View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

o
FEDERAL
0
RESERVE
SPECIALIZATION IN RISK MANAGEMENT
Supervision of Derivative Instrum ents

BANK

OF

o
C

l e v e l a n d

o

o
by
0

Jerry L. Jordan
President and Chief Executive Officer
Federal Reserve Bank of Cleveland

presented at
C O N F E R E N C E O N C O P I N G W I T H F I N A N C I A L FRAGILITY:
A G L O B A L PERSPECTIVE
Limburg Institute of Financial Economics (LIFE)
University of Limburg
Maastricht, T he Netherlands

September 9, 1994

PO BOX 6 3 8 7
C l e v e l a n d
O H

4 4

1 0

1

Specialization in Risk Management*
Supervision of Derivative Instruments
The financial press in the United States is having a great time this year telling
readers about the risks associated with financial derivatives. The casual reader might think
that some n ew risks have been invented, or, at a minimum, that our financial system is
riskier n o w than a few years ago. Neither conjecture is true. N e w risks have not been
discovered, and the financial market is not a riskier place.
Nor is it true that an overriding policy priority today is the need for new legislation
or new regulation to deal with derivatives. I realize that some people view the m o d e m
financial system as a house of cards that remains standing only when buttressed by wise
regulation and other government interventions. Naturally, these people view financial
innovations like derivatives as potentially destabilizing challenges to policymakers and
regulators.
M y own view is that financial innovation tends to be inherently stabilizing, not
destabilizing. For those many of us w h o view m o d e m financial systems - and, for that
matter, market economies based on private property and price systems - as inherently
resilient, financial innovations are welcomed as reinforcements of the natural discipline and
stabilizing forces at work in a market economy.1
Using derivatives to deconstruct risk into new categories does not create more
risk. The revised risk catalogue, including systemic risk, credit risk, counterparty risk,
settlement risk Herstatt risk, market risk, legal risk, and operating or management risk,
seems thicker, I know. For market experts, such terms as delta risk, g a m m a risk,
convexity risk, and volatility risk, have become common. The naive conclusion seems to
be that increasing the number of categories means increasing the amount of risk, and that
*

The writings of m y Federal Reserve colleagues Alan Greenspan, John LaWare, William
McDonough, Susan Phillips, and Peter A. Abken have been particularly useful in preparing
this paper. Members of the staff of the Federal Reserve Bank of Cleveland, in particular
Ed Stevens, have contributed significantly to the paper.
1 I have set forth elsewhere (Jordan, 1994) the view that organizations and institutional
arrangements that strengthen property rights and the role of prices in allocating productive
resources are enduring and wealth enhancing.

1

legislators and regulators should be mounting a counteroffensive.
The Congressional hearings on derivative instruments this past spring reminded m e
of m y paternal grandmother, raising a family in rural Texas about 70 years ago.
Apparently, m y father and m y uncle were what you might call “
adept at finding innovative
ways of entertaining themselves.”Recognizing this, m y grandmother frequently would
say to one of their older sisters, “ o find out what the boys are doing, and tell them to
G
stop it!”
I detect a similar tendency among well-intentioned legislators, w h o want to say
“
Stop it!” soon as they see that some participant has lost a bundle of money using
as
financial derivatives. Such a reaction stems from the idea that market participants are
propelling the financial system and the deposit insurance system into more risk. What
needs to be understood is that w e are seeing innovations in risk management, not
innovations in risk itself. The underlying positions of participants always involve
substantial risk, and there is substantial risk in any business endeavor. Innovations in risk
management should be welcomed, as when wheat farmers first learned to lock in the price
they would get for the current crop through the use of futures contracts.
In m y remarks today, I want to make the following points:
(1) Risk exists because there is uncertainty in the world. Successful innovations in risk
management, such as derivative instruments, do not make financial markets riskier.
(2) Increased specialization in the management of risks improves the functioning of
markets; resource allocations will be wealth enhancing as comparative advantages evolve
in identifying and managing risks.
(3) Supervision of financial activity can strengthen the ultimate discipline coming from the
marketplace; regulation of financial activity in a global environment can have unintended
consequences by forcing activity out of natural channels and by socializing risk.

Specialization and Risk
People often say they want to “
reduce risk,” minimize risk,” eliminate risk,”
“
“
or
“
avoid risk.” Such language suggests that risk is something undesirable, as i is for most
t
people. However, individuals shed risk largely by passing it along to someone else. For

2

the system as a whole, risk is unchanged-it is simply borne by someone else. This
‘
someone’ a y be a specialist w h o is better equipped to manage it or, in the case of public
m
policy, a citizenry that m a y or m a y not be aware that a risk has been socialized.
I find it helpful to think of financial innovations like derivatives in the context of
Frank H. Knight’distinction between “
s
uncertainty”
and “
risk,”
developed in his classic
1921 book, Risk. Uncertainty, and Profit2 At the heart of his analysis was a distinction
between uncertain situations, in which the probabilities of possible outcomes simply are
unknown, and risk situations, in which the probability distribution of potential outcomes of
an event is known.
Following Knight’usage, entrepreneurs are specialists w h o use their expertise to
s
transform genuine uncertainty about future events into risk —that is, from less-wellspecified into better-specified distributions of potential outcomes. Specialization allows
entrepreneurs to calculate expected values as a basis for cost estimates, supply decisions,
and market clearing prices. For example, farmers and bakers use the wheat futures market
as the production technology for transforming uncertainty about the market price of wheat
at harvest time into a known price and risky return when wheat is planted and when bread
marketing and distribution arrangements are made. Bankers and insurance companies, as
well as bond and stock mutual fund managers, use specialized knowledge and access to
information, supplemented by the law of large numbers, as their production technology in
transforming uncertainty about the outcome of unique business ventures into a risky return
on a portfolio of assets.
Physicists say that matter can be neither created nor destroyed, but its form can be
converted from solid to liquid to gas. Something similar can be said of uncertainty -- i
t
exists in nature and i can take a variety of forms. General uncertainty can be segmented
t
into identifiable risks that, in tum, can be transformed into alternative forms. For example,
interest rate risk can be converted into credit risk. Risk can be transferred from one party
to another. Specific risks can be decomposed into component parts, allowing types of risk
to be segmented or combined. And, risk can be “
managed.”

2 Knight (1971 [1921]).

3

Financial risk management has been evolving for centuries. For most of this time,
the pace of evolution was gradual. With the coming of the industrial revolution, however,
economic development was accompanied by financial development, reflecting more rapid
accumulation of innovations in financial contracts, institutions, and markets. O ne index of
financial development is the intermediation ratio, measuring the layering of wealth owners’
claims on intermediaries, atop intermediaries’
claims on borrowers, atop borrowers’
claims
to real capital. The seeming redundancy of claims was productive because banks,
insurance companies, and other types of intermediaries each developed innovations,
through unique specializations, that allowed them to make a profit while assuming risk and
offering wealth owners a more assured return.
More recently, communications technology and computing power have changed
the financial development process. Increasingly, derivative financial instruments allow risk
to be transferred and better managed without adding new layers of intermediary claims
atop the underlying real capital stock. Redistributing risk from less to more efficient
specialists means, in general, moving toward a more efficient allocation of risk-bearing
resources. Wealth is enhanced because, despite the trepidation of savers about interest
rate risk, for example, more houses are built as more mortgages are packaged into
collateralized mortgage obligations and sold in the global capital market. As a result, less
of something else m a y be built, but the world is a better place because markets have used
the new financial technology to offer savers a higher return per unit of risk, and consumers
a lower cost of consumption per unit of risk.
After all, if wealth were not being enhanced, w h y would anyone use the new
technology? T w o parties will enter into a contract only because each thereby realizes an
increase in the present value of wealth. One party increases its present value by assuming
a risk for which its specialization makes i fitted. The other party increases its present
t
value by shedding that risk in return for a fee, or in return for a risk that is more congenial
to its o w n specialization. Events are basically unchanged (ignoring, for the moment, any
negative externalities and any feedback of efficiency on growth), but resources are
allocated more efficiently. Old techniques for transforming uncertainty into manageable

4

risks are pushed aside as new specialists penetrate more broadly and deeply into the
market. The cost of risk-bearing falls for society, thereby enhancing wealth.

Derivative Financial Instrum ents
“
Credit risk”
management was the focus of attention in commercial banking in the
1980s. That followed a period in which “
interest rate risk”
and “
country risk”
were the
hot topics. Now, financial engineers are altering the financial intermediary process, using
finance theory and computer technology to divide risk into components that heretofore
were inseparable from underlying assets. Beyond that, globalization of financial markets
prevents anyone from monopolizing the benefits of derivatives, just as it cuts through
monopoly and regulatory specializations based on old technology.
Financial derivatives were being used long before the 1970s, when organized
exchange trading began. Thereafter, most derivative financial contracts were traded on
the organized commodities exchanges, so that holders of contracts were protected by the
exchanges themselves, as well as by Securities and Exchange Commission (SEC) rules.
The exchanges, as the counterparty to each contract traded, had a clear self-interest in
promoting integrity of trading and delivery, and that remains the approach to supervision
of exchange-traded derivative contracts today. The nature of contracts, however, has
expanded to include a wide variety of options indexed to financial-market measures like
stock market price indicators and even the monthly average overnight federal funds rate.
The most rapidly growing instruments in the derivatives market, and the locus of
recent innovation, are over-the-counter (OTC) contracts. These are outside the purview
of any exchange rules. They include ordinary currency and (largely) interest rate swaps,
plus a small portion in more exotic hybrid contracts. Typically, one or more of the
counterparties to an O T C contract is a commercial bank, with the contract tailored to the
idiosyncratic needs of the counterparties, much like a commercial loan. Most banks say
they enter into contracts simply as end users, to meet their ow n risk management needs.
A dozen or so of the largest money center banks, however, are counterparties to a very
large percentage of all O T C contracts because they act as dealers, tailoring contracts to

5

the needs of any customer. These banks must manage the net risk that results from their
dealer position, earning their income from a bid-ask spread.
A n y user of derivative contracts, whether a dealer or end-user, should be able to
demonstrate to management, directors, auditors, and shareholders, qualitatively, if not
quantitatively, the rationale for positions taken. This does not mean that all derivative
positions can be designated as either a hedge or a speculation, because derivative contracts
are not ordinarily isolated in a separate profit center.
Even a derivatives dealer will not necessarily try to run a riskless book of offsetting
derivatives exposures, but instead m a y want to use its dealer position to offset a risk
exposure elsewhere on its books. For end users as well as dealers, derivatives must be
part of a larger risk-management strategy. Current accounting practices, of course, do not
produce an integrated record of risk management This is w h y some supposed losses from
derivatives are not losses at all, but simply the offset to gains elsewhere in the business.
Offsetting the gains and losses leaves the neutral position the firm was trying to ensure by
off-loading certain types of risk. References to a firm’derivatives business as an actual
s
or intended profit center, on the other hand, suggests an intention to do more than
contribute an ingredient to risk management, perhaps by arbitraging some market niche, or
simply by engaging in informed speculation.

Supervision and Regulation
Derivatives are innovations that, like atomic energy and genetic engineering, can
be used for good or i l and be intended for good but have il effects through
l,
l
mismanagement. A n objective of government supervision is to ensure that innovation
takes place in an open environment, so that those affected can see h ow innovations are
being used and to what effect. Supervision in this sense stands in contrast to regulation,
by which government seeks to ensure good results by directing or delimiting the actions of
citizens1. The beauty of a market economy is that innovations like derivatives can be

3 For discussion and recommendations for strengthening the disciplining role of market
forces through appropriate supervision and regulation, see Jordan (1993).

6

expected to have good results, as the invisible hand of the market mediates among the self
interests of potential users, preventing unintended gains and losses over time by driving
the inept from the marketplace. This simple Smithian economic theory, however, seems at
variance with the publicity being given to some spectacularly large losses attributed to
derivatives activity in recent years. Current proposals to regulate derivatives are, at least
in part, a reaction to those losses.
One reason for losses undoubtedly is inexperience with the n ew engineering of
risk. Both experience and theory make good teachers and the experience with derivatives
has been teaching some valuable (or at least expensive) lessons. S o m e of these lessons
have been more like refresher courses, such as the notions that interest rates are not a one­
way bet, that leveraging a position leverages risk, and that undersupervised pockets of a
large organization are bound to invite agency problems. Other lessons seem
dumbfoundingly simple only with hindsight, like the fact that valuations of a derivatives
position can be extremely sensitive to overlooked or untested assumptions about things
like liquidity and cross correlations among asset returns.
Three additional important lessons are largely overlooked in the media. One is that
derivatives redistribute risk from one counterparty to the other. This means that the
counterparties to spectacular losers have been some substantial gainers w h o probably will
be unable to continue laying off risk on as favorable terms in the future as in the past. A
second lesson is that the recent spate of spectacular losses has been associated, for the
most part, with so-called “
exotic”
derivatives that comprise only a very small fraction of
the market. Of course, “
plain vanilla” T C currency and interest rate swaps and futures
O
can have very long maturities, so experience to date is not necessarily the whole story
there. A third lesson is about the usefulness of capital in paying for an education. All but
a slight fraction of losses on derivatives contracts to date have been absorbed from the
capital of the exposed party, its parent, or its sponsor, and not from any haircut on the
value of the contract to the party “ the money.”
in
Market discipline can be a powerful educator, as long as the experience of both the
gains and the losses from using derivatives remains with the contracting parties.
Moreover, the potential for gain and loss provides the incentive for concerted action to

7

change the infrastructure of markets. Law, accounting, and standard market practices
devised in the past need updating to cover the new derivative instruments. The 1993
recommendations of the Group of Thirty are one recent example of the self-interest of the
participants in the OTC derivatives market seeking the basis for more reliable evaluations
of risks.
Government can help in strengthening institutional arrangements that promote
market discipline. For example, markets operate best with reliable information, but most
observers agree that reliable information is not yet consistently available about derivatives
and derivatives positions. Supervisory authorities such as central banks are in a position
to help market participants develop common forms of disclosure, even by such simple
procedures, in the United States, as releasing CAMEL and BOPEC ratings.4
“Better information” actually covers a wide range of possibilities where
supervisors and participants have common interests. One opportunity for improvement is
communication internal to a firm. Because banks are the dominant institutions in the OTC
derivatives market, the existing bank examination process can serve as a useful check on
some seemingly obvious, but sometimes overlooked, matters where advice can substitute
for painful experience. Supervisory authorities can ensure that fundamental questions are
being addressed. Is a bank’s strategy to be only an end user, or also a dealer in
derivatives? Do staff, top management, and directors all have the same understanding of
the bank’s derivatives strategy? How consistently is that strategy being communicated to
shareholders and the public?
Another aspect of information where the interests of supervisory authorities and
market participants converge is in maintaining objective measures of risk for reliable
interbank comparisons. Inclusive evaluations of management capability, based on
something like the rating systems we use in the United States, are suitable for this
function. Basle risk-based capital standards represent significant progress toward
establishing global interbank comparability.

4 CAMEL and BOPEC are acronyms for the factors underlying regulatory risk ratings
used for banks (capital, asset quality, management, earnings, and liquidity) and bank
holding companies (bank, other subsidiaries, parent company, earnings, and capital).
8

A n obvious next step, already taken by the Basle Committee, will be to incorporate
off-balance-sheet risks into firm-specific capital adequacy measures. The federal banking
supervisors in the United States have such a proposal out for comment at the moment. In
the case of exchange-traded contracts, of course, risk is not so contentious an issue. Daily
marking-to-market and margin requirements protect the exchange, while the strength of
the exchange protects the in-the-money counterparty. O T C derivatives, in contrast,
cannot be marked to market directly, so another method of monitoring exposures must be
developed.
As the bank examination process evolves, I expect that the risk management of
derivative instruments used or offered by a bank will employ firm-specific stress testing of
the bank’capital adequacy. This will require a valuation model for O T C derivatives that
s
includes consideration of duration, counterparty concentrations, and liquidity.
Development of the appropriate simulation model should be left to the firm, but it must be
a well documented, comprehensive representation of its exposures, and must also be
flexible about the range of stress assumptions under which the model can be simulated.
The advantage of a mutual market and supervisory interest in folding off-balance-sheet
risk into capital adequacy is that supervisors will have the benefit of comparing many
different methods, and of requiring answers to challenging questions.
Not all proposals for government action in the derivatives market are as benign as
seeking better information. As a general rule, i is wiser to let market forces mete out
t
losses as well as profits, than to force everyone to follow suboptimal rules and socialize
losses. Perhaps the initial educational role of government supervision and examination
will dwindle in importance over time. Certainly there can be no permanent detailed
direction of derivative practices from supervisory personnel whose technical expertise,
while substantial, cannot be expected to match that of market players. Nonetheless, there
is a crucial role for oversight by the chartering authority or, more significantly, the
provider of deposit insurance, in protecting the public trust and the public purse.5 In
Ronald Reagan’phrase about arms control, “
s
Trust, but verify.”

' For a more detailed discussion of the roles of public authorities in supervision and
regulation, see Jordan (1993).

9

Systemic Risk
Financial innovations in general, and derivative financial instruments in particular,
m a y represent nothing more than n ew specializations being used to manage risk more
efficiently. Like A d a m Smith’pin maker, new risk specializations should increase
s
economic efficiency and human well-being, as its benefits are realized through market
trading. This assumes, however, that financial innovations have no negative externalities
detracting from the benefits of the rapid spread of new financial technologies -- that is,
that the marginal private and social costs of risk bearing are identical.
There appears to be widespread apprehension that the social costs of derivatives
exceed their private costs. Unrecognized in private cost, apparently, is the systemic aspect
of potential market collapse, reflecting n ew interdependencies generated by derivatives,
among counterparties, risks, and markets.
Of course, for as long as counterparties have had counterparties, credit risk has
had an element of interdependence —one party’repayment of debt to another was a
s
function of someone else’ability to repay a debt to the first party. These
s
interdependencies have been modeled according to three unique sources of potential
difficulty. T w o of these - manias and fragility - are viewed in much the same way today
as they were by Walter Bagehot in 1870. in Lombard Street. The third, systemic risk, is a
newer and still fuzzy concept of uncertain significance.
M an ias and bubbles, including th eir consequent panics

are one model of

interdependence among market participants. This model recognizes that economic agents
have a propensity for delusion about the return on particular investments, as a mass of
investors mutually support each other’belief -- first, in the impossible and then, in the
s
inevitable. Their c o m m o n delusion is a misreading either of the likely real return to
capital, or of the probability of cashing in a position before anyone else does. Walter
Bagehot described the phenomenon as when owners of savings “
find that ...specious
investments can be disposed of at a high profit, they rush into them more and more....So
long as such sales can be effected the mania continues; when i ceases to be possible to
t
effect them, ruin begins.” In m o d e m jargon, this is known as “ bigger fool theory.” A
6
the
6 Bagehot (1921), pp. 131-132.

10

recent example was when the Nikkei average went to 40,000. Another is the popularity of
the M M M enterprise in Russia, despite government warnings that it is nothing more than a
Ponzi scheme.
The explosive growth of O T C derivatives contracts conceivably could be classified
as a temporary mania, particularly from the point of view of those whose mismanagement
has produced spectacular losses. With hindsight, marginal private cost was apparently
seriously underestimated. Continued rapid growth of derivatives contracts at the pace of
the past several years certainly would begin to raise the mania flag. Even in a global
financial marketplace there must exist a finite limit to shiftable risk. For now, however,
that point does not seem to be in sight.
The second model based on interdependence isfra g ility , in the technical sense
used by writers like H y m a n Minsky. The fragility model produces debt/equity ratios that
are higher than is socially efficient, and that rise as an economic expansion proceeds.
Again, harking back to Bagehot, “
And in so far as the apparent prosperity is caused by an
unusual plentifulness of loanable capital and a consequent rise in prices, that prosperity is
not only liable to reaction, but certain to be exposed to reaction. The same causes which
generate this prosperity will, after they have been acting a little longer, generate an
equivalent reaction.” That i over the course of an economic expansion, financial
7
s,
markets become increasingly susceptible to instability in response to any random shock.
Thus, fragility is an endogenous feature of modern market economies. The risk levels of
all financial contracts are interdependent in that they jointly depend on the state of the
aggregate economy.
Systemic risk ,as

that term has come to be used, is like the mania and fragility

models in that interdependence creates the possibility of falling dominoes; all three models
exhibit that c o m m o n systemic characteristic. However, m o d e m discussions of systemic
risk do not emphasize the mass delusion of a mania, or the endogenous c o m m o n
association with the aggregate economy found in fragility. The systemic risk model seems
to postulate the existence of some new, third externality that makes private calculations of

7 Bagehot (1921), p. 146.

11

risk understate the true susceptibility of financial contracts to loss. In other words,
absence of incentives for participants to internalize all costs associated with certain
instruments creates a problem at the aggregate level that is not apparent at the micro level.
The presence of the externality invites government intervention to restore the equality of
private and social marginal cost.
A major difficulty with the systemic risk concept comes in trying to identify the
nature of this new form of externality that pushes private cost below social cost. S o m e
have argued that borrowing is like an internal combustion engine, polluting the financial
market atmosphere. W h e n I lend to you by reducing m y liquidity or otherwise accept
greater risk, I increase the probability that I will be unable to meet m y obligations to
others. This would represent a negative externality from the point of view of m y
creditors, if they were unaware of m y lending to you. The argument breaks down,
however, to the extent that, acting out of self interest, m y creditors are able to internalize
the supposed externality. Loan covenants, for example, protect a creditor from a debtor
entering into unforeseen debt or credit relationships. More generally, the expectation of
internalizing this potential externality is recognized in the eternal watchphrase, “ n o w thy
K
counterparty!”
In the case of derivatives, a variant of the pollution argument has emphasized the
concentrated dealer market. Each major dealer is the source of an interdependence among
the exposures of its worldwide circle of end users. Evaluations of counterparty risk
exposures to these end users should include a dealer risk, analogous to country risk, that
would be too trivial to notice in a less concentrated market. Similar allowances might be
made for interdependence arising from the use of c o m m o n operations centers, payment
networks, legal advisors, or credit rating services. Evaluating risk is not a simple matter;
i involves compound probabilities and cross correlations.
t
In general, the interdependence envisioned in the systemic risk model seems to
involve the sensitivities of many large counterparties to one another. Derivatives and
globalization of markets m a y indeed be producing more complex compound probabilities
of trouble. However, so too is the information age vastly expanding the ability to monitor
counterparties and markets. If sophisticated financial engineering can produce complex

12

derivative products, cannot that same sophistication estimate the resulting increasingly
complex compound probabilities of trouble?
The point is that there seems to be no reason to believe that the potential
externality of increasingly complex financial relationships has outdistanced an increasingly
powerful ability to internalize that potential externality. The commercial overhead of
m o d e m financial centers —including satellite-fed, on-line, worldwide information and
monitoring systems, armies of legal talent, and even on-site monitors from the rating
agencies —all reflect the substantial expenditures of firms trying to internalize their risk
exposures.
As long as economic agents are able to estimate compound probabilities of
failures, systemic risk is indistinguishable from normal credit risk. Knowing your
counterparty and your counterparty’counterparties, and even your counterparty’
s
s
counterparties’
counterparties, should lead to quality spreads in market prices, to prudent
loan loss reserves and capital from which to absorb losses, and to equality of the private
and social cost of risk.
Sup p ose that no w ed g e betw een private and social co st is inserted by the inability
to evaluate and control interdependency. T hen, either there is no w ed g e - in w h ich case
there is no p o licy basis for concern about system ic risk - or the w ed g e originates from
som e other source.
S uppose that m arkets are p erfectly able to, and d o, price the risk o f financial
d o m in o es falling in a system ic collap se, but that p olitician s, acting for society, are
unw illing to tolerate the con seq u en t lo sses to individual constituents. P olitician s therefore
im agine a w ed ge betw een private and social cost, b elievin g the latter to be ab ove its “true”
level. T he failure o f m arkets to internalize this im agined extra social cost w ould lead to a
p olitical perception that sp ecialized risk-m anagem ent products were being overproduced.
T he problem w ould not be derivatives; the problem w ould be that p olitician s have a low er
tolerance for risk than do market participants. The leg isla tiv e and regulatory corollary
w ou ld be a ch allen ge to reduce losses w ithout creating a moral hazard by su bsid izing risktaking.

13

Moral Hazard
Perhaps the most troubling aspect of the widespread apprehension about systemic
risk consequences of derivatives is the seemingly equally widespread conviction that
government - meaning central banks and deposit insurance providers - can be counted on
to prevent a systemic collapse. Even if government had no such intention, the conviction
that it would come to the rescue produces a growing externality in the form of
unmonitored, compound probabilities of trouble. Systemic risk becomes real, though it
need not be, but w h o bears the risk? Likely, it is taxpayers, through the central bank and
deposit insurance system, whose exposures would grow while private exposures would
decline. More important, what is the direction of causation? Does growing systemic risk
invoke central bank risk-bearing because i is socially more efficient, or does the apparent
t
willingness of central banks to bear risk allow markets to adopt financial specializations
whose systemic risk externalities are downloaded onto central banks?
Here is where central banks must tread very carefully. There is a moral hazard in
reassuring markets, or in allowing markets to believe incorrectly, that a lender of last
resort will act, and at low cost, to prevent a contagious spread of broken promises. If
systemic risk is becoming as worrisome as we are led to believe by some commentators, i
t
seems likely that the reason is not innovations in financial technology, but the moral
hazard of central banks’
implicit willingness to underwrite that risk.
Moral hazard can be a real danger in central banking, as was demonstrated during
the last great spurt of financial innovation during the cash management revolution of the
1970s. Prophets had foretold the coming of cashless transactions. What happened in the
United States, however, was that both cash transactions and cash balances were
eliminated. Increasingly over the 1970s, reserve balances were created by central bank
daylight credit, on demand, to accommodate transactions. Not until the early 1980s did
the Federal Reserve discover the extent to which its free daylight credit, rather than
someone’pre-existing, positive reserve balance, was the medium of wire payment; that a
s
large chunk of private payment system risk had been transformed into Federal Reserve
credit risk.

14

Efforts to reduce and manage Federal Reserve payment system risk in the past
decade largely have involved digging out from under a mountain of daylight credit
initiated during the cash management revolution. That story began with the central bank
inadvertently absorbing the externality of payments system risk, rather than creating
institutional mechanisms by which private parties would be led to internalize that risk.
N o w a risk management revolution is in full swing. The specter of growing
systemic risk is used to rationalize nonbank access to the discount window, to seek direct
nonbank access to Fedwire payment finality, and to call for regulatory guidelines for
derivatives contracts that could become the leverage for obtaining central bank assistance
when they prove to be flawed. The regulatory challenge is to avoid these snares.
Incentives must be created for participants to internalize risk. This is essential if w e are to
avoid socializing losses. As central bankers, our role is to supervise markets by spreading
information that promotes knowledgeable risk-management structures, while avoiding
wholesale reassurances that timely central bank money creation will ameliorate trouble.

Concluding Remarks
R apidly spreading use o f d erivatives su ggests that they are exp ected to add valu e
to those on both sid es o f contracts. D erivatives do not add to or subtract from the risks
that are inherent in a m od em financial system . T hey do, how ever, allow existin g
uncertainty to be borne m ore efficien tly. Financial innovations are to be w elcom ed as
basically wealth enhancing. A s my colleagu e at the Federal R eserve, John LaW are, has
said, “ D erivatives ... have been used primarily to contain risk....A ... useful definition o f
banking is that the banker essen tially m anages financial risks for his depositors. H is job is
to m anage risk, not avoid it”.*
T he inform ation age is changing the w ay risk is m anaged. “T his is a scien tific
revolution^ said former U .S . Secretary o f State G eorge Schultz in exp lain in g the
im p lications o f the inform ation age to M ikhail G orbachev (too late, as it turned out). H e
w ent on, “There w as a tim e when a governm ent could control its scien tific establishm ent

* LaW are (1 9 9 4 ), pp. 5-6.

15

and be basically successful. N o longer. T o keep up today and in the future means that
scientists will have to be in constant touch with the ‘
thinking community’
around the
world. A nd this is an information revolution. The inability of one nation to be
predominant in the international financial world is going to be repeated in field after field.
The key is going to be knowledge-based productivity... . 9
”
To the extent that derivatives are not well understood, surprises should be
expected and should be no cause for concern as long as information about derivatives use
is not hidden, and is matched by attention to the adequacy of liquidity and capital. A more
significant danger is that w e smother market incentives for counterparty scrutiny with
overly-generous central bank assistance.
S o m e of the advocacy of new legislation or regulation has been based on the view
that the entire financial system could be jeopardized by the losses sustained by a single
large participant in derivatives markets. M y view is that such vulnerability has not been
established. In fact, I believe that certain proposals might actually increase systemic risk
because they would penalize standard risk-hedging methods and change behavior to get
around the regulations. Those w h o argue that financial innovation calls for new
regulations should remember that new regulations are very often the stimulus for new
innovation.
The ultimate regulator of any economic activity is the market. In the case of
finance, the global marketplace is a powerful source of discipline. As w e consider
proposals for action by governmental authorities, I suggest that w e establish a litmus test.
Namely, in the words of Federal Reserve Chairman Greenspan, “ relevant question ..
The
.
is whether private market regulation is enhanced or weakened by the addition of
government regulation.”0
1

9 Shultz (1993), p. 893.
10 Greenspan (1994), p. 26.

BIBLIOGRAPHY
Abken, Peter A. “
Over-the-Counter Financial Derivatives: Risky Business?”
Federal
Reserve Bank of Atlanta, Econom ic Review, vol. 79, no. 2 (March/April 1994)
pp. 1-22.
Bagehot, Walter. L o m b a rd Street: A D escription o f the M o n ey M a rk e t, 14th ed.
London: John Murray, 1931.
Greenspan, Alan. Testimony before the Subcommittee on Telecommunications and
Finance of the Committee on Energy and Commerce, U.S. House of
Representatives, M a y 25, 1994.
Group of Thirty, Global Derivatives Study Group. “
Derivatives: Practices and
Principles,”
July 1993.
Jordan, Jerry L. “ Market Approach to Banking Regulation,”
A
forthcoming in The Cato
J o u rn a l ,vol. 13, no. 2 (Fall 1993), pp. 363-80.
Jordan, Jerry L. “
Economic Forces versus Monetary Institutions,”
forthcoming in James
A. Dorn, Roberto Salinas, and Michael Walker, eds., M o n etary Arrangem ents in
the Am ericas a fter N A F T A , Washington, D.C.: The Fraser Institute, 1994.
Knight, Frank H. Risk, U ncertainty, and P ro fit. Chicago: University of Chicago Press
1971 (1921).
LaWare, John P. Remarks presented at the American Bankers Association’ 1994
s
National Regulatory Compliance Conference, Washington, D.C., June 13, 1994.
McDonough, William J. Remarks presented at the Institute of International Bankers’
Annual Meeting, N e w York, M a y 4, 1994.
McDonough, William J. “
Financial Market Innovation - Practical Concerns,”
Remarks
presented before the Money Marketeers of N e w York University, N e w York City,
June 16, 1994.
Phillips, Susan M. “
Derivatives and Risk Management: Challenges and Opportunities,”
Remarks presented at the Federal Reserve Bank of Atlanta Conference on
Financial Markets, Coconut Grove, Florida, February 25, 1994.
Phillips, Susan M. Remarks before the Forum on Developing Country Access to
Derivatives Markets, The World Bank, Washington, D.C., June 30; 1994.

17

Shultz, George P. T u rm o il and T riu m p h : M y Years as Secretary o f State. N e w York:
Charles Scribner’Sons, 1993.
s
U.S. General Accounting Office. “
Financial Derivatives: Actions Needed to Protect the
Financial System,”
GAO/GGD-94-133, M a y 1994.

18