Papers
- Assessing the turnpike hypothesis with the Rawls criterion (
256.6 kb )
(with Vitali Belenky)
The Turnpike theorem for the model with the terminal criterion is one of the most important results in the linear theory of Neumann-Gale economic dynamics. Recently, however, considerable attention has been paid to a model with another criterion proposed by Rawls. To our knowledge, the `Turnpike Hypothesis' (TH) for the model with the Rawls criterion has not been assessed yet. The present paper gives the answer to this question. By constructing examples of regular technologies for which trajectories do not converge, or converge to different rays, we show that the answer is negative. This work rejects not only the TH, but also the strong hypothesis. Descriptions of direct and conjugate (dual) models and corresponding mathematical apparatus are given. The paper also indicates that a related paper Kaganovich (2000) - which shows that under the Rawls criterion there is a rolling-plan mechanism converging to the turnpike - contains an error.
- Staged Financing with a Variable Return ( 201.4 kb )
(with Andrew Wait)
This paper explores the hold-up problem between two parties (an entrepreneur and an investor) when one of the parties (the entrepreneur) is unable to commit not to repudiate the initial contract. To mitigate the hold-up we allow the parties to stage investments over time and derive the optimal investment path for a variable rate of return case. Our model predicts that neither positive wealth of the entrepreneur nor the lack of discounting ensures that all profitable projects proceed. The model is extended in several ways: first, both agents are allowed to repudiate the initial contract and second, new costs of staged financing are introduced.
- Treasure game ( 258.8 kb )
(with Alex Matros)
We study the investment choice of two agents in a R&D race where the necessary threshold investment for success is uncertain. The race is modeled as a multistage game with observed previous actions where the player's probability of success depends only on his investment in that period. We provide a complete characterization of a symmetric equilibrium of this game. We find that the investment process can be divided in two stages: in the first stage cooperative behavior of players can be supported as an equilibrium; in the second stage there is overinvesting in comparison with cooperative equilibrium.
- Market niche, flexibility and commitment ( 133.9 kb )
(with Suren Basov and Andrew Wait)
We study a market-entry game in which the potential entrants wish to coordinate their actions (i.e enter different market segments rather than compete directly). If: (1) the firms have an option to wait; and (2) each firm has a different reaction time after they have decided to wait, the unique outcome that survives the iterated elimination of weakly-dominated strategies favors the less flexible firms.
- Market entry with a second-mover advantage ( 274.3 kb )
(with Andrew Wait)
We study a market-entry game with a second-mover advantage. In the symmetric equilibrium, there can be a non-monotonic relationship between the probability with which a player will invest (entry) and the length of time until the deadline. Moreover, the probability of investment can move chaotically as the horizon is extended. In the limit when the interval between periods goes to 0 chaotic trajectories arise when the efficiency effect does not hold - that is, when the one-period monopoly profit is less than the total of the one-period duopoly profits. We also show that the presence of chaotic trajectories is associated with a smaller expected delay in entry.
- Bundling in auctions with complementary goods ( 158.2 kb )
(with Andrew Wait)
We examine when a revenue-maximizing auctioneer prefers to auction a homogenous product in one bundle (a single-object auction) as compared with selling the item in two or more shares. When the items are super-additive (complementary) the auctioneer always prefers a single-object auction. When the product is sub-additive (supplementary), the auctioneer is more likely to choose a share auction when there are a large number of potential bidders. When there are a small number of bidders the auctioneer will tend to prefer a single-object auction. Further, we design a graphical approach for analyzing the optimal number of shares when there is a large number of bidders.