Che materia stai cercando?

Anteprima

ESTRATTO DOCUMENTO

4 R C. M Z B

OBERT ERTON AND VI ODIE of the neoclassical model, the prescriptions and pre-

world, including central banks, can function with- dictions of these alternatives are manifestly sensitive

out the computer-based mathematical models of to the specific market frictions and posited behav-

modern financial science. Furthermore, the specific 7

ioral deviations of agents.

models that these institutions depend on to conduct Perhaps more latent is

their global derivative pricing and risk-management the strong sensitivity of these predictions to the

activities are based typically on the Black–Scholes institutional structure in which they are embedded.

option pricing methodology. There is a considerable ongoing debate, sometimes

So much for the past: What about the impending expressed in polar form, between the proponents

future? of these competing paradigms. Those who attack

the traditional neoclassical approach assert that the

overwhelming accumulation of evidence of anoma-

With its agnosticism regarding institutional struc- 8

lies flatly rejects it.

ture, neoclassical finance theory is an ideal driver They see a major paradigm

to link science and global practice because its shift to one of the new alternatives as essential for

prescriptions are robust across time and geopo- progress. Defenders of the neoclassical paradigm

litical borders. Future development of derivative- respond that the alleged empirical anomalies are

security technologies and markets within smaller either not there, or that they can be explained

and emerging-market countries could help form within the neoclassical framework, and that in

important gateways of access to world capital mar- either case, the proposed alternatives do not offer

9

kets and global risk sharing. Financial engineering a better resolution. That debate so framed is best

is likely to contribute significantly in the devel- left to proceed anomaly by anomaly and we say no

oped countries as well; as for instance in the major more about it here.

transitions required for restructuring financial insti-

4

tutions both in Europe and in Japan. Instead, we take a different approach. Rather than

choose among the three competing theoretical per-

spectives, we believe that each, although not yet of

But will the same intense interaction between the the same historical significance, can make distinc-

science and practice of finance continue with respect tive contributions to our understanding and each

to the new directions of scientific inquiry? has its distinctive limitations.

In neoclassical theory, institutions “do not mat-

3 The challenge to neoclassical finance ter” in the sense that equilibrium prices and the

allocation of resources are unaffected by specific

institutional structures. As long as markets are

With its foundation based on frictionless and effi- efficient and frictionless, one can use almost any

cient markets populated with atomistic and rational convenient financial system in a model for ana-

agents, the practical applicability of the neoclassical lyzing asset demands and the derived equilibrium

modeling approach is now challenged by at least two asset prices and risk allocations will be the same as

alternative theoretical paradigms. One, New Insti- in models with more realistic and more complex

tutional Economics, focuses explicitly on transaction financial systems.

costs, taxes, computational limitations, and other

5

frictions. The other, Behavioral Economics, intro- In criticizing neoclassical theory, proponents of

duces non-rational and systematically uninformed

6 both neo-institutional and behavioral finance often

behavior by agents. In contrast to the robustness

J O I M F Q 2005

OURNAL F NVESTMENT ANAGEMENT IRST UARTER

T D F S 5

HE ESIGN OF INANCIAL YSTEMS

neoclassical theory of finance will be approximately

posit the same simple financial institutional struc- correct for asset prices and resource allocations, after

ture in their models, and then proceed to show the endogenous changes in institutional structure

how the introduction of market frictions and devia- 11

have taken place.

tions from rationality can cause significant changes Furthermore, FSF can be used

in equilibrium allocations and asset price behavior. to predict likely changes in institutional structure

But this is not a valid argument. Unlike the friction- and to identify targeted changes in that structure

less and rational neoclassical case, there is no longer that might lead to more efficient allocations.

the invariance of optimal asset demands to insti-

tutional specifications. Hence, proper assessments, Many of the issues facing decision makers around

theoretical and empirical, of market allocational the world today are about institutional change. In

and informational efficiency and interpretations of China, for example, decentralization and privati-

apparent distortions on capital asset pricing from zation of large parts of the economy during the

behavioral and transactional dysfunctions cannot past decade have produced remarkable improve-

be undertaken without explicit reference to a real- ments in standards of living. Public officials and

istic modeling of the institutional environment. business leaders now see an urgent need to cre-

Thus, as major changes take place in the insti- ate a financial infrastructure to support contin-

tutional structure for trading financial assets and ued economic development. In Japan, officials

allocating risks, one would expect that the impact of are considering fundamental changes in the struc-

such frictions on asset prices would change. Indeed, ture of their banking system to overcome eco-

from the FSF perspective, the particular institu- nomic stagnation. And in Europe and the United

tions and organizational forms that arise within States, pension and Social Security reform has

the financial system are an endogenous response become a top priority. A critical issue every-

to minimize the costs of transaction frictions and where is controlling the risk of default by financial

behavioral distortions in executing the financial institutions.

10

functions common to every economy. As a con- Neoclassical theory generally serves as a good

sequence, in well-functioning financial systems, starting point in addressing such policy issues.

high transaction costs and dysfunctional cogni- It can identify properties of an efficient equilib-

tive dissonance among individuals may not have rium resulting from the assumptions of rational

a material influence on equilibrium asset prices optimizing behavior and perfect competition. In

and risk allocations. Therefore, from this per- the posited frictionless environment of neoclassical

spective, market-friction and behavioral predictions models, however, multiple alternative institutional

may not provide reliable insights about observed structures are possible to support the same equilib-

asset prices and resource allocations, but they will 12

rium asset prices and risk allocations.

be centrally important—along with technologi-

cal progress—in explaining the actual institutional

structure of the financial system and the dynamics For example, the celebrated Coase theorem shows

of its change. that in the absence of transaction costs, a variety

of organizational structures can result in optimal

13

resource allocation. In such an environment there

4 The functional synthesis would be no reason for firms to exist, since the

simpler neoclassical structure of atomistic agents

The central conclusion of FSF is that in well- interacting directly in competitive markets would

developed financial systems, predictions of the work just as well. As Coase shows, however, when

F Q 2005 J O I M

IRST UARTER OURNAL F NVESTMENT ANAGEMENT

6 R C. M Z B

OBERT ERTON AND VI ODIE The two fundamental tenets of FSF are:

transaction costs are brought into the analysis, then •

organizational structure matters. Some economic Neoclassical theory is approximately valid for

activities are best undertaken in large hierarchi- determining asset prices and resource allocations

cal firms, while other activities are best organized (albeit as a reduced-form model), but offers lit-

through atomistic markets. tle to explain which organizational structures

for production and performing various financial

Another well-known example of neoclassical functions and which particular market instru-

assumptions leading to indeterminacy in structural ments and financial intermediaries will evolve.

form is the celebrated M&M Propositions regard- Neo-institutional and behavioral theories are

14

ing the capital structure of firms. Modigliani and centrally important in analyzing the evolution

Miller prove that in the absence of transaction costs, of institutions including market instruments and

agency costs, and taxes, firms would be indiffer- financial intermediaries, but are unlikely to pro-

ent with respect to their financing mix between vide significant and stable explanations of asset

16

debt and equity. When these frictions are taken prices and resource allocations.

into account, however, a firm’s capital structure can

15

matter a great deal. 4.1 Example 1. Transaction costs and option pricing

In both examples—the Coase Theorem and the A quarter century ago, Hakansson (1979) wrote

M&M Propositions—the neoclassical model serves about the “Catch 22” of the option pricing model.

as a starting point for analysis of institutional His point was that if the conditions for Black–

structure. However, the neoclassical model alone Scholes pricing are satisfied, then the option is a

cannot in general identify the most efficient struc- redundant security with no social purpose; and if the

ture. The new institutional and behavioral theories conditions are not satisfied, then the pricing model

17

can be used to help identify features of the envi- is wrong. The seeming paradox can be resolved,

ronment that may make one structure superior however, by considering transaction costs.

to another in a particular setting at a particular

time. In reality most investors face substantial transac-

tions costs and cannot trade even approximately

Thus, the neoclassical model by itself offers some continuously. But in a modern, well-developed

limited guidance to decision makers seeking to financial system, the lowest-cost transactors may

understand and manage the process of institutional have marginal trading costs close to zero, and can

change. In FSF, neoclassical, institutional, and trade almost continuously. Thus, the lowest-cost

behavioral theories are complementary rather than producers of options can approximate reasonably

competing approaches to analyzing and managing well the dynamic trading strategy, and their cost

the evolution of financial systems. By employing all of replicating the payoffs to the option is approxi-

18

three modes of analysis, FSF can perhaps help pol- mately the Black–Scholes price.

icy analysts to choose among competing structural

solutions to real-world problems. As in any competitive-equilibrium environment,

price equals marginal cost. As is typical in analyses

Instead of attempting a highly formal development of other industries, the equilibrium prices of finan-

of FSF, which is still quite tentative, we frame its cial products and services are more closely linked to

synthesis of the different schools of thought using the costs of efficient actual producers than to inef-

a series of illustrative examples. ficient potential ones. The result in this context is

J O I M F Q 2005

OURNAL F NVESTMENT ANAGEMENT IRST UARTER

T D F S 7

HE ESIGN OF INANCIAL YSTEMS

management services at a much lower cost than

that high-trading-cost individuals can become cus- individuals can provide for themselves. Equilibrium

tomers of low-trading-cost financial intermediaries asset prices will, therefore, reflect the lower marginal

and buy options at nearly the same price as if those transaction costs of those financial-service firms and

individuals could trade continuously without cost. not the higher transaction costs of the individuals.

The underlying force driving the development of Neoclassical portfolio theory also offers some guid-

efficient institutional structures is Adam Smith’s ance in identifying the likely nature of the services

“invisible hand”—firms seeking to maximize their to be provided by financial intermediaries. The the-

profits in competitive product markets. Potential ory of portfolio selection tells us that in the absence

customers have a demand for the contingent pay- of transaction costs and with homogeneous expec-

offs associated with options, and profit-seeking tations, individuals would be indifferent between

financial firms compete to supply the options choosing individually among all assets and choos-

using the lowest-cost technology available to them. ing among a small number of optimized portfolios.

As marginal trading costs for the financial firms This is the classic “separation” theorem of port-

approach zero, equilibrium option prices approach 22

folio theory.

the Black–Scholes dynamic-replication cost. Thus, But in the presence of significant

we should find that with an efficient, well-developed information and transaction costs, the separation

financial system, over time, the neoclassical model theorem turns into an elementary theory of financial

gives the “correct” pricing as a reduced form, but intermediation through mutual funds.

options and other derivative financial instruments

and the institutions that produce them are certainly Mutual funds are the investment intermediaries

19

not redundant. that specialize in producing optimized portfolios

by gathering the information needed (expected

returns, standard deviations, and correlations

4.2 Example 2. Continuous-time portfolio theory among the full set of risky assets) and combining

them in the right proportions (the efficient port-

folio frontier). Because of economies of scale in

Our second example is closely related to the first gathering information, processing it, and trading

one, but carries it a step further. Consider the securities, the transaction costs for mutual funds

Intertemporal CAPM and the assumptions of fric- will be significantly lower than for individuals, so

tionless markets and continuous trading used in

20 individuals will tend to hold mutual funds rather

deriving it. It is well known that by introducing than trade in the individual securities themselves.

transaction costs into a model with an institutional

structure in which individuals all trade for them- This view also addresses the issue of heterogeneous

selves directly in the markets, one can get very expectations in the Capital Asset Pricing Model

different portfolio demand functions and thus very

21 by offering a justifying interpretation for its stan-

different equilibrium prices. But in the presence dard assumption of homogeneous beliefs: namely,

of substantial information and transaction costs it investors in mutual funds in effect “agree to agree”

is not realistic to posit that the only process for indi- with the return-distribution estimates of the pro-

viduals to establish their optimal portfolios is to fessionals who manage those funds. Furthermore,

trade each separate security for themselves directly since professional investors tend to use similar data

in the markets. Instead, individuals are likely to sets and methods of statistical analysis, their esti-

turn to financial organizations such as mutual and mates may be more homogeneous than would

pension funds that can provide pooled portfolio

F Q 2005 J O I M

IRST UARTER OURNAL F NVESTMENT ANAGEMENT

8 R C. M Z B

OBERT ERTON AND VI ODIE easy-to-understand, seamless solutions to complex

otherwise be the case if individuals were gathering

23 life-cycle risk management needs of households.

data and making forecasts directly for themselves. Households today are called upon to make a wide

In more realistically complete models of lifetime range of important and detailed financial deci-

portfolio selection, individuals may have complex sions that they did not have to in the past. For

optimal dynamic strategies. Here too, neoclassical example, in the United States, there is a strong

theory offers a useful starting point for a the- trend away from defined-benefit corporate pen-

ory of financial structure. As shown in Merton sion plans that require no management decisions

(1989), for every dynamic trading strategy there by the employee toward defined-contribution plans

exists an equivalent contingent contract or derivative that do. There are more than 9000 mutual funds

security. Black, Merton, and Scholes derived the and a vast array of other investment products.

option pricing model by showing that there is a Along with insurance products and liquidity assets,

dynamic trading strategy that replicates the pay- the household faces a daunting task to assemble

offs from a call option. That same approach applies

24 these various components into a coherent effective

to any derivative security. The contingent-claim- lifetime financial plan.

equivalence to dynamic portfolio strategies can be

derived by running the option-pricing derivation

25 Some see this trend continuing with existing prod-

“in reverse.” ucts such as mutual funds being transported into

technologically less-developed financial systems.

From contingent-claim-equivalence it follows that Perhaps this is so, especially in the more immedi-

a low-transaction-cost financial intermediary can ate future, with the widespread growth of relatively

sell to high-transaction-cost customers fully hedged inexpensive Internet access to financial “advice

(“immunized”) contracts that have the contin- engines.” However, the creation of all these alter-

gent payoffs associated with an optimized portfolio natives combined with the deregulation that made

strategy. The intermediary pursues the dynamic them possible has consequences: deep and wide-

trading strategy at its lower transaction costs and ranging disaggregation has left households with the

provides the specified contractual payoffs to its

26 responsibility for making important and techni-

customers. cally complex micro-financial decisions involving

risk—such as detailed asset allocation and esti-

Note that under this view of the process of financial mates of the optimal level of life-cycle saving for

intermediation, the products traditionally provided retirement—decisions that they had not had to

by investment management firms tend to merge make in the past, are not trained to make in the

with the long-term contracts traditionally produced present, and are unlikely to execute efficiently in the

by the life insurance industry. This convergence future, even with attempts at education.

transformation has been going on for many years

in the market for variable annuities in the United The availability of financial advice over the Inter-

States, although it has largely been motivated by the net at low cost may help to address some of the

tax-deferral advantages of annuities. information-asymmetry problems for households

with respect to commodity-like products for which

If this view is correct, then as transaction the quality of performance promised is easily ver-

costs continue to decline, financial intermediaries ified. However, the Internet does not solve the

will produce more complicated-to-produce prod- “principal–agent” problem with respect to more

ucts that combine features of investments and fundamental financial advice dispensed by an agent.

insurance. They will be customized to provide

J O I M F Q 2005

OURNAL F NVESTMENT ANAGEMENT IRST UARTER

T D F S 9

HE ESIGN OF INANCIAL YSTEMS

the production process while neoclassical models

That is why we believe that the future trend will shift remain valid as reduced-form predictors of equi-

toward more integrated financial products and ser- librium asset prices and allocations, we now offer

vices, which are easier to understand, more tailored an example of how behavioral factors can have

toward individual profiles, and permit much more

27 similar effects. As we know from the empirical

effective risk selection and control. studies done by Kahneman, Tversky, and other

behavioral scientists, people’s financial behavior can

Production of the new brand of integrated, cus- differ systematically from the neoclassical assump-

tomized financial instruments will be made eco- tions of rationality. In particular, it has been shown

nomically feasible by applying already existing that when individual choices depend on probabili-

financial pricing and hedging technologies that ties, subjective estimates of these probabilities are

permit the construction of custom products at often subject to large biases. It does not neces-

“assembly-line” levels of cost. Paradoxically, mak- sarily follow, however, that the market prices of

ing the products more user-friendly and simpler products whose demand depends on probability

to understand for customers will create consider- estimates—products such as insurance—will reflect

ably more complexity for their producers. The good those biases. To see why, consider the market for life

news for the producers is that this greater complex- insurance.

ity will also make reverse engineering and “product

knockoffs” by second-movers more difficult and Suppose that people systematically underestimate

thereby, protect margins and create franchise values their life expectancies. Then, if they are risk-averse

for innovating firms. Hence, financial-engineering (or even risk-neutral) the price they will be willing

creativity and the technological and transactional to pay for life insurance will be “too high” relative to

bases to implement that creativity, reliably and the actuarially fair price. For example, suppose that

cost-effectively, are likely to become a central the actuarially fair annual price is $20 per $10,000

competitive element in the industry. of insurance, but people would be willing to pay

$40 as their “reservation” price. What would be the

These developments will significantly change the likely institutional dynamics of price formation in

role of the mutual fund from a direct retail product this market?

to an intermediate or “building block” product

embedded in the more integrated products used Life insurance firms that enter this market early

to implement the consumer’s financial plan. The might earn large profits because they can charge

“fund of funds” is an early, crude example. The the reservation price of $40 while their underwrit-

position and function of the fund in the future ing cost will be the $20 expected loss. But others

will be much like that of individual traded firms will examine the mortality data, calculate the spread

today, with portfolio managers, like today’s CEOs, between price charged and the objective costs of

selling their stories of superior performance to supplying life insurance, and soon discover the

professional fund analysts, who then make rec- profit opportunity available. If there are no effective

ommendations to “assemblers” of integrated retail barriers to the entry of new firms, price competition

financial products. 28

will drive the price to the zero excess-profit point.

4.3 Example 3. Irrational pessimism/optimism Thus, in the long-run, competitive equilibrium,

life insurance prices will reflect the rational unbi-

Having given two examples of how transaction costs ased probabilities of mortality, even though every

can endogenously determine financial structure and buyer of life insurance has biased estimates of

F Q 2005 J O I M

IRST UARTER OURNAL F NVESTMENT ANAGEMENT

10 R C. M Z B

OBERT ERTON AND VI ODIE cost of capital that is the same as predicted in a

these probabilities. The institutional structure of neoclassical no-home-bias equilibrium.

providers of this risk-intermediating function and

its dynamics of evolution may be greatly affected Thus, the final equilibrium asset prices and allo-

by this behavioral aberration even though asymp- cations will be as predicted by neoclassical finance

totically it has no effect on equilibrium price and theory. However, the institutional structure in which

once again neoclassical pricing obtains, as a reduced

29 specific financial functions are executed may be

form. materially determined by investor home bias. Of

all possible institutional structures that are consis-

4.4 Example 4. Home bias tent with the neoclassical equilibrium, FSF looks

for the one that most effectively mitigates the dis-

Now consider the well-documented “home-bias”

30 tortionary effects of home bias. Thus, instead of

effect in portfolio selection. Several rational expla- mutual funds and other investment intermediaries

nations for this effect have been proposed in the exclusively serving the function of international

economics and finance literature—for example, diversification on behalf of US residents, home bias

higher information costs for foreign vs. domestic

31 may cause domestically based manufacturing and

shares. But suppose that the reason is indeed an service companies to perform this diversification

irrational bias against investing abroad. Thus, US function through direct investment.

residents prefer to invest in the shares of US corpo-

rations just because they are domiciled in the United Much the same story would be true at a more micro-

States. They, therefore, invest far less abroad than level for regional biases within a country’s borders.

is optimal according to the neoclassical model of For example, Huberman (1999) reports that people

optimal diversification. invest disproportionately in the shares of their local

Bell Operating Systems. Again, we argue that this

Does the posited behavioral “aberration” result behavior does not necessarily lead to a distortion in

in an equilibrium allocation different from the equilibrium prices of shares relative to the neoclas-

neoclassical prediction? sical prediction. However, this behavior would lead

one to predict that Bell operating companies located

Not necessarily. If US corporations were to invest in more investor-rich regions might branch out

only in US capital projects, then with investor home and invest directly in operating companies in other

bias the equilibrium cost of capital and expected less wealthy regions. Cross-regional diversification

return on shares for US companies would be lower would thus be performed by the operating tele-

than in the neoclassical equilibrium, and higher phone companies themselves rather than by mutual

for non-US projects and firms. However, with funds and other “pure” financial intermediaries.

value-maximizing managers and absent legislative

restrictions on investment, this equilibrium is not Note the operation here of the “invisible hand.”

sustainable. With the lower cost of capital for Each individual investor retains his/her home-

the shares of US corporations, US firms will find biased behavior, and firm actions are driven by the

that direct investments abroad will have higher net

32 motive of maximizing net present value, without

present value than domestic ones. Asymptotically requiring any explicit awareness of that behavior.

in the limiting case of no other imperfections except

investor home bias, US corporations would end up Recognition that endogenous institutional changes

issuing shares in the United States and investing may affect the influence of home bias on asset

overseas until they reach an asset allocation and

J O I M F Q 2005

OURNAL F NVESTMENT ANAGEMENT IRST UARTER

T D F S 11

HE ESIGN OF INANCIAL YSTEMS

“regret insurance,” then investors may be no more

prices, if that bias is behaviorally driven, suggests at risk of realizing regret from paying the premium

some interesting time series tests which compare than from the purchase of other standard forms of

the amounts of stock of companies held directly insurance, such as fire and theft protection on a

by “locals” who are not managers of the firms in house or car.

the 1950s, 1970s, and 1990s. One might expect

that the much larger institutional holdings of stocks Those regret-averse investors who would otherwise

in the latter periods would mitigate the home bias

33 hold sub-optimal portfolio strategies because of

effect. Much the same tests could be applied to strong regret aversion may well be willing to pay a

investments in local mutual fund groups that over premium price for such an option. The theory lay-

time have moved into investing in shares of foreign ing out the production technology and production

companies. cost for an intermediary to create look-back options

first appeared in the scientific literature more than

34 36

4.5 Example 5. Regret aversion two decades ago. Today, look-back options are

available widely over-the-counter from investment

and commercial banks.

Now consider another example from investing to

illustrate how institutions might respond to an irra-

tional behavior pattern by creating new financial The point of this example is to suggest that if

instruments. Suppose that people do indeed have regret aversion is indeed a significant behavioral

an aversion to feeling sorry after-the-fact for earlier phenomenon, then FSF theory predicts an insti-

investment decisions they made. If this behavioral tutional response in the form of creating products

trait is widespread, then we might expect to find a like look-back options. If regret is so widespread

demand in the market for “look-back” options. A that it affects equilibrium prices, then at a given

look-back call option gives its owner the right to buy point in time, one investor’s regret concern about

an underlying security at the lowest price at which selling a security is likely to mirror another investor’s

it traded during the term of the option. Similarly, a regret concern about buying that security. If so,

look-back put option gives its owner the right to sell a properly designed institution or market may be

the underlying security at the highest price at which able to “pair off ” these offsetting demands and neu-

35

it traded during the term of the option. tralize the regret effect on asset demands. Thus,

Thus, by the theoretically predicted incremental effect that

paying a fixed insurance-like premium, the investor this behavioral phenomenon might have had on

is assured of no regret from his investment deci- equilibrium asset prices and allocations in an insti-

sions during the subsequent period covered by the tutional environment without look-back options or

option, because he will buy the stock at the lowest another functionally equivalent institution can be

price (or sell it at the highest price) possible. There mitigated or eliminated entirely with their inclusion

is of course a prospect for regret from paying for 37

by institutionally rational intermediaries.

the option itself, if the ex post gain from the option

does not exceed its cost. However, such regret, if any,

may well be minimal because the premium is fixed 4.6 Example 6. Organizational design

in advance (bounding the amount of regret) and the

“base” price for comparison (if the investor had sold

or bought at some point instead of purchasing the In this example, we move from financial products to

option) is likely to be “fuzzy.” Furthermore, if the consider how organizational design itself might off-

marketing of the option frames it psychologically as set dysfunctional individual behavior and produce

F Q 2005 J O I M

IRST UARTER OURNAL F NVESTMENT ANAGEMENT

12 R C. M Z B

OBERT ERTON AND VI ODIE to attempt to alter the behavior of the individual

an end result that is in line with neoclassical pre- analyst.

dictions. For example, suppose that when making

investment decisions individually, analysts tend to

be optimistic and overconfident in their forecasts 4.8 Example 8. Sociological elements

38

for the securities they study. Let us suppose fur- 40

of behavioral finance

ther that when individual analysts, each of whom

has studied a different security, are brought together The preceding examples of behavioral distortions of

in a group and asked to form a group consensus efficient risk allocation and asset pricing all involve

regarding all of the securities, the bias is mitigated cognitive dissonance of individual agents. However,

39

or altogether eliminated. there is another dimension of potential behavioral

effects that is sociological in nature in that it derives

FSF theory would predict a strong tendency for from the social structure of the financial system.

asset-management and other financial-service firms Sociological behavior is neither under the control

to organize investment selections as a group process of individuals within that social structure nor a

including creating investment committees to eval- direct consequence of simple aggregation of individ-

uate the recommendations of individual security ual cognitive dysfunctions. A classic instance within

analysts and portfolio managers. The committees 41

finance is the Self-Fulfilling Prophecy (SFP),

would have the effect of mitigating the bias of the applied for instance to bank runs: a bank would

individual analysts. Consequently, there would be remain solvent provided that a majority of its depos-

little or no impact of this individual bias on actual itors do not try to take their money out at the same

investment choices and equilibrium asset market time. However, as a consequence of a public proph-

prices. esy that the bank is going to fail, each depositor

attempts to withdraw his funds and in the pro-

cess of the resulting liquidity crisis, the bank does

4.7 Example 7. Don’t change behavior; solve indeed fail. Each individual can be fully rational

with institutions and understand that if a “run on the bank” does not

occur, it will indeed be solvent. Nevertheless, as a

consequence of the public prophesy, each depositor

Now suppose it were possible to change the behav- decides rationally to attempt to withdraw his sav-

ior of individuals to make them less optimistic ings and the prophecy of bank failure is fulfilled.

and overconfident when analyzing individual secu- As we know, one institutional design used to off-

rities. Although such a change in behavior would set this dysfunctional collective behavior is deposit

eliminate the bias, it might be better not to tin- insurance. There are of course others.

ker with the behavior of individuals. The reason is

that although optimism and overconfidence are dys-

functional in the domain of security analysis, they “Performativity” or Performing Theory has been

may be functional in other domains vital to indi- employed as a mode of analysis with respect to the

vidual success. That is, there can be unintended accuracy of the Black–Scholes option pricing model

and unanticipated consequences of this action. By in predicting market prices of options, exploring

eliminating a person’s optimism and overconfidence whether the model’s widespread public dissemina-

in general, we may therefore do more harm than tion and use by option traders may have actually

good. Thus, it may be considerably better to rely caused market pricing to change so as to make the

42

on investment committees as a means of offsetting model’s predictions become more accurate. Other

the individual bias caused by overconfidence than recent work applying sociological analysis to finance

J O I M F Q 2005

OURNAL F NVESTMENT ANAGEMENT IRST UARTER

T D F S 13

HE ESIGN OF INANCIAL YSTEMS

5.3 Theory as a predictor of practice

issues includes studies of the sociology of arbitrage

and understanding the development of derivative

43

and other financial markets. As technology progresses and transaction costs con-

tinue to fall, finance theory is likely to provide

increasingly more accurate predictions and prescrip-

5 Elements of functional and structural tions for future product innovations. Combining

finance behavioral theory with neoclassical theory, together

with existing theory within New Institutional Eco-

In this section we review the main analytical ele- nomics, should produce better predictions and

ments of FSF as exemplified by the cases of the prescriptions for the kinds of institutional changes

preceding section. 45

to expect.

5.1 Functions are the “anchors” 5.4 Institutional rationality versus individual

irrationality

When studying the dynamics of financial systems,

it is best to adopt an analytical framework that treats Even when individuals behave in ways that are irra-

functions rather than institutions as the conceptual tional, institutions may evolve to offset this behavior

44

anchors. In this analytical framework the func- and produce a net result that is “as if ” the individ-

tions are exogenous, and the institutional forms are uals were behaving rationally. This is a version of

endogenously determined. Adam Smith’s “invisible hand.” Structural models

that include transactions costs, irrational behavior,

or other “imperfections” may give distorted predic-

5.2 The point of departure is the neoclassical tions when framed in a neoclassical “minimalist”

paradigm institutional setting of atomistic agents interact-

ing directly in markets. It is, therefore, essential

When analyzing changes in parts of the financial sys- to include the endogenous institutional response.

tem, a useful point of departure is the neoclassical Studies of the impact of transactions costs or irra-

paradigm of rational agents operating opportunis- tional behavior patterns would be greatly enhanced

tically in an environment of frictionless markets. if as a matter of format, they included a section

If existing prices and allocations fail to conform to on institutional designs that have the potential to

the neoclassical paradigm, it is helpful to focus on mitigate the effects of these patterns on prices and

why this is so. The possible causes of such a failure allocations. The resulting institutional design, if not

might be: already in place, can be seen as providing either

a prediction about the dynamics of future insti-

• Existing institutional rigidities, in which case tutional change or as a normative prescription for

we might consider applying institutional design innovation.

techniques to circumvent their unintended and

dysfunctional aspects or abolish them directly, if

the institutional sources are no longer needed. 5.5 Synthesis of public and private finance

• Technological inadequacies, which may disap-

pear over time as a result of innovation. The FSF approach has no ideological bias in select-

• Dysfunctional behavioral patterns that cannot be ing the best mix of institutions to use in performing

offset by institutional changes. financial functions. It treats all institutions, whether

F Q 2005 J O I M

IRST UARTER OURNAL F NVESTMENT ANAGEMENT


PAGINE

23

PESO

151.92 KB

AUTORE

Atreyu

PUBBLICATO

+1 anno fa


DETTAGLI
Corso di laurea: Corso di laurea magistrale in analisi economica delle istituzioni internazionali
SSD:
A.A.: 2011-2012

I contenuti di questa pagina costituiscono rielaborazioni personali del Publisher Atreyu di informazioni apprese con la frequenza delle lezioni di Valutazione e finanziamento dei progetti e studio autonomo di eventuali libri di riferimento in preparazione dell'esame finale o della tesi. Non devono intendersi come materiale ufficiale dell'università La Sapienza - Uniroma1 o del prof Cataldo Alessandro.

Acquista con carta o conto PayPal

Scarica il file tutte le volte che vuoi

Paga con un conto PayPal per usufruire della garanzia Soddisfatto o rimborsato

Recensioni
Ti è piaciuto questo appunto? Valutalo!

Altri appunti di Valutazione e finanziamento dei progetti

Metodi di capital budgeting
Dispensa
Contabilità economico - patrimoniale
Dispensa
Fondi di Ricchezza Sovrana - Pasca di Magliano
Dispensa
Matematica finanziaria - Consiglio
Dispensa