Are Bad Banks the Solution to a Banking Crisis by mifei

VIEWS: 334 PAGES: 43

									               Are Bad Banks the Solution to a Banking Crisis?*

                                          Jonathan R. Macey**

                                      SNS Occasional Paper No 82
                                                 June 1999

* This study is a background paper to a report evaluating Securum, the organisation that was created in 1992 to
manage the non-performing loan portfolio of Nordbanken, one of the largest banks in Sweden. In 1997 Securum and
SNS reached an agreement that SNS was to undertake an independent study of the operation of Securum and whether
an organisation like Securum is a suitable one to handle banking crises. SNS turned to three Swedish academic
researchers, Clas Bergström and Peter Englund, Professors of Finance, Stockholm School of Economics, and Per
Thorell, Professor of Law, Uppsala University (now at Ernst & Young), to undertake the evaluation. Funding has
been provided by Venantius, which in 1997 took over some remaining parts of Securum, and Nordbanken. The views
expressed in this paper are, of course, those of the author.
** Cornell Law School, Myron Taylor Hall, Ithaca, New York 14853, USA. E-mail:
1. Introduction
Sweden is only one of a large number of countries that have faced severe and systemic
problems in the banking sector during the past fifteen years. In all, ninety countries have
experienced significant episodes of banking collapse during this period. Twenty of these
episodes have produced bailout costs for governments in excess of ten percent of their
country’s GDP.1 As these bailouts are ending, finance ministries and international economic
organizations find themselves in the midst of a furious debate over the appropriate strategy
for resolving the Asian financial crises. The recent intervention by the Board of Governors of
the Federal Reserve System in the United States to bail out a large hedge fund, Long Term
Capital Management, has further contributed to the debate about the proper scope of
regulation in resolving troubled financial institutions.
       The Swedish government is quite proud of the way it handled the banking crisis of the
late 1980s. It is holding out the strategies utilized during this period as a model for the rest of
the world, particularly Asia.2 The critical feature of the Swedish strategy for resolving its
banking crisis was the splitting of distressed financial institutions into so-called “good” banks
and “bad” banks. This decision, which was the cornerstone of the Swedish approach to its
banking crisis, merits further study before it becomes the international model for future bank
failure procedures.
       The purpose of this article is to examine the good bank/bad bank strategy in general
and Swedish bank insolvency practices in particular in order to provide an independent
assessment of Securum’s performance and of the possibility of exporting the Securum
strategy to other economies facing financial crises. This Report is divided into three parts.
The first consists of a general discussion of the costs and benefits of the good bank/bad bank
strategy for resolving troubled banks. The second consists of an analysis of the performance
of Securum and its sister institution, Retriva, which were the entities established to take over
the bad assets of Nordbanken and Gota Bank, respectively. Lastly, the report considers

   Charles Calomiris, Harmful Bailouts, American Enterprise Institute, January 1998, On the
   See Lars Heikensten, Financial Crisis-Experiences from Sweden, speech at a Seminar
arranged by the Swedish Embassy, Seoul, Korea, July 15, 1998.

whether the Swedish experience with bad banks is generalizeable, that is, whether it can
readily be utilized in other bank failure situations.
        The major conclusions of this Report can be succinctly summarized. First, while the
good bank/bad bank strategy represents a viable strategy for many types of financial
institutions, it is possible to overemphasize the value of this strategy. Good bank/bad bank
restructurings simply involve the separation of the good assets of a bank from the bad assets
during a time of crisis, and calls for separate management of the disposition of these two
categories of assets. Thus, if one divides the period of a banking crisis into two parts, the
crisis prevention and crisis management stages, the good bank/bad bank strategy addresses
only the crisis management stage. Whatever the benefits associated with this crisis
management technique may be one cannot ignore the critical crisis prevention stage. This is
not to say that good bank/bad bank crisis management strategies are uninteresting or
unimportant. Rather, the point is that they are not a panacea. In fact, by definition, these
strategies are only used when problems exist. Clearly, the prevention of a banking crises is
always the first-best strategy.
        Crisis prevention requires early intervention. One of the primary findings of this
Report is that the earlier that bank intervention occurs, the more difficult it will be to identify,
categorize, sort, and value bad assets. Consequently, early intervention makes the task of the
managers of bad banks more difficult. In contrast, when intervention comes very late, after
the assets on a bank’s balance sheet have had time to deteriorate, even though the overall
costs of a bail-out become more costly, the task of the managers of bad banks is made
considerably easier by the fact that it is relatively easy to identify the bad assets.
        The main point made in this Report about the trade-off between early intervention and
the likelihood of success in implementing a bad-bank strategy implies that the fewer
resources that regulators devote to crisis prevention, the easier their job will be later when it
comes to crisis management. Thus, when evaluating the performance of a regulatory regime it
is probably a mistake to view crisis prevention and crisis management in isolation. These
components of banking crisis administration must be viewed together in order to completely
understand how well regulators have performed. While this observation may appear to be
obvious, analysis of the regulatory responses to the Swedish financial crisis has isolated crisis

prevention and crisis management, and has focused, understandably, on crisis management.3
          Similarly, when we evaluate Securum’s performance we must recognize the fact that
there may be incentives inherent in Securum’s good bank/bad bank strategy that could cause
the success of the program to be exaggerated. Specifically, those bank managers who were
responsible for selling off the bad assets may have had incentives to aggressively write-down
the value of those assets in the initial stage of crisis management, thereby causing the
economic success of the bail-out to be overstated. This Report suggests that competition for
the bad bank assets may provide a more objective measure of the bad banks’ balance sheets
with which a more accurate assessment of the crisis management stage program could be
          In addition, this Report observes that successful bank crisis management requires that
there be a solid political consensus among the various groups affected by the crisis. Because
effective (i.e. early) crisis prevention makes effective crisis management more difficult, and
ineffective crisis prevention makes the job of crisis management more manageable, it is clear
that it is far easier to achieve the political consensus necessary for an effective crisis
management stage once there is general agreement about the nature and scope of the problem
and the need for solutions. By contrast, at the crisis prevention stage it will be far more
difficult to achieve political consensus because there is likely to be meaningful disagreements
about the desirability of intervention. This conclusion stems from the observation that when
there is substantial disagreement about the value of a distressed bank’s assets, it will be
harder to reach political consensus about how those assets should be resolved.
          Of course, the problem of bank failures, and the bank failure crisis itself, are not
unique to Sweden. Over the past twenty years, countries all over the world have been forced
to find solutions to the problem of failed banks. According to Bengt Dennis, ex-governor of
Sweden's central bank, between 1980 and 1996, “more than two-thirds of the 181 members of
the International Monetary Fund experienced serious banking problems.”4 These failures have

      Lars Heikensten, supra note 2.
  Confidence Seen as Key Factor Swede Outlines Way to End Crisis, NATION (Bangkok),
Dec. 11, 1997. As reported:
       Dennis noted that Chile required 30 per cent of its gross domestic product
       during 1981–1987 to tame a crisis; Venezuela, 20 percent in 1994; Spain, 15–

forced policy-makers and academics to address the causes and consequences of bank failure.

2. Some Preliminary Observations About Bank Failure in Economic Theory
Contrary to popular belief, the failure of an insured depository institution does not reflect a
failure of the banking system as a whole or even a failure of the regulatory system that was
constructed in the aftermath of the wave of bank failures stemming from the collapse of the
stock market in 1929.5 As with other sectors of the economy, the failure of a financial
institution indicates one of two states of affairs: either the firm has not responded to market
forces with a satisfactory mix of price and product performance relative to its competitors in
the industry or else the firm’s product is not in sufficient demand by consumers to justify its
production in the first place. In either case, termination of the firm's operations represents a
net social gain. In a market economy, when an enterprise fails, the resources previously
devoted to a firm will find other, more efficient uses.6 As a result, the existence of failing
institutions may be a sign of health rather than a sign of malaise because it indicates either
that innovation is driving obsolete firms out of the industry, or that competition is driving

          20 percent in 1977–85; Mexico, 15–20 per cent in 1994; Sweden, 4 per cent in
          1990–1993; and the United States, 2 per cent in the 1980s. Thailand is
          expected to spend about Bt800 billion to end this crisis or around 15–20 per
          cent of GDP.
      See Tussing, The Case For Bank Failure, 10 J. L. & ECON. 129, 143 (1967).
   As financial intermediaries, banks directly affect the allocation of financial resources by
other firms. As Tussing notes:
        a financial intermediary has the potential not only for misusing the labor,
        invested capital, and other factor inputs it uses directly, but also for distorting
        the use of resources by business generally. If it is desirable that an inefficient
        manufacturing establishment fail, so that the resources devoted to it can be
        made more useful to society, it is doubly desirable that an inefficient bank fail.
        Not only will the resources directly devoted to it find more useful outlets, but
        the increased efficiency in financial processes which results can mean that
        loans are made more nearly in accordance with social priorities, that local and
        regional economic growth are not retarded by underbanking, and that
        interregional financial flows are not impeded through the petrification of inept
        locational decisions.
Id. at 146.

inefficient firms out of the market. These considerations suggest that in a developed market
system bank failure is not necessarily problematic.
       Concerns with bank failure, therefore, must be based on some special problems that
alter our customary reliance on the valuable economic role played by business failure. A
number of bank-specific concerns have been suggested: the fear of bank runs; the loss of
society's and depositors' wealth; and the costs of bank failure for the system of deposit
insurance, which itself is a response to perceived problems with bank failures. Yet, if the
government intercedes inefficiently, such as by keeping a failed institution alive for too long,
the costs of bank failure will rise unnecessarily, and the benefits may not fully be obtained.
       Similarly, even where bank failures serve some worthwhile market function, if the
market can find a means for minimizing the losses associated with a failed or failing bank and
maximize the value of performing assets in that bank by emancipating them from the binds of
the non-performing assets, the benefits of failures may not be obscured. The most important
such method has become known as the good bank/bad bank restructuring strategy. This
strategy minimizes the losses from bank failures, including the losses associated with the
erroneous deployment of bank assets. At the same time, the strategy can be combined with
elements of market discipline that ensure that the state does not either subsidize excessive
risk-taking or deprive market participants of the incentives to monitor bank management.

3. The Good Bank/Bad Bank Framework
For a failing bank, “the good bank-bad bank structure may represent the last, best chance for
assuring survival. It is unequaled as a means of quickly curing the debilitating problem-asset
disease.”7 Good bank/bad bank restructuring plans take a variety of forms, but all involve the
separation of performing and non-performing assets of a financially distressed bank into
separately managed financial institutions. Good bank/bad bank restructuring plans place low
quality assets into a special asset company, which is dubbed the “bad bank.” This “bad bank”
can take any one of a number of forms, including a liquidating bank, a partnership, a trust, or

   Timothy Hurley, Good Medicine: Bad-Bank Plans, AM. BANKER, June 17, 1992, at 4. See
also Peter S. Rose, The Good Word on “Bad” Banks, CAN. BANKER, Nov.–Dec. 1989, at 42
(arguing that the creation of bad banks have been responsible, in part, for the recovery of
several leading U.S. banks).

a corporation.8
       The other entity, the “good bank,” retains and manages the performing assets. This
strategy is intended to serve a number of business purposes:
                  The strategy allows a bank to jettison its troubled assets into a
                  limited-service bank or liquidating trust that operates as a
                  separate company. The net effect is to allow the “good” portion
                  of the bank to operate unencumbered by the “bad” bank’s

                  A liquidation plan is prepared for each asset in the bad bank,
                  and the asset values are discounted to current value. Investors
                  buy the securitized “bad” debt, betting that the proceeds of the
                  liquidation will be sufficient to repay the debt on time ...9

       The primary advantage of the good bank/bad bank structure is that the old bad bank
assets are dealt with separately from the good bank assets. It is far easier for a good bank to
succeed when it is unencumbered by the bad bank’s assets because the good bank has a lower
cost of capital. Thus, under the good bank/bad bank structure the good bank enjoys a better
comparative advantage.
       Early in the history of the deployment of the good bank/bad bank model, a variety of
alternative scenarios emerged through private ordering in the absence of regulatory
intervention. These strategies emerged to provide flexibility in dealing with the twin
challenges facing regulators and private-sector creditors who are trying to resolve a failed
bank. These challenges are: (1) to deal with the troubled institution’s non-performing and
under-performing (bad) assets; and (2) to avoid the costs and inefficiencies associated with
maintaining bad loans in moderately healthy institutions.

  U.S. Banking: Making the Good Bank/Bad Bank Structure Work, INT'L FIN. L. REV., April
1992, at 34.
  Mayer, Slicing the Mellon: Behind the Good Bank/Bad Bank Scheme, BARRONS, Aug. 15,
1988, at 100; Restructuring the Healthy Bank, BANKERS MAGAZINE, Nov.–Dec. 1988, at 35.

         In other words, the good bank/bad bank strategy can be employed with or without
actual failure, and with or without the participation of the government. In fact, there is some
indication that even regulators are beginning to realize benefits from shifting the management
of failed institutions to the private market.10 Thus, a number of alternative scenarios for
organizing the assets of insolvent and nearly insolvent banks have evolved. For example, a
bank might spin-off its non-performing assets into a separate institution. Alternatively, a bank
might be acquired and, following the acquisition, the low quality assets might be spun-off
into a separate entity while retaining the acquired bank as the repository for the performing
assets. Similarly, a bank might be acquired and split into two new institutions, each managing
separate types of assets.
         It is worth noting that each of these alternatives is available in both the private market
as well as under situations of public receivership. When the insuring agency takes control, it
could choose to manage the assets of an insolvent institution separately, or consolidate those
bad assets in one entity, such as the use of the asset management corporation Securum in
Sweden, or the use of the Resolution Trust Company (RTC) in the United States. While the
U.S. RTC was a wholly governmental entity, it is probably more accurate to view Sweden’s
Securum as quasi-private, quasi-public entity. Securum was established in 1992 to manage
the bad assets of Nordbanken. Securum took over troubled assets with an original book value
of SEK 67 billion, consisting of SEK 60 billion in loans and SEK 7 billion in real estate and
other tangible assets. These assets were acquired for SEK 50 billion (having been written
down by SEK 17 billion by Nordbanken prior to their transfer to Securum).
         Securum itself was financed with loans from Nordbanken of SEK 27 billion, and an
equity infusion of SEK 24 billion from the state of Sweden. As with other bad banks,
Securum’s mission was to manage and, ultimately, liquidate Nordbanken’s assets, recovering
as much as possible from debtors, and minimizing the total costs to the State and taxpayers.11

    INT'L FIN. L. REV., supra note 8, at 34 (indicating a trend toward a preference for the
private market for the management of troubled assets retained by the government).
      Securum AB, 1992–1997 (English Version) at 2.

4. A Brief History of the Good Bank/Bad Bank Strategy
This good bank/bad bank restructuring strategy first emerged on a large-scale in the mid-
1980s in the United States. In 1985, the FSLIC first used the strategy to deal with Westside
Federal Savings and Loan of Seattle. The FSLIC created a good bank called Mariner Federal
Savings and Loan, an entity later acquired at a substantial premium. At the end of 1986, the
FSLIC had employed this restructuring strategy with at least ten additional institutions.
Already in 1984, the FDIC bailed out Continental Illinois Corporation. The FDIC's
Continental bail-out plan, however, resembles a good bank/bad bank structure, as the FDIC
acquired $3.5 billion in bad loans and began to try to manage these loans by either selling or
collecting on them.12 In essence, Continental's non-performing assets were placed in a
governmental version of a bad bank.
          The 1986 acquisition of Texas Commerce Bancshares by Chemical New York
Corporation is another example of a deal that included the creation of a separate entity to
house the acquired bank's troubled loans.13 The Chemical acquisition was facilitated by a
measure of regulatory flexibility in obtaining the necessary capital requirements to receive
approval for the acquisition.14 The Chemical transaction illustrates the various means by
which the merger can be facilitated, such as offering an unleveraged pool of non-performing
assets as consideration.15 Also, in 1986, Midland Banks established a separate entity to deal

      Rose, supra note 7, at 43.
   Phillip L. Zweig & Steven Frazier, Chemical to Buy Houston Bank for $1.19 Billion,
WALL ST. J., Dec. 16, 1986 (describing acquisition of Texas Commerce Bancshares by
Chemical New York Corporation and the creation of a separate entity to hold a portion of
Texas Commerce's nonperforming loans).
    “[R]egulators may be willing to accept commitments to restore capital promptly following
the completion of a good bank/bad bank spin-off.” INT'L FIN. L. REV., supra note 8 at 34
(citing the Chemical/MHT transaction as evidence of regulatory willingness to allow
flexibility for achieving capital requirements following a restructuring). But, “they have far
from embraced the structure ... [and] the Federal Reserve has shown that it will closely
scrutinize any proposed transaction” and should the bad bank take the form of a trust,
partnership or corporation, the SEC will also place the entity under close scrutiny given the
implications the institution will have for the Investment Company Act. Id.
   Edward Herlihy et al., Financial Institutions—Mergers and Acquisitions 1996: Another
Successful Round of Consolidation and Capital Management, Corporate Law and Practice
Course Handbook Series, PLI Order No. B4-7179, Feb. 1997, at 386 n.6 (strategy utilized by

with the troubled loans of its Crocker National Corporation, facilitating the sale of Crocker to
Wells Fargo.16
            In connection with its 1988 recapitalization, Mellon Bank Corporation implemented a
very successful good bank/bad bank restructuring strategy. The Mellon strategy is often cited
as an important example of this concept. The Mellon strategy took the form of a liquidating
bank structure. Grant Street National Bank was chartered, taking the form of a limited-
purpose bank structure under voluntary liquidation pursuant to the National Bank Act. Grant
Street became the bad bank and was under jurisdiction and supervision of the Office of the
Comptroller of the Currency throughout the liquidation process.17
            The Mellon deal cost $166 million and absorbed $142 million in losses from the sale
of non-performing assets to the bad bank.18 The deal was financed through a sale of $513
million in junk bonds and other debt by Drexel Burnham Lambert, Inc.19 The new bank was
further capitalized by a $128 million contribution by Mellon itself.20
            The good bank/bad bank strategy continues to be utilized in the United States. For
example, a number of bank holding companies (most notably BankAmerica and First
Chicago) began probing the possibility of using bad bank securitization strategies for troubled

Chemical in its 1987 acquisition of Texas Commerce Bancshares and First Interstate in its
1988 acquisition of Allied Bancshares).
   C. Thomas Long & Jamie P. Bowman, Regulation of Savings and Loan Associations and
Saving and Loan Holding Companies and Related Issues, Corporate Law and Practice Course
Handbook Series, PLI Order No. B4-6872, Apr. 27, 1989, at 221-23.
    ”Grant Street was subject to various regulatory restrictions, including lending limits and
restrictions on holding foreclosed properties for longer than five years. Since Grant Street did
not accept deposits, did not grant credit to the public in the ordinary course of its business and
was not insured by the FDIC, it was not deemed to be a bank for Bank Holding Company Act
purposes. It did qualify as a bank, however, for Investment Company Act purposes, thus
exempting the activities of Grant Street from the provisions of that statute.” Herlihy et al.,
supra note 15 at 393.
   Stephen Kleege, Mellon Inspires ‘Bad Bank’ Plans for Property Portfolios, AM. BANKER,
Feb. 5, 1991, at 6.

real estate assets as early as 1992. Permutations of the good bank/bad bank strategy have also
surfaced where a full-fledged restructuring is too difficult or expensive.
       Since the initial innovation of the good bank/bad bank restructuring scheme in the
banking industry, other industries have borrowed the concept of spinning off bad assets and
liabilities, including the real estate industry,21 the insurance industry,22 and even construction

    Kleege, supra note 18, at 6; Lynn B. Sagalyn, Conflicts of Interest in the Structure of
REITS, REAL ESTATE FINANCE, Summer 1996, at 34 (describing the use of a good bank/bad
bank format in real estate); Adrienne Linsenmeyer, Solving the Real Estate Crisis, FINANCIAL
WORLD, Mar. 19, 1991 (describing the high frequency of proposals to solve problems in the
real estate market with a good bank/bad bank concept); Stan Ross & Dennis Yeskey, Real
Estate Relief, BEST'S REV. - PROPERTY-CASUALTY INS. ED., Jan. 1993, at 31 (discussing
option for life insurance companies to use the strategy to confront non-performing real estate
loans and real estate owned).
    Ruth Gastel, Insolvencies/Guaranty Funds, III INSURANCE ISSUES UPDATE, Dec. 1997
(describing the National Association of Insurance Commissioners' white paper and standards
for good bank/bad bank restructuring models in the insurance industry); A.M. Best Reviews
Emerging Insurer ‘Good Bank/Bad Bank’ Structures, Discusses Its Role and Rating
Implications, PR NEWSWIRE, Dec. 27, 1995 (describing the increasing use of the strategy in
the insurance industry); CIGNA Good Bank/Band Bank Debate Rages, BESTWIRE, Dec. 18,
1995 (discussing the arguments for and against the CIGNA good bank/bad bank plan);
Shifting Insurance Liabilities: A Risk to Policy-Holders?, CORP. LEGAL TIMES, Aug. 1996, at
46 (roundtable discussion, including discussion of the good bank/bad bank model in the
insurance industry); Ernst & Young, Who's Afraid of Environmental Liability?, MONDAQ
BUS. BRIEFING, June 18, 1996 (describing insurance restructurings using the good bank/bad
bank strategy); Jan H. Schut, Lloyd's Last Chance, INSTITUTIONAL INVESTOR, May 1996, at
113 (describing the reinsurance strategy of Lloyd's of London to runoff liabilities to Equitas);
Robert Stowe England, At the Brink, FINANCIAL WORLD, Nov. 21, 1995, at 70 (same); Lloyd's
of London: Solvent Abuse, ECONOMIST, Apr. 29, 1995, at 88 (same); Jan H. Schut, The
Emilco Enigma, INSTITUTIONAL INVESTOR, Mar. 1996, at 37 (describing insurer Emilco's
restructuring plan); John H. Snyder et al., The Industry at a Turning Point, BEST'S REV. -
PROPERTY, CASUALTY INS. ED., Jan. 1996, at 36 (insurance industry); Dave Lenckus, CIGNA
Critics Unmoved Despite Best Defense, BUS. INS., Jan. 1, 1996, at 1 (describing CIGNA's
restructuring plan loosely modeled on the good bank/bad bank strategy); Mary Jane Auer,
Spinning Off Liabilities Into a Separate Company May Bring More Pain than Gain,
INSTITUTIONAL INVESTOR, Nov. 1997, at 43 (criticizing the good bank/bad bank model for
dealing with insurance company liabilities, especially as used to deal with asbestos and
environmental liabilities).
   Brad German, Paradise Lost: California Builders Suffer Losses, BUILDER, May 1993, at
78 (describing the initial trial of the strategy in the construction industry).

       The good bank/bad bank strategy has been suggested or employed for banks in
numerous countries facing banking crises, including France,24 Germany,25 the Czech and
Slovak Republics,26 Thailand,27 China,28 New Zealand,29 and Brazil.30 Japan, with its

    Howard C. Gelbtuch & Takashi Kataoka, Real Estate Securitization Gaining Favor in
France and Japan, REAL ESTATE ISSUES, Aug. 1997, at 1 (“In France ... many banks began
setting up ‘good bank’/‘bad bank’ structures ... [E]quity investors have generally reacted
favorably to the French restructurings, pushing up share prices as problem real estate is
removed from a company's books.”); Brussels Irked by French Evasions Over Bank Rescue:
Taxpayers May Have to Dig Deep to Sort Out Credit Lyonnais, FIN. TIMES (London), Nov.
26, 1997, at 2 (discussing the separation of Credit Lyonnais's assets by stripping the non-
performing ones into a bad bank called the Consortium de Realisation, and the French finance
minister's call to have the good bank scrutinized differently than the bad bank).
   David Shirreff, The Ultimate Stress-Test, EUROMONEY, Sept. 1997, at 117 (describing a
consortium of big German banks securing a good bank/bad bank split to deal with the
Bayerische Kreditbank (BKB) problem).
     Good Bank, Bad Bank, EAST EUROPEAN INS. REP., Apr. 1997 (describing the Czech
National Bank's splitting of Agrobanka into solvent and insolvent parts); Michael S. Borish et
al., Banking Reform in Transition Economies, FINANCE & DEVELOPMENT, Sept. 1995, at 23
(describing the good bank/bad bank approach of the Czech and Slovak Republics to recover
non-performing loans as “feasible when governments are committed to fiscal prudence, rapid
bank privatization, and changes in bank management and governance.”).
    Siriporn Chanjindamanee, Tarrin to Seek Cabinet Nod for “Good Bank,” World Sources
Online, Inc., Emerging Markets Datafile, Jan. 13, 1998 (discussing Thailand's official
financial sector restructuring plan based on the Swedish good bank/bad bank model);
Technically Insolvent Thai Finance Firms Propose Joint Rehabilitation, AGENCE FRANCE
PRESS, Oct. 30, 1997 (describing proposal by 58 suspended finance firms to establish a good
bank/bad bank structure where the good assets would be grouped under license of
Chatphaibul Finance and the bad debts would be bought up at a large discount by Asset
Management Corp.).
   Gordon G. Chang, China, INT'L FIN. L. REV., July 1997, at 21 (describing $600 billion in
non-performing bank loans in China and the framework for a solution based on the good
bank/bad bank structure, yet also illustrating Beijing's unwillingness to adequately fund the
   In 1990, the Bank of New Zealand (BNZ) established, with limited success, a bad bank,
Adbro, while BNZ became a good bank. Terry Hall, Public Outcry Fails to Fend Off Bid for
BNZ—The Pain Caused by Australia’s Takeover of a Neighbor’s Financial Icon, FIN. TIMES
(London), Nov. 6, 1992, at 30.
    Jim Freer, Slow Adjustment: Brazil's Economic Policy, LATIN FIN., Dec. 1995, at 40
(anticipating use of the strategy in Brazil).

financial crisis partially motivated by non-performing bank assets, is just beginning to
consider the strategy and its potential within Japan's unique syndicate structure and critical
financial situation.31
             Perhaps the most well known good bank/bad bank strategy, however, was developed
in Sweden and launched there in 1992.32 As of 1995, costs and work force had been cut by 20
percent and Nordbanken had become Sweden's most profitable bank.33 With pounds 4.5
billion in bad debts, the Swedish government instituted a pounds 4 billion bailout. As of
1995, of that amount, the government was expected to recoup pounds 2 billion through
proceeds, taxes, and a special dividend.34
             In all, good bank/bad bank restructuring provides a unique means for disposing of
non-performing loans and returning an institution to health. One scholar has summarized
some of the benefits of the strategy:
                    The good bank-bad bank structure conveys three important
                    benefits to the banks and thrifts that use it: (1) It immediately
                    improves the bank's earnings power by improving the ratio of
                    earning assets to costing liabilities and lowering its operating
                    expense ratio (2) It immediately enhances the bank's ability to
                    raise capital from external sources at reasonable cost, by
                    improving the bank's earnings power and reducing the level of

    Jeanne B. Pinder, Headache for Japanese Investors, N.Y. TIMES, June 2, 1993, at D1
(Japanese banks “large-scale, syndicated variation of the ‘good bank, bad bank’ strategy” to
deal with problem real estate investments). But see Kenichi Ohmae, Five Strong Signals of
Japan's Coming Crash, WASH. POST, June 28, 1998, at C1 (discussing the difficulty of
isolating losses given the syndicate structures normally followed by Japanese banks and
advocating that Japan adopt the Swedish model of creating an institution to facilitate the
successful separation of the good assets from the bad); Is the Asian Flu Fatal? Asian
Economic Crisis, REASON, May 1998, at 18 (Brookings Institution economist Robert Litan
arguing that in Japan, “if they decided to take the good bank/bad bank approach, as a practical
matter there are very few people or institutions in Japan that are healthy enough to buy the
unhealthy stuff.”).
       Mary Brasier, Swedish Bank Massage, DAILY TELEGRAPH, Oct. 13, 1995, at 25.

                 nonperforming assets; [and] (3) It allows the bank to play
                 offense again. A high level of nonperforming assets has
                 enervating effects on a bank's ability to compete aggressively
                 and plan for the future. A return to competitiveness my be the
                 most important benefit that the good bank-bad bank structure

5. Hurdles to the Implementation of the Good Bank/Bad Bank Strategy: Securum’s Advantages
Several incentives drive the use of the good bank/bad bank strategy. Regulatory agencies have
an incentive to use the strategy when the losses to depositors or the government can be
decreased. Private entities have an incentive to increase returns (i.e., avoid losses) on troubled
assets through the benefits of separating the management of the assets. Private acquirors have
an incentive to realize the profits available from restructuring underperforming banking firms.
Additionally, in many countries, the tax advantages of spinning off bad assets provides
another incentive.36 A remaining good bank is interested in obtaining the benefits of the tax
code resulting from the spin-off, i.e. to record on its own books the loss inherent in the assets
transferred to the bad bank.
          Despite these incentives, there are a number of significant hurdles to implementing a
good bank/bad bank strategy. In particular, the transaction costs—such as legal and
underwriting fees—associated with establishing and operating a good bank/bad bank
structure may be high.37 Moreover, particularly where early intervention is being attempted by
the regulatory authorities, there may be considerable disagreement over which assets are bad
and which are good. However, in the case of Securum, the problem was not so much that
intervention came late (which is the typical problem that has plagued other countries that
have been faced with banking crises, including the U.S. and Japan), but that the crisis was
exaggerated. The available evidence suggests that there was overreaction and panic in the

      Hurley, supra note 7, at 4.
    Long & Bowman, supra note 16, at 221–23 (describing the potential tax benefits of
restructuring pursuant to the good bank/bad bank structure).
      Rose, supra note 7, at 47.

face of what were, in reality, temporary liquidity problems within the Swedish banking
industry. This atmosphere of crisis and panic made Securum’s job easier by creating a climate
in which it was possible to under-state the value of the assets being transferred to Securum.
By the time Securum was organized to accept the assets of Nordbanken, the prevailing view
was that Nordbanken was in acute financial distress and there was a clear consensus that the
majority of the loans taken by Securum were in default.38
         The bad bank strategy is easy to implement when late intervention occurs because the
assets taken by the bad bank are relatively easy to evaluate. In the case of Securum, the bad
bank strategy was similarly easy to effectuate, albeit for a different reason: the atmosphere of
crisis and panic surrounding the Swedish banking crisis made it possible for Securum to
aggressively write-off the value of the loans it was assuming. This made Securum’s job of
resolving these assets a more attainable task.
         Still another problem with establishing a successful bad bank is finding the
appropriate personnel. As the discussion below makes clear, one of the principle strategic
advantages to employing a bad bank strategy is that the human capital skills necessary to
manage and resolve the problems associated with troubled assets are entirely different than
the human capital skills that bankers must possess to deal with performing loans. A principle
justification for utilizing a good bank/bad bank strategy is to move bad assets into
institutional structures where qualified people who specialized in managing bad assets can
deploy their skills. However, for this strategy to succeed, it must be possible to recruit people
with appropriate industry-specific asset management skills to manage the bad assets. Indeed,
this was one of the principle justifications for Securum:
                One of the basic ideas when Securum was created was that it
                would need a different type of expertise than a bank if it were to
                enhance the value of assets in real estate, industrial companies
                and hotels. Securum recruited qualified personnel with
                industry-specific know-how to the various asset management
                companies and their boards of directors. Securum then

      Securum AB 1992–1997 supra note 11, at 13.

                    supported them as an active owner.39

Of course, finding such qualified, specialized personnel to work in a bad bank is not easy, and
in some cases it is impossible. The problem of attracting and recruiting suitable personnel is
particularly acute for bad banks because these institutions, when they are successful, will
convert bad bank assets into cash expeditiously and efficiently. This means that successful
bad banks will be dismantled relatively quickly. Indeed, one of the measures (although by no
means the only measure) of a successful bad bank is how quickly it is able to convert non-
performing loans into cash so that it can be dismantled. This, of course, adds to the difficulty
of finding adequate personnel for bad banks. And, where appropriate personnel cannot be
found, it is by no means clear that a good bank/bad bank strategy will be successful.
           Another obstacle to implementing bad bank strategies is the administrative and
technical hurdles to evaluating the assets that come into a bad bank. In this regard, Securum
had several advantages from the beginning that many bad banks do not enjoy. First, all of
Securum’s assets came from the same place, Nordbanken. This meant that, unlike bad bank
structures that accept assets from several troubled financial institutions, Securum was not
forced to cope with the burdens of rationalizing and coordinating a variety of computer
systems, filing systems, and information retrieval systems. Moreover, the documents that
Securum had to work with were more uniform than they would have been if they had come
from a variety of sources. This relieved the burden on the Securum personnel charged with
evaluating loan documentation. In addition, because all of Securum’s assets came from a
single source, it was possible for Securum to pattern its organizational structure after that of
Nordbanken’s. Securum did this by adapting both its finance company’s organization and its
real estate operations to Nordbanken’s regional structure. For other assets, Securum
established an organization with one unit to deal with both hotels and tourism, and one unit to
deal with industrial operations. By adopting an organizational structure that mirrored
Nordbanken’s, Securum was able to achieve “speed and flexibility” in the disposition of

       Id. at 15.
       Id. at 25.

          Another key advantage to having a bad bank dedicated to receiving the bad assets
from a single firm (Nordbanken) relates to the need to recruit qualified, knowledgeable
personnel. Securum recruited key personnel from Nordbanken. These people were not only
available, of course, but they also had both general expertise in dealing with problem loans,
as well as particularized knowledge of the particular assets coming into Securum. Of course,
it will not be possible to recruit individuals who are as familiar with the bad asset portfolio
when the bad bank must accept assets from a variety of financial institutions.
          Again in this context it is important to realize that late intervention makes it easier for
the bad bank to achieve its objectives. The potential problem with having a bad bank recruit
personnel from the troubled financial institution whose assets it is accepting is that some of
these personnel may have been involved in the original decision to grant the non-performing
loans that now constitute the assets of the bad bank. Where this occurs, potential conflicts of
interest arise since the personnel involved in the original decision to grant the loan may be
unwilling to recognize a problem or otherwise take responsibility for an earlier error.
          In the case of Nordbanken and Securum this problem was reduced because most of
the loans taken by Securum were already in default. Consequently, the Nordbanken personnel
hired by Securum generally were not the people who had granted the loans, but rather the
personnel who administered the loans. These people not only did not have the same conflicts
of interest as the people who had granted the loans, but also they had developed some
particularized expertise in resolving troubled loans. Even in this situation, however, it appears
that personnel problems arose within Securum, and that dealing with these problems often
was “costly from a financial standpoint.”41
          Still another special advantage that Securum enjoyed is an advantage that will not be
available to every bad bank. In particular, this advantage may not be available to bad banks in
developing economies. This advantage is that Securum only accepted assets above a certain
minimum size, SEK 15 million. Securum took over loans to just over 400 borrowers (3,000
loan commitments altogether). Securum’s task of resolving these troubled loans was made
easier to the extent that it is easier to administer a relatively small number of large loans than
to administer a large number of small loans.

      Id. at 29.

           Thus, while it is undoubtedly true that it is better to remove bad assets from problem
banks than to leave them within such banks, it is important to recognize: (1) the bad bank
structure is not cost-free; (2) the bad bank structure is not the only mechanism for
accomplishing the objective of removing bad assets from problem banks; and (3) in certain
contexts, notably that of Securum, the bad bank arrangement may work better than in other
contexts. The problems of implementing a successful bad bank strategy have led several
financial institutions to opt for less complex solutions to cope with their troubled loan
problems. These alternatives include individual asset sales, bulk sales to private investors
(Chemical bank), bulk sales through public or private auctions (as with California Federal and
Glendale), and other aggressive disposition and work-out strategies.42
           One of the most formidable barriers to a good bank/bad bank organizational structure
involves capital acquisition, including the demands by regulatory authorities that the bad
banks (and good banks) raise more equity capital in order to meet capital requirements for
continuing operations.43 Modern capital adequacy regulations serve to impede good bank/bad
bank projects. When troubled assets are separated and “marked to market on transfer to an
unconsolidated entity,” or when the bad bank's equity is dispersed to shareholders, capital will
be reduced. Restoration of this capital may be required in order to receive approval from
regulators. For example, in the case of the Mellon Bank transaction, $200 million in new
equity securities were sold concurrent with its announcement of its bad bank plan. Financing
for the Mellon strategy and its collecting bank, Grant Street, largely resulted from use of the
junk bond market. The decline of this market makes the private financing of bad bank
strategies much more difficult.44 Some of this difficulty is the result of regulators
unwillingness to approve certain types of financing structures necessary to implement the

    Herlihy et al., supra note 15, at 425 (strategy utilized by Chemical in its 1987 acquisition
of Texas Commerce Bancshares and First Interstate in its 1988 acquisition of Allied
       Rose, supra note 7, at 47.
       Herlihy et al., supra note 15, at 388.
   Id. Herlihy et al. explain this type of impediment in relation to Premier Bancorp and the
possibilities for future financing strategies:

           Given the difficult regulatory hurdles, as well as the logistical aspects of planning,
applying and financing the project, the process of implementing a good bank/bad bank
restructuring will be at best lengthy, and in all likelihood impossible without the active
support of the government.46
           The next sections review in somewhat greater detail the principal economic
advantages of the good bank/bad bank strategy. Determining what economic principles
motivate the attraction of the good bank/bad bank strategy is critical to understanding its

6. Specialization
Many advocates of the good bank/bad bank strategy argue that the primary economic
rationale behind the strategy involves the ability to jettison or dump non-performing assets,
consequently freeing an institution of the albatross of bad loans and reaping the tax benefits
of disposing of the bad assets at a loss. While this is undoubtedly true, it is a simplistic
characterization of the possibilities available in asset management following a good bank/bad
bank restructuring. It assumes that the performing loans will exhibit no net increase above
their market value at the time of restructuring. In other words, this explanation focuses too
much on the advantages to the good bank, and not enough on the advantages to the bad bank.

           In early 1990, Premier Bancorp abandoned its plans to spin off a collecting
           bank because of its inability to obtain the necessary mezzanine financing.
           Federal regulators had ruled that it would be unacceptable for Premier's main
           subsidiary bank to provide partial mezzanine financing for the collecting bank,
           and Premier was unable to obtain the needed financing from unaffiliated
           investors. Over time, third-party mezzanine financing may once again be
           available for good bank/bad bank structures.
             Even with improved market conditions, difficulties in financing good
           bank/bad bank structures are likely to persist. Despite improvements in the
           market for mezzanine financing, it may be difficult to finance a special asset
           company at full asset value levels. Structures involving larger equity
           components (possibly to be provided by existing shareholders) and/or risk
           sharing formulas may be required.
    Id. (reporting that it can take up to a year, during which market conditions could change to
the detriment of the strategy).

Moreover, this explanation relies too much on the artificial profit associated with the tax
advantages of restructuring.
          As suggested above, a better rationale for adopting a good bank/bad bank strategy is
that such a strategy permits both the owners of the good assets and the owners of the bad
assets to reap the benefits of specialized management in order to maximize the returns on
each type of asset. Alternatively, when the assets are grouped together, management must
constantly face conflict over the application of scarce firm resources to each.
          As discussed in the next section, the market’s ability to identify the quality of assets is
often low. Thus, non-performing assets may drag the good assets down as investors are
unable to delineate between the two, subsequently decreasing the value of both. When good
and bad assets are managed together, they act as rivals.
          George Stigler has described the inefficiencies that result from a firm’s management
of rival functions.47 He describes these rival functions as those for which ”the greater the rate
of output of one process, the higher the cost of a given rate of output of the other process.”48
Given finite resources, the processes within a firm will always compete. For example, this
problem arises “when the entrepreneur must neglect production in order to supervise
marketing.”49 A manager of a firm with large number of assets in both performing and non-
performing categories must necessarily neglect production of one to serve the other. It is
merely an application of the economic concept of opportunity costs. This rivalry occurs in
banking institutions with significant holdings of both good and bad assets. As Crockett
explains, eliminating this conflict can increase the performance of all assets involved:
                    Separating the workout and disposition of low-quality assets
                    from the healthy institution also eliminates the potentially
                    adverse effects of having both performing and nonperforming
                    assets managed by the same institution. For example, an

   See George J. Stigler, The Division of Labor is Limited by the Extent of the Market, 59 J.
POL. ECON. 185, 185, 187, 185–93 (1951), reprinted in GEORGE J. STIGLER, THE
   See Stigler, reprinted in TIM JENKINSON, READINGS IN MICROECONOMICS, at 144
      Id. at 142.

               aggressive collection program or a policy with strict limits on
               forbearance toward troubled borrowers may be consistent with
               maximizing the realization values of distressed loans and
               assets. Having a separate entity pursue this policy eliminates
               any potentially ill effects on the institution's reputation with
               prospective borrowers. Similarly, separate organizations make
               it possible to follow an optimal asset disposition policy
               independent of any concern about the effects of such a policy
               on the financial results that would be reported by a single

Thus, the task is to find a means by which rival processes can each see a net increase in
production without necessarily expending additional resources. This is precisely the economic
basis underlying the market’s tendency toward specialization. Remove the rivals, separate the
assets, and focus the attention on the particularized managerial task.
       Thus, specialization not only works because persons are able to concentrate their
energy on one task, but also because marketplace incentives prompt people to concentrate in
areas where they hold (or can develop) a particularized skill. This process leads to the more
efficient allocation of human resources and brings the concomitant increase in efficiency in
the efforts of those resources.51 In contrast to the separation of good and bad assets in
Securum, take the case of Swenska Handelsbanken. In Handelsbanken, the good and bad
assets were not segregated and the responsibility for existing assets remained with the old
managers, who aggressively fought for recognition of losses. Consistent with the Securum
experience, it is the economics of specialization that provides the strongest justification for
the good bank/bad bank strategy.

  John H. Crockett, The Good Bank/Bad Bank Restructuring of Financial Institutions,
BANKERS MAGAZINE, Nov.–Dec. 1988, at 34.
    Geoffrey C. Hazard, Jr., “Practice” in Law and Other Professions, 39 ARIZ. L. REV. 387,
396 (1997) (“Specialization of knowledge and technique has so many benefits for society as a
whole that it is an irreversible process, at least short of an atomic or astrophysical catastrophe
that returns us all to being cave people.”).

           By dividing asset responsibility between the banks, each is capable of capturing the
specific skills necessary for managing distinct categories of asset quality because the
management of bad assets requires different skills than the management of good assets.52
According to this “axiomatic” principle,53 it is the productive advantages of skill
specialization that drive the division of labor.54 Thus, performance is inversely proportional to
the number of multi-skill tasks undertaken by a single individual.55
           Similarly, the return on the expenditure of human capital resources is increased when
those resources are directed to a specific skill. This is what Adam Smith describes as the
division of labor improving the “dexterity of the workman.”56 Superior knowledge, training,
and experience in a specific field makes one better at his job. Thus, managers able to
specialize following a restructuring will not only direct their skills to managing a distinct
category of assets, but their skills in that process should also improve, assuming they were
required previously to develop skill in handling both bad assets and good assets. In other

       Crockett, supra note 50, at 34.
       Hazard, supra note 51, at 390 (describing Smith's theory of specialization as “axiomatic”).
    ADAM SMITH, WEALTH OF NATIONS 11 (Kathryn Sutherland ed., 1993). Smith articulates
the principle that the division of labor, causing skill specialization, increases productivity:
“The greatest improvement in the productive powers of labour, and the greater part of the
skill, dexterity, and judgment with which it is anywhere directed, or applied, seem to have
been the effects of the division of labour.” Id.
    Adam Smith articulates the relationship:
        The division of labour, however, so far as it can be introduced, occasions, in
        every art, a proportional increase of the productive powers of labour. The
        separation of different trades and employments from one another, seems to
        have taken place, in consequence of this advantage. This separation too is
        generally carried furthest in those countries which enjoy the highest degree of
        industry and improvement.
Id. at 13.
       Id. at 15. Smith describes the impact:
            First, the improvement of the dexterity of the workman necessarily increases
            the quantity of the work he can perform, and the division of labour, by
            reducing every man's business to some one simple operation, and by making
            this operation the sole employment of his life, necessarily increases very much
            the dexterity of the workman.

words, quality will increase.
         The quantity of work also increases from specialization. This advantage results from
savings in time related to specialization. A concentration of resources upon completion of a
distinct task over time increases the speed by which that task can be accomplished
effectively.57 Furthermore, no time need be used inefficiently (or wasted) dealing with the
difficult loans spun-off to the bad bank at the good bank.58
         Specialists also are far more likely to discover innovations unique to their task. As
Adam Smith explains:
                [T]he invention of all those machines by which labour is so
                much facilitated and abridged, seems to have been originally
                owing to the division of labour. Men are much more likely to
                discover easier and readier methods of attaining any object,
                when the whole attention of their minds is directed towards that
                single object, than when it is dissipated among a great variety
                of things. But in consequence of the division of labour, the
                whole of every man's attention comes naturally to be directed
                towards some one very simple object. It is naturally to be
                expected, therefore, that some one or other of those who are
                employed in each particular branch of labour should soon find
                out easier and readier methods of performing their own
                particular work, wherever the nature of it admits of such

This effect is especially important in the productivity of bad banks. Managers are faced with

    Id. at 16 (“Secondly, the advantage which is gained by saving the time commonly lost in
passing from one sort of work to another, is much greater than we should at first view be apt
to imagine it.”).
   Rose, supra note 7, at 44 (“It can reduce the operating expenses, especially management
and staff time, of the ‘good’ bank that sold its troubled loans. The time spent working out
problem loans can now be devoted more profitably to other staffing needs.”).
      ADAM SMITH, supra note 54, at 17.

currently non-performing assets and must turn them around. After restructuring, not only can
managers bring the special skills needed for managing these types of bad assets, the theory of
specialization predicts that their focus on this distinct managerial task allows for the invention
of innovative profit-maximizing techniques.60 When dealing with bad assets, innovation is
deeply needed. When those assets remain consolidated with good assets, however, the bad
assets are not as likely to get the necessary attention required to discover new ways to realize
profit from these assets.
          The banking industry in particular has long recognized the benefits of specialization in
lending and other banking functions. For example:
                         Securitization also permitted an uncoupling of
                         the component functions served by banks and
                         thrifts as financial intermediaries, allowing for
                         specialization by institutions with comparative
                         advantages in one or more of these functions.
                         For example, many banks and thrifts chose to
                         focus more on the origination function of
                         lending and less on the funding function by
                         securitizing the loans they originated, thus
                         enabling them to expand loan volume faster than
                         deposit growth. At the same time, the risks
                         inherent in the traditional intermediary functions
                         of banks and thrifts that became more
                         pronounced as a result of volatile interest rates
                         in the 1970s and 1980s, such as prepayment risk
                         and interest rate risk, could be better managed

       Rose, supra note 7, at 44. Rose argues:
          [T]he management of the ‘bad’ banks can specialize in collecting whatever
          value remains in its newly inherited portfolio of problem loans. This kind of
          specialization can cut operating expenses and avoid destroying community
          goodwill for the ‘good’ bank, because it no longer has to hound hundreds of
          customers about repaying their loans.

                       through asset securitization, which essentially
                       provides a direct match between the asset and
                       the provider of capital.61

This matching process is correlated with both the specialized institution and the consequent
increase in the ability to price assets in the market once they are separated according to
performance value.
       The good bank/bad bank structure separates two distinct types of assets,62 placing
them under separate management authority. According to Bengt Dennis, ex-governor of
Sweden's central bank, the good bank/bad bank model “helps focus management resources on

    Joseph C. Shenker & Anthony J. Colletta, Asset Securitization: Evolution, Current Issues
and New Frontiers, 69 TEX. L. REV. 1369 (1991) (citing E. CORRIGAN, FINANCIAL MARKET
STRUCTURE: A LONGER VIEW 28 (1987)) (extension of securitization process in unbundling of
credit intermediation into multiple discrete pieces, presumably reflecting benefits of
specialization); J. HENDERSON & J. SCOTT, SECURITIZATION 6 (1988) (noting that
“securitization may allow and encourage firms to specialize in that part of the loan-making
process where they can exploit their competitive advantage as a possible low-cost producer of
their particular portion of the loan-making chain”); J. ROSENTHAL & J. OCAMPO,
SECURITIZATION OF CREDIT 14, 227–28 (1988) (explaining that securitization breaks the
vertically integrated process of lending into discrete series of steps and permits firms to focus
on a limited number of roles in the process to build a competitive advantage); GREENBAUM
& THAKOR, Bank Funding Modes: Securitization Versus Deposits, 11 J. BANKING & FIN.
379 (1987) (noting that securitization decomposes traditional lending process into more
elemental activities, thereby permitting intermediaries to specialize in those more basic
activities in which they enjoy a comparative advantage)). See also Naomi R. Lamoreaux,
Information Problems and Banks' Specialization in Short-Term Commercial Lending: New
England in the Nineteenth Century, in INSIDE THE BUSINESS ENTERPRISE: HISTORICAL
PERSPECTIVES ON THE USE OF INFORMATION 161, 180 (Peter Temin ed., 1991).
    John H. Crockett, supra note 50, at 32.
        One source of gain is in economies from management specialization.
        Managing a portfolio of distressed assets is a different activity from managing
        a solvent thrift institution. Splitting a failed institution into two components
        may reduce the demands on higher level managers who would otherwise be
        faced with information gathering and decision making in each of the divisions.
        To the extent that this streamlining results in more timely decisions and
        lowered costs of managerial coordination, cash flow benefits will materialize.
Id. at 34 (emphasis added). See also JOHN H. CROCKETT, ON THE GOOD BANK/BAD BANK

two distinct and separate functions.”63 It allows an institution to bring special management to
assets with special needs.
            Furthermore, management can perform better when it has a strategy focused on
particular assets.64 The management of the good bank no longer needs to struggle with the
non-performing assets. The management of the bad bank can accept their limitations, but
also, facing the challenge, know that their successful efforts will be identifiable and can be
rewarded. Their objective is more apparent and greater certainty in the duration of their
institution's survival also diminishes the indifference or nervousness that can impede effective
management.65 Also, managers and investors can better appraise the value of the assets under
their control and, consequently, can provide more accurate information to the market.66
            Additionally, excessive risk taking and other forms of moral hazard, which often
cause troubles for a banking institution can be curtailed through the use of a good bank/bad
bank strategy. The separation of assets into identifiable performance categories permits more
effective monitoring of managerial performance.67 This should give managers a greater
incentive to manage effectively in order to maintain their positions.68

   Confidence Seen as Key Factor Swede Outlines Way to End Crisis, NATION (Bangkok),
Dec. 11, 1997.
    Crockett, supra note 50, at 32–34. Crockett explains:
        Uncertainty ... is not conducive to managerial initiative. Unclear objectives
        and diffuse responsibility make it more difficult to summon managerial action
        that will maximize the value of the firm. To the extent that the restructuring
        approach avoids these disincentives, it may lead to increased cash flows over
Id. at 34.
   Id. (“Similarly, a well-defined objective and a finite horizon for workout specialists lead
to more efficient management of bad assets.”).
    Alchian, Information Costs, Pricing, and Resource Unemployment, 7 W. ECON. J. 109,
110 (1969) (“Like any other production activity, specialization in information is efficient.
Gathering and disseminating information about goods or about oneself is in some
circumstances more efficiently done while the good or person is ... able to specialize (i.e.,
while specializing) in the production of information.”).
      Crockett, supra note 50, at 34. Crockett explains that managers will have more

7. The Communication of Information on Asset Quality
An additional economic justification for the implementation of a good bank/bad bank strategy
stems from investor reaction to the segregation of assets and management. First, investors can
respond to the existence of separated management and the benefits that accrue from that
specialization. Second, as discussed earlier in this part, monitoring is less difficult when
assets are separated.
       The basis for such enhanced monitoring largely results from the increase in available
information on the value of assets stemming from management targeted by performance
level. Prior to restructuring, investment in both good assets and bad is hindered by the
market's inability to identify the risks associated with investment.
       Akerlof describes this as the “lemons problem.”69 When investors cannot easily assess
the quality of goods offered by sellers and sellers are assumed to have superior knowledge of
such quality, inefficiencies result because market investors cannot trust that they are making
an economically sound investment. They are averse to the risk that they may be investing in a
lemon.70 Dishonest sellers have the ability to misrepresent the quality of goods (i.e., offer

identifiable responsibilities over assets and that the performance of these assets can more
effectively be evaluated:
        It is reasonable to expect that separate organizations will make the task of
        monitoring and evaluating managerial performance simpler. In turn, this will
        have favorable effects on the incentive structure for managers. ... By
        stabilizing the good bank, managers and employees will perceive the benefits
        from continuing routine activities such as loan origination and marketing to
        prevent the runoff of core deposits [as opposed to those managers who see
        themselves as mere temporary caretakers].
   George A. Akerlof, The Market for “Lemons”: Quality and the Market Mechanism, 84
Q.J. ECON. 488–500 (1970).
    Id. While this is an expected response of a rational economic actor, some scholarship in
cognitive psychology identifies additional factors that may motivate this result. Carol M.
Rose, Giving Some Back—A Reprise, 44 FLA. L. REV. 365, 374 n. 57 (1992) (citing Roger G.
Noll & James E. Krier, Some Implications of Cognitive Psychology for Risk Regulation, 19 J.
LEGAL STUD. 747, 752, 776–77 (1990) (suggesting that people are generally risk averse,
where “intuitive risk perception weighs potential losses more heavily than potential gains
(‘loss aversion’)”).

“lemons”) in a market plagued by large information disparities. These sellers can act
opportunistically, preying on those with inferior information.71 The inability of buyers to
distinguish between honest and dishonest sellers also decreases the confidence investors will
have in making an investment.
          Problems in both identifying asset value and monitoring lead to decreased prices.
Because such a decrease affects honest and dishonest sellers alike, the honest sellers will no
longer find it profitable to remain in that market,72 depressing prices even further as the
confidence in asset management decreases.73 Akerlof's theory of the effects of severe quality
uncertainty is an extension of Gresham's Law which states that bad money tends to drive out
good money in an economy where both are in use.74 This mis-allocation of resources results
in deadweight social losses.75 A significant amount of scholarship has explored the
application of the Akerlof lemons hypothesis in a variety of fields.76

    Several articles provide examples illustrating that consumers without information, if they
purchase at all, tend to purchase low-quality products at high quality prices, absent the ability
to distinguish between them. See, e.g., Lucian A. Bebchuk, Litigation and Settlement Under
Imperfect Information, 15 RAND J. ECON. 404 (1984); Gregg Jarrell & Sam Peltzman, The
Impact of Product Recalls on the Wealth of Sellers, in EMPIRICAL APPROACHES TO
CONSUMER PROTECTION ECONOMICS 377 (Pauline M. Ippolito & David T. Scheffman eds.,
1986); Benjamin Klein & Keith B. Leffler, The Role of Market Forces in Assuring
Contractual Performance, 89 J. POL. ECON. 615 (1981); Sam Peltzman, An Evaluation of
Consumer Protection Legislation: The 1962 Drug Amendments, 81 J. POL. ECON. 1049
(1973); Carl Shapiro, Premiums for High Quality Products as Returns to Reputations, 98 Q.
J. ECON. 659 (1983); Thomas S. Ulen, The Coasean Firm in Law and Economics, 18 J. CORP.
L. 301 (1993). See also generally JACK HIRSHLEIFER & JOHN G. RILEY, THE ANALYTICS OF
   A similar analysis indicates that good managers will tend to leave troubled institutions as
profits fall and their talents can yield greater salaries elsewhere.
  See George A. Akerlof, The Market for “Lemons”: Quality Uncertainty and the Market
Mechanism, 84 Q.J. ECON. 488, 488–90 (1970).
      See id. at 489.
      Id. at 488.
   See, e.g., T. Markus Funk & Daniel D. Polsby, Distributional Consequences of Expunging
Juvenile Delinquency Records: The Problem of Lemons, 52 WASH. U. J. URB. & CONTEMP. L.
161 (1997); Walter Kamiat, Labor and Lemons: Efficient Norms in the Internal Labor Market
and the Possible Failures of Individual Contracting, 144 U. PA. L. REV. 1953 (1996)

          A good bank/bad bank restructuring alleviates the lemons problem facing many
troubled financial institutions. The segregation of assets allows investors to more accurately
verify the value of each bank's assets and better identify dishonest managers due to the
decreased monitoring costs associated with the separation. Capital market investors will be
less fearful that the good bank will fail.77 Additionally, when the bad bank becomes
controlled by a specialized management team that will gain more accurate information about
valuing and collecting loans, many investors will be willing to pay for that information,
thereby raising the value of the bad bank's assets.78
          And, perhaps most importantly, the fact that the bad bank management was not
involved in the granting of the original troubled loans adds a large measure of credibility to
the process. Where the managers of the bad bank are not associated with the original credit
decision, it stands to reason that they are far more likely to recognize fully the nature and
extent of the losses of the assets being held by the bad bank. There are two reasons for this.
          First, while the managers who originally made the loans may be reluctant for both
economic and psychological reasons to write-off the bad loans on their banks’ books, new
managers suffer no such perverse incentives. Once a loan decision has been made, an ample
body of literature in social psychology shows that this earlier approval will strongly affect the
later actions and decisions of the people who originally made and approved the loan. In
particular, the “theory of escalating commitments predicts that banks’ officers will come to
identify with the loans they have made, and that future decisions regarding those loans will be
made so as to be consistent with the earlier decisions.”79

(applying Akerlof theory to explain the absence of “just cause” contracts in the non-union
(Victor P. Goldberg ed., 1989) (noting that the economic term “[a]dverse selection now refers
to any situation in which an individual has knowledge about his own quality (the goods he
sells, his ability to perform, his health status) while whomever he is dealing with knows only
about the characteristics of the average member of the group”); William Samuelson,
Bargaining Under Asymmetric Information, 52 ECONOMETRICA 995 (1984) (examining
behavior of buyers holding different degrees of information).
      Rose, supra note 7, at 44.
      Id. at 45.
      T. Gilovich, “How We Know What Isn’t So: The Fallability of Human Reason in

            From an economic perspective, bank managers also have incentives to decline to
recognize the problems associated with loans they previously have made. Managers may
abstain from writing down troubled loans they have made in order to preserve their
reputations, their bonuses, and even their jobs. Obviously, taking corrective action to write-
off a bad loan reveals the error associated with a previous decision, and the bankers
associated with that earlier decision will be understandably reluctant to do that.
            Second, the new managers of the bad bank may have incentives of their own to write
down the value of bad bank assets. These incentives to mark-down bad loans aggressively in
some ways mirror the incentives of the previous managers to decline to make such mark-
downs. By aggressively writing down the value of the assets they acquire, managers of bad
banks can improve their own performance. Consider the typical ways in which the managers
of bad banks are evaluated. They are evaluated by: (1) how quickly they can dispose of the
assets under their control; (2) how the price received for the asset relates to the (newly written
down) value of the asset; and (3) whether the bailout cost more than expected. All of these
criteria suggest that managers will want to be aggressive initially in seeking write-downs of
assets. Assets that have been written down aggressively can be sold easily at prices equal to or
greater than their newly assigned values. Setting an initially high estimate of the cost of a
restructuring project makes it less likely that new money will be needed. In other words,
aggressively writing down the value of the assets of a bad bank causes expectations for the
assets to be low. And low expectations are always easier to meet than high expectations.
            This analysis seems to fit the case of Securum very well. First, Securum worked to
insure the impartiality of its personnel by ”trying to ensure that no one worked on the
divestment of loan commitments that he had helped to arrange.80 This strategy addressed
Akerlof’s “lemons problem” by ensuring Securum’s “credibility outside the company.”81 In
fact, decision-makers within Securum who were tied to previous loan commitments were
assigned to other jobs, and in some cases forced to resign from Securum. Second, Securum

Everyday Life,” D.G. MYERS, ”SOCIAL PSYCHOLOGY” 86 (1992); M. Rabin, Psychology and
Economics, 34 J. ECON. LIT. 11–46 (1998).
      Securum 1992–1997 supra note 11, at 30.

implemented an aggressive, incentive-based compensation plan. Securum’s top management
acknowledges that “[t]he newly appointed management first had to convince potential
employees to make themselves available and then motivate them to work as quickly as
possible, despite the fact that this meant they would work themselves out of a job more
quickly.”82 Interestingly, the rapid pace with which Securum divested itself of its assets is
cited as one of the primary indicia of its mission’s success.83
          Finally, it is noteworthy that Securum also cites the fact that “[t]he cost to taxpayers of
the financial crisis has been reduced significantly compared with what had been expected
when the crisis was at its worst point.”84 In other words, Securum is able to claim success
because the original estimates of the problem were so severe. Securum clearly beat the
expectations game. Securum, together with its sister organization, Retriva, was originally
capitalized with equity of SEK 27.8 billion. Initially, this entire sum was expected to be
consumed over the course of the lives of these two institutions. However, Securum was able
to return SEK 17.8 billion of the funds originally received.
          The point here is not that Securum did a bad job. The arguments are far more modest.
The first argument is that the organizational structure of Securum created clear incentives for
the firm initially to be extremely aggressive in its valuations of the assets that came under its
control. The second point is that, in light of this perverse incentive structure, it seems
inappropriate as a matter of good social science to use the original estimates of these assets’
values as the benchmark for evaluating Securum’s performance. A more objective benchmark
would seem appropriate; however, given the opacity of banks’ balance sheets, it is not at all
obvious how a more objective measure might be obtained where there is only a single bidder
for the assets of a troubled bank. Perhaps a better approach would be for regulators to attempt
to foster competition among bidders for the assets of a bad bank. In this way, the market
pressure of competition to acquire the assets would, at least to some extent, counter the
incentives of purchasers systematically to give initial underestimates of the value of the bad
bank’s assets.

      Id. at 31.
      Id. at 2.
      Id. at 3 (emphasis added).

           Despite these caveats, the process of moving assets from a troubled bank to a good
bank/ bad bank structure does increase the information about the nature of the assets in both
the good bank and the bad banks. Despite the biases of the managers of the bad bank
discussed above, the new estimates of value for the assets in the bad bank are almost certainly
going to be more accurate than the previous (book) values, primarily as a consequence of the
managers’ increased expertise and efficiency under the segregated crisis management
           This increase in information due to such restructuring and the concomitant benefits to
valuation of assets held by both the bad and good banks should attract investment. A study by
Flannery and Sorescu concluded that asset quality tends to have a greater impact on the
market than profitability.85 Investors will invest in assets they are able to value with a
significant degree of accuracy. Evidence tends to show that investor response has empirically
been positive in relation to bad bank restructurings and similar transactions.86

8. Incentives
Apart from benefits from specialization and from reductions in the cost of monitoring
discussed above, an additional benefit from establishing a good bank/bad bank structure is
that such a structure can make it easier to establish compensation systems that reflect both the

   Mark J. Flannery & Sorin M. Sorescu, Evidence of Bank Market Discipline in
Subordinated Debenture Yields: 1983–1991, 51 J. FIN. 1347, 1373 (1996)(“Asset quality and
market leverage show the most consistent effect on SND spreads, with profitability less
important and interest rate risk generally insignificant.”).
       INT'L FIN. L. REV., supra note 8, at 34. The following states this positive response:
           Recent bank merger transactions have illustrated that large special reserves can
           be established in connection with strategic transactions without unduly adverse
           consequences. Such reserves have generally been well received by the market,
           indicating that the capital lost as a consequence of a special reserve can be
           readily replenished if the institution is otherwise healthy. The market may also
           be more receptive to bad bank structures. ... Recent experience has also shown
           that collecting bank activities can be highly profitable. Several large banks
           (e.g., NationsBank, Banc One, Fleet/Norstar) have achieved significant
           success in managing troubled assets for their own account and for the account
           of third parties, including the RTC and FDIC.

skills and training involved and the value the use of those skills brings to the bank.87
Specialization and the division of labor facilitate adjustments to salaries based on individual
productivity.88 Those who produce more, earn more.89 In the case of bank management, those
who manage a greater return on assets should also earn more.
          As discussed earlier, performance of a bank is difficult to tie to specific assets or to
specific decisions when an underperforming bank consolidates the management of both its
good and bad assets. “[T]he ‘good’ bank and the ‘bad’ bank can each establish its own unique
salary schedule and bonus plan with less internal conflict over who deserves or doesn't
deserve a raise.”90 In other words, an incentive structure is more easily enforced under a
system of specialized and segregated management. With a properly functioning incentive-
based pay system, managerial performance should improve. And, as discussed above, this
seems to have been the case with Securum. The generous compensation packages did, in all
probability, contribute to the success of the organization. Of course, such compensation
packages are more easily defended in private firms than in governmental entities. However,
the job insecurity of Securum employees caused by the brief duration of their expected
employment periods makes it possible to justify what might otherwise appear to be above-
market compensation arrangements.
          Although the generous compensation packages given to bank managers heighten the
incentive to undervalue bad bank assets during the initial stage of crisis management, this is
not to say that the compensation packages are inappropriate. In fact, as discussed above, once
a system is implemented under which the bad bank assets could be more accurately valued—
such as with a competitive bidding system, bad bank managers might well perform even
better throughout the crisis management stage. By basing compensation to a greater extent on
the actual efficient disposition of the bad assets, and to a lesser extent on the managers’
subjective valuation of the bad assets, the compensation packages might well provide even

      ADAM SMITH, supra note 54, at 97–98.
      VON MISES, supra note 88, at 625–34.
      Peter S. Rose, supra note 7, at 44.

greater productivity gains than were achieved by Securum.

9. A Brief Look at the Good Bank
Non-performing assets have high maintenance and carrying costs. Eliminating these costs
immediately improves the position of the good bank. With restructuring, under-performing
assets are less likely to be written off as total losses. In other words, if the assets are kept
together, the emphasis will be on the performing assets. The gains from those assets will, in
effect, cover (or subsidize) the losses attributable to the bad assets. If the assets are split into a
good bank/bad bank structure, the performing assets will bring gains alone, and even
assuming no gains from specialization, the value of those assets should increase because they
will no longer be “guilty by association” with the bad assets. Furthermore, because the bad
assets will, presumably, be intensively managed once they are split off, they should,
independent of any gains from specialization, bring greater gains than if coupled with the
performing assets.
        A number of scholars have studied the financial outcomes of voluntary corporate
restructurings. Many studies have demonstrated, with statistical rigor, that such plans are
responsible for immediately increasing the value of those restructured firms.91 There is little
reason, other than perhaps the peculiar regulatory posture governing banks, to believe that
good bank/bad bank restructurings would not bring similar results.92

    See, e.g., James A. Miles & James D. Rosenfeld, The Effect of Voluntary Spin-Off
Announcements on Shareholder Wealth, 38 J. FIN. 1597–1606 (1983); Kathrine Schipper &
Abbie Smith, Effects of Recontracting on Shareholder Wealth, 12 J. FIN. ECON. 437–67
(1983); Gailen L. Hite et al., The Separation of Real Estate Operations by Spin-Off, 12 AM.
REAL ESTATE & URBAN ECON. ASSN. J. 318–32 (1984); Gordon J. Alexander et al.,
Investigating the Valuation Effects of Announcements of Voluntary Corporate Selloffs, 39 J.
FIN. 503–17 (1984); James D. Rosenfeld, Additional Evidence on the Relation Between
Divestiture Announcements and Shareholder Wealth, 39 J. FIN. 1437–48 (1984); Pren C. Jain,
The Effect of Voluntary Sell-Off Announcements on Shareholder Wealth, 40 J. FIN. 209–24
(1985); Gailen L. Hite & James E. Owers, Security Price Reaction Around Corporate Spin-
off Announcements, 12 J. FIN. ECON. 409–36 (1983).
    The Banking Crisis and Financial Reform; Norway; Industry Overview, OECD ECON.
SURVEYS—NORWAY, Mar. 1993, at 54. With effective communication of asset quality, the
strategy will increase returns:
        The danger is that the inefficiencies described above will accumulate, and the
        franchise values of state-owned banks deteriorate. A solution to this problem

10. Early Intervention
As indicated at the outset, one of the principle observations of this Report is that there is an
interesting trade-off between the timing of regulatory intervention in a banking crisis and the
likelihood of success of a good bank/bad bank strategy. In particular, this Report maintains
that, although early intervention can greatly reduce the costs associated with a bank bailout, it
also makes the task of distributing assets between the good bank and the bad bank more
difficult. This stems from the fact that bank assets are, by their very nature, highly opaque.
Consequently, for many bank assets there may be a wide range of uncertainty about the value
of these assets until there is an actual default. Thus, while early intervention is of vital
importance in reducing the costs of a bank bailout, the later the intervention, the less
controversy there will be over which assets will be allocated to the bad bank, and the easier
the task of the managers of the bad bank will be.
             Securum’s job was made easier by the fact that a majority of the loans taken over by
Securum were in default, and that those that weren’t in default “almost without exception
soon would be in default.”93
             The reason why early intervention is so important is because bank assets that are not
dealt with quickly tend to deteriorate in value very rapidly. This is a universal phenomenon
that exists in every banking crisis situation. The problem of rapid deterioration of assets was
also true in the case of Securum, as it was thought that the value of these assets would have
dropped very quickly if not taken over right away.94
             Part of the reason why bad bank assets rapidly deteriorate is explained by the moral
hazard facing the managers of the distressed bank whose assets are subject to being moved to

        may be found in the partial or marginal privatisation. ... [One option] is the so-
        called ‘good bank-bad bank’ method in which profitable operations and
        customer relationships are separated from doubtful assets, creating two legally-
        distinct banks. Presuming information about asset quality can be
        communicated effectively, interests in both banks can be sold to private
        investors, although only the ‘good’ bank will attract much of a return.
Id. (writing in the context of Norway's banking system).
       Securum 1992–1997, supra note 11, at 29.

the bad bank category. Part can be explained by the fact that the managers of distressed banks
tend to ignore their banks’ bad assets and focus only on the good assets because they do not
want to recognize themselves or draw attention by others to the fact that the bad assets are
dropping in value. Also, managers of distressed banks may lack the expertise to deal with
these assets in the first place.
        While there is little, if any, debate over the general policy question of whether early
intervention is desirable (clearly it is), there is some debate over the extent to which early
intervention is possible in light of the opacity of bank assets. The available evidence suggest
that early intervention is, in fact, possible.95 Evidence suggests that the market provides a very
good early warning of bank problems. For example, a study by Pettway and Sinkey of six
bank failures that occurred between 1973 and 1975 found that the market signaled problems
with all of these banks an average of thirty-three weeks before regulatory agencies placed
them on problem lists.96 Similarly, a study by Johnson and Weber indicates that the addition
of a bank to a regulator's problem bank list does not cause a significant market reaction,97
indicating that the market has already reacted to this information by the time the regulators
act. Similarly, when the names of banks on the U.S. Comptroller of the Currency's list of
problem banks were leaked to the press there was no significant market reaction to the
disclosure, indicating that the market already knew the regulator's information. And a study
by Shick and Sherman showed that stock prices of bank holding companies began to decline
fifteen months before regulators recognized that the subsidiary banks were experiencing
financial trouble.98

    Anderson and Viotti reach a similar conclusion about the lack of prompt crisis recognition
in Sweden. See Martin Anderson & Staffan Viotti, “Managing and Preventing Financial
Crisis—Lessons From the Swedish Experience,” PENNING- OCH VALUTAPOLITIK, No. 1,
   Pettway & Sinkey, Establishing On-Site Bank Examination Priorities: An Early-Warning
System Using Accounting and Market Information, 35 J. FIN. 137, 145 (1980).
  Johnson & Weber, The Impact of the Problem Bank Disclosure on Bank Share Prices, 8 J.
BANK RES. 179, 180–82 (1977).
  Shick & Sherman, Bank Stock Prices as an Early Warning System for Changes in
Condition, 11 J. BANK RES. 136 (1980).

          Because the market can identify underperforming or soon to be insolvent banks long
before failure, implementing a good bank/bad bank strategy prior to insolvency could work to
prevent that which might be a foregone conclusion without the strategy. Early intervention
could prevent additional damage to a bank’s assets. Thus, regulators should utilize the market
as a means for identifying and correcting problems before they ever reach the federal insurer.
Rather than create regulations that discourage either a spin-off (or takeover followed by a
spin-off ) into a good bank/bad bank system, regulators should encourage the use of any
strategy that can minimize the damage to a troubled institution and the damage to the
taxpayers. To fully meet this task, regulators should encourage both a market for corporate
control for banks and auction markets for the bad assets of distressed banks.
          The specialization factors encouraging good bank/bad bank restructurings are quite
similar to the factors involved in a takeover. After all, takeovers are motivated by an
acquiring institution’s belief that it has a superior ability to manage the acquired institution. In
other words, it has a comparative advantage. When a bank takeover is attractive only when
the acquiring institution can establish a good bank/bad bank system following acquisition, the
cost of accomplishing that takeover is especially high because the acquiring institution must
overcome the regulatory hurdles it faces in both acquisition and restructuring. Thus,
impediments to an open market for bank control diminish the instances of both control
transactions as well as good bank/bad bank restructurings.
          Because intervention can occur early on in a bank’s crisis, “[t]he good bank-bad bank
structure is most effective at the time it first becomes apparent that the institution cannot
generate core-operating profits until non-performing assets are significantly reduced.”99 It is
important to note that many poorly functioning banks could be averted from failure by
preemptive reorganization. Takeovers (including takeovers that come in the form of auctions
organized by regulators) are perhaps the most efficient means by which this is accomplished,
where the acquiring institution is able to identify under-performing banks and intervene to
reorganize the acquired institution.
          In other words, the motivations that have led to successful corporate reorganizations
could also be accomplished in the banking industry—partially through utilizing the good

      Hurley, supra note 7, at 4.

bank/bad bank strategy—if the market for bank control could be improved.
          These insights about takeovers are relevant in this context because the available
literature in financial theory indicates that weak and failing firms (especially those with weak
management teams) are particularly likely candidates for takeover, because they are likely to
provide the greatest opportunity for arbitrage profits by superior management teams. They are
prime candidates for an acquiring firm able to provide separate, specialized management
under the good bank/bad bank scenario. Indeed, in an insight that is particularly relevant for
banking law and policy, Professor Dewey observes that takeovers are “merely a civilized
alternative to bankruptcy or the voluntary liquidation that transfers assets from falling to
rising firms.”100 Thus, when properly viewed as a substitute for bankruptcy, a robust market
for bank control is seen as a sign of the vigor of competition in a particular industry, rather
than of its decline.101 Professor Henry Manne, in a seminal article, made a substantial
extension of Dewey's earlier insights. He observed that takeovers are not only efficient when
the failure of a firm is imminent, but also “before bankruptcy becomes imminent in order to
avoid that eventuality.”102 Manne goes on to point out that “if mergers were completely legal,
we should anticipate relatively few actual bankruptcy proceedings in any industry which was
not itself contracting. The function so wastefully performed by bankruptcies and liquidations
would be economically performed by mergers at a much earlier stage of the firm's life.”103
Manne's defense of the market for corporate control catalogues a host of advantages to the

   Dewey, Mergers & Cartels: Some Reservations About Policy, 51 AM. ECON. REV.
PAPER & PRAC. 255, 257 (1961).
     Id. This does not mean that takeovers of banks will only occur when failure is imminent.
Rather, takeovers are likely to take place whenever a firm is being run by inept or dishonest
management, even if the firm happens to be profitable. See Haddock, Macey & McChesney,
Property Rights in Assets and Resistance to Tender Offers, 73 VA. L. REV. 701, 709 (1987).
It stands to reason, however, that all else being equal, firms with inept or dishonest
management are more likely to fail than firms with competent, honest management.
Takeovers also are likely to occur when a firm's assets could be redeployed to more valuable
uses than the ones selected by incumbent management. Macey, State Anti-Takeover
Legislation and the National Economy, 1988 WIS. L. REV. 467, 469 (1988).
    Henry Manne, Mergers and the Market for Corporate Control, 73 J. POL. ECON. 110,
111 (1965).
       Id. at 112.

economy from a regulatory policy that encourages takeovers. A free market for corporate
control would bring about more efficient management of firms, increased protection for non-
controlling investors—stockholders, depositors, and other creditors—and a more efficient
allocation of resources.104
       The arguments in support of a robust market for corporate control apply with even
greater force to banks than they do for corporations generally. The monitoring function that
the corporate control market provides for all firms is a supplement to the monitoring
conducted by unsecured creditors in the case of the public corporation where creditors are not
protected by the government. Because depositors have very little incentive to engage in
monitoring under the current regulatory regime, which protects virtually all deposits, the need
for a robust market for corporate control is even more acute in banking than in other sectors
of the economy.
       The observations of Dewey and Manne that the market for corporate control both
serves as a substitute for bankruptcy and reduces the incidence of insolvency are particularly
relevant for the banking industry. Outside of banking, the creditors and shareholders who
make (uninsured) investments in a firm that becomes insolvent bear virtually all of the costs
of the insolvency. But if a bank becomes insolvent, much of the costs are borne not by
shareholders and uninsured creditors of the failed institution, but by the state. This provides
an even stronger argument for facilitating a robust market for corporate control—including
state-sponsored auctions—in order to reduce both the incidence of bank failures, and the costs
of those failures that do occur.

11. Conclusion
Bad banks are a useful and important strategy for dealing with the problems of failing banks.
The Swedish approach to its banking crisis involved perhaps the most committed and
pervasive division of ailing financial institutions into a good bank and a bad bank ever
attempted. Once in the bad bank, the asset received careful attention so that it could be
packaged and improved in order to be made attractive to potential buyers as quickly as

   See Frank H. Easterbrook & Daniel R. Fischel, Corporate Control Transactions, 91
YALE L.J. 698, 698 (1982); Haddock, Macey & McChesney, supra note 102, at 709.

       There is no question that the utilization of a bad bank framework is superior to leaving
bad assets in the distressed banks that made the original commitments. The use of the bad
bank framework such as that utilized in Securum facilitates a more accurate (albeit not
perfect) assessment of the real value of the assets of the troubled bank. In addition, the
utilization of the bad bank structure allows for the deployment of highly specialized experts
to bring their particular skills to bear on resolving the assets taken by the troubled bank.
       It is the conclusion of this Report, however, that a proper evaluation of the Swedish
experience with the financial crisis of the early 1990s would be seriously incomplete if it
focused exclusively on the crisis management conducted by Securum. Instead, it is important
to examine the entire response sequence, which includes not only the crisis management
conducted by Securum—and Sweden´s Bank Support Authority—, but also the crisis
prevention carried out by the Ministry of Finance, the Sveriges Riksbank (Swedish central
bank), and the Financial Supervisory Authority.
       The delays associated with lack of early closure in the crisis prevention phase of the
Swedish financial crisis had two consequences. First, the delays increased the overall costs
associated with the crisis by permitting bank assets to deteriorate prior to coming under
Securum’s control. Thus, regardless of how efficiently Securum functioned, the overall crisis
management component cannot be viewed as a success without also considering the costs
associated with the lack of early intervention. Second, the delay made Securum’s job easier
by lowering the difficulty of identifying, categorizing, sorting and valuing the bad assets. Put
simply, when there is early intervention, the decisions about which assets to allocate to the
bad bank and which to retain in the good bank are more difficult. While the causes of the
delay may be political in nature, rather than resulting from an inherent flaw in the good
bank/bad bank strategy, it is important to recognize the potential costs associated with the
delay when assessing Securum’s overall performance.
       Thus, delay in intervening to deal with a financial crisis may raise the overall costs of
resolving the crisis, but it makes the task of the managers of the bad bank easier. This is
particularly true where—as in the case of Securum—the bad bank is evaluated on the basis of
the difference between what was expected when the crisis was at its worst point and the
ultimate value received for assets by the bad bank. This evaluation scheme creates an

incentive structure in which the bad bank’s performance is a straightforward function of the
value of the assets that come under its control. All else equal, the lower the initial values, the
better the bad bank’s ultimate performance will be judged to have been. In the case of the
Swedish financial crisis, the problem was not so much that intervention came late: the
problem was that the extent of the crisis was exaggerated. The available evidence suggests
that there was overreaction and panic in the face of what were, in reality, temporary liquidity
problems within the Swedish banking industry. This atmosphere of crisis and panic made
Securum’s job easier by creating a climate in which it was possible to understate the value of
the assets being transferred to Securum.
        It is important to stress once again that this is not meant as a blanket criticism of
Securum. The point is not that Securum did a bad job. Rather, the point is simply that taking
an estimate of the value at the time they come into the bad bank does not permit an unbiased,
market-tested mechanism for establishing an initial benchmark against which to evaluate the
bad bank’s performance. This Report suggests alternatives such as setting up an auction, or
facilitating a takeover or merger or similar control transaction that would ameliorate this
problem. As things stand with the case of Securum, it will be extremely difficult to obtain an
objective, empirical measure of the success of the organization because it will not be possible
to know for certain how well the assets would have performed in some alternative ownership
        This Report offers some reasons why Sweden’s experience with Securum might not
be easily exportable to other economies currently undergoing financial problems. First, the
sort of specialized human capital skills that were brought to bear in Securum to deal with the
distressed assets put under its control simply may not be available in other economies
experiencing financial crisis. It is the availability of such specialized talent that forms the core
economic justification for establishing a bad bank structure. Absent such qualified personnel
with industry specific know-how, it is not obvious that a good bank/bad bank structure could
be made to work.
        Second, Sweden is an unusually honest country. Before introducing a Securum-like
structure in another country, one would have to consider whether it is possible to establish an
independent, entrepreneurial organization empowered to dispose of assets of significant value
without the problem of corruption destroying any value-added that might be derived from that

organization’s activities. The same point can be made about the assets that came into
Securum’s control. Securum established an irregularity committee to deal with fraud and
other irregularities in its loan commitments. However, while approximately half of the loan
files were subject to some form of criticism, fraud was proven in only a limited number of
cases. This relatively low level of fraud in connection with the assets coming into Securum’s
possession reduced Securum’s initial business risks and lowered the adverse financial
consequences associated with Securum’s activities. It is not obvious that this happy
experience would be replicated in other contexts. Along these lines, it is particularly
noteworthy that many of Securum’s employees came from Nordbanken. This significantly
reduced the “learning curve” associated with becoming familiar with Securum’s assets.
However, in a situation where fraud and self-dealing are more pervasive, this sort of hiring
pattern would not be possible.
        Third, as mentioned above, certain other factors contributed to Securum’s ability to
operate efficiently. Securum took a relatively small number of large assets. Such assets were
easier to manage than a larger number of small assets would have been. Securum, like its
sister entity Retriva, took all of its assets from a single entity. This entity was Nordbanken in
the case of Securum and Gota Bank in the case of Retriva. This rather unique situation
contributed to the speed and flexibility with which these bad banks were able to operate. The
bad banks could adapt themselves to the organization structure and geographic organization
of the banks from which they obtained their assets. And Securum was able to compensate its
personnel both generously, and on a performance basis. This, too, may be difficult to replicate
in other cultures.
        The rather unique nature of the Swedish financial crisis can be illustrated with
reference to the issue of crisis recognition. This Report stresses that the most important
variable in controlling the costs of a financial crisis is early crisis recognition. It is highly
unlikely that the rather peculiar Swedish experience with early recognition will ever be
replicated elsewhere. Unlike most jurisdictions (including the U.S. and Japan, as well as most
of the rest of Asia), where both regulators and bankers ignored the severity of the banking
crisis and engaged in regulatory forbearance by declining to close banks that should have
been closed, in Sweden the banks themselves voluntarily went to the government, described

the nature of the crisis and the extent of the problem, often in exaggerated terms.105
          And, as suggested above, if anything, Swedish regulators and bankers exaggerated the
severity of the problems facing the Swedish banks, rather than ignoring the problem for years
as was done in many other countries. As I observed in a 1994 Report:
                        [f]ar from ignoring the problem, in Sweden there
                        was a general overreaction. ... Unlike the
                        situation in the U.S. and internationally. ...
                        where regulators had to fight bankers in order to
                        obtain control of troubled banks, Swedish banks
                        have applied to the government for regulatory
                        assistance voluntarily. ...106

In contrast to the regulatory responses to the banking crisis in other countries, the response to
the crisis in Sweden was quick and more severe than truly necessary given the true nature of
the problems. In other words, the error bias in most countries has been in the direction of
ignoring the severity of the banking crises with which they have been faced. In Sweden, by
contrast, the error bias has been in the direction of overreaction. This Report has raised the
possibility that this overreaction may provide some explanation for the apparently exemplary
performance of Securum.

    Jonathan R. Macey, “The Future Regulation and Development of the Swedish Banking
Industry,” SNS Occasional Paper No. 56, May, 1994.
       Id. at 66.


To top