Semiparametric tests of conditional moment restrictions under weak or partial identification

We propose two new semiparametric specification tests which test whether a vector of conditional moment conditions is satisfied for any vector of parameter values θ₀. Unlike most existing tests, our tests are asymptotically valid under weak and/or partial identification and can accommodate discontinuities in the conditional moment functions. Our tests are moreover consistent provided that identification is not too weak. We do not require the availability of a consistent first step estimator. Like Robinson [Robinson, Peter M., 1987. Asymptotically efficient estimation in the presence of heteroskedasticity of unknown form. Econometrica 55, 875–891] and many others in similar problems subsequently, we use k-nearest neighbor (knn) weights instead of kernel weights. The advantage of using knn weights is that local power is invariant to transformations of the instruments and that under strong point identification computation of the test statistic yields an efficient estimator of θ₀ as a byproduct.     

On the lowest-winning-bid and the highest-losing-bid auctions

Theoretical models of multi-unit, uniform-price auctions assume that the price is given by the highest losing bid. In practice, however, the price is usually given by the lowest winning bid. We derive the equilibrium bidding function of the lowest-winning-bid auction when there are k objects for sale and n bidders with unit demand, and prove that it converges to the bidding function of the highest-losing-bid auction if and only if the number of losers n−k gets large. When the number of losers grows large, the bidding functions converge at a linear rate and the prices in the two auctions converge in probability to the expected value of an object to the marginal winner.     

Exact and asymptotic tests for possibly non-regular hypotheses on stochastic volatility models

We study the problem of testing hypotheses on the parameters of one- and two-factor stochastic volatility models (SV), allowing for the possible presence of non-regularities such as singular moment conditions and unidentified parameters, which can lead to non-standard asymptotic distributions. We focus on the development of simulation-based exact procedures–whose level can be controlled in finite samples–as well as on large-sample procedures which remain valid under non-regular conditions. We consider Wald-type, score-type and likelihood-ratio-type tests based on a simple moment estimator, which can be easily simulated. We also propose a C(α)-type test which is very easy to implement and exhibits relatively good size and power properties. Besides usual linear restrictions on the SV model coefficients, the problems studied include testing homoskedasticity against a SV alternative (which involves singular moment conditions under the null hypothesis) and testing the null hypothesis of one factor driving the dynamics of the volatility process against two factors (which raises identification difficulties). Three ways of implementing the tests based on alternative statistics are compared: asymptotic critical values (when available), a local Monte Carlo (or parametric bootstrap) test procedure, and a maximized Monte Carlo (MMC) procedure. The size and power properties of the proposed tests are examined in a simulation experiment. The results indicate that the C(α)-based tests (built upon the simple moment estimator available in closed form) have good size and power properties for regular hypotheses, while Monte Carlo tests are much more reliable than those based on asymptotic critical values. Further, in cases where the parametric bootstrap appears to fail (for example, in the presence of identification problems), the MMC procedure easily controls the level of the tests. Moreover, MMC-based tests exhibit relatively good power performance despite the conservative feature of the procedure. Finally, we present an application to a time series of returns on the Standard and Poor’s Composite Price Index.     

Heteroskedasticity, autocorrelation, and spatial correlation robust inference in linear panel models with fixed-effects

This paper develops an asymptotic theory for test statistics in linear panel models that are robust to heteroskedasticity, autocorrelation and/or spatial correlation. Two classes of standard errors are analyzed. Both are based on nonparametric heteroskedasticity autocorrelation (HAC) covariance matrix estimators. The first class is based on averages of HAC estimators across individuals in the cross-section, i.e. “averages of HACs”. This class includes the well known cluster standard errors analyzed by Arellano (1987) as a special case. The second class is based on the HAC of cross-section averages and was proposed by Driscoll and Kraay (1998). The ”HAC of averages” standard errors are robust to heteroskedasticity, serial correlation and spatial correlation but weak dependence in the time dimension is required. The “averages of HACs” standard errors are robust to heteroskedasticity and serial correlation including the nonstationary case but they are not valid in the presence of spatial correlation. The main contribution of the paper is to develop a fixed-b asymptotic theory for statistics based on both classes of standard errors in models with individual and possibly time fixed-effects dummy variables. The asymptotics is carried out for large time sample sizes for both fixed and large cross-section sample sizes. Extensive simulations show that the fixed-b approximation is usually much better than the traditional normal or chi-square approximation especially for the Driscoll-Kraay standard errors. The use of fixed-b critical values will lead to more reliable inference in practice especially for tests of joint hypotheses.     

The Neuer Markt as an institution of creation and destruction

The market for publicly traded equity is the heart of a modern capitalist system. In particular, the market for newly listed firms is in turn a bellwether for the public equity market because the entry gives firms expanded access to capital, allowing them to emerge and grow. On the other hand, stock markets also serve as an institution to reallocate productive resources from non-surviving to surviving firms. We describe the Neuer Markt—a special segment of the Frankfurt Stock Exchange in Germany—as an institution to foster the creation of new firms and their destruction.

Do external diversity practices boost focal firm performance? The case of supplier diversity

Based on the resource-based view, we propose that external diversity practices such as supplier diversity may affect firm performance. We find that the relationship between supplier diversity and short-term performance (i.e. productivity) is moderated by context such that firms in declining industries experience positive productivity effects while firms in munificent industries witness negative effects. For longer-term profitability (i.e. Tobin's q), we do not find support for a positive relationship between supplier diversity and long-term performance. However, positive supplier diversity effects emerge in munificent environments. Overall, in support of the strategic human resource management approach, we conclude that the effect of external supplier diversity on firm performance is contingent upon environmental munificence, which documented the necessity to include supplier diversity as a relevant component of a comprehensive diversity and equality management system.     

Construction of D-optimal s 1 × s 2 × s 3 factorial designs using graph theory

In this paper, the maximum determinant of the associated 0-1 matrix in D-Optimal saturated main effect plans for 3× s_2 × s_3 factorials, is derived by the use of Graph theory and Combinatorics. The present work is related to a problem suggested by Chatterjee and Narasimhan (2002). Using the theoretical results, we also give the designs for s₃≥s₂ + 1. This research was supported by the State Scholarships Foundation of Greece.     

Sustaining Collusion in Growing Markets

The impact of demand growth on the collusion possibilities is investigated in a Cournot supergame where market growth may trigger future entry and the collusive agreement is enforced by the most profitable ‘grim trigger strategies’ available. It is shown that even in situations where perfect collusion can be sustained after entry, coping with a potential entrant in a market which is growing over time may completely undermine any pre‐entry collusive plans of the incumbent firms. This is because, before entry, a deviation and the following punishment phase may become more attractive thanks to their additional effect in terms of delaying entry.     

The economics analysis of a <Emphasis Type="Italic">Q</Emphasis>-learning model of cooperation with punishment and risk taking preferences

The aim of this paper is to better understand how cooperation mechanisms work in the context of a Q-learning model. We apply a learning reinforcement model to analyse the conditions needed to have a stable cooperative equilibrium when people take part in a common project and could take advantages of free-riding. Our results show that a stable equilibrium can be reached thank to mechanisms of punishment, but the final result strongly depends on the risk-taking individuals’ preferences. In particular, we find that the penalties will be effective only with people having high exploration rates,namely with people able to adapt their strategies and learn to cooperate. Otherwise, it is possible to have an unstable equilibrium with cooperation until individuals have a very high intrinsic motivation to cooperate, whatever the others do.     

The stock performance of America's 100 Best Corporate Citizens

We consider the stock performance of America's 100 Best Corporate Citizens following the annual survey by Business Ethics. We examine both possible short-term announcement effects around the time of the survey's publication, and whether longer-term returns are higher for firms that are listed as good citizens. We find some evidence of a positive market reaction to a firm's presence in the Top 100 firms that are made public, and that holders of the stock of such firms earn small abnormal returns during an announcement window. Over the year following the announcement, companies in the Top 100 yield negative abnormal returns of around 3%. However, such companies tend to be large and with stocks exhibiting a growth style, which existing studies suggest will tend to perform poorly. Once we allow for these firm characteristics, the poor performance of the highly rated firms declines. We also find companies that are newly listed as good citizens and companies in the Top 100 but outside the S&P 500 can provide considerable positive abnormal returns to investors, even after allowing for their market capitalization, price-to-book ratios, and sectoral classification.     

Tests of risk premia in linear factor models

We show that statistical inference on the risk premia in linear factor models that is based on the Fama–MacBeth (FM) and generalized least squares (GLS) two-pass risk premia estimators is misleading when the β$\beta$’s are small and/or the number of assets is large. We propose novel statistics, that are based on the maximum likelihood estimator of Gibbons [Gibbons, M., 1982. Multivariate tests of financial models: A new approach. Journal of Financial Economics 10, 3–27], which remain trustworthy in these cases. The inadequacy of the FM and GLS two-pass t$t$/Wald statistics is highlighted in a power and size comparison using quarterly portfolio returns from Lettau and Ludvigson [Lettau, M., Ludvigson, S., 2001. Resurrecting the (C)CAPM: A cross-sectional test when risk premia are time-varying. Journal of Political Economy 109, 1238–1287]. The power and size comparison shows that the FM and GLS two-pass t$t$/Wald statistics can be severely size distorted. The 95% confidence sets for the risk premia in the above-cited work that result from the novel statistics differ substantially from those that result from the FM and GLS two-pass t$t$-statistics. They show support for the human capital asset pricing model although the 95% confidence set for the risk premia on labor income growth is unbounded. The 95% confidence sets show no support for the (scaled) consumption asset pricing model, since the 95% confidence set of the risk premia on the scaled consumption growth consists of the whole real line, but do not reject it either.     

Collusive Networks in Market‐Sharing Agreements in the Presence of an Antitrust Authority

This paper examines how the presence of an antitrust authority (AA) affects market‐sharing agreements made by firms. These agreements prevent firms from entering each other’s markets. The set of agreements defines a collusive network, which is pursued by antitrust authorities. This paper shows that in the absence of an AA, a network is stable if its alliances are large enough, and in the presence of an AA, more competitive structures can be sustained through bilateral agreements. Antitrust laws may have a procompetitive effect, as they give firms in large alliances more incentives to cut their agreements at once.     

Inference with dependent data using cluster covariance estimators

This paper presents an inference approach for dependent data in time series, spatial, and panel data applications. The method involves constructing t and Wald statistics using a cluster covariance matrix estimator (CCE). We use an approximation that takes the number of clusters/groups as fixed and the number of observations per group to be large. The resulting limiting distributions of the t and Wald statistics are standard t and F distributions where the number of groups plays the role of sample size. Using a small number of groups is analogous to ‘fixed-b’ asymptotics of [Kiefer and Vogelsang, 2002] and [Kiefer and Vogelsang, 2005] (KV) for heteroskedasticity and autocorrelation consistent inference. We provide simulation evidence that demonstrates that the procedure substantially outperforms conventional inference procedures.     

Collusive equilibria with switching costs: The effect of consumer concentration

I study the ability of two competing firms to set collusive prices in markets where consumers have switching costs. In consumer markets (with a small number of consumers), I find an antifolk result in which collusion, in the presence of switching costs, does not arise with patient firms. Patient firms compete aggressively and consumers expect a large utility. A collusive equilibrium is unstable because a deviating firm incorporates the future consumer utility in its deviating price. Also, consumers have a strategic impact so, with the prospects of large utility, they decide to switch to destabilize the firms' collusive agreement. These results do not eventuate in markets with a large number of consumers. In mass markets (a continuum of consumers), a single consumer lacks a strategic impact to destabilize a collusive agreement and a deviating firm cannot appropriate the consumer utility when deviating from collusion. Collusion, then, becomes straightforward to achieve. We show that for any number of consumers, switching costs make collusion easy to sustain.     

Multiple capital inputs, Q, and investment spending

Despite their explicit treatment of dynamics and solid theoretical basis, investment models based on the Brainard-Tobin Q have recorded a generally disappointing empirical performance. When the Q model is expanded to recognize the possibility that the value of the firm depends on two or more capital inputs with differing adjustment cost technologies, the econometric equation following from optimizing behavior includes Q as well as a set of additional explanatory variables. The importance of these omitted variables is assessed, and the capital homogeneity assumption for equipment and structures implicit in Conventional Q models is rejected. The Multi-Capital Q model is then extended in two ways: (i) adding inventory, research and development, and labor as quasi-fixed factors and (ii) exploring the sensitivity of the instrumental variables estimates to normalization. We conclude that the Multi-Capital Q model is a useful extension that overcomes an important omitted variables problem in the Conventional Q framework.     

Experimentelle Forschung in der Budgetierung – Lügen, nichts als Lügen?

Budgeting is one of the most important instruments for planning, control and coordination in decentralized firms. Participative budgeting allows a firm’s headquarters to make use of the divisional managers’ superior information when formulating budgets, but, simultaneously, gives the managers the opportunity to benefit from their superior information and to build slack into their budgets. This paper presents experiments analyzing the behavior in participative budgeting processes and explores the question how the actors’ behavior is influenced by pure self-interest on the one hand and non-standard preferences e.g. for the well-being of others or for honesty on the other hand.     

Implementing efficient market structure

Abstract. This article studies the design of optimal mechanisms to regulate entry in natural oligopoly markets, assuming the regulator is unable to control the behavior of firms once they are in the market. We adapt the Clarke–Groves mechanism, characterize the optimal mechanism that maximizes the weighted sum of expected social surplus and expected tax revenue, and show that these mechanisms avoid budget deficits and prevent excessive entry. Received: 7 May 2001 / Accepted: 24 June 2002 We would like to thank seminar participants at Bonn and Berlin, in particular Peter Bank, Wieland Müller, and Urs Schweizer, the two anonymous referees, and the associate editor for most useful and exceptionally detailed comments. Financial support was received by the Deutsche Forschungsgemeinschaft, SFB 373 (“Quantifikation und Simulation ?konomischer Prozesse”), Humboldt–Universit?t zu Berlin.