This article was downloaded by: [Washington State University Libraries ]On: 25 October 2014, At: 01:22Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,UK

International Economic JournalPublication details, including instructions for authorsand subscription information:http://www.tandfonline.com/loi/riej20

Minority Ownership, Deferral,Perverse Intrafirm Trade andTariffsProfessor Chander Kant aa Seton Hall UniversityPublished online: 28 Jul 2006.

To cite this article: Professor Chander Kant (1995) Minority Ownership, Deferral,Perverse Intrafirm Trade and Tariffs, International Economic Journal, 9:1, 19-37

To link to this article: http://dx.doi.org/10.1080/10168739500000003

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all theinformation (the “Content”) contained in the publications on our platform.However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, orsuitability for any purpose of the Content. Any opinions and views expressedin this publication are the opinions and views of the authors, and are not theviews of or endorsed by Taylor & Francis. The accuracy of the Content shouldnot be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions,claims, proceedings, demands, costs, expenses, damages, and other liabilitieswhatsoever or howsoever caused arising directly or indirectly in connectionwith, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes.Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expresslyforbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

INTERNATIONAL ECONOMIC JOURNAL Volume 9, Number 1, Spring 1995

MINORITY OWNERSHIP, DEFERRAL, PERVERSE INTRAFIRM TRADE AND TARIFFS

CHANDER KANT* Seton Hull University

When the foreign subsidiary has minority local ownership and the MNF engages in transfer pricing, its intrafirni exports are always from the country with the higher marginal cost. Further, permitting defxrral from home taxation of non-repatr~ated foreign profits changes the nature of intrafirm trade from efficient to perverse even when the foreign subsidiary is fully-owned by the MNF. lntrafirm trade differs significantly from that between unrelated buyers and sellers, and tariffs on such trade (when it is perverse) can restore global production efficiency. [F12]

1. INTRODUCTION

Foreign direct investment has been increasing rapidly throughout most of the world. A'significant part of foreign direct investment is in horizontally-integrated activities (where these activities could be production of an intermediate o r a final good, o r a service); and a large part of international trade now consists of internal transactions between units o f multinational f i rms (MNFs) located in different countries. For example, almost 40% of United States imports and exports in 1984 were between U. S . firms and their foreign affiliateslparents (see?Little, 1987); approximately half of cars sold by Honda Motor Company in the U. S . are now built in the U. S. ; there has been a phenomenal growth in multinational banking, insurance and other financial services recently; and many resource-extracting or r e s o u r c e - b a s e d i n d u s t r i e s l i k e c r u d e - o i l o r p e t r o l e u m p r o d u c t s a r e o f t e n "multinational" in character. Also, a large number of foreign subsidiaries are not wholly owned by the MNFs; and many home countries defer home taxation of non- repatriated foreign profits.

Foreign direct investment has recently been analyzed in the literature using both partial and general equilibrium models of the multinational firm.' These models invariably assume or conclude that if the M N F undertakes internal real transactions, it

* This paper has benefited from helpful comments by three anonyniouslrefereek for this Journal and from seniinar participants at the N.B.E.R.

' See Horst (1 97 I). Katrak (1977 and 1980). Eden (1 978). Batra and Hadar (1 979), Batra and Ranlachandran (1980), Itagaki (1981), Markusen (1984), Helpman (1984). Ethier (1986). Kant ( 1988a and 1988b), Horstmann and Markusen (1 987 and 1989), Ethier and Horn (1 990), Stoughton and Talmor (forthcoming) and Gresik and Nelson (forthcoming). Most of these models consider or allow for an horizontally-integrated MNF.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

20 CHANDER KANT

is because of real cost differences, However, intrafimi trade can itself be used by the MNF as an 'international conveyor of profits'.' Then, factors like minority local ownership of the foreign subsidiary or deferral froni home taxation of non-repatriated foreign profits can themselves induce the MNF to engage in internal real transactions.

The purpose of this paper is to cxamine real implications of the two factors mentioned above; i t . , whether minority ownership and/or cleferral lead the MNF to export intrafirm froni the higher cost country. Its Section 2 presents the model, defines minority local ownership and exaniines the implications of foreign tax credit and deferral provisions for the effective statutory tax rate on profits earned in tlie foreign country. In Section 3 we analyze whether intrafirm tradc is perverse (i.e. originating from the higher niarginal cost country) when the home country permits deferral froni lionie taxation ol' non-rcp~ttrinted foreign profits. Sectio,n 4 presents different results when either deferral is not permitted or it is not in the MNF's interest to defer repatriation of its foreign profits. Section 5 exaniines how thc i~itroduction of tariffs changes the results given in tlie previous two sections, and compares quotas to tariffs. General iniplications of the results itre presented in Section 6.

2. THE MODEL

Consider an MNF prorlucing and selling a product in two countries. This product may be any of the following: a find good, an intermediate good, a service. I t also exports part of its output from its unit in one country to that in the other. We perniit intrafirni trade in either direction. i.c. the MNF's exporting i l l l i t coi~ld either be in the home country (coi~ntry I ) or in the foreign country (country 2) . The technology which makes such strategy optinium will be rlcscribcd belo\v. The MNF has market power in tlie t \ v ~ countries, and the industry in which i t operates has nun-insignificant barriers to entry. We wish to examine tlie reill iniplications of internal real transactions in tlie presence of niinority foreign ownership and/or deferral.

Let n,. I , s,. .v,. K, (s,). C;. (s;). I - , . c,. ;uid c', represent gross pure profits, statutory profit tax rate, sales, output, total revenirc and total cost functions. niarginal revenue and marginal cost functions and slope of the marginal cost function, respectively. in country I , ancl let T, = ( I - t,). \vlicre i = 1. 2.' First consicler that the MNF exports a part of output of its unit in tlie home country - cirllecl tlie parent firm - to its subsicliary in the lorcign country. Then,

' The terms "iltternal real transactions" ant1 "i~l(r;rl'irm tradc"arc usecl interchangeably in this paper.

' I t is u.orth emphasizing that pure ~~rol'irs hcrc mcans total revenue minus total cost (inclutling opportunity cost of ownetl capital) mlnus normal profits in the respective country. Further. we ;rssulne that accrleratcd cr~pit;rl cost recovery allowa~lces. investment tau credits. other tletluctions or escmptio~is, and rctluced [ a s rates or outright subsidies granted for particular activities or regions in most tau coclc's make the tasable profits npprosimately equal to pure profits so that the sta~utory profit tax rate is Ic\,ictl and rerrlized on pure profits.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

n, = R, (s, ) - C, (s, + 111) + p111 ,

n2 = R2 ( s 2 ) - CZ (sZ - 111) - pill ,

where 111 are the MNF's imports into country 2 from coutitry 1 and y is the transfer prics. Clearly, s, = s, + 111, and .v2 = s, - 111. Assume .s, > m > 0.

These individual pure profit functions are restated below for internal real tri~nsactions flowing in the other direction. Let m' represent amount of the foreign subsidiary's output which is now exported intrafirm to the parent firm. Then, .v, = .s, -

111'. .v, = .s2 + III'. s , > 111' > 0 (by assumption) nncl

Note that ( I ' ) and (2') can alternatively be obtained by substituting it1 = - i r r ' in ( I )

ancl (2 ) .

Minority O~vliership

Consider minority local ownership of the subsidiary. This coi~ltl Ije due to foreign ownership restrictions (in at least some sectors o r inclustrics) inlposetl by most countries. Some of the more prominent of these countries are: Argentina. Australia. Brazil, Canada. Colombia, Egypt, Fr:uice, Incba, Indotlesin, Japan, Korea. Mcxico. Spain, Sweden ancl Venezuela. See Coopcrs and Lyhrand (1993). Minority local ownership has been introduced by Kant (1988b iund 1990) by a straight forwartl extension of the earlier M N F models. Let k be tlic proportion of the foreign subsicliary owned by the MNF, and assume

The lu\ver liniit on k is imposed so that the M N F has full control over its foreign subsidiary: and the MNF owns its foreign subsicliary fitlly \vhen k = I.'

-' This paper cloes 1101 go into the bargaining problem between local minority sliarclioltlers and the parent firm. For :in analysis of this problem. see Stoughton and T:~lmor (forthconiinp). Lee (1090) oncl Miyagiwa (1992) also examine the effects of foreign shareholding. But. their model assumes foreign equity investment to be-portfolio rather than direct so that eitlier country's firm masimizcs its own profit function rathcr than the joint or the global profit function.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

22 CHANDER KANT

Foreign Tax Credit Scheme with Deferral

Under deferral, profits of a subsidiary are taxed by the home country only when such profits are repatriated to honie. Tlien, under the foreign tax credit scheme (assumed in this paper), the effective statutory rate on repatriated foreign profits is equal to sum of the two countries' statutory profit tax rates minus the foreign tax credit given by the home country. This is so because the home country taxes the repatriated foreign profits at its tax rate but gives a tax credit for foreign taxes paid. The rate at which the tax credit is given is the s~naller of the two statutory tax rates.

Note that the effective statutory rate and the statutory rate are the same for home profits. Let t; represent the effective statutory profit tax rate levied on pure profits earned in country 2. We wish to rclatc the magnitude of t; to those of the statutory tax rates, t,. First consider some non-repatriation. The MNF gains by not repatriating any part of its foreign profits only if t , > t,. Let q represent.the proportion of foreign profits repatriated, where 0 5 q < I. The Iionie tax on repatriated profit is then t , q q and the foreign tax credit received is t,r/rr,. Clearly.

ti = 1, + t,q - t,q = t,, , where

t,, = t: + (t, - t,)q .

Now consider following two situations concurrently. Consider either that t, 5 t, (so that the MNF does not gain by not repatriating its foreign profits fully), or t , > t, but the honie country does not perniit deferral frv111 honie taxation of non-repatriated foreign profits. In either case, for

t, t2 , t; = (t, + t,) - t, = t, , and for

t, < t, , ti = (1, + t2) - t , = t, .

To summarize, we have the following three cases: (A) t, > t2 and non-full repatriation (i.e., q < I). Then, t; = t,,; (B) r, 2 t, and either full repatriation (for t , = t,) or non- deferral from home taxation of non-repatriiited foreign profits (fort, > 1,). In that case, ti = t,; and (C) t, < t, where t; = t,. Case (A) is discussed in the following section. p n the other hand, different results for the other two cases are presented in Section 4.

3. t, > t, WITH DEFERRAL

First the case of minor~ty foreign ownership of the subsidiary and then that of full

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

INTRAFIRM TRADE AND TARIFFS

ownership by the MNF over its foreign subsidiary are discussed.

A. Minority Local Ownership

Let n represent the MNF's global net profit function. Then,

n = 'I;n, + T,,kn,, where

GI = (1 - tT1) = (1 - t l ) + (tl - t ? ) (1 - q) > 0 .

Consider first that the direction of intrafirm trade is from the parent firm to the foreign subsidiary so that the two individual profit functions are defined by (1) and (2), respectively. Then, the partial derivative of n with respect t o p is

TI, = ('I; - &,k)m = Ttn where T = ('I; - T,,k).

But T can also be stated as :

The two terms on the right hand side of (9) are respectively positive and negative, and the sign of T is indeterminate. Assume it to be non-zero. First consider the T > 0 case. An increase in y increases the parent firm's (the subsidiary's) export revenues (import costs). But, only the k-proportion of the subsidiary's import costs are borne by the MNF [with (I - k) proportion borne by local shareholders in the foreign country], while the export revenues accrue wholly to the parent firm. Therefore, an increase in p transfers pure profits from local shareholders in the foreign country to the parent firm. This positive effect - called the minority local ownership effect - is captured by the first tern1 on the right hand side of (9). On the other hand, (t, - t,)(l - q)k measures the global profit tax savings by the MNF due to non-full repatriation of its share of foreign profit. An increase in p , by decreasing ~r,, reduces this tax-saving effect and hence decreases IT. This negative effect-called the deferral effect-is measured by the 1- k(t, - t, )(I - q)] term in (9).

When T c 0, a decrease in p decreases the parent firm's (the subsidiary's) export revenues (import costs). By analogous reasoning as above, the minority local ownership effect now means a shift of pure profits from the parent firm to local shareholders in the foreign countyy. The deferral effect is now interpreted as follows: ;decrease in y, by increasing ~r,, increases the tax savings effect of not repatriating ~r, fully to the home country, and hence increases n.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

Clearly. in the case ol' a positive (~legati\.c) 1'. the minority locitl o\vncrsliip cl'fcct clominates (is clominatccl by) the tlcfcrrnl cfl'cct. Il'therc were 110 prospect ol'a tr~unskr pricing penalty, in the kwrner (latter) casc the M N F \vo~rlcl hc intercstccl in increasing (clccrcasing) its transkr pricc as mi~cli irs pohsiblc. I-i)llo\cing Kant ( I9XXa). consicler now that the MNF faces the prospect of imposition ol' it pc~ialty where the probability. a, of tlie penalty depends on the transfer pricc tleclitrccl by it. Assume that tlie government has a guiclelinc that tlie transfer price slioirlcl he cqilal to ihc itrm's length price. i, . I f the MNF charges F there is 110 threat 01' the penalty, ancl cr = 0. 11' i t cloth not charge the arm's length price, a transfer pricing penalty can he imposed on i t ; i.c. in that casc u is positive.

The probability of imposition uf the pcnalty clepcncls on the clivcrgcnc~c bct\vccn the transfer pricc charged ancl !lie arm's length pricc. i.c.. ( = ( 1 - . Let 1 7 ~ . he tlic transfer price which triggers the transkr pricing penalty with certainty. i.c. where cr = I . As 1) gets closer to I), , tlic probitbility ol' promulgation of tlic pen;tlry incrcascs, itlid IISSLII~IC i t iric~~e;~scs ;I( an increasing rate:. Thus, lor increase in the transkr pricc (ITP) casc.

, /?,I . b < 1) < / I , . 0 < (x( 1) - ;J) < I . a'( / J - ;)) > 0. u"( 1) - ;) > 0.

\vhile for clecreasc in the translkr pricc (DTP) casc.

, / i , l . b > 1, > I>,.. 0 < a ( / ) - ;)) < I . a ' ( / ) - ;)I < 0. [x"(/) - F) > 0. ( I I )

CIeitrly.

sign or'(/) - f i ) = sign T. (12)'

Let 5 > 0 represent the triunslcr pricing pcnitlty. Tlicn. the cspcc.tccl Ios\ clue to tlic

pcnirlty is :

( a ( / ) - f i )+O.[ l - a ( / ) - ; ) I = ( a ( / ) - ; ) ) > O . (13)

Thcrcforc, the ol>jcctivc I'ilnction ol' the MNF. 4. itncl thc first-orclcr c~orlclition \\;it11

respect 10 /I itre:

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

INTKAFIKM I'KADE ANDl't\I<lFFS

(h,, = f i l l - {a'(/) - r,, = o .

The litst cqitatiotl clil'l'crs llom similiu ccl11;ttions ill other MNI; nloclcls ill slio\ving that the clcri\lativc of the MNIz's ohjccti\lc I'unction with respect lo 11 clocs not have the same sign for all pcrmissihlc v:~lucs o f / ) . Clearly, the MNF mity not find it optimal to charge thc corncr or the limiting transl'cr pricc. In the 'I'> 0 cnsc. tlie first tcrm on the right hand side of ( 15) is positive, ancl the MNI: has inccnti\lc to incrci~se its transli'r price. But, on increase in the transfer pricc :tlso incrc;t<cs tlic probability of the translbr pricing penalty. This negative cfl.ect on (h is mcasurcil hy the 1- 5d (1) - j ,) / term in (15). Only if tlic I'orrner effect ilominatcs the latter lor all permissible values of p. will the optimum transfer pricc be the corner or the limiting transfer price. But, due to the increasing probability of the penalty (which increases at an increitsing rate). the more likely outcome is that the optimum transfer pricc is in the interior, i.e., between y iuid p,..

Similarly. for tlie DTP case. Hirslilcifcr (1957) prvvccl that the arm's length price. a, equals c,. Thus, in tlic ITP (DTP) cnsc, the MNF's o p t i m ~ ~ m transfer price. 1'. is greater (less) than the exporting coi~t~try 's marginiil cust. TIILIS.

sign (1) - (:I ) = sign T.

Now, analyze the first-ortlcr coliclitioti with rcspect to i~ttrafirm trade. Call intrnfirm trade efficient if the t \ \ 1 o marginal costs ;ire ccluul in ecluilihrium. On thc otliel. hiuicl. term i t pcr\,crsc (rcstrictcd) i t ' t l ~ c MNIZ's profits arc ~ilaxiniizetl at a Icvcl 01' itltcr~ii~l tr;~~is;~ctic)~ls S L I ~ I I tIi;it tllc cx~)ort.i~lg l l ~ l i t IS tlic higher (Io\ver) tllitrgillitl cust. Whether intral'ir~ll trailc is el'l'icicnt, pcrvcrsc or rcstrictccl is csanlinctl below by stating the partial cleri\:ati\'e 01'4 with respect to 111 ;IS:

\\lhilc the scconcl orcler (sufficient) conelition \\ l i t11 respect lo 111 is:

If clifferences in procluction functions ;inel i~ ip~~t - rcn t i t l s it1 the two countries iWe

" The first ortlcr conditions with respect 10 .A, a ~ l t l .v, give the I'amiliar contlition t l i ; ~ l

marginal cost cql~nls tnargitlal rc\lcnilc ill cacll coulttry. A clt;tnge i n In first cIi;~ligcs 1.,. To tnointnin marginal cost and m;~rginal reveltile rcluality. . r , ~tit~st c11i111ge. Furtlier. 111 ;IISO ;~t't'ects the first term on the right Iiantl side of (15). and licncc I J ; ancl from (17). 1, also affects r r r .

Thi~s. 111. . r l , sl atid 1) are all deterrni~~ctl si~nitltancously.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

26 CHANDER KANT

allowed, then (18) requires that long-run marginal costs must be increasing at equilibrium in at least one country. Thus, one set of necessary and sufficient conditions under which the M N F will find it optimum to horizontally integrate in the two countries and undertake intrafirni trade is that the home country has decreasing or constant costs (increasing or constant returns to scale) at equilibrium, the foreign country has increasing costs (decreasing returns to scale), and the k-proportion of after-tax rate of increase in ~narginal cost in the foreign country is greater than the after-tax rate of decrease (if any) in marginal cost in the home country.

On the other hand, if production functions and input-rentals are identical in the two countries, then ( 1 8) rules out increasing or constant returns to scale (decreasing or constant costs) at equilibrium in both the countries. Helprnan (1984) and Markusen (1984) assume that inputs like R & D, advert is ing, market ing, dis t r ibut ion, management, finance and organization used in one plant or facility serve at zero marginal cost additional plantslfacilities, and thereby give rise to increasing returns to scale. This assumption is likely to be quite valid for R & D. But inputs like advertising, marketing, distribution, management, finance and organization cannot serve additional plantlfacility located in a possibly distant country (with likely different language, culture, legal system and industrial relations) a t zero marginal cost.' Further, if a l l s tages of product ion before final sale (in particular, if transportation and retailing costs) are included, decreasing returns to scale (increasing costs) from "production" (as interpreted above) of a product/vari~ty from a single facility globally are also quite likely.*"

Now we analyze equation (17). Consider again the ITP case first. In view of (16), both increase in the transfer price and internal real transactions shift profits from the foreign to the home country [and the first term in (17) is unambiguously positive]. Therefore, lower marginal cost in the exporting country, i.e. a negative (c, - c2 ), is not

a necessary condition (although it is a sufficient condition) for the MNF to export from there (for $,,, to be positive initially), ant1 the M N F finds such trade profitable even if c, - c,. Further, a necessary condition for the M N F to achieve its trading

' Markusen (1984) recognizes that increasing returns to scale may be weak, and horizontal integration takes place due to factor-intensity differences.

T r a d e theory generally considers triunsport costs and tariffs together as if they have identical effects on comparative advantage. But marginal tarilT costs remain constant while marginal transport costs are likely to increasc as a single productlvariety of a product is transported to ever farther consunling centers. In addition. retailing costs are incurred at numerous final sale points distributed all ovcr the globe; and are likely to be significantly positive in non-purely competitive markets. Then. transport and retailing costs can be considered as natural barriers to increasing rctul-11s to scale of "producing" a productlvariety from a single plantlfacility globally.

' Horstmann and Markusen ( 1 989) also use transportotion/tariffnltiriff costs to explain horizontal integration. In addition. Ethier and Horn (1990) suggest that increasing costs of "managerial control", and possibly increasing costs of interl'ncing with host country with different language. culture. legal system and industrial relations can lead to horizontal integration.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

INTRAFIRM TRADE AND TARIFFS

equilibrium is:

(c, - c2) = ( y - c,)TIT,,k,

that is, only if internal real transactions are perverse. In the TTP case being discussed, the minority local ownership effect dominates the deferral effect. Clearly, in this case, minority local ownership over the subsidiary encourages intrafirni trade (by either increasing home production or decreasing foreign production or both) beyond the efficient level. Exporting from country 1 wastes real resources while, at the same time, increasing the global profit tax burden on the MNF. But the shift of pure profits from local shareholders in the foreign country through intrafirm trade more than compensates the MNF for both these negative effects. Lastly, the higher are the transfer price and the proportion, q, of foreign profits repatriated, and the lower are the proportion, k, of the foreign subsidiary owned by the MNF, and the statutory profit tax rate differential, ( t , - r,), the greater is the extent of perverse internal real transactions.

Consider now the negative T case. As shown above, in this case .the M N F s optimum transfer price is ,snlaller than c,, and results similar to the ITP case follow. In this case, both decrease in the transfer price and intrafirm trade [by creating negative net export revenue, - (p - c,), in the home country] shift profits from the home to the foreign country. But, the gain due to deferral from home taxation at the higher home rate of profits shifted to the foreign country dominates both the resulting shift of pure profits to foreign shareholders and the increased overall production costs. Further, the lower are the transfer price and q, and the higher are k and ( t , - t2) , the greater is the extent of perverse intrafirm trade.

Now consider that internal real transactions'are in the other direction, i.e. from the foreign subsidiary to the parent firm, In that case, the expression for the MNF's global net profit function remains the same as (6). But, the individual pure profit functions, n, and n2, in (6) are now defined by (1') and (27, respectively [rather than by (1) and (2) as earlier]. The partial derivative of n with respect to p is now:

TI, = - (T - Glk)t711 = - Ttn' = T 'tn', where

If T' is positive, i.e., if T is negative and the deferral effect is greater than the minority local ownership effect, the MNF increases its transfer price on its exports (now) from the foreign subsidiary so as to shift profits to the foreign country. Vice versa when T' is negative. The definition and properties of the probability of penalty function, the expression of the objective function and the first order

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

2 8 CHANDER K A N T

conditions with respect to s , and s , are the same as before. The expression and the analysis for $,, is siniilar with T ' r ) ~ ' replacing Till. The arm's length price now equals c2, and

sign ( p - c,) = sign T '.

Further,

and the discussion in the two paragraphs following (18) still holds. Lastly. restate $,,,.

so that at the trading cquilibriun~

Similar conclusions as above i~bout pcrversc internal real tri~nsactions hold. Exporting intrafirm from country 2 wastes real resources. But, in the ITP case, reduction in the MNF's global profit tax burtlcn clue to shift of profits to the lower- tax foreign country more than compensates the MNF both for the liiglicr overall production costs and tlie shift of profits from the parent firni to foreign local shareholders. Vice versa i n the DTP case. Further, in the ITP (DTP) case, the higher (lower) are I), k and ( I , - I , ) ; ancl tlic lower (higher) is (1, the greater is the extent of perverse intrafirm trade.

It may be noted that the parent firm nerd not be wholly ownetl by the MNF. Sirnilar results follow when tlie MNF cloes not own the parent firm fully but its controlling owners own a greatcr sliiirc ol' the parent firm than of the subsidiary. Then, the fc)llowing proposition summarizes the above discussion:

PROPOSITION 1: ( A ) 1Vlre11 tlrc, MNF tlc[c,r.\ ~-eptrtrio/iorr o f ( . so l rrr o r 1111 of) i t s foreig11 pr(~f i t c111i1 \ I : / I ~ I I its ~ o ~ r t r u / / i ~ ~ , y O I I , I I ~ , I . S o 1 1 . 1 1 ( I , q ~ ~ ~ l t c ~ r . s / I ~ I ~ ~ J (!/' 1 1 1 ~ ~ L I I . C , I I I

fir111 t1rr111 o f t / ~ i ~ f i ) r i ~ i , q ~ ~ . s I I / ) . s ~ ~ / ~ ~ I I ~ J , / o \ i ~ , 1 . I I I [ I I . ~ ~ I I ~ I / ( ~ ) . v t i l l 1/11> l,.i-porti~l~y c.01111t1.y i,s lrot rr rlece.s.srll:\; co~lditio~r fi)r 111c MNF t o ~r~rtlo-ttrke ~ I I / C I . I I ~ I / 1.1.ol t~-t~~r.vtr~~tio~r.s, r111tl

i1.s 1)rofit.s c ~ r i ~ I I I ~ I . V ~ I ~ I ~ Z ~ Y / o111,v $ tlro 1,.~/101~fi11g ( . ~ I I I I I I . ~ 11l1.s 11113 11i~qIlc~r 111clrgi11c11 ( 'OSt .

( B ) Fllrt11e1; i l l 1/10 ITP ccl.se, u~11e11 / / r e 1101110 [ / i ) l . e i ~ ~ l ] ( Y ) I ~ I I I I - , ~ i.s tl1t) c.v/)ortill,q

c o ~ o r ~ r y , tlri~ Iri,y/rer ~ r c > t/ri> t~.cl~l.sfc>r / ) I .~( ,L , 01111 (1 [ ~ I I I A k [ I I I ~ I ( 1 , - /:)I, (11tc1 t 1 1 ~

/u\~:er rlre k ( I I I ~ ( I , - 1:) [i.s q ] . / / l c , ty~.ci~tc,r i.s 1/11, 1>.~tt)11t of /~c,~.~~l,r.so I I I I ~ ( ( / ; ~ I I I

trtrrli~. 011 tllr o/lrrr h o ~ l r l , i l l tlrc, LITP g:,rr.siJ. \i./lc,rr r.1-po1.t.v ol.igi11trtr /j.olrr tllr I I U I J I ~ [forei<y/r] ( . O I ~ I I I I . ~ . t11c L , . V / ~ I I / 01' tI1e per~>e~..ve i11ter11c11 re( / / I ~ L ~ I I . S [ I ( ~ I ~ O I I . S 1,s

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

INTRAFIRM TRADE AND TARIFFS 29

greater the lower ure the trunsfer price ur7d q [and k and (t, - /?)I, o t ~ d higher are k nncl (t, - t,) [ i s q1.I'

B. Whole Ownership of the Foreign Subsidiary:

Consider now that the foreign subsidiary is wholly-owned by the MNF (with t, > t, and deferral). In that case, k = I , the minority foreign ownership effect vanishes and T is clearly negative. When the parent firm (foreign subsidiary) is the exporting unit, the MNF is now only interested in decreasing (increasing) its transfer price. In either case, profits are shifted to the lower tax foreign country.

The conclusions about perverse intrafirm trade stated in Proposition 1 A hold even if, as now, the extent of ownership by the controlling owners in the two units of the MNF is the same (except that now internal real transactions shift profits within the MNF only rather than also fromlto foreign shareholders). Proposition I B is modified as follows: when exports originate from the home (foreign) country the extent of the perverse intrafirni trade is greater the lower (higher) is the transfer price, tlie lower is q, and the higher is (ti - t,).

As an illustration of this phenomenon, consider that U.K. perniits deferral from its taxation of non-repatriated foreign profits, and British Petroleuni exports petroleum products to its fully-owned subsidiary in lower-tax Singapore. Under these conditions, British Petroleun~ decreases the transfer price (on this internal trade), pure profits are shifted to Singapore, and marginal cost of production is higher in U.K. at equilibrium. Clearly, gain due to deferral from U.K.'s taxation of profits shifted to Singapore through intrafirm trade conipensates British. Petroleum for exporting intrafirm from its higher marginal cost unit.

4. CASES WITHOUT DEFERRAL

This section presents different results for the other two cases. These are: ( B ) ti = t , which results when I, 2 t, and either the MNF repatriates its foreign profits to tlie home country fully (for t , = 1,) or the home country does not permit deferral from home taxation of non-repatriated foreign profits (for I , > t,); ancl (C) t ; = t, which occurs when t , < 1,. Give a wider meaning to (1: let (1 now represent tlie proportion of foreign profits taxable at home (whether repatriated or not). Clearly, cl = 1 for both

"' Although the model pl.esented above envisages the production of only one good (whether final or intermediate) of service, identical results follow i f the M N F produces both the intermediate and the final good or the service 2nd engages in internal real transaction of the product which is produced in both the countries; except that when both the intermediate and the final good are produced in both the countries and the final gootl is expol.ted' intl.afil.rn. for the MNF to achieve its optimum. the coniposite tnarginal cost (i.e. the sun1 of marginal costs of producing the intermediate and the final good) must be higher in the exporting country.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

CHANDER KANT

(B) and (C).

A. Case t , 2 t,with No Deferral

In this case, the effective statutory profit tax rate equals t , for profits earned in either country. Now the deferral effect vanishes. Equate q = I and substitute t , for t,, in various expressions in Section 3. When the foreign subsidiary has minority local ownership, Tequals ( I - t ,) ( I - k), and is unambiguously positive. In that case, when the exporting unit is the parent firm (foreign subsidiary), the MNF increases (decrkases) its transfer price to a level higher '(lower) than the marginal cost in the home (foreign) country. Now, the shift of pure profits to the MNF from foreign shareholders (irrespective of the direction of intrafirm trade) more than compensates the MNF for exporting from the higher marginal cost country; and the lower is k, the greater is the extent of the perverse intrafirm trade.

For example, consider that the profit tax rates in Germany and France are equal; and Siemens, Germany exports air-traffic management systems to its not-wholly owned subsidiary in France. Then, Siemens increases its transfer price; and the the shift of pure profits from local shareholders in France through intrafiml trade more than compensates Siemens for exporting from its higher marginal cost German unit.

On the other hand, when the foreign subsidiary is wholly owned by the MNF, the foreign local ownership effect also disappears, T equals zero, and the MNF would have no reason to engage in transfer pricing. In fact, in view of the probability of the penalty function with positive expected value, it chooses the a m ' s length transfer price, ;. Further, irrespective of the direction of internal real transactions, marginal cost must be lower for such exports to originate in a country, and thc MNF achieves its optimum only if the two marginal costs are equal; i.e., only if intrafirm trade is efficient.

The last conclusion, and those from Section 3.B above can now be combined. The last paragraph above shows that internal real transactions are efficient when k = q = I; while Section 3.B demonstrates that deferral (with k = I ) results in perverse intrafirm trade. Thus, we can state:

PROPOSITION 2: Corlsider that the foreign subsidiary is ivhollv-otvnecl by the MNF: Then. yerniittirig deferrril fiorn hotlie tir.ratior~ qf rlor7-repcrtr-iiitrcl foreigr~ pr4t .s. changes the nrrtlrrtJ ofintrr$rm trride fro171 efllcier~t to perverse.

B. Case ti < t2

In this case, we substitute t , for t , , while, as before, we equate ci to 1. Again, when exports originate from the home (foreign) country, the MNF only increases (decreases) its transfer price, and profits are shifted to the home country in either case. Proposition 1A continues to hold and is, in fact, strengthened because now global

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

. .

INTRAFIRM TRADE AND TARIFFS

profit tax savings (since the home country has the lower tax rate) also compensate the MNF for exporting from the higher marginal cost unit. In addition, we have that the greater is (t, - t,), and the lower, is k, the.greater is the extent of the perverse internal. real transactions. Lastly, even.if the foreign subsidiary is wholly owned by the MNF (and there is no deferral), we have perverse intrafirm trade.

As an example, consider the export of cars by Ford, U.S.A. to, say, its wholly- owned subsidiary in the highey-tax U.K: Ford increases its transfer price; and perverse internal real transactions are explained by the shift of pure profit from Ford's wholly-owned unit in U.K. to Ford, U.S.A. ' '

Box 1 below summarizes the conclusions about perverse intrafirm trade.

Box 1. Nature of Intrafirm Trade

Minority Ownership

Whole Ownership

We also have the following generalization of Proposition 1A; and state Proposition 3 (which is somewhat similar to Proposition I B):

PROPOSITlON 1A': The MNF need not have lower tnurgirzal cost in the exportrng country to export intrcrfirm (it finds internal real tran.sactions profitable even if the two marginal costs are equal), and it acllieves its rnaxirn~rm profits on(v when the exporting country heis the higher rnarginal cost. The only exception to these results is when t , 2 t,, the foreign subsidiciry is wholly o~vnecl by the MNF nricl the home country does not permit deferrer1 fro111 hori~e ta.ration of non-repatriated foreigri p rofits.

PROPOSlTlON 3: The extent of perverse internal trcinscictions is greciter; the grecrter the optimum transfer price diverges froin the rncirginal cost in the exporting country When t , 2 t,, 0 I q < 1 arid k = I , this extent is greciter the lower is the proportion, q.

tl < t2 (4 = I )

Perverse

Perverse

t , 2 t2

Deferral

Perverse

Perverse

NO Deferral (q = 1 )

Perverse

Efficient

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

3 2 CHANDER KANT

offorsigrz profits repcrtricrted by tlre MNF: crtici tlre higher is ( I , - I?) . Wlzerr (1 = I utlcl I /2 < k < I , it is greater the lower i.s tlre yroportiot~, k , of rlze fotzigtz .sub.sir/icrrjl owrred bv the MNF: Lrrstlv, ,vlrerr t2 > I , , I / i s greater tile gr-eoter i s ( t , - t , ) .

5. TARIFFS AND QUOTAS

Consider now tariffs ant1 quotas as policies to counteract perverse intrafirm trade. First consider tariffs. Then compare quotas to tariffs. To give a motivation for introducing tariffs in our model, we first show the effect of tariffs both when the foreign subsidiary is \vliolly owned by the R4NF and either there is full repatriation or there is no deferral, i.e., k = (1 = I . First consider t , 2 t,. Recall in this case t; = 1,. For thc casc of cxports originating from the parent firm, we have

where xi is the MNF's global net profit function under the stated conditions ancl T is . the N L I ~ ~ 1 1 o r ~ ~ 1 1 1 tariff rate on imports in country 2. The M N F decreases its transfer

price so as to save on tariff payments. The clefinition and properties of the probability of the penalty function reniain unchangccl. Lct $ represent the objective function under these conditions. Then,

Siniilirr exprcssion can be derived when exports go from the foreign subsicliary to the parent lirm iuicl the home country has a tkrriff on imports. In either citse. tlie exporting country must Iiave lower marginal cost both Ibr the MNF to export flow there, and at tlie trading ecluilibrium ( i t . . we have rcstrictccl internal real transactions). Cleiuly. restricted intrafirm trade is smaller than efficient int~.afirm trade. Tliis suggests that imposition of tariffs in a situation of perverse intcrnal rc;rI trirns;tctio~i~ (i.e. when tlic volume of intrafirm trade is greater than the "efficient" Icvcl) may clirliinate sucli perverse trirde. Tliis is anirlyzccl helo\\) first for tlic clclkr~;tl casc and then li)r tlic cases \vitlio~~t cleferrrrl.

t, > tz with Deferral

Let rr represent thc MNF'x glohal net prol'it function in tlie prcscncc of all of the following: tariffs, minority forcign o\\inership and deferral provisions. Define r . = ( 1 + r). ancl redefine gross profits in country 1 as:

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

INTRAFIRM TRADE AND TARIFFS 33

rr = T,rl + T2,k7r;,

= (T - T,,kr)tn = (7; - T,,kr*)tn ,

where T is defined by (9) above. From (7) ,

(since t, > t,). Then, from (8"), the MNF decreases its transfer price when kz' is either greater than or equal to one. When it is smaller, its.exact transfer pricing strategy is ambiguous. The objective functign is now termed qY, and its partial derivative with respect to m can be stated as:

Similar expression can be derived when internal real transactions are in the other direction and the home country has a tariff on imports. Irrespective of the direction of intrafirm trade, for both the ITP and the DTP cases, a positive sign of this partial derivative has no implications for the signs of both (c,r* - c,) and (c, - c,): in spite of tariffs, intrafirni trade may be perverse. Identical conclusions hold for the trading equilibrium.

Cases Without Deferral

Conditions for ITP and DTP are different in these cases. In case (B), ti = t , , and the equation corresponding to (8") is:

( I - k)-in (8"') gives the effect of minority foreign ownership on nr (and is positive) while (- ks) gives that of tariffs (and is negative). I f the former effect is greater (smaller) than the absolute value of the latter, i.e. k f is smaller (greater) than 1, the MNF increases (decreases) its transfer price.

In case (C), t ; = t 2 ; andthe partial derivative of lr with respect to p, obtained by replacir~g T,, by T, in equation (8'7, is:

Now, the MNF increases its transfer price when k f is either less than or equal to one.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

34 CHANDEK KANI'

When i t is greater, i t may either increase o r clecrease its 11. However, conclusions about pcr\icrsc internal real transactions i n both cases (B)

ancl (C) - i.e.. when (1 = I -are identical to those prcsentecl l i ~ r the clefcrral wi th tariffs

case ithove. Then. we have the follo\ving proposition:

PROPOSITION 4: T/IO i ~ ~ ~ / ~ o . v i / i o ~ ~ (!f / t r r i [ f i OI I ~ / I I I . ~ I J ~ I . I I I I I I I / ) O ~ I . Y /7>, OI IO r r / ~ i l 1 /10

M N F , / i . o ~ r i 111r o l l r c , ~ . c . t r ~ r I c r r t l l o 111e r.t,.sfor.rr/iorl of e[j"i(: ic~lt I I I ~ ~ , I . I ~ ~ I / 1.etr1 t ~ . t r ~ r . s t r c ~ t i o / r . r

f ro111 1 1 1 ~ i 1 1 i t i t 1 1 . s i I ~ i t r l i o ~ ~ o / ' ) I * / ~ ~ , I I J 1 1 1 0 , ~ t11-c, I ) ~ I I , ~ ~ I - . s P

No\v we compare import quotas to tariffs. Although the effect o f an import quota o n the volume ill' intraf i rm tr i~dc, ancl on wlicthcr cl'l'icient internal real triuisactions arc restoreel, is the same as that o f t i~r i f fs . thc I 'ollowing consiclcrations mitkc quotas an

infer ior policy. First consider who gets the quota rents. II'quota rights are given away to importers or i f they i ~ r c auctionccl. then imporrers o r the impor t ing country's

government. rcspcctively. get the clLlota rcnts. But, i f thcy arc given away to tl ic export ing country las irndcl- the widely prev i~ lcnt \,oluntury export restraint ( V E K ) agrecmcntsl. then quota rents accrue to exporters. In that case, import quotas arc more costly to the import ing country than ccluivalcnt tariffs.

Further, i n our muclcl, the un i t i n tl ic impor t ing co i in t ry hiis market power.

Under such situation. as shown hy H l i i i ~ w i i t i ( 1965). this unit \\.ill cliargc higher

prices i ind proclucc Ichs i~ncler quota protection t l i i ~ n unclcr ta r i f f protection. Lastly. tar i f fs are easier to acl~ninistcr than cluotas. Icacl to Ichs C ~ O ~ ~ L I P I I O I ~ . IohI>yil ig ;111cl

c h i ~ s i ~ i g 01' the \a luahlc cluota rcnts ant1 act at rhe source 01' this pcl.\,c.rse illtr;il'irm

triiclc."

" ' r l . : ~it'l's : i ~ , t : it IIIC sok~~.ec 01 ' II~~\,L\I.>' ~IIII.:I~II.III t~.:~clc i l l IIIL, l~oI lo\vi~~; \v:iy. ~ c I ~ \ c I . ~ L ~ ~ ~ i t r ; ~ ( i r ~ ~ i II.;ICIC~ ;iri\c\ ITC'C.;III~C' II1c MNF's i11tc1.11:11 I.L';II II.:III~:IL.I~~II~ sIi111 ~ I I ~ C prol' i~\ 1'1.0111 L.IIIIL.~ IIIC Ii)rc1;11 sl~;~rcl~i~l[lcrx 111. l1lt, I~igI le~. IXS I'OI.LY;I~ C,II~IIIII.~. (11.. IIIICILY ~lcl 'c~.r:~l I.IXIII~ IIOIIIC I:I~:I~IOII

01' ~~o~~-rc~>:~lr i ; i tccl l'(>rcig11 1 ~ 1 0 l ~ i t . 1'1.0111 IIIC III;~IL~I, [:I\ IIOIIIC L.O(IIIII-~. DLIC t o :III! IIIIC f { ) r IIIO~C j

01' Ihe\c g:ri~lc. ~ntr;rl'ir~n rrarlc 1s inc,reasccl hcy~l r~r l II~C cl'l'icicnt Ic\.el. 'I'llc iilc~.c.asc in 1l1c h1NF's por~icln olcosrs ill tile i ~ n l ) o ~ - t l ~ ~ g L.OIIIIII~ (luc 10 t:lriI'l\ L.OIIII~~I.-;ICI\ :ill ~IIc"(c g : l i ~ ~ \ 10 111~

h4NF. L:~stly. a suhsicly o n such c.\pol-ts \\ors~*n\ I '~S\.~I ' \C~ ii~tt.; lI '~~.n~ II.:ICIC hcc;i~~sc i~ ~\~I;II~CIS 111rra1'il.m esl,orts t'u~.thcr hcyoncl rhc cl'l'~c.icn~ Ic\.cl.

Tlicsc rcsulta i~nply II~:I~ link;~gcs h c t \ \ ~ c c ~ ~ L~OIIIC~IIC cos~\ :IIICI cli~.cctiol~ 01' i~ l t l - ;~hr~ l l tr;idt' hrcakdown eit1ic.r \\,IICII the MNI- c)l~cr;~tcx u ~ ~ c l e ~ il~.I'crr:~I I)ro\,l\loli\. or i t ~loes 1101 1'~1Ily 0 \ ~ 1 1 i t s

l'orciyn s~~h>iil iary: :IIILI C~IICIUS~~I~S cIr:~wii OII 1111' :1x~11111~~11011 01' :I I'k~ll!. (I\\ IICC~ suhxicI~;~ry or tlic ;1hhCnce ol'prol' i~ I:IS CIII~I~~~C~II~I:IIS C:IIIIIO~ he ~ I . \ . \ I I I I I , , ~ 10 c \ ~ c ~ i ~ l 10 111c \~oI-ILI 01' ~ ~ ~ l ~ s i ~ l i ; ~ ~ ~ i c s ivi111 111111orily IOI.~I;II o \ \ , I I~ I .> I~I~ 01. 01' prol' i~ I;I\ LI~~'I'L~I~L~I~II:II\.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

INTRAFIKM TRADE AND 'I'ARIFFS 35

efficient to perverse even when the foreign subsidiary is fully-owned by the MNF. Further, if tlie foreign subsidiary has minority local ownership and the MNF engages in transfer pricing, its intrafirm exports are always from the country witli the higher marginal cost." 'Since most tax codes have deferral provisions, ancl since less than wholly-owned subsidiaries constitute a large ant1 increasing proportion of total MNF activities, perverse intrafirm trade cannot be ignored.

Ethier (1986) has recently suggested that internalization is the only one of the three key elements-ownership advantages, locational consit lerations and internalization of international transactions - explaining foreign dircct investment not already incorporated into trade theory. This paper suggests that shifting pure profits from local portfolio eqyity investors in the Coreign country or i~clupting overall tax-niininiisation stratcgies can tlicmsclves bc powei-ful niotivcs I'or undertaking internal transactions; and i t draws real implications of such internal transactions. Hcllcincr and Lnvcrgne ( 1979) ancl t.lelpm:in and Krugman ( 1985) pose the question (but clo not provide ;In answer) whetlicr intr:~firni trade differs significantly from that between unrelated buyers and sellers. This paper gives iin affirmative answer to that clucstion.

This paper also points out sumc policy clilcmmas 1i)r governments in tlie world populntecl by MNF's. We have shown that tarifl's can restore el'ficicnt intr:~firm tratlc from a situation where it is perverse otlicrwisc: ant1 stnterl \\lliy tariffs arc superior to quotas. An altcrnotive COLIICI bc to ensure that the M N F al\v;~ys charges a triuisli.r pricc equal to tlie arm's length pricc. Although succcss in 1i)rcing the transti-r pricc to equal tlic marginal cost ol' the exporting unit, eliminates per\Jcrsc internill real transactions. such s~lcccss will be difficult to achieve: i t is very h;u-tI I'or governments .to cleterminc the i~rm's Icngtli price. . For cxnmple. the U.S. has onc 01' tlic riiost stril1,gcnt regi~l:~tions (Section 482 of tlic tax coclc) o n transl.cr pricing, Still. the U.S. Internal Revenue Service is fighting a losing hattlc to cnl)rcc complia~isc \villi tlic\c regulations.'.'

Lastly, although our tliscussion Iios hecn in terms of a rni~ltin:~tion:~l I'irn~. (lie part of our :ill:~lysis above pertaining to less than wholly ownccl suhsicliary nl>plics to any (purely) clomcstic firm which lii~s two or more units witli the con troll in^ owners having :I greater ownership in one u n i t than the othcr. With the rcccrlt surge of mergers nncl acquisitions i n many coillltries. tllcrc are likely to he considcrohlc inccritivcs to engage in internal tr:~nsac.tiolis \\~liicli violntc real c,o\t :~cl\~iuitagcs.

Hatra. K . N . ancl Hatlar. .I.. "Theory of the Multinatioli;~I Firm: I'iscll \tcrsus I=lo;~ting E,\c.li:i~~gc R;i~es..' O . ? / i 1 1 t 1 E(.OII(IIII~(, / - ' ( 1 ~ ) ( ~ 1 ~ ~ . Ic11y 1070. 25s-00.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

36 CHANDER KANT

and Ramachandran, R., "Multinational Firms and the Theory of International Trade and Investment," Americcrr~ Ecorlor~ric Kevievv, June 1980, 278-90.

Bhagwati, J., "On the Equivalence of Tariffs and Quotas," in Robert E. Baldwin, et al. eds, Trcrde, Groivtl~ crrld the Baltrnce of Payments, Chicago: Rand McNally, 1965.

Coopers and Lybrand, , Ir~ter/~crtiot~nl Tcrs Sirmrncrries, John Wiley: New York, 1993. Eden, L., "Vertically Integrated Multinationals: A Microeconomic Analysis," The

Ccrt~adiai~ Joi~rr~crl of Economics, August 1978, 534-46. Ethier, W. J., "The Multinational Firm," The Qirarterly Joirrr l~l of E C O I I O I ~ ~ C S ,

November 1986, 805-33. and Horn, H., "Managerial Control of International Firms and Ratterns

of Direct Investment," Jo~rrr~nl of Ir~terr~ntiorrcrl Ecor~oniics, February 1990, 25-45.

Goldberg, F. T. Jr., Statement before the Cor~rr~ritter UI I Oversiglrt, House Coniniittee on Ways and Means, U.S. Congress, July 10, 1990.

Gresik, T. A. and Nelson, D. R., "Incentive Compatible Regulation of a Foreign -owned Subsidiary," Jo~rrrrrrl of lrrterrlcrtiot~trl Econor~rics, forthcoming.

Heck, P. G., Stcrter~ler~t before the Cor~~r~rittee or1 Oiwrsight, House Committee on Ways and Means, U.S. Congress, July 10, 1990.

Helleiner, G. K. and Lavergne, R., "lntra-firm Trade and Industrial Exports to the United States," 0,~forcl Birlletir~ of Ecorrorlrics trrrcl Stntistics, November 1979, 297- 31 1 .

Helpman, E., "A Simple Theory of International Trade with Multinational Corporations," Jo~rrtlul qf Politice11 Ecor~ot~r?:. June 1984, 45 1-7 1 .

and Krugman, P. R., M o ~ k ~ t Strirc,trrrc) crrlet Forpig11 Trcrde, M.I.T.: Cambridge. Massachusetts, 1985.

Horst, T., "The Theory of the Multinational Firm: Optimal Behavior under Different Tariff and Tax Rates," Jorrrrral of Politico1 Ecorror~~>~, SeptemberlOctober 197 1 , 1059-72.

Horstniann, I . J. and Markusen, J. R.. "Strategic In\lesiments and the Development of Multinationals," Irrtrrrrcrtioricrl Ecor~ornic Koisieic:, February 1987, 109- 12 1 .

, "Firm-specific Assets and the Gains from Direct Foreign Investment," Ecorlornictr, February 1989, 4 1-48.

Itagaki, T., "The Theory of the Multinational Firm under Exchange Rate Uncertainty," Ccrr~aclirrr~ .loirrrrnl of Ecorlorlric.~, May 198 1. 276-97.

Kant, C., "Endogenous Transfer pricing and the Effects of Uncertain Regulation," Jol l r l l~ / of lrrter1~tr1ior1crl E ~ O I I O I I I ~ C . ~ , February 1988, 147-57.

, "Foreign Subsidiary, Transfer Pricing and Tariffs." Soutl~err~ Ecorlorllic

.loirrrrcrl. July 1988. 162-70. . "Multinational Firms ~uid Go\lernnient Ke\/enues." Jo1rr11t11 qf Plrblic

Ecor~orrrics, July 1990, 137-47. Katrak, H.. "Multinational Monopolies and Commercial Policy." O.~:fortl G~orroriric

P(rpc>r.s. July 1977, 283-9 I .

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4

INTRAFIRM TRADE A N D TARIFFS 37

, "Multinational Monopolies and Monopoly Regulation," Osforcl Economic Papers, November 1980,453-66.

Lee, S., "lnternational Equity Markets and Trade Policy," Jorrr~lcrl qf I~rrerrrcrtion~l Econorr~ics, August 1990, 173-84.

Little, J. S., "lntra-Firm Trade: An Update," New Englcrr~cl Econornic Review, March 1987,46-5 1.

Markusen, J. R., "Multinationals, Multi-Plant Economies, and Gains Froni Trade," Jo~rrr~crl of lr~terr~crtior~~l Ecorlo~nics, May 1984, 205-26.

Miyagiwa, .K., "International Shareholdings and Strategic Export Policy, lrrterrrcrtior~crl Ecorlo~r~ic Joirrrlnl, Autumn 1992, 37-42.

Stoughton, N. M. and Talmor, E., "A Mechanism Design Approach to Transfer Pricing by the Multinational Firm," Eirroye~~~l Ecorrornic Review, forthconling.

Mcrilir~g A~1clres.s: Professor Chciiider Kcr~rr, W Parrl Still~rrn~r Scliool of Blrsirre.ss. Setorl Hcrll U ~ r i ~ ~ e r s i ~ Soirtlr Orctrrge, New Jersc~y 07079-2692, U.S.A.

Dow

nloa

ded

by [

Was

hing

ton

Stat

e U

nive

rsity

Lib

rari

es ]

at 0

1:22

25

Oct

ober

201

4