Baltic Sea Region Territorial Monitoring System

Territorial Cohesion - Synthetic analysis

The Gini Concentration Ratio (GCR) as well as the Atkinson index measure inequality among regions in the BSR. In terms of economic volume, here measured through GDP, the BSR has witnessed a trend of increasing concentration and inequality in relative growth patterns up till the year 2010, where after a weak opposing trend is discernible. A rather similar trend can be observed also regarding the distribution of the BSR population, where the largest regions have steadily increased their population share. Moreover, from 2010 and onwards this trend has been even more pronounced as rural out- and emigration particularly from the eastern BSR continues unabated. In terms of distribution of jobs, the concentration patterns are not unequivocal. The GCR indicates an increasing concentration up till 2009, after which a steep opposite trend follows, only to be reversed again in 2010. In terms of job concentration, the Atkinson index however tells a slightly differing story. The sensitivity parameter (ε value) of the Atkinson index enables giving greater emphasis to low or high performers. That parameter is here set at 0.8, which gives greater weight to changes in small regions. Following that, one may observe a trend of reduced disparities up till 2008. After that, inequality in terms of jobs tends to increase particularly at the expense of the smallest regional labour markets of the BSR.

A very similar pattern is discernible also when examining the Kuznets or 80/20 ratio, which measures the difference between the top and bottom quintiles of regions in terms of volume of GDP, number of jobs, and population. The trend in favour of the largest regions has been most stable regarding population. In terms of economic volume (GDP), the development has been somewhat balanced till 2007, whereafter a rapid shift in favour of largest regional economies took place. After 2010 this trend has been slightly reversed. In terms of employment, the largest BSR regions have increased their chunk of jobs up till 2008, after which slower polarisation can be observed.

The Sigma convergence indicator measures the development of disparities in general between all BSR regions by means of the coefficient of variation, which is simply the standard deviation divided by the mean. Also this indicator tells the tale of increasing regional disparities in the area. Both economic volume (GDP) as well as population are by and large increasingly concentrated in the BSR. In terms of jobs, the development has been more even.

However, the substantial scale differences in the variation between the three variables in the graph above are partially hiding the actual development in terms of employment. In the figure below, the development of total BSR employment is depicted by the blue line and the regional variation in employment by the red line. This comparison tells us that when the number of jobs in the BSR increased, that increase was beneficial for most regions in the area (red line downward sloping). However, following the financial crisis of 2008 and the subsequent reduction in number of jobs, that reduction hit smaller regional economies much harder than the larger ones, indicated by the upward sloping red line. More alarming still, after 2009, when employment in the BSR again is on the rise, the larger regions appear to be the ones benefiting from this development most, whereas smaller regional economies tend to lag increasingly behind, further aggravating the disparities in new job creation in the BSR.

In contrast to Sigma convergence, Beta convergence addresses disparity reduction via a catch-up process by the poorest performers. Concerning the level of economic wealth, regions with a low GDP/capita have been in a continuous catch-up process throughout the period. This is exemplified in the graph below, where, on average, the poorer the region is (in terms of GDP/capita), the faster has been its total GDP growth. Also regarding population, the larger the region in terms of population, the faster has been its growth up till 2009. In 2009-2011 this trend was somewhat slowed down, but after 2012 again extremely unfavourable to the smallest regions. However, employment patterns do to a large extent not reveal any statistically significant correlation between labour market size and its growth. In the years 2006, 2008 and 2010 such a correlation exists, but in those years the development has been in favour of the largest regional labour markets hence indicating decreasing convergence.

(C) ESPON BSR-TeMo, RRG, 2013