Estimating feed utilisation matrices using a cost function approach

In this paper a multiple-output cost function framework is proposed to construct national feed balances or feed utilisation matrices (FUMs). The framework is applied to the Belgian compound feed industry. For estimation purposes a Symmetrie Generalised McFadden (SGM) cost function is selected. The cost function is estimated using readily available time-series data for the period 1962–88. Unlike previous studies based on duality theory, this study exploits the properties of nonjointness in animal feed production to establish a complete FUM. The allocation of feed ingredients among different livestock categories as well as the composition of various compound feeds are identified. Also own- and cross-price elasticities of demand for feed ingredients by type of livestock are reported.     

Interaction styles of street-level workers and motivation of clients: a new instrument to assess discretion-as-used in the case of activation of jobseekers

This paper proposes a re-conceptualization and a measurement instrument for street-level workers’ interaction styles. Interaction styles are a relevant lens giving insight into how discretion is used and how street-level behaviour affects clients’ motivation and engagement. The re-conceptualization builds on a revision of May and Winter's interaction style concept from the perspective of the psychological self-determination theory. Data from 349 caseworkers of the Flemish employment service were collected via an online survey and analysed with factor and latent class analysis. Findings support a four-dimensional interaction style concept and reveal seven types of caseworkers along these dimensions.     

-learning agents in a Cournot oligopoly model

Q-learning is a reinforcement learning model from the field of artificial intelligence. We study the use of Q-learning for modeling the learning behavior of firms in repeated Cournot oligopoly games. Based on computer simulations, we show that Q-learning firms generally learn to collude with each other, although full collusion usually does not emerge. We also present some analytical results. These results provide insight into the underlying mechanism that causes collusive behavior to emerge. Q-learning is one of the few learning models available that can explain the emergence of collusive behavior in settings in which there is no punishment mechanism and no possibility for explicit communication between firms.     

Determinants of foreign direct investment in Cambodia

This paper analyses the determinants of the factors that might influence inward FDI in Cambodia by referring to its economic, geographic, and political characteristics. Using exclusive unbalanced panel data sets during 1995–2005, for both approved and realized FDI for, respectively, seventeen and fifteen home countries, the estimation results show that the determinants of approved FDI and realized FDI are somewhat similar. The FDI home country's GDP, its bilateral trade with the host country and the exchange rate have a positive impact on inward FDI flows into Cambodia. As expected, geographic distance negatively affects the level of FDI inflows in Cambodia.     

Home Employment Effects of EU Firms’ Activities in Central and Eastern European Countries

We examine whether or not affiliate production in Central and Eastern Europe (CEEC) affects factor demand in the EU, at the investing firm and sector levels. Using firm level data, we estimate parent labor demand elasticities for a number of manufacturing sectors, following a flexible cost function approach. We find evidence of inter-sector heterogeneity, but not of a substantially greater impact in low-skilled intensive sectors. Labor demand in the EU is affected by FDI in the CEEC, both at the investing firm and sector levels. It has a significant sector and non-sector component.     

The structure of personal influence of the Belgian holding companies: A quantitative analysis

In this paper, a model of the input-output type is developed which enables the quantitative analysis of personal influence between firms, induced by simultaneous presence in the boards of directors. The empirical part is confined to the system of the most important Belgian holding companies. The structure of personal influence in this system is tackled by means of the calculation of matrices of so-called direct and total communication coefficients. Subsequently, these matrices are dichotomised using appropriate thresholds, and quasi-block-diagonalised. The resulting groups of holding companies are checked with a priori information and with the results of clique detection.