News

Yavor Alexiev

A little over a month ago, IME traced the dynamic of the increase in income and unemployment levels in the different districts of Bulgaria.[i] The goal was to determine how the labor market in the various regions reacts to the combination of rising wages and continuously high unemployment. In that analysis, we divided the regions in four groups:

• Group 1: Districts with slowly increasing wages and falling unemployment in 2012: Burgas, Vidin, Gabrovo, Dobrich, Kyustendil, Pleven, Smolyan and Haskovo. In these districts unemployment in 2012 was falling, while at the same time the annual average level of wages increased at a slower rate than the national average.
• Group 2: Districts with slowly increasing wages and rising unemployment in 2012: Blagoevgrad, Varna, Veliko Tarnovo, Vratsa, Lovech, Pernik, Ruse, Sliven, and Yambol. The combination of low rate of increase in wages and the continuous rise of unemployment is evidence of the lack of flexibility on the labor market in these regions.
• Group 3: Districts with rapidly increasing wages and rising unemployment in 2012. In these regions the wages in 2012 have increased at a rate higher than the national average, which is accompanied by the rising unemployment. These are Montana, Pazardzhik, Plovdiv, Razgrad, Silistra, Sofia (capital), Stara Zagora and Targovishte.
• Group 4: Districts with rapidly increasing wages and falling unemployment in 2012 – there are only two districts, in which the rate of wage increases is higher than the national average, without negatively affecting the labor market. These are Shumen and Sofia (district).

Keeping these four groups, in the current analysis we will take a look at what happened with the annual average employment rate of the population above 15 years old.

In times of continuously high unemployment, monitoring the employment level is of crucial importance to the interpretation of the labor market processes, because it shows what part of the population in working age actually has a job. By itself, the fall in unemployment in some regions (if not accompanied by increase in employment) doesn’t show real recovery of the labor market.

Graph 1: Change in the employment rate in the districts with slowly increasing wages and falling unemployment in 2012 (Group 1), percentage points

Source: NSI, IME’s calculations

In seven of the eight districts, in which in 2012 wages increased at a rate that is lower than the national average we see a rise in employment. Burgas is the region with the largest number of created jobs – almost four thousand. In Dobrich and Smolyan, the number of employed has increased by 3.3 and 2.9 thousand respectively. Only in Kyustendil employment is shrinking and in 2012 the employed decreased by 1.7 thousand people.

This is also the group with the most favorable labor market development in 2012, manifesting itself in decrease in unemployment and increase in employment. The slower rise in income in these regions helps people keep their jobs and in the same time leads to creating new ones.

Graph 2: Change in the employment rate in the districts with slowly increasing wages and rising unemployment in 2012 (Group 2), percentage points

Source: NSI, IME’s calculations

In five out of the nine regions, in which in 2012 we see a simultaneous increase in unemployment and a slow rise in income, employment is also decreasing. This is evidence of a serious lack of flexibility in the labor markets in these regions.

Most worrisome is the situation in Sliven. The minimal increase in income in combination with falling employment and rising unemployment, can be seen as a sign that the labor market does not seem to be able to cope with the administrative increase of wages. This increase in wages affects most of all the districts with lower salaries, such as Sliven.

In addition to Sliven, the rate of increase of the average annual wages is lowest in Blagoevgrad, Veliko Tarnovo, Pernik, Ruse, and Yambol. These are also the districts that show the biggest improvement in terms of employment – in four of them employment increases.

Graph 3: Change in the employment rate in the districts with rapidly increasing wages and rising unemployment in 2012 (Group 3), percentage points

Source: NSI, IME’s calculations

In four of the eight regions, in which in 2012 income levels increased faster than the average national rate and unemployment also rose, we see a fall in the level of employment as well. Only Plovdiv shows a significant increase in the number of employed, but even this district is below the level of 1 percentage point. We also have to note here that in this group the districts that showed an increase in employment in 2012 (Plovdiv, Silistra, and Montana), are also the ones with the lowest deviation from the average national rate of wage increase. In the other five regions wages have a much faster rate of increase, which has a negative effect on employment.

Graph 4: Change in the employment rate in the districts with rapidly increasing wages and falling unemployment in 2012 (Group 4), percentage points

Source: NSI, IME’s calculations

In 2012 there are no significant changes in the rate of employment in the two districts with decreasing unemployment and rapidly increasing wages. Despite the increase in income and employment in Shumen, the district remains one of those with the highest unemployment.

Graph 5: Distribution of the four groups of districts according to the change in the rate of employment of the population above 15 years old in 2012, percentage points.

Source: NSI, IME’s calculations

As is evident, employment in 2012 increased by more than one percentage point only in districts with lower than the national average rate of monthly wage increase.

There is an increase in the employment level in 11 out of the 17 districts, in which wages in 2012 increased at a rate lower than the national average. The slow increase in income in some of the poorest districts in 2012 continues to aid the increase in employment in the same districts since the start of 2013. But in other districts, even the slow rise in wages cannot positively influence employment. In the same in five out of the ten districts, in which the increase in wages is higher than the national average, the average annual level of employment in 2012 waslower than the one in 2011.

Groups 1 and 2 (districts with slower rise of wages) include four of the six regions, in which in the first quarter of 2013 there was an increase in the number of employed in comparison to the end of the fourth quarter of 2012. These are Vidin, Gabrovo, Kyustendil, and Pernik. With the exceptions of Montana and Shumen, employment in the first quarter of 2013 fell in all of the districts of Bulgaria.

Yavor Alexiev

The recently held parliamentary elections in Bulgaria have once again been a reflection of the effect diverging socio-economic conditions exert upon electoral moods. The election results of the three leading parties – GERB (Citizens for European Development of Bulgaria), BSP (Bulgarian Socialist Party) and DPS (Movement for Rights and Freedoms), exhibit a more or less robust relationship with three main socio-economic indicators – ethnicity, age and employment rate.[1]

DPS takes advantage of the “ethnicity” factor

As expected, the distribution of votes according to administrative regions has been significantly affected by the ethnic origin of the local population. In regions where over 30% of those who declared to consider themselves as Turks during the last national census in 2011, DPS won between 28.3% (Silistra) and 59.8% (Kardzhali), while nationwide, the party managed to win as much as 11.3% of electoral votes nationwide. Another observation exemplifying the role of the ethnicity factor in shaping DPS’s electoral support is that in some regions in Western Bulgaria, which have common socio-economic characteristics with the above-mentioned regions, DPS managed to secure only modest electoral support, amounting to as little as 1.1% (Vidin) and 0.7% (Kiustendil).

Sources: CIC (Central Electoral Commission), NSI (National Statistical Institute), IME calculations

BSP takes advantage of the “AGE” factor

In order to illustrate the relationship between BSP’s election results and the age factor, we use data of the Bulgarian National Statistical Institute (NSI) concerning two indicators - the number of citizens aged above 65 years per region [2] and the ratio of citizens above 65 over citizens in the 15-64 age group (i.e. the coefficient of age dependency).

Out of the nine regions, where the latter ratio exceeds 33.3%, i.e. there is more than one person aged 15-64 years per three people over 65 years of age, the results show that BSP took the upper hand in eight of them. The only exception is Gabrovo, where GERB took 37.2% of the votes, which also happens to be the highest regional result of the former ruling party.

Sources: CIC (Central Electoral Commission), NSI (National Statistical Institute), IME calculations

GERB takes advantage of the “Employment” factor

In the majority of the administrative regions, out of the four parties that made it to Parliament, GERB can be deemed to be the only one the electoral support of which exhibits a positive correlation with district employment rates.[3] GERB is the second political power after BSP in the five regions where unemployment is highest in the country – Vidin, Lovech, Montana, Vratsa and Pleven. Accordingly, in regions where the unemployment rate is lower than or close to the average for the country, GERB wins the highest number of votes, with the notable exceptions of Stara Zagora and Pernik.

Sources: CIC (Central Election Commission), NSI (National Statistical Institute), IME calculations

[1]The featured graphs represent a conditional comparison of various socio-economic indicators and election results in different administrative regions. These are under no circumstances to be considered as proof of the electoral moods of citizens based on their ethnic origin, age or socio-economic status; rather, these depict common tendencies in the shaping of political preferences of groups of people living in a different social, economic, political and cultural context.

[2]The five regions where DPS’s electoral dominance can be primarily attributed to the ethnicity factor have been excluded from the comparison between the age dependency ratio and the results of BSP.

[3]The five regions where DPS’s electoral dominance can be primarily attributed to the ethnicity factor have been excluded from the comparison between the employment rate and GERB’s results.

Yavor Aleksiev

In 2012, the unemployment rate at national level has kept increasing, reaching 12.3%. Despite this, the average monthly salary of employees under labor contracts has increased by 8.6% on annual level. Even accounted for inflation (3% annual average in 2012), actual wage growth is still staggering at 5.4%.

Overall, in 2012 the unemployment rate increased in 18 districts and decreased in 10. But only in two of those the decrease of unemployment rate was accompanied by higher than the average national growth rate of monthly wages.

The main issue here is how the labour market in the various districts of the country has responded to the combination of rising wages and persistently high unemployment. If we take a closer look at the dynamics of unemployment and wages by districts in 2012, we can differentiate four main types of districts:[1]

Slowly increasing wages and falling unemployment rate

Source: NSI, IME calculations

In eight districts: Burgas, Vidin, Gabrovo, Dobrich, Kyustendil, Pleven, Smolyan, and Haskovo, unemployment has fallen, and at the same time the average wage on annual level has increased, although at a lower rate than the national average. The slow rise of wages in some of the country’s poorest districts (Vidin, Smolyan, Haskovo, and Kyustendil) has resulted in a decrease of the unemployment rate.

Slowly increasing wages and rising unemployment rate

Source: NSI, IME calculations

Unemployment increased in nine districts, although wages in them grew at a lower rate than the national average. These were: Blagoevgrad, Varna, Veliko Tarnovo, Vratsa, Lovech, Pernik, Ruse, Sliven, and Yambol. In most of these districts, nominal average monthly wage increase was minimal, being less than 3%in Ruse, Sliven, and Yambol. Especially striking was the case of Varna, where unemployment rose by 6%. The combination of slow wage growth and a continued rise of unemployment rate testified to the lack of flexibility in the labour markets of those districts.

Rapidly increasing wages and rising unemployment rate

Source: NSI, IME calculations

In eight districts, wages increased by a higher-than-the-national-average rate, and in the same time unemployment rose as well. Among those districts appear the ones with the lowest annual average unemployment rate – Sofia (cap.) and Stara Zagora. The significant rise of average wages in Razgrad in combination with increased unemployment possibly indicates the layoff of low-skilled employees.

Rapidly increasing wages and falling unemployment rate

Source: NSI, IME calculations

The only two districts where the wage increase has outpaced the national average, without a negative effect on the labour markets, are Shumen and Sofia (district). The latter continues to close the gap with Sofia (capital), while the only districts with higher wage levels are Stara Zagora and Vratsa. Sofia (district) is the district with the third lowest unemployment rate after the capital and Stara Zagora. Meanwhile, Shumen remains one of the districts with the highest unemployment rate in the country in 2012.

What’s next?

The effects of the recent increase of the minimum wage and social insurance thresholds is yet to be evaluated. The fact that in most districts the wage growth rate, which is higher than the national average, goes hand in hand with an increase in the unemployment rate, indicates that the labour market needs more time before we can speak of a true recovery.

The simultaneous increase in the unemployment rate and wages, observed in many districts, demonstrates the lack of flexibility in local labour markets. Their low adaptability is caused by administrative interference, namely the constant increase of minimum social insurance thresholds and minimum wages, which in turn hinders the downward wage adjustment and fuels the layoff of workers.

The project “Regional Profiles: Indicators of Development” is carried out with the financial support of the America for Bulgaria Foundation.

Yavor Alexiev

The results of the Labour Force Survey for 2012 and the revised labour market statistics for 2011, published about a month ago, reveal some interesting trends on district level.

To start with, the employment rate of the population aged 15+ has been revised downwards in only three districts-Pazardzhik, Blagoevgrad and Smolyan (Table 1). In Varna and Sofia (city) the rate of employment has remained unchanged, while in all other districts the rate of employment has increased. The average increase of employment for the country as a whole is by 1 percentage point – from 45,6 to 46,6%.

Table 1: Effect of revision on the employment rate for the population aged 15+ by districts, difference in percentage points as compared to preliminary data for 2011

Source: NSI

What stands behind the increase of employment?

The increase of the employment rate is a result of the modified procedure [1] for estimating the weights, applied since the beginning of 2012. It is also a consequence of the correction in the estimates of the population that followed the 2011 Census. The Census information has helped establish the actual demographic status [2] of the districts, which has led to a more accurate assessment of the economic activities of population. Graph 1 depicts a part of the effect arising from the census of the average annual population by districts on revising the employment data.[3]

Graph 1: Revision of employment data in 2011, and difference between regular estimates of the annual average population and Census 2011 results.

Source: NSI, IME calculations

The Graph shows the underestimated employment rate for the period 2002-2011 in most districts due to higher estimates of the population by the National Statistics Institute. The bad news is that this is the statistical effect of a population and hence – labour force shrinking much faster than the rate of job creation. The population has diminished much quicker than projected in 26 out of the 28 districts. The only exceptions are Varna and Sofia where it increases faster than expected.

What happened in 2012?

According to the last published data of NSI for 2012, employment increased in 16 districts and declined in 11. The only region where change is not observed is Stara Zagora.

Table 2: Year-on-year changes in the employment rate in 2012, difference in percentage points as compared to revised data for 2011.

Source: NSI

Despite this positive tendency in employment rates, the number of jobs increased in only 9 districts. In 2012 the number of those employed in the district of Burgas increased by 3.9 thousand, in Plovdiv – by 3.5 thousand, in Dobrich – by 3.3 thousand and in Kyrdzhali and Smolyan - by around 3 thousand. The trend in Varna is alarming because the employment rate there decreased by 2,3 percentage points, while the number of employed people – by 10 thousand. In other words, about one third of the jobs lost in 2012 are in the in district of Varna, and roughly the same number of job losses occurred in the districts of Sofia and Pazardzhik.

Graph 2: Employment rate dynamics and number of employed 2012-2011

Source: NSI

Data on the developments of the labour market on regional level for the last two years do not provide reasons for joy in 2013. At this moment the increase in employment is not the result of jobs’ recovery, but of the quick decline of the population, including the economically active people. The 2011 Census data shows a higher employment than earlier estimated precisely because the number of people of working age decreases faster than expected. The actual job recovery is yet to come.

The project “Regional Profiles: Indicators of Development” is carried out with the financial support of the America for Bulgaria Foundation.

[1] See the methodology of the study “Labour Force Survey”, pages 5-6 (http://www.nsi.bg/img/MRUBRIK/LFS_Methodology.pdf)

[2] The projection for the demographic development of the population in each district is important in determining the employment rate, because its value affects the weights in the sample, on the basis of which the economic activity of the population is estimated.

[3] The red line in Graph 1 depicts the difference between preliminary data on employment rates for the population aged 15+ for 2011 and the revised information based on the Census. The blue line represents the registered deviation from the increase or decrease of the population aged 15+ in each district, arising from the annual demographic estimates of NSI. The data is derived when from the dynamics of the average annual population for the period 2002-2010 we subtract the difference between the annual average value for the period 2002-2010 and the data for 2011 alone. The latter have a corrective function because they are based on the Census 2011. For instance, the average annual population of Plovdiv for the 2002-2010 period dropped by 1845 people on average per year. In 2011 alone, as a result of the census, an annual decline by 16 865 people is observed. The difference between the 2011 values and the average value for 2002-2011 period is 15 020 people. After that, the number of people aged 15+ (within those 15,020) is estimated based on the share of the population aged 15+ in the starting year 2001. Their percentage in the entire population aged 15+ for 2001 is actually the change of the average annual population aged 15+ for the entire period 2002-2011 that has missed by the regular annual demographic estimates of NSI.

Of course, the values thus derived are conditional due to the constant changes of the age characteristics of district population. But in view of the tendencies for increase of the average age in all districts, the share of people is unlikely to be lower than the values obtained using this method.

Desislava Nikolova

Since 2012, IME has been conducting a comprehensive study on the socio-economic development of the districts in Bulgaria. For the objectives of this research, we collect and analyze data on regional level for more than 50 various indicators. One of them is the average annual wage of the employed people (under labour and professional contracts).

What can be easily seen from the data on the 28 districts are the enormous and constantly growing differences of the work pay between districts. While in 2000 the gap between the best-paid districts and the ones with lowest wage levels was around 1.5 times, in 2011 the difference widened to more than 2 times. It is interesting to note that the city of Sofia steadily and unconditionally preserves its leadership throughout the entire 2000-2011 period. The bottom of the ranking in the beginning of the period is occupied by Smolyan (until 2005), followed only shortly by Haskovo (2006) and Blagoevgrad (2007). Since 2008 the district with the lowest pay has been Vidin, which is hardly a surprise having in mind the overall economic lagging behind of the north-western region and especially of the Vidin district.

Table 1: Difference in wages of employed personnel in the best-paid and worst paid district (in times)

Source: NSI, calculations by IME

The data show not only the growing difference between the highest and lowest paid districts, but also the constantly widening gap between district-level annual wages and their country average level. The process is clearly illustrated by the dynamics of two related statistical indicators - the variance and the standard deviation. The standard deviation shows the extent of deviation of regional values from the country average level. Its dynamics for the 2000-2011 period clearly shows that the difference with the country average is increasing. While the standard deviation was less than 500 BGN in the beginning of the period, in 2011 it already surpassed 2000 BGN, hence it has increased more than 4 times. Although we operate with nominal values, inflation is not to blame -inflation for the 2000-2011 period (December to December) is 77.4%, while the nominal increase of the standard deviation is more than 400%.

Table 2: Standard deviation of wages by region

Source: IME’s calculations

In general, regional wage data confirm one of the main conclusions in our study, namely that differences in the socio-economic development among Bulgaria’s districts are intensifying. This process is observed in the period of economic growth up until 2008, as well as in the crisis and stagnation period that followed. The discrepancies are especially evident between the centre, i.e. the city of Sofia, and the rest of the country - meaning that the country can be easily divided into Sofia on the one hand and the remaining territory, on the other. �This tendency becomes more acute over the years.

It is about time to start considering the increasing gap between districts in the planning and implementation of government policies, especially regarding income policies and measures against the informal economy. What we witness every single year is the constant increase of minimum social security thresholds by sectors and occupations for everyone without taking into account the region’s characteristics. While this increase might have a generally low effect on the labour force in Sofia, where wages are already above the new threshold, in less developed regions like Vidin, for instance, it could be harmful for a part of the employees, as the new requirement increases labour costs and might lead to job cuts.

Unfortunately, up to this moment the government has not published an analysis on the effects of the thresholds’ annual increases on employment or on the informal economy, although such a report is quoted in the latest version of the National Reform Program, uploaded recently on the website of the Ministry of Finance. The quotation of the above-mentioned analysis that can be found in the program, says: “During the crisis, the increase of minimum incomes for low-skilled labour has weak negative effects upon the dynamics of the socially insured people by economic activities. A more serious negative impact is observed in a regional aspect between the change of the ratio between minimum incomes and average wages and the dynamics of those employed. In the districts, where the increase of the ratio between minimum thresholds and the average wage is greater than the country average, the decline in the number of employed people tends to be greater than the economy’s average, with few exceptions”. The quotation testifies that even the Ministry of Finance identifies a greater negative effect from the increase of thresholds on employment in some regions. The situation with the minimum average wage is analogical - its constant increase (by 29% within a little more than a year, between September 2011 and January 2013) most likely gives the boot to some employees or sends a number of low-skilled ones into the grey economy. Most likely the negative effects are greater in the regions with a smaller difference between minimum and average wages.

The problem is whether these numbers, the economic theory and the empirical analysis of the Ministry of Finance will be finally considered at the annual discussions on social insurance thresholds or while politicians make their usual promises for further increases of the minimum wage? A differentiation of thresholds and minimum wages by regions or by districts and their tying-up to local labour conditions (local levels of employment, salaries, unemployment and other relevant indicators) might lessen the negative effects, if the elimination of thresholds and minimum wages are politically unacceptable. It might be good to think about such a change in policies, if we do not want to continue losing jobs.

The project “Regional Profiles: Indicators of Development” is carried out with the financial support of the America for Bulgaria Foundation.

IME Hosted a Roundtable Titled "Regional Development of Bulgarian Districts - Approaches for Analysis and Key Issues" (29 January 2013)

The IME hosted a round-table discussion titled "Regional Development in Bulgaria - Analytical Approaches and Key Issues." The event took place on January 29, 2013 at the ”Sredets” Hall of Grand Hotel Sofia.

The discussion was attended by government officials, as well as representatives of NGOs, business associations and other stakeholders whose work is relevant to the issue of regional development.

The IME team presented the results of the study "Regional Profiles: Indicators of Development" to the expert audience and received useful recommendations regarding the further development and direction of the project, given the upcoming releases in 2013 and 2014.

The project “Regional Profiles: Indicators of Development” is carried out with the financial support of the America for Bulgaria Foundation.

Alexander Tzvetkov PhD, Regiostat

The goal of the multidimensional analysis of regional development, using Kohonen’s Self-Organizing Maps (SOM) is to track and analyze the region’s common socio-economic development during the 2008-2010 period, using a system of indicators grouped in seven categories: economy, infrastructure, demography, education, healthcare, environment and social environment.

Two “fake regions” have been established for the purposes of this analysis. One is a “perfect” region, which scores best in all indicators in the same time. The other one is a “worst” region, which scores worst. These “fake regions” are used as reference points and benchmarks for assessing the development of the 28 regions during this period. The current state and development of all 28 regions is assessed as positive or negative in comparison with these benchmarks. Afterwards the regions have been grouped in clusters, formed by Kohonen’s self-organizing maps. This allows us to compare all regions simultaneously showing the degree of difference, the development direction – whether there is a growth or a decline in development, the dynamics of the said development and the presence/lack of cohesion between the regions.

The main conclusion from the analysis is that there is no cohesion developing between the regions. On the contrary – there is a region – Sofia (capital) – which stands out among all the rest in terms of socio-economic development and this difference is increasing during the research period.

The common socio-economic state of Sofia (capital) is the best, as can be seen by its position closest to the “perfect” region on the map. Varna, Burgas and Plovdiv regions are also close to it, but despite their relatively good development, they fall far behind Sofia (capital). This is depicted by the color on the cluster border between Sofia (capital) and the three other regions – a highly-saturated color, showing a big difference.

As per the development of the economy, the highly-saturated color in the lower left corner, where Sofia (capital) is located shows that it is the most developed. Varna follows, but even though it is the second in the country, it falls far behind Sofia (capital). The worst-developed regions are Silistra, Razgrad, Vidin and Shumen, since the color of the map in their location is not saturated.

Gabrovo has the most intensive dynamics when it comes to the development of infrastructure – 2010 is highly saturated in comparison with 2008.

Vidin, Lovech and Montana have the worst demographic state during the entire period – and the least saturated portions of the map. Gabrovo has very pronounced and worsening negative trends during the period.

The regions with the worst education systems are Sliven, Targovishte, Razgrad and Silistra (least saturated on the map).

The rather pale color of the maps demonstrates that Razgrad and Targovishte have the worst healthcare during the entire period.

The region with the most environmental issues during the entire period (as can be seen by the lowly saturated color on the map) is Stara Zagora. The rather saturated color on the map, showing good environmental conditions, denotes Gabrovo as the leading region in the country in this respect during the three analyzed years.

Unlike the other categories (economy, for instance), the social environment indicators have a more pronounced dynamics. The color saturation on the map shows that in 2008 Haskovo, Pernik and Gabrovo had the best social environment in the country. However, they demonstrate negative development trends, because in the following years their color of the map is less saturated.

The project “Regional Profiles: Indicators of Development” is carried out with the financial support of the America for Bulgaria Foundation.