Estimating Dynamic Models of Imperfect Competition

Bajari, Patrick; Benkard, C. Lanier; Levin, Jonathan

doi:10.2139/ssrn.537443

Cited by 238 publications

(404 citation statements)

References 41 publications

Supporting

Mentioning

402

Contrasting

Unclassified

Order By: Relevance

“…Because the number of such variables is usually large, one has to deal with the curse of dimensionality, which increases data requirements and poses computational challenges. In this article, I overcome these issues by using a data set on thousands of mergers within one industry, and by applying recent advancements in the estimation of dynamic games (see Bajari, Benkard, and Levin, ; hereafter “BBL”). Moreover, modelling of mergers in a dynamic framework introduces several conceptual issues including simultaneous merger bids for a single product and multiproduct bids by a single acquirer.…”

Section: Introductionmentioning

confidence: 99%

Estimation of cost efficiencies from mergers: application to US radio

Jeziorski

2014

The RAND J of Economics

View full text Add to dashboard Cite

This article estimates fixed-cost efficiencies from mergers using a dynamic oligopoly model in which mergers and repositioning of products are endogenous. The inference is based on revealed preference approach selecting cost synergies that rationalize observed merger decisions. The estimates can be used to assess the total welfare impact of retrospective and counterfactual mergers. The framework is applied to estimate cost efficiencies after the 1996 deregulation of U.S. radio industry. Within the period of 1996 to 2006 the cost savings resulting from mergers amount to $1.2 billion per year (equally split across economies of scale and within-format cost synergies). as well as participants at numerous seminars.1 In this article, I use the terms merger and acquisition interchangeably to mean any change of ownership of a part of or a whole company. 816Copyright C 2014, RAND. JEZIORSKI / 817 one I infer that the presumed cost efficiencies are too large. On the other hand, when the model predicts no merger, but the data indicate one, I infer the presumed cost efficiencies are too small.Implementing the proposed cost estimator requires robust long-run predictions of gains from mergers, which are obtained using a dynamic model with endogenous mergers and product characteristics. In contrast, previous empirical work analyzes mergers in a static framework and treats market structure as given (see Nevo, 2000;Pinkse and Slade, 2004;Ivaldi and Verboven, 2005). Such static models are useful in addressing the short-run impacts of mergers but do not account for resulting long-run changes in the market structure. Benkard, Bodoh-Creed, and Lazarev (2008) evaluate a longer-run effect of a merger on market structure but still treat it as an exogenous one-time event. The proposed dynamic framework builds on the above methods accounting for dynamic processes such as self-selection of mergers, follow-up mergers leading to merger waves, and postmerger product repositioning.Modelling and estimating models with endogenous mergers pose econometric and computational challenges. To evaluate a potential merger, both acquirer and acquiree must take into account the ownership structure and characteristics of all active products. Because the number of such variables is usually large, one has to deal with the curse of dimensionality, which increases data requirements and poses computational challenges. In this article, I overcome these issues by using a data set on thousands of mergers within one industry, and by applying recent advancements in the estimation of dynamic games (see Bajari, Benkard, and Levin, 2007; hereafter "BBL"). Moreover, modelling of mergers in a dynamic framework introduces several conceptual issues including simultaneous merger bids for a single product and multiproduct bids by a single acquirer. This study addresses the former issue by modelling players' moves as sequential with bigger owners moving first, and the latter issue by approximating multiproduct mergers with a series of highly correlated product-by-product acq...

show abstract

Section: Introductionmentioning

confidence: 99%

Estimation of cost efficiencies from mergers: application to US radio

Jeziorski

2014

The RAND J of Economics

View full text Add to dashboard Cite

show abstract

“…It is also related to recent literature on the identification and estimation of dynamic game models (e.g. Pesendorfer and Schmidt‐Dengler, 2008; Aguirregabiria and Mira, 2007; Berry, Ostrovsky and Pakes, 2007; Bajari, Benkard and Levin, 2007). While we do not focus on dynamic games here, one contribution that we make is the consideration of situations where agents have both continuous action spaces and continuous state spaces.…”

Section: Introductionmentioning

confidence: 93%

Pairwise-Difference Estimation of a Dynamic Optimization Model

Hong

Shum

2009

Review of Economic Studies

View full text Add to dashboard Cite

We develop a new estimation methodology for dynamic optimization models with unobserved shocks and deterministic accumulation of the observed state variables. Investment models are an important example of such models. Our pairwise-difference approach exploits two common features of these models: (1) the monotonicity of the agent's decision (policy) function in the shocks, conditional on the observed state variables; and (2) the state-contingent nature of optimal decision making which implies that, conditional on the observed state variables, the variation in observed choices across agents must be due to randomness in the shocks across agents. We illustrate our procedure by estimating a dynamic trading model for the milk production quota market in Ontario, Canada. Copyright © 2009 The Review of Economic Studies Limited.

show abstract

“…Three main sets of parameters need to be identified: buyers' utility function, sellers' unobservable quality, and sellers' cost parameters. The first two are estimated using a three‐step procedure, whereas the third is identified using a two‐step method similar to Bajari, Benkard, and Levin ()…”

Section: Identification Proceduresmentioning

confidence: 99%

“…The five moments are the number of Powersellers, number of registered stores, number of sellers who have both statuses, and two moments from the demand function (equation ):

\begin{matrix} true \\ left ({\overset{r}{true}}_{10} - {\overset{r}{true}}_{00}) / β_{p} - ({\overset{r}{true}}_{01} - {\overset{r}{true}}_{00}) / β_{s} = 0 \\ left ({\overset{r}{true}}_{10} - {\overset{r}{true}}_{00}) / β_{p} - ({\overset{r}{true}}_{00} - {\overset{r}{true}}_{10} - {\overset{r}{true}}_{01} + {\overset{r}{true}}_{11}) / β_{p s} = 0 . \end{matrix}

The five parameters to be estimated are the quality thresholds for Powerseller and store statuses,

μ_{p}

and

μ_{s}

; the coefficient of quality in the utility function of buyers,

β_{r} / α

; the parametric variable that converts the index of persistent quality found in the previous step to that value; and the variance of random shocks to quality. The estimation procedure will be discussed in detail in Section 5. The next step is estimating the cost parameters of sellers using a two‐step estimator method similar to the method in Bajari, Benkard, and Levin (). The method uses the basics of revealed profit to estimate the deep parameters of the model and, in this case, to estimate the cost parameters: the average monthly cost sellers should pay to become a store on eBay,

c^{s}

, and the average cost of obtaining an iPod for sellers, c .…”

Section: Identification Proceduresmentioning

confidence: 99%

“…This relationship allows the parametric estimation of qualities using the quantity choices of sellers over time. In the next step, using the approach of Bajari, Benkard, and Levin (), I identify sellers' cost parameters. This procedure is based on the assumption that the observed data are the outcome of sellers' maximization problem based on their actual cost function; therefore, the conditions for optimality of policy functions can be used to back out cost parameters.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reputation and adverse selection: theory and evidence from eBay

Saeedi

2019

The RAND J of Economics

View full text Add to dashboard Cite

How can a marketplace introduce mechanisms to overcome inefficiencies caused by adverse selection? In this article, I use a unique data set that follows eBay sellers to show that reputation is a major determinant of price variations. I develop a model of sellers' dynamic behavior where sellers have heterogeneous qualities unobservable by buyers. Using reputation as a signal of quality, I structurally estimate the model to uncover buyers' utility and sellers' costs and underlying qualities. I show that removing the reputation mechanism increases low‐quality sellers' market share, lowers prices, and consequently reduces sellers' profit by 66% and consumer surplus by 35%.

show abstract

Estimating Dynamic Models of Imperfect Competition

Cited by 238 publications

References 41 publications

Estimation of cost efficiencies from mergers: application to US radio

Estimation of cost efficiencies from mergers: application to US radio

Pairwise-Difference Estimation of a Dynamic Optimization Model

Reputation and adverse selection: theory and evidence from eBay

Contact Info

Product

Resources

About