From Estimation Results to Stylized Facts: Twelve Recommendations for Empirical Research in International Activities of Heterogeneous Firms

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes.You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. licence. www.econstor.eu econometric studies is to uncover stylized facts that hold over space and time, and that can both inspire theoretical models that are based on "realistic" assumptions, and inform policy debates in an evidence-based way. Which results from the thousands of empirical estimates reported in the literature on the micro-econometrics of international firm activities do we consider as convincing? Based on my own experience from the last twenty years I use the opportunity of this lecture to make twelve recommendations that, hopefully, will help to find the right way on the thorny road from estimation results to stylized facts. I will deal with the following topics: comparisons of means vs. comparisons of distributions; extremely different firms, or outliers; unobserved heterogeneity; simultaneous occurrence of differences across quantiles, outliers, and unobserved heterogeneity; heterogeneous effects of international firm activities on firm performance; replication; within-study replication by international research teams; meta-analysis; and talking to practitioners. As a first step this is fine. But one should not stop here. As Moshe Buchinsky (1994: 453) put it: "'On the average' has never been a satisfactory statement with which to conclude a study on heterogeneous populations." An empirical study of heterogeneous firms should look at differences in the whole distribution of the variable under investigation between groups of firms, not only at differences at the mean. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicatedTo illustrate this point I will look at productivity differences between exporters and firms selling on the national market only. The data used were collected in personal interviews with firm owners or top managers. The population covered encompasses all manufacturing establishments with at least 5 employees in the German state of Lower Saxony. From this population a random sample (stratified by industry and size classes) was interviewed. Detailed information on the data set and how it can b...

Export Experience of Managers and the Internationalisation of Firms

“…Following Wagner's () guidelines, we remove firms that are in the bottom and top percentile of the labour productivity distribution to reduce the influence of outliers. However, we do not strictly adhere to Wagner's rule of focusing only on firms that have more than 20 employees, chiefly.…”

“…This ‘knowledge spillover’ and ‘managerial know‐how’ are the novel variables that we introduce in an otherwise standard empirical trade analysis of export initiation (Bernard and Wagner, ; Wagner, ). We find that managerial capabilities are as important as productivity in overcoming the liability of selling on international markets.…”

Credit Constraints and Exports: A Survey of Empirical Studies Using Firm Level Data

“…In most cases, however, firm level data are strictly confidential, and as a rule these data can only be used on computers located inside the statistical agencies that are in charge of collecting the data. The data cannot cross borders, and often they cannot be accessed by citizens 6 For a comprehensive discussion of this topic see Wagner (2011) 8 of a foreign country (who are not liable to jurisdiction in case of violation of privacy in the country where the data are located). Within-study replication using firm level data from various countries, therefore, usually cannot be performed by one author (or a team of authors) from one country.…”

Regional effects or none? Firms' profitability during the Great Recession in Finland

“…Further, we perform the regressions separately for manufacturing sectors and for service sectors (for which export status of firms is available). In line with previous literature (Wagner 2011), we drop the first and last percentile of the observations due to the large effects that possible extreme outlier observations can have on the results. 20…”