We have analyzed a model inwhich the effects of vertical integration on consumer and overall
welfare depend on the underlying market structure. We have shown that, perhaps surprisingly,
vertical integration is more likely to be procompetitive exactly when the market structure consists
of a small number of nonintegrated rivals. More generally, in our model vertical integration is
procompetitive under fairly wide circumstances because efficiency effects due to a reduction
in the oligopsony distortion tend to dominate foreclosure effects. Because of this, even vertical
integration that leads to full foreclosure of the rivals can be procompetitive. However, vertical
integration can also increase consumer surplus and decrease total welfare because final output
may be produced at higher costs after integration. With regards to the incentives to vertically
integrate, we find that a small amount of integration is always profitable despite the free-riding
problem the integrating firm faces. In addition, the private incentives to integrate tend to be too
weak when the number of rivals is small and excessive when it is large. Our numerical results also
indicate that – within the confines of our model – the effects of seemingly intuitive but ultimately
misguided policy recommendations can be sizeable.
Our model considers the case in which upstream suppliers are perfectly competitive whereas
downstream firms have full oligopsony power, leaving the open question for future research of
what happens when input suppliers exert market power as well. We expect that the main effects
will still be at work if input suppliers have some limited bargaining power. If a downstream firm
integrates with an upstream supplier, the newly integrated firm owns more capacity units and
therefore will purchase more capacity from other upstream suppliers on the input market. At the
same time, it will produce more infra-marginal units of output and thus utilize its capacity less
intensively than nonintegrated downstream firms.
In contrast, the effects we identified will not be at work if downstream firms have no market
power on the wholesale price as is the case, for example, in the well-known model of Salinger
(1988), in which only upstream firms can influence the market price. Our model can therefore be
seen as analyzing the other end of the spectrum, in which market power in the input market is
purely oligopsonistic instead of oligopolistic, so that the two models are complementary to each
other.
We also note that the policy implications of models inwhich only upstream firms have market
power in the intermediate good market differ from ours. For example, in Inderst and Valletti
(2011a), the foreclosure effect becomes more pronounced if the downstream market consists
of a smaller number of firms, because the integrated firm then comes close to monopolization
of the market. A similar effect is present in Salinger (1988). As a consequence, an important
determinant for policy implications on vertical integration is the distribution of market power in
the upstream market. Our main policy implication that, if our model is perceived as a plausible
description of the relevant markets, antitrust authorities should be wary of vertical integration
in less concentrated industries is therefore most relevant if downstream firms have considerable
market power in the intermediate good market.
Our main result can be tested empirically using data from industries with different levels of
market concentration. The prediction of our model is that the likelihood that efficiency effects
dominate is increasing in market concentration. Existing empirical studies on the competitive
effects of vertical integration have focused on highly concentrated industries and found that
efficiency effects dominate foreclosure effects in almost all cases (see Lafontaine and Slade,
2007). Our results indicate that this finding could be overturnedwhen examining less concentrated
markets
We have analyzed a model inwhich the effects of vertical integration on consumer and overallwelfare depend on the underlying market structure. We have shown that, perhaps surprisingly,vertical integration is more likely to be procompetitive exactly when the market structure consistsof a small number of nonintegrated rivals. More generally, in our model vertical integration isprocompetitive under fairly wide circumstances because efficiency effects due to a reductionin the oligopsony distortion tend to dominate foreclosure effects. Because of this, even verticalintegration that leads to full foreclosure of the rivals can be procompetitive. However, verticalintegration can also increase consumer surplus and decrease total welfare because final outputmay be produced at higher costs after integration. With regards to the incentives to verticallyintegrate, we find that a small amount of integration is always profitable despite the free-ridingproblem the integrating firm faces. In addition, the private incentives to integrate tend to be tooweak when the number of rivals is small and excessive when it is large. Our numerical results alsoindicate that – within the confines of our model – the effects of seemingly intuitive but ultimatelymisguided policy recommendations can be sizeable.Our model considers the case in which upstream suppliers are perfectly competitive whereasdownstream firms have full oligopsony power, leaving the open question for future research ofwhat happens when input suppliers exert market power as well. We expect that the main effectswill still be at work if input suppliers have some limited bargaining power. If a downstream firmintegrates with an upstream supplier, the newly integrated firm owns more capacity units andtherefore will purchase more capacity from other upstream suppliers on the input market. At thesame time, it will produce more infra-marginal units of output and thus utilize its capacity lessintensively than nonintegrated downstream firms.In contrast, the effects we identified will not be at work if downstream firms have no marketpower on the wholesale price as is the case, for example, in the well-known model of Salinger(1988), in which only upstream firms can influence the market price. Our model can therefore beseen as analyzing the other end of the spectrum, in which market power in the input market ispurely oligopsonistic instead of oligopolistic, so that the two models are complementary to eachother.We also note that the policy implications of models inwhich only upstream firms have marketpower in the intermediate good market differ from ours. For example, in Inderst and Valletti(2011a), the foreclosure effect becomes more pronounced if the downstream market consistsof a smaller number of firms, because the integrated firm then comes close to monopolizationof the market. A similar effect is present in Salinger (1988). As a consequence, an importantdeterminant for policy implications on vertical integration is the distribution of market power inthe upstream market. Our main policy implication that, if our model is perceived as a plausibledescription of the relevant markets, antitrust authorities should be wary of vertical integrationin less concentrated industries is therefore most relevant if downstream firms have considerablemarket power in the intermediate good market.Our main result can be tested empirically using data from industries with different levels ofmarket concentration. The prediction of our model is that the likelihood that efficiency effectsdominate is increasing in market concentration. Existing empirical studies on the competitiveeffects of vertical integration have focused on highly concentrated industries and found thatefficiency effects dominate foreclosure effects in almost all cases (see Lafontaine and Slade,2007). Our results indicate that this finding could be overturnedwhen examining less concentratedmarkets
การแปล กรุณารอสักครู่..
