Measuring the Influence of Networks on Transaction Costs Using a Nonparametric Regression Technique

Geraldine Henningsen, Arne Henningsen, Christian H.C.A. Henning

    Research output: Working paper/PreprintWorking paperResearch

    480 Downloads (Pure)


    All business transactions as well as achieving innovations take up resources, subsumed under the concept of transaction costs. One of the major factors in transaction costs theory is information. Firm networks can catalyse the interpersonal information exchange and hence, increase the access to non-public information so that transaction costs are reduced.Many resources that are sacrificed for transaction costs are inputs that also enter the technical production process. As most production data do not distinguish between these two usages of inputs, high transaction costs result in reduced observed productivity. We empirically analyse the effect of networks on productivity using a cross-validated local linear non-parametric regression technique and a data set of 384 farms in Poland. Our empirical study generally supports our hypothesis that networks affect productivity. Large and dense trading networks and dense information networks and household networks have a positive impact on a farm’s productivity. A bootstrapping procedure confirms that this result is statistically significant.
    Original languageEnglish
    PublisherInstitute of Food and Resource Economics, University of Copenhagen
    Number of pages20
    Publication statusPublished - 2013
    SeriesIFRO Working Paper


    • Information networks
    • Transaction Costs
    • Non-parametric estimation
    • Productivity analysis


    Dive into the research topics of 'Measuring the Influence of Networks on Transaction Costs Using a Nonparametric Regression Technique'. Together they form a unique fingerprint.

    Cite this