Appendix A. Sequential Entry
Proof. Step 1. Start in period n. Let the remaining capacity be kn = kn-1 — x^-↑ ■ There are
two subgames; either player n enters or stays out of market n. In the subgame with entry, the
unique Nash equilibrium is {xt^ (kn), -r^}, where xt^ (kn) = min {kn,x™} and "xrn = βe (xtβ (kn)).
If player n decides to stay out of market n, we have the following limit Nash equilibrium [x™', θ},
where ^ym (x™, θ) = 0. The unique Nash equilibrium in the subgame with no entry is {x)(λ (k), 0},
where x™ (kn) = min {kn,x™}. From S, it follows that x™ > x™.
Step 2. Player n would enter if kn < x and stay out as long as kn ≥ x. To deter entry, player m
would need kn ≥ x. Now, assume that enough unused capacity remains to deter entry. Rewrite the
equilibrium output of player m in period n as a function of kn-ι and x%-ι, i.e. x™ (kn-ι, x%-ι ) =
min {kn-i — xrβ-1,χm}.
Step 3. Working backwards to period n — 1, we have two subgames; either player n — 1 enters or
stays out of market n — 1. First, capacity kn-ι would ensure a successful commitment by player
m in market n — 1 to an output x™—i, if and only if:
°v °x™ °V
-----(T™ 1,β1 (T™ 1)) + —ɪ ∙ -— (x™ (k,T™ 1), 0) ≥ 0 (13)
я„,™ ∖ n-1, ∖ n- 1)) а^т Ягг-т ∖ n V , n-1 ) , ) — V >
°xn-1 °xn-1 °xn
Now, ∂χm = —1 if k ≤ x™ + τm and g∂⅛ (k, 2:^-1 ) = 0 if k > x™ + xrβ^. To deter entry, player
m has to commit to T in the subgame with entry. The following inequality must be satished:
°V °V
-°^(x, в™(T)) = ° °c, 0) (14)
If xn-1 > 0, it follows from (S) that xrβ > x ⇒ kn-1 > 2T. If (D) holds with equality, i.e.
T = x™, then the LHS of equality [14] is equal to zero and the equality is satished if and only if
, ~ =rn , , ~ , =rn
kn-1 — x = x ⇒ kn-1 = x + x .
Step ^. Working backwards to period n — 2, we have two subgames; either player n — 2 enters or
stays out of market n — 2. First, capacity kn-2 would deter entry if:
°v ° V
∂x^- (t, в™ (T)) = ∂x^
°xn-2 °xn-1
° ° °V
(x™.,, 0 = (x™, 0)
∖ n- 1 ’ / rnrm × n ’ '
^x >x ^n
(15)
26
More intriguing information
1. Monetary Policy News and Exchange Rate Responses: Do Only Surprises Matter?2. Testing for One-Factor Models versus Stochastic Volatility Models
3. The name is absent
4. Peer Reviewed, Open Access, Free
5. TOMOGRAPHIC IMAGE RECONSTRUCTION OF FAN-BEAM PROJECTIONS WITH EQUIDISTANT DETECTORS USING PARTIALLY CONNECTED NEURAL NETWORKS
6. The name is absent
7. On the origin of the cumulative semantic inhibition effect
8. The name is absent
9. The name is absent
10. On Social and Market Sanctions in Deterring non Compliance in Pollution Standards