1 Advances in Political Economy - Department of Political Science

EDITOR’S PROOF

A Collective-Action Theory of Fiscal-Military State Building 55

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

403

404

405

406

407

408

409

410

411

412

413

414

1.3 Investment in fiscal centralization

First, notice that when fiscal capacity is centralized the ruler maximizes fiscal trans-

fers by choosing the socially optimal amountG∗. Each corporation faces the fol-

lowing participation constraint when the ruler proposes centralization:

v(xi,y) ̄ +θαiy(G,y) ̄ −t≥v(xˆi,y) ̄ +θαiy

(

f(g),ˆ y ̄

)

−ˆgi−ˆτi−ei. (4)

The right hand side of (4) is constant and given by the equilibrium policy pro-

file{ˆτ,x,ˆgˆ}. The corporations can refuse centralization and force the ruler to

keep fiscal capacity fragmented. Summing overn and solving fort, we ob-

tainnt≤

∑

i[v(xi,y) ̄ +θαiy(G,y) ̄]−C, whereCis a constant. It follows that

uCR(x, G)≤

∑

i[v(xi,y) ̄ +θαiy(G,y) ̄]−C−c(x, G)−F. Therefore, the ruler

sets the maximum upper bound on net fiscal transfers by choosingG∗as defined

in (3).

Two conditions must hold for fiscal centralization to occur. First, the participa-

tion constraint in (4) must hold for at leastn ̄corporations. The corporations can

refuse centralization and the ruler has no credible threat but to preserve fiscal frag-

mentation. Second, the ruler’s payoff must be higher under centralization than under

fragmentation. If the ruler’s payoff given the tax payment necessary to obtain com-

pliance fromn ̄corporations is less than the payoff from{ˆτ,x,ˆ gˆ}, the ruler does not

propose centralization.

Substituting in (4)for(τˆi,xˆi,g)ˆ and solving fortwe obtain the maximum tax

payment that each corporation is willing to pay in exchange for the optimal level of

the public good:

tiM≤v(xi,y) ̄ +θαi

[

y

(

G∗,y ̄

)

−y

(

f( 0 ,g−i),y ̄

)]

, fori= 1 ,...,n. (5)

A couple remarks about this maximum tax payment are in order. First, the ruler

obtains higher maximum payments from those corporations who benefit more from

the public good (αi). Second, the ruler can obtain compliance from corporationiat

a tax payment higher than the maximum in (5) by compensating with private goods

(a higherxi) or if the prospects of economic activity increase.

I first obtain the SPNE assuming the ruler can collect corporation-specific tax

payments and provides the socially optimal amount of private goodsx∗. The con-

straints in (5) bind for alli, otherwise the ruler would be able to increase his

payoff by increasing the tax payment for some corporations. Lett∗i ≡v(xi,y) ̄ +

θαi[y(G∗,y) ̄ −y(f ( 0 ,g−i),y) ̄]be the binding constraint in (5)fori. The follow-

ing proposition gives the condition under which policy profile{t∗,x∗,G∗}is an

equilibrium forn ̄=n, wheret∗=(t 1 ∗,...,tn∗).

Proposition 1At the SPNE,the ruler proposes policy profile{t∗,x∗,G∗},all cor-

porations accept and the ruler increases fiscal centralization if the probability of a

threat is such that:

θ≥

F+c(x∗,G∗)−[

∑

iei+c(x

∗,G)ˆ]

[Y(G∗,y) ̄ −Y(G,ˆ y) ̄]

∑

iαi/n

. (6)