Carlo Ciccarelli (p), Matteo Gomellini, Paolo Sestito

Italy is among the countries most subject to an aging process that is reducing, and will further reduce in the future, the share of youths in its overall population. Somehow paradoxically, youths, albeit having acquired a status of “scarce factor”, have not improved very much their relative standing in the society (see for evidence Rosolia and Torrini, 2016) .

This paper presents first evidences on the relationship between demographics and manufacturing productivity growth in Italy over the last 100 years.

We focus upon the two decades at the beginning and the end of such a period (1901-1911 and 2001-2011), characterized by very different demographic (a growing population with a large share of children and youths versus a population with more deaths than births and a shrinking share of youths) and economic conditions (a booming decade versus a period of stagnation, culminated in a large crisis).

We exploit, within each decade, the differences existing across provinces in both demographic and economic conditions and by the use of an IV approach – so as to take into account that the age composition of the population may be affected, through migrations and other channels, by economic factors – we estimate a causal link stemming from the age structure of the population to productivity growth.

In both periods, a higher share of youths turns out to be associated to a higher growth of productivity in manufacturing: the effect is sizable and the impact is similar over both the decades, although weaker in the most recent one.

According to the estimated coefficients, an increase of five percentage points in the initial value of the share of young people was associated to an increase in manufacturing productivity growth of about 0.8 percentage points per year between 1901-1911 and 0.2 percentage points per year between 2001-2011.

We used the estimated coefficients to quantify how much in the next forty years labor productivity will be affected by the declining share of the people aged 20-40 (IMF, 2015, performs similar projections for aggregate productivity).
In the hypothetical scenario where the share of young people is the sole driver of productivity in manufacturing, the latter will decrease by about four per cent each decade between 2011 and 2051.

Frank Van Oort, Martijn Burger, Spyridon Stavropoulos (p)

Smart specialisation is among the most prominent policy concepts in Europe for stimulating regional economic growth. A burgeoning empirical literature on one of the central building blocks of the concept – related variety – confirms that this concept is guiding in regional structural change and resilience, yet a large heterogeneity in effects over regions, industries, and innovative technologies suggests that more insight is needed in aspects explaining this heterogeneity. Recent research therefore turns to micro-foundations of relatedness and productivity, and captures regional heterogeneity in terms of technological readiness and institutions.
In this paper, we relate regional structural composition – related and unrelated variety – to firm level productivity in European regions, applying a Cobb-Douglas production function framework and using firm-, industry, and regional-level mixed hierarchical (multilevel) models. Data are obtained from the ORBIS-Amadeus database, providing financial and economic information on firms in almost all European countries, including balance sheets and income statements items, and a wide range of performance indices. Following earlier research, regions are characterised to their capacities in terms of knowledge accessibility, knowledge absorption and knowledge diffusion. Applying regional regimes, we estimate models on firm labour productivity, controlling for firm, regional and industry variability.
Our analyses indicate that regional related variety has a positive impact on firm productivity in European regions, especially for firms in high-tech and medium-tech regions – regions focused on by the smart specialisation strategy. These outcomes have implications for European policies on competitiveness and smart specialisation – as firms embedded in regions without these technological and institutional circumstances are systematically worse off in terms of productivity, and catching-up is not obvious for such regional economies. This regional heterogeneity complements firm heterogeneity, and makes best practices for all regions difficult to identify.

Alberto Gude (p), Sandy Dall'Erba

Spatial effects on regional growth have been thoroughly studied, but so little is known about their strength against inter-industry forces in the production frontier approach. The limitations are twofold: sectoral data is frequently aggregated, and sectoral fragmentation is completely overlooked. Both Marshallian and Jacobian externalities have proven their importance on efficiency and productivity. Therefore, this paper extends the traditional aggregate production function including interregional and inter-industry dependencies.

In order to disentangle the true data generating process, the SLX model is introduced in the framework developed by Hsiang (2010) for environmental economics, using Conley (1999) method for correcting serial and spatial correlation. This is the simplest model producing flexible spillovers (Halleck Vega and Elhorst, 2015).

The geographic matrix has been widely used to capture spillovers, even though core and dynamic regions are not determined by the geometric centre. Moreover, spatial dynamics differ among industries, as well as the relationship structures behind them. Consequently, this paper proposes two possible economic matrices to overcome these limitations based on trade flows and migration flows. New economic geography highlights that trade on intermediate goods works as a key role, so intermediate goods weight the influence of each industry.

The Spanish heterogeneity and spatial structure are a challenging framework to test our model. The panel dataset comprises annual observations at the NUTS 3 level for 48 Spanish provinces over the period 2001 to 2013 across 5 industries (i.e. excluding Canary Islands and the non-market activities).

Michael Lahr (p), Hyunjoo Jang, Erik Dietzenbacher

Compared to most other OECD nations, productivity in South Korea has risen fairly slowly. Some researchers have suggested this is due to the slow pace of regulatory reform there compared to that going on in other nations. But it could also be due to lifestyle changes of Korean households, international competitive pressures (e.g., the rise of China's economy), shifts in the nature of the nation's capital investments, and changes in the composition of the nation's exports, among other possibilities. We examine such potential proximate causes across regions of South Korea using a structural decomposition approach. In this paper we apply, for the first time, a shift-share version of structural decomposition analysis developed elsewhere by Dietzenbacher and Lahr (2017). We use data from for regions of South Korea for 2003, 2005, 2010, and 2013 as published by the Bank of Korea. These data are quite detailed, containing 16 regions with 82 industries per region. We interpret the findings on the proximate causes of productivity change using seven (7) components by broad sector by region in light of known interregional relocations, international trade agreements, regulatory and other differences across the regions of South Korea.