1. see Humphrey, T. and Keleher, R., “The Lender of Last Resort A Historical Perspective” (1984) 4(1) Cato Journal 275, 282. How unfitting that the heirs of the man who named the function—Sir Francis Barings—lived to see the Bank of England failing to rescue Barings in 1995!Google Scholar

2. From a central bank's point of view, the nature of its lender of last resort role involves three different aspects: (1) The discount rate at which the central bank lends, acting in its capacity as lender of last resort, is an instrument of monetary policy. In Germany the tightening and easing of the “lombard rate” (i.e. the rate at which the Bundesbank lends emergency funds to banks against collateral) has been used as an instrument of monetary policy. In the US the Federal Reserve System has made use of the discount window lending in monetary policy. However, it is argued that “for the sake of economic stability, it is important that the Fed distinguish at all times between its monetary policy and its LOLR function. The lender of last resort should replace liquidity lost as a result of customer demands; it should not force additional liquidity into the system” (see Benston, G. et al. , Safe and Sound Banking: Past Present and Future (1986), p.113). Regulation A, which governs the extension of credit by Federal Reserve banks, states: “While open market operations are the primary means of affecting the overall supply of reserves, the lending function of the Federal Reserve Banks is an effective method of supplying reserves to meet the particular credit needs of individual depository institutions. The lending functions of the Federal Reserve System are conducted with regard to the basic objectives of monetary policy and the maintenance of a sound and orderly financial system.” (2) The lender of last resort is an instrument of banking supervision in a “crisis situation” stage. As part of its micro-prudential functions, the central bank acting as lender of last resort provides assistance to a bank (or banks) suffering from a liquidity crisis. (3) The lender of last resort is a service provided by the central bank in its capacity as bankers' bank.Google Scholar

3. Under a currency board arrangement, however, the LOLR role of the central bank becomes limited (in the case of a mixed central bank/currency board system) or disappears altogether. For an interesting discussion on this issue, see G. Caprio, M. Dooley, D. Leipziger and C. Walsh, “The Lender of Last Resort Function under a Currency Board. The Case of Argentina”, World Bank Policy Research Working Paper No. 1648. Sept. 1996.

4. see Thornton, H., An Enquiry into the Nature and Effects of the Paper Credit of Great Britain (1802)Google Scholar and Bagehot, W., Lombard Street A Description of the Money Market (1873)Google Scholar. Recent studies of the work of Thornton and Bagehot on the LOLR are found in Humphrey, T., “The Classical Concept of the Lender of Last Resort” (1975) 61(1) Federal Reserve Bank of Richmond Economic Review 8, Humphrey and Keleher, op. cit. supra n.1Google Scholar, and Bordo, M., “The Lender of Last Resort Some Historical Insights” (1990) 76(1) Federal Reserve Bank of Richmond Economic Review 18.Google Scholar

5. See Bagehot, idem, p.56: “And at the rate of interest so raised, the holders—one or more—of the final Bank reserve must lend freely. Very large loans at very high rates are the best remedy for the worst malady of the Money Market when a foreign drain is added to a domestic drain.” Bagehot distinguished between actions to follow in the face of an external drain (a decline in the bank's gold reserve induced by a balance of payments deficit): raise bank rate, and actions to follow when threatened by an internal drain: lend freely.

6. see Lastra, R., Central Banking and Banking Regulation (1996), pp.266–268, 127.Google Scholar

7. see Kaufman, G., “Bank Contagion: A Review of the Theory and Evidence” (1994) 8 J. Financial Services Research 123–150Google Scholar. Kaufman, further elaborates his views in “Bank Failures, Systemic Risk and Bank Regulation” (1996) 16(1) Cato Journal 39. In this latter study he concludes that “bank instability is more a regulatory phenomenon than a market phenomenon”. However, he concedes that “systemic risk may exist without government regulation” and that there may be a role for the central bank as LOLR in such a case. Furthermore, he also suggests that since the success of “macroeconomic policies to achieve stability and avoid price bubbles” is highly questionable, “backup prudential policy is desirable”.Google Scholar

8. See D. Schoenmaker, “Contagion Risk in Banking”, LSE Financial Markets Group, Discussion Paper No.239, Mar. 1996.

9. See Kaufman, (1996), op cit. supra n.7, at p.7Google Scholar. Kaufmann also notes: “Indeed the evidence for countries that do not have explicit government insurance indicates that they generally have implicit 100 percent insurance.” P. Molyneujt in a paper, “Banking Crises and the Macroeconomic Context”, presented at the EBRD Seminar, “Bank Failures and Bank Insolvency Law in Economies in Transition”, in Oct. 1997, says that “everybody knows” the central bank will provide emergency liquidity assistance to what he calls “core banks”. Caprio et al., op cit. supra n.3, at p.6, have found that in all currency board regimes with no explicit LOLR, in instances of crises “some LOLR appears to have emerged”.Google Scholar

10. see Greenspan, A., “Remarks at a Conference on Risk Measurement and Systemic Risk”, Board of Governors of the Federal Reserve System (1995), p.7.Google Scholar

11. The restoration of confidence is also likely to require a change in macro-economic policy. Furthermore, if the crisis has been magnified because of a weak bank supervisory structure (e.g. in Venezuela in 1995) or because of supervisory and regulatory mistakes (e.g. the US thrift crisis in the 1980s) those issues also need to be redressed in order to restore confidence in the banking system.

12. In my opinion, the Bank of England's decision not to provide liquidity assistance to the already insolvent Barings in 1995 (though criticised in part) should be credited on the grounds that it did not promote moral hazard, thus enhancing the reputation of London as a financial centre where market discipline is fostered.

13. In theory, some argue that if an institution goes bust, there will always be another institution to pick up the pieces; in practice, however, some central bankers contend that, in a condition of panic, the central bank needs to keep things going until a market is found for the assets of the assisted institution.

14. While the test of illiquidity is rather straightforward (i.e. lack of liquid funds), the test of insolvency is often more complicated. As H. Schiffman explains in the paper he presented to the Oct. 1997 EBRD Seminar (supra n.9): “In commercial bankruptcy there are two traditional definitions of insolvency: failure to pay obligations as they fall due (equitable insolvency); and the condition when liabilities exceed assets (balance sheet insolvency).” Economists usually refer to insolvency following the balance-sheet insolvency test, i.e. when net worth is negative. (An institution may be liquid but insolvent, so long as its cash flow is positive.) However, an economically insolvent bank need not be declared legally insolvent by the responsible agency and may be offered instead financial assistance. A bank is considered to have failed when the competent authorities order the cessation in its operations and activities. If the bank is declared legally insolvent at the exact moment the market value of its net worth reaches zero, direct losses are suffered only by shareholders. If that declaration occurs when the market value of its net worth is already negative, losses will accrue not only to shareholders but also to uninsured creditors and/or to the insurance fund. See Lastra, op. cit. supra n.6, at p.135.Google Scholar

15. see Schwartz, A., “The Misuse of the Fed's Discount Window” (1992) 75(4) Federal Reserve Bank of St Louis Review 58–69.Google Scholar

16. As acknowledged, CAMEL is the rating system that US regulatory agencies use to assess bank soundness. CAMEL is an acronym which stands for Capital adequacy, Asset quality, Management, Earnings and Liquidity.

17. As Schwartz, op. cit. supra n.15, at p.64, recalls: “The bank was insolvent when its borrowing began and insolvent when its borrowing ended. The loans merely replaced funds that depositors withdrew, the inflow from the Reserve Bank matching withdrawals.”Google Scholar

18. Goodhart, C. and Schoenmaker, D., “Institutional Separation between Supervisory and Monetary Agencies”, LSE Financial Markets Group, Special Paper No.52, 04 1993, p.18.Google Scholar

19. I am quoting from pp.29–30 of a draft paper by C. Goodhart (1997), “Setting Standards is the First Step: Maintaining Them is the Harder Part”, cited with the permission of the author.Google Scholar

20. The names of the banks that receive LOLR loans but do not fail are not likely to be publicised. Indeed, any publicity might erode frail confidence, possibly causing a run on the institution (the very effect the central bank is trying to prevent through its LOLR assistance).

21. 12 U.S.C. 221 et seq.

22. FDICIA was enacted on 19 12 1991, Public Law 102–242 (105 Stat. 2236–2393).Google Scholar

23. According to s.38 of the Federal Deposit Insurance Act as amended by FDICIA, an institution is “undercapitalized” if “it fails to meet the required minimum level for any relevant capital measure” (emphasis added).

24. In this case, the 60-day period may be extended for an additional 60-day period (emphasis added).

25. For the purposes of FDICIA, “A depository institution is viable (emphasis added) if the Board or the appropriate Federal Banking Agency determines, giving due regard to the economic conditions and circumstances in the market in which the institution operates, that the institution: (i) is not critically undercapitalized; (ii) is not expected to become critically undercapitalized; and (iii) is not expected to be placed in conservatorship or receivership.”

26. An institution is critically undercapitalised, according to s.38 of the Federal Deposit Insurance Act as amended by FDICIA, when “it fails to meet any level specified under subsection c.(3)(a)” (i.e. when the leverage ratio of tangible equity is less than 2% of total assets). Emphasis added.

27. See Case 172/80 Züchner v. Bayerische Vereinsbank [1981] E.C.R. 2021. The Court's decision made it clear that the banking sector is exempted from the competition rules only to the extent that any anti-competitive conduct by banks is imposed on them by the monetary authorities.Google Scholar

28. Art.92 of the EC Treaty. See generally Ileana Simplicean-Stroia, “Study of the State Aid Policy in the European Community: The ‘Illegal’ State Aid Problem” (1997) 3 J.Int. Legal Studies 87.Google Scholar

29. Member State non-compliance with the procedural rules of Art.93(3) of the EC Treaty.

30. See (1997) O.J. C166/33 (Case T–62/97)Google Scholar. The applicant challenged the decision of the Commission of 26 July 1995 which gave conditional approval to the aid granted by France to Crédit Lyonnais (Case T–32/96). For a recent analysis of the approval by the Commission of the latest rescue plan for Crédit Lyonnais, see Financial Times, 21 05 1998, p.38 and 22 05 1998, p.2.Google Scholar

31. See European Parliament written question No.P–1216/96 by Giorgio Ruffolo on several related aspects of the application of Art.92 to banks, published on 15 10 1996 (1996) O.J. C305/92.Google Scholar

32. Sec Financial Tunes, 7 Oct. 1997. For a comprehensive study of the legal status of the Landabanken, see Gruson, M. and Schneider, U., “The German Landesbanken” (1995) Columbia Business L.Rev. 337–446.Google Scholar

33. See Fuiancial Times, 14 10. 1997, p.15: “An Uncertain Case for State Aid. The UK's Financial Law Panel is examining the legal position of banks.”Google Scholar

34. See Lastra, op cit. supra n.6, at pp.145–156Google Scholar (“Who should supervise the banking business”). See also Goodhart and Schoenmaker, op cit. supra n.18. The relationship between the central bank and the banking system involves two distinct yet intimately related functions, for the central bank can act as “bankers' bank” and as “banking supervisor and regulator”. The boundaries between these two functions are at times blurred. As Collyns, C., “Alternatives to the Central Bank in the Developing World”, International Monetary Fund, Occasional Paper No.20 (1983), p.3, remarks: “The functions of banker to domestic commercial banks and of regulator of these banks' behaviour are closely linked.” The lender of last resort role of the central bank is an example of this complexity. In the UK the Bank of England has a long non-statutory tradition in its role as “bankers' bank”, but its banking supervisory functions have had a short statutory history: from 1979, when the first Banking Supervision Act was passed, to 1997, when the newly elected Labour government announced that a single financial regulatory agency was to be set up, thus stripping the Bank of its formal supervisory responsibilities (though the Bank retains its LOLR role as well as general responsibility for financial stability). It should be noted that while as “bankers' bank” the central bank typically acts as a “bank”, as formal “banking supervisor” it acts as a “public authority” with a statutory role, backed by legislation.Google Scholar

35. This point is emphasised by W. Hogan in his critique of the Australian Wallis Report “What Direction from Wallis” (1997). The Final Report of the Financial System Inquiry (familiarly known as the Wallis Report) was published by the Australian Government Publishing Service in Mar. 1997.Google Scholar

36. see Goodfriend, M. and King, R., “Financial Deregulation, Monetary Policy and Central Banking” (1988) 74(3) Federal Reserve Bank of Richmond Economic Review 3, 13.Google Scholar

37. For a brief exposé of the structure, goal and functions of the FSA see the “Pathfinder Prospectus” on the FSA, released on 28 Oct. 1997Google Scholar. For a comment and analysis on the FSA, see Parliamentary Brief, 01. 1998, pp.28–34, interview with the FSA Chairman, Howard Davies.Google Scholar

38. In particular, paras.5–9 of the Memorandum refer to information gathering and sharing and paras.10–13 refer to rescue operations.

39. Each institution will have nominated representatives, which can be contacted, and meet, at short notice.

40. See para.13 of the Memorandum. Indeed, if fiscal resources need to be committed—i.e. taxpayers' money—then the Chancellor of the Exchequer should have a say in deciding whether or not to provide support to the troubled institution/s.

41. 12 C.F.R. 201.3(d). However, “the rate applicable to such credit will be above the highest rate in effect for advances to depository institutions. Where the collateral used to secure such credit consists of assets other than obligations of, or fully guaranteed as to principal or interest by, the United States or an agency thereof, an affirmative vote of five or more members of the Board of Governors is required before credit may be extended.”

42. 12 C.F.R. 201.7 (C.F.R. revised as of 1 01 1996).Google Scholar

43. See J. Sachs, “Do We Need an International Lender of Last Resort?”, paper presented to the Study Group on Private Capital Flows to Developing and Transitional Economies at the Council of Foreign Relations on 5 Oct. 1995. Sachs argues that international bankruptcy procedures modelled upon Chapters 9 and 11 of the US Bankruptcy Code would be the best response to cope with Mexican-type crises. Sachs's proposals, which include the reorganisation of the IMF to act as a kind of international bankruptcy court rather than as a lender of last resort to member governments, overlook important legal and constitutional aspects, including inter alia the difficulties of enforcing international law and the differences in national legislation regarding insolvency law and liquidation procedures. For instance, at the EC level—where a coherent degTee of harmonisation has been achieved in other banking aspects—a proposed directive on the reorganisation and winding up of credit institutions is still the subject of much controversy. A bankruptcy procedure for developing countries analagous to Chapter 11 of the US Code is also recommended in a report published by B. Eichengreen and R. Portes in 1995, under the title “Crisis? What Crisis? Orderly Workouts for Sovereign Debtors”, CEPR. See also the Group of Ten Working Party's Report, “The Resolution of Sovereign Liquidity Crises”, May 1996, and Kenen, P. (Ed.), “From Halifax to Lyons: What Has Been Done about Crises Management?” Essays in International Finance, 10. 1996.Google Scholar

44. Krugman, P., “Will Asia Bounce Back?”, mimeo (1998) puts the blame on bad policies and bad bankingGoogle Scholar. Sachs, J. and Radelet, S., “The Onset of the East Asian Financial Crises”, mimeo (1998) puts the blame on financial panic, therefore justifying the possible role for an international lender of last resort. The 1998 World Bank Global Development Finance Report also blames the lendersGoogle Scholar. Indeed, as A. Fraga, op cit. infra n.49, at p.53, noted after the Mexican crisis in 1995: “If Mexico is thought to have borrowed so much, it is also fair to ask why markets were so lax in providing the financing.”Google Scholar

45. In addition, other measures to prevent future crises, such as the desirability of controls over short-term capital inflows, have also been proposed following the turmoil in East Asia. See e.g. Stiglitz, J., “Boats, Planes and Capital Flows”, Financial Times, 25 Mar. 1998. The flows that might be subject to controls are those which exhibit four characteristics: foreign currency denominated, short-term, fixed interest, financing flows.Google Scholar

46. The Core Principles for Effective Banking Supervision, adopted by the Basle Committee on Banking Supervision in Sept. 1997, are a step in this direction.

47. Gianviti, F., General Counsel of the IMF, in “The IMF and the Liberalization of Capital Markets” (1997) 19 Houston J.Int.L. 773, 774–775Google Scholar, claims that the Mexican debt crisis of 1994–95 revealed some “major weaknesses in the current approach to liquidity crises”, including “the lack of adequate preventive mechanisms to detect potential liquidity crises at an early stage and avoid their occurrence or limit their magnitude”. He explains that the resolution of external debt problems due to a major capital outflow is not the responsibility of the Fund, but the responsibility of the country facing this outflow. He further argues that, though in an age of liberalisation of capital markets, these principles may seem antiquated, they are still in force and must be observed. However, it should be noted that discussions are currently under way to extend the IMF's jurisdiction to capital movements, through an amendment of the IMF Articles of Agreement. See e.g. IMF Survey, 27 04 1998, p.121.Google Scholar S. Cross—in private correspondence—suggests that the IMF should shift more of its attention and effort to the capital side of every member's situation; as part of the surveillance process, every member would have to project the capital inflow-outflow consequence of its prospective current account position, and say how it expects to finance its forthcoming surplus/deficit, and the likely consequences for the capital markets. The IMF can then give or withhold its blessings on the member country's plans.

48. Of course, the IMF will need to strike a fine balance between its role as a confidential adviser to members and its potential role as an international financial watchdog (see Financial Times, 30 03 1998, p.3).Google Scholar This debate at an international level resembles the domestic debate in the UK in the 1970s when the Bank of England, till then an honest broker for commercial banks in its capacity as “bankers' bank”, also adopted a more formal role as bank supervisor, with the passage of the 1979 Banking Act.

49. Fraga, A., “Crises Prevention and Management: Lessons from Mexico”, in Kenen, (Ed.), op. cit. supra n.43, at p.54, has indicated that the IMF should act as “the permanent auditor of countries, who should voluntarily submit themselves to examination in order to lower their borrowing costs”. Fraga also proposes that Art.IV consultations be sup- plemented by quarterly reviews to enhance the credibility of the data released under the IMFs initiatives on better disclosure of country data (i.e. the special and general data dissemination initiatives).Google Scholar

50. A. Greenspan, in his remarks before the 34th Annual Conference on Bank Structure and Competition of the Federal Reserve Bank of Chicago, 7 May 1998, refers to cross-border inter-bank funding as “potentially the Achilles' heel of the international financial system” and a “significant source of systemic risk”. He points out that “creditor banks expect claims on banks, especially banks in emerging economies, to be protected by a safety net and, consequently, consider them to be essentially sovereign claims”. He suggests that some sort of discipline should be imposed on creditor and debtor banks: “For example, capital requirements could be raised on borrowing banks by making the required level of capital dependent not just on the nature of the bank's assets but also on the nature of their funding. An increase in required capital can be thought of as providing a larger cushion for the sovereign guarantor in the event of a bank's failure … Alternatively, the issue of moral hazard in interbank markets could be addressed by charging banks for the existence of the sovereign guarantee, particularly in more vulnerable countries where that guarantee is more likely to be called upon and whose cost might deter some aberrant borrowing. For example, sovereigns could charge an explicit premium, or could impose reserve requirements, earning low or even zero interests rates, on interbank liabilities. Increasing the capital charge on lending banks, instead of on borrowing banks, might also be effective. Under the Basle capital accord, short-term claims on banks from any country carry only a twenty percent risk weight. The higher cost to the lending banks associated with a higher risk weight would presumably be passed on to the borrowing banks.”

51. M. Feldstein in an article published in the Financial Times, 5 Mar. 1998 (“Trying to do too much”) claims that if the purpose of the IMF packages for Korea, Thailand and Indonesia was “to act as a lender of last resort in order to stop financial panics and the runs by creditors, the IMF's funds would have had to be available for immediate disbursal, not held back until these countries demonstrated their willingness to carry out major structural reforms”.

52. See Financial Times Survey, “Asian Financial Markets”, 27 04 1998, p.33.Google Scholar

53. See IMF Survey, 12 01 1998, p.7. This facility has at the international level some of the features that Thornton and Bagehot described for the LOLR at the domestic level. It should be noted that both Thornton and Bagehot wrote their important contributions to the understanding of the LOLR in the 19th century, when crises were mostly confined to the national arena.Google Scholar

54. See ibid. The IMF also intends to bring to completion at an early date ongoing studies of mechanisms that could effectively limit the need for official financing in cases of crisis of market confidence.

55. Guitián, M. argues in “Conditionality: Past, Present and Future”, Staff Papers, Vol.42, No.4, International Monetary Fund, Dec. 1995, that the interpretation of conditionality is not independent of the international economic regime in place.Google Scholar

56. See IMF Survey, 15 12 1997 for the approval of the Korean arrangementGoogle Scholar, IMF Survey, 17 11 1997 for the approval of the Indonesian stand-by arrangementGoogle Scholar, and IMF Survey, 17 09 1997, for the approval of the Thai stand-by arrangement.Google Scholar

57. Fraga, op. cit. supra n.49, at p.53, himself a managing director of Soros Fund Management, notes: “investors behave myopically, each one perhaps thinking that it will be possible to exit ahead of the rest Moral hazard on the part of investment bankers (who hope to earn underwriting fees) and of Wall Street traders and mutual fund managers (who are paid bonuses every year) also helps to explain the Mexican case.” Feldstein, op. cit. supra n.51, also claims that “using IMF funds to pay off loans to private creditors weakens the incentive of lenders to be cautious in future international slending”.Google Scholar

58. As M Wolf explains in an article, “Capital Punishment”, Financial Times, 17 03 1998Google Scholar: “If a commercial bank in Thailand borrows dollars, no central bank will provide it with the currency they may need. Similarly, if Thailand as a country borrows dollars, there is no lender of last resort to assist it if creditors suddenly want their money back. The IMF—which cannot print US$—lacks the resources to perform this role.”

59. I thank C. Goodhart for bringing this issue to my attention.

60. See JP Morgan, “World Financial Markets”, 27 03. 1998, p.13. If investors remain free to make off with their winnings, they should also pick up the tab for their losses.Google Scholar

61. However, in principle, the country which receives a stand-by arrangement is expected to pay back to the Fund the amount received plus a rate of interest in a timely fashion.

62. The liquid resources of the IMF consist of usable currencies and SDRs held in the General Resources Account. When supplementary resources are needed to prevent or cope with a crisis the IMF can borrow under the General Arrangement to Borrow (credit lines from 11 industrial countries), an associated agreement with Saudi Arabia, and under the recently created New Arrangement to Borrow (credit line with 25 participant countries). The IMF can also get more resources through an increase in quotas and through an SDR allocation (provision of international reserves to its members in proportion to their quotas). The last increase in quotas—by 45%—was approved by the IMF Board of Governors on 23 Dec. 1997. See IMF Survey, 12 01. 1998, p.7.Google Scholar

63. As Kindleberger, C., A Financial History of Western Europe (1984), p.273, reminds us: “The record [in financial markets] shows displacement, euphoria, distress, panic and crises occurring decade after decade, century after century.”Google Scholar

64. Sec e.g. Financial Times, 1 Oct. 1997, “Global Payments Bodies to Merge”Google Scholar