• 検索結果がありません。

Chapter 6

International Separation of Ownership and Management

6.1 Introduction

Chapters 4 and 5 concern the ownership and management structures of the firms. To the best of my knowledge, no paper has examined the traditional strategic export policies in the presence of both international cross shareholding and separation of ownership and man-agement. This chapter combines the analyses in the previous two chapters and discusses how the strategic subsidization incentives are affected by managerial delegation when the shares of the firms are internationally owned by the residents of both countries, i.e., in the presence of international separation of ownership and management. This chapter at-tempts to study how the complexity of managerial decision process and cross shareholding structure alter the standard welfare implication of strategic export promotion policies.

The works related to this chapter is summarized in the following table.

decisions, respectively. Section 5 compares the equilibrium results and their values in Chapters 4 and 5. Section 6 shows two special cases of symmetric and partial ownership structures. The concluding remarks are summed in section 7.

domestic owner’s subsidy and leads to fiercer competition in the output market. The domestic firm’s profit gain shrinks and the dividend given to the foreign firm decreases.

The second part (1−σi)xi shows thesubsidy outflow effect, which has the same expression as in the case without managerial delegation.

The third part (1 σj)xjP(X)rSiD∂S∂seDi

i shows the dividend suppression effect through managerial delegation. Further, note that riDS ∂S∂seDi

i < ∂x∂seDi

i , and hence, man-agerial delegation scales down the decrease in the dividend from the shared foreign firm.

The above three negative parts weaken subsidy incentives in the presence of cross-country shareholding.

Denote RiE(sj;σ) as country i’s reaction function, where superscript E represents the values under international separation of ownership and management.

An increase in the domestic residents’ ownership share over the domestic firm (or the foreign residents’ ownership share over the foreign firm) strengthens both countries’ subsidy incentives, leading to an increase in the optimal export subsidy. The results follow from

RiEσi(sj;σ) = ∂RiE(sj;σ)

∂σi =−∂2Wi(s;σ)/∂σi∂si

2Wi(s;σ)/∂s2i = ∂πieD(s)/∂si

2Wi(s;σ)/∂s2i >0, RiEσ

j(sj;σ) = ∂RiE(sj;σ)

∂σj

=−∂2Wi(s;σ)/∂σj∂si

2Wi(s;σ)/∂s2i = ∂πeDj (s)/∂si

2Wi(s;σ)/∂s2i >0, where use was made of (5-16) and (5-17).

6.3 Effects of Cross-Country Shareholding

6.3.1 Equilibrium Government Subsidy

Denote sEi (σ) as the equilibrium government’s subsidy of country i. The full-game Nash equilibrium subsidy profile is thus defined as a solution to

sEi (σ) =RiE(sEj (σ);σ) (i, j = 1,2;j ̸=i).

Under the symmetric cost conditions, that ci =cj =c,

sEi (σ) = (16σi+ 13σj10σiσj18)(a−c)

2[2411(σi+σj) + 5σiσj] . (6-3) The optimal subsidy rate is dependent on the cross shareholding structure (σi, σj). From Assumption 4.1, the denominator in (6-3) is positive, i.e., 2411(σi+σj) + 5σiσj >0, and hence sEi (σ) is positive if and only ifσj > 181316σ10σi

i. Figure 6.1 illustrates (σ1, σ2) for which country 1’s government finds zero subsidy optimal with origin (0.5,0.5). In view of Figure 6.1, when the domestic shares of both firms are large enough, each country’s government subsidizes the firm; otherwise, export tax is the optimal policy.

Without cross-country shareholding, Das (1997) showed that sEi (1,1) = sDi = 141(a c) > 0 in (5-20), and each government always has a positive incentive to subsidize its exports. In the presence of cross-country shareholding, (6-3) shows that ∂sEi∂σij)

k >0(k = i, j), which yields

sEii, σj)< sEi (1,1).

Proposition 6.1. Given managerial delegation, the presence of cross-country shareholding weakens both countries’ subsidization incentives, i.e., sEii, σj)< sEi (1,1).

6.3.2 Equilibrium Owner’s Subsidy Equivalent

In view of (5-10), the owner’s subsidy equivalent is always positive in the duopolistic market. Solving for the owner’s subsidy equivalent in the equilibrium yields

dEi (σ) =deDi (sEi (σ), sEj (σ)) = 3(2−σj)(a−c)

2[2411(σi+σj) + 5σiσj] >0. (6-4) It is easy to show that ∂dEi(σ)

∂σi > 0. Increasing domestic ownership strengthens the domestic owner’s subsidy incentives.

Differentiating (6-4) with σj yields

∂dEi (σ)

∂σj = 3(2−σi)

2[2411(σi+σj) + 5σiσj]2(a−c)<0.

The above equation is equivalent to ∂dEi(σ)

∂(1σj) >0. Increasing the shares of foreign equities owned by domestic residents also increases the domestic owner’s subsidy equivalent.

Comparing dEii, σj) with dEi (1,1) yields2

dEii, σj)−dEi (1,1) = 3(2−σj)(a−c)

2[2411(σi+σj) + 5σiσj] 3(a−c) 14

= 3[1011σij + 5σiσj]

14[2411(σi+σj) + 5σiσj](a−c).

Thus, dEii, σj) < dEi (1,1) if only if σi < 1011j

j, or equivalently, ∆σi = σi σj <

5(1σj)(2σj)

11j . The owner’s subsidy equivalent may be larger in the presence of cross-country shareholding if the two firms’ domestic share difference is large enough.

6.3.3 Equilibrium Output

Under the linear demand function, the equilibrium output yields xEi (σ) = 2dEi (σ) = 3(2−σj)(a−c)

2411(σi+σj) + 5σiσj.

The same result holds for the equilibrium output, i.e., xEii, σj)< xEi (1,1) if and only if σi < 1011j

j.

Proposition 6.2. Given managerial delegation, the presence of cross-country shareholding may increase firm i’s equilibrium output if firm i’s domestic share σi is large enough.

dEii, σj)TdEi (1,1) ⇐⇒ xEii, σj)TxEi (1,1) ⇐⇒ σi T 10j

11j.

Note that if σi < 1617, cross-country shareholding always lowers firm i’s equilibrium output, i.e., xEii, σj)< xEi (1,1) irrespective of values of σj.

6.3.4 Equilibrium Total Subsidy

Solving for total subsidy, it yields

SiE(σ) = sEi (σ) +dEi (σ) = (8σi+ 5σj iσj 6)(a−c) 2411(σi+σj) + 5σiσj . SiE(σ) is positive if and only if σj > 2(35(1σi)

i). (σ1, σ2) for S1E(σ) = 0 are depicted in Figure 6.1. Although the owner’s subsidy equivalent is always positive, total subsidy may become negative when foreign residents’ ownership in the home firm’s shares is large enough. That is, the government’s optimal tariff outweighs the owner’s optimal subsidy.

Since ∂SiE(σ)

∂σk >0(k =i, j), SiEi, σj)< SiE(1,1) always holds.

2dEi (1,1) is equivalent to ˆdDi in (5-22).

Next, I examine the effects of managerial delegation under cross-country shareholding.

Without managerial delegation, the optimal government subsidy (see Dick (1993), Welzel (1995)) yields

sCi (σ) = (16σi+ 12σj12σiσj15)(a−c) 3320(σi+σj) + 12σiσj . sCi (σ) > 0 is positive if and only if σj > 12(11516σσi

i). (σ1, σ2) for sC1(σ) = 0 are shown by the dashed line in Figure 6.1, which lies between the lines for sE1(σ) = 0 and S1E(σ) = 0.

Figure 6.1 shows that without managerial delegation, the government is more likely to subsidize its own firm.

(0.5,0.5) σ2

σ1 1

1 0.8

7/11

2/3 5/6

S1E >0

S1E >0

S1E <0 sE1 <0

sE1 <0

sE1 >0 0.9

0.6 S1E = 0

sE1 = 0

sC1 = 0 3/4

0.9

(1,1)

Fig. 6.1: sE1 and S1E

Without managerial delegation, the equilibrium output yields xCi (σ) = 2(3j)(a−c)

3320(σi+σj) + 12σiσj. 104

6.4.1 Optimal Subsidy

Comparing sEi (σ) withsCi (σ) yields

sEi (σ)−sCi (σ) = (16σi+ 13σj 10σiσj 18)(a−c)

2[2411(σi+σj) + 5σiσj] (16σi+ 12σj12σiσj15)(a−c) 3320(σi+σj) + 12σiσj

= 4(1ij2(117216σi+ 32σ2ij + 126210σi+ 32σi2

2[2411(σi+σj) + 5σiσj][3320(σi+σj) + 12σiσj] (a−c).

(6-5) Since the denominator in (6-5) is positive, the sign ofsEi (σ)−sCi (σ) is determined by the numerator. Define

f(σ) = 4(1i2j (117216σi+ 32σi2j+ 126210σi + 32σ2i (6-6) which is a quadric equation of σj. The discriminant for f(σ) is given by ∆(σi)def= (117 216σi+32σi2)216(1i)(126210σi+32σi2). There are four solutions for ∆(σi) = 0, but because of Assumption 4.1, I only consider the range aroundσi (0.5,1). Since ∆(0.5)>0 and ∆(1) <0, there exist at least one solution satisfying ∆(σi) = 0 for allσi (0.5,1) in view of the intermediate-value theorem.

Lemma 6.1. For all σi (0.5,1), there exists a unique σi =bσ satisfying ∆(bσ) = 0.

Proof. (Reduction to absurdity) If there are two solutions σa, σb (0.5 < σa < σb < 1) satisfying ∆(σa) = ∆(σb) = 0, by the mean-value theorem, there must existσc a, σb) satisfying ∆c) = 0. Since ∆′′i) = 64(941960σi + 192σ2i) > 0 for all σi (0.5,1),

i)> 0 for all σi > σc. This leads to ∆(σi) >0 for all σi > σb, which contradicts the result that ∆(1)<0.

From above and Lemma 6.1, it yields

∆(σi)T0 ⇐⇒ σi Sbσ , ∀σi (0.5,1). (6-7) Since 1i <0 in (6-6), from the results in (6-7), sEi (σ) vs. sCi (σ) can be shown for the following three cases.

(I)When σi >bσ, ∆(σi)<0 holds.

f(σ)<0 ⇐⇒ sEi (σ)< sCi (σ)

(II)When σi < bσ, ∆(σi) > 0 holds. There exist two real roots σji) and σji) satisfying fi, σj) = f(σi, σj) = 0.



f(σ)>0 ⇐⇒ sEi (σ)> sCi (σ) when σji)< σj < σji) f(σ) = 0 ⇐⇒ sEi (σ) = sCi (σ) when σj =σji) or σji) f(σ)<0 ⇐⇒ sEi (σ)< sCi (σ) when σji)< σj or σj > σji), where σji) = 117216σi+32σ

2 i+

∆(σi)

8(1i) and σji) = 117216σi+32σ

2 i

∆(σi) 8(1i) .

(III)When σi =σ,b ∆(σi) = 0 holds and σj(bσ) = σj(σ).b

{ f(σ)<0 ⇐⇒ sEi (σ)< sCi (σ) when σj ̸=σj(σ)b f(σ) = 0 ⇐⇒ sEi (σ) =sCi (σ) when σj =σj(σ)b  

The values of the two roots, σji) andσji), for ∆(σi) = 0 yield the following lemma.

Lemma 6.2. For all σi (0.5,σ),b σji) and σji) satisfy:

(i) σji)<0 , σji)>0.

(ii) σji)T1 ⇐⇒ σi Sσm. (iii) σji)T0.5 ⇐⇒ σi Tσn. Here 0.5< σm < σn<bσ <1.

Proof. See the Appendix.

The curves forσ21) and σ21) obtained from Lemma 6.2, are depicted in Figure 6.2.

The two curves intersected at σ1 =σ, whereb σ2(σ) =b σ2(bσ). Figure 6.2 is largely divided into two areas by the curves of σ21) and σ21). The left area shows sEi (σ) > sCi (σ) and the right area shows the opposite one; the curves represent sEi (σ) = sCi (σ). Note that regardless the value of σ2, if σ1 > bσ, sE1(σ) < sC1(σ) always holds; and if σi < σm, sE1(σ)> sC1(σ) always holds.

Proposition 6.3. Given cross-country shareholding, managerial delegation may raise or lower the governments’ optimal subsidy rates depending on the cross shareholding structurei, σj)whenσm ≤σi bσ. Furthermore, if the domestic shareholding ratio is small enough that σi < σm, managerial delegation always strengthens the government’s subsidization incentive; if the domestic shareholding ratio is large enough that σi >bσ, the government’s subsidization incentive is always weakened under managerial delegation.

Without cross shareholding, as shown in the previous chapters, managerial delegation always weakens the government’s subsidization incentive. When the domestic shareholding ratio is large, a small fraction of foreign shareholding does not change this result. However, when the domestic shareholding is small enough, a nearly half, the large portion of foreign shareholding induces the government to tax the exports. Under managerial delegation, the negative cross-rent shifting and dividend suppression effects are dampened, and as such, the government’s tax incentive is also weakened.

(0.5,0.5) σ2

σ1 1

σn 1

σm σb

sE1(σ)< sC1(σ) sE1(σ)> sC1(σ)

σ21) σ21)

σ2(bσ) = σ2(bσ)

Fig. 6.2: Values of sE1(σ) vs. sC1(σ)

6.4.2 Owner’s Subsidy Equivalent and Total Subsidy

dEi (σ)> sCi (σ) and SiE(σ)> sCi (σ) hold for any given (σi, σj). With managerial delega-tion, both the owner’s subsidy equivalent and total subsidy always result in higher subsidy rates regardless of the cross shareholding structure.

6.4.3 Output Decision

Comparing xEi (σ) with xCi (σ) yields

xEi (σ)−xCi (σ) = 2(1−σi)(2729σj+ 8σ2j) +σj

[2411(σi+σj) + 5σiσj][3320(σi+σj) + 12σiσj] >0.

Given σk(k = i, j) > 0.5, xEi (σ) > xCi (σ) holds. Managerial delegation increases the equilibrium output irrespective of the cross shareholding structure.

Proposition 6.4. Given cross-country shareholding, managerial delegation always in-creases each firm’s equilibrium output, i.e., xEi (σ)> xCi (σ).