Investment in health and economic inclusion of South Eastern European countries.

Frasholli, Denada ; Hysa, Eglantina

Introduction

Health is an invaluable asset for human beings. Being healthy and living a long life are the most important goals of every individual. Good health has a significant importance for human existence and is a very important source of well-being. It is a key factor in a person's ability to develop his skills and knowledge and allows individuals to fully live their lives, without shortcomings or deficiencies. Health problems prevent people from performing their daily activities and are reflected as obstacles in the development of individual's potential during their entire lives.

According to economic theories, health is directly related to education and vice-versa. Meanwhile both, health and education are much correlated to the investments done in these sectors. On the other side, investments are related to the economic capacities of countries. Based on this perspective, rich countries are supposed to invest more and par consequence to have healthier population and skilled labor force, which means additional possibilities for investing more and more. Conversely, poor countries are supposed to have serious problems in investing for health and education and par consequence they suffer from different issues related to these two sectors.

In individual context, health is seen as a predictor for personal and economic development of everyone that possesses it. It is considered as basic factor in worker's productivity and individual's capacity to learn and grow intellectually. In economic context, health and education are the foundations of human capital, which is the basis of an individual's economic productivity (Shultz, Becker, 2001).

The macroeconomic evidence confirms that countries with the weakest conditions of health and education have a much harder time achieving sustained growth than do countries with better conditions of health (World Health Organization, 2001). Therefore, investment in health is an important strategic action for poverty reduction and a contributor to the general well-being of the population, leading to higher levels of economic growth in the long run.

This paper will study the causality from health investment to economic growth. The following sections will provide a review of the literature related to this topic and evidence of the causality from health investment to economic growth. The performance of a regression analysis of GDP per capita in SEE countries run on health expenditure per capita over the period 2000-2011 is taking place, in order to better understand how and at what extend GDP per capita varies with the changes in health expenditure per capita. The countries used in this regression analysis are: Albania, Bosnia and Herzegovina, Croatia, Former Yugoslav Republic of Macedonia, Montenegro and Serbia, excluding Kosovo since there are no available data. After analyzing the ANOVA table with the respective results, a brief conclusion of the research is being summarized.

Literature Review

One of the main issues in the field of health economics is to establish the direction of the causality between health and income. An informal explanation of this causality is: "a lot of people who otherwise wouldn't be poor are, simply because they are sick; however, few people who otherwise would be healthy are sick because they are poor" (Harold, 1978). There are several benefits that result from good health. Good health can enhance educational outcomes, both through school attendance and performance (Bloom, Canning, 2000; Schultz, 1999; Baldacci et al., 2004). Investing in health can improve individual productivity; healthy individuals are more likely to be efficient at assimilating knowledge, have stronger mental and physical capabilities and, in consequence, obtain higher productivity levels, and hence higher incomes (Strauss and Thomas, 1998). Healthy populations tend to have higher productivity due to their greater physical energy and mental clearness (Bloom, Canning, 2000). There is a general agreement among authors that better health can increase labor supply and productivity, ultimately leading to an increase in income (Muysken, Yetkiner, Ziesemer, 2003).

In the analysis of the impact of health on economic growth, Sorkin (1977) concluded that health, seen through reductions in mortality, had an important impact on economic growth during the early twentieth century. The relation between health expenditures and gross domestic product was studied by Heshmati in 2001 in a research through generalized Solow model. The variable representative of health status in the growth function was health expenditure.

After analysis, he concluded that health expenditures have a positive and significant impact on gross domestic product growth. A more general explanation of health impact on growth is given by Robin Swift. He argues that the improved health can positively impact economic growth via several channels, which include the rise in total GDP, but more significantly, through long term increases in both human and physical capital that in turn stimulate productivity and GDP per capita". There are two main approaches that the economists try to find some results. The first one is the effect of GDP in the health expenditure in a certain country. The other one is the amount of the health expenditure and its impact on the GDP growth on an economy. This study deals with the second issue, the observation of health expenditure in the overall GDP of a country.

Contribution of Health to Economic Growth

It is well recognized that economic growth leads to better health, because wealth means better nutrition and increased capability to invest in health. Anyway, this relation also runs in the other direction, which means better health contributes to economic growth. Investment in health is increasingly seen as a means of achieving economic growth. Good health, which results from health investments, contributes to the achievement of economic growth through: better efficiency or high labor productivity of healthy employees, reduced treatment burden, higher incentive to invest in education and training to obtain better skills, improved human capital, higher domestic saving and investment, higher rates of foreign investment and lower "dependency ratio".

Healthy employees are more efficient and productive in their work than others that suffer from diseases. The output per hour worked of healthy employees is higher compared to unhealthy ones. This is due to better physical and mental abilities of healthy employees and their reduced incentives to take days off work. As a result, better health leads to increased production as well as profitability of the firms. An indirect positive impact in the worker's productivity has the improved health of family members, which allows for less time devoted to caring for dependants and more time to work.

As people become healthier, their motivation to continue education and being equipped with better skills increases. Healthy individuals have better abilities to learn because of reduced diseases and better nutrition, generally miss fewer days from school and complete higher levels of education compared to unhealthy ones. What's more, if good health is a predictor of higher life expectancy, healthier individuals would have more incentive to invest in education and training, as the rate of depreciation of the gains in skills would be lower (Strauss, Thomas, 1998). Motivation by better health, better skills and education contribute to a growth in the human capital base which is a determinant of economic growth. Consequently, investment in health is an important predictor of economic growth.

The health status of an individual has potential impacts on the income level and its distribution between consumption, savings and investment. Healthier individuals will be more encouraged to save because of the expectation of a higher life expectancy. They also have more resources available to save, since they devote a small amount of resources to their health and usually save for their retirement.

These savings are a potential fund for investment. In addition to this, companies are more likely to invest when workforces are healthier or better educated. Investments in health with the target to prevent illnesses or treat them in the early stages are an efficient way to avoid larger expenses related to the illnesses and their possible complications. Individuals and also governments would be able to spend less for illnesses if investments to prevent them would have been made. This would provide more capital available for different investment purposes. Better health, increased labor productivity and control of diseases encourage foreign investment. Health improvements attract higher levels of foreign investment due to lower health risks for employees and higher growth opportunities for the companies. Increases in the level of foreign investment positively impact economic growth.

One way reduced poverty can be achieved is by investments in sexual and reproductive health. These investments result in demographic changes that could favorably lead to economic growth. The outcomes of these investments are generally smaller sized families due to lower fertility rates and a reduction of the "dependency ratio", which is the ratio of active workers to dependants. This allows for increased saving for individuals and increased investment opportunities to get higher profits. In the population level, increased national savings enhance economic growth by providing funding for investment.

Methodology

This study investigates the relation between health expenditure per capita and GDP per capita in South Eastern European Countries over the period 2000-2011. The goal is to infer the effect of health expenditure per capita on GDP per capita. It is important to know the effect that an increase in health expenditure per capita has on GDP per capita, while holding all other factors fixed. A regression of GDP per capita on health expenditure per capita is run in order to explain the first in terms of the second or in studying how GDP per capita varies with changes in health expenditure per capita. The data for the regression are taken from World Bank database.

Regression Analysis

Rivera and Currais estimated the relationship between health and economic growth of The Organization for Economic Co-operation and Development (OECD) member countries over the period 1960-1990 and showed that countries with higher health expenditures had higher economic growth. Other authors have also made research on the impact of health expenditures on economic growth. A study in thirty-three developing countries over the period 1990-1998, using the generalized Solow growth model and panel data model, resulted in a positive and significant impact of health expenditure on economic growth (Mojtahed, Javadipour, 2004). Based in the previous studies, higher levels of health expenditure per capita are associated with higher levels of GDP per capita. This study also expects the same relationship between the variables for the South Eastern European Countries. It is of great importance the analyses of both variables' trend for each country, which has been drawn in the graphs below. Albania, Bosnia and Herzegovina, the Former Yugoslav Republic of Macedonia, Montenegro and Serbia show the same trend of GDP per capita. They experience a constant increase from year to year for the period 2000-2008 and a decrease after 2008 followed by an increase in 2011. But it is not exactly the same trend for Croatia. It has experienced rapid and larger increases in GDP per capita from 2000 to 2008, and a higher fall during 2008-2010. The health expenditure per capita trend is almost the same for all SEE countries as that of GDP per capita.

[GRAPHIC 1 OMITTED]

The Scatterplot of GDP per capita and health expenditure per capita shows the picture of this relationship for years 2000-2011 for South Eastern European Countries. All these countries show positive relationship between health expenditure per capita and GDP per capita. The main conclusion based on this graph is that there exists a "unification" of these countries in health expenditure per capita--GDP per capita relationship. Croatia shows a quick recover of its economy which has been reflected as huge amounts of expenditure in health sector.

[GRAPHIC 2 OMITTED]

The other countries of the region show positive trends through years, theirs economy has been recovering. Even if the GDP of Albania, Bosnia and Herzegovina, Macedonia, Montenegro and Serbia have increased from 2000 to 2011, this growth rate was very slow and par consequence, the health expenditure in these countries is seen to be slow too. Meanwhile, all these countries show almost the same trend by showing that they have almost the same performance and they belong to the same region, having the same difficulties and challenges.

Next, the research demonstrates the regression equations and its' results. The regression equation is as follow:

y = [[beta].sub.0] + [[beta].sub.1]x + [mu]

(eq. 1)

Where:

y--is the depended or explained variable, in this case GDP per capita in current US $.

[[beta].sub.0]--is the intercept parameter, sometimes called the constant term; it is the predicted value of y when x = 0.

[[beta].sub.1]--is slope parameter in the relationship between y and x, holding the other factors in u fixed; it measures the predicted change in y for one-unit change in x.

x--is the independent or explanatory variable, in this case health expenditure per capita in current US $.

[mu]--is the error term or disturbance in the relationship, represents factors other than x that affect y. It stands for the "unobserved".

This equation is used for all the SEE countries data. It has been applied for each country in specific. The results founded help us comparing the countries with each other. The regression tries to explain the relation between GDP per capita and the independent variable, health expenditure per capita. Health expenditure per capita is expected to be statistically significant and have positive coefficients. An increase in this independent variable should increase the GDP per capita. The regression results are founded in table below.

Regression results and output tables for each country are founded in Appendix. Each table is statistically called an ANOVA table. The variation in the dependent variable is separated into two components: the explained variation and unexplained variation.

The total degrees of freedom for each equation is (n - 1) = 11 since totally there are 12 observations. The degree of freedom for regression is k, the number of independent variables. The degrees of freedom associated with the error term is n - (k + 1) = 10. The SS refers to the sum of squares, or the variation.

Total variation = SS total = [SIGMA][(Y - [bar.Y]).sup.2]

Error variation = SSE = E [(Y - [??]).sup.2]

Regression variation = SSR = [sigma][([??] - [bar.Y]).sup.2] = (SS Total - SSE)

MS refers to the mean square and is obtained by dividing the SS term by the df term. Thus, MSR, the mean square regression, is equal to SSR/k, and MSE equals SSE/ [n -(k + 1)].

It is likely that the estimation can include some error. The error in the predicted value of the dependent variable is measured by the multiple standard error of estimate.

[s.sub.y.12...k] = [square root of [SIGMA][(Y - [??]).sup.2]/n - (k + 1)] = 14.305

Another measure of the effectiveness of the regression equation is the coefficient of multiple determination, which is the proportion of the variation in the dependent variable, Y, that is explained by the set of independent variables [x.sub.1], [x.sub.2], [x.sub.3], ... [x.sub.k]

The coefficient of multiple determination, [R.sup.2], take the values from 0 to 1, which is the percent of the variation explained by the regression. The closer [R.sup.2] is to 1, the stronger the association between Y and the set of independent variables, [x.sub.1], [x.sub.2], [x.sub.3], ... [x.sub.k].

The ANOVA table is used to calculate the coefficient of multiple determination. It is the sum of squares due to the regression divided by the sum of squares total.

[R.sup.2] = SSR/SSTotal = 0.981

As the number of independent variables in the regression model increases, the coefficient of multiple determination increases. Even if the additional independent variable is not a good predictor, its inclusion in the model decreases SSE which in turn increases SSR and [R.sup.2]. In this case [R.sup.2.sub.adj] is used to measure the effectiveness of multiple regression models.

[R.sup.2.sub.adj] = 1 - SSE/n - (k + 1)/SSTotal/n - 1 = 0.976

The overall ability of the independent variables [X.sub.1], [X.sub.2], ... [X.sub.k], to explain the behavior of the dependent variable Y can be tested. Two tests of hypotheses are considered. The first one is called the global test, which investigates the possibility that all the regression coefficients are equal to zero.

It tests the overall ability of the set of independent variables to explain differences in the dependent variable. The null and the alternative hypothesis are as below:

[H.sub.0]: [b.sub.1] = 0 [H.sub.1]: [b.sub.1] is not = 0

The test statistic used is the F distribution is calculated to be the following:

F = MRS/MSE = SSR/k/SSE/[n - (k + 1)] = 208.76

By comparing F-stat to critical value of F, the null hypothesis is rejected,

Health expenditure per capita is found to be statistically significant for all the countries. In all of the cases related to the regression analysis, P-value is found to be less than 5 percent, showing the significance of the "health expenditure" variable in the respective equation. R square adjusted is another indicator showing that Gross Domestic Product is explained by health expenditure exactly by the amount of that value. For example, GDP of Albania and Serbia are 99 percent explained by "health expenditure" variable, GDP of Bosnia and Herzegovina and Croatia are 98 percent explained by "health expenditure" variable, GDP of Montenegro and Macedonia are explained by 97 percent and 95 percent respectively by "health expenditure" variable. There is a strong positive relationship between these two variables for all countries. Health expenditure per capita in Albania, as a measure of investing in health, explains slightly more of the variation in GDP per capita than in the other countries. On the other hand, Macedonia and Albania are the countries where health expenditure has the highest impact on economic growth. For the Albanian case, one percent increase in health expenditure, the GPD of this country increases by around 17 percent, while the GDP of Macedonia is affected by 19 percent. Croatia and Montenegro have an increase of GDP by 11 percent for a one percent increase of health expenditure. Meanwhile, Bosnia and Herzegovina and Serbia remain the countries less affected by the health expenditure; one percent increase in the health expenditure shows around 9 percent increase in their GDP.

Conclusion

Research indicated that health can actually drive or lead to economic growth. Good health is important both at individual and economic context. At the individual context, good health is a determinant of economic productivity. At the economic context, it is a determinant of human capital, which positively affects productivity and economic growth. As the review of the literature indicated, the impact of health in economic growth is extensively analyzed by many authors and there is a general agreement among them that better health effects positively the economic growth of a country. Investments in health contribute to the achievement of economic growth in a variety of ways, such as by higher labor productivity of healthy employees, reduced treatment burden, improved human capital, higher domestic saving and investment, higher rates of foreign investment and lower "dependency ratio". The study of the data for the South Eastern European Countries case supports the theory that health investment leads to economic growth. Research results reveal that there is a statistically significant positive relationship between health expenditure per capita and GDP per capita. The relationship between health expenditure per capita and GDP per capita is found to be slightly stronger in Albania, followed by Serbia with a very small difference. The findings of this study suggest that policy-makers interested in promoting the economic growth of a country should consider the role that health investment plays in it. The more contribution to the health sector, the more the GPD of South Eastern European Countries is supposed to be enhanced.

Appendix: E-views Outputs

Albania

Dependent Variable: Y
Method: Least Squares
Date: 01/15/14 Time: 14:12
Sample: 1 12
Included observations: 12
Y=C(1)+C(2)*X

                     Coefficient   Std. Error   t-Statistic   Prob.

C(1)                 -175.9593     71.03645     -2.477029     0.0327
C(2)                 17.24291      0.397252     43.40548      0.0000

R-squared            0.994720      Mean dependent var         2707.588
Adjusted R-squared   0.994192      S.D. dependent var         1143.518
S.E. of regression   87.14557      Akaike info criterion      11.92405
Sum squared resid    75943.50      Schwarz criterion          12.00487
Log likelihood       -69.54429     Durbin-Watson stat         2.657200

Bosnia and Herzegovina

Dependent Variable: Y
Method: Least Squares
Date: 01/15/14 Time: 14:31
Sample: 1 12
Included observations: 12
Y=C(1)+C(2)*X

                     Coefficient   Std. Error   t-Statistic   Prob.

C(1)                 583.1715      135.0020     4.319725      0.0015
C(2)                 8.800493      0.419361     20.98546      0.0000

R-squared            0.977797      Mean dependent var         3139.561
Adjusted R-squared   0.975577      S.D. dependent var         1289.851
S.E. of regression   201.5773      Akaike info criterion      13.60124
Sum squared resid    406334.3      Schwarz criterion          13.68205
Log likelihood       -79.60741     Durbin-Watson stat         1.163349

Croatia

Dependent Variable: Y
Method: Least Squares
Date: 01/15/14 Time: 14:40
Sample: 1 12
Included observations: 12
Y=C(1)+C(2)*X

                     Coefficient   Std. Error   t-Statistic   Prob.

C(1)                 1376.965      408.0474     3.374523      0.0071
C(2)                 11.74771      0.491241     23.91434      0.0000

R-squared            0.982815      Mean dependent var         10430.05
Adjusted R-squared   0.981096      S.D. dependent var         3837.021
S.E. of regression   527.5553      Akaike info criterion      15.52540
Sum squared resid    2783146.      Schwarz criterion          15.60621
Log likelihood       -91.15238     Durbin-Watson stat         0.436508

Macedonia

Dependent Variable: Y
Method: Least Squares
Date: 01/15/14 Time: 14:47
Sample: 1 12
Included observations: 12
Y=C(1)+C(2)*X

                     Coefficient   Std. Error   t-Statistic   Prob.

C(1)                 -1476.947     332.6863     -4.439459     0.0013
C(2)                 19.13914      1.318370     14.51727      0.0000

R-squared            0.954700      Mean dependent var         3215.173
Adjusted R-squared   0.950170      S.D. dependent var         1223.475
S.E. of regression   273.1115      Akaike info criterion      14.20865
Sum squared resid    745898.9      Schwarz criterion          14.28947
Log likelihood       -83.25189     Durbin-Watson stat         0.692166

Montenegro

Dependent Variable: Y
Method: Least Squares
Date: 01/15/14 Time: 14:54
Sample: 1 12
Included observations: 12
Y=C(1)+C(2)*X

                     Coefficient   Std. Error   t-Statistic   Prob.

C(1)                 250.1458      233.3168     1.072130      0.3089
C(2)                 11.12278      0.551981     20.15064      0.0000

R-squared            0.975964      Mean dependent var         4470.322
Adjusted R-squared   0.973561      S.D. dependent var         2190.841
S.E. of regression   356.2340      Akaike info criterion      14.74006
Sum squared resid    1269027.      Schwarz criterion          14.82088
Log likelihood       -86.44039     Durbin-Watson stat         1.374570

Serbia

Dependent Variable: Y
Method: Least Squares
Date: 01/15/14 Time: 13:41
Sample: 1 12
Included observations: 12
Y=C(1)+C(2)*X

                     Coefficient   Std. Error   t-Statistic   Prob.

C(1)                 523.7757      96.85058     5.408080      0.0003
C(2)                 8.770967      0.227282     38.59067      0.0000

R-squared            0.993330      Mean dependent var         3813.941
Adjusted R-squared   0.992663      S.D. dependent var         1858.165
S.E. of regression   159.1641      Akaike info criterion      13.12876
Sum squared resid    253332.0      Schwarz criterion          13.20958
Log likelihood       -76.77256     Durbin-Watson stat         1.025289

y--GDP per capita in current US $ (2000-2011)

x--Health expenditure per capita in current US $ (2000-2011)

References:

Bloom, D. E., Canning, D. (2000). Population Does Matter: Demography, Growth, and Poverty in the Developing World, New York: Oxford University Press.

Schultz, T. P. (1999). Health and Schooling Investments in Africa. Journal of Economic Perspectives, American Economic Association, vol. 13(3), 67-88.

Baldacci, E. et al. (2004). Social Spending, Human Capital, and Growth in Developing Countries: Implications for Achieving the MDGs. IMF Working Paper, wp/04/217, Washington DC.

Heshmati, A. (2001). On the causality between GDP and health care expenditure in augmented Solow growth model. Stockholm School of Economics Working Paper in Economics and Finance 423.

Luft, H. S. (1978). Poverty and Health: Economic Causes and Consequences, Cambridge: Ballinger Publishing.

Mojtahed, A., Javadipour, S. (2004). An investigation of the effects of health expenditure on economic growth (case study of developing countries). Iranian Journal of Economic Research, (19), 31-54.

Muysken, J., Yetkiner, I. H., Ziesemer, T. (2003). Labour Productivity and Growth. In Hagemann H., Seiter S., Health, Growth Theory and Growth Policy. London: Routledge.

Murphy, K. M., Topel, R. H. (2003) Measuring the Gains from Medical Research: An Economic Approach. Chicago: The University of Chicago Press.

Nordhaus, W. D. (2002). The health of nations: The contribution of improved health to living standards. NBER (National Bureau of Economic Research) Working Paper Series, Cambridge. Retrieved from: http://www.econ.yale.edu/~nordhaus/homepage/health_nber_w8818.pdf.

Rivera, B., Currais, L. (1999). Income Variation and Health Expenditure: Evidence for OECD Countries. Review of Development Economics, 3(3), 258-267.

Shultz, T., Becker, G. (2001). Macroeconomic and health: investing in health for economic development, Report of the Commission on Macroeconomics and Health, WHO, Geneva.

Sorkin, A. L. (1977). Health Economics in Developing Countries. Lexington, MA: Lexington Books.

Strauss, J., Thomas, D. (1998). Health, Nutrition and Economic Development, Journal of Economic Literature, vol. 2, 766-817.

Swift, R. (2011). The Relationship between Health and GDP in OECD Countries in the very Long Run. World Bank. (n.d.). Retrieved from: http://www.worldbank.org/.

World Health Organization (2008). Health systems, health and wealth: Assessing the case for investing in health systems. Retrieved from http://www.euro.who.int/__data/assets/pdf_file/0009/83997/E93699.pdf.

World Health Organization (2001). Macroeconomics and Health: Investing in Health for Economic Development. Report of the Commission on Macroeconomics and Health. Retrieved from http://www.who.int/.

Article Info

Received: April 29 2015

Accepted: May 30 2015

Denada Frasholli, Associate Professor PhD, Epoka University, Faculty of Economics and Administrative Sciences, Phone: 0035542232086, E-mail: dfrasholli10@epoka.edu.al

Eglantina Hysa, Lecturer, PhD, Epoka University, Faculty of Economics and Administrative Sciences, Phone: 00355672057557, E-mail: ehysa@epoka.edu.al.

Table 1. OLS Regression of GDP per Capita on Health
Expenditure per Capita

                       AL        BA        HR

Health Expenditure

Coefficient:          17.24     8.80      11.74
P-value:              0.00      0.00      0.00
                     (43.41)   (20.98)   (23.91)
R2_adj.               0.99      0.98      0.98

                       MR        SR        MK

Health Expenditure

Coefficient:          11.12     8.77      19.14
P-value:              0.00      0.00      0.00
                     (20.15)   (38.59)   (14.52)
R2_adj.               0.97      0.99      0.95

Note: The values in brackets are t-values