On Antitrust Enforcement and the Deterrence of Collusive Behaviour

This paper examines an antitrust enforcement policy of using significant price changes in an industry as a sorting mechanism for the allocation of resources devoted to policing collusion. That is, in either responding to complaints or initiating investigations on their own, I examine the issue of whether an antitrust enforcement policy of inferring possible collusion from significant prices changes is effective in deterring collusion given that antitrust officials have no direct knowledge of the costs of individual firms. Using the imperfect information repeated game of Green and Porter (1984), I show that this investigation strategy if coupled with uniform costs being borne by firms, can reduce the expected profits from the collusive agreement: however, unless the punishment is large enough, it will be ineffective in reducing the frequency of collusion. More importantly, it can have the undesirable effect of reducing the output agreed to by firms, if firms choose quantities, or raise collusive prices if firms are choosing price. Moreover, if the enforcement policy is anticipated by firms, the punishment mechanism adopted to support collusion will be altered to offset the policy.     

Collusion Under Imperfect Monitoring with Asymmetric Firms

We explore the effects of asymmetries in capacity constraints on collusion where market demand is uncertain and where firms’ sales and prices are private information. We show that all firms can infer when at least one firm's sales are below some firm‐specific ‘trigger level.’ When firms use this public information to monitor the collusive agreement, price wars may occur on the equilibrium path. Symmetry facilitates collusion but, if price wars are sufficiently long, then the optimal collusive prices of symmetric capacity distributions are lower on average than the competitive prices of asymmetric capacity distributions. We draw conclusions for merger policy.     

The Small Firm in a Quantity Choosing Game: Some Experimental Evidence

We demonstrate with a grim trigger strategy that the small firm should be more willing to collude tacitly as its market share declines; large firms should be less willing to cooperate. The small firm is not a maverick. The intensity of rivalry between two firms with asymmetric market shares is studied in experimental markets. Treatments give duopolists (1) 50% shares, (2) a 60 or 40% share, and (3) an 80 or 20% share. Choices for the small firm in the latter treatments are not significantly larger than the collusive choice. Irrespective of relative size, firms in all three market environments exhibit collusive behavior.     

Identifying Participants in a Price-fixing Conspiracy: Output & Market Share Tests Reexamined

If there is a cartel agreement among a subset of firms in an industry, it should be predicted that all firms in that industry will increase prices. Nevertheless, industry prices alone should not indicate that a particular firm is guilty of that conspiracy. According to the output test and its market share variant – proposed by Blair and Romano – if the output or the market share of the firm that claims to be innocent in the collusive activity rises in response to the price increase, that firm's claim should be accepted as true. Using a collusive variant of the dominant firm model, this paper shows that these are not robust tests to reveal either innocence or guilt, and characterizes cases where they may pardon a guilty firm (Type I error) or indict an innocent firm (Type II error). This paper also shows that a market share test can not be used to prove a dominant firm's intent for predatory pricing.     

Capacity signals and entry deterrence

If potential entrant firms are well informed they will generally not believe incumbent's threats to expand output when experiencing entry. But this expectation underlies most excess capacity models. We demonstrate an asymmetric information equilibrium in which potential entrants rationally fear output expansion by oligopolists with excess capacity. Less effective collusive oligopolies may be destabilized (expanding output) upon entry. One symptom of less effective collusion is excess capacity. Hence excess capacity becomes a signal of the potential for output expansion. In a rational expectations equilibrium this signal may also be mimicked by oligopolies which would not otherwise carry excess capacity.     

COLLUSION AND RESEARCH JOINT VENTURES*

I examine the question whether cooperation in R&D among firms producing similar products leads to product market collusion. Suppose that firms engage in a stochastic R&D race while maintaining the collusive equilibrium in a repeated‐game framework. Innovation under non‐cooperative R&D leads to an inter‐firm asymmetry, destabilizing collusion in pre‐discovery and post‐discovery periods. Innovation sharing under cooperative R&D preserves the symmetry and also increases total profit, thereby facilitating collusion. However, welfare may increase with cooperative R&D. I also examine the condition for collusion under licensing and compare the results.     

Modeling oligopolistic interaction

It is shown that standard versions of overt and tacit collusion under quantity-setting oligopoly are formally equivalent. The two approaches are combined to model collusive behavior when firms expect rivals to react to output changes. The combined model yields a structural equation for firm-level price-cost margins that is consistent with empirical studies using line-of-business data.     

ALTERNATING MONOPOLY AND TACIT COLLUSION*

This paper considers the use of the alternating monopoly strategy (AMS) as a (tacit) collusion device. We show that firms may choose this strategy in particular environments, when other collusive strategies are also feasible. In particular, we stress how the presence of an observable move (entry), distinct from the competitive stage (price setting), can serve as a coordination device, reducing monitoring costs in incomplete information environments. The paper thus shows that AMS may be preferable to the classic market sharing strategy (MSS) and in some cases it is the only collusive equilibrium.     

Single Bidders and Tacit Collusion in Highway Procurement Auctions

Collusion in auctions can take different forms, such as refraining from bidding. Certain aspects of highway procurement auctions facilitate collusive outcomes. We collect data on asphalt paving auctions conducted in Kentucky from 2005-2007. We determine the potential service area of each asphalt plant and potential bidders for each paving project. We analyze firms’ bid participation decisions, including variables affecting costs as well as competitive and strategic effects. In many geographic markets where firms face only a few identifiable rivals, county boundaries serve as a coordinating mechanism for softening competition, significantly influencing firms’ decisions whether and how much to bid.     

The R&D performance in Taiwan's electronics industry: a longitudinal examination

R&D investment has been widely regarded as an important input for firms, particularly for high‐tech firms, to achieve competitive advantage within their industry. Hence, a number of high‐tech firms are now investing substantial amounts into R&D. Since R&D efforts enable firms to raise the competitive advantage, one noticeable and interesting issue expected to know is the degree to which R&D investment influences firm output performance. In Taiwan, much greater emphasis is also being placed into R&D investment in the high‐tech industries; however, R&D output performance has never been seriously examined within this sector. Since the island's electronics industry is widely regarded as the most promising industry in the ‘high‐tech sector’, and is expected to place greatest emphasis on its R&D efforts, we take the electronics firms as our analytical sample. This paper therefore sets out to estimate the impact of R&D on firm performance, in terms of productivity growth and the rate of return on investment, within the electronics industry in Taiwan, whilst also examining the Schumpeterian hypothesis, that R&D performance is an increasing function of firm size. Our examination of R&D performance is based on a panel sample of 83 large electronics firms, completely balanced over the period from 1994 to 2000, with series data of R&D capital also being constructed. Based upon the extended Cobb‐Douglas production function, a random effects model is developed with the estimations revealing that the output elasticity of R&D is around 0.19 and the average rate of return on R&D is around 22%. These findings clearly demonstrate that investment in R&D by these electronics firms has had an impact on their competitive advantage. Compared to the findings of previous studies, where the analytical unit of data was at firm level, here the rate of return on R&D is consistent with similar estimates for the US and UK, but lower than those for Japan. However, our estimations do not provide support for the hypothesis that the impact of R&D on productivity is an increasing function of firm size.     

SCREENING AND MERGER ACTIVITY*

In our paper, the target of a proposed merger, by setting a reserve price, is able to screen prospective acquirers according to their (expected) ability to generate merger‐specific synergies. Both empirical evidence and many merger models suggest that the difference between high and low‐synergy mergers becomes smaller during booms. Thus, a target's opportunity cost for sorting out relatively less fitting acquirers increases and, hence, targets screen less tightly during booms, which leads to a hike in merger activity. Our screening mechanism not only predicts that merger activity is intense during booms and subdued during recessions but is also consistent with other stylized facts about takeovers and generates novel testable predictions.     

Vertical integration and downstream collusion

We investigate the effect of a vertical merger on downstream firms’ ability to collude in a repeated game framework. We show that a vertical merger has two main effects. On the one hand, it increases the total collusive profits, increasing the stakes of collusion. On the other hand, it creates an asymmetry between the integrated firm and the unintegrated competitors. The integrated firm, accessing the input at marginal cost, faces higher profits in the deviation phase and in the non-cooperative equilibrium, which potentially harms collusion. As we show, the optimal collusive profit-sharing agreement takes care of the increased incentive to deviate of the integrated firm, while optimal punishment erases the difficulty related to the asymmetries in the non-cooperative state. As a result, vertical integration generally favors collusion.     

Collusion Under Risk Aversion and Fixed Costs

We analyze collusion under demand uncertainty by risk‐averse cartels that care about the utility derived from profits. With sufficient risk aversion and non‐trivial fixed operating costs, it becomes difficult for cartels to collusively restrict output both when demand is low and marginal dollars are highly valued, and when demand is high and potential defection profits are high: output relative to monopoly levels becomes a U‐shaped function of demand. Greater risk aversion or higher fixed operating costs make collusion more difficult to support in recessions, but easier to support in booms.     

Scientific foundation,patents, and new product introductions of biotechnology and pharmaceutical firms

Absorptive capacity (ACAP) has long been used in the research of technology firms as an indication of knowledge acquisition. This paper links knowledge input using new product introductions (NPI), to commercial output, addressing one of the key criticisms of research and development expenditures as a measure of an organization's effort. We propose that firms with a strong basic science foundation such as biology and chemistry seek to advance their scientific knowledge in addition to developing and selling innovative products. In contrast, corporations with a more applied focus forego fundamental scientific advancement and concentrate their actions on the development of innovative products. Results from a sample of 487 firms over a 10‐year period indicated that firms with an applied science foundation had greater numbers of NPIs, both breakthrough and innovative, suggesting that a solutions‐based approach had greater value in terms of innovative output. The pursuit of science may build knowledge, but a clear link between increased ACAP and innovative output was not found. Our results also suggest that basic science firms have a positive relationship with both breakthrough and incremental NPIs using merger and acquisition activities. Applied science firms found university alliances useful in generating NPIs.     

Do Multi‐Plant Firms Reduce Misallocation? Evidence from Canadian Manufacturing

Using Canadian plant‐level data, this paper shows that, depending on the industry, the differences in the average plant‐level productivity and cross‐plant allocation of resources between multi‐plant and single‐plant firms account for 1 to 15 per cent of the industry‐level TFP. A large part of this contribution stems from more efficient cross‐plant allocation of resources, measured by the covariance between plant size and productivity, in the pool of plants in multi‐plant firms compared to the pool of plants in single‐plant firms. There is less dispersion in the marginal products of the inputs, and thus less misallocation, in industries in which multi‐plant firms account for a larger share of output. The patterns found in the cross‐plant distribution of productivity and size are also consistent with better allocative efficiency among plants in multi‐plant firms than among plants in single‐plant firms.     

Service Regulations,Input Prices and Export Volumes: Evidence from a Panel of Manufacturing Firms

Using a panel of firm‐level data from Spanish manufacturers, this study shows that better service regulation reduces the price of intermediate inputs paid by downstream firms. The beneficial cost effects of services reforms extend to both large and small‐to‐medium sized corporations (SME’s), but the former tend to enjoy greater gains. This feature also manifests itself in international markets. We find evidence of an input cost channel through which service regulations affect the volume of exports of large manufacturers, while the evidence of such a channel is weaker for SME’s. Our estimates indicate that, from 1991 to 2007, large firms increased their volume of exports by an average of 20 per cent as a result of the direct input cost effect of services reforms, such that the firms that benefited the most typically belonged to industries more dependent on service inputs. Furthermore, convergence to the ‘best practice’ regulatory framework in services would have raised exports at least by an additional nine per cent. We conclude that firm size is relevant for the connection between services reforms, intermediate input prices and export volumes.     

Flagging cartel participants with deep learning based on convolutional neural networks

Adding to the literature on the data-driven detection of bid-rigging cartels, we propose a novel approach based on deep learning (a subfield of artificial intelligence) that flags cartel participants based on their pairwise bidding interactions with other firms. More concisely, we combine a so-called convolutional neural network for image recognition with graphs that in a pairwise manner plot the normalized bids of some reference firm against the normalized bids of any other firms participating in the same tenders as the reference firm. Based on Japanese and Swiss procurement data, we construct such graphs for both collusive and competitive episodes (i.e when a bid-rigging cartel is or is not active) and we use a subset of graphs to train the neural network such that it learns distinguishing collusive from competitive bidding patterns. With the remaining graphs, we test the neural network’s out-of-sample performance in correctly classifying collusive and competitive bidding interactions. We obtain a very decent average accuracy of around 95% or slightly higher when either applying the method within Japanese, Swiss, or mixed data (in which Swiss and Japanese graphs are pooled). When using data from one country for training to test the trained model’s performance in the other country (i.e. transnationally), predictive performance decreases (likely due to institutional differences in procurement procedures across countries), but often remains satisfactorily high. All in all, the generally quite high accuracy of the convolutional neural network despite being trained in a rather small sample of a few 100 graphs points to a large potential of deep learning approaches for flagging and fighting bid-rigging cartels.     

Interfirm Bundled Discounts as a Collusive Device

This paper investigates whether and how firms competing in price with homogeneous goods (i.e., Bertrand competitors) can achieve supernormal profits using interfirm bundled discounts. By committing to offering price discounts conditional on the purchase of a specific brand of other differentiated good, the homogeneous good suppliers can separate consumers into distinct groups. Such brand‐specific discounts help the firms relax competition and attain a collusive outcome. Consumers become worse off due to higher effective prices. Our result shows that in oligopolies it is feasible to leverage other's market power without excluding rivals.     

Cartel pricing dynamics with cost variability and endogenous buyer detection

This paper characterizes collusive pricing patterns when buyers may detect the presence of a cartel. Buyers are assumed to become suspicious when observed prices are anomalous. We find that the cartel price path is comprised of two phases. During the transitional phase, price is generally rising and relatively unresponsive to cost shocks. During the stationary phase, price responds to cost but is much less sensitive than under non-collusion or simple monopoly; a low price variance may then be a collusive marker. Compared to when firms do not collude, cost shocks take a longer time to pass-through to price.     

Cooperative R&D and the Value of the Firm

In this paper, we investigate the stock price responses of listed firms in the U.S. markets to announcements of R & D collaborations. We find that abnormal returns of stocks are significantly positive after R & D collaborations are announced. The positive stock price response towards the R & D cooperation initiations can be partially explained by the nature of the collaborations and the characteristics of the participating firms. We also find that the stock prices of rival firms respond negatively to announcements of R & D cooperation. This result seems to support the hypothesis that cooperative R & D improves economic efficiency of the cooperative firms that gain competitive advantage. We do not find evidence supporting the hypothesis that R & D cooperation creates collusive, anticompetitive effects in the product market.