The proposed study will develop a goal-programming framework to facilitate implementation of Development Agencies [DA] policies for credit extension to and evaluation of micro credit banks [MCB] which in turn provide credit as well as rudimentary business training and basic social services to the poor lacking collateral. Because the hierarchy of and tradeoffs among goals embedded in these policies vary over time and in different places, it is expected that such a framework will provide valuable aid to the policy maker in the DA.

Microcredit banks [MCB] extend group-based credit to the poor needing funds to invest in home-based enterprises. The borrower is too poor to have collateral required for credit from formal sources such as commercial banks. Each member of a group is responsible for the loans to both him[her]self and other group-members. Credit extension is typically integrated with [a] a savings program, and [b] provision of rudimentary business training as well as basic social services related to health, nutrition, children’s education, and family planning. This integration improves loan repayment behavior, poverty alleviation, and social development along health, nutrition and education dimensions.

A MCB typically obtains funds from development agencies [DA] such as the USAID and the World Bank at subsidized interest rates, and loans these funds at or near the market rates. The spread between these two rates is around 10- 15%. A DA providing funds to a MCB on a recurrent basis requires assessment of the MCB on several dimensions:

It should be noted that measurement or quantification of the above criteria set and their targets is subject to varying degrees of difficulties. For instance, bad debt experience is easily measurable and can be readily assessed against the benchmark yielded by alternative forms of lending in similar environments or obtained from historical experience. On the other hand, the rate of migration to urban areas can only be observed in the long run and would require control of other factors that also influence migration. Leaving aside the measurement problems, the policy maker encounters several additional challenges pertaining to the above criteria sets and their benchmarks:

• It is extremely unlikely that the observed value of a target will coincide with the benchmark determined a priori. How should the deviations from the benchmark be viewed especially since deviations in either direction from benchmarks are unlikely to be symmetrically desirable or undesirable? A given magnitude of improvement in bad debt experience with respect to the benchmark may be only mildly desirable, whereas an equal magnitude of deterioration in bad debt experience may be strongly undesirable.

• It is also possible that deviation from the benchmark for a goal may be undesirable in itself; however, when it is viewed in light of higher achievement of a preferred goal, such a deviation may be tolerable. For instance, higher overhead leading to barely break-even profits may be acceptable due to improved sanitary conditions in a given community. But can such tradeoffs among criteria that are not necessarily inconsistent be identified a priori? More pertinently, can the benchmarks be modified at the planning stage in order to reflect goal hierarchy and potential conflicts among goals?

• It is highly unlikely that determination of optimal benchmarks reflecting priorities in a given situation will have general applicability over time or space. Thus, urban migration may be less of a concern in a sparsely populated country than in, say, Bangladesh or India. Or, enhanced living standards over time may lead to reduced fertility and hence less focus on birth control but greater concern for improved technology or better diet. Similarly, even when benchmarks targets are set, performance evaluation of a MCB may have to reflect a change in the environments. The Cambodian example of 30% bad debt experience in a given year may need to be assessed against the adverse impact on borrowers’ incomes because of changed intensity of the civil war.

Differing needs as well as varying environments in which a particular MCB is functioning suggests that a DA cannot utilize a set of uniform standards applicable to all MCBs seeking financial support. Customizing these standards both for financing initially and later assessing efficacy of the MCB becomes critical for a DA. This study proposes to help a DA in these tasks by developing a goal programming [GP] framework.

The goal programming [GP] framework requires a modified linear programming [LP] approach [Mehta, 1974, pp.160-4]. An initial LP formulation of the problem is as follows: Max CX subject to AX B; X where C, A, and B are vectors or matrices of parameters whose values are known and constant. X is the vector of decision variables to be determined. Specifically, the objective function CX may represent the benefits for the household, business and community due to credit extension as well as the contribution margin for the micro credit bank on credit extension. The constraints may represent the size of potential clientele, the maximum level of bad debt experience, minimum level of literacy, health, and nutrition standards to be achieved by the MCB. Obviously, greater benefits for the community are possible only through more extensive credit extension and larger overhead for the MCB that would reduce its profitability and require more funds from the DA.

The basic premise of the GP formulation is that not all constraints are inviolable, and some constraints that embody a policy makers goals can be modified; indeed, the policy maker knows the goals, their hierarchy, and their provisional targets [initially appearing in the LP formulation] as well as directional desirability of deviations from these target values .The GP formulation requires modification of both the policy constraints thus identified and the original objective function [CX] as equality constraints so as to include the initial target values as well as deviations from these targets. Knowledge of the hierarchy of goals and their directional desirability permits quantification of penalties [rewards being the negative penalties] for deviations from each target. The new objective function is then to minimize the penalties associated with deviations.

The solution to the problem cast in the GP framework yields optimal levels of targets for the identified goal set that reflects the hierarchy of the goals, interrelated targets and their permissible deviation ranges, as well as potential tradeoffs among them. To the extent that this hierarchy and the potential tradeoffs among goals vary for a MCB over time, the optimal target values will also be unique for a MCB at a given time. DAs that provide funds to MCBs functioning in different environments face the challenge of determining (a) what to expect realistically from a particular MCB if it is provided funds; (b) whether such funds allocation individually and aggregately is consistent with the DAs own mission; and (c) whether the actual performance of the MCB meets the expectations identified in (a). Because tasks in (a) and (c) are unique for a MCB, the research effort is aimed at undertaking a pilot project that addresses these tasks in a goal-programming framework.

• Khandker, Shahid, B. Khalily and Z. Khan, Is Grameen Bank Sustainable? working paper, 1994, World Bank. Washington, D.C.

• MkNelly, Barbara and C. Waterip, Impact Evaluation of Freedom from Hungers Credit with Education Program in Thailand, working paper (1993) Freedom from Hunger Foundation, CA.

Project financing is a specialized method of financing large projects. The projects are risky and capital-intensive infra structure or resource-based. This financing technique allocates financial risk among creditors, borrowers, investors, and operating companies involved in the project. It is a type of privately placed debt. The debt is tied to the fortunes of the project thus minimizing the risk exposure of the parent company. An example of Project Financing is the technique used by the British Petroleum (BP). In 1972, the BP needed to pay for the development of huge Forties Field in the North Sea. Other examples of ‘project financing’ have been the financing of Alaska Pipeline. This financing technique is being used extensively in developing countries because of the necessity of undertaking infrastructure development. For example, among non-Muslim nations, in 1995 there were 31 projects for $ 7 billion, 29 projects for $ 37 billion, and 120 projects for $ 96 billion under consideration in Mexico, India, and China respectively. Among the Muslim nations corresponding figures were 14 projects for $10.9 billion, 15 projects for $24 billion and 12 projects for $6.2 billion under consideration in Indonesia, Malaysia and Turkey respectively. An analysis of the large projects under consideration in the developing countries in 1995on the basis of project financing shows the 11% of the total number and 14% of the total dollar investment were located in the Muslim world.

All Muslim countries are included in the category of developing countries. Their infrastructure needs to be developed for their economic development. Project financing being an efficient financing technique for infrastructure development can play a significant role in the development of Muslim countries.

1. Project financing as practiced in the western world involves debt financing. This results in paying and receiving interest. Islam prohibits interest-based transactions. According to a verse in Quran, "Those who devour interest do not rise except as rises one who Satan has smitten with insanity. That is because they say: ‘Trade also is like interest; whereas Allah has made trade lawful and interest unlawful’."

his prohibition on interest has made the use of project financing difficult. Very few Muslim countries have been able to use project financing. Some examples are a power project in Pakistan, a real estate development project in Iran and a highway construction in Malaysia. A greater use of this method of financing will satisfy the financing needs for the infrastructure development of Muslim countries leading to their economic development. Islam does not permit the charging of interest. Western banks, because of risk, are not willing to provide funds on the basis of equity financing. Thus what is religiously acceptable to Muslims is not economically acceptable to the western lenders.

2. Islamic investors tend to have their money invested short-term. A large part of their profit-sharing deposits are used to finance trade and inventory. Thus Islamic banks have two alternatives: either to run a big gapping risk or leave the field of long-term financing business. Gapping risk is magnified by the fact that Islamic banks do not have an inter-bank market where they could swap short and medium tem deposits against long-term investments. Securitization of investment vehicles is another method of narrowing the gapping risk. However, as of now this technique has not been employed because of lack of an appropriate derivative security.

3. Islamic financing has to be approved by a Sharia board. Because of varying interpretation of Sharia, a certain type of security that is Islamically acceptable in one country may not be acceptable in another. This restricts the standardization of securities leading to designing different investment vehicles for projects in different countries.

1. To suggest an integral model for using project financing acceptable to both the lenders (mostly western) and the project sponsors who would not agree to interest based transactions, and

2. To suggest a new equity security that will have the characteristics of debt security for the lenders and an equity security from the point of view of a firm located in the Muslim world.

It is a theoretical paper. Not enough data are available to run any statistical techniques to prove or disprove the utility of the suggested model. Examples of some of the recent transactions, such as Hub River Project in Pakistan, Construction of an Airport in Malaysia, and A Housing Project in Iran culminating into some recent big projects in Muslim countries in conjunction with financial instruments Islamically acceptable have been used to arrive at the theoretical model. Innovative Securities used by various Muslim countries to circumvent the problem of paying interest, particularly the ones listed below are analyzed:

(a). Bonds with Warrants sold by Bank Islam Malaysia for Petrona , Malaysian Oil Co.

(b). Petroleum Securities Corporation issued Petroleum bonds in Islamic version. They carry no interest rate. They are paid dividends linked to the sale of oil by the issuing firm.

(c). Tehran municipality issued participation certificates for finance to support a commercial development project. The securities are linked to the project and a 20% return to a four-year bond is expected. The rate is considered to be comparable to the rate available on similar risk projects.

(d). Use of a simultaneously callable, puttable and convertible common stock designed in this paper to satisfy both the lenders and the borrowers religious and financial concerns.

Security: The security is callable at the discretion of the sponsors. The call will be deferred to a date when the amount invested by the investors (banks, etc.) has been fully amortized with a predetermined rate of return. The call price will equal the balance of the investment if it is partially amortized.

The security is puttable at the discretion of the investors. The put price will equal the original investment minus the principal amortized, again using a predetermined rate of return. This put will be exercised if the sponsors fail to pay according to the established pay out scheme. The security on putting will automatically convert to an interest-free loan to the sponsors. This debt will have a lien on the property of the project. It may also be guaranteed by the sponsors or the government. Guaranteeing a loan is permissible in Islam particularly when the guarantor is wealthier than the debtor. According to Quran, if a debtor is in a straitened position then he should by given reprieve. This implies that the debt should not be called immediately. However, if the sponsors have committed a fraud then the investors may take legal action and/or exercise their right to attach the property of the project. The security is convertible into interest free loan at the discretion of the investor to debt if the investors do not get their investment back with a predetermined rate of return. This is a common security. Investors will be stockholders in the project. They will have access to full information. The chances that the sponsors or the subsidiary created to develop the project will commit a fraud or deviate from the contractual obligations is minimized.

Pricing of the security: The cash flows are known with considerable degree of certainty. The product or the service is presold in many cases. Only a discount rate is needed to price the security. The security is callable only after all of the financial requirements of the investor have been met. This will mean no premium for this feature. The security being puttable at par and convertible to an interest free loan will be deemed less risky by the market. Hence a reduced discount rate should be used to price it. Another reason for a lower discount rate is the establishment of a sinking fund that is commonly required by banks for long-term loans.

Combination of solutions 1 and 2 is likely to overcome financial hindrances to the use of project financing.

3. Interpretation of Sharia: At present the possible solution to this problem is creation of a single religious board in a country or a group of like-minded countries. As more and more of the projects are analyzed on religious grounds it is expected that commonalties among various countries will emerge.

To overcome the gapping risk, Islamic Investment Companies could raise project-specific funds or create special pools of funds that are used for project financing. Since the investors know the purpose for which the funds are being raised, the gapping risk is minimized. Also it is logical to set up unit trust or mutual fund types of investment vehicles aimed at long-term investors. Kuwait Investment Company and some Islamic Financial institutions are in the process of testing these techniques. To tap the resources of western institutions the use of a puttable, callable and convertible security should be examined carefully. It is likely to be accepted by both the investors and Muslim firms.

Open Market Stock Repurchases The Canadian Experience

During the 1980s, U.S. firms announcing open market stock repurchase programs earned favorable long-run stock returns. Recently, concerns have been raised regarding the robustness of these findings. This comes at a time of explosive growth in the adoption of repurchase programs, both in the U.S. and worldwide. This study further investigates long-run performance following open market repurchase announcements using a recent sample of Canadian programs. Undervaluation appears as important factor in motivating Canadian repurchases as in the U.S. Moreover, as in the U.S. the Canadian stock market seems to discount the information contained in program announcements. Toronto Stock Exchange listed firms announcing programs between 1989 and 1995 show excess performance measured relative to Fama-French (1993) Three factor model of approximately .5% per month for a period of three years following the announcement. Canadian value stocks announcing repurchase programs as well as firms announcing large repurchase programs show particularly favorable abnormal long-run performance.

This paper investigates the announcement effects of both rights issues and cash offerings by the Dutch firms for the period 1985 -1996. We further test signaling and market feet back models. Finally, the long-term performance of firms the issued seasoned equity offerings is investigated.

From the total of 91 firms that went public, 31 firms issued seasoned equity or rights issue. The initial return for reissuing IPO firms is 8.89%; that is 1.67% higher than the equivalent return for non-reissuing firms. Reissuing firms are more likely to issue a larger proportion of their total equity. They issue on average 40.0% of their total equity as opposed to 32.1% offered by non-reissuing firm. Reissuing firms also offer a larger amount of equity at the IPO, Ÿ 1031.6million versus Ÿ 78.1million for non-reissuing firms.

The average return on a portfolio of all IPOs, bought on the first day of the calendar month following the IPO month and held until the 48th month, is 1.66%. The reissue group has an average holding period return of 29.65% against a return of -12.97% for the no-reissue group. The same trend is supported when we use the wealth relative, where the no-reissue firms perform 25% less than the market in the same period. The reissuing firms on average outperform the market by 7%. This results are opposite of what have been found in other countries.

We further investigate returns for the reissue group by splitting them into the firms that return to the market with rights issues and firms that return to the market with seasoned equity offers (SEOs). The latter group has an increasing return from the first month until the twelfth month after the IPO. This positive relation holds in relation to the market, where the SEOs outperform the market by 37%. After this period, the companies that issue rights perform better than the companies that issue seasoned equity. Moreover, we find the market reacts less unfavorable on the announcement of a right issue than on the news of a seasoned equity offering.

This paper develops a multi-period model of an international economy, in which a domestic country has restricted foreign ownership in domestic assets. Also, short selling is limited either explicitly (government regulations) or implicitly (stock exchange collateral requirements). Although Free and Restricted shares (within the same share class of the firm) are equivalent in every way with the exception of ownership eligibility, the combination of these two restrictions can result in a premium for the shares available to both foreign and domestic investors (Free shares) above the domestic investor-only-shares (Restricted shares). The bulk of the attention in the existing theoretical and empirical literature has focused on the premium on Free shares, whereas the resulting equilibrium asset holdings has received tangential treatment. This paper looks at equilibrium asset holdings of domestic and foreign investors.

Perhaps even more puzzling than the existence of premium-priced Free shares is the holding of these shares by domestic investors. Why would a domestic investor be willing to pay a premium for Free shares, given the availability of lower priced Restricted shares? Indeed, the equilibrium holdings of Free shares merits attention in its own right. On this issue, however, existing single period models are not capable of yielding rich, dynamic results. Within the multi-period model found in this paper, domestic investors will hold Free shares in the final decision period only if its price is identical to the Restricted share price. However in prior periods, domestic investors are observed holding the more expensive Free shares, despite the fact that identical dividend distributions and voting rights are obtained with either type of share. The former prediction provides a testable implication of the model, whereas the latter agrees with the empirically observed portfolio holdings of domestic investors.

In many regards, this work bridges frictional asset pricing theory with international investment theory to provide insights and evidence into both that cannot be obtained separately. Indeed, the short selling restriction breaks down the Euler equation into an inequality, such that asset prices are determined by the maximum valuation of the asset's payoffs across all investors eligible to invest in the asset. Due to the simple fact that only one price path is observed for a given payoff stream, the impact of short selling restrictions is hard to quantify or observe empirically. However with the existence of foreign ownership restrictions and separately traded Free and Restricted shares, we have the unique opportunity to observe the distinct prices assigned to a given payoff by foreign and domestic investors. The paper proceeds to explicitly relate the model to the expanding literature on frictional asset pricing. Although not necessary for a premium to develop, asset pricing within this model may be sub-linear, for example. That is, an investor with the maximum valuation of an asset's combined payoffs across all future states of nature will not necessarily have the maximum valuation of the asset's payoff in each individual state of nature. However, scaling an asset's payoffs will simply scale the asset's value.

To provide empirical evidence, this paper scrutinizes premiums on Free shares in Sweden for the period 1985 - 1993, at which time the Swedish government required firms to remove the foreign ownership barrier. Given the distinction between Free and Restricted shares no longer existed after 1993, the Stockholm Stock Exchange eliminated the dual listing of these shares. As a result, 1993 can be taken as the final period in the model, when domestic investors should have eliminated their holdings of premium priced Free shares. These premiums and the corresponding portfolio behavior of foreign and domestic investors provide support for this model, while raising additional questions for future research in the area.

Both academic and professional literature seem to agree that the netting clauses in a swap contract tend to reduce the credit (default) risk for both parties and therefore must be preferred over gross settlement. This risk reduction merit of netting has become a common understanding of academics, practitioners and regulators, and their feeling towards netting has been reinstated in headings such as "Happiness is a Full Net" (Tremble and Sarwa, Euromoney, April 1991). Its wider practice is also being promoted by International Swap Dealers Association, Inc. (ISDA) internationally and is endorsed by academic studies that analyze default risk implication of swaps.

The netting provision of the 1992 ISDA Master Agreement specifies that on any date due amounts in the same currency and in respect to the same transaction can be netted and only the difference be paid to the party who was to receive the larger amount. In case of more than one transaction, the net amount is to be determined in respect of all amounts due on the same date and in the same currency. It is the contention of the market that exchange of difference checks as a result of netting instead of gross amounts will result in risk reduction for the parties unambiguously.

The framework in the articles that have partially analyzed the implications of netting on default risk is an efficient and integrated world market where swaps are zero sum games (for example, Baz and Pascutti, the Journal of Derivatives, Winter 1996, Duffie and Huang, Journal of Finance, July 1996, Sorenson and Bollier, Financial Analyst's Journal, June 1994.) Usually, one of the parties to the swap is practically risk free (the dealer). Thus, the emphasis is on the risky party's chance of defaulting on its debt payments prior to the due payments on the swap. More importantly, when it comes to the comparison of default implications of gross versus net settlement, these authors fix the amount to be swapped. Naturally, the net settlement with lower probability of default given a notional amount fares better since default risk is to be avoided. Unfortunately, the models have not allowed the swapped amount (the notional principal) to be one of the endogenously determined decision variables.

On the other hand there is a line of study by Usmen (Financial Management, Summer 1994) which has looked at the default structure from a different perspective. The model in Usmen is cast in a mildly segmented capital market which has been empirically shown to be a closer approximation to the world capital markets of today (for example, Bonser-Neal, Brauer et al., (Journal of Finance, June 1990). Another distinction in Usmen’s study is that the amount to be swapped is determined along with the swap rate (price) in a value-maximizing framework where both parties are allowed to be risky. The result about default risk that emerged in that study was somewhat surprising. The default risk of swaps in a segmented capital market is a commodity of which more may be desirable to less. The desirability of default depends on market conditions (segmentation per se) and the firms' specific default structures. Usmen study dealt with currency swaps and analyzed gross settlement. The present paper extends that analysis to incorporate net settlement and compares it to gross settlement in a context where the optimal swapped amounts (notional amounts) are also determined.

The analysis is cast in a one-period, two-country state preference framework where the investors are risk neutral but have heterogeneous expectations. There are two value-maximizing firms that enter into a currency swap for their financing needs. Value differentials (the added value the firms create by entering into a swap contract) are computed under two different scenarios: gross settlement and net settlement. In either case, the resolution in insolvency is the limited-two-way payment method. With this resolution, the non-defaulting party is absolved from any payment if the other party is insolvent. The author favors this resolution method because of the shortcomings of the alternative full-two-way payment method. (In the full-two-way payment method the parties commit to their swap obligations irrespective of the other party's solvency, consequently, full-two-way payment may be biased against the higher credit party, in most cases a dealer.

In the value differential equations there are two decision variables, namely, the swap rate and the swapped amount (notional principal). Both firms are allowed to be risky. Examination of the value differential equations reveals that the difference net settlement brings over gross settlement is in the content of the default set (the set of states where the contract will be void because one or both firms are insolvent). Comparing the default sets resulting from each type of settlement, it is apparent that when the amount to be swapped is fixed the default set is smaller with net settlement implying lower probability of default. However, if swapped amounts (notional principal) are allowed to vary, the resulting changes in the contents of the two default sets are very different. Since there is more room to increase the notional amount (beyond the capacities of the counterparties), there is a greater chance of including states where exchange rate outcomes are favorable to both parties. This increases the chances of increasing the differential values beyond that of a default free swap. In this analysis what is important with risky swaps is not the number of states (the probability of default), but rather the particular states in which default occurs. Thus, selection of the method of settlement might be used to better position oneself in terms of choosing the states where default occurs.

A numerical example illustrates the above default implications of net versus gross settlement. Value differentials are computed under each scenario by using market and firm specific parameter values. In the computations the decision variables are varied to search for the optimal pair of the swap rate and the swapped amount that maximize the added values of the swap counterparties. When the swap rate is kept low and the swapped amount is varied, gross exchange is not viable since value differentials for both firms are always negative. However, with net settlement, the situation is reversed. At a certain notional amount both parties gain from the currency swap. It turns out that net exchange is preferred over gross exchange not because it is less risky at a certain level of notional amount but because it allows a different default structure (default set). Surprisingly, the reason for selecting the net settlement option is not its risk reduction attribute but rather the flexibility it gives in designing the default structure. It can do so because the notional amount is an imaginary value that can be extended beyond the earning capacity of the counterparties involved.

With the present paper, the author attempts to clarify an issue related to the net settlement provision of a swap contract. The common belief in the market that netting reduces the default risk and therefore should be desired is put into a different perspective. By choosing a mildly segmented capital market as a backdrop and by allowing the swapped amount (the notional principal) to vary (in contrast to the common analytical framework), the present paper demonstrates an additional merit of net settlement. While it is true that for a fixed swapped amount, net settlement necessarily reduces the probability of default over gross settlement, the risk reduction merit might not be desirable. By increasing the notional principal beyond the capacities of gross settlement, one can alter the default structure of the contract such that the overall gain to the counterparties supersedes those of risk free swaps as well as risky swaps with gross settlement. The gains are market and firm specific, therefore, in designing a swap contract, net versus gross settlement is another option to be evaluated along with the choice of the swap rate and the swapped amount.

The North American Freed Trade Agreement (NAFTA) is the most significant trade treaty in the history of the United States. Initially, the original agreement (implemented in 1989) included only the United States and Canada (the U.S.-Canada Free Trade Agreement). NAFTA was formed with the entry of Mexico on January 1, 1994. If California was a separate nation, it would be the seventh largest economy in the world. Therefore, the impact of NAFTA on California is significant within this three-nation Agreement. This study focuses on the impact of NAFTA on small California firms.

It is well known in the business and economic literature that most new job formation comes from small firms. In the late 1970s, Professor David Birch of MIT analyzed small businesses using information derived from Dunn and Bradstreet. With a sample of 5.6 million firms, he concluded: that companies will less than 100 employees created 80% of net new jobs in the United States economy during the 1970s (Gilder, 1984). This was later found to hold true for the early 1980s as well (Osteryound and Newman, 1993).

This project has two main objectives. The first objective is to determine the reasons why small California firms export goods and services to other countries. The literature cites several reasons why firms export (Czinkota, et al., 1995). For example, is the profit motive the only reason for exporting or are there other reasons? Does the firm export because it believes it is selling a unique product or do the firm's products have a technological advantage over competitors? Does the firm's management possess special knowledge about foreign customers or market situations? Is the firm exporting because it is close to foreign customers and ports?

Of most importance to this study is the relationship between the reasons firms export, generally, and the reasons they export to NAFTA countries. Since approximately 80 percent of the firms surveyed sell some of their exports to other NAFTA countries, do the reasons for exporting depend on the percentage of exports firms sell to Mexico and/or Canada? In other words, does NAFTA make a difference with respect to the firm's motivations to sell goods abroad. The second objective of this paper deals with the impact of NAFTA on small California firms. For example, have exports to Mexico increased because of NAFTA, and if so, by how much? Have sales to Canada increased because of NAFTA, and if so, by how much? Has the number of the firm's employees increased or decreased because of NAFTA? Have firms made direct investments in Mexico or Canada because of NAFTA or do they plan to do so in the future?

A questionnaire was sent to the Presidents and CEO's of 600 exporting firms listed in the California International Trade Register. Most of the responses in the survey instrument were based on a six-point Likert scale. These firms represent six SIC categories. Past experience has shown that it is better to sample many firms in a few industries than a few firms in a large number of industries. Also, because the great majority of new employment occurs in small firms, the firms that have been selected have 200 employees or less. The survey instrument was divided into three main sections:

1. General Information, (2) Reasons for Exporting, and (3) Firms that export to Canada and/or

The data was examined with respect to firm size, using the number of employees and asset size as the measures of size; and by SIC codes. Of most importance, the data was also examined according to the percentage of exports traded with NAFTA vs. non-NAFTA countries. Although this study deals exclusively with small California firms, do the "reasons" that companies have for exporting differ with respect to whether the firm has fewer employees (< 50 employees) or a greater number of employees; or according to SIC code? The strength of the "reason" to export have also been investigated with respect to the percentage of exports to Canada and Mexico. As mentioned before, are the "reasons" for exporting dependent on or do they change because of NAFTA?·

As another example, for small California firms that trade with other NAFTA countries, is the change in the number of employees hired related to the size of the firm? For example, have firms with < 50 employees been affected to a greater extent than firms with more than 50 employees. How does the number of employees hired differ with respect to SIC code? Also, how does the mix (or percentage) of exports to non-NAFTA vs. NAFTA countries impact these relationships? The data has been analyzed using one- and two-factor analysis of variance and other statistical procedures.

Trade agreements have been around from the beginning of civilizations. They have brought both prosperity and advances in civilizations as well as conflict and war. Although there are many trade agreements in the Americas, NAFTA, North American Free Trade Agreement, draws the most attention. The two main direction of NAFTA is free trade and free investment. The three countries involved are Canada, Mexico, and United Stated. It is assumed that Chile will be the next country to enter NAFTA.

Our question is how to develop a trade association of the Americas. This is referred to Free Trade Association of Americas (FTAA). How should this be accomplished? Should other countries in the Americas enter NAFTA or form other trade agreements? The author approach will be the application of an Input, Process, Output (IPO) model. The model is a static view the system. Great care is taken in identifying the system boundaries versus limited attention to the dynamic nature of the system.

Inputs are defined in three groupings: controllable, inputs, and uncontrollable. Controllable inputs are viewed as variables that can be changed in the near future. Inputs are the variables that go into the system. The uncontrollable variables are not changeable in the near future. It is important to mention that this model does not look at or consider the future in detail. The processes are centered on people, hardware, software, procedures, and data. For information systems, these areas can be quite extensive. For trade agreements, processes tend to identify conflicts. The outputs are what is desired. Generally it is quality, performance, and very measurable constructs.

The argument for trade agreements is similar to the global economy. If barriers to trade are removed then products will be manufactured where the costs are least. There will be a division of labor in the world and the ability for economies to specialize. Therefore, quality of life will improve for everyone. There are several assumptions in this systems approach example. First, it is assumed that competition and free trade is best for the consumer. Secondly, the leadership is forming trade agreements is not identified and it is assumed that favoritism does not exist. Of course, the systems approach could be applied for a specific country or region's point of view and this would lead to what is best for the stakeholders. However, the latter approach would be much harder since detailed preferences are needed as to what subgroup in the country or region should benefit and by how much.

The major input variable is the trade agreement. In the case of NAFTA, the agreement concerns trade and investment among the United States, Canada, and Mexico. (1) NAFTA also includes barriers to trade in agricultural products. Import sensitive sectors are protected up to 15 years. A major concern is sanitary and phytosanitary measures. Other issues are export subsidies, internal support and grade and quality standards. Another major issue is rules of origin. The controllable variables are viewed as something that can be negotiated. The author includes reduce tariffs and barriers, labor movement, one currency, export subsidies, rules of origin, and import sensitive outputs. To some extend retaliation and side agreements are controllable variables.

The uncontrollable variables are displacement, resistance, pollution, labor standards, health and safety, anti dumping rules, risk for investors, risk for nation, education, and tax structure. In many ways, the uncontrollable are the real problems to overcome for implementation of a trade agreement. As an example of resistance, let us consider Washington State apples shipped to Mexico. If they must go through a port of entry in Texas the cost of shipping is much higher then a port of entry in California.

Risk for investors and risk for nation is difficult to eliminate. The assumption with trade agreements is changes will take place in each country. Individuals or companies should invest in new products and services. If agree upon changes are not implemented or are prevented, the risk of loss to individuals and nations is great. The uncontrollable variables are major causes of risk and therefore risk can not be reduced.

1. People: A free market is best for the citizens given the risk issues are resolved. If only certain segment of society is to benefit, then a stakeholder's approach can be taken. 2. Hardware: The physical facilities and infrastructure should be viewed as hardware. Additionally, components of the agreement require new hardware especially the health and safety issues. 3. Software: Software to facilitate the trade agreement is needed. For example, rules of origin and anti dumping regulations require complex data collection and analysis. 4. Procedures: Procedures are the checks and balances. Most of the requirements in trade agreements are procedural based. The hope of course is that the future will have few procedures. 5. Data: Data is limited and the information derived from data is more limited. There are plenty of reports (2,3) but the information does not give a clear picture. For example, it is not clear what percent of the exports to Mexico from the US simple off set imports from Mexico to US of the same item. The information needed is a clear indication of net growth in production for a given economy.

The outputs include Investment movement and Trade movement; this is certainly true for NAFTA. The authors assume the following three outputs: 1) improved quality of life, 2) lower average cost, and 3) improved efficiency. The author feels that the system approach can be helpful in clarifying the issues and communication of the trade agreement. System theory assumes that controllable variables can become inputs variables with time. The author wonders how many people entering a trade agreement would want this. Secondly, the uncontrollable variables are the major barriers to successful implementation of trade agreements. It is unlikely that uncontrollable variables will ever become input variables. With this prospective, the author wonders if one major trade agreement is worth the effort. It would appear that a variety of more limited trade agreements would be the best strategy.

We propose filtering historical simulation by GARCH processes to model the future distribution of assets and swap values. Options’ price changes are computed by full re-evaluation on the changing prices of underlying assets. Our methodology takes implicitly into account assets’ correlations without restricting their values over time or computing them explicitly. VaR values for nonlinear portfolios are obtained without linearising them. Historical simulation assigns equal probability to past returns, neglecting current market conditions.

Current methods of evaluating the risk of portfolios of nonlinear securities are unsatisfactory. Delta-gamma hedging becomes unstable for large asset price changes or for options at the money with short maturities. Monte-Carlo simulations assume a particular distributional form, imposing the structure of risk that they were supposed to investigate. Moreover, they often use factorisation techniques that are sensitive to the ordering of the data.

To overcome the above limitations we propose to adapt the methodology of Barone-Adesi, Bourgoin & Giannopoulos (1997). We propose to model changes in asset prices to depend on current asset volatilities. Asset volatilities are simulated to depend on the most recently sampled portfolio returns. Our simulation is based on the combination of GARCH modelling (parametric) and historical portfolio returns (non-parametric). Historical returns are adapted to current market conditions by scaling them by the ratio of current over past conditional volatility.

The scaled returns are the basis of our simulation. To simulate a pathway of returns for each of a number of different assets over next 10 days we select randomly 10 past sets or "strips" of returns, each return in a strip corresponding to an asset’s price change which occurred on a day in the past. Thus each strip of returns represents a sample of the co-movements between asset prices. We then iteratively construct the daily volatilities for each asset that each these strips of returns imply according to our GARCH model.

The European banking system is experiencing a phase of vehement change. It is true that the effects of an implementation of a single European currency are in the centre of the current discussion. But the globalization of financial markets and the new banking technologies exert a far greater influence on the future European banking scene. They trigger extensive merger processes and revolutionise the marketing of standardised financial products.

For decades, the European banks have operated in an oligopolistic banking market with intuitive profit margins on interests and commissions as well as agreements on good behaviour in advertising policy. The time of tacit terms is now coming to its end and is being replaced by a growing international competition in which more and more often banks are forced to calculate the cost of their chosen loss leader products in such a way that they can just cover the marginal costs. From a bank internal point of view, the globalization and growing competition of conditions in the deposit and credit business may have a taste of cannibalism. In reality however, the European banking industry is now simply on its way to normality with regard to competition and adaptation of new technologies, a way that industry has long since gone. The fact that the banks are now forced to be more efficient and cost-conscious is of advantage to the bank clients and ultimately supports the European banking industry for international competition.

When canvassing new clients, the good manners will decline due to tighter competition in the banking industry. Banking terms and conditions are especially suitable for comparative advertising. It can be expected that in Europe as well the ban on comparative advertising will be questioned. Irrespective of that, in future the banks will increasingly criticise the competitors’ services, in order to present themselves in a positive way.

The already broken out merger fever in Europe will significantly change public finance. Following economic forces as well as the spirit of the globalization, strategic alliances will be formed, cost-reducing programmes will be developed, and branches will be closed and replaced by multifunctional terminals. The interdependencies and mergers are growing particularly in the insurance industry and banking industry. Due to synergy effects, many branches will be shut down so that the banks’ profitability will be increased. As a result of globalization in Europe the big Cultural Revolution in bank organisation has finally started. But the European banks should have learned from Mao’s Cultural Revolution. The idea of a Cultural Revolution in bank organisation is convincing, for all organisations, be it government, military, church, industry or banks, tend to petrify and therefore have to be questioned and restructured continuously. However, bank employees are not chess pieces that can be moved at will, that can be sacrificed and used again in the next game at the same position, which is why changes in the organisation structure as highly sensitive measures require intensive preparation.

The people are the bank’s most important resource. This is true for bank customers and bank employees. A reading device can be replaced, but it is very difficult to regain a lost bank customer and to re-motivate an employee whose self-esteem is hurt. In Europe, many people talk about the bank customers’ declining loyalty, but the bank employees’ loyalty will change as well because of the globalization and Americanisation of the bank organisation. The employee’s loyalty to the employer and the identification with the bank and the team will be replaced by a growing concentration on one’s own advantages. The intrinsic motivation will be displaced by financial motivation.

Far too long had the European banks allowed their organisation structure to become encrusted, they avoided and repressed competition in investment banking. When the globalization of the banking industry finally forced them to change their policy, they threw out their babies with the bath-water instead of providing fresh bath-water. Many European bank employees have lost their feeling of a bank-specific corporate identity. In general, managers of savings banks and cooperative banks that are strongly integrated in the regional social life and less influenced by globalization, identify more intensively with their institute than the branch managers of big banks.

Today, very different cultures collide in banks. This is not only because of regional differences and globalization but also because of functionally caused differences. We get an idea of the imminent cultural conflicts within the bank organisation if we compare e. g. conurbation to rural areas, traders to controllers, sales staff to back office staff, technology freaks to innovation sceptics.

The imperfect restructuring of the European banking industry that has already begun is far better than a continuation of the status quo. However, these measures could have been implemented more successfully if the banks had requested more flexibility from its staff earlier and if the management had been prepared to face innovative international changes. Instead, they defended positions that were not tenable such as the prohibition of money market funds and ways to prepare a balance sheet that are unfavourable to shareholders.

Corporate banking will become more significant, for big groups of companies, as money market players, will ask critically for the added value of bank services and in many cases will prefer corporate banking. International groups will further reduce the number of their bank relations and will increase their bargaining power over banks by setting up corporate banks as an alternative. Not only industrial producers but banks as well are facing the question of make or buy of financial services. This is especially true for smaller banks in investment banking but also in the field of the growing external payments which can partly be dealt with at more favourable prices by electronic banks.

Tighter price competition and the possibilities to reduce costs by means of new technologies strengthens the efforts to rationalise and lead to a mechanisation of major parts of the routine business. Expert systems and neural nets will dominate the credit rating that can be standardised. Image processing, Internet and flexible working hours lead to document handling partly done by home-workers and banking without office facilities. Together with an increased commission-based payment of bank employees, this results in the trade unions’ deprivation of power. In routine business, the lowest handling fees determine future customer orders which also results in a complete anonymity of more than seventy-five per cent of all bank services which, at the same time, causes problems with regard to customer loyalty. With the bank customers’ as well as the bank employees’ increased flexibility of working-hours, the importance of individual customer advice outside bank facilities and bank opening times is growing in the non-routine business. This results in an increased provision-based payment of front office employees that can hardly be granted to people working in the back office.

Bank stocks remain attractive because of rationalisation, concentration, staff reduction and a new performance orientation in the context of a shareholder value orientation. However, to the employees the bank’s attractiveness as an employer declines. Flexibility, mobility and capability of learning will become the employee’s most important assets in the competition for work.

Presently, the European direct banks are facing far higher obstacles to customer canvassing than initially expected. Therefore, they need more time with high expenditures on advertising to cross the break even point. If they finally get a return, the client will ask his branch for comparable online banking conditions. The electronic banking by direct banks will largely be integrated into their parent company and will only survive in some niches. However, meanwhile the customer will have become accustomed to pay market prices for advice and to pay for online transactions in accordance with activity-based criteria.

There is no structure of bank organisation that is the best possible ad infinitum, for organisation structures as well have to adjust to the requirements of the age. Therefore, it is pointless to discuss today whether the European banks’ strictly hierarchically-oriented line organisation of the past was a case of mismanagement. It was a sign of the time. Of importance is that with it the European banks earned well and achieved high growth rates. Today, they need new organisation and personnel structures.

International investing has been promoted by professional managers and academicians on the ground that low correlation among world equity markets allows for portfolio risk reduction. With the advances in communication technology and reduction of transaction cost, this paper hypothesizes that the benefits of diversification may be eroding due to increasing correlation among world markets. The question this paper addressed is whether the benefits to international investments still remains.

To determine whether the correlation among global stock markets has changed, we focus on the relationships among the following markets: the U.S stock market versus developed markets (Europe, Australia and Japan); the U.S stock market versus the emerging market (Asia, South American and Africa); the developed markets versus the emerging markets. This study spans twenty-three years from 1976 to 1997.

It has been suggested that during periods of high market volatility, the correlation among global stock markets appears to increase. If this is true, then the protection provided by diversification is reduced when it is most needed. This study therefore examines the relationships among world markets during periods of high volatility resulting from critical events in the global economy. These events could be political or economic. The intention is to learn the degree to which market reactions diverge or converge in response to these events. We hypothesize that the reaction among world markets could differ based on the nature of the events. The insights gained from this research should provide a better understanding of the benefits to international diversification.

Data Sources: Morgan Stanley Capital International (MSCI) data on emerging and developed markets for the past twenty-three years are used to test our hypothesis. MSCI indices cover over 1,400 securities in 23 emerging markets and 2,700 securities in 22 developed markets and include about 60% of all stocks trading in these markets. Similar criteria for selecting the indices are used for both emerging and developed markets, allowing for meaningful comparisons across all markets. Twenty countries, which best represent the characteristics of their groups, are employed to emulate the developed and emerging markets.

Methodology: In order to test the initial hypothesis of the changing correlation over time, the twenty-three year period are divided into five overlapping sub-periods to obtain moving averages. This process helps to isolate any significant changes in the trend of the correlation. Monthly data is used for this analysis. To test whether there are spillover effects to other markets from one market that is initially exposed to high volatility, the period surrounding critical events, both political and economic, are examined. The critical events are identified from the historic headlines in the yearly almanac. Some examples are the recent Asian market crash, the Mexico currency devaluation at the end of 1993, the Soviet Union's break-up of 1992 and the Gulf War of 1990.

To test the market sensitivity to critical events, weekly and daily data from MSCI are used. Thirty-six events, both political and economic, have been identified over the twenty-three year period. World market repercussions to these events are studied to determine whether markets diverged or converged in their reactions.

Currency exchange rates are closely tied to inter-country trade of goods and services as well as capital flows. The collapse of the Bretton Woods system or the fixed exchange rate system in 1973, along with the coinciding growth in global trade and greater mobility of capital have all contributed to an increase in exchange rate volatility. Concerns about exchange rate volatility have prompted central banks to actively intervene in foreign currency markets in an attempt to stabilize the exchange rate. An explicit assumption behind central bank intervention is that interventions can minimize deviations of the actual exchange rate from a target implied by fundamentals and/or that central banks can dampen the short-term volatility of exchange rates. Since the mid-1980s, governments of the major industrial countries have taken a very active role in the management of currencies. This policy shift has stimulated much debate about the feasibility and desirability of using intervention in foreign exchange markets to limit movements in exchange rates. Conflicting evidence has been amassed to support and oppose the efficacy of Federal Reserve (Fed) interventions in stabilizing exchange rates. For instance, Ballie and Humpage (1992) found that intervention between February 1987 and February 1990 was associated with an increase in dollar/Deutschemark ($/DM) and dollar/Japanese yen ($/Y) exchange rate volatility. On the other hand, Dominguez (1993) examined the impact of actual U.S. intervention for the period 1985 to 1991 and found that the Feds actions compressed the daily volatility associated with the $/DM and $/JY exchange rates. More recently, Bonser-Neal and Tanner (1996) provide empirical evidence suggesting that central bank intervention does not have any significant effect on volatility.

This study makes two important contributions to the existing literature on central bank intervention. First, previous studies examine the effect of intervention operations on volatility estimates derived from the spot currency market. This study, on the other hand, estimates conditional volatilities of $/DM and $/JY from foreign exchange futures prices. The strong dependence of futures prices on expectations suggests that the currency futures are a good candidate for a study of the interaction between volatility and new information on monetary policy contained in the Federal Reserve intervention operations. Second, the statistical approach in past studies raises a serious methodological issue. All of these studies make the a priori presumption that central bank actions influence exchange rate volatility, and none empirically test the direction of the relation between volatility and intervention. Under these circumstances, use of OLS estimation would yield biased and inconsistent estimates. Although there are strong theoretical reasons as to why intervention actions may influence volatility, there can also be some equally plausible reasons that central bank intervention is a response to the volatility conditions in the exchange rate market. For example, it is possible that central banks in an attempt to lessen the deleterious economic and financial consequences of increased volatility in the currency market step-up its intervention operations.

This paper addresses the shortcomings in previous studies by investigating the short-run dynamics between the Feds intervention actions and the conditional volatility in the $/DM and the $/Y futures prices for the period 1985 to 1993. The estimation technique employed, which is a definite departure from existing studies, places minimal restrictions on the explicit structure of the relationship between volatility and intervention activity. A two-step methodological procedure is undertaken. First, conditional variance is obtained from the GARCH (1,1) model for the return of the two currencies. In the second step, a time-series VAR model is employed to identify the dynamic relationship between the conditional volatility estimate (obtained from the GARCH (1,1) procedure) and the Fed=s intervention actions.

Results from the study indicate no relationship between Fed=s intervention activity against the DM, and the $/DM conditional volatility during the 1985 to 1993 period. However, intervention is associated with negative changes in the $/Y volatility during the 1985 to 1993 period as a whole, and specifically during the January 1, 1985 to February 21, 1987 Plaza period and the February 21, 1987 to December 31, 1989 Louvre period. This supports the signaling hypothesis which posits that interventions resolves uncertainty among market participant by providing information about future monetary policy. Furthermore, the results suggest a strong positive feedback effect (bi-directional causality) flowing from volatility to intervention. During the post-Louvre period (January 1, 1990 to December 31, 1993), we find Feds intervention to exacerbate the volatility in the $/Y, without a corresponding feedback relationship. The sign reversal may be attributed to the breakdown of the Louvre Accord, which brought about lower cooperation and coordination among central banks to achieve exchange rate stabilization.

Given the general lack of consensus among economists about the purpose and effect of central bank interventions, the issues surrounding this debate seem to be a promising area for further research in the futures arena. For instance, it will be interesting to (i) test whether the character of the spot/futures price relationship differs on intervention days from that on nonintervention days, and (ii) establish a link between the basis (futures and spot spread) and speculative trading behavior on the one hand, and Fed intervention actions on the other. These issues are left for further examination.

The emerging high growth economies of Asia such as Hong Kong, Taiwan, South Korea, Singapore, Malaysia, Thailand. Indonesia and the Philippines have attracted the international investors for portfolio diversification over the last several years. However, the current Asian foreign exchange crisis reflects the fickleness of International investors, who are eager to invest their money in Asian countries when the returns are high and quick to withdraw it at early signs of trouble. And it reflects a bow to the inevitable: economic growth at near-double-digit rates had to slow someday. The ongoing volatility in the global financial and the capital markets have originated from Thailand in July of 1997. Six months into the Asian financial crisis, there is still plenty of uncertainty about how big a mess it will ultimately generate. Asia, with 30% of the world economy, has weaker banks, weakening currencies and lower inflows of capital. In the month ahead, this trend of weakness in Asia and strength elsewhere could set the course of economic events. This weakness is partly the result of a drop in foreign investment. Also important is a drop in the flow of the countries huge domestic savings, often channeled through banks into major projects. Thus the Asian foreign exchange crisis along with its linkage highlights the crucial role that finance plays in capitalist.

While the linkages and intertemporal relationships among the developed financial and the capital markets have been examined extensively, no attention has been paid towards the linkage between the foreign exchange markets of the South and the Southeast Asian countries, especially with respect to current turmoil in these economies. The objective of this paper is to provide new evidence on the question of interdependence among the above eight countries of Asia and the Mexico on the basis of daily exchange rate series. Mexico has been included in our study because peso value has also been volatile for the last few years. We investigate whether there are co-movements and bi-directional linkages among these foreign exchange markets since the beginning of the present foreign exchange crisis in Asia. If such interdependence is detected between two markets, it will also be important to assess the respective role of the countries in these driving forces, and more generally try to investigate the dynamic interrelationships that might exist between the different foreign exchange markets. We are also concerned with the trending behavior of the nine markets, namely those of Hong Kong, Taiwan, the Philippines, Singapore, Malaysia, Thailand, South Korea , Indonesia, and Mexico. We try to explore the degree to which their respective market exchange rates exhibit common long-term stochastic trends and the degree to which these common trends are directed by one of these market rates.

Since cointegration implies that nonstationary times series such as exchange rates move stochastically together toward some long-run stable relationship, the existence of cointegrating relationships among various markets has a direct implication in terms of the existence of common trends among these rates. In other words, if the random walk component behavior is a realistic hypothesis for the rates of these various countries, do these components differ across the national markets, or are the various markets sharing random walk components? On the other hand, cointegration also implies the existence of a Granger causality property between the rates. This framework is attractive if the purpose is to study both co-movements and dynamic relationships.

Thus our study focuses on the eight foreign exchange markets of Asian countries and Mexico which got more international attention recently because of the ongoing currency crisis. The currencies are Hong Kong dollar, Singapore dollar, Malaysian ringgit, Taiwan dollar, Thailand baht, Indonesian rupia, South Korean won, The Phillippines peso, and the Mexican peso. The period of investigation covers from July 2 through December 2, 1997. All exchange rates are in per U.S. dollar and are obtained from the Wall Street Journal. Rates are quoted on the New York foreign exchange selling rates and are based on trading among banks in amounts of one million U.S dollar. The series were transformed into natural logs and are graphed in Figures.

Empirical results are comprised of Augmented Dicky-Fuller as well as the Phillips-Perron tests of stationarity, Johansen multi variate tests of cointegration, and the Granger causality tests. The level series are supportive to unit root hypothesis for all currencies because both the ADF and PP tests could not reject the null hypothesis of unit root. However, first differenced series are found to be stationary for all currencies except the South Korean won and the Taiwan dollar which converge to zero unit root after second difference. We then tested for cointegration by applying the Johansen multi variate test to these rates. We began the analysis with various stochastic trends in data by performing VARs of lags interval 1 to 4. The tests results show that the series are cointegrated up to 4 lags because the null hypothesis of no cointegration has been rejected at the conventional level of significance. This is the case with different assumptions about linear and quadratic trend or no trend in the data. We also tested the null hypothesis no ‘Granger Causality’ at the 1 percent and the 5 percent level of significance unto several lag intervals. Statistical results show that the null hypothesis of one market does not ‘Granger Cause’ other market has been rejected at the above levels of significance with 7 or 14 days lag.

Based on our findings, we found that there exists a rather highly integrated Pacific Rim foreign exchange markets. We also observe that there exists Granger causality between the Mexican peso and some of the Pacific Rim currencies such as the Hong Kong dollar, Taiwan dollar, Malaysian ringgit, Thai baht, and the South Korean won. It is not unlikely given the growing volume of their trading and investment relations and because of the governments efforts to liberalize their economies with reforms of financial and the capital markets. Based on our present analysis it reveals that Hong Kong dollar, Singapore dollar and the Malaysian ringgit are the dominant currencies in the region in terms of lead-lag relationships. In order to provide further information concerning the on going foreign exchange crisis in these countries, this research is extending the sample period and the final report will be available at the conference. It is expected that the findings out of this research will be helpful to international fund managers as well as individual investors willing to reduce their portfolio risks and enhance the expected returns.

One of the greatest obstacles in the development of emerging capital markets is the resolution of information uncertainty. Such markets have difficulty attracting debt and equity capital due to the fact that reliable information is not available to foreign investors. This problem is exacerbated by the fact that many of these developing countries have experienced periodic difficulties in their debt payments over the last two decades.

In the early 1990s many developing countries began the process of opening their financial markets to foreign investment. To mitigate the information uncertainty problem, these countries turned to Moody’s and Standard and Poor’s (S&P’s) rating agencies, which are considered credible external certifiers of government and private credit risk information (i.e., in the past these agencies only provided credit ratings in major industrial countries). In response to rating demand, Moody’s and S&P’s began publishing international ratings of firms’ credit quality in emerging market countries. Most of these countries already had domestic rating agencies; however, unlike domestic ratings, international bond ratings are based on comparisons to similar firms in different countries. Because they offer investors interested in international diversification valuable information that was not previously available, issuers gain access to funding sources worldwide and possibly more attractive financing terms.

To the authors’ knowledge no other research has been published on this relatively new international credit development. The present study seeks to fill this gap in the literature by focusing on the wealth and risk effects of these new international ratings on firms in Mexico. Mexico is interesting because it is a prime example of a developing country with previous debt problems that recently has made dramatic steps in opening its economy to international trade, privatizing business firms, and implementing market reforms to promote free enterprise. Additionally, the U.S. has historically been a major investor in Mexico and recently has increased its economic ties to Mexico. In 1989 Mexico rescheduled its debt to U.S. banks under the Brady Plan, and in an effort to attract foreign investment, it began negotiating a trade agreement with the U.S. in 1990. These negotiations culminated in the 1994 North American Free Trade Agreement (NAFTA) between Mexico, the United States, and Canada. NAFTA has greatly increased international scrutiny on financial market, as well as economic and political, events in Mexico. It has also motivated Moody’s and S&P’s to issue the largest number of international ratings in Mexico among the developing countries in Latin America (e.g., in 1995 there were 65 published ratings by S&P’s in Latin America, with 25 of these ratings going to Mexican entities and the remainder spread out among Argentina, Brazil, Chile, Columbia, Panama, Uruguay, Venezuela).

We hypothesize two potential information effects of the international ratings on private firms and government-owned entities in Mexico. First, due to the resolution of information uncertainty, demand on the part of international investors in these emerging market securities will increase and tend to tend to positively impact the prices of rated bonds, as well as the stock prices of rated firms. We will refer to this potential positive impact as the wealth effect hypothesis. Second, consistent with Ross (1989), an increase in the flow of information in an efficient market will tend to cause security price movements to become more frequent and therefore more volatile. This price risk hypothesis is particularly relevant to emerging markets in which information flows can be scarce and consequently infrequent trading occurs. In the present context security prices are expected to become more volatile after the introduction of Moody’s and S&P’s ratings in Mexico if information flows are enhanced by the ratings.

By implication, since firms in Mexico and other emerging market countries benefit from international ratings in terms of higher bond and stock prices, the new ratings not only attract foreign investment interest but also tend to lower costs of capital in developing countries.

The currency swap market has grown rapidly since its inception with the currency swap between IBM and the World Bank in 1981. Earlier literature has concluded that the rationale for the currency swap consisted in arbitraging the comparative advantages that different borrowers enjoyed in different capital markets. The basis for this conclusion was the segmentation of capital markets due to institutional constraints such as capital controls or unequal access to information across capital markets. During the last decade, however, such institutional and informational barriers across capital markets restricting capital flows have weakened with increasing integration of capital markets.

In this paper, we show that firms have reason to use currency swaps even when institutional and informational barriers vanish. When the cashflows are generated in foreign currency by a foreign project, firms could use foreign-currency debt. However, we show that if the firm's cashflows are sensitive to exchange rate changes, the financing alternative of combining the use of currency swaps with home currency debt to synthesize foreign currency debt is preferred by the firm. The sensitivity of the firm's cashflows to exchange rate changes is called economic exposure. The level of economic exposure of a firm is unobservable to the outside investors but known to the managers of the firm. We show that such asymmetry of information about economic exposure results in the dominance of currency swaps over foreign currency debt.

Suppose a U.S. company wanted to issue debt denominated in Japanese yen. It could either issue yen denominated debt directly or issue a dollar denominated debt and use a currency swap to exchange the dollar denominated debt for yen denominated debt. The role of the currency swaps can then be simply described as synthesizing foreign currency debt. To simplify the analysis, we assume that the firm only receives foreign-currency denominated cashflows. When the foreign-currency debt is synthesized using home currency debt and currency swaps, the default risk is established based on expected foreign currency cashflows which are translated in home currency terms. When currency swaps are used together with the home-currency debt, the home currency riskless interest rate is effectively replaced by the foreign currency riskless interest rate. Therefore, pricing the home currency debt will involve accounting for both default risk and exchange rate risk. Foreign-currency debt will be priced based only on default risk in foreign currency terms and the foreign currency riskless interest rate. The firms compare the total cashflows net of cost of debt in home currency terms to choose between the foreign currency debt and synthesized foreign debt.

In the presence of currency risk alone currency swaps do not dominate foreign currency debt. If there is currency risk but the cashflows of the firm are not sensitive to exchange rate changes (i.e., there is no economic exposure), the default risk assessed for both foreign currency debt and synthesized foreign debt is identical. Thus, both debts are identically priced. When firms face economic exposure, the negative correlation between the cashflows and exchange rates gives rise to a difference between the pricing of foreign-currency debt and the synthesized foreign debt. We show that this difference renders the use of currency swaps most appropriate for the firm facing economic exposure.

Conventional explanations for the use of currency swaps rely on international capital market segmentation which renders foreign currency denominated debt more expensive. Artificial capital market barriers lead to differential transaction costs in domestic and foreign currency debt markets. This transaction cost differential could be the result of a front-end cost differential between domestic and foreign bonds, or a search-cost differential among these bonds. For example, one argument for the existence of currency swaps is based on international capital market segmentation due to international investment restrictions. In contrast we argue that even if international capital markets get integrated (so that transaction costs are eliminated), product market imperfections cause the cashflows of the firm to be sensitive to exchange rate changes. We show that the use of currency swaps maximizes the value of such firms for the shareholders. We also demonstrate that in the absence of transaction costs, if firms do not face economic exposure, foreign-currency debt, home-currency debt, and synthesized foreign debt are equivalent financing alternatives.

We do so by discussing the conventional transaction cost based explanation as a special case of our economic exposure based analysis. We show that when there are transaction costs and when currency risk does not result in any economic exposure for the firm, currency swaps do not dominate foreign currency debt as well as home currency debt. They are however useful in providing covered interest arbitrage when there are positive deviations from UIRP. We conclude that in an integrated and frictionless capital market, synthesized foreign debt is equivalent to foreign currency debt and home currency debt and, therefore, currency swaps are redundant securities.

There are several empirically testable predictions of our analysis. Firstly, during periods of positive UIRP deviations, we should observe a greater use of currency swaps by firms in the home country. Fore example, positive UIRP deviations in the data on the US $ as the home currency and the JPY as the foreign currency and the respective riskfree interest rates would imply that U.S. firms use currency swaps involving the US$ and JPY over other financing alternatives. Secondly, comparing firms across different industries, we should find that firms sampled from industries facing imperfect competition should be using currency swaps more than the firms sampled from industries which have perfect competition.

The currency swap alternative dominates home currency borrowing only when the UIRP deviation is positive. Fore example, when the foreign country maintains the risk free interest rate at a level lower than the home country risk free interest rate or intervenes in the exchange rate market to depreciate the value of the foreign currency, the UIRP deviations are positive. Since the levels of economic exposure are not observable, firms with either type of projects will use currency swaps in a mildly segmented capital market where there are such positive UIRP deviations. Even under asymmetric information, foreign currency debt is dominated by currency swaps. We conclude that integration of capital markets will not necessarily result in more foreign currency denominated borrowing but can result in greater use of the currency swaps because of imperfect product markets and economic exposure to foreign exchange risk.

The expansion of world trade, the increasing integration of the world's money and capital markets, and improved telecommunications have dramatically changed how firms do business in the post-Bretton Woods era of floating exchange rates. To be successful in this new business environment, both domestic firms engaged in foreign trade and multinational enterprises must clearly understand certain key relationships between interest rates, exchange rates, and the forward market for foreign exchange. Failure to understand these key relationships could, ultimately, lead to the failure of the firm itself. At a minimum, not understanding the relationships between these critical variables will damage a firm's bottom line and/or subject the firm to unnecessary risks. The current instability in the Asian foreign exchange and equity markets serves to confirm the inherent challenges of doing business overseas.

This study employed daily data obtained from 1) The Wall Street Journal (i.e., spot as well as forward exchange rates based on the "best available prices" and 2) The Economist (i.e., money market and Eurocurrency interest rates) to evaluate certain key theoretical relationships and to expand the body of applied knowledge available to firms actively involved in international business. The data include observations for thirty consecutive, workable trading days for a recent period in 1997. Data for the Canadian dollar, the French franc, and the Swiss franc were evaluated. More specifically, this study performed the following tests:

1. The International Fisher Effect. Assuming that nominal interest rates include components reflecting the "real" rate of interest as well as expected inflation (commonly called the "Fisher effect" in the domestic market), the n-period interest rate differential between two countries should accurately predict the n-period appreciation/depreciation of the exchange rate between the two countries' currencies. This would be necessary to maintain purchasing power parity (i.e., the two currencies have the same value in terms of a common numeraire) as well as identical effective rates of return across countries.

To test the validity of the theory, actual percentage appreciations/depreciations (for 30-days, 90-days and 180-days) across several major currencies were computed and then compared, ex post, to the (de-annualized) interest rate differentials existing 30, 90 and 180 days prior, respectively. In effect, "expected" interest rate differentials based on currency appreciations/depreciations were compared, after the fact, to actual interest rate differentials. Because of the inevitable ambiguity as to which interest rate to use as the predictor of currency movements, this study tested differentials in both (n-period) money market and eurocurrency rates.

2. Covered Interest Arbitrage. In an efficient market, observed premiums/discounts in the forward market for foreign exchange should accurately reflect existing (n-period) interest rate differentials. If this were not true, arbitrageurs would find risk-free profit opportunities.

To test the validity of this hypothesis, 30, 90, and 180-day premiums/discounts were compared to contemporaneous, de-annualized interest rate differentials (i.e., 30, 90, and 180-day rates). In effect, "expected" interest rate differentials based on forward premiums/discounts were compared, contemporaneously, to existing interest rate differentials. This second test of interest rate parity employed data for the same time period and currencies as the first test (#1, described above). Again, because of the ambiguity surrounding the interest rate, differentials in both (n-period) money market and eurocurrency rates were evaluated.

The results reveal that neither the International Fisher Effect nor Covered Interest Arbitrage hold. While some portion of the observed discrepancies may be explained by a number of practical considerations (such as transactions costs and taxes), it appears that these factors are insufficient to fully explain the observed violations of the International Fisher Effect and Covered Interest Arbitrage theories. Hence, the data do not support these fundamental tenets of international finance. (Efforts to identify and quantify the practical considerations which can explain small deviations from the International Fisher Effect and Covered Interest Arbitrage present interesting and potentially productive topics for future research.)

3. Efficiency of the Forward Market. This study also tested if the n-period forward rates for the various currencies were accurate (unbiased) predictors of future spot rates. Data for the same thirty consecutive, workable trading days (as tests # 1 and 2, above) were employed. The results reveal an interesting point in reference to forward rates as unbiased predictors of future spot rates. While 30-day forward rates underestimated future spot rates for all three currencies, with the exception of the Canadian dollar, which was mixed, the 90-day and 180-day forward rates overestimated the future spot rates for all three currencies.

Finally, this study examined the short-term financing implications for multinationals of cases of consistent overestimation or underestimation of future spot rates by forward rates. The results indicate that multinationals, on average, can benefit in the case of overestimation (in terms of financing costs) by leaving their foreign currency position open. However, in the case of underestimation (in terms of financing costs) their foreign currency position should be covered.

The stock prices of multinational corporations (MNCs) are determined by cash flows from foreign countries, which, in turn, are greatly affected by exchange rates. Cornell (1983) and Wolff (1988) provide empirical results for a direct relationship between exchange rate and economic activity. Others find a linkage between stock price and economic activity (Chen, Roll, and Ross (1986), Mandelker and Tandon (1985), Fama (1981)). These findings support an implicit link between stock prices and exchange rates. Solnik (1987) finds a negative relation between real domestic stock returns and real exchange rate changes. Ma and Kao (1990) note that domestic currency appreciation negatively affects the domestic stock price movement for an export-dominant economy and positively affects an import-dominant economy. Jorion (1990) reports a moderate relation between the rate of return on U.S. multinational firms’ stocks and the rate of change in a trade-weighted value of the U.S. dollar over 1971-87. The relationship between exchange rates and stock prices have been frequently discussed in the financial press (e.g., Wall Street Journal (1993, 1995), Bollen (1995)).

¦ y_t = „ + - y_t-1 + 3 ( ¦ y_t-j +§ _t (1)

j=1

where y is the series being tested, and k is the number of lagged differences to capture any autocorrelation. K is chosen so that the Ljung-Box Q-statistic fails to reject the null hypothesis of no serial correlation in the residuals (§ ). The test is pseudo t-statistic for the null hypothesis that - = 0. The results show that the exchange rate and stock index series share similar temporal properties (i.e., nonstationary in the levels having one unit root), and thus two price series are tested for cointegration. If two price series are cointegrated, the OLS regression may be biased toward higher R² values, lower Durbin-Watson statistics, and lower standard errors. Then the t- and F-statistics become invalid. Cointegration test is to examine stable long-run relations between two variables. The market mechanism generating the two series may exhibit short-run drift, but in the long run the two series will approach one other. The market forces guiding these movements will correct for any discrepancy that may exist and attract them back together to ensure a long-run stationary equilibrium.

If two series are cointegrated, the intertemporal relations between the two variables are examined within the framework of an Error-Correction Model (ECM). The ECM captures both the short-run dynamics between time series and their long-run equilibrium relationship, providing a suitable framework to examine the Granger-causality. The ECM also captures the partial adjustments one variable makes to a shock caused by another variable. The causal relations between the time series suggest the short-run dynamic adjustments needed by the levels of the variables to reach positions of long-run equilibrium. Two cointegrated variables may have the joint error correction representation.

The final prediction error (FPE) criterion (Akaike (1969)) is used to determine the optimal values for the Ns. Short-run dynamics between the two price series are examined by the coefficients. If one (or more) of the, coefficients is statistically significant, movements in exchange rates will have a short-run effect on the stock market. If one of the coefficients is statistically significant, the stock market will have a short-run effect on exchange rates. But the existence of a long-run relationship between the stock market indices and exchange rates hinges crucially on the significance of and. Since S_t and F_t are cointegrated, the Z term (deviation from the long-run relation) must include both variables. Then either or will be nonzero and statistically significant. After the appropriate lag structure is identified, we estimate the system for each of the six countries using a seemingly unrelated regression approach to gain efficiency. Long-run dynamics are captured by the parameters and.

While the propensity among firms in recent years has been towards an expanded utilization of discounted cash flow methods for evaluating investment proposals, evidently practitioners still feel comfortable using less sophisticated procedures, particularly payback. Surprisingly, according to the surveys payback is being used by over 50 percent of the largest firms in the United States as either the primary or secondary method (less than 10 percent as the primary method). In the Oblak-Helm multinational survey, payback was used by 72 percent of the companies for primary and ancillary investment decision making.

The global popularity of payback undoubtedly stems from of its substantial intuitive appeal (i.e., it tells how long it takes to pay for an investment). This paper proposes a generalized time-adjusted cash flow payback approach which is theoretically consistent with the assumptions of the net present value capital budgeting technique, yet phrased in terms that practitioners may find more understandable.

The generalized time-adjusted payback decision rule develops as follows:

n