This draft: February 2009
Abstract:
This paper reexamines the effect of the regulatory regime on both penetration and coverage of broadband access to the internet. The framework also allows for an evaluation of different public policy measures such as subsidization of broadband demand and supply. A welfare analysis asks what the optimal regulatory regime is and whether and how high-speed access to the internet should be subsidized. Using an approach similar to Valletti et al. (2002), the paper highlights the importance of population density for whether firms invest to provide internet access. The analysis reveals a trade-off between coverage and penetration.
Keywords: Broadband, coverage, penetration, investment, population density
JEL: L 51, L 96, L12
- Margin Squeeze in Fixed-Network Telephony Markets - competitive or anticompetitive?
This draft: November 2008
Joint work with: Wolfgang Briglauer and Anton Schwarz
Abstract:
This paper looks at the effects of different forms of wholesale and retail regulation on retail competition in fixed network telephony markets. We explicitly model two asymmetries between the incumbent operator and the entrant: (i) While the incumbent has zero marginal costs, the entrant has the wholesale access charge as (positive) marginal costs; (ii) While the
incumbent is setting a two-part tariff at the retail level (fixed fee and calls price), the entrant can only set a linear price for calls. Competition from other infrastructures such as mobile telephony or cable is modelled as an ‘outside opportunity’ for consumers. We find that a
horizontally differentiated entrant with market power may be subject to a margin squeeze due to double marginalization but will never be completely foreclosed. Entrants without market power might be subject to a margin squeeze if the wholesale access price is set at average
costs and competitive pressure from other infrastructures increases. We argue that a wholesale price regulation at average costs is not optimal in such a situation and discuss retail minus and deregulation as potential alternatives.
Keywords: access regulation, foreclosure, margin squeeze, telecommunications, fixed
networks
JEL L12, L41, L42, L50, L96
- Do we (still) need to regulate fixed network retails markets?
This draft: November 2008
Joint work with: Wolfgang Briglauer and Anton Schwarz
Abstract:
In the beginning of fixed network liberalisation in Europe in the late 1990s, the main concern of regulators was to lower calls prices. This was done by introducing wholesale regulation and promoting service based competition. Some years later, the concern of some regulators turned from too high calls prices to too low calls prices which might ‘squeeze’ entrants out of the market. We look at a simple model in which this development is explained by increasing competitive pressure from an ‘outside opportunity’, e.g. mobile telephony. We conclude that a margin squeeze is not necessarily used by the incumbent as a device to drive competitors out of the market and increase market power but can also result from increased inter-model competition. If this is the case, we argue that regulators should consider alternatives to cost oriented access prices such as retail minus or complete deregulation.
Keywords: access regulation, vertical integration, foreclosure, price squeeze,
telecommunications, fixed networks
JEL: L12, L41, L42, L50, L96
- Diffusion of new technology – The case of multiple generations
This draft: December 2006
Joint work with: Thomas Astebro, University of Toronto
Abstract:
We model adoption decisions by competitive firms when successive generations of a new technology become available over time. Profit-maximizing firms choose which generation to adopt and at which adoption date. The model accounts for leapfrogging and simultaneous adoption of different generations. Leapfrogging occurs if a potential user does not adopt the state of the art technology but adopts the next generation. In the simultaneous adoption case some adopt the new generation while others still adopt an old generation. The two mentioned patterns are shown to be equilibrium outcomes that depend on exogenous parameters. Both overlap and leapfrogging may arise from the same parameters. As a consequence, firms of different sizes may adopt a generation at the same time. The model predicts that leapfroggers, who are the smallest firms in the industry, may well adopt the new generation before medium and large firms and that large firms are likely to adopt both generations. Empirical analysis on the adoption of two generations of machine tools (NC and CNC) by U.S. metalworking plants show that there is indeed substantial leapfrogging: 26% of all plants in the sample had adopted CNC but not NC by 1993. There was also overlap in adoption: 68% of the adopters of CNC adopted during 1981-1993; 53% of the adopters of NC adopted that technology during the same period. We find that the non-adopters of both technologies (NC and CNC) are the smallest, the adopters of NC but not CNC are on average larger, the leapfroggers are still larger and the largest plants, on average, adopt both technologies. Leapfroggers adopt CNC approximately one year earlier than adopters of both technologies. Empirical results are broadly consistent with model predictions.
Keywords: Diffusion, Leapfrogging and CNC-machine tools
JEL-Classification No.: L13, O31 and O33
Joint work with: Anna Hammerschidt
The final version is published in: Bulletin of Economic Research, 61(2), 179-188
Abstract:
This paper shows that R&D cooperation leads to the monopoly outcome in terms of price and quantity if demand is unit-elastic. If the demand function exhibits an upper bound for the willingness to pay, R&D cooperation is inferior to a scenario in which firms cooperate both in their R&D and their output decision.
Keywords: R&D cooperation, spillovers, cartelization
JEL-Classification No.: L13, O31
Abstract:
Strategic (capital) investments and the strategic choice of product characteristics are among the most important devices to manipulate market positions in a favorable way. This article examines optimum firm behavior as a function of cost parameters, market size, and barriers to entry. Synthesizing the economics and the business literature, it discusses under what conditions excessive entry deterrence, second-mover advantage as well as delegation of entry deterrence emerge in the same framework. The paper shows that both the number of firms and the equilibrium prices may be non-monotonic in market size. Larger markets may exhibit higher prices.
Keywords: Hotelling model, entry deterrence, over-investment, under-investment, capital investment, strategic location choice
JEL-Classification No.: L11, L13
Joint work with: Klaus Gugler, University of Vienna
This draft: August 2003
We show that for a spatially differentiated economy reduced product variety is the likely outcome of mergers except in cases where exit costs in relation to (outlet specific) fixed costs are high. Our empirical analysis of the Austrian retail gasoline market confirms that increases in concentration reduce product variety. Ignoring this product variety effect is likely to lead to an underestimate of market power in structural merger analysis.
Keywords: spatial product differentiation, retail gasoline, mergers, concentration
JEL-Classification No.: L11, L13, L90
This article reexamines sequential entry of firms in a Hotelling model of spatial product differentiation and corrects some results of Neven (1987). Contrary to Neven, I show that the pattern of locations is generally asymmetric in the case of a duopoly. Profits are non-monotonic in market size, even in the range where the number of firms does not change. The firm that bears the "burden" of entry deterrence gains from lower barriers to entry as long as entry deterrence is possible. Equilibrium profits of all firms may be larger in situations in which more firms are active.
Keywords: Hotelling model, entry deterrence, strategic location choice
JEL-Classification No.: L11, L13
This paper shows that vertical foreign direct investment will reduce prices but the aggregate welfare effect is unambiguously positive only under free market entry. Using a standard model of imperfect competition, we develop this result by considering two different cases. In the first case, the total number of firms is fixed, and we show that national and multinational firms may coexist. In the second case, we allow for market entry, and we focus on situations in which either only national or only multinational firms are active. Furthermore, we discuss impact effects on labor demand. We show that a decline in foreign wages increases domestic employment.
Abstract:
This article adds technology choice to a free-entry Cournot model with linear demand and constant marginal costs. Firms can choose from a discrete set of technologies. This simple framework yields non-existence of equilibrium, existence of multiple equilibria and equilbria in which ex-ante identical firms choose different technologies, as possible outcomes. I provide a full characterization of the parameter sets for which these outcomes arise. The (non-)existence problem disappears if vertical market size is large. Non-existence is largely a 'small number' phenomenon. Asymmetric equilibria emerge either because of indivisibilities or due to similarity of different technologies in terms of the average costs realized.
Keywords: Cournot equilibrium; existence; market size, heterogeneity; integer constraint
JEL Classification: D43; L13
This draft: April 2002. Abstract:
This article introduces technology choice into a Hotelling model of spatial competition. This yields two entry deterrence devices, as well as complex strategic choices for the firms and a rich picture of industry structure. Depending on cost parameters and market size, firms may choose to over-invest or to under-invest. Industry structure is typically asymmetric either in terms of the locations chosen or the technologies used or in both. I find excessive entry deterrence, second mover advantage as well as delegation of entry deterrence. Both the number of firms and the equilibrium prices may be non-monotonic in market size. Larger markets may exhibit higher prices.
Keywords: Hotelling model, entry deterrence, overinvestment, underinvestment
JEL-Classification No.: L11, L13
Size: ~300k
Joint work with: Walter Elberfeld, University of Cologne
This draft: June 2001. German Economic Review, Vol. 3, 2002. Abstract:
We introduce technology choice into a model of monopolistic competition and analyze the structural effects of changes in market size. A larger market leads to the adoption of a large scale technology. If a technology switch occurs, the number of firms decreases, and a rationalizing effect arises: individual and aggregate output increases; prices fall. This need not benefit consumers since a technology switch is associated with a decrease in product variety.
Keywords: technology choice, monopolistic competition, shakeout, variable elasticity of substitution
JEL-Classification No.: L10
Size: ~235k
This draft: January 2002.
Published in: International Journal of Industrial Organization, 2002, Vol. 20, pp. 1409 -1436. Abstract:
This paper adds two elements to a standard model of monopolistic competition: First, the number of potential entrants is limited in each period and increases only over time. Second, the potential entrants differ with respect to the consumers’ valuation of the variant they could offer. It is shown that the resulting simple model exhibits a rich dynamic structure covering cases like the product life cycle, a path dependent equilibrium and the traditional textbook case of entry. The welfare analysis confirms the view that you can’t have too much entry. Even entry of 'inefficient' firms improves welfare.
Keywords: Industry evolution, product life cycle, path dependence
JEL-Classification No.: L10
Size: ~ 575k
This draft: October 1998
A slightly revised version has been published in The RAND Journal of Economics, Vol. 30, 1999, pp. 679-693. Abstract:
This paper analyses the adoption and diffusion of new technology in a market for a differentiated product with monopolistic competition. It is shown that in a noncooperative equilibrium ex-ante identical firms adopt a new technology at different dates. This equilibrium can be described by a simple distribution function. For non-identical firms, the conditions are stated under which a positive relationship between firm size and speed of adoption exists. It is demonstrated that increased competition often promotes diffusion. Diffusion is shown to occur more slowly in the noncooperative solution than in a constrained social optimum.
An older, more extensive and slightly different version from August 1997 is available here.
Keywords: Adoption, diffusion, monopolistic competition
JEL-Classification: O31
Size: ~ 83k (old version:211k)
This draft: March 1998
Published in: International Journal of Industrial Organization, Vol 18, 2000, pp. 369-379.
Abstract:
In this note the circumstances under which ex ante identical firms will choose different adoption dates are clarified. In particular, conditions under which ‘diffusion’ will arise in both open-loop and closed-loop games are identified. Furthermore, it is shown that the rents of nonidentical firms are not equalised, even if pre-emptive adoption is possible. Finally, an example is given in which the reduction of the uncertainty associated with the implementation of the new technology leads to a postponement of the adoption by the late adopter.
JEL classification: O31, O32
Keywords: Adoption, diffusion, pre-emption, rent equalisation.
