Impact of Employed Labor Force, Investment, and Remittances on Economic Growth in EU Countries

This paper analyzes the evolution and influence of gross domestic product per capita, labor force participation rate, gross fixed capital formation, and personal remittances on economic growth for European Union (EU) countries, using data from the World Bank (1996–2019) and from Eurostat (the first two quarters 2019–2020). The study has three components: statistical analysis, analysis of the evolution for each country and EU, and estimation of the impact on economic growth rate by using a linear multifactorial regression model for 1996–2019, 1996–2008, and 2009–2019. The model was validated by econometric techniques. The long-term causal relationship between exogenous and endogenous variables was validated by the Granger test. The results of the study show a differentiated evolution of the indicators, and that all indicators are severely affected by the 2008 financial crisis and the debut of the COVID-19 crisis in early 2020. The model used shows the significant positive influence of labor and investment, and the minor effect of remittances on economic growth. In the context of the COVID-19 epidemic, the results of the study could be arguments to be considered for the redesign of economic policies at European Union level.


Introduction
With the change of political regime in the early 1990s, substantial transformations have taken place around the world. Thus, many countries in the world with developing economies have turned into open markets with free flows of capital and goods, including labor. Globalization has had a strong effect on Eastern European countries, recording a massive labor movement and increased investment and living standards in developing countries.
The integration of Eastern European states into the European Union (EU) was followed by an increase in the process of labor migration from developing countries to developed countries that has brought about profound changes throughout the EU [1].
Due to demographic and economic imbalances, people will continue to migrate to developed countries as long as there is a demand for labor. A consequence of the free movement of labor is the emergence of significant flows of remittances, which can have a macroeconomic impact on both countries, those that receive the labor and the countries of origin. Remittance flows across Europe are vital for developing countries if they are used primarily for investment, so that the impact of remittances leads to the economic growth of these countries [2].
(G_GDP), because it is a more consistent measure than the nominal GDP growth rate, being adjusted with inflation, and it indicates how much GDP grew in one year compared to the previous year.
The labor force represents the number of employed people, but also the unemployed who are looking for a job; the participation rate in the labor force refers to the number of people available for work, as a percentage of the total population, which is a measure of the active labor force of an economy. We considered it appropriate to analyze this indicator because it can be influenced by shortand long-term economic trends.
Gross fixed capital formation, also called "investment," is a measure of the net gross investment (purchases, less sales) in fixed capital assets by enterprises, government, and households in the domestic economy during a financial year. It provides information about the future evolution and the growth model, being used to analyze the trends of investment activity over time.
Personal remittances include personal transfers (all current transfers in cash or in kind made or received by resident households to or from nonresident households) and compensation of employees (income of frontier, seasonal, and other short-term workers who are employed in an economy in which there are nonresidents and residents employed by nonresident entities). Personal remittances received (% of gross domestic product) is the flow of personal remittances expressed as a percentage of gross domestic product.
Numerous studies have addressed the indicators proposed for analysis, either individually or in aggregate, obtaining different results, both depending on the indicators tested and the countries and periods studied.
The accession of Eastern European states to the EU has led to significant flows of remittances that had a strong macroeconomic impact on both migrant and home countries [13].
The impact of remittance flow on economic growth has been the subject of numerous studies around the world, and based on the results obtained for different groups of states and geographical positions [14], one can see that remittances can have a positive impact, a negative impact, no impact, or an insignificant one in the development of states.
Examining the effects of remittances on economic growth showed that remittance flows were beneficial for economic growth in different groups of countries [15][16][17][18] or for individual countries [19]. In addition, the empirical analysis of the relationship between economic growth and remittances showed that remittance flows of workers could have a negative effect on economic growth, depending on the degree of development and geographical location of developing countries [20][21][22][23][24], an insignificant effect [25], or none [26,27]. By conducting a study for 68 developing countries, research reported in [28] considers that the impact of remittances on economic growth is difficult to quantify.
Based on the analysis of the relationship between remittances and GDP growth per capita, a negative effect of remittances was identified in a first phase of state development, after which the effect becomes positive [29].
Regarding the relationship between economic growth, employment, and remittances, there is no consensus on the contribution of remittances to economic growth and new job creation. The study of the impact of remittance flow on the employment rate shows that they can lead to employment growth and thus economic growth [30,31]. Another study [32] find that GDP growth, using the example of Slovakia, does not necessarily imply an increase in the employment rate. Research reported in [33] investigates the impact of remittances on the workforce for 122 developing countries and finds that remittances lower the female labor force participation rate, but does not affect the male labor force participation rate. Results by [34] show that remittances and economic growth have a significant positive effect on employment, while investments have a significant negative effect on it.
Several studies show that remittance flows to developing countries were more stable than other financial flows, even when the world economy was affected by the global financial crisis of 2009 [35,36]. The analysis of the effect of remittances on financial development and implicitly on economic growth for developing countries with a significant flow of remittances shows that there are strong, significant, and positive links between remittances, investments, and economic growth [2,[37][38][39][40].
Remittances can be mobilized in productive investments that lead to economic growth [41][42][43][44]. According to [45], financial flows can directly support economic growth, and their instability can aggravate the instability of the growth rate.
Gross capital formation (GCF) can improve the employment rate and economic growth [46]. Analyzing the relationship between these indicators shows a positive long-term relationship between gross capital formation with both economic growth [47,48] and the employment rate of labor [49].
Some studies have highlighted the importance of remittances from migrant workers as one of the main sources of external financing (representing a significant percentage of GDP) for economic growth, investment, and income distribution in beneficiary economies [25,38].
Globally, a massive decline (around 20%) in remittances is expected in 2020 due to the economic crisis induced by the COVID-19 pandemic, which is considered the strongest decline in recent history. At the same time, global money transfers to low-and middle-income countries are expected to fall by 19.7%, which is a significant loss of funding for many vulnerable households in their countries of origin. In Europe, remittance flows will fall even more sharply, with a projected decrease of 27.5%, as a direct consequence of the economic crisis triggered by COVID-19 and subsequent global bottlenecks that have lowered wages and abolished jobs [50].
Based on research conducted by authors on the influence of various factors on GDP, for independent states or grouped by region, we consider it appropriate to analyze the impact of investment, employment, and remittances on GDP growth.
The aim of this research is to empirically assess the potential effects of macroeconomic factors of economic growth and to determine the impact of the four indicators on the long-term GDP growth rate in the member states of the European Union (EU) during the periods 1996-2019, 1996-2008, and 2009-2019, by using multifactorial regression as the research method. We first aim to carry out an economic analysis of the evolution of indicators (G_GDP, GDP_C, GCF, Labor and Rem_GDP) in three periods: 1996-2019, 1996-2008 (before the global economic and financial crisis), and 2009-2019 (after the crisis) for all EU countries. The analysis is performed on each state. As the current pandemic started at the beginning of the year, we realized that, in fact, all of the results obtained from the study are severely affected, so we extended the research to 2020 and aimed to analyze the first effects of the COVID-19 crisis, even though the year 2020 is still unfinished and the COVID-19 crisis remains fully active. For this analysis we considered the data available for the first two quarters of 2020 and compared them to the first two quarters of 2019, for EU-27 countries (excluding UK).
After analyzing the evolution of the indicators, we move on to identifying the best regression model, based on which we continue to estimate the influence of remittances, the employment rate, and investments on the GDP growth rate. The assessment of the impact of these indicators on the well-being of a country is done individually for each country.
We then identify the best regression model based on which will go further to estimate the influence of remittances, employment rate, and investment on the growth rate of GDP. The processing is performed using econometric techniques provided by the software EViews. The results show that economic growth is influenced by the employment rate, investments that make an important contribution to the economic growth of each country, and worker remittances that contribute to economic growth in their countries of origin, although to a lesser extent than the other analyzed indicators.
Thus, the conclusions reported in this study represent a significant contribution to existing literature, primarily through the aggregate analysis of indicators for EU countries, as there is no such analysis in the literature. The results obtained from the estimation could be points of departure for debates leading to the adoption of the best decisions at EU level, but also at the level of national policies, on increasing investment and the employment rate. At the same time, the analysis of the comparative evolution of the indicators for the first two quarters of 2020 and 2019, which shows that the crisis caused by the COVID-19 pandemic will have an important effect on investments, employment, remittances, and their long-term economic impact, may underlie the adoption of budgetary and political measures to provide support to EU member states. Section 1 begins with a review of relevant studies on economic growth, remittances, employment and investment, and the relationship between them, demonstrating the importance of such a study. Section 2 presents the data sets, model, and estimation methodology that were used to support the analysis. Section 3 presents empirical investigations and the results of the analysis, consisting of three components. First, in order to create an overview of how these indicators evolved during the period of interest, a statistical analysis of the indicators is performed. Second, the best model for estimating the evolution and trends for indicators is identified, starting from the linear polynomial regression model and then adjusting it by differentiation and various econometric tests. Finally, an estimation of the model obtained is performed and then the estimation results are interpreted. Section 4 contains the conclusions of this study, including discussion on the economic interpretation of empirical results and possible directions for national governments or European policies related to the issues addressed.

Data, Models, and Methodology
The macroeconomic indicators used for the analysis follow:
The values of the indicators for the period 1996-2019 are taken from the World Bank [51] and are analyzed at the level of each country. For the first two quarters of 2019 and 2020, the data related to the studied indicators were taken from the Eurostat database [4]. First, we performed a statistical analysis of the analyzed indicators and their evolution, both at the level of each country and at the level of the European Union, and then compared the value of the indicators.
Then, starting from the representation of GDP through a Cobb-Douglas-type production function, which involves the decomposition of GDP dynamics into the contributions of capital, labor, and productivity of factors [15,52], we proposed to estimate the relationship between labor, investment and remittances, and economic growth. Thus, we considered as a dependent variable the growth rate of GDP, to which we added as independent variables the real GDP per capita, labor force participation rate, gross fixed capital formation, and personal remittances.
To identify a viable regression model, two scenarios were tested, each involving testing the stationarity of the data series, estimating the stationarity of their parameters, testing the model hypotheses, and interpreting the results.
The model we started from follows: where • y is the growth rate of real GDP per capita, the dependent variable, G_GDP; • a is the real GDP per capita, GDP_C, and b is labor force participation rate, Labor; • c is the gross fixed capital formation (% of GDP), GCF; • d is personal remittances received (% of GDP), Rem_GDP; • i represents the EU country and t is the reference time period; • the u i parameter allows for specific fixed effects; • ε it represents the estimated residues that represent deviations from the long-term relationship.
We tested the stationarity of the data at nominal values, and then due to the fact that most economic time series are far from stationary when expressed in their original units of measurement, we performed the logarithm of the dependent variables, obtaining Model (2).
We used cointegration tests, panel error correction models, and Granger causality tests to determine the long-term causal relationships and causal direction between G_GDP and other indicators.
Transformations such as logarithms can help to stabilize the variance of a time series [53]. We performed the estimation using the least squares method and tested the validity of the model, its degree of reliability, and the statistical significance of the parameters included in the model. After validation, the model was corrected to meet the assumptions of the multiple linear regression model. Thus, by applying these techniques, the model obtained follows: Differencing helps stabilize the mean of the time series by removing changes in the level of the time series, and therefore eliminating (or reducing) trend and seasonality [54].
Theoretical models were estimated and data processed using the econometric processing and analysis software, EViews 9.5. Thus, all the tables and figures are based on our own processing of the database, created by using the values of the indicators from the World Bank and Eurostat.

Statistical Analysis of Data
The first step of the study consists in statistical analysis of data. Each variable included in the model is analyzed in order to highlight its economic evolution in each country and compared to the EU, during the period for which the data were selected.
The annual GDP growth rate (G_GDP) allowed us to perform analyses and comparisons of the dynamics of economic development both over time and between the economies of EU states of different sizes. From an economic point of view, it can be said that at the level of the EU states, the economic growth registers a sinuous evolution, experiencing periods of significant economic growth at the level of all states (very large before the crisis of 2008) and, conversely, significant decreases of the economic growth rate (periods of economic crisis, 2009) ( Figure 1). Sustainability 2020, 12, x FOR PEER REVIEW 6 of 33 We used cointegration tests, panel error correction models, and Granger causality tests to determine the long-term causal relationships and causal direction between G_GDP and other indicators.
Transformations such as logarithms can help to stabilize the variance of a time series [53]. We performed the estimation using the least squares method and tested the validity of the model, its degree of reliability, and the statistical significance of the parameters included in the model. After validation, the model was corrected to meet the assumptions of the multiple linear regression model. Thus, by applying these techniques, the model obtained follows: Differencing helps stabilize the mean of the time series by removing changes in the level of the time series, and therefore eliminating (or reducing) trend and seasonality [54].
Theoretical models were estimated and data processed using the econometric processing and analysis software, EViews 9.5. Thus, all the tables and figures are based on our own processing of the database, created by using the values of the indicators from the World Bank and Eurostat.

Statistical Analysis of Data
The first step of the study consists in statistical analysis of data. Each variable included in the model is analyzed in order to highlight its economic evolution in each country and compared to the EU, during the period for which the data were selected.
The annual GDP growth rate (G_GDP) allowed us to perform analyses and comparisons of the dynamics of economic development both over time and between the economies of EU states of different sizes. From an economic point of view, it can be said that at the level of the EU states, the economic growth registers a sinuous evolution, experiencing periods of significant economic growth at the level of all states (very large before the crisis of 2008) and, conversely, significant decreases of the economic growth rate (periods of economic crisis, 2009) ( Figure 1).   The crisis was visible in the EU-28 since 2008, amid a significant slowdown in GDP growth. The recovery at EU-28 level meant an increase in the GDP index (based on chained volumes) of around 2% in 2010 and 2011, after which GDP fell in 2012 and 2013, before gradually registering higher positive rates of variation.
The decrease is obvious at the average G_GDP (see Appendix A, Table A1); thus, if in the period 1996-2008 the averages were between 1.06% (It) and 6.81% (Es), in the period 2009-2019 the average values were between −1.89% (Gr) to 4.34% (Ir). The significant decrease in G_GDP compared to the before the crisis period is shown in Figure 2. The crisis was visible in the EU-28 since 2008, amid a significant slowdown in GDP growth. The recovery at EU-28 level meant an increase in the GDP index (based on chained volumes) of around 2% in 2010 and 2011, after which GDP fell in 2012 and 2013, before gradually registering higher positive rates of variation.
The decrease is obvious at the average G_GDP (see Appendix A, Table A1); thus, if in the period 1996-2008 the averages were between 1.06% (It) and 6.81% (Es), in the period 2009-2019 the average values were between −1.89% (Gr) to 4.34% (Ir). The significant decrease in G_GDP compared to the before the crisis period is shown in Figure 2. Emerging countries had the highest annual GDP growth rates in 2019 compared to 1996 (over 4% in Bulgaria, Hungary, Lithuania, Poland, and Romania), and the lowest rates were recorded in Sweden (0.11%), Germany, and Luxembourg (with growth rates of about 0.3%), and Italy, Malta, England, and Finland (below 0.1%).
There is a significant reduction in the 2019 growth rate (1.36%) compared to 1996 and obviously there will be a significant decrease in 2020 due to the COVID-19 crisis, which has severely affected the economic growth of all countries. Thus, following the analysis of the first two quarters of 2020 compared to the same period of 2019 (Figure 3), it is observed that the year 2020 began with a relatively slow decline and then intensified in the second quarter, in all EU countries. At EU level (excluding the UK as it left the EU on 1 January 2020) in the first quarter of 2020, the decline in GDP was 4.4%, and in the second quarter compared to the same quarter of 2019 it was 16.65%.
At individual EU states level, in the first quarter of 2020, there is a slight increase compared to the same period of 2019 in most states (16 states). In the second quarter there is a drastic decrease in all EU states, the largest decrease recorded for Spain (20.81%), followed by Croatia and Greece, with over 17%.
For the 1996-2019 time interval, the average calculated at EU level (until 2020) identifies a level of economic growth of 1.53% per year. The homogeneity coefficient exceeds the value of one, revealing a completely heterogeneous structure, asymmetrical to the right, and strongly leptokurtic, with several values concentrated around the average, and high probabilities for extreme values. There is the same trend in all states, respectively, a strong heterogeneity, asymmetric to the right, and leptokurtic (see Appendix Table A1). Emerging countries had the highest annual GDP growth rates in 2019 compared to 1996 (over 4% in Bulgaria, Hungary, Lithuania, Poland, and Romania), and the lowest rates were recorded in Sweden (0.11%), Germany, and Luxembourg (with growth rates of about 0.3%), and Italy, Malta, England, and Finland (below 0.1%).
There is a significant reduction in the 2019 growth rate (1.36%) compared to 1996 and obviously there will be a significant decrease in 2020 due to the COVID-19 crisis, which has severely affected the economic growth of all countries. Thus, following the analysis of the first two quarters of 2020 compared to the same period of 2019 (Figure 3), it is observed that the year 2020 began with a relatively slow decline and then intensified in the second quarter, in all EU countries. At EU level (excluding the UK as it left the EU on 1 January 2020) in the first quarter of 2020, the decline in GDP was 4.4%, and in the second quarter compared to the same quarter of 2019 it was 16.65%.
At individual EU states level, in the first quarter of 2020, there is a slight increase compared to the same period of 2019 in most states (16 states). In the second quarter there is a drastic decrease in all EU states, the largest decrease recorded for Spain (20.81%), followed by Croatia and Greece, with over 17%.
For the 1996-2019 time interval, the average calculated at EU level (until 2020) identifies a level of economic growth of 1.53% per year. The homogeneity coefficient exceeds the value of one, revealing a completely heterogeneous structure, asymmetrical to the right, and strongly leptokurtic, with several values concentrated around the average, and high probabilities for extreme values. There is the same trend in all states, respectively, a strong heterogeneity, asymmetric to the right, and leptokurtic (see Appendix A Table A1).  In 2008, compared to 1996, GDP growth rate shows the relative heterogeneity in SL and heterogeneity for the other countries. Twenty-four states have a slight asymmetry, more pronounced to the right for Bg and UK, and two states have a slight asymmetry on the left (Ge and Sl). The distribution is leptokurtic for three states (Bg, Ir, UK), presenting higher probabilities for the extreme values, while for the other states the distribution is platykurtic, with values dispersed in a larger interval around the average (see Appendix A, Table A1).
Compared to 2009, in 2019 GDP growth rate shows heterogeneity for all 28 EU member states. Twenty-three states have a slight asymmetry, while other five have a more pronounced asymmetry, to the right (Dk, Es, Lt, UK) or to the left (Ir). The distribution is leptokurtic for 17 states (At, Be, Cz, Dk, Es, Fi, Fr, Ge, Hu, Ir, La, Lt, Lu, Nl, Sk, Sl, UK) with higher probabilities for extreme values, while for the other states the distribution is platykurtic, with values dispersed in a larger range around the average (see Appendix A, Table A1).
In order to assess living standards in the EU countries, we analyzed GDP per capita, thus eliminating the influence of the absolute size of the population. Following the analysis of the evolution of the GDP per capita for the period 1996-2019, for the EU-28 states, we found a significant increase of this indicator at the level of all states (Table 1).  In 2008, compared to 1996, GDP growth rate shows the relative heterogeneity in SL and heterogeneity for the other countries. Twenty-four states have a slight asymmetry, more pronounced to the right for Bg and UK, and two states have a slight asymmetry on the left (Ge and Sl). The distribution is leptokurtic for three states (Bg, Ir, UK), presenting higher probabilities for the extreme values, while for the other states the distribution is platykurtic, with values dispersed in a larger interval around the average (see Appendix A, Table A1).
Compared to 2009, in 2019 GDP growth rate shows heterogeneity for all 28 EU member states. Twenty-three states have a slight asymmetry, while other five have a more pronounced asymmetry, to the right (Dk, Es, Lt, UK) or to the left (Ir). The distribution is leptokurtic for 17 states (At, Be, Cz, Dk, Es, Fi, Fr, Ge, Hu, Ir, La, Lt, Lu, Nl, Sk, Sl, UK) with higher probabilities for extreme values, while for the other states the distribution is platykurtic, with values dispersed in a larger range around the average (see Appendix A, Table A1).
In order to assess living standards in the EU countries, we analyzed GDP per capita, thus eliminating the influence of the absolute size of the population. Following the analysis of the evolution of the GDP per capita for the period 1996-2019, for the EU-28 states, we found a significant increase of this indicator at the level of all states (Table 1). The most significant increases in GDP per capita are noticeable in the emerging states of Estonia, Latvia, Lithuania, and Romania (over 460%), and the smallest increases were in Italy and Greece ( Figure 4).  In 2008, the first year of the economic crisis, compared to 2007, GDP per capita increased in almost all states (except for Ir, which fell by 5.09%), although in a much lower percentage than in previous years. The value of GDP_C increased throughout the analyzed period (1996-2019). The largest increase is recorded for all states after the economic crisis (2009-2019). The evolution of the average GDP_C during 1996-2019 is represented in Figure 5. 120,000.00 In 2008, the first year of the economic crisis, compared to 2007, GDP per capita increased in almost all states (except for Ir, which fell by 5.09%), although in a much lower percentage than in previous years. The value of GDP_C increased throughout the analyzed period (1996-2019). The largest increase is recorded for all states after the economic crisis (2009-2019). The evolution of the average GDP_C during 1996-2019 is represented in Figure 5. In 2008, the first year of the economic crisis, compared to 2007, GDP per capita increased in almost all states (except for Ir, which fell by 5.09%), although in a much lower percentage than in previous years. The value of GDP_C increased throughout the analyzed period (1996-2019). The largest increase is recorded for all states after the economic crisis (2009-2019). The evolution of the average GDP_C during 1996-2019 is represented in Figure 5.  The homogeneity coefficient has a value that reveals a relatively heterogeneous structure of GDP_C values for both the EU and most member states (14 states). For Italy and Greece, the value of the homogeneity coefficient shows us that they have a relatively homogeneous structure and for the others the homogeneity coefficient shows a heterogeneous structure. All states have a platikurtic distribution with values scattered over a larger range around the average, most with an asymmetry to the left (22 and EU) or to the right (Cy, Fi, Gr, Nl, Sl, Sp).
In 2008 compared to 1996, GDP_C shows relative homogeneity for At, Be, Dk, Fr, Ge, It, Mt, Nl, Pt, Sw, and UK; relative heterogeneity for Bg, Cr, Cy, Cz, Fi, Gr, Hu, Ir, Lu, Po, Sk, Sl, and Sp; and heterogeneity for other countries. One state has a slight right asymmetry (Ir), and the other ones have a slight left asymmetry. The distribution is platykurtic, with values scattered in a larger range around the average for all 28 EU member states.
In 2019 compared to 2009, GDP_C shows homogeneity for Fi, Gr, It, and Nl; relative homogeneity for At, Be, Bg, Cr, Cy, Cz, Dk, Es, Fr, Ge, Hu, Lu, Mt, Po, Pt, Sk, Sl, Sp, Sw, and UK; and relative heterogeneity for Ir, La, Lt, Ro. One state has a slight right asymmetry (Ge), while the other states have a slight asymmetry to the left. The distribution is platykurtic, with values scattered in a larger range around the average for all 28 EU member states.
Analyzing GDP/capita in the first two quarters of 2020, we find that in the first quarter, at EU-27 level, this indicator decreased by 0.80% compared to 2019, and in the second quarter by 12.39 compared to the same period of 2019. At state level, there is a sharp decrease in the second quarter for all states, with the largest decrease in Spain (21.31%) and the least decrease in Ireland and Lithuania at around 3% (Table 2). Another indicator of appreciation of the level of economic development is the employed population. The labor force participation rate at EU level in 2019 was above the value in 1996, increasing by 3.92%, but at states level the situation is very different, with the evolution of this indicator varying between countries (Table 3).  Analyzing Labor by periods, there is a decrease in the labor force in 2009 compared to 2008 in eight states (Cy, Cz, Dk, Hu, Ir, Pt, Ro, Sl), but the mean value in the following period shows an increase in population occupied in most states ( Figure 6). For the economically developed states, however, we find a decrease in Labor throughout the period 1996-2019; this decrease is accentuated in the period following the 2009-2019 crisis. Italy has the lowest employment rate at around 49%, followed by Croatia at around 50%. The highest averages are recorded by economically developed countries (Sw, UK, Nl, Dk, Ir, Pt), mainly due to the attraction of labor force from emerging EU countries.
Fifteen countries have a labor force participation rate higher than EU; the highest increases in 2019 compared to 1996 are recorded in Croatia (31.37%) and Malta (28.18%), and the lowest increase is recorded in Finland (0.11%). It should be noted that in seven countries (Sk, Fr, Cz, Es, Po, Dk, Ro) the labor force participation rate decreased, and the most significant decrease was in Romania with a percentage of 14.9% (Figure 7).
The continuous increase in labor force participation rates in the EU contrasted with a decrease in population of 1.3%.
For the 1996-2019 interval, the statistical analysis of the indicator shows a homogeneous distribution of the labor force participation rate over the studied period; the indicator does not show extreme variations either at the EU level or at individual level. Italy has the lowest employment rate at around 49%, followed by Croatia at around 50%. The highest averages are recorded by economically developed countries (Sw, UK, Nl, Dk, Ir, Pt), mainly due to the attraction of labor force from emerging EU countries.
Fifteen countries have a labor force participation rate higher than EU; the highest increases in 2019 compared to 1996 are recorded in Croatia (31.37%) and Malta (28.18%), and the lowest increase is recorded in Finland (0.11%). It should be noted that in seven countries (Sk, Fr, Cz, Es, Po, Dk, Ro) the labor force participation rate decreased, and the most significant decrease was in Romania with a percentage of 14.9% (Figure 7). a percentage of 14.9% (Figure 7).
The continuous increase in labor force participation rates in the EU contrasted with a decrease in population of 1.3%.
For the 1996-2019 interval, the statistical analysis of the indicator shows a homogeneous distribution of the labor force participation rate over the studied period; the indicator does not show extreme variations either at the EU level or at individual level.  The continuous increase in labor force participation rates in the EU contrasted with a decrease in population of 1.3%.
For the 1996-2019 interval, the statistical analysis of the indicator shows a homogeneous distribution of the labor force participation rate over the studied period; the indicator does not show extreme variations either at the EU level or at individual level.
The average labor force participation rate also shows that most countries (18) have a higher average than the EU average; Sweden has the highest value (66.28%) and Italy has the lowest value (48.79%). In 2019, the trend is largely maintained; the labor force participation rate at EU level was 57.58%, with Sweden having the highest labor force participation rate (73.36%) and Italy the lowest (49.89%).
Greece, the Netherlands, and Malta have a leptokurtic distribution, sharper than the normal distribution, with more values concentrated around the average and showing extreme values. The other countries have a platikurtic distribution, with homogeneously dispersed values around the average. Bulgaria, Czechia, Estonia, Finland, Germany, Hungary, Romania, and Slovakia show a slight asymmetry to the left, and the others a slight asymmetry to the right.
In 2008 compared to 1996, labor force participation rates are homogeneous for all 28 EU member states. Sixteen states have a slight asymmetry to the right (At, Be, Dk, Fi, Fr, Gr, Hu, It, Lu, Mt, Nl, Pt, Sk, Sl, Sp, Sw), while the other states have a slight asymmetry to the left. The distribution is leptokurtic for one state (Sw), presenting higher probabilities for the extreme values, while for the other states the distribution is platykurtic, with values dispersed in a larger interval around the average.
For 2019 compared to 2009, labor force participation rates show homogeneity in all 28 EU member states. Eight states have a slight (Fr, Hu, It, Lu, Po, Ro, Sk, Sp) or a more pronounced (Sw) asymmetry to the right, while the other states have a slight asymmetry to the left, except for Ir where the left asymmetry is more pronounced. The distribution is leptokurtic for two states (Ir and Sw) presenting higher probabilities for the extreme values, and for the other states the distribution is platykurtic, with values dispersed in a larger interval around the average.
Regarding 2020 (Figure 8), the effect of the COVID-19 pandemic on the labor force participation rate is imperceptible in the first quarter; however, the decrease in the rate is evident for the second quarter in almost all states (except Croatia), the largest decrease recorded by Spain (almost 7%).
asymmetry to the right, while the other states have a slight asymmetry to the left, except for Ir where the left asymmetry is more pronounced. The distribution is leptokurtic for two states (Ir and Sw) presenting higher probabilities for the extreme values, and for the other states the distribution is platykurtic, with values dispersed in a larger interval around the average.
Regarding 2020 (Figure 8), the effect of the COVID-19 pandemic on the labor force participation rate is imperceptible in the first quarter; however, the decrease in the rate is evident for the second quarter in almost all states (except Croatia), the largest decrease recorded by Spain (almost 7%).
The main factor of economic growth along with the final consumption by the population is the gross fixed capital formation. Across the EU, GCF growth was only 1.3% in 2019 compared to 1996; this was due to the fact that in a number of important countries the value of investments in GDP remained at about the same value or even decreased when compared to 1996. Greece had the largest decrease in GCF (45%), followed by Sk (34%) and Cz, Pt, Cy, and Mt with over 20%. In contrast, the highest increases were registered by Bulgaria with 310% and Ireland with 119.4%, which, however, also had very strong variations (Table 4). The main factor of economic growth along with the final consumption by the population is the gross fixed capital formation. Across the EU, GCF growth was only 1.3% in 2019 compared to 1996; this was due to the fact that in a number of important countries the value of investments in GDP remained at about the same value or even decreased when compared to 1996.
Greece had the largest decrease in GCF (45%), followed by Sk (34%) and Cz, Pt, Cy, and Mt with over 20%. In contrast, the highest increases were registered by Bulgaria with 310% and Ireland with 119.4%, which, however, also had very strong variations (Table 4).
Analyzing GCF in the first year after the crisis, it is observed that in most states the value of investments decreased. The emerging states had higher increases (Bg 16.6% and Ro 5.5%) than the developed states (between 1-3%). After the crisis period (2009-2019), there is a moderate increase in investment in 16 states (At, Be, Cz, Dk, Es, Fi, Fr, Ge, Hu, Ir, La, Lt, Mt, Sk, Sw, UK), the highest increase in Ireland (105.97%) and the lowest La and Fi, around 3%. The other states registered decreases, the largest decrease registered by Greece with 45.08%.
The largest investments are obviously made until the crisis, averaging around 20% of GDP (excluding the UK, 17.66%). The average values for the period 1996-2019 are lower than those for the period 1996-2008, which means that gross fixed capital formation in the EU-28 has not fully recovered since its dramatic collapse in 2009, although it has shown relative increases each year ( Figure 9).
Thus, it is observed that at the level of EU member states (Figure 10), there are large differences in terms of investment intensity, which is reflected, in part, in the different stages of economic development of countries, as well as the dynamics of growth in later years. The average of GCF at EU level for the period 1996-2019 is 21.44%; 15 states have higher average values than the EU average, the highest averages registered by Estonia (28.05%) and Czechia (27.99%).  Thus, it is observed that at the level of EU member states (Figure 10), there are large differences in terms of investment intensity, which is reflected, in part, in the different stages of economic development of countries, as well as the dynamics of growth in later years. The average of GCF at EU level for the period 1996-2019 is 21.44%; 15 states have higher average values than the EU average, the highest averages registered by Estonia (28.05%) and Czechia (27.99%).
In 2019, gross fixed capital formation (in current prices) as a percentage of GDP was 21.64% at EU-28 level, with 13 countries having higher GCF shares than the EU average, with the highest values in Ireland (43.4%), followed by Hungary (28.6%), Estonia and Czechia (over 25%), and the other 15 countries registering values lower than the EU average, with the lowest level in Greece (11.42%).
In 2019 compared to 1996, according to the analysis of the homogeneity coefficient, it is observed that 13 states and the EU as a whole have a homogeneous structure of GCF values, 11 countries have a relatively homogeneous GCF value, and the other four states (Croatia, Cyprus, Estonia, and  in Ireland (43.4%), followed by Hungary (28.6%), Estonia and Czechia (over 25%), and the other 15 countries registering values lower than the EU average, with the lowest level in Greece (11.42%).
In 2019 compared to 1996, according to the analysis of the homogeneity coefficient, it is observed that 13 states and the EU as a whole have a homogeneous structure of GCF values, 11 countries have a relatively homogeneous GCF value, and the other four states (Croatia, Cyprus, Estonia, and Portugal) have relatively heterogeneous values.  In 2019, gross fixed capital formation (in current prices) as a percentage of GDP was 21.64% at EU-28 level, with 13 countries having higher GCF shares than the EU average, with the highest values in Ireland (43.4%), followed by Hungary (28.6%), Estonia and Czechia (over 25%), and the other 15 countries registering values lower than the EU average, with the lowest level in Greece (11.42%).
In 2019 compared to 1996, according to the analysis of the homogeneity coefficient, it is observed that 13 states and the EU as a whole have a homogeneous structure of GCF values, 11 countries have a relatively homogeneous GCF value, and the other four states (Croatia, Cyprus, Estonia, and Portugal) have relatively heterogeneous values.
GCF for most EU member states and the EU as a whole has a platikurtic distribution, with the exception of Bulgaria, Hungary, Ireland, Latvia, Lithuania, and Romania, which have a leptokurtic distribution. Half of the countries and the EU have a slight asymmetry to the left, and the other 14 states have a slight asymmetry to the right.
For 2008 compared to 1996, GCF shows homogeneity (for At, Be, Cz, Dk, Fi, Fr, Ge, Gr, Hu, It, Lu, Mt, Nl, Po, Pt, Sl, Sw, UK), relative homogeneity (for Cr, Cy, Es, Ir, Lt), and relative heterogeneity (La, Sk, Sp) or heterogeneity (Bg and Ro). Six states have a slight asymmetry to the right (Bg, Gr, It, Lu, Mt, Sp), while the others have a slight asymmetry to the left. The distribution is platykurtic, with values scattered in a larger range around the average for all 28 EU member states.
In 2019 compared to 2009, GCF shows homogeneity (in At, Be, Cr, Cz, Dk, Es, Fi, Fr, Ge, It, La, Lt, Lu, Nl, Po, Ro, Sk, Sl, Sp, Sw, UK), relative homogeneity (in Bg, Hu, Mt, Pt), or relative heterogeneity (Cy, Gr, Ir). Ten states have a slight asymmetry to the right (Cy, Es, Fi, La, Lt, Lu, Nl, Po, Ro, UK), while the other states have a slight asymmetry to the left, except for Cr where it is more pronounced. The distribution is leptokurtic for three states (Bg, Cr, Sl), presenting higher probabilities for the extreme values, while for the others the distribution is platykurtic, with values dispersed in a larger interval around the average.
At EU-27 level, the first quarter of 2020 marked an increase in the share of GCF in GDP of 3.4% (Table 5). Regarding the share of investments in GDP by state, in the first quarter of 2020 compared to 2019 there is a decrease in most states (except for Ro, Po, Nl, La, Lt, Ge, Cy, Cr, At), the largest decrease being registered by Luxembourg (12.94%). In the second quarter compared to the same period of 2019, 11 countries mark an increase in GCF in GDP, but for EU-27 there is a decrease of 6.5%. The largest decrease is recorded by Ireland (71%). Given that remittances represent a capital inflow into the country of origin, remittances can affect the country's economic balance and the macroeconomic variable GDP. Remittances are an important source of foreign transfers to developing countries and can be considered a financial mechanism for development. The trend of this currency inflow in the countries of origin obviously illustrates that the value of remittances decreases visibly [8,55], and the strategic analysis of this trend, especially due to the COVID-19 pandemic cannot be optimistic-foreign exchange inflows into countries of origin will fall drastically. The economic and geopolitical situation of the host countries is only one of these factors, but the UK's exit from the EU (1 January 2020) will clearly have a negative impact on remittance inflows to the external balance of payments of emerging countries.
The share of remittances in GDP represents an increase of 100.58% in 2019 compared to 1996. The largest increases in Rem_GDP were, as expected, in the emerging countries Lt, Ro, Sk, Es, Cz, Bg, Hu, with the share of remittances in GDP increasing from around 0.03% to over 3% (Table 6, Figure 11). A decrease in Rem_GDP was recorded in six countries (Gr, Ir, Lu, Pt, Sl, Sp) with the largest decrease recorded in Portugal (92.42%) and followed by Greece (84.73%). The share of remittances in GDP has increased significantly in emerging countries due to the free movement of people in the EU, which has led to an increase in labor migration to developed countries. In the pre-crisis period, developed countries received massive labor from emerging countries, which led to an increase in the value of remittances sent to their countries of origin, the largest increase recorded by Lithuania, followed by Estonia. In the post-crisis period, their share in GDP decreased but remained significant.
The analysis of the average share of remittances in GDP shows a significant increase after the The share of remittances in GDP has increased significantly in emerging countries due to the free movement of people in the EU, which has led to an increase in labor migration to developed countries.
In the pre-crisis period, developed countries received massive labor from emerging countries, which led to an increase in the value of remittances sent to their countries of origin, the largest increase recorded by Lithuania, followed by Estonia. In the post-crisis period, their share in GDP decreased but remained significant.
The analysis of the average share of remittances in GDP shows a significant increase after the crisis (2009-2019) ( Figure 12); most states recorded increases of average Rem_GDP, except for Pt and Gr, which recorded significant decreases. UK, Sp, Ir, and Dk generally remained at the same values. In the pre-crisis period, developed countries received massive labor from emerging countries, which led to an increase in the value of remittances sent to their countries of origin, the largest increase recorded by Lithuania, followed by Estonia. In the post-crisis period, their share in GDP decreased but remained significant.
The analysis of the average share of remittances in GDP shows a significant increase after the crisis (2009-2019) ( Figure 12); most states recorded increases of average Rem_GDP, except for Pt and Gr, which recorded significant decreases. UK, Sp, Ir, and Dk generally remained at the same values. In 2019 compared to 1996, the share of remittances in GDP shows homogeneity in Luxembourg, relative homogeneity in four states (At, Be, Dk, Fi), relative heterogeneity for four states (Cr, Fi, Ir, In 2019 compared to 1996, the share of remittances in GDP shows homogeneity in Luxembourg, relative homogeneity in four states (At, Be, Dk, Fi), relative heterogeneity for four states (Cr, Fi, Ir, Nl), and heterogeneity for the other states. Ten states have a slight asymmetry and Finland a more pronounced asymmetry to the right, and the other states generally have a slight left asymmetry (except Pt). The distribution is leptokurtic for six states (At, Fi, Fr, Ir, Pt, Sp) presenting higher probabilities for the extreme values, and for the other states the distribution is platikurtic, with values dispersed over a larger range around the average.
In 2008 compared to 1996, Rem_GDP shows homogeneity (in Be, Lu), relative homogeneity (in Cr, Fr, Ge), relative heterogeneity (in At, Dk, Fi, Hu, Ir, It, Nl, Sl), and heterogeneity for the other countries. Seven states have a slight (As, Cr, Dk, Fi, Hu, Lu, NL) or more pronounced (Fr) asymmetry to the right, while the other states have a slight asymmetry on the left, except Cy where it is more pronounced. The distribution is leptokurtic for six states (Cy, Fi, Fr, Ir, It, La) presenting higher probabilities for the extreme values, while for the others the distribution is platykurtic, with values dispersed in a larger interval around the average.
In 2019 compared to 2009, Rem_GDP shows homogeneity (in At, Be, Es, Fi, Ge, Lu, Pt), relative homogeneity (in Bg, Dk, Fr, It, La, Mt, Nl, Po, Sk, Sw), relative heterogeneity (in Cr, Cy, Hu, Ir, Lt, Sp), and heterogeneity for the other countries. Thirteen states have a slight (At, Cz, Dk, Fr, Ge, Hu, Ir, It, La, Pt, Ro, Sl, Sp) or more pronounced (Fi) asymmetry to the right, while the other states have a slight asymmetry to the left. The distribution is leptokurtic for one state (Fi) presenting higher probabilities for the extreme values, while for the others the distribution is platykurtic, with values dispersed in a larger interval around the average.

Analysis of the Stationary Value of the Indicators
For stationarity analysis of indicators, we started from Model (1) and verified the stationarity of the data series using the Dickey-Fuller test, the option with individual effects and the automatic lag selection. The results of the unit root test, LLC (Levin-Lin-Chu), performed on each country, show that the null hypothesis, the existence of a common unitary root, cannot be rejected, as these processes are partly nonstationary.
The results of the ADF (Augmented Dickey Fuller) tests for the first difference, performed for the indicators at nominal values, lead to the conclusion that the null hypothesis of the existence of a unit root can be rejected for all countries for the G_GDP variable, where the processes are stationary at a 1% confidence level. Unit root tests for the first difference showed that for the other four indicators, the processes are stationary for most states, the null hypothesis being rejected at a level of confidence differentiated from 1% to 10%.
Since there is a difference between the orders of magnitude, we proceeded to test a model in which we logarithmed the four independent variables, obtaining Equation (2): We checked the stationarity of the variables in logarithmic form. The results of the ADF-Fisher level tests were not significant. Thus, the first difference was tested, leading to the following conclusions: for GDP_C the null hypothesis of the existence of a unit root can be rejected for most states (except Greece and Spain) at a confidence level differentiated from 1% to 10%; for Labor, variables are stationary in all countries at a confidence level of 1-10%; and for GCF there is no single root in the 1-10% confidence interval. Therefore, processes are stationary for most states (with the exception of Italy, Portugal, Spain, and the EU as a whole). Regarding Rem_GDP, the states with the nonstationary processes are Austria, Estonia, Germany, Greece, and Sweden, otherwise the states are stationary and differentiated between 1-10%.
We then continued to stationary the nonstationary series (GDP_C, Labor, GCF and Rem_GDP), making the difference twice (second order of integration), and the newly generated series were tested again with the augmented Dickey-Fuller statistical test. The results show that for each of the variables tested, t-statistic is less than t-critical and p-value associated less than 5%, so the null hypothesis is rejected and the alternative hypothesis accepted, so the series have no unit roots, thus they are stationary (see Appendix A, Table A2).
Therefore, the multiple linear difference Model (2) is validated, as follows: To further validate Model (3), we used the Granger causality test. Taking into account Equation (3), only the causality of the variables Labor, GFC, and Rem_GDP toward GDP was reported in Table 7. The null hypothesis "GCF/Labor/Rem does not Granger-cause G_GDP" can be rejected if the probability of the test is inferior to a certain confidence level.   Considering the relevance level of 5%, it is observed that the probabilities attached to the t-statistic test are lower than this level, and thus the coefficients are statistically significant, being significantly different from 0. F-statistic for the proposed model has values with very low probabilities below 0.001, and therefore it is accepted that the multiple linear regression model studied is valid overall (see Appendix A, Table A3). The model passes the error autocorrelation tests according to Durbin-Watson (DW, around 2); thus, there is a positive serial correlation in most states. The determination ratio (R-squared) has values generally over 75%, showing the percentage by which the influence of significant factors is explained. Over 75% of the variation of G_GDP is explained by the variation of the dependent variables. The value of approximately 50% for adjusted R-squared also confirms the validity of the model.
At the 5% relevance level, it is observed that the probabilities attached to the t-statistic test are lower than this level for 14 countries in the period 1996-2008 (Cz, Dk, Ee, Ge, Gr, Ir, La, Lt, Nl, Po, Pt, Ro, Sk, Sp) and for 15 countries in the period 2009-2019 (Cr, Cy, Dk, Ee, Ge, Gr, Hu, Ir, It, La, Lt, Lu, Pt, Sp, UK), and thus the coefficients are statistically significant. The F-statistic for the proposed model has a very low probability value, below 0.001, and therefore it is accepted that the multiple linear regression model studied is generally valid (see Appendix A, Table A3).
The model passes the error autocorrelation tests, according to the Durbin-Watson test (DW around 2), so there is a positive serial correlation in most states for the two time periods. The determination ratio (R-square) has values generally over 68% for the period 1996-2008 and over 79% for the period 2009-2019, which shows the percentage that explains the influence of significant factors. Thus, over 68% (period 1996-2008) and over 79% (period 2009-2019) of the G_GDP variation is explained by the variation of the dependent variables. The value of about 50% for the adjusted R-square also confirms the validity of the model (except for Cr, Hu, and UK during 1996-2008, and Be during 2009-2019).
Next, we test the homoscedasticity of the model with the White test and the results show that the random errors are homoscedastic because the probability attached to the F-statistic is higher than 5% (see Appendix A, Table A4). In conclusion, the model chosen passes the error autocorrelation tests.
The economic analysis of the estimation results in Table 8, and comparison with the ones in Tables 9 and 10 for the two intervals 1996-2008 and 2009-2119, respectively, lead to the following conclusions: • The parameter sign of the indicator DL_GDP_C (β_1) is positive for all countries. This means that in the analyzed periods, keeping the other variables constant, the GDP growth rate increases with the increase of GDP per capita; at the same time, a decrease of GDP per capita would lead to a decrease of GDP growth rate. The labor force participation rate variable has a positive sign for 12 EU countries (Be, Cy, Dk, Fi, Gr, Hu, Ir, It, Lu, Mt, Pt, Sw), which means that keeping the other variables constant, the labor force participation rate contributes positively to economic growth, in accordance with the results found by [31,34]. In other countries and across the EU, the negative coefficient shows that when the employment rate increases, the economic growth rate decreases, according to the results found by [32]. Referring to the two intermediate periods, although the number of positive coefficients is close to that of the entire period (13 for the first period, 15 for the second one), the structure changes significantly. Thus, for some countries the impact of Labor became positive (Cr, Es, Gr, Ro, Sk, Sl)-mostly emerging countries, while for others, the impact of Labor became negative (Cy, Cz, Fi, Fr, Lt, Lu, Nl, Po, Sw).

•
The positive signs in Table 8 for the share of investments in GDP variable in most states (except Austria and Luxembourg) show that increased investments have an important contribution to the economic growth of each country, in accordance with the results found by [2,37,39], while decreasing investments would lead to a decrease of growth rate. The results in Tables 9 and 10 show differences during the two intermediate periods. Thus, for some countries the impact of share of investments in GDP became positive (Bg, Es), while for others, the impact of this indicator became negative (At, Cr, Fi, Gr, Ir, Sl). The coefficient of the Rem_GDP variable has a positive sign for emerging countries, suggesting that increasing worker remittances contribute to economic growth in these countries, consistent with other studies [15][16][17][18]. In the other countries this variable has a negative coefficient; decreasing remittances do not have an impact on economic growth, similar to the results reported previously [20][21][22][23][24]. Comparing to the two intermediate periods, there are only five differences of signs for the first period, and only 11 similar signs for the second period, showing the 2009 crisis causing a major change in the impact of remittances on economic growth rate. Thus, for some countries, the impact of increasing remittances become positive (Be, Bg, Cy, Gr, It, La); while for others, decreasing remittances have a positive effect on GDP growth rate (At, Cz, Es, Hu, Lt, Mt, Po, Sw, UK).

Discussions, Conclusions, and Proposals
The analysis of the evolution of the GDP per capita showed a significant increase at the level of all states, especially for the emerging ones, the largest increases recorded after the 2008 crisis.
From an economic point of view, it can be said that at the level of EU states there are considerable reductions in the GDP growth rate after the financial crisis of 2008, followed by a slight recovery, while in 2019 there is a substantial decrease for all states except Denmark, followed by a dramatic decline in 2020 due to the COVID-19 crisis that has severely affected the economic growth of all countries.
Obviously, the COVID-19 crisis will require that the EU community reinvent itself, and move on to reconstruction. The crisis is already visible in the EU-27, but only after the 2020 country reports will we be able to realize how hard all EU states have been affected. Thus, already against the background of a decrease in the GDP growth rate, there will be a considerable reduction in the GDP growth rate, which will be followed by a decrease in GDP per capita.
Although the labor force participation rate at EU level in 2019 increased compared to 1996, the evolution was very different between states. The employment rate decreased for emerging countries due to migration to developed countries, while it increased for the developed countries.
Regarding the GCF indicator, following the analysis we found that its value remained approximately at the same level or even decreased compared to 1996, failing a full recovery after its dramatic collapse in 2009, although it showed relative increases in each year. The largest investments were made until the crisis, and after the onset of the crisis, the value of investments decreased in most countries. At the level of EU member states, there were large differences in the intensity of investment determined by their degree of economic development. Investment values for the period 1996-2019 were lower than those for the period 1996-2008, which means that gross fixed capital formation in the EU-28 had not fully recovered after its dramatic collapse in 2009, although it showed relative increases every year.
The share of remittances in GDP is growing significantly in all countries in the post-crisis period, but mainly in emerging countries due to the free movement of people in the EU. The Rem_GDP analysis illustrates that its value is declining, and the economic and geopolitical situation of the host countries will clearly have a negative impact on remittance inflows into emerging countries.
After the statistical analysis, we estimated the relationship between remittances and economic growth based on several indicators. Thus, we proceeded to test multiple linear regression models, and determined the best model and then performed the economic analysis on the results. The estimate shows that economic growth is influenced by the employment rate, investments make an important contribution to the economic growth of each country, and worker remittances contribute to economic growth in the countries of origin, although to a lesser extent than the other analyzed indicators.
In conclusion, we can say that the decrease of the employed population and investments are the main factors that lead to the reduction of the GDP growth rate. Observing the downward trend of G_GDP over the analyzed period and the current crisis caused by COVID-19, we can conclude that the average GDP growth rate will be insufficient to close the development gaps of emerging countries compared to the EU average. One solution for improving the potential for long-term economic growth would be to increase investment, especially by attracting European funds.
At the same time, policies are needed to stimulate labor force participation in economic activity, to rebalance the demographic structure and its judicious distribution at the territorial level, and to invest in the education and development of human capital, so that each EU member state can experience economic development and sustainable growth.
The outlook for the value of remittances sent to countries of origin remains uncertain, as the impact of COVID-19 on the economic growth of EU countries is not yet known. In the past, remittances have been countercyclical, with workers sending more money home in times of crisis, but the COVID-19 pandemic has affected all countries, creating additional uncertainty, dramatically slowing migrant payments to households in developing countries. The shock of reduced remittance flows will have a severe and lasting impact on households that depended on this income to meet basic needs.
The COVID-19 epidemic is a major shock to the European economy, and EU member states have already adopted or are in the process of adopting budgetary and political measures to provide support to sectors that are particularly affected and implicitly to citizens. To this end, the European Commission has set up the "Investment Initiative" in response to COVID-19 to help member states finance their measures to respond to the COVID-19 crisis. The initiative combines the mobilization of immediate financial support from the Structural Funds to meet the most urgent needs and with the application of maximum flexibility in the use of funds. Acknowledgments: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflicts of Interest:
The authors declare no conflict of interest.

Random Errors F-Statistic Probability
Greater than 0.05