Bayesian Hypothesis Testing: a Reference Approach

For any probability model M={p(x|θ, ω), θεΘ, ωεΩ} assumed to describe the probabilistic behaviour of data xεX, it is argued that testing whether or not the available data are compatible with the hypothesis H₀={θ=θ₀} is best considered as a formal decision problem on whether to use (a₀), or not to use (a₀), the simpler probability model (or null model) M₀={p(x|θ₀, ω), ωεΩ}, where the loss difference L(a₀, θ, ω) –L(a₀, θ, ω) is proportional to the amount of information δ(θ₀, ω), which would be lost if the simplified model M₀ were used as a proxy for the assumed model M. For any prior distribution π(θ, ω), the appropriate normative solution is obtained by rejecting the null model M₀ whenever the corresponding posterior expectation ∫∫δ(θ₀, θ, ω)π(θ, ω|x)dθdω is sufficiently large. Specification of a subjective prior is always difficult, and often polemical, in scientific communication. Information theory may be used to specify a prior, the reference prior, which only depends on the assumed model M, and mathematically describes a situation where no prior information is available about the quantity of interest. The reference posterior expectation, d(θ₀, x) =∫δπ(δ|x)dδ, of the amount of information δ(θ₀, θ, ω) which could be lost if the null model were used, provides an attractive nonnegative test function, the intrinsic statistic, which is invariant under reparametrization. The intrinsic statistic d(θ₀, x) is measured in units of information, and it is easily calibrated (for any sample size and any dimensionality) in terms of some average log‐likelihood ratios. The corresponding Bayes decision rule, the Bayesian reference criterion (BRC), indicates that the null model M₀ should only be rejected if the posterior expected loss of information from using the simplified model M₀ is too large or, equivalently, if the associated expected average log‐likelihood ratio is large enough. The BRC criterion provides a general reference Bayesian solution to hypothesis testing which does not assume a probability mass concentrated on M₀ and, hence, it is immune to Lindley's paradox. The theory is illustrated within the context of multivariate normal data, where it is shown to avoid Rao's paradox on the inconsistency between univariate and multivariate frequentist hypothesis testing.     

Doing good matters to consumers: the effectiveness of humane-oriented CSR appeals in cross-cultural standardized advertising campaigns

This investigation contributes to the advancement of international advertising theory by developing a model of cross-cultural consumer responses to humane-oriented corporate social responsibility (CSR) appeals in advertising and by testing it using structural equation modelling with data from six countries (Germany, Switzerland, Austria, France, USA, and Chile). The model includes ad-, advertiser-, product-, and behaviour-related variables.

Based on survey data from 924 subjects, the model received support in all six countries which suggests that humane orientation is a favourable cultural dimension for advertising purposes, even in countries in which humane orientation is less highly valued. The higher the perceived degree of humane orientation in an advertisement, the more positive was the evaluation of the ad, and the more likely subjects were to evaluate the advertiser as being socially responsible. These two variables positively influenced attitudes towards the product and, consequently, behavioural intentions. Some support was found for a moderating influence of country on the relationship between perception of humane orientation in the ad and the evaluation of that ad and on the relationship between perception of humane orientation in the ad and perceived degree of social responsibility of the advertiser. Implications for the use of humane-oriented CSR appeals in international advertising are discussed. Results suggest that international advertisers may pursue a standardized campaign in multiple markets when incorporating humane-oriented CSR appeals. Limitations and directions for further research are addressed.     

Trade, technology, and the rise of the service sector: The effects on US wage inequality

This paper uses a multi-sector version of the Ricardo-Viner model of international trade to quantify empirically the effects of technological changes, international trade, changes in the sectoral composition of the economy, and other factors on the US wage premium. The main finding of the paper is that changes in the sectoral composition of the economy were the most important force behind the widening of the wage gap, accounting for about 60% of the relative increase in wages of skilled workers between 1970 and 1996. In essence, capital was reallocated to sectors where it is relatively complementary to skilled workers.     

Do droughts have long-term effects on water consumption? Evidence from the urban area of Barcelona

This article examines the long-term effects of droughts on water consumption using data of municipalities of the urban area of Barcelona. Two important characteristics of the sample of municipalities are the relatively low water consumption in the pre-drought period and the fact that indoor uses are clearly predominant. Controlling for prices, income and various socio-demographic factors, we find a reduction in water consumption not only during the drought episode but also in subsequent periods. These permanent reductions in water use may be explained by household behavioural changes, technological changes or structural water policy changes.     

Capital Budgeting in Multidivision Firms: Information, Agency, and Incentives

We examine optimal capital allocation and managerial compensationin a firm with two investment projects (divisions) each runby a risk-neutral manager who can provide (i) (unverifiable)information about project quality and (ii) (unverifiable) accessto value-enhancing, but privately costly resources. The optimalmanagerial compensation contract offers greater performancepay and a lower salary when managers report that their projectis higher quality. The firm generally underinvests in capitaland managers underutilize resources (relative to first-best).We also derive cross-sectional predictions about the sensitivityof investment in one division to the quality of investment opportunitiesin the other division, and the relative importance of division-leveland firm-level performance-based pay in managerial compensationcontracts.     

Capital Structure and Regulation: Do Ownership and Regulatory Independence Matter?

We study the effect of ownership structure and regulatory independence on the interaction between capital structure and regulated prices using a comprehensive panel data of publicly traded European utilities. We find that firms in our sample tend to have a higher leverage if they are privately controlled and regulated by an independent regulatory agency. Moreover, the leverage of these firms has a positive and significant effect on their regulated prices, but not vice versa. Our results are consistent with the theory that privately controlled regulated firms use leverage strategically to obtain better regulatory outcomes.     

How does owners’ exposure to idiosyncratic risk influence the capital structure of private companies?

This paper identifies the owner's exposure to idiosyncratic risk as an important determinant of the demand for loans and the capital structure of private companies. The analysis is based on a sample of small and medium-sized companies from the United States. The exposure to idiosyncratic risk is approximated by the share of personal net worth invested in one company (SNWI). Exposure to idiosyncratic risk increases the cost of equity capital, since higher equity returns are required as compensation. This therefore makes bank financing more attractive. We find that SNWI increases both the demand for new bank loans and leverage substantially.