Franco Mariuzzo

Updates Management in Mobile Applications. iTunes vs Google Play

This paper focuses on a specific strategy that developers of mobile applications may use to stimulate demand: the release of updates. We start with a stylised theoretical analysis to describe the developers decision to release an update. Its predictions are then tested by using an unbalanced panel with the top 1,000 apps in iTunes and Google Play for five European countries. We show that while in iTunes updates increase the rate of growth of downloads, in Google Play their effect is not significant. We argue that the lack of quality control by Google Play can lead to an excess of updating. We also find that the past performance of the app influences the decision to release an update, but only in iTunes. This finding is in line with our theoretical analysis and can again be interpreted on the basis of the different way of governing the release of updates in the two stores.

What Can Merger Retrospectives Tell Us? An Assessment of European Mergers

In this review of retrospective European merger studies we provide a discussion of the price effect of analysed mergers and examine whether the antitrust agency made the right decisions. We find that remedied mergers, on average, were not followed by a price-increase, suggesting that, in our sample, merger interventions were effective at eliminating problems. High market concentration was more likely to lead to higher post-merger prices, although remedies were able to reduce post-merger price-increases, even in concentrated markets. We look at a number of reasons why prices may increase post-merger and find little evidence of genuine agency errors.

Commodity Market Dynamics and the Joint Executive Committee, 1880–1886

Using weekly spot and future commodity prices in Chicago and New York, we construct expected transportation rates for grain between these two cities, expected inventory levels in New York, and realized errors in the expectations of such variables. We incorporate these exogenous commodity market dynamics into Porters (1983) structural modeling of the Joint Executive Committee Railroad Cartel. As in Porter, we model marginal cost as a parametric function of (instrumented) output, among other factors. Unlike Porter, we model pricing over marginal cost as a nonparametric function of a set of variables, which include expectations of deterministic demand cycles and cartel stability. We estimate the pricing and demand equation simultaneously and semiparametrically. Our estimated weekly markups during periods of cartel stability are shown to reflect optimal collusive pricing over deterministic business cycles, as modeled in Haltiwanger and Harrington (1991). Periods of cartel instability are proven to be triggered by realized mistakes in expectations of New York grain prices.

Demand Estimation and Merger Simulations for Drugs: Logits v. AIDS

We uses ADHD drugs sales data from 2000-2003 and compare estimates of elasticities and merger simulations from three different demand models. Models include logit, random coefficients logit, and conditional AIDS demand model with multistage budgeting. The magnitude of cross-price elasticities is large in the third model in comparison to the first two, and some of the cross-price elasticities are estimated to be negative. Hypothetical merger simulations show large price effects for the multistage AIDS model in comparison to the discrete choice models.We uses ADHD drugs sales data from 2000-2003 and compare estimates of elasticities and merger simulations from three different demand models. Models include logit, random coefficients logit, and conditional AIDS demand model with multistage budgeting. The magnitude of cross-price elasticities is large in the third model in comparison to the first two, and some of the cross-price elasticities are estimated to be negative. Consequently, hypothetical merger simulations show large price effects for the multistage AIDS model, both for the merging firms as well as for the competitors, in comparison to the other two discrete choice models.

QUANTIFYING THE DETERRENT EFFECT OF ANTICARTEL ENFORCEMENT: ESTIMATING DETERRENCE

Evaluations of the consumer harm caused by cartels are typically partial because they do not attempt to quantify the impact of deterrence, or acknowledge that the CA does not root out all anti-competitive cases. This paper proposes a broader framework for evaluation which encompasses these unobserved impacts. Calibration of this framework is challenging because one cannot rely on estimates for cases which have been observed to make deductions about those that have not - an example of the classic sample selection problem which is endemic across much of the empirical Industrial Organisation literature. However, we show how empirical findings, already available in the existing literature, can be plugged into a Monte Carlo experiment to establish bound estimates on the magnitudes of cartel-induced consumer harm. Lower bound (i.e. cautious) estimates suggest that (i) the harm detected by the CA really is only the tip of the iceberg, accounting for only a small fraction (at most one sixth) of total potential harm; (ii) deterrence is at least twice as effective as detection as a means for removing harm; and (iii) undetected harm is at least twice as large as detected harm. Under less cautious, but very plausible, assumptions, all three effects could be much greater than this.

Cartel Enforcement and Deterrence Over the Life of A Competition Authority

This paper searches for evidence that, as a Competition Agency (CA) builds up experience in cartel enforcement, this feeds back into the business community to deter future cartel formation. We present two simple models, focussing respectively on composition and frequency deterrence, which describe how the feedback would work. The ideal outcome is that, over the long-run, the number of cartels detected by a successful CA will follow an inverted U-shaped time path: its propensity to detect increase, but the number of cartels out there to be detected decrease. Empirically, we try to simulate the long-term dimension by using an international panel of CAs. Although comparable data are only available for a relatively short time period (2006-2014) we hope that longer-run effects are captured by including in the panel CAs at very different stages in their life cycles. We find evidence of the predicted inverse U shape, and interpret this as consistent with an increasingly strong feedback from enforcement to deterrence as the CA evolves over the years.

Feasibility study on the microeconomic impact of enforcement of competition policies on innovation: final report

While there is an extensive literature on the relationship between market competition and innovation at the aggregate level, there are far fewer case-specific accounts of this relationship, particularly on the effects of specific competition policy interventions on innovation. The objective of this study was to offer a detailed literature review, develop a methodological framework for ex post evaluating such impacts, collect data, and provide preliminary results on how innovation was affected by the 2012 consolidation of the HDD market (involving the Seagate/Samsung and Western Digital/HGST mergers). Following a Schumpeterian trichotomy, we evaluate the impact of this consolidation on three layers of innovation: research spending, patent activity, and the characteristics of newly marketed products. Our results show a heterogeneous response, with one of the acquiring firms, Seagate, displaying increased innovation on all three levels. This is consistent with an interpretation that, notwithstanding the increased concentration, the HDD market remains contestable, mainly from other storage technologies (as predicted by the European Commission in 2011). We found no impact on Western Digital’s R&D spending, or product characteristics, but an increase in their patent activity. The study provides detailed explanations for these results, together with important methodological contributions on how to use R&D, patent and product characteristics data to ex-post evaluate the impact of competition policy on innovation.

To upgrade or not to upgrade? the release of new versions to survive in the hypercompetitive app market

Very low entry barriers and an exceptionally high degree of competition characterize the market for mobile applications. In such an environment one of the critical issues is how to at- tract the attention of users. Practitioners and developers are well aware that managing app updates (i.e., releasing new versions of an existing app) is critical to increase app visibil- ity and to keep users engaged, disguising a hidden strategy to stimulate downloads. We use unbalanced panel data with characteristics for the top 1,000 apps on iTunes and Google Play stores, for five European countries, to empirically in- vestigate publishers’ strategies concerning the release of up- dates. We find that only in the case of iTunes updates boost downloads and are more likely to be released when the app is experiencing poor performance. We interpret this finding as evidence that the lack of quality control by Google Play leads to an excess of updating of Android apps.

Post-Merger Price Variation Matters, So Why Do Merger Retrospectives Ignore It?

The price effect of past mergers has been extensively researched over the past two decades. The overwhelming majority of these studies estimate the over-time average price effect of the merger. Merger guidelines agree that mergers should be approved if market dynamics, such as entry, eliminate negative welfare effects. Estimating price averages ignores key information about the post-merger dynamics of prices and is unable to identify if post-merger prices eventually revert to pre-merger levels. We provide evidence from a set of Monte Carlo experiments to show how serious this problem might be. Firstly, potentially all the studies that concluded - estimating post-merger over-time averages - that the merger led to a price increase, could have been wrong, and in fact the merger price increase disappeared within a reasonable time. Similarly, up to half of the studies that concluded that the merger did not increase prices could have been wrong in their conclusion.

Entry Limiting Agreements for Pharmaceuticals: Pay-to-Delay and Authorized Generic Deals

Pay-to-delay deals involve a payment from a branded drug manufacturer to a generic maker to delay entry. In return, the generic receives a payment and/or an authorized licensed entry at a later date but before the patent expiration. We examine why such deals are stable. We combine the first mover advantage for the first generic with the ability of the branded manufacturer to launch an authorized generic to show when pay-to-delay deals are an equilibrium outcome. Policy simulations show that removing the ability to launch authorized generics will deter such deals but removing exclusivity period for first generic will not.