13.07.2015 Views

School of Economic Sciences - Washington State University

School of Economic Sciences - Washington State University

School of Economic Sciences - Washington State University

SHOW MORE
SHOW LESS
  • No tags were found...

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

AppendicesAPro<strong>of</strong>sAssumptions L i (M ij ) = L i Y j (M ij ) = Y j θ j (M ij ) = θ j .Justification. Suppose governor i may take at most one trade mission to country j. Furthersuppose the parameters and fundamentals, Y j , w i , and τ ij , are such that this governor chooses togo on the mission. I show this decision will not change to zero missions when both the benefit andthe cost <strong>of</strong> the mission go to zero,limb→0˜X ij M b ij − ˜X ij = 0 and limdi →0 d ig j (M ij − 1) = 0.The notation above matches that in the paper, yet should be viewed generally enough to hold inform when the assumptions do not hold. Let d i and b converge to zero along the same sequence.Compare the rates <strong>of</strong> convergence. Use l’Hôpital’s rule:limz→0˜X ij M z ij − ˜X ijzg j= limz→0˜Xij M z ij log M ijg j (M ij − 1)= ˜X ij log M ijg j (M ij − 1) > 0.Therefore near the limit, the convergence rate is equal. Since the governor chooses to go on a trademission initially and the rate <strong>of</strong> convergence is constant in the limit, the governor does not want tochange her mind at any point in the sequence.Lemma 1. Equilibrium aggregate exports from i to j as a function <strong>of</strong> trade missions,X ij (M ij ) = ˜X ij × M abij ,where a =γσ−1 − 1 > 0 and b > 0, is increasing in M ij.26

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!