_ 2÷ε-[2-(n-l)ε-,⅜tεε.t (43)
‘ (2+ε)(2-nε)
This confirms, as required, that xk is increasing in t-k,, the sum of the access costs of all firms
excluding firm k. By contrast, the sign of the effect of own access cost on sales appears to
be ambiguous. However, it can be shown that it must be negative. The substitutability
parameter ε in (41) cannot be a primitive one, since if it was then demand for good k would
increase without bound as the number of goods increased. It is shown in Neary (2000) that,
if demands are generated by a symmetric quadratic utility function with substitutability
parameter e, then ε and e are related as follows: ε=e/[ 1+(n-1 )e]. Making this substitution,
the coefficient of tk in (43) can be seen to be unambiguously negative, as required. It is also
sufficiently negative to offset the effects of an increase in t-k=(n-1 )tk.: the result in (18)
continues to hold.
A.2 Cournot Competition with General Demands
Return to the Cournot case, with homogeneous products, but now allow for a general
demand function: p=p(X). The first-order condition for firm k is: bxk=p-tk; where b is
(minus) the demand slope: b≡-p'(x). Summing over all n-1 firms and totally differentiating
gives:
(n+2+r)bdx = -dt (44)
where r≡xb'/b is a measure of the concavity of the demand function, and the coefficient of
dx must be positive for stability. Next, totally differentiating the first-order condition for firm
k:
bdx,k = - (l+akr)bdx - dtk (45)
25
More intriguing information
1. Developmental Robots - A New Paradigm2. Structure and objectives of Austria's foreign direct investment in the four adjacent Central and Eastern European countries Hungary, the Czech Republic, Slovenia and Slovakia
3. Chebyshev polynomial approximation to approximate partial differential equations
4. The name is absent
5. An alternative way to model merit good arguments
6. The name is absent
7. The name is absent
8. Fiscal federalism and Fiscal Autonomy: Lessons for the UK from other Industrialised Countries
9. Permanent and Transitory Policy Shocks in an Empirical Macro Model with Asymmetric Information
10. The name is absent