, the system collapses.

The fix is simple. Stop banks borrowing from one another on secured terms. Immediately the traditional LIBOR market will be reestablished and quotes can reflect reality. But the ancillary benefits are even more important. With a negative pledge, the market would have to take over the regulators’ job (which they have performed so ineptly) of monitoring the credit health of counterparties: other banks will do this more efficiently. Resolution with pari passu treatment of all creditors will provide confidence that depositors in a reasonably well capitalised bank will not lose, while under the current system no sensible person can deposit money without security or a Government (implicit or explicit) guarantee. That is the unsustainable iceberg of which the LIBOR scandal is only the visible tip.

And you know what you have to do with eggs that float.


From → Finance

  1. LIBOR rates are based on unsecured funding – the actual market rates at which banks lend to each other without security. The problem was that those rates were self-reported. I’m not sure how preventing other forms of funding would help. Depositors still can’t assess asset quality, and without access to secured funding there will be more risk of rapid deterioration and bank runs. One reason for banks holding liquid government bonds is so that they can quickly access cheap secured funding even in difficult market conditions. Your proposal would remove that possibility, forcing them to sell bonds outright into a falling market. Once sold, those bonds will still be unavailable to other creditors if they default, so the loss to creditors is unchanged, but arguably the probability of default is higher.

  2. I understand the point – Andy Haldane also thought that it was “sometimes convenient” to allow secured borrowing. But it’s a curate’s egg. Once depositors cease to have the benefit of a negative pledge or an implicit guarantee they are mad to keep funds unsecured with a bank. And everone can’t be secured when the leverage ratio goes above the reciprocal of the collateral haircut percentage. There are better answers to the liquidity problem (such as BoE lines) rather than ones which inherently jeopardise the solvency. When I started in banking 30 years ago no bank ever gave security. Of course bankers weren’t paid so much either.

  3. But a negative pledge doesn’t prevent outright sale. I’m not sure I agree that depositors are in a worse position because a bank has engaged in a repo (sale with obligation to repurchase) rather than simply sold the assets outright. Given that the assets will be subject to haircut, there will typically be additional cash returned by the lender to the borrowing bank in the even of the borrower’s default. BoE lines are themselves secured (& Bagehot’s classic statement of the role of central bank as lender of last resort emphasised need to lend only at penal rates against good collateral). And whilst secured borrowing has become more common as repo market has expanded, it isn’t new. The repo market – which was always mainly inter-bank – is about a century old.

    I think the underlying problem – which you raise in another post – is that it’s unreasonable to expect retail depositors to monitor the credit quality of banks. Therefore deposit insurance is required, which it’s right for banks to pay for based on their deposit base and credit quality. I don’t think that preventing secured borrowing makes asset quality better (given that across the financial sector secured lenders and secured borrowers net out – the lenders are improving their creditworthiness by taking security). And I don’t think it makes the task of assessing credit quality easier or more palatable for small investors.

    • The first lesson in credit training is never lend unsecured (to any kind of company) when others have security. That’s true whether the counterpary is an industrial company or a bank, and it’s true whether the security is a charge or a repo. The arithmetic as to why is set out in the footnote to my post You’re completely right that we can’t monitor credit risk but the problem is that neither can the regulators particularly when the position changes hugely overnight (repos etc turn over far faster than a portfolio of corporate loans). We’ve ended up with the ludicrous position of massive amounts of assets being rehypothecated multiple times. While in principle selling assets and realising security might yield the same, in practice in any resolution where the parcel has been passed a dozen times, each time with funding breakage and realisation cost indemnities, a far greater part of the asset pool will be eroded when a bank goes bust and all the intermediaries make their deductions before what’s left is handed to the liquidator. It is the invisible chain of costs that makes security so pernicious. And because someone else has security, every other bank needs it. The evil here is the presentation of equity as a cushion to support creditors when in fact it too has dematerialised in supporting haircuts. Just like clause 9 of the Banking Reform bill, this de facto subordination is fatal to any possibility of banks standing up without the HMG support – whether implicit or explicit, and ensure that the Vickers programme will fail.

  4. Even that rule is contextual. A syndicated revolving credit facility to a highly rated company will almost certainly be unsecured, but the same company might have granted security over its fleet of company cars in a leasing agreement, security over specific property under a commercial mortgage, sold its receivables to a factor, securitied cashflows or granted security over a stand-alone project or subsidiary. It’s even greyer for financial corporations. If they sell a bond outright and sell a put option on the bond, then other creditors are no better off than under a repo – in each case the bank receives cash on the bond sale, but the creditors face a potential additional claim against the bank’s assets in default, either from the repo counterparty or the option buyer. Banks have a lot of options for raising liquidity when their credit quality deteriorates – sale of assets, repo, securitisation – any of which might disadvantage creditors. If you stop banks from grating security (or trading repos, which are legally outright sale rather than granting security interest) you won’t do anything to stop the other ways in which depositors can be disadvantaged. But you’ll introduce new systemic risk if banks are forced to lend unsecured to each other, and the short term impact is likely to be freezing weaker firms out of the market entirely.

    The chain of rehypothecation can be de-risked by settling for cash rather than physical assets (e.g. in event of repo default, a net claim is presented rather than a demand for physical settlement – the asset is sold into the market rather than put back to the seller).

    • But all of these just reinforce my point: no one in their right mind would deposit unsecured with a bank that is allowed to do this, except in the context of implicit state support. That is why the Vickers programme will fail. And you’ve illustrated my other point in evidence to these commissions, viz. that the ring fencing rules, once written down, will be easily circumvented.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: