May 16, 2008
Can central banks go broke?
Those of you with time on your hands and a interest in spending some it it in a basement being beaten with a rubber hose, may want to take a look at my recent Centre for Economic Policy Research Policy Insight No. 24, “Can Central Banks Go Broke?”
In that paper, I ask whether it matters if a central bank suffers a large capital loss. Can the central bank become insolvent? How and by whom or by what institution should the central bank be recapitalised, if its capital were deemed insufficient? These are relevant questions not just in Zimbabwe and Tajikistan today, but wherever central banks have taken on or may be asked to take on large exposures to private credit risk. This includes the USA, the Euro Area, the UK, Iceland and many other advanced industrial countries.
Special problems arise where private financial institutions (or the central bank and the Treasury itself) have large liquid foreign exchange-denominated liabilities or index-linked debt. Examples of countries that have an internationally active banking system with very large foreign currency-denominated liabilities and assets (including a significant amount of short-term liabilities) include Iceland (whose currency does not have an international reserve currency role) and the UK, where sterling is a much-diminished global reserve currency, with 4.7% of the total global stock of reserves at the end of 2007, against 26.5% for the euro and 63.9% for the US dollar.
The central bank’s lender of last resort and market maker of last resort capacity is diminished if the liquidity needed by the banking system it is responsible for, is foreign exchange rather than domestic currency. The ability of the national sovereign (central bank, Treasury, sovereign wealth funds etc.) jointly to beg or borrow the foreign exchange resources required to provide credible lender-of-last-resort and market-marker-of-last-resort-support to their banking system and financial system, then becomes crucial for the financial viability of the banking sector, including the central bank.
Even if the exposure is domestic-currency denominated, the lender-of-last-resort and market-maker-of-last-resort-role may conflict with maintaining price stability. In that case it is essential that the central bank is backed fiscally by the Treasury, that is, by the tax payer.
Fiscal backing of the central bank by the Treasury/ministry of finance, creates special issues whenever there is significant cross-border banking activity. Lender-of-last-resort and market-maker-of-last-resort-responsibilities are informal and flexible. The range of eligible counterparties/beneficiaries at the discount window, in repos and other open market operations, and for special liquidity facilities that can be created at the drop of a hat, is discretionary and can be varied at will. The presumption that a foreign subsidiary, unlike a branch of a foreign bank, will be supported by let alone bailed out by the central bank of the host country (the country where the subsidiary is registered) is not a firm legal convention.
The potential problems concerning which central bank is responsible - as lender of last resort, market maker of last resort and bailer out of first resort - for which kind of bank (or even non-bank financial institution), is likely to become acute in the Euro Area, once there are banks incorporated as Societas Europaea (SE), which will not have any clear national identity. Perhaps the national central banks of the Euro Area (which one assumes to be backed by their national fiscal authorities) will continue to implement the lender-of-last-resort-role for banks incorporated as SE but domiciled/located in their traditional jurisdiction. Perhaps instead, the ECB assumes lender-of-last-resort and market-maker-of-last-resort-functions directly for SEs. In either case the fiscal backing of the ECB and even that of national central banks is a murky area. Will, say, the Banca d’Italia (backed by the Italian Treasury) bail out an Italian bank that was taken over by a Dutch bank which is now itself owned by a British, a Spanish and a Belgian bank?
The failure to sort out the ambiguities concerning the distribution of the fiscal burden that may arise through bail-outs of banks operating in multiple Euro Area nation states puts a large question mark behind the effectiveness of the Euro Area financial stability arrangements. The Euro Area has proven itself to be capable of handling a banking sector liquidity crisis. The institutional arrangements, including the fiscal burden sharing key, for handling a banking sector insolvency crisis are opaque at best, non-existent at worst.
We must know who would recapitalise the ECB should it suffer a material capital loss, and through what mechanism this would occur. The shareholders of the ECB are the 27 EU national central banks. Only 15 of these national central banks belong to the Eurosystem. Central banks in any case have only limited and qualified deep pockets. We must know how the Euro Area or EU tax payers are lined up to recapitalise the ECB, should the need arise. This is not an issue that should be sorted only when a banking sector solvency crisis hits the Euro Area.
And the answer to the final question is ???…
Posted by: Snori | May 16th, 2008 at 5:06 pm | Report this commentLet’s keep it simple and start the ball rolling with some sort of pro rata metric — pick your own. Of course, this does seem to put the cart before the horse as the much more immediate issue is proper regulation of these entities before the go bust…
Posted by: SS | May 16th, 2008 at 6:11 pm | Report this commentI thought the BIS (Bank of International Settlement) in Basel and the IMF where overlooking central banks and acting as lender of last resort for its members. Then again the same “what if they go bust” question could be asked of these institutions I suppose… Ultimately it is always the tax payers who’s the lender of last resort. Yes, but what if…
Posted by: Christian G. | May 16th, 2008 at 10:06 pm | Report this commentUnfortunately, this is symptomatic of the under-researched problems overhanging the Euro zone, which the markets are for the most part brushing aside. To the recent credit problems (still multiplying in Europe) should be added the immense budgetary imbalances of many weaker Eurozone members, currently generally overlooked.
Unfortunately, while I must agree with Mr. Buiter, I feel that there is a significant chance that any comprehensive reform to the Eurozone financial stability arrangements may already been left too late for effective implementation.
We are most likely facing a major retrenchment of the Euro exchange rates, probably preceded by a period of increased other European (including Swiss) currency volatility and vulnerability
Posted by: Gerassimos Rocos | May 17th, 2008 at 12:27 am | Report this commentI suppose since there is no real discipline to money’s rate or final devaluation, the fiat money growth curve looks like a hyperbola function.
A Central Bank goes broke when people no long desire to trade with the currency.
Fortunately for CB’s paper money seems to possess a magical hold over people and no matter how much CB’s debauch it, people still cradle it like a baby as they go to buy their bread.
One of the firms printing Weimar Republic Marks submitted an invoice for the work to the Reichsbank for 33 quintillion Marks..
Fool me once… shame on… uh… shame on… you. Fool me can’t get fooled again!
Posted by: Cameron King | May 17th, 2008 at 1:58 am | Report this comment“Après-nous le Déluge!”. History repeats itself.
Posted by: Augustin | May 17th, 2008 at 4:23 am | Report this comment“Après-nous, le Déluge!” History repeats itself!
Posted by: Augustin | May 17th, 2008 at 4:25 am | Report this commentA correct analysis,but too late. The dynamics of the widely spread unbalances will prevent a rational solution not only for the ECB but for all Central banks.
The people that witnessed the German hiperinflation are almost all dead.
So are the people who witnessed the 1930 Depression, a word the press did not want to mention till very recently (the famous D word)
We have a recent case where witnesses are still alive, The Argentine inflation of the last 50 years.
Let’s find out what really happened from the people who lived it and let us use that experience. Everything we are talking about today has happened many many times.
Confucius said: The essence of knowledge is having it to apply it
Posted by: george eckstein | May 17th, 2008 at 5:40 am | Report this commentThe only way to avoid these problems with fiat paper is to go back to some kind of GOLD-STANDARD
or COMMODITY-STANDARD.
FIAT PAPER means nothing look at Zimbabwe.
If it was not the USA the dollar would be the same as the ZIMBABWE currency.
Posted by: Ronald A | May 18th, 2008 at 12:51 pm | Report this commentAnd also that can happen with the EURO.
It is very wrong that central banks still sell their gold and swap it for paper.
The possibility that the central bank could go bust and require government recapitalisation is just one reason why I believe that central banks should have no financial stability responsibilities whatsoever, including any lender (or market maker) of last resort function. My main concern is conflict of interest. When financial instability threatens, there must be a temptation for a central bank with some responsibility for financial stability to run an easier monetary policy than otherwise and take a risk with price stability. Di Noia and Di Giorgio (in International Finance 1999) provided evidence that central banks with supervisory responsibilities have a worse inflation performance. Moreover, some routine central bank operations involve transactions with private sector institutions, which must present the temptation to use private information obtained in the course of financial stability surveillance to steer clear of (and thereby pass on) problems. Naturally, these routine operations may be conducted less efficiently if the staff involved are also responsible for surveillance themselves. Perhaps there were coordination problems in dealing with the onset of the Northern Rock crisis, but these could be as easily dealt with by completing the transfer of financial stability responsibilities to the FSA as bolstering this side of the BoE. As for the lender of last resort function, this is really a fiscal policy, whose costs and risks should be unequivocally assigned to government. I suspect that one reason that governments like the central bank to serve as lender of last resort is that modest losses can be absorbed by an obscure reduction in seigniorage.
Posted by: Tim Young | May 19th, 2008 at 5:29 pm | Report this commentTim, I have a long of sympathy for these views. If lender of last resort and market maker of last resort operations produce losses for the central bank, it is key that these should be explicitly fiscalised, that is transferred to the Treasury and the tax payer.
I would therefore propose that any assets acquired by the central bank in the performance of its LLR and MMLR functions that have material credit risk attached to them, be transferred straight to the Treasury or some other designated public body guaranteed by the Treasury.
Whether the operational performance of the LLR and MMLR functions should be performed by a public entity other than the central bank is less important than the immediate transfer of assets subject to default risk from the central bank’s books to the tax payer, where they belong.
Posted by: Willem Buiter | May 20th, 2008 at 6:17 pm | Report this commentI have two innocuous questions/comments about this very fine piece. First, the view (also attributed to Ed McKelvey) that the Fed is not affected by “going broke” because it can still print money is one sided. Should the Fed choose to fight inflation, it could run out of assets to sell for that purpose. Since last August, the Fed has lent out its government securities, many with no requirement that they be returned, and by it has sterilized its private credit extensions. Available government securities are soon going to be less than half that available last August. This is not to mention the loss of credibility and reputation that would follow on large credit losses at Fed.
Posted by: Jack Tatom | May 24th, 2008 at 8:49 pm | Report this commentSecond, I think your description of the Bear “loan” is inaccurate. The “loan” is to an SIV that the Fed will create and own and that the Fed contracted with Black Rock to manage, not a loan to JP Morgan Chase. It really purchases an equity position in an SIV owned by the Fed, and that SIV will use the proceeds to buy the bad paper of Bear Stearns, presumably in the next week or so and this will allow Chase to close the acquisition of Bear. There is a crediting of interest to the Fed by the SIV, but all profits, or losses, after the first billion to Chase, go to the Fed.
The answer for this question should be NO, if we are talking about the Fed and the reason is simple: They can print dollars !!?.
Posted by: Rogerio Martins | September 17th, 2008 at 9:08 pm | Report this comment