THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT IN LITHUANIA

Transcription

1 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT IN LITHUANIA A Thesis Presented to the Faculty of Finance Programme at ISM University of Management and Economics in Partial Fulfillment of the Requirements for the Degree of Bachelor of Finance by Mantas Traigys Advised by Inga Miliauskienė January 2017 Vilnius

2 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 2 Traigys, M., The impact of remittances on financial sector development in Lithuania. [Manuscript]: Bachelor Thesis: Finance. Vilnius, ISM university of Management and Economics, Summary The aim of this thesis is to evaluate how remittances influence the development of financial sector in Lithuania. In order to achieve this aim, firstly the situation analysis of trends of remittances and financial development of Lithuania in the period were examined. Secondly, previous studies regarding other factors influencing financial development and models for estimation of the link between remittances and financial development were analyzed. Furthermore, empirical research for several dimensions of financial development were conducted. Finally, conclusions about existing relationship and recommendations for further research were presented. Research methodology used in this paper consisted of analysis of scientific literature and previous empirical researches, unit root test for stationary data, ordinary least squares regression and statistical testing for autocorrelation, heteroskedasticity and functional misspecification. Results indicated that remittances have positively influenced depth of financial institutions and financial markets in Lithuania, but does not have any significant impact on efficiency of financial institutions or financial markets. efficiency. Keywords: Remittances, financial sector development, financial depth, financial

3 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 3 Traigys, M., Migrantų piniginių perlaidų įtaka Lietuvos finansinio sektoriaus vystymuisi. [Rankraštis]: bakalauro baigiamasis darbas: finansai. Vilnius, ISM Vadybos ir ekonomikos universitetas, Santrauka Šio darbo tikslas - nustatyti kaip migrantų piniginės perlaidos daro įtaką Lietuvos finansinio sektoriaus vystymuisi. Pirmiausia, tam, kad tikslas būtų pasiektas, buvo atlikta metų situacijos analizė. Tiksliui pasiekti buvo nustatytos piniginių perlaidų ir finansinio vystymosi Lietuvoje tendencijos. Taip pat buvo išanalizuota teorija nagrinėjanti kitų veiksnių įtaką finansiniam vystymuisi, bei teorija nagrinėjusi sąryšį tarp piniginių perlaidų ir finansinio vystymosi. Vėliau buvo atliktas empyrinis tyrimas, siekianti atrasti sąsajas tarp piniginių perlaidų ir finansinio vystymosi krypčių. Galiausiai, buvo pateikti empirinio tyrimo rezultatai, suformuluotos išvados ir pateikti pasiūlymai tolimeesniems tyrimams. Tyrimo metodologija naudota šiame darbe: mokslinės literatūros ir ankstesnių tyrimų analizė, stacionarumo tyrimas, trumpiausių kvadratų regresija ir statistinis testavimas nustatant autokoreliaciją, heteroskedastiškumą ir funkcinę mispecifikaciją Tyrimo rezultatai nustatė, kad migrantų piniginės perlaidos turi teigiamą itaką finansinių institucijų ir finansinės rinkos dydžiui, bet neturi jokios reikšmingos įtakos finansinių institucijų ir finansinės rinkos efektyvumui. Raktiniai žodžiai: Piniginės perlaidos, finansinis vystymasis, finansų sektoriaus dydis, finansų sektoriaus efektyvumas.

4 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 4 Contents List of tables... 7 List of figures... 7 Introduction Situation analysis General information Remittances trends in Lithuania Financial development in Lithuania Theoretical justification and research methodology The concepts of remittances and financial development Evidence and methodology in previous researches Financial depth of institutions Financial depth of markets Financial efficiency of institutions Financial efficiency of markets Financial access and financial stability Choice of the model Research methodology Control variables Data for the model... 32

5 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 5 3. Empirical research and results Workflow of research Regression model Augmented Dickey Fuller test Model for depth of financial institutions Model for efficiency of financial institutions Model for of financial markets Research implications, limitations and further outlook Conclusions References Appendix A Appendix B Appendix C Appendix D Appendix E Appendix F Appendix G Appendix H Appendix I Appendix J... 59

6 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 6 Appendix K Appendix L Appendix M... 62

7 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 7 List of tables Table 1. List of independent and dependent variables of the model Table 2. Augmented Dickey Fuller test for non-differentiated and differentiated data Table 3. OLS models for variables of financial depth of institutions Table 4. OLS models for variables of financial efficiency of institutions Table 5. OLS models for variables of financial efficiency of institutions Table 6. The results of various tests for models List of figures Figure 1. Personal remmitances as % of GDP, Foreign direct investment as % of GDP, GDP growth as annual percentage change in Lithuania in year Figure 2. The structure of migrants remmitances in Lithuania, Latvia, Estonia and Poland in Figure 3. Financial depth of Lithuania in year Figure 4. Financial efficiency of Lithuania in year Figure 5. Financial stability of Lithuania in year Figure 6. Financial access of Lithuania in year Figure 7. Financial aid, foreign direct investment and remittances in the world in year Figure 8. Net migration of Lithuania in Figure 9. Personal remittances, received (% of GDP) by country in Figure 10. Foreign direct investment, net inflows (% of GDP) by country in Figure 11. Financial development framework Figure 12. Private credit by deposit money banks to GDP (%) by country in

8 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 8 Figure 13. Deposit money banks' assets to GDP (%) by country in Figure 14. Stock market capitalization to GDP (%) by country in Figure 15. Bank net interest margin (%) by country in Figure 16. Bank overhead costs to total assets (%) by country in Figure 17. Stock market turnover ratio by country in Figure 18: Bank Z-score by country in Figure 19. Stock price volatility by country in Figure 20. Bank branches per 100,000 adults by country in Figure 21. ATMs per 100,000 adults by country in

9 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 9 Introduction The influence of remittances on national economy is highly debated topic in recent years. A considerable number of research (Paola Giuliano and Marta Ruiz-Arranz, 2005; Cooray, A. 2010) was made in order to estimate the impact of remittances on economic growth and financial sector progress on developing and developed countries all over the world. The potential of remittances to reduce poverty and FCI has been discussed in various research (Gupta, Pattillo & Wagh, 2007; Imai, Gaiha, Ali and Kaicker, 2012). Also, remittances can increase financial development of the country by boosting number of deposits with financial institutions, therefore, expanding country s financial system. In addition to that, remittances can increase economic growth in countries with developed financial sectors. Lithuania is one of the developed countries, which are extremely influenced by remittances. According to Statistics department of Lithuania nearly 590 thousand people emigrated from Lithuania in the last 15 years and remittances constituted 4.4% of GDP in Thus, the inflow of remittances makes Lithuania the second most dependable country on international remittances after Latvia in the European Union. Such dependency on remittances might have influenced the size and efficiency of financial sector in Lithuania. Despite previous researches on remittances impact on economy in developing countries, just little research was made to analyze the influence of remittances on financial sector s progress in developed countries, for instance Lithuania. The problem, raised in this thesis: what is the impact of remittances on the financial sector s development in Lithuania? The aim of the thesis is to evaluate the impact of remittances on financial development of Lithuania. The following objectives were set to achieve the determined aim: To overview the growth tendencies of remittances in Lithuania from.

10 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 10 To examine the financial sector development in Lithuania. To overview academic literature on the issue how remittances influence financial development of other countries. To examine the methodology for empirical study and select other variables that influence financial development of Lithuania To conduct an empirical research using selected data in order to identify the impact of remittances to financial sector s variables. Analyze the results, draw conclusions and provide suggestions for further research. Methodology of the research The analysis of scientific literature and previous empirical research was used to determine correct variables. Statistical data analysis was conducted in order to obtain stationary data observations for further analysis. Finally, ordinary least squares model was applied for time series data. All models and analysis were conducted using Gretl program. The significance of the research There was not many similar researches on evaluation of financial development made in Lithuania. Therefore, the results and conclusions of this paper may be significant for students as a background for future researches.

11 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT Situation analysis 1.1 General information Remittances plays an important role in world s economy. Received personal remittances were constantly increasing yearly and reached its peak of 533,125 Billion USD in Thus, remittances are considered as the second largest external financial source after foreign direct investment in developing countries (Giuliano and Ruiz-Arranz, 2005). Also remittances, unlike foreign direct investment or financial aid is stable financial source (see Appendix A). Remittances declined only in 2009 as a result of financial crisis. However, decrease is insignificant when compared to sharp decline of foreign direct investment or financial aid. The reason for constant increase of remittances is explained by the fact that migrants money transfers have cumulative effect as new migrants arrive and transfer money, older emigrants do not stop sending money home (Kasnauskienė and Buzytė, 2011). Consequently, the yearly growth of remittances does not follow the trends of migration. Futhermore, remittances tend to be counter cyclical and thus, can decrease the probability of financial crises (Bugamelli and Paterno, 2005). On the other hand, Giuliano and Ruiz-Arranz states that remittances follow cyclical behavior in economies with less developed financial systems (2005). According to World Bank, remittances tend to reduce poverty, increase spending on health and education and improve investment of small business and entrepreneurs. A positive association between remittances and financial sector can be explained in two ways. First of all, remittances sent by migrants tend to boost the demand of financial services by transfer. Secondly, remittances can be used to finance entrepreneurs who are not eligible for credit in banks. The link between remittances and development of financial sector is investigated in large amount of studies. Research by Aggarwal, et al. (2010) investigated the link of remittances to

12 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 12 financial development of developing countries by finding positive correlation between remittances and bank deposits and credit to GDP. Futhermore, Sami (2013) found that remittances inflow expedite the development of banking sector in Fiji. Also, remittances can boost availability of banking services thus increasing population s tie with formal financial system (IMF, 2005). Lastly, the study of 98 countries remittances relationship to financial development by Cooray (2010) concluded that remittances have positive link with changes of financial sector size and efficiency. On the other hand, Aggarwal et al. (2006) argue that remittances do not boost bank deposits, if nation does not trust the financial institutions of the country. Also, research by Acosta, Lartey and Mandelman (2007) concludes that remittances can cause Dutch disease phenomena in the country, which is unable to absorb them. To author s knowledge two studies regarding remittances influence on economic growth of Lithuania were conducted. The study by Kumar and Stauvermann (2014) investigated the impact of remittances on output per workers in Lithuania, and study by Kasnauskienė and Buzytė (2011) determined the influence of remittances on economies of Lithuania and Poland. The first study did not include financial development factors in the analysis, whereas the second study used one of the financial development factors as an independent variable in order to evaluate the impact on economies Remittances trends in Lithuania Emigration is a long ranging and highly debated issue in Lithuania. Emigration exceeds immigration by at least people per year from 1996 to 2014 (see Appendix B). Such shocking trend tends to be used as a hot in political discussions and in the media. However,

13 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 13 both, politicians and the media usually demonize migration (and especially emigration) by presenting only negative sides of the process ignoring the positives ones. In this thesis one of the positive sides of emigration consequences remittances and its impact on financial sector s development in Lithuania is analyzed. Remittances in Lithuania were constantly increasing from 1996 to This is a logical result due to constant migration tendencies discussed above. It confirms the theory that remittances have cumulative effect because of new and old emigrants transfers to Lithuania. Furthermore, remittances as percent of GDP of Lithuania were increasing and reached all time high of 4.51% in When compared to other Baltic states and Poland (see Appendix C) remittances of Lithuania ranks second in the magnitude remittances as % of GDP is lower than remittances in Latvia, but significantly higher than remittances in Estonia or Poland. Another factor that might influence financial sector s development is foreign direct investment (FDI). For Lithuania, Poland and Latvia (see Appendix C), yearly changes of FDI follows quite similar trend. The exception is Estonia, which tendencies of foreign direct investment are extraordinary. In 2009, the year after financial crisis, FDI in Estonia increased in 2009 and was equal to 9.5% when FDI in Latvia was negative and FDI in Lithuania was equal to 0.048%. When comparing trends of factors influencing financial development in Lithuania in (Figure 1), significant differences can be seen between variables. Remittances as % of GDP were not impacted by financial crisis in From 2008 to 2009 remittances as of % GDP even increased from 3.27% to 3.31%. Such trend confirms the research of Bugamelli and Paterno (2005) which concluded that remittances tend to be counter cyclical and it rejects the results of Giuliano and Ruiz-Arranz (2005) research, which stated that remittances follow cyclical behavior in economies with less developed financial systems.

14 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 14 Source: World Development Indicators database provided by World Bank 15,0% 10,0% 5,0% 0,0% -5,0% ,0% -15,0% -20,0% Personal remittances, received (% of GDP) GDP growth (annual %) Foreign direct investment, net inflows (% of GDP) Figure 1. Personal remmitances as % of GDP, Foreign direct investment as % of GDP, GDP growth as annual percentage change in Lithuania in year On the other hand, foreign direct investment follows more volatile trend. Also, this factor seems to be pro cyclical as it follows similar trend compared to annual GDP growth. As GDP declined by ~16% in 2009, FDI decreased by 3.94% and was equal to 0.05% of GDP, These results indicate that foreign direct investment are pro cyclical to economic situation in Lithuania and thus, they are absolute opposite in comparison to remittances of Lithuania. Amongst countries, migrants remittances are calculated differently. However in this thesis evaluation of remittances are taken from World Bank. It calculates remittances as a combination of 3 variables: workers remittances, compensation of employees and migrants transfers. The further explanation of these concepts will be provided in Theoretical justification and research methodology section. As the figure 2 below indicates, workers remittances in Lithuania in 2011 formed 82% of migrants remittances.

15 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 15 Source: World Development Indicators database provided by World Bank 100% 90% 18% 80% 70% 52% 60% 50% 40% 82% 100% 85% 30% 20% 10% 0% 48% 15% 0% Lithuania Latvia Estonia Poland Workers' remittances Compensation of employees Figure 2. The structure of migrants remmitances in Lithuania, Latvia, Estonia and Poland in 2011 It is the largest proportion when compared to Latvia, Estonia and Poland. In Estonia, 85% of remittances consisted of compensation of employees, while in Latvia all migrant s remittances were composed of it. It is worth noting that migrants transfers were not recorded in any examined country. Kasnauskienė and Buzytė state that workers remittances are the most correct variable since it shows the periodical and unpaid transfers from one country to another (2011). On the other hand, compensation of employees causes inaccuracies in calculations because it is hard to distinguish the proportion of money are spend abroad and send home. Therefore, such conclusion is favorable for this thesis as larger part of migrants remittances in Lithuania consist of workers remittances. Theoretically, there are two ways how remittances are consumed. Firstly, it is either are used for consumption to buy clothing, food, health care and etc. or are used as a source for investment and savings. Remittances might be used as another source of income for low income

16 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 16 households strongly affected by financial crisis in However, only 4% of Lithuanian population are living behind the poverty line, so it is safe to assume that bigger share of remittances are used for investment, rather than consumption to fulfill basic needs. Therefore, remittances inflows to Lithuania should have some positive effect for economic growth or financial sector s development of the country. Lastly, according to Kasnauskienė and Buzytė (2011) remittances are calculated by using database of Formal Remittances service providers (FRSP). This database consists of datasets from banks, credit unions and other internet banking providers. However, almost 50% of remittances are transferred using informal channels such as acquaintances, family members and etc. Therefore, the results and data of remittances might be misleading and do not reveal the real extent of remittances in Lithuania Financial development in Lithuania Financial sector development is a crucial part of economic development of the country. It benefits economy by increasing capital accumulation and technological progress. Furthermore, financial development tends to diminish inequality by creating ability for low income households to access financing. Financial development will appear when financial sector lowers its transaction costs, optimizes the availability of information and thus improves key functions of financial sector ( Financial development, n.d.). However, it is very difficult to evaluate financial sector development of the country because it has wide range of dimensions. Therefore, in this thesis, framework developed by World Bank s Global Financial Development Database (see Appendix D) will be used to measure financial sector development. It examines two elements of financial sector - financial markets and financial institutions. Further, it defines four dimensions of development: financial depth, access, efficiency and stability. Few variables and

17 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 17 ratios of Lithuania from every dimension will be examined in this section. Also, they will be compared with variables of Latvia, Estonia and Poland. Dimensions and variables will be explicitly explained in the Theoretical justification and research methodology section. Financial Depth shows the size of financial sector compared to the economy. In order to measure financial depth of financial institutions, empirical studies mentioned previously tend to examine private sector credit to GDP. This variable correlates with long-term economic growth and poverty reduction. Another variable that can be measured is total banking assets to GDP it also includes credit to government and other banking assets. For financial markets development evaluation mainly stock market capitalization to GDP is used. As the Figure 3 below indicates, Financial Depth of institutions variables were constantly increasing until , but started declining since then. On the order hand, stock market capitalization to GDP started declining in 2007 and were more affected by financial crisis it plummeted to 10% and did not show any signs of recovery (World Bank does not provide results of this variable after 2012). Source: Global Financial Development database provided by World Bank. Financial Depth of Lithuania Private credit by deposit money banks to GDP (%) Deposit money banks' assets to GDP (%) Stock market capitalization to GDP (%) Figure 3. Financial depth of Lithuania in year

18 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 18 Then compared to other countries (see Appendix E and Appendix F), Lithuania tends to follow very similar trend over years. On the other hand, the size of financial institutions relative to economy in Lithuania seems to be lowest of all 4 countries. While results of Latvia and Estonia in 2010 reaches 95-96% and are similar to high-income developed countries (private credit to GDP average of 103%, according to World Bank), Lithuania is stuck with only 60% of GDP. Another dimension is financial efficiency which measures the cost of services provided by financial intermediates. The main variable used for evaluating efficiency of financial markets is stock market turnover ratio. To estimate efficiency of financial institutions previous studies (Cooray, 2010) mainly used two variables: Bank net interest margin and Bank overhead costs to total assets. Source: Global Financial Development database provided by World 30,00 25,00 20,00 15,00 10,00 5,00 0, Bank net interest margin (%) Bank overhead costs to total assets (%) Stock market turnover ratio (%) Figure 4. Financial efficiency of Lithuania in year

19 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 19 As the figure 4 above shows, financial institutions efficiency was gradually increasing over the years as bank overhead costs and net interest margin were declining. However, stock market turnover ratio were varying sharply and was decreasing over recent years. When compared to other Baltic States and Poland (Appendix F and Appendix G), Lithuania seems to have the most efficient financial institutions as both efficiency variables were declining most rapidly. On the other hand, results of stock market turnover ratio (see Appendix G) shows that Lithuania has the second least efficient financial markets in comparison to other 3 countries. Third dimension describing financial sector development is stability. It shows how financial system is able to cope with financial distress. Main variable indicating stability of financial institutions is bank Z-score. It has negative link to insolvency of financial institutions, thus a higher z-score shows a lower probability of insolvency (Martin Čihák et al., 2012). In order to evaluate stability of financial markets, variable of Stock price volatility is usually used. Source: Global Financial Development database provided by World Bank. Figure 5. Financial stability of Lithuania in year Stock price volatility and Bank Z-score trends in Lithuania are quite similar (see Figure 5). They both were hugely influenced by financial crisis in Stock price volatility spiked by

20 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 20 13% from 2008 to 2009 and Bank Z-score decreased 4 times. When comparing financial efficiency of Lithuania to other countries (see Appendix H), results are different. Lithuania and Latvia have similar Z-scores, but Latvia was less affected by financial crises (as well as Poland). On the contrary, stock price volatility of all 4 countries followed very similar trend. All of them were highly affected by financial crisis, but managed to reduce the stock price volatility in recent years. Last dimension which describes financial sector development is financial access. It indicates the ability of people (firms and individuals) to access financial services. In the empirical studies, mainly two variables were used to evaluate accessibility of financial sector - the number of bank accounts per 1000 adults and Market capitalization excluding top 10 companies to total market capitalization. However, none of these variables are available in the case of Lithuania. Therefore, other two variables, ATMs per 100,000 adults and Bank branches per 100,000 adults will be used in order to measure accessibility of financial sector. Also, both variables evaluate financial institutions since there is no data available to measure access to financial markets in Baltic States. As the Figure 6 shows, amount of bank branches per adults was increasing until 2007 and remain stable since then. As for ATMs per 100,000 adults, there was significant growth until 2010 and sharp decrease after Amount of ATMs did not fluctuate much from 2011 to 2014.

21 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 21 Source: Global Financial Development database provided by World Bank. 70,00 60,00 50,00 40,00 30,00 20,00 10,00 0, Bank branches per 100,000 adults ATMs per 100,000 adults Figure 6. Financial access of Lithuania in year When compared to other Baltic States and Poland (Appendix I), Lithuania stands in the middle as their amount of bank branches per 100,000 adults was higher than in Estonia, but lower than in Latvia and Poland. As for ATMs per 100,000 adults, results of Lithuania are the lowest. Value of the variable in Lithuania in 2014 was equal to 51, while second lowest result of Latvia was equal to Such results show that people in Estonia, Latvia or Poland can access ATMs more easily.

22 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT Theoretical justification and research methodology 2.1. The concepts of remittances and financial development According to Eurostat Remittances is household income being generated by economic activity in another than the home economy, which subsequently is transferred to or earned on the account of the household in the home economy ( Glossary: Remittances, n.d.). As mentioned previously, remittances are sum of 3 variables: workers remittances, compensation of employees and migrants transfers. Such remittances calculation system was used until 2008 when IMF released Balance of Payments Manual (BPM6). There it announced new definition of remittances. Now it consists of personal transfers and compensation of employees. However, part of the data which will be used in this thesis are older than 2008, therefore earlier definition of remittances is going to be implemented. In BPM5 workers remittances was defined as transfer made by migrants employed and resident in the compiling economy to their relatives in their country of origin (Glossary: Workers remittance, n.d.). Compensation of employees is defined as the total remuneration, in cash or in kind, payable by an enterprise to an employee in return for work done by the latter during the accounting period (System of National Accounts (SNA), 2008, p. 189). It consists of two components: salaries and social insurance contributions paid by employer. Because remittances of Lithuania does not include (or data are not collectable) migrants transfers, the concept of it will not be explained. Financial sector is the set of institutions, instruments, markets, as well as the legal and regulatory framework that permit transactions to be made by extending credit (Directorate, n.d.). World Bank describes five functions of a financial sector:

23 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 23 (i) producing information ex ante about possible investments and allocate capital; (ii) monitoring investments and exerting corporate governance after providing finance; (iii) facilitating the trading, diversification, and management of risk; (iv) mobilizing and pooling savings; and (v) easing the exchange of goods and services. ( Financial development, n.d.) Financial development will appear when financial sector lowers its transaction costs, optimizes the availability of information and thus improve the functions described above ( Financial development, n.d.). In order to evaluate financial sector development of Lithuania the framework of World Bank s Global Financial Development Database will be used and implemented (see Appendix D). It classifies financial development into 4 categories: Financial Depth It shows the size of financial sector (financial institutions and financial markets) in comparison to the economy. Economy is measured by the economic output ( Financial Depth, n.d.). Financial efficiency It shows the ability of financial institutions and markets to grant services at the lowest costs (Sobiech, 2015). Financial access World Bank defines it as The ability of individuals and firms in an economy to access financial services (Global Financial Development Report, 2013, p. 47). Financial stability

24 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 24 It indicates resilience of financial institutions and markets to stress ( Financial stability, n.d.). Remittances, as the stable income source can provide the opportunity for large part of the population to use formal channels of financing (banks and etc.) instead of the informal ones. Therefore, remittances can increase size and access of financial institutions and markets. Also, while bigger part of remittances tend to be spend on consumption, some part of it can be saved or invested in stock markets or other areas. Thus, remittances might have influence on the market capitalization of countries. Remittances can have influence on providing means for people to access finance in rural territories (Cooray, 2010). Further Cooray found that remittances allow banks to overhead costs and net interest margins, thus achieving greater efficiency (2010). In addition to that, remittances tend to be counter cyclical and can decrease the probability of financial crises (Bugamelli and Paterno, 2005). Therefore, remittances can have an impact on financial stability Evidence and methodology in previous researches Financial depth of institutions The impact of remittances on financial depth was analyzed in numerous researches. While the main variables used to estimate influence on depth of financial institutions were ratios of Private credit by deposit money banks to GDP and Deposit money banks' assets to GDP, methodology of the previous researches are quite different to one another. Ratio of Private credit by deposit money banks to GDP shows how much resources are granted to the private sector by the domestic money banks compared to the GDP (World Bank, 2016) and Deposit money banks' assets to GDP indicates how much assets are controlled by deposit money banks compared to GDP of a country (World Bank, 2016). Domestic money banks stands for financial institutions which can receive transferable deposits.

25 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 25 Cooray (2010) used annual data from 1990 to 2007 for 98 low to medium income countries. Research used ratios of deposit banks assets to GDP and private sector credit to GDP in order to find the relationship between remittances and depth of financial institutions. Also, research implemented several control variables, such as the level of per capita income, ratio of exports to GDP, the ratio of FDI to GDP and a dummy variable for the exchange rate in order to evaluate the financial development of the country. As the estimation methods, research used pooled OLS and system GMM methods. The results of OLS methods showed significant positive correlation between remittances and financial depth as 1% increase in remittances led to 0.04% boost of bank deposits and 0.03% raise in private credit to GDP. Futhermore, Gupta, Pattillo and Wagh (2007) investigated how remittances impact the financial depth of 44 Sub-Saharan Africa countries. The methods of research was random and fixed effects panel regressions which used unbalanced panel composed of six time periods of 5 years average ranging from 1975 to The financial depth was estimated by ratios of bank deposits to GDP and M2 to GDP. The control variables consisted of log of GDP, per capita GDP, Inflation (GDP deflator as % of GDP), exchange rate, ratio of imports and exports to GDP and the sum of FDI and financial aid to GDP. The research found that remittances has positive and significant relationship to ratios of bank deposits to GDP and M2 to GDP. Moreover, Oke (2011) estimated the relationship between remittances and financial depth variables: ratios of money supply to GDP and private credit to GDP for Nigeria. Research used OLS regression and GMM estimation for the annual data from 1997 to Remittances, financial depth variables and control variables, such as the annual change of GDP deflator, ratio of trade to GDP and lag value of remittances were taken in the log form (exceptions were exchange rate and financial liberation variables). Such transformation of variables was made in order to prevent spurious regression from the model. The results of estimations showed that both financial depth variables are significantly and positively correlated

26 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 26 against remittances and one percent increase in remittances boosts ratios of money supply to GDP by 0.09 and credits to GDP by 0.1. Similar research methodology was used by Goschin (2013) to estimate the link between remittances and economic growth in Romania using annual data from 1994 to Research used ADF test to check if variables are stationary and OLS regression for the estimation. While the results of this research is not handy, the methodology could be very useful for the development of the model in this thesis. Masuduzzaman (2014) used annual data from 1981 to 2013 to evaluate workers remittances impact on financial development in Bangladesh. Research used annual data from 1981 to The proxy for development was financial depth, more specifically, the ratios of M2 to GDP, private domestic deposit to GDP and bank credit to GDP. The control variables were log of real GDP (country size), inflation (annual % change of GDP deflator), real interest rate and the ratio of sum of imports and exports to GDP as the measurement of openness in economy. To estimate if time series are stationary Augmented Dickey Fuller (ADF) and Philips Perron (PP) was used. Afterwards, Johansen co-integration test was implemented. It showed that workers remittances have positive and significant influence on all 3 variables of financial depth. Similar research to Masuduzzaman (2014) was made by Sibindi (2014). It evaluated the relationship between remittances per capita and financial development (per capita broad money) in Lesotho using annual data ranging from 1975 to Research also used PP and ADF tests for stationarity as well as Johansen Test for cointegration. It determined the positive and significant link between remittances and financial development in Lesotho. Furthermore, Chia Yee Ee (2014) estimated the linkage between remittances, financial depth (domestic credit provided by the banking sector to GDP) and economic growth in Malaysia. It also used annual

27 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 27 data from 1984 to 2013, as well as ADF and Johansen co-integration test. This research also determined the positive relationship between remittances and financial depth Financial depth of markets The main variable of financial market depth is the ratio of stock market capitalization to GDP. It shows the value of shares in stock market as a percentage of GDP (World Bank, 2016). Several studies showed that there is positive relationship between remittances and depth of financial market. Mandaci, Aktan Gumus and Tvaronavičienė (2013) estimated stock market development of 30 countries in the period of by using Random-effect SUR estimation. Research found that remittances, as well as, FDI and credit provided by banking sector to GDP have statistically significant positive effects on market capitalization as a proxy for stock market development. 1% increase in workers remittances improved market capitalization by 0.06%. Shahzad, Adnan, Ali and Raza (2014) used panel data of five South Asian countries in the period of to investigate the impact of remittances on financial development. Financial development was determined by 8 different factors 2 insurances variables, 3 stock market development variables (Stock market capitalization to GDP, Stock market total value traded ratio to GDP, Stock market turnover ratio to GDP) and 3 variables of financial institutions development (Liquid liabilities to GDP, Deposit money bank assets to GDP, Private credit by deposit money banks and other financial institutions to GDP). Principal component analysis (PCA) was used to create single variable of financial development index. Further, pooled OLS and GMM dynamic system estimation models with control variables such as GDP as log, GDP per capita in log, exports (% of GDP), Inflation, and FDI were implemented. Results of OLS estimation showed that 1% increase of workers remittances boosted financial development by

28 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT percent. GMM model indicated positive and significant relationship between remittances and the index of financial development as well. Aigheyisi and Edore (2013) used annual data from 1991 to 2011 for Ghana and 1989 to 2011 for Nigeria. Research implemented ordinary least squares and Cochrane-Orcutt estimations in order to determine the linkage between Market capitalization ratio and Foreign Financial Resources Inflows such as FDI, remittances, Foreign portfolio investment, Financial aid and public external debt. The results of OLS method failed the autocorrelation test, thus the Cochrane-Orcutt method was used. It showed that the link between remittances and stock market capitalization, as the proxy for stock market development is positive and significant in both, Nigeria and Ghana. On the contrary, El-Nader and Alraimony (2013) used Johansen cointegration test and found that growth of remittances to GDP in Jordan is negatively correlated to Stock market capitalization to GDP in the period of 1990 to El-Nader and Alraimony (2013) explained it by the idea that marginal propensity to consume in Jordan was high and steady in that period while the most of the remittances were used for acquire education, accommodation and land. On the other hand financial crisis stopped capital flows in the country Financial efficiency of institutions In order to estimate efficiency of financial institutions, mainly two variables are used: bank overhead costs and net interest margin. Overhead costs shows operating expenses of a bank in comparison to all assets held by bank (World Bank, 2016) and net interest margin shows Accounting value of bank's net interest revenue as a share of its average interest-bearing (total earning) assets according to World Bank (Global Financial Development Report, 2013, p. 45),

29 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 29 thus the decline in these two variables leads to the overall increase in efficiency of financial institutions. By author s knowledge, two researches evaluated the relationship between the financial efficiency of institutions and remittances. Cooray (2010) research, which was mentioned above, using GMM and OLS methods estimated that the 1% increase in remittances results in the decline of 0.003% in overhead costs and decrease of 0.004% in net interest margins. Sobiech (2015) used overhead costs to total assets, deposit interest rate and interest rate spread as the proxies for institutional efficiency. These 3 variables with other 5 variables as proxies for financial depth were implemented to form a financial development indicator of developing countries from 1970 to Research was made to identify if remittances have impact on economic growth through the financial development index while using system GMM and quasi-maximum likelihood estimations. However, it is still useful since it evaluates the link between workers remittances and financial development Financial efficiency of markets Main variable estimating the efficiency of financial markets is stock market turnover ratio. It the division between value of shares traded and average market capitalization (World Bank, 2016). By author s knowledge, only one research analyzed the link between efficiency of financial markets and remittances. The research by Shahzad, Adnan, Ali and Raza (2014) described previously, used stock market turnover ratio as one of 8 indicators to create the financial development index. This index had the positive correlation with remittances. Thus it indicated that remittances have positive influence on financial efficiency of markets in Pakistan, India, Sri Lanka, Bangladesh and Nepal.

30 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT Financial access and financial stability To author s knowledge, no previous researches were made regarding the link between remittances and financial access or financial stability. While some researches vaguely examined remittances and some variables of access to banking sector, they did not include any variables which were examined in situation analysis section (z-score and stock volatility for financial stability or bank branches per and ATMs per 100,000 adults). Also, variables of financial access lack data points and could create various limitations for the model. As for financial stability, both variables have sufficient data points, however no research was made in order to estimate the link between them and remittances. Therefore, both, financial stability and financial access will be excluded from further analysis. Analysis and model will examine the link between remittances and two dimensions of financial sector development financial depth and financial efficiency in Lithuania Choice of the model Research methodology For the estimation of relationship between remittances and financial sector variables OLS regression will be used. Such estimation model was used in several researches described previously (Cooray, (2010); Shahzad, Adnan, Ali and Raza (2014); Aigheyisi and Edore (2013); Oke (2011)). Other type of research method could be Johansen s co-integration test. However, such model in previous researches was mainly used as a method for founding a relationship between financial development and remittances and then used in Engle-Granger method to find a link between remittances and economic growth through financial development. In addition to that, is hard to interpret the results of Johansen s co-integration test and it cannot be used if variables are integrated at the different orders.

31 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 31 Futhermore, it would be convenient to use Principal Component Analysis as it was used in Shahzad, Adnan, Ali and Raza (2014) research. It would provide an opportunity to convert 6 financial development variables into one financial development index for the further estimation. However, there is not enough observation points of financial development variables for Lithuania (annual data ranging from 1996 to 2014, thus 19 observation points.). Thus, with such small sample results of Principal Component Analysis might be inconsistent and incorrect. Moreover, while bigger part of previously analyzed researches used cross sectional data of two or more countries, dataset of this thesis consists of time series data. Therefore, Augmented Dickey Fuller test will be used to found if variables of model is stationary. If variable is found to be non-stationary, the first difference or logarithmic transformation of variables will be taken in order to make it stationary. Such transformation prevents model from spurious regression, which stands for the relationship of two or more variables which are strongly correlated together (R squared equals to 0.95 or more) due to coincidence while in reality they are not causally related. In conclusion, the research will consist of ADF test to find if variables are stationary, 6 OLS regression estimations, since 6 financial development variables will be used. After that, OLS regressions will be checked for autocorrelation, heteroskedasticity and functional misspecification Control variables Despite various methodologies used in researches previously described, nearly all of them implemented similar control variables when determining the relationship between financial development variables and remittances. Thus, control variables defined below will be implemented in the model of this thesis too.

32 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT 32 The ratio of FDI to GDP. Used in Cooray (2010), Gupta, Pattillo and Wagh (2007), Shahzad, Adnan (2014). GDP per capita (constant 2010 US$) in terms of log. Used in Cooray (2010), Gupta, Pattillo and Wagh (2007), Shahzad, Adnan, Ali and Raza (2014). Exports of goods and services (% of GDP). Implemented in Cooray (2010), Gupta, Pattillo and Wagh (2007), Shahzad, Adnan (2014) and Masuduzzaman (2014). Inflation as the Consumer Price Index (annual %). Used in Gupta, Pattillo and Wagh(2007), Giuliano and Ruiz-Arranz (2005) Data for the model The relationship between remittances and variables of financial sector development in Lithuania will be tested using two of four dimensions, financial depth and financial efficiency. The ratios of Private sector credit to GDP, total banking assets to GDP, stock market capitalization to GDP will be used as proxies for financial depth and overhead costs, net interest margin and stock market turnover ratio will be used as proxies for financial efficiency. The annual data from 1996 to 2014 for these variables will be taken from Global Financial Development Database of World Bank. Remittances as the % of GDP and control variables will be taken from World Development Indicators dataset of World Bank. Exception is Inflation variable which is taken as consumer price index is taken from World Bank and annual percentage change is calculated by author.

33 THE IMPACT OF REMITTANCES ON FINANCIAL SECTOR DEVELOPMENT Empirical research and results 3.1. Workflow of research Third sector of the thesis will consist of few parts. Firstly, the model of the research will be presented. Then, time series data will be checked for stationarity using Augmented Dickey Fuller test. If needed, first or second differences of variables will be taken to make data stationary. Further, various OLS regression models will be calculated. They will be divided into 3 parts. Firstly, two variables of financial institutions depth will be modeled, secondly two variables of financial institutions efficiency, and lastly, two variables of development of financial markets will be examined. Such breakdown of variables will be applied because it will give clarity on research, since some financial development variables lack data points. In conclusion, all models will be checked for autocorrelation, heteroskedasticity and functional misspecification. Regression model used in this thesis: 3.2. Regression model FD i = β 1 Remittances + β 2 FDI + β 3 GDPperCapita + β 4 Exports + β 4 Inflation + ε Where all the variables are explained in the Table below. Dependent variables were described in previous sector. For the estimation of FD 1 and FD 2 variables annual data from 1996 to 2014 will be used. However, the year of 2009 is excluded, since this data point is missing in World Bank database. Also, it was not found in any other dataset available online (at least in free databases). For the estimation of FD 3 and FD 4 variables annual data from 1996 to Finally, for FD 5 and FD 6 variables, which shows depth and efficiency of financial markets, OLS regression model will be implemented using annual data from 1996 to Further datapoints