By Richard Alford, a former New York Fed economist. Since then, he has worked in the financial industry as a trading floor economist and strategist on both the sell side and the buy side.
In “Lombard Street” published in 1873, Bagehot specified the purpose of a Lender of Last Resort (LOLR) as forestalling bank panics in fractional reserve banking systems. Bagehot also provided criteria that define LOLRs, which remain relatively unchanged. In fact, the Bagehot criteria have become something of a mantra: Lend freely at penalty rates against good collateral to illiquid but solvent banks. Given Bagehot’s purpose and definition, has the crisis of 2008 provided a test of the Fed as an LOLR? If so how well did the Fed perform? What are the ECB’s responsibilities as the LOLR in Europe in 2011?
To evaluate the Fed as a LOLR, one must consider how well it fulfilled the four requirements as set out by Bagehot:
1. Lend freely
2. At a penalty rate
3. Against good collateral
4. To illiquid but solvent banks
The Fed lent freely. Discount Window lending has been the vehicle by which the Fed has traditionally fulfilled its responsibilities as the LOLR. Discount Window borrowing did increase. The Discount Window loans to depository institutions peaked at close to $112,000million in late October 2008. (They were $126million in October of 2007.)
The Fed also increased other extensions of collateralized credit by to a variety of financial institutions and markets that were part of the shadow banking system. The Fed did so in order to prevent a run on the shadow banking system, the effects of which would have been similar to a run on the banking system itself. (This expansion of the list of institutions with access to the Fed as a LOLR is not unprecedented. By legal mandate, the Fed is not only the LOLR to the banking system. The Fed has also been the lender of last resort to thrifts.)
One set of vehicles was very close to the central bank’s traditional role as the Bagehot LOLR providing short-term liquidity to the banking system. The Term Auction Facility peaked near $493,000M in March of 2009. The Fed also provided short-term liquidity to primary dealers and other broker dealers through the PDCF and TSLF. Fed loans under these programs peaked near $147,700 in October of 2008.
A second set– the ABCPFF, AMLF, and MMIFF– provided liquidity directly to select borrowers and investors in money market instruments. The ABCPFF and the MMMFF (plus “deposit”-like insurance provided by the Treasury), was aimed at preventing a “banking” crisis in the MMMF component of the shadow banking system. Credit extensions via the ABCPFF and the MMMFF peaked near $145,900M in October of 2008.
However, foreign entities had also been active participants in the US shadow banking system. The Dollar denominated assets of banks not domiciled in the US is roughly the same size as the Dollar assets of the US commercial banking sector. Foreign banks, particularly European banks, activities in Dollar-denominated markets helped determine credit conditions in the US. Prior to the crisis, foreign institutions had borrowed Dollars short-term to fund positions in US assets, e.g. MBS. If foreign institutions without access to the Fed Discount Window had been unable to access market-sourced liquidity, they then would have had been forced to sell Dollar assets. As a result, the domestic efforts aimed at supplying liquidity to US-based agents to prevent the liquidation of assets would have been compromised, If not for naught. US institutions would have had to mark the prices of assets down and report lower net worth-a recipe for further pressure on US institutions. The same concerns are operative today, given the Euro crisis.
From this perspective, the Fed currency swaps with foreign central banks (who in turn lent the Dollars to institutions with Dollar liquidity problems and no access to the Fed Discount window) are consistent with its role as a LOLR. The Fed ignored geographical borders to supply the required liquidity to foreign–based participants just as it had ignored precedent and regulatory borders to supply liquidity to US institutions without access to the Discount Window. Fed currency swaps peaked in December of 2008 near $582,750M.
Like Discount Window borrowings, the lending programs required collateral. The swap lines were collateralized with foreign currency and backed by the central banks. (Given the questions that exist about the future of the Euro and hence the ECB, the “credit” risk embedded in the swap lines with the ECB are higher than they were in 2008.)
The interest rates varied across the lending programs. Some were linked to the Discount Rate which floats 100 basis points above the Fed funds rate. One was linked to the Fed Funds rate. Others were set by auction processes. The speed with which the loans made under these programs were unwound suggests that the interest rate was a penalty rate or administered in fashion that led counterparties to behave as if they were penalty rates.
These two set of vehicles closely adhered to the Bagehot requirements. The Federal Reserve ended the MMIFF in October 30, 2009. The AMLF, PDCF, and TSLF were closed on February 1, 2010.
The Fed also moved well beyond the Bagehot criteria. It created a third set of lending vehicles which were directed not at supplying liquidity to illiquid but solvent institutions, but rather at supporting asset prices and supplying liquidity and credit to sectors from which the banking and shadow banking systems had withdrawn.
The Commercial Paper Funding Facility, CPFF, was instituted to insulate the real economy from funding problems in the commercial paper component of the shadow banking system. At its peak in January of 2009, the Fed had lent close to $350,000M against commercial paper. TALF, the Term Asset-Backed Securities Loan Facility, was launched in March of 2009. It was designed to the consumer and small business asset backed securities market. TALF peaked near $48,000M in March of 2010. These CPFF and the TALF represent a clear break from the Bagehot criteria: they were not loans to illiquid but solvent financial institutions that performed banking functions. They may have been desirable, even necessary, tools for the Fed to have used to stabilize the financial system and the real economy, but they were not part of a LOLR function.
QE1, aka “credit easing”, was also directed at narrowing credit spreads and maintaining function. The Fed efforts to maintain the asset backed securities market (CPFF and TALF) have more in common with “credit easing” than the LOLR. Hence, the CPFF and TALF could be viewed as part of unconventional monetary policy and not as part of the LOLR function.
The Fed also extended credit to support specific institutions:
1. The Fed loan to Maiden Lane LLC to facilitate the purchase of Bear Stearns by JPMorgan
2. Fed Loans to Maiden Lane II and III as part of the package for AIG
The purpose of these “loans” was not to aid an illiquid but solvent institution. The Fed arranged and helped finance a merger in the case of Bear Stearns and a restructuring in the case of AIG. The Fed was performing a role that would have fallen to the FDIC had Bear Stearns and AIG been depository institutions. Furthermore, while it is the Fed’s position that the acquisition of the assets held in the Maiden Lane vehicle was de jure a “non-recourse loan”, the Fed’s acquisition of the assets was de facto indistinguishable from an outright purchase of the assets. This structure and the Fed’s position presumably reflect:
1. the perceived necessity of permanently removing the assets from the balance sheet of the entity,
2. the Federal Reserve Act which prohibits the Fed from buying the types of assets in question,
3. Congressional disapproval of a purchase of assets at a non-market price (there was no market) with the significant downside risk.
Furthermore, it has never been the purpose of an LOLR to finance corporate takeovers or the de facto nationalization of firms. Hence, the Fed was not acting as an LOLR in either the JPMorgan acquisition of Bear Stearns or the de facto nationalization of AIG, although one could argue that it was advancing financial stability. The necessity of the Fed involvement is also questionable. The Fed did not finance the government moves to stabilize Fannie and Freddie nor did it finance the government role in the bailout of GM and Chrysler.
Did the Fed fulfill its responsibility as an LOLR during the crisis of 2008? Yes. It lent freely to institutions in the banking and shadow banking systems that were experiencing liquidity problems. It lent against collateral and at rates that encouraged borrowers to unwind borrowings, which allowed for the sun setting of the facilities. Did the Fed also engage in other activities that supported market function and specific institutions? Yes, but these should not be viewed LOLR activities. Did the Fed violated the Bagehot criteria when it “lent” for purposes other liquidity and to insolvent institutions (including depository institutions) in support of the “too big to fail” doctrine? Yes.
Does the Fed as the LOLR have a duty to buy and support the prices of Treasury issued debt as some have suggested? The Fed pegged the prices (and yields) of Treasuries from the start of WWII until 1951. When after a contentious battle with The Executive branch, the Fed-Treasury Accord of 1951 ended the Fed’s pegging of the prices /yield on Treasury issues. The move has been viewed as a desirable and required step to insure a monetary policy that was free to pursuit the goals of full employment, growth, price and financial stability unencumbered by a need to maintain an unchanged yield on Treasury issuance. I do not recall anyone ever asserting that the Fed in achieving this independence had abrogated its responsibility as a LOLR. It may be desirable for the Fed to engage in another round of QE, but will be another instance of unconventional monetary policy and not an exercise as the LOLR.
Europe is in the midst of a financial crisis and the ECB is being called upon to act as an LOLR and support the prices of the debt of various peripheral countries. It may be wise, or not, for the ECB to buy the sovereign debt of the peripheral countries. It might be consider part of monetary policy or unconventional monetary policy, but given history and the definition of the LOLR it is hardly an LOLR function. An unlimited commitment to monetize the sovereign paper would imply that the ECB has surrendered any ability to control either quantity of reserves in the system or manipulate short-term interest rates. In short, its ability to execute monetary policy will have been abandoned. It would in effect become an off-budget financing arm of a non-existent and at the same time dysfunctional pan-Europe Finance Ministry.
This highlights a political dimension. While the ECB’s resistance to monetizing the debt or new issuance of debt by the peripheral countries reflects a variety of legal factors, it may also reflect the ex post criticism directed at the Fed. The Fed has been criticized by many for becoming the off-budget financing arm of the Treasury. The Executive and Legislative branches acted quickly enough to support GM and Chrysler with on budget appropriated funds. With a short bridge from the Fed, the Executive and Legislative branches moved to buttress Fannie and Freddie, again via enabling legislation. However with little popular support for a company brought down by playing with derivatives, the Fed alone financed, oversaw and took the heat stemming from the de facto nationalization of AIG. The Fed quickly became a political punching bag for decisions made at AIG, including the decision to pay all counterparties 100 cents on the Dollar. (Somehow after exceeding its mandate as a central bank and LOLR, the Fed decided it did not want to try its hand at being a bankruptcy court and allocating assets across counterparties.)
The ECB is in a position similar to that which the Fed found itself in with AIG on the brink. It is not the role of central banks or LOLRs to disguise the failures of fiscal policy, the architecture of the EZ or the political system any more than it is to bailout insolvent companies. The Fed is paying a political price: the Fed and monetary policy have become political footballs. It could have been avoided if the Fed simply said we will do with AIG what was done with Fannie and Freddie, i.e. the Fed provide limited support support in the very short-term with an agreed upon understanding that Congress and the President will provide the long-term solution and financing as well as an exit for the Fed.
The ECB can avoid both a near-term Euro crisis and the problem faced by the Fed, if it announces that it will buy the sovereign debt of the peripheral countries only if it can achieve a public agreement with the governments that stipulates the conditions under which it will buy the debt, a predetermined schedule, and ceiling the purchases and an predetermined date for an exit. The agreement will make clear that the ECB will provide help for a limited period of time in which the political process will have time to forge a permanent solution to the crisis. It will also put the focus back on elected official as well as give them a time limit and a reason to agree upon a solution to the underlying structural problems.