Page 105 - Azerbaijan State University of Economics

76. Ibrahim Niftiye: Descriptive Analysis of Employment in Azerbaijan: Possibilities of the Dutch Disease

77.
78.
The main elements of the applied descriptive statistical analysis are the minimum value,
maximum value, range, mean, median, mode, standard deviation, variance, coefficient
of variation, skewness, and kurtosis values. The descriptive analysis is performed on
three levels: natural value, which is a thousand persons per sector; year on year growth;
and cumulative growth, both being in percentage points. As a concluding part of the
results section scatterplot matrix and Pearson’s R correlation analysis have been added
to check the correlation between the pre-selected economic sectors and real effective
exchange rate (REER is an essential economic indicator among the Dutch disease
studies, so the analysis provides initial conceptualization of the hypothesis).

3. RESULTS
Table 1 summarizes the descriptive analysis of employment in Azerbaijan between
2000–2018 in the mining, manufacturing, agriculture, and services sectors in terms of
employed persons (left-hand side) and year on year percentage changes (right-hand
side). The expected outcome of recent trends in the economy of Azerbaijan that align
with the Dutch disease model is that manufacturing and agriculture would shrink, while
mining and services would expand alongside REER appreciation, output, and trade
booming in the mineral sectors. Accordingly, the mining sector had the lowest range
(7.30 thousand persons, standard deviation – 2.72 thousand persons, coefficient of
variation – 0.07). Thus, the mining employment data did not denote any extreme
positive skewness (0.33) and the kurtosis value (-1.69), which indicates that mining did
not experience any significant employment decrease or increase. This points to the fact
that resource movement in its “direct de-industrialization” form is not observable in the
Azerbaijan economy.

Meanwhile, Table 1 shows the highest range in services by 392.80 thousand persons
alongside its positive linear increase over time (see. Fig. 5), which indicates the
“indirect de-industrialization” form of resource movement. With 14 years of
uprising against five years’ decrease, services have a coefficient of variation by 0.11,
reaching a maximum amount of labor of 1288.60 thousand persons from 895.80
thousand. The negative skewness (-0.44) and negative kurtosis (-1.04) of services
reveal that the stable and upward increase in the employment of services was in fact
accompanied by the booming period, suggesting an indirect de-industrialization of
the resource movement effect.

The manufacturing and agriculture sectors with the same increased (9) and decreased
years (10) possessed the highest coefficient of variations of 0.11 and 0.24, respectively,
pointing to higher spread data distribution, as well as volatile developments. The range
was higher in manufacturing (58.40) than in agriculture (53.60), indicating the
decreased share of agriculture in employment. Comparing the timely progress in Figure
105