Northern Rock has, for the most part, slipped from the attention of most Americans, and that’s a mistake, because the bank’s failure and resoluton is a useful object lesson. Admittedly, England has a different bank regulatory regime than the US. The biggest difference, until recently, was that retail deposits were only partially guaranteed and it took a relatively long time for depositers to get their insurance payments from the government. That, plus England’s tough insolvency laws, virtually guaranteed a bank run.
Nevertheless, Northern Rock provided the specter of regulators operating without a playbook. As LSE professor Willem Buiter calls it, they did a good job on their remedies for the asset side of the balance sheet, but were way too generous on the liability side. That’s a pattern we are likely to see repeated if there is another financial panic: overreacting and overcorrecting.
Buiter also argues that had the Bank of England had a discount window that was more like the Fed’s, the crisis could have been averted. I’m not at all certain about that. The mere use of the discount window is a sign that an institution is in trouble, and banks in the US are very loath to use it. With England’s former deposit regime, knowledge of the use of what are essentially emergency facilities might well have provoked a run. It thus seems a further refinement is that regulators need to delay disclosure of the use of the discount window by a very large amount of time, at least a month and better yet two or three months. By that time, the borrower will have either gotten past its stress period or have perished.
John Kay of the Financial Times proposed a different remedy, namely, more generous deposit insurance combined with provisions to assure that guaranteed deposits aren’t used to fund risky activities. Banks might, similar to a suggestion made by McKinsey’s Lowell Bryan, be required to use guaranteed deposits to invest only in segregaed, safe assets.
But Kay and Buiter are ultimately worried about the smae issue: making sure central banks don’t wind up bailing out bank shareholders.
Two announcements have been made during the past couple of days about official support for Northern Rock – one relating to the asset side of its balance sheet and one relating to the liability side. As regards its assets, Northern Rock is now being provided with additional facilities enabling it to borrow through the Bank of England on a secured basis against all of its assets, rather than just against prime mortgage collateral as was the case up to this announcement. As regards its liabilities, the government’s guarantee of its deposits has now been extended to included new deposits.
The asset-side measure makes good sense and brings the Bank of England’s lender of last resort policy closer to what I have advocated for a while. The liability-side measure is likely to compound the earlier mistake.
By accepting the bulk of Northern Rock’s assets as collateral for borrowing from the Bank of England through the Liquidity Support Facility that was purpose-built for the Northern Rock bail-out, the Bank of England is getting close to turning the Liquidity Support Facility into something approximating the Federal Reserve System’s (primary) discount window. The Bank of England’s own discount window, its Standing lending facility, is a pale and anaemic shadow of the Fed’s discount window, because it only accepts extremely high-grade and already utterly liquid securities as collateral. All the Bank of England does at its Standing lending facility is maturity transformation. It exchanges long maturity (duration) liquid assets for very short maturity (duration) liquid assets. It does therefore not provide liquidity in any meaningful sense.
The Fed can, provided it decides that exceptional circumstances prevail, accept at its discount window as collateral absolutely anything it deems fit. When the (private) Bank of New York (back in the 70s I believe) needed to access the discount window of the New York Fed overnight because of some technical glitch (I think they had to borrow $23 bn), they offered as collateral the entire bank, including the building and the furniture.
If an asset can be valued, it should, properly valued and subject to the appropriate haircut, be acceptable as collateral at the discount window. The central bank should insist on sufficient ‘over-collateralisation’ (in addition to the penalty rate it charges for discount window borrowing) to make sure that the tax payer can expect to benefit from the transaction.
If the Bank of England had operated a similar sensible policy at its discount window in August and September 2007, there would have been no need to create the Liquidity Support Facility the Bank dreamt up for Northern Rock….
Of course, the Fed then went and rather spoilt it, by reducing the spread of its discount rate over its policy rate (the Federal Funds target rate) from 100 basis points to 50 basis points; this is pure pandering to the profits of those institutions that are already able and willing to borrow at the discount window; it would have made more sense to raise the discount rate spread over the policy rate by 50bps (to 150 basis points) to underline the Fed’s commitment to a Bagehotian lender of last resort model: lend freely (against collateral that would be good during normal times, but may have become illiquid during turbulent market conditions) but at a penalty rate.
While I am happy about the actions of the Bank vis-à-vis the asset side of Northern Rock’s balance sheet, I am appalled at the Chancellor’s decision to extend the deposit guarantee at Northern Rock to new deposits. This encourages Northern Rock to try to attract new deposits using above-market deposits rates, as long as these are below the penalty rate charged on borrowing from the Liquidity Support Facility. What is especially outrageous about both the old and the new guarantee is that it covers not only retail deposits, but also wholesale deposits and most unsecured lending to Northern Rock.
Why should the unsecured wholesale creditors of Northern Rock get any protection at all? There is no social justice (widows and orphans) argument to support this intervention, nor an efficiency argument – the wholesale creditors to Northern Rock should be expected to be able to pay the cost of verifying its financial viability. No public purpose is served by subsidising, through ex-post insurance, the ‘rate whores’ that are likely to make up the bulk of the wholesale creditors of Northern Rock. Municipalities, charities and professional and institutional investors that were happy to pocket the slightly above-market interest rates offered by Northern Rock should not be able to dump the default risk (whose anticipation/perception was the reason for the higher rates) on the tax payer.
In its statement introducing the deposit guarantee, the Treasury said “Since it would otherwise be unfair to other banks and building societies, the arrangements would not cover any new accounts set up after 19 September, other than re-opened accounts as set out above.” Apparently, it now is no longer unfair or it does not matter that it is unfair. The Treasury statement says that Northern Rock will pay a fair price for the guarantee. We shall see. No doubt a small army of mathematically gifted Treasury civil servants are busy pricing the contingent claim represented by the deposit guarantee for Northern Rock. If the customary lack of openness and transparency of the Treasury prevail, we will never get the information to judge whether Northern Rock paid a fair price for the guarantees extended by the state to its creditors.