The results of the authors’ own calculations show that the function (1b) best captures the causal dependencies.2 The purely academic problem would be to

test additional quarterly pools of data, but the limited statistical base does not allow one to make calculations on a monthly basis. However, even such a simple study enables rejecting simultaneous analyses and those with the annual shift, for the sake of discerning the effects of employment downturn mainly in a sixmonth term. A typical situation is the presence of non-employment economic growth in the developed countries. The explanation lies in the formula-expressed relation between the product, employment and its attribute – productivity (the model is univariate because it uses only one Smithian factor of production – work):

Pt = Zt * Wt (2)

where

Z is employment W is labor productivity P is gross domestic product

which, when converted, becomes:

r = rz + rw + rz * rw (3)

where

r is GPD dynamics rz is employment dynamics rw is labor productivity dynamics

and after some simplification:

r ≈ rz + rw (3a)

This means that the condition of employment growth in the economy is r > rw. Labor productivity dynamics may be regarded as an autonomous independent variable, although in reality it is affected by several factors (mainly innovation and technology, organizational and market development, incentive system, working hours and working conditions, training and experience of employees). Work efficiency dynamics is much less subject to fluctuations than production, sales, demand, employment and unemployment. In the richest countries, productivity grows the most slowly. In the new EU member states (N12) reducing the gap it is significantly faster, while it is the fastest in Asian countries which experience the Industrial Revolution. Stability in the work efficiency dynamics is not absolute in nature, and thus the change is slow. Nevertheless, one can assume that rw, and thus the b parameter in function (1) in the EU27, oscillate around 2 percent in the long term. This was the case in 1974-89 (Commission of

the European Communities 1993, p. 12) and nowadays this is confirmed by the recent observations of the time series. GDP growth of 2.1 percent in 2010 corresponded to the stability of unemployment rate of 9.7 percent in 2010-11 among the economically active population, while the next year’s growth of 1.6 percent led to an increase in unemployment in 2011 and 2012. The conclusion is explicit – the economic growth of between 0 and 2 percent in the EU is usually accompanied by a decline in employment and rise in unemployment. In order to overcome the serious problem of unemployment in the EU, the region should have a few years of stable growth rate of 3 percent. At a rate of 2 percent one cannot expect the situation to improve. The presented data are averaged, and the sensitivity of particular national labor markets to changes in the economic situation may be different. It may best be exemplified by Poland, the only country in the EU to avoid the 2009 recession. In 2007-12 Poland had the fastest economic growth and at the same time recorded a decline in the labor market – in contrast to Germany, showing progress of employment (Austria and Malta also made slight progress; Figure 14.2). The decline in the labor market in Bulgaria, Cyprus, Lithuania and Spain is much higher than one might have expected from the general trend of development. In contrast, the labor market in Germany, Austria, Belgium, Finland and Italy is superior in the context of GDP changes. The same data presented in the form of scattered points enable presentation of the nature of the regression function, which illustrates the level of inverse proportionality of the two phenomena (Figure 14.3). A more detailed analysis of annual changes in GDP and unemployment, taking into account the effect of half-yearly displacement (i.e. changes in unemployment rates in the years 2009, 2010, 2011, 2012 were associated with the interpolated dynamics of GDP, respectively, in the second half of 2008 and the

first half of 2009, the second part of 2009/first part of 2010, etc.) enabled making 108 observations (27 countries times four periods). One may distinguish the following types of macroeconomic responses:

A GDP increases (does not decrease), the unemployment rate decreases. B GDP increases (does not decrease), the unemployment rate is increasing

(not decreasing). C GDP declines, the unemployment rate is increasing (not decreasing). D GDP declines, the unemployment rate decreases.