- Supplement to "What Model for Entry in First-Price Auctions? A Nonparametric Approach"
Marmer, Vadim, University of British Columbia - Economics, vmarmer@interchange.ubc.ca
Shneyerov, Art, Concordia University, achneero@alcor.concordia.ca
Xu, Pai, University of Hong Kong, paixu@hku.hk
The paper contains supplemental material for Marmer, Shneyerov, and Xu (2010) "What Model for Entry in First-Price Auctions? A Nonparametric Approach."
Creation Date: 2010-06-18 Revision Date: 2010-06-18
- Weak Identification in Fuzzy Regression Discontinuity Design
Lemieux, Thomas, University of British Columbia, tlemieux@interchange.ubc.ca
Marmer, Vadim, University of British Columbia - Economics, vmarmer@interchange.ubc.ca
We consider weak identification in the fuzzy regression discontinuity (FRD)
model. In this model, the treatment effect is identified through a
discontinuity in the conditional probability of treatment assignment. Weak
identification corresponds to the situation where the discontinuity is of a
small magnitude. When identification is weak, we show that the usual t-test based on the FRD estimator and its standard error suffers from
asymptotic size distortions. To eliminate those size distortions, we propose
a modified t-statistic that uses a null-restricted version of the standard
error of the FRD estimator. Simple and asymptotically valid confidence sets
for the treatment effect can be also constructed using the FRD estimator and its null-restricted standard error.
Creation Date: 2010-05-15 Revision Date: 2010-05-26
- A Folk Theorem for Competing Mechanisms
Peters, Michael, University of British Columbia - Economics, peters@econ.ubc.ca
Troncoso-Valverde, Cristian, University of British Columbia,
We prove a folk theorem for games in which mechanism designers compete
in mechanisms and in which there are at least 4 players. All allocations
supportable by a centralized mechanism designer, including allocations
involving correlated actions (and correlated punishments) can be supported
as Bayesian equilibrium outcomes in the competing mechanism game.
Creation Date: 2010-05-13 Revision Date: 2010-06-10
- On the Revelation Principle and Reciprocal Mechanisms in Competing Mechanism
Games
Peters, Michael, University of British Columbia - Economics, peters@econ.ubc.ca
This paper provides a set of mechanisms that we refer to as emph{reciprocal
mechanisms. }These mechanisms have the property that every outcome
that can be supported as a Bayesian equilibrium in a competing mechanism
game can be supported as an equilibrium in reciprocal mechanisms. In this
sense, reciprocal mechanisms play the same role as direct mechanisms do
in single principal problems. The advantage of these mechanisms over
alternatives like the universal set of mechanisms
is that they are conceptually straightforward and no more difficult
to deal with than the simple direct mechanisms used in single principal
mechanism design.
Creation Date: 2010-05-13 Revision Date: 2010-05-13
- Bias Due to Input Source Substitutions: Can It Be Measured?
Diewert, Erwin, University of British Columbia - Economics, diewert@econ.ubc.ca
Nakamura, Alice O. , University of Alberta, alice.nakamura@ualberta.ca
Once a business opts to purchase rather than produce an input, it can also change the source from which the product is procured. Producer price index programs face problems in dealing with price changes associated with sourcing changes. We present measures for price index bias due to sourcing substitutions. We begin with highly simplified cases to convey the rationale for our approach, and then show how the measures could be generalized. We also explain related aspects of the industry accounts. This material makes it clear that the growth of outsourcing and the related increases in domestic and foreign sourcing substitutions pose important challenges for statistics agencies.
Creation Date: 2010-05-12 Revision Date: 2010-07-13
- Decompositions of Profitability Change using Cost Functions
Diewert, Erwin, University of British Columbia - Economics, diewert@econ.ubc.ca
Profitability at a period of time is defined as the value of outputs produced by a production unit divided by the corresponding cost. Using some earlier work by O’Donnell, the paper provides a decomposition of profitability growth over two periods into various explanatory factors. The explanatory factors are: the change in the production unit’s cost efficiency, an index of output price growth, returns to scale, technical progress and an index of input price growth. If output prices for the production unit are not available, the paper suggests that marginal costs could be used as prices to weight outputs in each period. Using marginal or average costs to weight outputs leads to a methodology for evaluating the productivity performance of nonmarket production units that is similar to the methodology suggested in the Atkinson report.
Creation Date: 2010-05-12 Revision Date: 2010-07-13
- Serial Defaults, Serial Profits: Returns to Sovereign Lending in the Age of Philip II
Drelichman, Mauricio, University of British Columbia - Economics, drelichm@interchange.ubc.ca
Voth, Hans-Joachim, ,
Philip II of Spain accumulated debts equivalent to 60% of GDP. He also defaulted four times on his short-term loans, thus becoming the first serial defaulter in history. Contrary to a common view in the literature, we show that lending to the king was profitable even under worst-case scenario assumptions. Lenders maintained long-term relationships with the crown. Losses sustained during defaults were more than compensated by profits in normal times. Defaults were not catastrophic events. In effect, short-term lending acted as an insurance mechanism, allowing the king to reduce his payments in harsh times in exchange for paying a premium in tranquil periods.
Creation Date: 2010-04-01 Revision Date: 2010-04-01
- Comment on Understanding PPPs and PPP Based National Accounts
Diewert, Erwin, University of British Columbia - Economics, diewert@econ.ubc.ca
The paper discusses the three major problems that Deaton and Heston identify that make it difficult to construct consistent estimates of real output across countries: (i) the choice of a multilateral index number formula; (ii) weak national accounts estimates of country expenditures on the 155 major expenditure categories that are identified in the World Bank’s International Comparison Program and (iii) the fact that countries at different levels of development consume products that are not consumed widely by other countries in the comparison and thus the country price levels that are determined in the comparison project may not be reliable.
Creation Date: 2010-03-16 Revision Date: 2010-07-13
- Malmquist and Törnqvist Productivity Indexes: Returns to Scale and Technical Progress with Imperfect Competition
Diewert, Erwin, University of British Columbia - Economics, diewert@econ.ubc.ca
Fox, Kevin J. , University of New South Wales - Economics, K.Fox@unsw.edu.au
Caves, Christensen and Diewert proposed a method for estimating a theoretical productivity index for a firm using Törnqvist input and output indexes, augmented by exogenous estimates of local returns to scale. However, in order to implement their method, they assumed that the firm maximized revenue in each period, conditional on the observed input vector in each period, taking output prices as fixed. This assumption is not warranted when there are increasing returns to scale. Thus in the present paper, it is assumed that the firm solves a monopolistic profit maximization problem when there are increasing returns to scale and the results of Caves, Christensen and Diewert are modified in accordance with this assumption.
Creation Date: 2010-02-17 Revision Date: 2010-07-13
- Measuring Productivity in the Public Sector: Some Conceptual Problems
Diewert, Erwin, University of British Columbia - Economics, diewert@econ.ubc.ca
In many sectors of the economy, governments either provide various services at no cost or at highly subsidized prices. Examples are the health, education and general government sectors. The paper analyzes three possible general methods to measure the price and quantity of nonmarket government outputs. If quantity information on nonmarket outputs is available, then the first two methods of price valuation rely on either purchaser based valuations or on cost based valuations. If little or no information on the quantity of nonmarket outputs produced is available, then the method recommended in the System of National Accounts 1993 must be used, where aggregate output growth is set equal to aggregate input growth. The paper also discusses various methods of adjusting for quality change.
Creation Date: 2010-02-17 Revision Date: 2010-07-13
Home