Regular readers have known since October 2010 that "yes" was the answer since this blog ran the Free Lunch series of posts (see here, here, here and here).
These posts not only looked at the issue of "are the central banks safe", but then carried the analysis forward to examine "how the characteristics that make central banks safe could be used to address the issue of solvency in the financial system".
It is nice to have Wall Street confirm my analysis on central bank safety.
I expect Wall Street will shortly turn to the logical next issue of how to use the characteristic that makes central banks safe to address solvency in the financial system. When they do so, they will confirm another element of my blueprint to save the financial system - the element being that central banks can supply all the funding to support an insolvent bank not provided by the bank's depositors.
The Free Lunch posts ran over a year ago, but then regular readers expect me to be ahead of the curve given my December 2007 comments preceding the financial institution blow-up phase of the financial crisis.
Central banks have adopted ever-more unconventional policies since the financial crisis erupted in 2007. Chiefly, their extraordinary responses have taken the following forms.
First, the Fed and the Bank of England have purchased government securities, ‘quasi’ government securities (e.g., eligible mortgage-backed securities in the US) or credit instruments on a scale not seen (outside of Japan) during the post-war era. Even the ECB has dabbled in bond-buying via its securities market program. Second, all three central banks have acted as large-scale lenders of last resort to banks. Since 2008, that designation includes investment banks that have acquired banking licenses, at least in the US. Finally, in their capacity as liquidity providers to the financial sector several central banks— among them the ECB and some of the national central banks within the Eurozone—have significantly lowered eligible collateral standards for banks seeking funding.
All that activity makes some folks nervous. In particular, two questions arise. First, could a central bank go bust and precipitate a liquidity crisis? Second, if a central bank gets into trouble, who stands behind it?
Below, we explore both questions. We conclude that, for the most part, fears of central bank insolvency leading to a collapse of liquidity and failure of the payments system are wide of the mark. But it is possible to imagine a central bank suffering large losses, enough to wipe out its capital and hence warrant recapitalization.
That’s where matters get complicated and potentially problematic. Recapitalization by the fiscal authority—where it exists—may require legislation, which introduces politics and potential delays into the picture. So even in the US or the UK, where national governments backstop the central bank, the politics of ‘bailing out’ the Federal Reserve or the Bank of England could become messy, potentially leading to a period of elevated market uncertainty.
But the greater challenge resides in the Eurozone. No central fiscal authority exists to backstop the ECB. And some of the ECB shareholders—the 27 national central banks of the EU—might also themselves become financially stressed in a scenario where the ECB faces large losses.
In the end, investors might conclude that given the unimaginable alternatives, EU national governments would step in to recapitalize the ECB and, where necessary, their national central banks. But the history of the Eurozone crisis has not offered many reassuring examples about the speed and effectiveness of Eurozone political decision-making. In terms of crisis management, the lesson from Europe thus far has been that, indeed, crisis precedes management. Investors could be forgiven for wondering whether central bank recapitalization might not require a crisis first.
Why disorderly central bank default is unlikely
Central banks can become insolvent. Infrequently they do. But it is unusual for their losses, per se, to impair the functioning of the financial system. Indeed, losses at central banks are not the same things as losses among ordinary banks.
For one, central banks aren’t forced to mark-to-market their holdings or to provision against changes in the probability of credit losses. Dodgy assets can be held at par until losses materialize or until maturity.
But the most crucial distinction is that central banks borrow with the money they (and they alone) print. That money is fiat—irredeemable in anything but itself. To fund its Treasury or mortgage-backed securities purchases in recent years, for instance, the Fed merely credited banks’ accounts at the Fed with the dollars it printed (electronically, of course).
That makes the central bank unique. Its creditors cannot change the terms on which it borrows. As a result, the capital position of central banks is all-but irrelevant—it neither affects the central bank’s cost of funds, nor its ability to fund itself.
The privileged position of issuing fiat money enables central banks to operate with skinny capital. The Fed’s capital is $50bn—not much when compared to a balance sheet over $2.5 trillion. Were it a bank, on the other hand, the Fed’s capital ratio of less than 2% would have already landed it in bankruptcy court. Moreover, central banks can tolerate write-downs. In part, that is because they don’t require capital to borrow. But equally, it is because they are moneymaking machines in a different sense as well—they make fat profits (known as seigniorage). Central banks enjoy unparalleled net interest margin (borrowing at near-zero interest rates [Central banks will incur borrowing costs to the extent they pay interest on reserves, but those ‘borrowing rates’ are typically very low—much lower than the yield on the assets they hold and much lower than any private sector financial intermediary could hope to attain.], while investing in much higher yielding government or credit securities). Last year, for example, the Fed earned $76.9bn in profit, more than its total capital base.
Typically, central banks return all but a small fraction of their earnings to the government. In bad times, however, those earnings could be used to offset realized losses, bolstering the capital of the central bank.
But the key point is this one—even if the central bank incurred sufficiently large losses to create a negative equity position, that outcome would not change its ability to borrow via securities purchases. In short, even negative equity (technical insolvency) would not prevent a central bank from performing its customary open market operations nor its lender of last resort function.[Some observers have noted that the ECB’s low seigniorage revenues do not provide the same ‘earnings-based’ capital cushion available to other central banks. But the essential point is that capital does not determine the central bank’s ability to perform its mandated functions.] So when does a central bank actually go bust? The answer is when it cannot make payments, which would be the case if the central bank borrowed in foreign currency (i.e., money which it cannot ‘print’). That is not the case for the Fed, BoE or ECB—the liabilities of all three central banks are almost exclusively in their own currency.[In the event a country exits the Eurozone, its central bank would issue new national fiat currency, effectively re-denominating its liabilities.]
So if central bank capital is all but irrelevant, why have it? Alternatively, why worry if it is eroded by losses?
The answer may be, in part, optics.[The statutes of central banks may also require it.] No one, not even a central banker, wants negative net worth. But more subtly, it is also undesirable to incentivize central banks to maximize seigniorage (i.e., to earn their capital). After all, given a fixed net interest margin, maximizing seigniorage is about boosting assets ‘under management’. In turn, that requires creating excess money, which raises the risk of inflation. So the preferred public policy is to endow central banks with capital rather than to compel them, however infrequently, to earn it. Lastly, the source of central bank capital—the national government—acts a subtle yet powerful reminder that the central bank is not utterly independent, but ultimately is answerable to the taxpayer.
The Eurozone is different
That’s why recapitalization is probable if a central bank suffers large enough losses to wipe out its capital.
So who recapitalizes the central bank? And why is the Eurozone different?
Insofar as central banks are a part of government [Technically and historically, this is not quite right. Central banks historically may have public-private roots. The Fed is a quasigovernmental organization and, as noted earlier, the ECB shareholders are the 27 national central banks of the EU. But insofar as they deliver public goods (a medium of exchange and, hopefully, price stability) and because they have (near) monopoly issuance of money, they are commonly assumed to be part of government.], the taxpayer ultimately stands behind them, at least where a central (federal) government exists. In the event necessary, it is widely assumed that the Treasury (or a European Finance Ministry) would recapitalize its national central bank.
Yet the act of doing so is a fiscal transfer, making it subject to legislative approval.It isn’t far-fetched, therefore, to imagine that if a central bank required a capital infusion, politics would intrude. To be sure, even if recapitalization were held up by politics, the central bank could perform its mandated duties, as previously noted. But political intervention could have other unsettling byproducts, such as de jure or de facto restrictions on the central bank’s operational independence (or even its revocation altogether).
The Eurozone is a special case because there is no central (or federal) government that stands ready to recapitalize the ECB. Moreover, some of the ECB shareholders (the national central banks) might find themselves in a pinch at the same time that the ECB needs a capital top-up. That’s because some Eurozone national central banks have similar or worse ‘risk asset’ exposures than the ECB. For example, via the Emergency Liquidity Assistance (ELA) facility, several national central banks have extended considerable collateralized lending to banks in their countries, reportedly accepting even weaker collateral than the ECB has in its own operations.Accordingly, it is possible that a series of Eurozone sovereign or banking defaults could simultaneously erode the capital position of the ECB and those of some of its shareholding national central banks. That outcome would imply that central bank recapitalization would have to be led by a subset of creditor countries (i.e., Germany). That’s potentially a problem—recent history reminds us that Europe’s creditors have a proclivity for prevarication where asymmetric bailouts are involved.
Summary and conclusions
Central banks have taken on more risk in recent years. That’s been necessary and highly desirable during the most severe financial crisis since the 1930s, followed by the ‘great recession’. One shudders to think of the consequences of the alternative—‘Austrian’ central bankers running the show.
Yet the actions of central bankers in recent years may yet have undesired consequences. Central banks have claimed to have exercised prudence in demanding sufficient collateral and adjusting ‘haircuts’ to the value of collateral. But with central bank balance sheets swollen relative to their capital and a second recession underway in Europe, credit losses could mount. Stuff happens.
To emphasize, the risk is not that large central bank losses would impair the ability of the monetary authorities to provide liquidity, conduct open market operations, target policy rates, or safeguard the payments system. Rather, in the event that losses wipe out too much of their capital, the chief risk becomes the intrusion of politics into central banking. It might even bring about the end of independent central banks.