Frequently we hear stories of economic growth propelled by increasing government expenditures. An expansionary fiscal policy achieved with extensive government spending is usually enough to stimulate consumption in times of recessions. On the other hand, in such a situation, the resources are not allocated efficiently, and the growth may not be sustainable in the long run.
When we take a look at Azerbaijan’s budget expenditures, except for a brief halt during 2009 and 2017, the expenditures keep escalating. It is natural that the budget is getting larger when considering that the economy is also expanding. However, during 2001 – 2017 period, Azerbaijan’s economic growth averaged 9.5% while the same indicator for budget expenditures is 22.8%. In most of the years (except for 2009 and 2014 – 2017), the growth in state budget outpaces the growth in economy. Thus, it makes sense to ask: Is Azerbaijan’s economic growth simply increasing government expenditures in a different form?
While gathering data on state budget expenses, I have decided to analyze the categories classified by State Statistic Committee separately: Economy, Education, Healthcare, Social Security and Arts (there are other categories as well, such as science or judiciary but I decided to focus on these five above). Among these, the category that particularly interests me is Economy, as it involves expenditures towards construction. By the end of this paper, I hope to explain why construction is especially interesting.
For now, I have constructed a graph, showing “Economy” expenditures and GDP growth of Azerbaijan:
Graph 1: GDP Growth and expenditures towards Economy
Source: State Statistical Committee of Azerbaijan (SSCA)
When we look at the data, clearly there is some form of relationship between the two variables. However, this is not enough to justify that Azerbaijani economy is boosted by expenditures, because the relationship might be the other way around: when the economy is booming, government spends more money on construction. In order to further analyze the issue, I am going to use a simple measure: Correlation Coefficient (CC). Essentially it is a measure that shows how two variables are related. In other words, are there any correlation between them and if so, how strong is it? The coefficient varies between -1 and 1 and is calculated using the following formula:
In this formula, n denominates sample size, while x and y represent our variables. In my calculations, x is going to be our GDP growth each time, and y will represent expenditures towards Economy, Education, Healthcare, Social Security and Arts. Results of the calculations are given on the table below:
Table 1: Correlation Coefficients of different categories
Source: SSCA, Author’s calculations
One thing that we need to keep in mind for not misinterpreting the coefficients is trend. The economy of Azerbaijan has been growing, along with its population, leading to a natural growth in most of the expenditure categories, including the ones given in the table above. The positive trends of both economic growth and specific expenditure categories mean that our correlation coefficient may be quite large, but that does not necessarily imply causation between the two.
To get a more comprehensive understanding of the topic, I will use another statistical indicator: Coefficient of Variation (CV). It is calculated by dividing the standard deviation by mean. A large CV indicates that the category has varied considerable during the period, which may give us a better insight. In the table below, you can see the calculation results:
Table 2: Coefficients of Variation of different categories
|Expenditure Category||Economy||Education||Healthcare||Social Security||Arts|
|Coefficient of Variation||27.091||0.935||0.966||1.359||3.151|
Source: SSCA, Author’s calculations
First off, the relationship between expenditures towards Arts and GDP growth minuscule as expected (CC = 0.064). This category also has a very marginal share in the overall budget (only 1.44% of all expenditures goes towards Arts), thus even if it did ascent during economic growth, it would still bear little importance to the Azerbaijani economy.
The other three categories, Education, Healthcare and Social Security all have some sort of social importance, meaning that government cannot cut them substantially even if the need arise. Because these are expenditures directed towards population, they all should have upward, stable trend in the last two decades. This is indeed the case with Education and Healthcare expenditures. Fairly high correlation coefficients along with a small coefficient of variation means that these expenditures have been steady and had an overall increasing trend.
However, the Social Security is not like the others, it has a small Correlation Coefficient of just 0.32. Once I took a look at the graph of these three categories, it was apparent that unlike Healthcare and Education, Social Security had something strange going on: it is more volatile and even experienced noticeable declines from time to time. Yet, the reason behind is most probably the efforts to ease the burden of Social Security expenditures on the state budget and make the State Social Protection Fund (SSPF) more financially independent. The diminishing government spending towards Social Security is presumably compensated by increasing revenues from rest of the sources of the Fund.
Graph 2: Timeline of “social” expenditures
Source: State Statistical Committee of Azerbaijan (SSCA)
As expected, Economy have larger CV value because unlike the others, it does not bear a social importance. Thus, when there is a need to balance the budget (to decrease expenditures), it is the first one to be experience a cut. It has a correlation coefficient of 0.80 and coefficient of variation equal to 27.1. Clearly, these expenditures vary a lot and have some sort of relationship with economic growth of Azerbaijan. I believe that this category is the one that rises the most when the economy needs a push and also is the one to shrink the most when the budget expenditures needs to decline. Here are a number of reasons why:
- When there is a need to cutback expenditures, the category of “Economy” is the easiest one because it bears almost no social importance (unlike Education, Healthcare and Social Security) and the long term consequences would not be dreadful. I mean sure, we might miss a couple of parks and other glamorously large governmental buildings, but the majority of the population would not even feel a slight change in their lives. Compare this now to a hypothetical situation where Social Security expenditures are cut considerably, leading to a very challenging situation for the society’s most marginalized groups: elderly, disabled and others who are dependent on a fixed amount from the government for their survival.
- When the economy is stable and needs some aid, government must decide where to put the funds. There are multiple options, but category of “economy” is perhaps the most attractive one. Firstly, the category mainly includes construction, which is a very good way of inflating the Gross Domestic Product. If the government has 100$ and a choice between investing it in education or construction, it always has a natural tendency to go for the latter. The reason is that, firstly, the 100$ invested in construction will immediately become a subject of multiplier effect and will contribute more to the GDP at the end of the year while the allocated resources towards education will take time to realize. Secondly, there is no doubt that a 1$ invested in education gives more than 1$ invested in construction, but the “payback period” in case of education is much higher than that of construction. A building can be built from scratch in 2 years and be ready to operate. A pupil needs at least 15 years of primary, middle, high school and Bachelor education (on average) before he/she can bring any significant effect back. Such a long payback time demotivates the government for a number of reasons. And lastly, the funds allocated towards construction have the most likelihood to be engaged in corruption, because public oversight mechanisms over construction costs are extremely weak.
The Coefficients of Correlation values told us that Economy, Healthcare and Education expenditures are correlated with GDP growth of Azerbaijan. When it comes to Healthcare and Education, their Coefficients of Variation were quite meager, indicating that they do not fluctuate much over time. Looking at their timeline, one can easily see that this is the case, and the reason for high CC values is simply the overall increasing trend.
The category of Economy, on the other hand had a sizeable CV value, meaning that it varies remarkably. Whenever the economy needs a boost, or there are abundant windfall oil revenues to spend, this category comes to the stage, because: (a) it bears no social importance, (b) has a shorter payback period than other investments and (c) has frail public oversight.
Thus, keeping all of the above mentioned factors in mind, it can be said that Azerbaijan’s economic growth gets a significant momentum from expenditures towards Economy (in other words, construction), which is undesirable. In a different post, I will explain why this is the case.
 These are expenditure categories, such as expenditures towards Economy, towards Healthcare and so on.
 The formula that I have used is called Pearson’s correlation coefficient, which is the most widely used method to calculate correlation coefficient.
Multiplier effect takes place when an extra $100 is spent in economy, which becomes $100 extra revenue of somebody else who will now spend more as well and so on. At the end, the $1 will turn into a much larger amount in Gross Domestic Product.