Martina Bazzoli (p), Silvia De Poli, Enrico Rettore, Antonio Schizzerotto

In recent years, there has been a growing demand for interventions in support of employment – such as, for example, training courses targeted on the unemployed. However, given the economic crisis and the deterioration of public expenditure, it is imperative for the public authorities to focus on the most effective support measures to ensure tangible benefits and the efficient use of taxpayer’s money.
This paper complements the empirical literature on the impact evaluation of training courses (Kluve 2010; Card et al. 2015). We focus on long vocational training courses for unemployed people implemented in the province of Trento in 2013 and 2014.
Before 2013, applicants who fulfilled some prerequisite were selected to the training programme according to order of registration, until the maximum number of places available was reached (i.e. first come, first served). Starting from 2013, in Trentino, the rules for the selection to training programmes have changed. Applicants, who fulfill the requirements, have to participate to an entry test and only those obtaining a score higher than a threshold are selected into the programme. The new selection rules would allow us to evaluate the effects of the courses, comparing participants and not participants near the cut-off point needed for the admission to the course, using the Regression Discontinuity Design.
Indeed, to try to extend the external validity of our study, we estimate the correlation between the score and the probability of being employed after the training course and we find that the two variables are not correlated. As demonstrated by Angrist and Rokkanen (2016), if there is no correlation between the score, which determined the admission to the programme, and the probability of being employed after x months from the beginning of the course it is possible to expand the analysis to the whole sample of treated and controls.
Our analyses relied on data from two administrative archives and from a survey conducted on people enrolled in the entry test for admission to the training programmes.
We find a positive impact on the probability of being employed one year after the programme, but this effect disappears after 18 months. Therefore, there is an effect only in the short period.

Elena Kotyrlo, Alexander Zaytsev (p)

Using difference-in-difference approach we assess the impact of Russian counter-sanctions - Russian 2014`s agri-food embargo and government development programs - on agricultural dynamics in Russian regions for 2014-2017 period.
Favorable conditions for the growth of Russian agriculture and food industry have been developing since 2014. In addition to 2014`s agri-food embargo and a double ruble devaluation, which strengthened the position of Russian agricultural producers, government expanded agricultural subsidizing by 18% during 2013-2017. At the same time, new constraints for agricultural development have appeared. The first is contracted consumer demand and the general stagnation of the Russian economy due to the double drop in oil prices and EU/US financial sanctions. The second is an increase of production costs due to the strong dependence of Russian agriculture and the food industry on imported machinery, equipment and some intermediate consumption products, reaching up to 80-90%. As a result, Russian agro producers faced multidirectional incentives.
The empirical strategy of our research is based on the difference-in-difference approach (DD) developed by Card (1990) and the use of the generalized synthetic control group (GSCM) developed by Xu (2017) and extending the synthetic control group (Abadie and Gardeazabal 2003; Abadie, Diamond, and Haimmueller 2010).
We consider agriculture as a treatment group. Electric power industry, construction and hotel services, as well as the transport industry are included into the control group. The countersanction («treatment») period includes 2014-2017 years. “Pretreatment” period is 2005-2013. The parallel trend assumption is tested.
The FE-model without weights shows an acceleration of agricultural regional dynamics in the counter-sanction 2014-2017 period by 4.9 percentage points in comparison with the control group dynamics. Positive dynamics in the counter-sanction period is manifested in agriculture precisely against the background of a general slowdown in economic development.
But when weighing on the regional structure of industries (by so taking into account the different regional scales), changes in agricultural growth rates during the counter-sanctions period become insignificant, as well as dynamics in the selected industries in total. This means that, the regions - large producers of agricultural products, experienced no, or much smaller effect from counter-sanctions and devaluation.

Key words: agri-food embargo, agriculture, US/EU-sanctions, Russia, difference-in-difference, synthetic control group

Don Webber (p)

The movement in discourse from regional development to territorial development is a critical shift in thinking that puts local production at the forefront of analysis. It enables and facilitates a seismic shift in thinking from mainstream macroeconomic-type economic growth models that assume away power and institutional forces to a contextually-focussed analysis that cannot escape the interrelationships between local firms, local consumers, local advisers, local policy makers, local trends and fashions, and local development. This article focusses on the role that firm-level Gross Value Added figures, and their territorial aggregate, can play in shaping local development policy, but in order to contend with policy in this area we need to be more explicit about what firm-level GVA figures actually show and what local forces determine them.
The article presents a new ‘territorial productivity model’ that is grounded on firm-level observations, local accounting data, shaped by institutional arrangements, moulded by implicit contracts between firms and their consumers, sculptured by managerial objectives that include demand management, and facilitated by the relative success of marketing campaigns. It takes explicit account of local arrangements and forms of cooperation between local actors including those in conflict and cooperation and supposedly innocent bystanders.
The article contends that inter-territorial productivity levels can be regarded as an indication of the relative success of local producers in creating goods that are in relatively high demand. This is in stark contrast to some understandings of GVA figures that imply greater allocative and productive efficiency or more efficient resource use. Nevertheless, local authorities and territorial policy makers continue to be charged with enhancing GVA productivity figures due to the connotations it has with employment and wealth. The article highlights that the use of firm-level and territorially-aggregated GVA data to guide territorial economic policy will be biased and should be treated with care. It underscores that territorial productivity levels reflect demand aspects of an ostensibly supply-side measure and that there is impurity in the wage-led growth concepts.