# Currency swaps, financial arbitrage, and default risk.

17 Another alternative to swap financing may be issuing foreign securities. The distinguishing aspect is that swaps are governed by contract law instead of securities law. This difference in legal treatment can have significant default implications.18 See appendix for proof.

19 Incomplete markets are pointed out as an explanation for the existence and evolution of the swaps market in Smith, Smithson, and Wakeman (1986) and in Arak, Estrella, Goodman, and Silver (1988) even though no analysis of the issue in that context is provided. Note that market incompleteness other than market segmentation may cause the stated condition in Result 1 not to hold. Many recent studies have addressed the analysis of interest rate swaps. In these studies, there is an ongoing debate concerning the importance of financial arbitrage in explaining the existence of the interest rate swap market. Some that favor financial arbitrage as a motive argue that the quality spread differential between two markets presents a financial arbitrage opportunity that can be shared to the benefit of both parties. (See Beidleman (1985, 1992), Bicksler and Chen (1986), and Felgran (1987).) As a result of comparative advantage, their borrowing costs are unambiguously reduced. Implied (but not stated) in these arguments is that both parties increase their value through swapping. Other studies argue against this view by noting that the very process of exploiting this kind of opportunity should soon eliminate it. (See Smith, Smithson, and Wakeman (1986, 1988), Turnbull (1987), and Wall and Pringle (1989).) The second argument against financial arbitrage is that the swap should be a zero-sum outcome even if financial arbitrage is the source of their economic benefits. These studies rightfully dismiss the importance of financial arbitrage in an efficient, complete, perfect, and integrated world capital market.

Currency swaps, however, have been largely ignored in the swap literature. This study attempts to fill that gap. It focuses on currency swaps exclusively in the belief that substantial differences exist between currency swaps and interest rate swaps that warrant a separate analysis.(1)

With currency swaps, one has to frame the analysis within an international capital markets setting. Even if the generally accepted view is that domestic capital markets are integrated (efficient), there is growing empirical and theoretical support that international capital markets are segmented (inefficient). Segmentation of international capital markets may be caused by government restrictions on international portfolio holdings of individual investors. In most countries, there are limitations on individual investor's access to foreign securities. Thus, it is plausible to assume segmented capital markets as an environment in which firms conduct their international financial and real activities.

In such an environment, however, firms can design strategies to reap the potential profits created by restrictions on portfolio holdings of individuals because corporate projects are not subject to the same type of restrictions.(2) Firms have an advantage over individuals because they have income generated in both markets (real activities) and can design securities (claims against their income) that will have differential valuation in the two markets. One such transaction is a currency swap, which firms can use as a financing alternative.

This paper analyzes how firms with foreign projects can exploit international capital market segmentation via currency swaps. I show how capital market segmentation may lead to non-zero outcomes and why non-zero outcomes do not necessarily yield positive gains for both parties as put forward in the comparative advantage arguments based on rate differentials.(3) In default-free currency swaps, heterogeneous expectations are sufficient to achieve this result.

Another concern in the swap finance literature is a true understanding of the default implications of this contract. Both the counterparties and the intermediary are interested in knowing how to assess the credit risks inherent in swaps. To this end, Cooper and Mello (1991) attempt to price the default risk premiums in swaps. They assume a perfect and integrated capital market in which swaps are zero-sum games. They obtain somewhat counterintuitive results, such as wealth transfers from stockholders to bondholders. Within their framework, the analysis of swap default also falls short of depicting the interactions of default risk with capital market imperfections, the main finding of this study. In this study, I find that in a segmented international capital market, firms can enhance shareholder wealth with risky currency swaps beyond what they could achieve through a default-free swap.(4) This result implies that credit risk is not an increasing function of the probability of default, as would be the case in a perfect and integrated domestic or world market.(5) In currency swap credit risk assessments, a second factor, the covariability of exchange rates and default states, plays a key role.(6) The second factor may offset the effect of the risk of default at certain times, and the swap partners may enhance value by defaulting. Thus, the interaction of foreign exchange risk and default risk is crucial to credit risk assessments for a currency swap. This finding is supported by an options analysis of default in currency swaps. I show that the default option can best be viewed as a "hybrid" currency call option that is firm-specific and can have positive or negative values. My analysis suggests that the counterparties can design the risky currency swap so that this option has positive value to both parties to the swap.

The analysis of this paper follows Thomadakis and Usmen (1991), which focused on international capital structure decisions of firms operating in segmented markets. Likewise, the present paper analyzes the benefits that firms can reap through currency swaps in a similar setting. The key in both analyses is that firms are able to overcome barriers to arbitrage because they have specific projects (projects with a particular pattern of cash flows across states, thus with a particular default structure) enabling them to exploit specific arbitrage opportunities across the same states. Thus, only those firms with these specific projects could benefit from the state-dependent arbitrage profit opportunities.

I. Elements of the Model and Financial Arbitrage

Assume a simple two-country world--the domestic country and the foreign country. The capital market is complete, efficient, and perfect (no transactions costs and no taxes) in each country. Further assume that investors are risk neutral. Risk neutrality is assumed to simplify the exposition and to derive results that are expressed in terms of familiar expectation forms. With risk aversion, similar results could be obtained. There are constraints on international asset holdings, in the form of capital controls, and restrictions on short selling by investors of both countries. This assumption is realistic and is crucial for the results. It allows the capital markets of the two countries to be segmented, which gives rise to the potential for financial arbitrage to exist across the markets. International market segmentation due to international investment restrictions is, by now, well-documented. (See, for example, Bonser-Neal, Brauer, Neal, and Wheatley (1990), Errunza and Losq (1985), Eun and Janakiramanan (1986), Gultekin, Gultekin, and Penati (1989), and Jorion and Schwartz (1986).) Another crucial assumption is that domestic and foreign investors hold heterogeneous beliefs about the probabilities of future states of nature. Under risk neutrality, this assumption is necessary to have the independent gains for both parties to be positive. With risk aversion and homogeneous beliefs, different risk adjustment factors would have served the same purpose.(7)

Within the above framework, the following notation is defined, where S is a set of states, s is an element of S, and B is any subset of S:

p(s), p*(s) = state probabilities assessed by domestic and foreign investors, respectively. The probability of an event B is P(B) = [summation over B] p(s) and P*(B) = [summation over B] p*(s) in the domestic and foreign market, respectively.

r, r* = one plus the risk-free rate in the domestic and foreign markets, respectively.

X(s), X*(s) = state-contingent cash flows of assets denominated in domestic and foreign currency at t = [Tau], respectively.

[e.sub.0] = spot currency exchange rate at t = 0, expressed in units of domestic currency per unit of foreign currency.

e(s) = state-contingent spot currency exchange rate at t = [Tau], expressed in units of domestic currency per unit of foreign currency.

The no-arbitrage condition for a foreign asset with an income of X*(s) and its domestic perfect substitute yielding X*(s)e(s) is

[summation of] X*(s) ([e.sub.0]/r* p*(s) - e(s)/r p(s)) = 0. (1)

In other words, whenever a domestic investor sells short the foreign asset, converts the proceeds into the domestic currency, and purchases the domestic perfect substitute, there should be zero profit. A condition sufficient for Equation (1) to hold is(8)

([e.sub.0]/r* p*(s) - e(s)/r p(s)) = 0 for all s. (2)

Note that Equation (2) expresses a state-contingent, no-arbitrage relationship between the prices of primitive securities that are denominated in different currencies. Alternatively, a positive value for the expression on the left-hand side of Equation (2) would imply potential arbitrage profits in the short sale of a pure claim denominated in the foreign currency coupled with the simultaneous purchase of the perfect substitute in the domestic country. Likewise, a negative value would indicate arbitrage potential in the simultaneous purchase of a foreign pure claim and sale of its perfect substitute in the domestic market. In the absence of capital controls and short selling restrictions, the above relationships, seen in Equation (2), should strictly hold for every state. In such an integrated international market where all domestic and foreign assets are traded by both sets of investors, no arbitrage will ensure that Equation (1) always holds. However, once one introduces restrictions on trading of certain assets and impediments to short selling, there will be assets for which Equation (1) fails to hold. A necessary and sufficient condition for the expression on the left-hand side of Equation (1) to be different from zero for some assets is

([e.sub.0]/r* p*(s) - e(s)/r p(s)) [is not equal to] 0 for some s. (3)

Condition (3) is my description of segmented international capital markets.(9)

A weaker condition that might hold even in a segmented international capital market is an integral condition known as Uncovered Interest Rate Parity (UIRP). Looking at Equation (1), the no-arbitrage condition for a riskless foreign asset is given by

D*[Sigma]([e.sub.0]/r* p*(s) - e(s)/r p(s)) = 0, (4)

where D* is the certain payoff on the foreign riskless asset. If the left-hand side of Equation (4) has a nonzero outcome, it would indicate arbitrage potential in trading the foreign riskless asset for the domestic investor. Equation (4) can be rewritten in the familiar form under risk neutrality as

[e.sub.0]/r* = Ee(s)/r, (5) (5)

or as

r/r* = Ee(s)/[e.sub.0], (5[prime])

where E is the expectations operator in the domestic market. Equation (5[prime]) states that under risk neutrality, the price ratio of two riskless assets predicts the rate of change in exchange rates when opportunities for arbitrage profit cease to exist.

Arbitrage can take place for riskless assets in the foreign country, such as government bonds and government-insured bank deposits, with the result that Equation (5) holds, but this condition does not imply the absence of arbitrage opportunities for all risky foreign assets. There is partial segmentation. To see this, consider Equation (1) again. It can be rewritten as

[e.sub.0]/r* E*X*(s) = Ee(s)/r EX*(s) + cov(X*(s),e(s))/r, (6)

where E* is the expectations operator in the foreign market. It should be clear from Equation (6) that UIRP would imply no arbitrage only for those risky foreign assets for which X*(s) and e(s) are independent and only if expectations are homogenous, E* = E.

In the above discussions, uncovered interest rate arbitrage was presented from the point of view of the domestic investors. The arbitrage opportunity concerned the risk-free foreign asset and its perfect substitute, a risky asset in the domestic market. Similarly, the no-arbitrage condition for the domestic riskless asset paying D and its risky perfect substitute in the foreign market, denominated in foreign currency, is

D[Sigma](p(s)/[e.sub.0]r - p*(s)/e(s)r*) = 0, (7)

which is equivalent to

[e.sub.0]/r* = 1/rE*[e.sup.-1](s). (8)

Equations (4) and (7) are trivially equal to each other and to zero in a complete and integrated market in which Equation (2) holds. The same statement holds true for Equations (5) and (8). One would also expect them to be simultaneously equal to zero in a segmented market whenever capital controls and short selling restrictions apply only to risky assets. However, under risk neutrality, UIRP will not hold in both markets unless Ee(s) = 1/E*[e.sup.-1](s). Clearly, this condition is possible only in a framework of heterogeneous expectations due to Jensen's Inequality.

In general, Equations (4) and (7) do not imply each other in segmented capital markets. To show this, reexpress the terms in Equation (7) into the domestic currency by multiplying through by [e.sub.0] to rewrite it as

-D[Sigma] 1/e(s) ([e.sub.0]/r* p*(s) - e(s)/r p(s)) = 0. (9)

In Equation (4), there are potential profits as domestic investors arbitrage foreign riskless assets and their risky domestic perfect substitutes. Equation (9) demonstrates the potential for arbitrage profits, denominated in domestic currency, when foreign investors trade in domestic riskless assets and their risky foreign perfect substitutes. It is evident from Equation (9) that if UIRP holds in the domestic market (and Equation (4) is valid), it will not necessarily hold in the foreign market. To see the significance of this result, suppose one set of investors, say foreign investors, are restricted from trading the domestic risk-free asset. Domestic investors, however, can trade the foreign risk-free asset as well as their own. Trading by the unrestricted domestic investors will ensure that Equation (4) holds, but this does not imply the validity of Equation (9). Since the foreign investors are restricted, arbitrage opportunities will still be possible in trading the domestic risk-free asset in the foreign market. Obviously, if both sets of investors are restricted from trading the risk-free asset of the other country and/or there are restrictions on short selling, neither Equation (4) nor Equation (9) may hold. It is also interesting to note here that any positive deviation from equality in Equation (4) may not be offset by a negative deviation from equality in Equation (9). The presence of 1/e(s) in Equation (9) will make the magnitudes different but not necessarily the signs. The arbitrage profits of domestic and foreign investors do not wash away. This is so because the riskless assets in the two markets are not perfect substitutes. It might also be possible that both deviations have the same sign. This would simply imply that arbitrage profits exist for the sale or purchase of both domestic and foreign risk-free assets.

I next consider the possibility of the forward sale and purchase of both currencies for the time period [0,[Tau]]. Note that it is only possible to contract a fixed amount of future sale of a currency; there are no state-contingent forward contracts. Thus, I can carry the arbitrage argument through the risk-free assets. Note that with a forward currency market, the foreign asset and its domestic perfect substitute are both riskless. Covered interest rate arbitrage operations in domestic and foreign markets yield

[e.sub.0]/r* = f/r (10)

and

1/[e.sub.0] r = 1/fr* (11)

respectively, where f is the forward rate for exchange of foreign currency into domestic currency at t = [Tau]. Assume that both sets of investors have equal market access and sufficient funds. When all arbitrage opportunities cease to exist, f equals [e.sub.0] r/r*, the interest rate parity forward rate. However, with constraints on trading and short selling, this relationship may not hold. This might not be a far-fetched observation, if one recalls the empirically verified deviations from covered interest rate parity (CIRP). Frenkel and Levich (1975, 1977) attributed these deviations to the existence of transactions costs. However, more recent studies (see Bahmani-Oskooee and Das (1985) and Deardorff (1979)) conclude that transactions costs cannot by themselves explain these deviations. They attribute it to the existence of factors other than transactions costs. One of these factors was suggested by Keynes (1923) and later Einzig (1937): The institutional constraints that limit a trader's position taking, whether for arbitrage or for other purposes.

Finally, I note that fluctuating exchange rates since the abolition of the Bretton Woods Agreement on fixed exchange rates, and increasingly volatile US interest rates, since the 1979 U.S. Federal Reserve Board policy change, coupled with restrictions on trading and short selling, may have increased the likelihood that Condition (3) for the existence and persistence of misalignments of exchange rates and interest rates holds. Casual observation of the past evolution of the swap market reveals that currency swaps emerged in response to fluctuating exchange rates, while the volume of interest rates swaps surged with increased interest rate volatility. Furthermore, a main theme in the aforementioned studies on capital market segmentation is that government-imposed regulations on investor access to investments, such as restrictions on foreign investments by domestic investors and restrictions on domestic security holdings by foreigners, has resulted in an international capital market that is segmented. As a result, potential arbitrage opportunities that cannot be eliminated by restricted investors persist. Casual empiricism suggests that in most countries where there is swap activity, these restrictions may be binding for individual portfolio transactions but not for the swap transactions of firms.

II. The Currency Swap Contract

The analysis of the currency swap contract will deal with a pair of firms. To ease the presentation without sacrificing generality, I let one of these firms be a domestic firm with domestic shareholders and a foreign subsidiary. The foreign subsidiary will generate cash flows next period that are denominated in a foreign currency, X*(s), but it needs to be financed in the foreign currency today. Similarly, the other firm is a foreign firm with foreign stockholders and a domestic subsidiary promising X(s) and is in need of domestic currency. I must emphasize that these two firms are chosen to be symmetrical in terms of their cash flow generation opportunities in order to ease the analysis of the swap contract. Symmetrical, here, means that both parties have cash flows in the undesired (foreign) currency, which they would like to convert to the desired (domestic) currency. For example, one of the firms can be a U.K. multinational parent who has a U.S. subsidiary and the other firm can be a U.S. multinational parent with a U.K. subsidiary. U.K. parent has U.S. dollar-denominated cash inflows and outflows, but its desired currency is British pounds. Symmetrically, the U.S. parent's desired currency is U.S. dollars, but due to its U.K. operations, it has cash inflows and outflows denominated in British pounds. The cash flows may be tied to their respective fixed-rate liabilities in their home currencies. This is the usual set up in most plain vanilla currency swaps.

Consider a currency swap contract entered into by these firms. Under this arrangement, there will be a transfer of different currencies between two parties at t = 0, and a state-contingent reversal of cash flows at t = [Tau],(10) at some rate of exchange that is negotiated in advance at t = 0.

Since these contracts are non-standard and specialized, the best means of enforcing performance is not clear to market participants. However, currency swap agreements are generally designed so that a party that elects, at its discretion, not to pay its obligation on the due date cannot realize an economic gain. The Interest Rate and Currency Exchange Agreement issued by the International Swap Dealer's Association (ISDA) contains a provision that makes the defaulting party liable for full payment of amounts due under contract, as well as for additional compensation to cover the future loss of the other party, as long as it is solvent. Thus, for the purpose of this paper, it is reasonable to assume there is no incentive for either party to default at its discretion at t = [Tau], when the true state of nature is revealed, as long as both parties are solvent. For example, at t = [Tau] the exchange rate may be unfavorable to the domestic firm, hence it owes more to the counterparty, the foreign firm. Thus, the domestic party is liable to default. However, under the currency swap agreement, if it does default, it is legally liable for the current and future losses of the other party. In effect, both parties will be indifferent to this outcome. Thus, nonpayment on a due date when both parties are solvent is not a problem in practice, and it is indistinguishable from closing out the contract or marking it to market. As far as modeling default goes, these states are nondefault states because the contractual provisions are in effect.

To define the default states in the model, I consider the cases where one or both parties are insolvent. Since the ultimate recovery in the event bankruptcy occurs is uncertain, I use a simple bankruptcy model that is consistent with the provisions of the ISDA swap documents. In my model, I assume that gross amounts(11) are exchanged and that if one or both parties are insolvent and cannot perform their obligations, the contract is void. No payments are made, and each party ends up with its own cash flow. I believe this resolution is the most sensible for swaps for two reasons: (1) it is stated in the Interest Rate and Currency Exchange Agreement that if one or both parties are, or are about to become, insolvent, this will constitute a true default state and the swap agreement will necessarily terminate; and (2) swaps were introduced as legal innovations that were designed to be superior to parallel, or back-to-back, loans, for which there was a questionable fight of offset in the event of default by one party to the arrangement. Swaps, however, are based on contract law, which states that if one party does not perform, the contract is void and the other party has no obligation to perform.(12)

The notion that swaps are legal innovations is very crucial to modeling their treatment under default. An alternative modeling approach would be to assume that the bankruptcy courts would hold that the swap agreement remains fully effective and that both parties were still fully liable for their payment obligations after one party becomes insolvent. For example, if the exchange rate outcome is favorable to the defaulting bankrupt party, the other party fulfills its obligation and in return joins other creditors in seeking compensation. This approach has been adopted in Cooper and Mello (1991) and Solnik (1990). In their modeling of swap default, if one counterparty defaults, the other party receives nothing even if the swap has negative value to the insolvent party and payment is due to the solvent party. However, the solvent party will have to make the payment whenever the swap has positive value to the defaulting party and money is owed to it. Thus, the insolvent party is released from its obligation, but the solvent party is still liable if payment is due. This type of modeling overlooks the fact that nonperformance by one party releases the other party from having to perform its obligation. In my model, if one party is insolvent and cannot perform, the other party, although solvent, does not perform either. Litzenberger (1992) notes the validity of this approach to modeling swap default based on the ISDA swap agreement provisions.

I let [D.sub.0] and [D*.sub.0] denote the amounts exchanged at t = 0, expressed in domestic and foreign currencies, respectively. For simplicity, I assume that the initial exchange takes place at the current spot rate [e.sub.0], i.e., [D.sub.0] = [D*.sub.0][e.sub.0]. The amounts promised to be exchanged at t = [Tau] for certain are denoted by D and D* in domestic and foreign currencies, respectively. The decision on D and D* will result in an exchange rate e* = D/D*, which may be different from the spot exchange rate [e.sub.0] and the theoretical forward rate f = [e.sub.0]r/r*.(13)

Note, however, that the current market treatment of swap cash flows upon bankruptcy is far from settled. There are conflicting views concerning the enforceability of the provision in the standardized Interest and Currency Exchange Agreement allowing the nondefaulting party to walk away when payment is due to the defaulting party. In some recent cases involving insolvency of one of the swap parties, the nondefaulting party has enforced the clause and refused to pay the due amount to the dismay of some market participants and regulators. In other cases, the nondefaulting party has voluntarily made the payment when the swap had positive value to the insolvent party.(14) While disputes relating to this clause could be settled in court, market participants are not inclined to do so and try to avoid legal action as much as they can. Meanwhile, the enforcement of this clause and its implications need to be made clearer to the swap market. Some dealers and the ISDA would prefer to retain the clause, but others feel that the market is better served without this provision.

III. Currency Swaps in Integrated Capital Markets

First, I look at the swap agreement from the perspective of the domestic firm. I assume the cost of the investment opportunity in the foreign market is known. I denote it I*. The firm has slack in domestic currency, L. The source of the slack is not specified, but I assume that the amount of slack will allow the firm to undertake the investment. One possibility is that proceeds from issuing domestic securities can be the source of slack, in which case a liability in the domestic currency is created. Thus, the domestic firm could issue domestic debt and convert the proceeds to foreign currency. I claim that the firm will engage in a currency swap contract to finance the investment only if the contract al lows the firm to create value above that the finn could realize by using its slack and spot foreign exchange markets.

Furthermore, without loss of generality, I assume a single group of claimants in the domestic capital market, the stockholder. Introduction of other domestic claimants will have no effect on the results of the paper since domestic capital markets are complete and integrated.(15)

First, I value the foreign investment when slack is used. This serves as a useful benchmark against which to compare the value of the same opportunity had it been financed by a currency swap. The value of the firm with slack financing is

V = (L - [e.sub.0] I*) 1/r [Sigma]X*(s)e(s)p(s). (12)

Under the swap agreement, the value of the same opportunity is

[Mathematical Expression Omitted]

where B is the set of states in which the contract is effective. Specifically, B = {s:X*(s) [is greater than or equal to] D* and X(s) [is greater than or equal to] D}. B[prime]=S - B will be the set of states in which the contract is void.(16)

Denote [Mathematical Expression Omitted] as the difference in value due to using a currency swap instead of slack to finance the investment. This function, [Delta]V, measures the difference in value between engaging in a swap and the alternative of using the domestic capital markets and the spot currency markets.(17) Recalling that [D*.sub.0] [e.sub.0] = [D.sub.0] and D*e* = D, I obtain

[Mathematical Expression Omitted]

It is interesting to note that bringing in the integrated capital markets condition embodied in Equation (2) will not result in [Delta]V = 0. Applying Equation (2), Equation (14) reduces to

[Delta]V = D* {e*/r - ([e.sub.0]/r*) + ([e.sub.0]/r* P*(B[prime]) - e*/r P(B[prime]))}. (15)

Similarly, the difference in value for the foreign firm translated into the domestic currency can be expressed as

[Mathematical Expression Omitted].

I can use Equation (2) to obtain

[Mathematical Expression Omitted].

Note that Equation (14) can be rewritten as

[Mathematical Expression Omitted].

Since Equations (17) and (18) are exactly offsetting, at any negotiated e*, the benefit to one party will come at the expense of the other. Specifically, [Delta]V = -[Delta]V* for all swaps.

Result 1. A necessary(18) and sufficient condition for all currency swaps to have a zero-sum outcome is

[e.sub.0]/r* p*(s) - e(s)/r p(s) = 0 for all s.

Note that the above condition can only be true in a complete and integrated international capital market. This is the type of capital market in which the arguments against financial arbitrage presented in Smith, Smithson, and Wakeman (1986, 1988) and Turnbull (1987) are valid. In fact, Result 1 is an extension of the proposition in Turnbull (1987) applied to currency swaps. In such a market, financial managers should look for motivations for swaps other than financial arbitrage. The present paper, however, argues that capital markets are segmented and that deviations from the state-contingent parities are persistently observed.(19)

IV. Currency Swaps in Segmented Capital Markets

After stating that swaps are only relevant in segmented capital markets, I now analyze currency swaps first as default-free and then explore the default implications. It is well-accepted in the literature that swaps represent portfolios of forward contracts. Accordingly, this study models the default-free currency swap as a forward contract. It clarifies some of the findings of previous research that primarily dealt with default-free interest rate swaps. Then I consider how default risk affects currency swaps. My analysis reveals that currency swaps with default risk are state-contingent forward currency contracts. Thus, no publicly-traded forward contract can substitute for the currency swap with default. Since default is firm-specific, currency swaps with default risk can be tailor-made for each counterparty. Specifically, the counterparties can choose the states in which the swap should be effective. There are some practical benefits here. More specifically, my analysis shows how the financial managers of corporations should design risky currency swaps.

A. Default-Free Swaps

Assume that investors in both markets can freely trade the risk-free asset in the other market and that their expectations concerning future exchange rate variation are such that Ee(s) = 1/E*[e.sup.-l](s). Under these conditions, UIRP holds in both domestic and foreign markets simultaneously. Using these parity conditions, Equations (16) and (18) can be written for a default-free swap as

[Delta]V = D*(e*/r - [e.sub.0]/r*) (19)

and

[Delta]V* = D*([e.sub.0]/r* - e*/r). (20)

Once again, [Delta]V = -[Delta]V*, and all default-free swaps (at any negotiated e*) have a zero-sum outcome.

Result 2. Necessary(20) and sufficient conditions for all default-free swaps to have a zero-sum outcome in an international capital market that admits arbitrage opportunities in risky assets are

(i) [e.sub.0]/r* = Ee(s)/r

(ii) [e.sub.0]/r* = 1/rE(*)[e.sup.-1](s)

Note that conditions (i) and (ii) together imply that Ee(s) = 1/E*[e.sup.-1](s). Thus, under risk neutrality, heterogeneous expectations are necessary to achieve the above proposition. A common insight among practitioners is that swaps at origination have zero economic value. My result shows that this insight is valid for default-free swaps only if UIRP holds in both domestic and foreign markets. Admitting that UIRP may not hold in long-dated maturities, even for currencies such as U.S. dollars and deutsche marks, it is evident that swaps rarely have zero economic value even at origination.

Suppose there are forward markets for the two currencies and that all arbitrage opportunities in this market are exploited through covered interest rate arbitrage operations.(21) This results in f = [e.sub.0]r/r*. Using this condition again yields Equations (19) and (20).

Result 2a. With the existence of a forward currency market, a necessary and sufficient condition for all default-free swaps to have a zero-sum outcome is f = [e.sub.0] r/r*.

I have shown that a default-free swap has a zero-sum outcome under certain conditions in the international currency and asset markets. In segmented capital markets, however, none of the above conditions may hold. Therefore, I choose to analyze default-free swaps using Equations (16) and (18) and assuming the existence of deviations from interest rate parities.

To obtain the total gain from a default-free swap contract, I add Equations (16) and (18) and let B[prime] represent the null set. The total gain function, T[Delta]V [is equivalent to] [Delta]V+[Delta]V*, is

T[Delta]V = D*{e*/r [Sigma](1-e(s)/e*)p(s) + [e.sub.0]/r*[Sigma](1 - e*/e(s))p*(s)}. (21)

Equation (21) can be written equivalently as

T[Delta]V = D*{[Sigma]([e.sub.0]/r* p*(s) - e(s)/r p(s)) - e*[Sigma] 1/e(s) ([e.sub.0]/r* p*(s) - e(s)/r p(s))}. (22)

Note that the two terms in braces in Equation (22) are the domestic and foreign UIRP relations. Thus, a default-free swap contract pools together the deviations from UIRP in the domestic and foreign markets. The first sum in Equation (22) is the deviation from UIRP in the domestic market ([DUIRP.sub.domestic]). Similarly, the second sum shows the deviation from UIRP in the foreign market ([DUIRP.sub.foreign]). Also note that T[Delta]V depends on D* and e*. This result may appear counterintuitive since one expects the size of the arbitrage gain to be independent of e* and expects e* to divide the gain between the two parties. However, the swap is a mutual borrowing agreement where the domestic party borrows in the foreign country promising D*, and the foreign party borrows in the domestic country promising D, while D and D* are related to e*. Thus, the borrowing decision variables D and D* can be equivalently expressed as D* and e*. Similarly, one can argue that the size of the gain from a risk-free swap should be dependent on the amounts swapped, D* and D, not on e*. But since e* = D/D* by definition, the size of the gain can be stated as a function of D* and e* as well. This leads to the following implications of Equation (22):(22)

1. If deviations from UIRP in both markets are positive, T[Delta]V will be positive and strictly increasing in e*, but limited by the risk-free borrowing potential of the parties. Thus, e* = X[(s).sub.min]/X*[(s).sub.min].

2. If deviations from UIRP in both markets are negative, T[Delta]V will always be negative. Thus, there is a disadvantage to swap contracts.

3. If deviations from both UIRPs are of opposite signs, e*= [absolute value of] [DUIRP.sub.domestic]/[DUIRP.sub.foreign] will yield a zero-sum outcome.

4. If deviation from [UIRP.sub.domestic] is positive but deviation from [UIRP.sub.foreign] is negative, for T[Delta]V to be positive e* [is less than] [absolute value of] [DUIRP.sub.domestic]/[DUIRP.sub.foreign]. This implies the lower the value of e*, the higher the gains from default-free swaps.

5. If deviation from [UIRP.sub.domestic] is negative and deviation from [UIRP.sub.foreign] is positive, T[Delta]V will be positive as long as e* [is greater than] [absolute value of] [DUIRP.sub.domestic]/[DUIRP.sub.foreign]. This implies the higher the value of e*, the higher the gains from default-free swaps.

These implications tell us when it pays to enter into default-free swaps and show us how to determine the swap rate in certain cases. At other times, we can only identify the upper and lower bounds for the swap rate. In any case, it is implied that to maximize benefits, the swap rate, determined by the amounts swapped D* and D, should be different from the current spot exchange rate, contrary to customary currency swap practice.

It is evident from the above implications that whenever a possibility for positive T[Delta]V is discovered, D* and D, or D* and e*, should be increased (decreased) to the limit. However, we must note that positive T[Delta]V does not imply that individual gains are both positive. Individually, the domestic firm will want to set e* as high as possible, whereas the foreign firm will desire the lowest possible e*. Thus, any negotiated e* will divide the total gain between the two parties.

Alternatively, the market participants may search for an e* such that both parties have positive gains, specifically [Delta]V [is greater than] 0 and [Delta]V* [is greater than] 0 independently. Examination of Equation (21) reveals that [Delta]V [is greater than] 0 only if e* [is greater than] Ee(s). The similar condition for [Delta]V* is that e* [is less than] 1/E*[e.sup.-1](s). Thus, the following result can be stated:

Result 3. In a default-free swap contract [Delta]V and [Delta]V* will both be positive only if 1/E*[e.sup.-1](s) [is greater than] e* [is greater than] Ee(s).

Again, I should point to the significant role played by heterogeneous expectations under risk neutrality. The above inequality would never hold if the expectations were the same.(23) Thus, implicit in the studies that argue for unambiguous gains for all parties due to comparative advantage (e.g., Beidleman (1985), Bicksler and Chen (1986), and Felgran (1987)) is the assumption that the parties have different expectations. This assumption is somewhat implied in Arak, Estrella, Goodman, and Silver (1988) whose results center around the different information sets borrowers and lenders have.

The choice for optimal e* can be approached in two ways by the counterparties. One alternative is where the individual gains are both positive, as stated in Result 3. However, it is also possible that they can choose e* and D*, thus the swapped amounts D* and D, so as to maximize Equation (22) regardless of relative shares in the total gain. This solution may dominate the previous one in terms of total gain, but will result in one of the parties being worse off. For the latter solution to be feasible, the counterparties would have to devise sharing rules for the total gain such that each counterparty is at least as well off as in the former solution. This necessitates a transfer of funds from the party that is made better off to the counterparty. The transfer can take place at the present time t = 0, for example, by exchanging currencies at a rate different from [e.sub.0].

Financial managers engaging in default-free currency swaps whenever a window of opportunity opens should be wary of at least two possibilities. First, they have to make sure that the total gain is positive by examining deviations from UIRP in domestic as well as in foreign markets. Second, they have to make sure that the swap exchange rate e* is chosen such that both parties end up with positive gains. It will be a rare occasion that this rate will coincide with the current spot exchange rate, [e.sub.0]. Thus, the common practice of reversing cash flows at the original exchange rate might be unwise in designing currency swaps. Furthermore, financial managers should explore sharing rules for the total gain other than the swap rate when one party is worse off but the total gain is greater than it would be if both had positive gains.

B. Swaps with Default Risk

In a swap contract, besides e*, there is a second decision variable, the swapped amount. Evidently, when the financial gains from swapping are positive, it is to the advantage of the swap parties to increase the contract amounts as much as possible. In this section, I analyze swaps with default risk and argue that the option not to reverse the deal in some states at t = [Tau] is an independent source of value.

I investigate default implications in a swap contract by examining Equations (16) and (18). I also compute the total gains by adding them together to obtain

[Mathematical Expression Omitted].

Upon examining Equations (16), (18), and (23), the following results become apparent. In a given state, if there is a gain or loss in the default-free term, that gain or loss will be reversed if the contract becomes void due to default in that particular state. Specifically, looking at Equation (16), one can see that, for any s in the set B[prime], what is positive in the first term is negative in the second. A further explanation proceeds as follows. If the swap contract is set such that it is effective in every state of the world, there will be states where (1 - e*/e(s)) [is greater than] 0, which are favorable to the foreign firm. But in other states, it will be negative and unfavorable. If B[prime] turns out to be the subset where (1 - e*/e(s)) is always positive, the firm loses that gain by defaulting. Thus, default must be avoided. On the other hand, if B[prime] is designed so that in those states s in B[prime] in which the finn would have lost in the default-free swap, then default becomes desirable. The reason is simply the fact that, by avoiding the loss, the firm will end up with a higher value than it would in a default-free swap. Similar arguments can be made from the perspective of the counterparty and total gains. Also note that in any state s in B[prime], whenever one party gains, the other party loses. However, as explained in Section I, the gains and losses are of different magnitudes. This implies that the total gain, T[Delta]V, of a swap with default risk will never be equal to zero in an imperfect market in which ([e.sub.0]p*(s))/r* [is not equal to] (e(s)p(s))/r for s in B[prime]. Furthermore, this result is independent of whether UIRP holds in both markets. Thus, the counterparties should search for the optimal D* for any given e* where their gains are maximized independently or jointly.

The essence of firms being able to reap the benefits of international financial arbitrage when individual investors cannot lies in the fact that firms may have specific real assets in foreign markets and can issue claims against the incomes these real assets will generate. In other words, financial activities of firms are closely linked to their real activities. Thus, a particular project owned by a firm will generate a specific cash flow pattern across the states that can screen the favorable and unfavorable potential arbitrage profits in a certain way. In particular, swaps with their unique default structures are a very convenient way of using the state-contingent potential arbitrage profits to the benefit of both counterparties, as explained in the preceding paragraph. In other words, risky swaps are a unique mechanism devised by firms that can benefit from inefficiencies in capital markets caused by non-swap transactions (individual portfolio holdings). Swaps with their unique security design can overcome barriers that cannot be eliminated by individual portfolio rebalancing.(24)

Inspection of Equations (16) and (18) will show that the optimal D* will differ for each party. Hence, under this criterion, the swap contract will not be feasible without an intermediary. The alternative criterion is where the counterparties optimize jointly. As discussed in connection with default-free swaps, joint optimization implies a search for a common set B[prime] where T[Delta]V is maximized, regardless of individual shares. I will have more to say about these solutions and criteria later in the paper in the context of a numerical example. Here, I choose to investigate further the implications of the default option on the value of a currency swap contract.

I introduce the following function:

[[Chi].sub.B[prime]] = [1 whenever s belongs to B[prime]

[0 elsewhere

Thus, Equation (18) can be restated as

[Delta]V = D*{e*/r [Sigma](1 - e(e)/e*)p(s) + e*/r [Sigma](e(s)/e* - 1)[[Chi].sub.B[prime]]p(s)}. (24)

Further simplification of the terms yields

[Delta]V = D*{(e*/r - Ee(s)/r)(1-P(B[prime])) + cov(e(s),[[Chi].sub.B[prime]])/r}. (25)

The expression in Equation (25) is revealing. If there were any gains to a default-free swap, the gains would be partially lost due to the probability of default, as expected.

Surprisingly, there is a second term, the covariance of exchange rates and default states. As long as that covariance is positive, it will offset the value lost due to default, and an optimal B[prime] that maximizes [Delta]V can be attained.

Similarly, Equation (16) can be restated as

[Delta]V* = D*{([e.sub.0]/r* - [e.sub.0]e*E*[e.sup.-1](s)/r*)(1 - P*(B[prime])) + [e.sub.0]e*/r* cov*([e.sup.-1](s),[[Chi].sub.B[prime]])}. (26)

Clearly, similar arguments can be made concerning the gain of the counterparty.

Result 4. In a swap contract probability of default is not a sufficient measure of the impact of default on value gains of each counterparty. Covariability of exchange rates and default states may play an offsetting role.

The joint optimization case where T[Delta]V is maximized can be represented as

T[Delta]V = D*{[([e.sub.0]/r* - Ee(s)/r) + e*(1/r - [e.sub.0]E*[e.sup.-1](s)/r*)] + [e*(1/r P(B[prime]) - [e.sub.0]E*[e.sup.-1](s)/r* P*(B[prime])) - [e.sub.0]/r* P*(B[prime]) - Ee(s)/r P(B[prime])] + [[e.sub.0]e*/r cov*([e.sup.-1](s).[[Chi].sub.B[prime]]) + 1/r cov(e(s),[[Chi].sub.B[prime]])]}. (27)

T[Delta]V is the sum of three terms in braces. The first term is the gain had the swap been default-free. The other two terms appear due to the possibility of default. The second term is exclusively based on the probability of default. However, the value of this term can be either positive or negative. Thus, it is ambiguous how the total gain will be affected by the probability of default.(25) Finally, there is a third term that depends solely on the covariances of exchange rates and default states in the two countries. Also note that the analysis of the paper points to potential gains from swap default and the factors that affect it. The opportunities for such gains may not be pervasive due to the current practice of swap financing. This is an empirical question to be addressed in future research. However, my analysis has implications for designing swap contracts that will exploit potential gains associated with swap default risk.

Default risk is one of the reasons why swaps cannot be totally standardized. Default risk has intrigued the swap industry since its inception. Practitioners have tried to sidestep the credit risk complications by either matching the credit risk of counterparties or by contracting with a third party that has specialized in bearing credit risk, such as a dealer or a bank. Unless this intermediary assumes the credit risk of each counterparty, the end users will choose to stay with counterparties that have high credit ratings. Also, current industry practice is to quote the same swap rate to counterparties with different credit ratings. This paper brings a new outlook to swap default risk considerations. Swap market participants need not be so conservative concerning credit ratings. It may be possible that a AAA-rated swap party gains more when it enters into a swap with a BBB-rated counterparty than with an intermediary. The gain depends not only on the probability of default, in a negative way, but also on a second factor, covariance of e(s) and the default states. The desired covariability might be achieved by swapping with a counterparty other than a riskless one. Also, for each pair of swap counterparties, there is a different swap rate, e*, that maximizes their gains and that depends on the amounts swapped, and thus on their specific default structures. It will be suboptimal for all parties to use the same swap rate. My results do not diminish the function of an intermediary. My model allows for intermediation for reasons other than credit specialization. The intermediation function is indispensable to match quantities swapped since optimal swap amounts may be different for different swap counterparties.

C. Default on a Currency Swap as a Call Option

The default features of currency swaps are highly suggestive of an option contract. To analyze the option-like properties of swap default, look again at Equation (14). Note that the value of a default-free swap contract is identical to that of a forward contract. This is no surprise since I modeled the swap agreement after a forward contract. What is of interest is the second term that emerges due to the possibility that one of the counterparties might default.

Recall that the payoffs to a call option written on a domestic currency can be expressed as max[0,(e(s) - e*)], where e* is the exercise price. Specifically, the payoffs to this call option are positive in those states in which e(s) [is greater than] e* and are equal to (e(s) - e*). These payoffs appear in the second term of Equation (14). Had B[prime] been equal to {s: e(s) [is greater than] e*}, those states that are unfavorable to the domestic firm, the value of the default option would be identical to a call option on the domestic currency. However, this is not the case, and B[prime] = {s:X*(s) [is less than] D* and X(s) [is less than] D*e*}. Evidently, B[prime] is dependent on nonperformance of the counterparties based on their cash flows and the amount swapped, rather than on the relationship between e* and [e.sub.[Tau]](s). In particular, B[prime] is a collection of states where [Mathematical Expression Omitted]. This implies that the call value may be positive or negative depending on the e* chosen and on the cash flows of the counterparties.

Thus, one can interpret this default term as a "hybrid" call option, where the payoffs are identical to those of a call on the domestic currency but are effective in those states in which the counterparties default.(26) A perfectly analogous argument can be made for the foreign firm.

Result 5. The value gain of a counterparty due to financing through a currency swap is equal to the value of a default-free forward contract on the domestic currency plus the value of a "hybrid" currency call option.

The value of this call option can be positive or negative. The task of financial managers is to design the swap contract so that this call option has positive value to both parties.

V. Numerical Example

By assigning plausible values to the market parameters concerning the [Delta]V and [Delta]V* functions, I next show how swap counterparties can increase shareholder wealth in segmented international capital markets.

The following numerical example uses the data presented in Table 1. In the computation of [Delta]V, [Delta]V*, and T[Delta]V, there are two decision variables controlled by the counterparties, namely, the swapped amount in foreign currency, D*, and the rate at which the exchange occurs at t = [Tau], e*. In my discrete example, I let D* represent the discrete cash flows of the foreign investment opportunity, that is D* = X*(s) for a particular s.

Implicit in the given data is that deviation from [UIRP.sub.domestic] is -0.08 and deviation from [UIRP.sub.foreign] is +0.1805. From Equation (22), we know that any e* [is greater than] [absolute value of] -0.08/+0.1805 or e* [is greater than] 0.4432 will yield T[Delta]V [is greater than] 0 for default-free swaps. Furthermore, Result 3 implies that in the range 0.594 [is greater than] e* [is greater than] 0.566, both [Delta]V and [Delta]V* will be positive. Thus, in the numerical example, I let e* equal 0.4, 0.58, and 0.8, respectively. For each value of e* chosen, there is a series of D values corresponding to the D* values described above since D = D*e*. The cash flows presented in Table 1, with these D* and D values, determine the breakdown of S into subsets B and B[prime].

Table 1. Data States p(s) p*(s) e(s) X(s) X*(s) 1 0.19 0.30 0.60 75 150 2 0.09 0.02 0.30 100 200 3 0.13 0.08 0.45 125 250 4 0.21 0.25 0.65 50 100 5 0.26 0.30 0.70 150 300 6 0.12 0.05 0.40 25 50 r = 1.05 r* = 1.09 [e.sub.0] = 0.5 Ee(s) = 0.566 E*[e.sup.-1](s) = 1.683

The results of the computations for each e* chosen are presented in Table 2. The values of [Delta]V, [Delta]V*, and T[Delta]V for different cases of default are detailed in the same table. I first examine the default-free cases where e* is the only decision variable. When e* is low, the foreign counterparty gains at the expense of the domestic counterparty. However, the total gain is negative, which makes it impossible for the contract to be implemented. At high e*, the situation is reversed, and the resulting total gain is positive. A swap contract is possible, but sharing rules have to be developed since one of the parties receives negative gains. Only when e* = 0.58 do both parties to the swap agreement augment value, simultaneously depicted by positive values of [Delta]V and [Delta]V* in panel B of Table 2.

The more interesting issue concerns the possibility that the counterparties can do better by increasing the amounts swapped. In that case, the swap is no longer default-free. A number of cases emerge. In general, each party can obtain a better outcome at some higher level of swapping. This is intuitive since at some level of D* default occurs in the states that are unfavorable to a particular party. However, optimal amounts of swapping associated with the best outcomes for each party do not necessarily match. Looking at panel B of Table 2, the domestic firm would like to set D* = 150, while the foreign firm would independently like it to be 50. However, total gain will be maximized at D* = 150. Noting that [Delta]V* at D* = 150 is almost as large as T[Delta]V at D* = 50, the solution to this example can reasonably be D* = 150.

Finally, it seems that both parties will be better off by exchanging at a higher rate than e* = 0.58. This is suggested by the higher value achieved for the best outcome of T[Delta]V when e* = 0.8. It should also be noted that the best outcome corresponds to a lower level of swapping. This can be a superior alternative only if the counterparties can develop sharing rules such that both can benefit from the swap. Obviously, these considerations lead to inclusion of an intermediation function and transactions costs, which are left for future research.

Table 2. Value Gains from a Currency Swap(a) Panel A. Assuming that e* = 0.4 D* D s[is an element of]B[prime] [Delta]V [Delta]V* T[Delta]V 25 10 [Phi] -3.95 +3.75 -0.20 50 20 [Phi] -7.91 +7.50 -0.41 100 40 6 -15.81 +14.99 -0.82 150 60 4,6 -16.22 +15.88 -0.34 200 80 1,4,6 -14.38 +12.00 -2.38 250 100 1,2,4,6 -20.12 +15.76 -4.36 300 120 1,2,3,4,6 -22.29 +17.64 -4.65 Panel B. Assuming that e* = 0.58 D* D s[is an element of]B[prime] [Delta]V [Delta]V* T[Delta]V 25 15 [Phi] +0.33 +0.27 +0.60 50 29 6 -0.36 +1.05 +0.69 100 58 4,6 +0.68 +0.86 +1.54 150 87 1,4,6 +1.56 +0.67 +2.22 200 116 1,2,4,6 -2.73 +2.60 -0.13 250 145 1,2,3,4,6 -7.43 +5.90 -1.53 300 174 S 0 0 0 Panel C. Assuming that e* = 0.8 D* D s[is an element of]B[prime] [Delta]V [Delta]V* T[Delta]V 25 20 [Phi] +5.57 -3.97 +1.60 50 40 6 +8.86 -6.79 +2.07 100 80 1,4,6 +11.10 -6.34 +4.76 150 120 1,2,4,6 +10.21 -7.22 +2.99 200 160 S 0 0 0 250 200 S 0 0 0 300 240 S 0 0 0 a Ruling out the cases where T[Delta]V [is less than] 0, the best outcome, if positive, in each column is boldfaced.

VI. Conclusion

In this paper, I developed a model to measure the value gains of the counterparties in a currency swap. The model links the value gains to state-contingent interest rate/exchange rate disparities. Given these market conditions and expected cash flows, the swap counterparties decide on the swap rate and the amount to be swapped. In an integrated market all swaps are zero-sum games. In a segmented market, all default-free swaps are zero-sum games conditional on a two-way uncovered interest rate parity. Swaps with default in a segmented market can create value gains beyond that of a default-free swap. With the possibility of default, two sources of uncertainty--cash flows and exchange rates--interact. The default option, state by state, becomes a screening device for the favorable and unfavorable exchange rate outcomes. Evidently, a chance to eliminate the unfavorable outcomes emerges.

My findings suggest that financial managers should look for both risk-free and risky arbitrage opportunities when they design currency swaps. Risk-free arbitrage does not necessarily imply positive gains to both parties at any swap rate. The optimal swap rate can be approximated such that it lies between the expected exchange rates in the two markets. Even if risk-free arbitrage is not possible, financial managers should look to benefit from risky swap contracts. If they can tailor currency swaps--state-contingent forward currency contracts--such that default states and exchange rates are positively correlated, they will be able to enhance shareholder wealth.

Finally, my results have empirical implications for successful, wealth-enhancing, as well as unsuccessful, wealth-reducing, swap outcomes. Any empirical test should examine the wealth gains to both parties under a swap agreement and distinguish between cases where both are positive, both are negative, or they are mixed. There are three empirical implications. First, for default-free swaps, a positive market reaction to a swap announcement is more likely if the differential effective yields on risk-free borrowings in the two countries are significant and have the desirable signs. Note that these deviations are pairwise country-specific and thus can be more easily tested with observations from the same country pair. Second, again with default-free swaps, a positive market reaction is more likely for both parties whenever the swap rate is bounded by expected exchange rates that are different in the two countries. Otherwise, the market reaction is more likely to be negative for both or mixed. Third, with risky swaps, empirical tests may be misleading if they ignore a second factor besides differential effective yields as a source of positive market reaction. My study shows that a positive market reaction is more likely with risky swaps if the covariance of exchange rates and default contingencies is positive. This implication is project-specific. Those projects that tend to generate low income when the respective home currencies are undervalued will be more likely to cause a positive market reaction for both parties.

1 For example, principal amounts are exchanged in currency swaps but not in interest rate swaps, which has a significant effect on default risk exposure. Furthermore, in a pure interest rate swap (i.e., in the same currency) the source of additional uncertainty (besides cash flow uncertainty) is interest rate fluctuations. In a credit swap (interest rate swap with two currencies involved), however, there is both interest rate and exchange rate uncertainty along with cash flow uncertainty.

2 This is not to say that foreign direct investment by firms is free of government regulation. To illustrate, consider the following limitations imposed on foreign corporate and noncorporate investors.

In the United States, projects involving natural gas imports into the United States must be approved by the Economic Regulatory Agency (ERA), the Department of Energy, and the Federal Energy Regulatory Commission (FERC). In their review, the agencies can limit the value of imports to protect U.S. public interest.

In Canada, investments requiring review must pass the "net benefit to Canada" test and must not threaten "cultural heritage and national identity." According to the Investment in Canada Act, investments requiring review by The Foreign Investment Review Agency (FIRA) include any "direct" acquisitions of Canadian businesses by non-Canadians with assets of $5 million or more and any "indirect" acquisitions involving assets over $50 million. For Americans, the thresholds are $150 million for direct acquisitions and $500 million for indirect acquisitions. These rules do not apply to investments that acquire control of Canadian businesses engaged in the production of uranium, financial services. Transportation services, or a cultural business.

In Japan, The Foreign Exchange and Foreign Trade Control Law restricts imports and exports of certain categories of goods as well as goods originating in certain countries. According to this law, inward direct investment is also regulated and is subject to review. Direct investment in several specified industries (agriculture, fisheries and forestry, 50% or greater ownership in mining) may not be permitted. Foreign participation in other industries, such as banking, financial services, utilities, and insurance, is also regulated under special statutes.

Similar restrictions apply in other countries. Details are available from the author upon request.

3 This result emerges because the present paper, contrary to previously cited work, bases its analyses on present values rather than rate differentials. As Turnbull (1987) has also suggested, the latter analysis can be misleading if financial instruments differ in design and risk. Thus, instead of invoking rate differentials in swapping, the concept of value creation is invoked since any cost savings due to rate differentials should necessarily enhance shareholder wealth. Furthermore, Wall (1987) argues that cost savings may be related to avoidance of agency costs associated with long-term debt via an interest rate swap. This source of cost savings is a separate issue and may interact with cost savings or dissavings due to financial arbitrage. In case of savings due to financial arbitrage, savings on agency costs will further enhance value. However, if for a party there are dissavings due to financial arbitrage. the benefit of any agency cost savings may be wiped out.

4 This is parallel to the well-known result that risky debt may create value by completing the market (Senbet and Taggart (1984)). In contrast, my analysis assumes complete markets; default risky swaps create value due to market segmentation.

5 Solnik (1990) finds that the markup an intermediary charges in contracting a swap with a risky client is a function of the subjective probability of default alone. This is due to his assumption that interest rates and default risk are uncorrelated.

6 Studies that address the issue of the stochastic interdependence of cash flows and exchange rates in the firm's financing decision are rare in the finance literature. Recent examples are Thomadakis and Usmen (1991, 1994) in which the firm's project financing decisions are examined.

7 For results with risk aversion in a similar setting, see Thomadakis and Usmen (1991). A related paper, Tessitore (1994), considers real decisions of oligopolistic firms in segmented markets in which risk aversion factors may differ.

8 Note that with homogeneous beliefs, p(s) = p*(s) and the following condition implies that e(s) must be a constant. Thus, perfect, integrated markets in which investors are risk neutral and hold homogeneous beliefs about state probabilities is inconsistent with flexible exchange rates. This should not be a surprise since the market described above will be identical to a domestic market where the unit of currency is the same in nominal terms, [e.sub.0] = e(s) = 1. A fixed exchange rate regime in which e(s) = [e.sub.0] r/r* would also be compatible with this market structure.

9 Note that this condition is not a result of market incompleteness. The international capital market, as well as the national capital markets, are assumed to be complete in our study.

10 A currency swap contract may include a multiperiod exchange of payment streams. The exchange of payments on each payment date can be viewed as a state-contingent forward contract. For this reason, the swap contract can be viewed as a portfolio of state-contingent forward contracts; that is, [Tau] may stand for any one of the scheduled payment dates. Moreover, the initial exchange may not take place. I include that feature in my model because I later suggest its use as an instrument for sharing the gains from swaps. Its inclusion does not play a crucial role in the results that follow.

11 Exchange of gross amounts instead of the better-known procedure of net exchange is the rule with currency swaps. Nevertheless, exchange of gross amounts is not a necessary restriction for the model to work. In my model, one of the decision variables is the swapped amount. Net exchanges can always be made bigger to match the gross amounts of the model if notional principal amounts are increased when it pays to do so.

12 For a further discussion, see Beidleman (1985, 1992) and Tessitore and Usmen (1993).

13 The default-free swap contract is indistinguishable from an off-market, implicit forward foreign exchange contract except that it is a longer-term instrument; the average swap has a 7 to 10 year maturity. The theoretical forward rate may be observed in the currency markets where there are no obvious covered interest arbitrage opportunities. However, in long-term markets, which are the concern of this paper, the theoretical no-arbitrage relationship (f = [e.sub.0] r/r*) may not hold. Thus, there is room for negotiation in the forward rate specified in the swap contract. Moreover, swaps with default risk are state-contingent forward contracts, and there is no reason for the swap rate to equal the theoretical forward rate that is consistent with risk-free covered interest arbitrage.

14 For a fuller discussion, see Shirreff (1991).

15 In the set up of this paper, swaps would have no direct bearing on the cash flows received by the domestic creditors of the firm. Debt claims have seniority over other claimants, such as off-market swap counterparties, and in my model, when the firm is bankrupt, no payment is received from the solvent counterparty. This approach stands in contrast to Cooper and Mello (1991), who assume that the nondefaulting party makes a payment even if the other party is bankrupt. This payment naturally adds to the liquidation value of the firm in some states, and it is the source of the wealth transfer to bondholders in their paper. In my model, there are no wealth transfers to bondholders.

16 This bankruptcy condition implicitly assumes that the parent will not support the payment obligation of the subsidiary. It may be argued that under certain circumstances the parent will back up the payment obligation of the defaulting subsidiary to maintain its reputation even if it is costly in the short run. The analysis of this reputation effect is left for future research.

Printer friendly Cite/link Email Feedback | |

Title Annotation: | includes appendix |
---|---|

Author: | Usmen, Nilufer |

Publication: | Financial Management |

Date: | Jun 22, 1994 |

Words: | 11389 |

Previous Article: | Bankruptcy costs and the financial leasing decision. |

Next Article: | Default risk, firm characteristics, and the valuation of variable-rate debt instruments. |

Topics: |