• No results found

Remittances and the level of small and madium sized enterprise start-ups

N/A
N/A
Protected

Academic year: 2021

Share "Remittances and the level of small and madium sized enterprise start-ups"

Copied!
43
0
0

Loading.... (view fulltext now)

Full text

(1)

J

Ö N K Ö P I N G

I

N T E R N A T I O N A L

B

U S I N E S S

S

C H O O L JÖNKÖPING UNIVERSITY

R e m i t ta n c e s a n d t h e l e v e l o f

S m a l l a n d M e d i u m s i z e d E n t e r

-p r i s e s ta r t - u -ps

Master thesis in Economics

Author: Elin Glommen Andersson (860405) Tutor: Johan Eklund

Carl-Henrik Olaison Jacobsson Jönköping June 2009

(2)

i

Acknowledgements

I would like to thank James Dzanis at Jönköping business school for providing me with helpful literature and data.

Finally would I like to thank Johan Eklund and Carl-Henrik Olaison Jacobsson at Jönköping business school who have guided and advised me through this D-thesis.

May 2009, Jönköping, Sweden Elin Glommen Andersson

(3)

ii

Master thesis in Economics

Title: Remittances and the level of Small and Medium sized enter-prises start-ups

Authors: Elin Glommen Andersson

Tutor: Johan Eklund

Carl-Henrik Olaison Jacobsson

Date: June 2009

Subject Remittances, Small and Medium sized enterprises and Business-es environment

Keywords Remittances, Firm size, Capital and Property Rights

JELClassifications: D69, O11, L25,D24, D23

Abstract

This thesis within economics is examining the impact that remittances could have on the level of new small and medium sized enterprise start-ups. Remittances could be seen as a capital flow and would therefore increase the level of new SME start-ups but is this really the case?

A model is developed with a panel data set over 45 countries all across the globe over a two year period. Six businesses environment variables are included in the regressions to see how the businesses environment affects the level of new SME start-ups. This model is also used when testing if the relationship between remittances and the level of new SMEs are stronger in the middle income countries than in the lower income countries. The descriptive statistics shows that both remittances and the number of new SME`s have increased from 2003 to 2005. The level of new SME`s have increased with a larger percentage share in the middle income countries relative to low income countries. The results from this thesis are somewhat difficult to interpret. Although there seems to be the case that remittances are not affecting the level of new SME start-ups when in-cluding all the countries in the same regression. As the countries are divided into two groups one can see a stronger relationship between remittances and the level of new SME started in the low income countries than in the middle income countries. One can also see that credit right and the cost of starting a new business is strongly related to the level of new SME.

(4)

iii

Table of Contents

1 Introduction ... 1

1.1 Purpose ... 3

1.2 Outline ... 3

2

Previous research & Theory ... 4

2.1 Remittances and Investments ... 4

2.2 Capital Limitations and Micro firms start-ups ... 6

2.3 Business environment ... 8

3 Empirical Framework ... 12

3.1 Method ... 12

3.2 Hypothesis ... 13

3.3 Data and Limitations ... 13

4 Empirical results ... 15

4.1 Descriptive Statistics ... 15 4.2 Regression results ... 16 4.2.1 Model 1 ... 16 4.2.2 Model 2 ... 17 4.2.3 Correlation ... 18 4.2.4 Model 3 ... 20 4.2.5 Model 4 ... 20 4.2.6 Remaining results ... 21

4.2.7 Low and middle income countries ... 21

5

Conclusion ... 24

6 References ... 25

Appendix 1 – Countries GDP per capita in 2003 ... 28

Appendix 2 - Variable Table ... 29

Appendix 3 – Regression results when normalizing GDP...31

Appendix 4 - Remittances ... 32

Appendix 5 - New businesses started ... 34

(5)

iv

Figures

Figure 1: Remittances in US$ per capita...15

Figure 2: New SME started per 1000 inhabitants...15

Figure 3: Remittances in US$ per capita in low and middle income countries...16

Figure 4: SME`s start-up per 1000 inhabitants low and middle income countries……16

Tables Table 1: Hypotheses on variable fluctuation...11

Table 2: Regression results 1...18

Table 3: Correlation matrix...19

Table 4: Regression results 2...21

Table 3: Regression results for low and middle income countries...23

Equations Equation 1: Remittances per capita...12

Equation 2: New businesses started per 1000 inhabitants...12

Regression model General Regression model...12

Regression model 1...16

Regression model 2 ...17

Regression model 3...20

Regression model 4...20

(6)

1

1

Introduction

Remittances have been a part of the economic development ever since individuals started to emigrate; many of the developed countries have during certain time-periods of their history, been dependent on remittances to be able to grow. Large parts of the European countries were dependent on their remittances sent from their emigrants in the 19th and 20th century to be able to reach the level of economic growth they have today (Maimbo and Ratha, 2005). As the developed part of the world increases their gross domestic product (GDP), more individuals from the developing parts of the world has taken the decision to immigrate to the western world. As the emigration has increased so has the remittances level, the remittances level of today is the highest in history and remittance will continue to grow with exceptions for the times during crises and depres-sions. Remittances at a macro level have shown to have an important effect on improv-ing development, remain economic stability, palliatimprov-ing economic shocks and reduce poverty (Maimbo and Ratha, 2005).

Remittances can have different definitions but the one that this paper will be referring to is the one stated by the international monetary fund (IMF).

“When migrants send home parts of their earnings in the form of either cash or

goods to support their families, these transfers are known as workers’ or migrant remittances” (Ratha, 2005).

The World Bank has estimated the growth of remittances to around 30 percent yearly and the total level of remittances for 2006 was estimated to 250 billion US$ (Ratha and Shaw, 2005). This is twice the level of foreign aid and the second largest financial in-flow to many developing countries, remittances can perhaps be seen as the largest source of external financing in developing countries. The World Bank in 2001 estimated that 17 percent of Haiti‟s GDP was represented by remittances and for Somalia 40 per-cent of GDP was represented by remittances during the 1990s. When analysing the data concerning remittances it is clear that some countries are dependent on remittances for their economic survival.

The money transferred as remittances is sometimes referred to as migradollars. These migradollars can have great impact on the receiving countries or communities for eco-nomic development if they are invested into productive projects. In the study of Mexico urban areas, it was found that 27 percent of the small firms are relying on remittances from abroad (Woodruff and Zenteno, 2001).

The importance of SME„s (Small and Mediums size enterprises) on the economic de-velopment is different in different countries. The definitions and measurement of SME varies between countries but the definition that will be used in this thesis is dependent of numbers of employers. A firm that has less than 250 employers will be considered as a SME (Ayyagari et al, 2003). In Ukraine, Belarus and Azerbaijan only 5.5 percent of the workforce is employed in SME„s but in Chile, Greece and Thailand more that 80 percent are employed in SME`s. The level of employees working in small and medium sized companies tends to increase as GDP increases (Ayyagari et al, 2007).

Remittances tend to be a much more stable source of capital than private flows. Remit-tances is largest in low-income countries if seen as a share of total international flows

(7)

2

but the largest part of remittances counted in US$ are sent to middle income countries (Ratha, 2005).

The largest part of the remittances that is transferred to the developing countries is transferred through informal sectors. An informal transfer is a transfer that does not in-clude any documentation and will because of this not show in the national account. It has been estimated that remittances through the informal sector can range from 35 per-cent to 75 perper-cent of the recorded flow through the formal sector. (Freund and Spata-fora, 2005). In 2004 the World Bank estimated that the remittances that is transferred through institutions between countries and communities are over 100 billion US$ a year.

In the Dominican Republic, around 25 percent of the households are receiving remit-tances. From the 25 percent of household receiving remittances 29 percent is stating that they are receiving the significant share of income from remittances. Only 16 percent of the individuals declare to use the receiving remittances in capital accumulations projects in the Dominican Republic (Amuedo-Dorantes and Pozo, 2006). This shows that a sig-nificant level of households in the Dominican Republic is dependent on remittances. The question of whether or not remittances is reducing poverty and increasing economic development for the receiver countries has been answered in many articles, the largest parts of the researches have come to the same conclusion that remittances will increase economic development and decrease economic inequality and poverty. See Cox and Ureta, 2003, Lòpez-Còrdova et al, 2004, Stark et al, 1988. Although a few authors found that remittances did not affect growth and poverty in a positive direction and they actually stated that remittances could have a negative effect for the countries See Chami et al, 2005. This statement will be discussed further in the theory and previous research section. How remittances is effecting the economy is important for many reasons; one is the huge amount of remittances that is sent today and how important this money transfer seems to be for households around the world for survival and for the economy as a whole as it could decrease poverty.

Another reason is that we today are facing a financial crisis that is spreading around the world, this is affecting more economies than any crisis has ever done before. A crisis normally leads to a decrease in the remittances transfer but it also leads to a decrease in FDI and foreign aid to many developing countries, this decrease in FDI and foreign aid will make the remittances even more important as an income source. The poor part of the world will probably suffer harder and more deeply than the west world without any own government that can support and help their own citizens, they are standing before a deep economical depression. (Lin, 2008)

Remittance obviously plays a large role in the development for many countries but is it helpful and useful as the individuals in the developing countries take the decision to start a new business? Many studies, for example Chami, 2005, Martin, 1991 and Diaz-Briquets, 1991 find that in the poorest households the remittances are used for daily ba-sic consumption goods as food, health care and education. Although in the richer households, the remittance may be used for investment in small businesses, as collateral or to pay for external debt services. Remittances can be used either for consumption or for investments. If used for investments remittances can be expected to have a positive effect on firm start-ups.

(8)

3

One of the most important variables when starting a firm is the access to capital; in de-veloping countries with low and middle high incomes, this is a huge problem to the en-trepreneur. Many of the developing countries are based on a system where a loan from a bank is almost impossible to receive and if the entrepreneur does not have any or a small amount of already existing capital this difficulties with access to capital can stop the development of new SME`s. As individuals from developing countries immigrates and due to altruism or other reasons start to send remittances to their home countries the access to capital is increasing and the ability to start a SME would increase.

This paper will have its focus on how the level of remittances is affecting the level of SME start-ups. The level of SME`s start-ups is strongly dependent on capital limitations which are dependent on the business environment, this relationship will be further dis-cussed in the next section.

1.1 Purpose

The aim of this thesis is to see if remittances are used for investment into small and me-dium sized enterprises start-ups. This thesis will test which effect, if any, remittance will have on the level of new SME`s registered, it will also show if this relationship is stronger in middle income countries in comparison with low income countries.

1.2 Outline

The rest of this paper is organized as follows. Section 2 presents previous research and gives a theoretical base to the paper on how and why remittances should affect the numbers of small and medium sized enterprises. Section 3 consists of the method used for the empirical research in the paper and states the hypothesis. Section 4 is used to analyze the empirical results that are received through descriptive statistics and regres-sions. The last, section 5, is the conclusion of the thesis, which is followed by a list of references.

(9)

4

2

Previous research & Theory

Remittances have shown to affect a country’s economy in many different areas such as growth, GDP per capita, social wealth and education but does it affect investments into small and medium sized enterprises start-ups? How is the level of SME start-ups af-fected by the businesses environment such as access to capital? These questions will be theoretically answered in the following section.

2.1 Remittances and Investments

Glytsos discussed in 2003 in his article, what aggregate income effect remittances have, he found in his studies that remittances have large effects on growth, standard of living and capital formations. The negative effect on aggregate income should be the import generating effect but the positive effects outweigh this. He also found when studying Greece that the majority of remittances are used for final consumption of goods (Glytsos, 1993). Taylor and Wyatt claims in their article that remittances will increase investments in new agricultural techniques, future immigrations and education. Remit-tance is also found to decrease poverty (Taylor and Wyatt, 1996).

As immigrants face the decision to transfer part of their wages to their home country this is done due to several reasons but they can be generally divided into three groups. The first one is that the immigrant simply cares about the family and friends that is left behind in their home village, and wants to create a better future for them. The second one is that the immigrant sees an advantage of creating investment that they can use as or if they return to their home country, the last one is due to family and social pressure from the individuals and insurances when or if the immigrant will return to their home country. The two later cases is dependent on if or when the immigrant will return to his or her home community, if the incentives is to stay in the new country or community for the rest of their lives the incentives to transfer remittances to their home countries will decrease (Dustmann and Mestres, 2007).

In a model developed by Chami et al (2005) remittances are assumed to be based on al-truism which is defined as; when the immigrant leaves the country and takes the deci-sion to send home remittances to their family and friends because of his consigns or ob-ligations. As the immigrant receives his wage and deduct his cost of living in the new country, he will decide how much he can remit.

The emigrant‟s purpose of the money that he is transferring, to his family and friends, is to increase their income and for the receiver to be able to use a part or the whole amount to consumption of daily goods if necessary. The part of the remittances that might be left over after the most necessary consumption has been done should be used for in-vestments for the future as for example to education, savings or businesses start-ups, whilst the receiver in many cases uses all the money for daily consumption or as labour substitution. Labour substitutions arise when an individual think that the labour work they perform is low paid, the incentive to work is then week and as they receive a rela-tive large amount of remittances they will reduce their labour hours. The individual will in this case cut down or stop working and instead use the remittances to a larger extend for daily consumption which will lead to that remittance is used as a substitute to wage. If this situation occurs a moral hazard problem has arisen, it is difficult for the emi-grated individual to be able to control how the remittances are used. In this situation it is

(10)

5

calmed by Chami that the receiver will decrease their workforce as remittances is in-creasing and the total income of the individual will decrease, the welfare would not be maximized and remittances cannot be seen as a capital flow (Chami et al, 2005).

Instead of keeping the labour income constant and use part of the remittance received for investments, parts of the remittances could be used to decrease the workforce. This will in the long run decrease the economic development and could in the short run de-crease the amount of SME start ups.

In the article; remittances precepts and business ownership in the Dominican Republic, the author‟s researching is if remittances are increasing business ownership or not. The outcome shows that remittances does not increase ownership of business but if the fam-ily or friends in the home country own businesses this will increase the remittances send from the emigrants (Amuedo-Dorantes and Pozo, 2006).

Beck T in 2005 explores the relationship between economic growth, poverty and the size of the SME sectors. This article shows results of a positive relationship between the size of the SME sectors and a decrease in poverty as well as an increase in economic growth (Beck T et al, 2005).

Households that are in a bad financial situation normally tend to plan their consumption on the day that they receive their income due to uncertainty; this is the situation for many households in the developing countries. In each period, the individuals have to decide if they should consume their whole income or save part of it for investments. If they decide to consume they will not achieve the benefit that investment yields as a pos-sibility to avoid the decrease in consumption that arise when the economy face a setback and/or increased consumption in the future (Adams, 1998).

The households in many developing countries cannot rely on the financial system when borrowing money due to corruption and bad financial systems so the access to capital is limited. The problems mentioned above can be solved with remittances from family or friends that are living abroad or in different areas of their own country. The access to remittances can change the household‟s decisions to invest more and thereby achieve asset accumulation. For many households the rate of time is considered more important that the rate for interests meaning that the individual thinks that he benefits more from consumption today than to invest the money to earn interests and consume in the future. This leads to many individuals‟ unwillingness to save or invest (Adams, 1998).

Adams found in 2005 that households that receive international remittances spend 2.2 percent more on household expenditures than the households that do not receive any remittances. This increase in household expenditure is mostly spent on consumption from a world standpoint. Although, from the individual immigrant‟s standpoint, this in-crease in household expenditure, can be seen as an investment where the individual can be expecting some future financial return. The increased remittances are important for the economy as a whole affecting the wages, employment and business opportunities. Adams also states that remittances have a positive effect on the area of the recipient in terms of health care, education and poverty. To sum up his finding; small proportions of the remittances goes to savings and investments in houses, land and/or buildings (Adams, 2005).

If the receiver does not use their remittances for labour substitution or consumption, part of the received remittances could be used for investments. Part of these investments

(11)

6

can be used for investments into new SME`s, these kinds of investments are likely to increase the individual‟s long-term income resulting in that the whole economy would gain in reduction of poverty and increased economic development.

2.2 Capital Limitations and Micro firms start-ups

In order to start a firm any individual will have to have access to capital, in many devel-oping countries this is a larger issue as the business environment, government and cor-ruptions can aggravate the accessibility to credit.

Remittances is argued not to be used as productive capital accumulation, economic freedom and economic development. It is instead argued to promote dependency of the receiver nations and only to be used to increase daily consumption (Martin, 1991 and Diaz-Briquets, 1991).

Self-employment can be a way out of the poor life for many individuals in low and middle-income countries as it has been recognized as an effective tool to increase em-ployment, economic development, income and decreasing poverty, but many individu-als from the developmental part of the world have no or very little access to capital (Azad, 2005).

The model developed by Evans and Jovanovic, 1989 is raising the issue of individuals‟ access to capital and entrepreneurial ability. The individual have to have a higher bene-fit or wage if they are self-employed than they would if they received the labour worker wage pay off. As the individual decide to be self-employed, he or she has to collect capital for the firm‟s start up costs. In the early stage of the enterprise it will suffer, from capital constrains since many institutions are not willing to lend to the firms in need of capital. Self-employments will decrease among those entrepreneurs that have higher constraints. To collect the capital, personal savings and loans from family and friends are necessary for the entrepreneur, since this is the largest source of capital in-vested in new SME`s (Evans and Jovanovich, 1989).

A micro firm is usually self-financed by the entrepreneur. In the part of the world where self financing is impossible for the majority of the population, money lenders are re-strictive with their loans and charges unreasonably high interest rates in fear of losing money (Azad, 2005).

The entrepreneurs in SME`s are often limited by capital. The cash flow that a firm can generate in a specific period is partly dependent on the level of own invested capital. As a capital constrained firm is entering the market its level of profit will be dependent on the level of own capital invested. If the firm can invest a higher level of own capital into the firm other costs can be excluded such as relatively high interest rates and adminis-trative cost. This cost could decrease the profit for the firm and in the long run force the firm out of business. Some of the entrepreneurs that are willing to be self-employed are not able to collect the own capital needed to start up a firm and are therefore forced to be employed instead. The capital needed to decrease this limitation is partly dependent on personal savings and loans from family and friends and this capital is to a large ex-tent received from remittances that is received through an earlier working period abroad or family members that have immigrated. (Evans and Jovanovich, 1989).

The individual that is remitting money tends to have problems controlling how the money is used by the receiver. If the money is used in a process of starting up a SME

(12)

7

this can be a way for the individual that is transferring money to have a larger control over how the money is invested. It can also be a more desirable way for the entrepre-neurs in the developing countries to receive money through their family and friends to be able to overcome the capital constrains that through loans from the formal sector which tend not to be properly processed.

The article; Remittances and microenterprises in Mexico written by Woodruff C and Zenteno R, is discussing which effect capital constrains have on microenterprises where one of the largest difficulties is to separate supply of credit from demand of credit. They claim that remittances have a significant effect on level of invested capital due to larger access to capital for Mexico microenterprises. Approximated one third of the invested capital in micro enterprises is connected to the immigration to US and remittances lead-ing to that. Remittances increase the decision to start an enterprise(Woodruff C and Zenteno R, 2001).

The article written by Cull and Xu in 2005 focuses on small and medium sized firms and their access to capital. The authors find that one cannot reject that micro firms do not lead to an increase in GDP and reduction in poverty (Beck and Demirguc-Kunt, 2006). A small firm is more sensitive to a poor financial system, bad property rights and access to capital compared to larger firms. A small firm is more likely to reinvest their profits if the above variables are correctly operated (Cull and Xu, 2005).

Micro firms around the world in low and middle income countries are to a large extend small firms owned by families and the firms are usually the only source of income for the family. Micro firms are a break up from the informal sector with labour intensive work, small-scale production, self-employment, lack of access to markets and lack of credits due to high cost of borrowing and almost no supply of credit for the individuals. Banks are denying small entrepreneurs loans because of the fact that they are unable to collect collateral and are facing high risks, other limitations can also be for example the lacking order in documents and account records in many SME`s, (Azad, 2005).

Mobilization of saving could solve part of the problem of access capital. The savings from many households that receive remittances are never transferred into the formal sector, where these savings can be used for larger investments. Mobilize savings is a hard and costly process where two larger obstacles have to be overcome; the first one is asymmetric information where the individuals can feel insecure, they feel that they do not have total control over their savings. The second obstacle is the transaction costs as-sociated with collecting savings. However, if these obstacles are to be overcome the economic development can increase and exploit economies of scale and thereby increas-ing the capital access. Poolincreas-ing of savincreas-ings can also lead to technology improvements and innovations (Levine, 2004).

A lot of the remittances can never be used for investments since the level of capital for the individual is too small to invest, a reform of mobilizing savings of the remittances could have a great effect of the level of new SME`s since it pools the savings, but to be able to enforce this system a country has to have a good operating formal sector. An in-dividual that has incentives to start up a firm but has too little own capital and is not able to borrow from family and friends will, if the formal sector is working properly, benefit from the mobilization of capital.

(13)

8

According to Levine 2004 a financial system should include five functions; allocate capital and produce information, monitor investments, facilitate management risk, mo-bilize savings and ease of exchange of goods and services. How well the different coun-tries financial system is providing these revises to their citizens is varying, but one thing is clear that if a country is suffering from a bad financial system the individuals that want to save or invest will suffer. Poor financial systems results in credit limitations es-pecially for poor people that are unable to collect collateral to be able to loan from banks this can keep the poor individuals in many developing countries from engaging in profitable investment opportunities (Levine, 2004).

The level of SME`s start-ups will be strongly related to the entrepreneurs‟ personal wealth, wage rates, cost of capital and entrepreneurial ability. It is also important to consider the economic conditions that are present as the firm starts and during the forth-coming years. The optimal capital stock will also be dependent on which industry the firm is operating in (Woodruff and Zenteno, 2001).

2.3 Business environment

Business environment has a large effect on the level of SME start-ups since it effects the individual‟s decision to start a new firm. The business environment is a measure of the business climate surrounding the firm. It includes different parts that have a major im-pact on SME start-ups. In this thesis business environment will include Business dom, Property rights, Cost of starting a business, Cost of closing a business, Fiscal free-dom, Investment freedom and Access to credits.

When an individual is economical free, he or she has total control over his or her own labour and property. The definition on economic freedom is the following;

“Encompass all liberties and rights of production, distribution, or consumption of goods and services. The highest form of economic freedom should provide an abso-lute right of property ownership; fully realize freedoms of movement for labour, capital and goods; and an absolute absence of coercion or constraint of economic liberty beyond the extent necessary for citizens to protect and maintain liberty it-self” (Miller and Holmes, 2009).

All governments around the world are imposing constrains on economic activities; these constraints are diminishing the production, distribution, consumption and service pro-vided to the market. Price control is a constraint that is often used by the government with clearly shown effects on the equilibriums of supply and demand on the economic market (Miller and Holmes 2009). SME are significant more sensible than larger firms to an unstable and underdeveloped business environment. This leads to less SME`s started in countries with less economic freedom which tends to be the situation in many developing countries (Ayyagari et al, 2007).

Third world countries tend to look at the western countries and see a fast and traced road out of poverty. As some of the developing countries tend to copy larger parts of the western legal system to implement and to try to enforce a more controlled and reliable system they meet new and maybe even larger problems than before often pressured by the developing countries. Some developing countries have implemented reforms that have created markets where it is extremely hard to set up a firm because of high admin-istrative costs, long waiting time to be able to start a business and high legal costs. As

(14)

9

the long and maybe corrupt process of starting a business is finished if it even gets to this point, the firm that has started has had huge costs and a loss of income during the period and is almost doomed to fail. The copying of the western legal system will in most cases not increase access to capital because it does not have the same presump-tions as the developed world had during their development and it could lead to more capital needed to be able to start a new SME`s in the long run leading to a decrease in SME`s start-ups (Soto H, 2000).

Business freedom is a positively related concept to SME`s; when business freedom creases many variables such as access to capital changes in a positive direction to in-crease the level of SME start-ups (Miller and Holmes, 2009).

Many of the individuals that are willing to be self-employed in the developing parts of the world lives in countries where the government is heavily controlling the market and have huge constraints that work as obstacles for the economic development. As indi-viduals receive remittances they will be able to overcome some of these constraints and thereby the level of SME start-ups will increase as long as they keep their labour con-stant (all else equal).

Property rights is one of the components included in the concepts of businesses

envi-ronment and measures to which degree the government is protecting the individual‟s properties. If the property rights for personal wealth, future return and capital are not protected the entrepreneurs will not have the incentives to reinvest their profit into the firm. Due to this the firms, if they are started at all, will not generate profit (Beck at el 2005). The largest motivation for economic activities is the accumulations of personal property. As the motivation to engage in economic activities decreases, the economic development will decrease. To protect the private property an effective and honest legal system, which is treating all individuals the same and are available for all individuals, is needed (Miller and Holms, 2009). If the property rights are high in a country this will lead to an increase in business activities and individuals would save and invest higher level of their received remittances. When the investments are increasing the level of SME`s start- ups might also increase.

Credit market rights is a regulation that is restricting the access to capital for SME`s and

are thereby restricting the firms to enter new market leading to a restriction of economic freedom. This variable reflects domestic credit markets including five components; ownership of banks, competition, extension of credits, avoidances of regulations that lead to a negative real interest rate and interest rate controls. The two first components measure to which extend the banks are private owned and whether or not foreign banks are permitted to compete. The later three components is measuring to which extend credit is supplied to the privet persons and the controlled of interests rates (Gwartney, J and Lawson, R. A., 2005). As the credit market rights increases the easier it will be for the individuals to access capital which is one of the most important variables when an individual takes the decision to invest in a new business. If the access to credit increases so might the level of new SME`s do.

The cost of starting up a new business is measured relatively to income. They include

the legal fees, cost of enforcing and developing contracts that arise when starting and register a company. A higher entry cost might result in less firm start-ups and less firms to enter the sector or to be able to enter new markets. Lower entry costs on the other hand could work as an incentive to increase the firm start-ups (Beck et al, 2005).

(15)

10

The cost of closing a business. This can be seen as an exit cost and measures all the

costs that are included when a firm must exit the market. Low bankruptcy costs indicate a faster and less expensive process when a company has to leave the market, which leads to the most efficient outcome (Beck et al, 2005). As cost of starting up a new business and the cost of closing a business are increasing more capital is needed to in-vest in an own business. If remittances is received and used for inin-vestments in small and medium size enterprises, individuals could overcome the capital limitations and might because of the remittances start up a new business.

Fiscal freedom is referring to the individual‟s freedom to run their business and control

their income and wealth, as governments are imposing fiscal burdens through taxes that individuals cannot avoid the governments are generating profit to themselves and this decreases the revenue for the individuals. The marginal tax rate an individual has to pay to the government is the government‟s cut of the profit from the individual‟s next unit of work, whatever remains after this cut is the actual reward for the individual‟s effort. As the taxations are increasing the revenues for the individual are decreasing and the in-centive for the individual to start a business decreases (Miller and Holmes, 2009). Taxa-tion levels on small and medium sized businesses in many developing countries are high relatively to their income level. Even if an individual is receiving remittances his or her incentives to invest the capital in a micro firm are low due to the low profit that will re-main in relation to the work effort as the government is imposing high taxes.

Investment freedom is a measurement of how the government controls the individual‟s

investment. Capital should be used for the best investments where it can generate the largest return. As the government is imposing regulations for both internal and external capital in and out flow to the countries this freedom is decreased and access to capital is limited for both the demanders and suppliers of capital. (Miller and Holms, 2009) If the economic freedom is increasing it will lead to more individuals investing in SME`s be-cause they would need less capital and have higher incentives to engage in economic ac-tivates. This would lead to an increase in the economic development and decrease in poverty in the specific country as a whole. It also benefits the individuals that can invest their money in the investments opportunities that they found most profitable.

If the businesses environment is changed in the preferred direction this can lead to an increase in the level of new SME`s, individuals might invest their money if they feel se-cure that the money invested can generate profit in the future and even more important that government will not control or steel the investment or the profit.

(16)

11

From the theories in the theoretical sections one can expect the variables to affect the level of new SME`s in the way stated in the table 1 below;

Table 1: Hypotheses on variable fluctuation

Variable Abbreviation Expected

signs The number of NBS in 1000

inhabitants

(Dependent variable) NBS

Remittances in US$ per capita

REM +

Cost of Starting a New Business in % of Income

STRT -

Cost of Exciting a Business in % Estate

EXIT -

Property Right PROP +

Credit Index CRED +

Freedom of Investment INFE +

Fiscal Freedom FIFE +

As will been shown later in the empirical framework, remittances will show to have the opposite effect than was predicted from the main theories.

(17)

12

3

Empirical Framework

This section will explain the method and hypotheses that will be used to achieve regres-sion results and descriptive statistics. It will also explain which data will be used and what limitations that have aroused due to data constraints.

3.1 Method

Remittances transactions are growing and are thereby having a larger impact on the de-veloping countries‟ economies. In this thesis 45 countries are included, the countries are chosen from two different income groups; low and middle income countries. These countries are chosen due to availability of data and are located all around the world and are placed into groups based on the level of income per capita in year 2003. The coun-tries with an income per capital over US$ 1846 are considered a middle income country while the countries with a lower income per capita are considered to be a low income country; the countries are listed in appendix 1. The World Bank have ranked the coun-tries somewhat different from the division that is used in the analysis although to get a equal amount of countries in both groups was US$ 1846 used as the number where the countries were divided. The following formulas was used to calculate average remit-tances per capita for all countries and the average level of new SME`s started per capita for all countries to perform descriptive statistics. These formulas have also been used to be able to compare the two income groups separately.

Remittance per capita =Total remittanceses in us$ / Total populations (1)

New businesses started per 1000 inhabitants = Total NBS/ Total population in 1000 (2)

As one might notice the term NBS (new businesses started) is now used instead of SME this might lead to some confusion but the two terms is measuring businesses start ups in the same way. This new term is used since it is used in the data set where the data over businesses start ups are collected.

To test which and how strong relationships remittances and new SME`s start-ups have a panel data set will be used including 45 countries over a 2 year period, a total of 90 ob-servations. A panel data set is used to be able to record the changes in SME start –up level that have occurred as remittances are increasing or decreasing.

At first an ordinary least square panel regression will be used for all countries after this a regression with a fixed effect towards time will be used. The regression that could be used for fixed effect towards countries is not included since the large number of dum-mies will give biased results. The number of remittances and the level of NBS have been normalized with respect to 1000 inhabitants. There will also be regression models based on the correlations between the variables; these models will be excluding the variables that are correlated with each other which are found by using correlation tables. After this the same regressions as above will be used to be able to analyse the two dif-ferent income groups to see if there is a stronger relationship between remittances and the level new SME start-ups in the middle income countries than in the low income countries. The following regression formula will be used;

(18)

13

New business started= β1+β2remittancest+ β3cost of starting businesst+β4costof

clos-ing a business+ β5property right indext+β6credit market indext+ β7investment freedom

indext+β8fisical freedom indext+ut

3.2 Hypothesis

The aim of this thesis is to see if remittances are used for investment in small and me-dium sized enterprises start-ups. It will also test which effect remittances will have on SME start-ups in low and middle income countries around the world.

 Is the number of SME start-ups increasing as remittances are increasing?

 Is the relationship above stronger in the countries with middle income than the countries with low income?

3.3 Data and Limitations

The data is collected from some of the largest organisations in the world and in appen-dix 2 one can see the variable table that is describing the specific data variable and also which source it is collected from and how many years that is available. In appendix 4, 5 and 6 the specific number for each variable for each country can be found.

Remittances data is collected from The World Bank‟s data statistics. One has to take into consideration when analysing the data of remittances that this variable is difficult to estimate and measure in a more correct way due to the fact that much of the remittances are sent to the developing countries through informal sectors and is never recorded in any accounting report. The level of new SME started data is also difficult to measure because many of the firms started is not registered due to high costs of registration and is therefore not included in any reports. These two measurement constraints can lead to biased results in the regressions.

One limitation in this thesis is the amount of years that is available in our panel data set. The data set for new SME started is a fairly new source and it is difficult to collect data from years prior to year 2003 for a specific country. A data set that included all the 45 countries over both years and measured in the same units was not available. The most ranking data over some of the businesses environment indicators is in many reports only presented for specific years (not every year). The data for informal sector variable was not found and was therefore excluded from the analysis although it would have been very interesting to include this variable in this thesis.

Remittances and new SME start-ups are measured in terms of US$, costs of starting a new business is measured in terms of percent of income, the cost of closing a business is measured in percent terms of estate. The credit market regulation is measured in an in-dex whit a scale from 1 to 10 where 10 would indicate total freedom. The property rights, fiscal freedom and investment freedom are measured in an index of economical freedom with a scale from 0 to 100 where 100 is total economic freedom. One problem that arose was that the changes in the indexes for each country over the two years were very small which creates problems when analysing the results from the panelled regres-sion.

A panel data set with only two years time period can result in difficulties with the pro-cedure of the regressions. The OLS model has high restrictions on the model and can

(19)

14

therefore be showing biased results between the independent and dependent variable. In the fixed model the model is less restricted but this model also has problems for exam-ple a problem with a low degree of freedom because of the introduction of many dummy variables and too many variables can also cause multicollinearity. (Baltagi, 2008)

Even though the panel model can include many problems in interpretations it is a good model that has many advantages. It can take into account the individual heterogeneity that is often present in a panel data set. It can also give more informative data, less co linearity among variables, more variability and more efficiency. The panel data set is also better when one wants to study the dynamics of changes (Gujarati, 2003). The use of panel data set when only including two time periods can be questioning but to be able to study the changes over time and since data is only availably for 2 time periods it seems to be the most useful regression method to use for this thesis.

(20)

15

4

Empirical results

This section will include two parts, one that is reporting the descriptive statistics and the second part will include and explain the results from the regressions. It will test for the possible relationship between remittances and the level of new SME.

4.1 Descriptive Statistics

This section will graphically show how remittances and SME`s start-ups have increased during the period of 2003 to 2005 although it will not say anything about the relation-ship between the variables of remittances and SME`s start-ups. By analysing the num-bers of average remittances per capita in US$ in figure 1 one can see that these numnum-bers have increased between 2003 and 2005 from US$ 112 to US$ 149 this is an increase by 37 US$. Figure 2 measures the numbers of SME`s start-ups per 1000 inhabitants this model shows an increase in SME`s start-ups per 1000 inhabitants from 0.85 in 2003 to 1.07 in 2005. The figures show an average increase by 33 percent in remittances per capita and a 26 percent increase in SME`s start-ups per 1000 inhabitants during the years of 2003 to 2005. One can see from the number in figure 1 and 2 that there is a lar-ger percentage increase in remittances per capita than in SME start-ups.

Figure 1: Remittances in US$ per capita Figure 2: SME start-ups per 1000 inhabitants.

The numbers in the diagram can show a relationship that as remittances increases so does the level of new SME but it could also just be a random coincident.

From figure 3 and 4 one can see an increase in both remittances and the level of new SME start-ups from the years of 2003 to 2005 in both low and middle income countries. The remittances have increased by around 60 percent in the low income countries and almost 150 percent in middle income countries this is an increase that is expected since the total level of remittances in the world is drastically increasing. One can also expect to see a larger increase of remittances in middle income countries since the remittances is measured in US$ per capita. Another aspect is that the purchasing power for each American dollar is less in the middle income countries relatively to low income coun-tries. One also see a larger increase of SME start-ups in the middle income countries this could be due to the theory that low income countries would be more dependent of their remittances for daily consumption than middle income countries.

(21)

16

Figure 3: Remittances in US$ per capita in low Figure 4: SME`s start-ups per 1000 inhabitants in

and middle income countries low and middle income countries

(In figure 3 and 4 the blue bar is representing low income countries and the red bar represents the middle income countries.)

4.2 Regression results

All the regressions in this section search for the relationship between remittances and the numbers of NBS. All the values for remittances and the level of NBS are normalized with respect to populations in terms 1000. The relationship between remittances and the level of NBS is due to the theory believed to be heavily influences of the businesses en-vironment that is present in a specific country. The business enen-vironment variables will be included in the regressions in different combinations in order to avoid invalid coeffi-cient estimates.

4.2.1 Model 1

The first model used only tests for the relationship between remittances and the level of NBS registered, this model excludes all business environment variables. The regression result from an ordinary least square panel regression which is assuming the intercept and the slope to be the same for the different countries but the error term to capture the differences over time and individual firms. The following model is used;

NBSt = β1+β2REMt + ut

From the results in model 1, which can be found in table 2, we cannot reject the null hypotheses due to a low F-value this implies that there is no change in the numbers of NBS registered as the level of remittances is increasing. The R-square adjusted is nega-tive. This table is showing how the relationships between the NBS and remittances is changing as one extra control variable is included. After performing the original model 1 a step wise introduction of each control variables will be preformed to search for mul-ticollinearity

(22)

17

4.2.2 Model 2

The second model that is analyzed is the regression that included all the control va-riables.

NBSt = β1+β2REMt+ β3STRTt+β4EXITt+ β5PROPt+β6CREDt+ β7INFEt+β8FIFEt+ut From the results in model 2.1 one can see a negative relation between cost of starting a new businesses and the level of NBS started with a significant value. This result shows that an increase in costs of closing a business with US$ 1 would decrease the level of NBS with 0.0006 enterprises per 1000 inhabitants. One can also see that if the credit right index would increase by 1 unit the level of NBS would increase by 0.28 units. These increases sound large but it is not likely to happen since it is difficult for the credit right index to increase a whole unit over a short period of time. An increase in this magnitude would need many years to develop or a drastic change in the government to be complete. When comparing the F-value from the model with the F-critical value one can reject the null hypothesise stating that there is a change in the level of NBS as we increase remittances.

The significant values of the model are in line with the theories started in the business environment section. A higher entry cost might result in less NBS since the entrepre-neur need more capital to be able to start the businesses. (Beck el at, 2005). As the credit right index is increasing the regulations of the credit market is relaxed which would make it easier for the entrepreneurs to access capital and thereby the investment in NBS would increase (Gwartney, J and Lawson, R. A., 2005).

The second version of model 2 will be an OLS panel regression with fixed effect to time, although the two versions of model 2 will be based on the same regression model. Model 2.2 is tested by using an OLS with fixed effect of time where the slope is con-stant but the intercept is varying over time. This model is used to see if a time effect oc-curs as for example technology changes during the time period. The null hypothesis for this model can be rejected based on the F-value but since the remittances coefficient is still not significant this would again show that remittances have no effect on the level of new SME started, the significant variables are starting a new business and credit rights. These could be the reasons why the null hypothesis can be rejected although shown that remittances have no effect on the level of new SME started.

One can also see that the relationships between the level of new SME`s started, credit market index and cost of starting a new business is as expected from the method section and significant. The variable is due to this following the theory in the section of busi-ness environment. The variables of starting a new busibusi-ness and credit right have almost the same values as in model 2.1 so the effect of a 1 unit increase would be as stated above.

The three regressions in model 2 are in some way miss specified since the results are in-consistent. This might be solved by excluding variables that are correlated with each other from the same regression. Which variables that are correlated will be answered in the next section.

(23)

18 Table 2: Regression results 1

* t-value is significant at the 0.05 level (2-tailed).

4.2.3 Correlation

To test for correlation between the variables a correlation matrix is used that is pre-sented in table 3, this is done due to the results in the previous used regressions, since the regression results show that the model is violating the classical assumption making the coefficients estimated invalid which can be due to the model being misspecified, or suffering from other difficulties.

One can see correlations between remittances and property rights, fiscal freedom and credit rights, this could be due to problems in measuring the data this is also the case of the correlations between the dependent variable and some of the control variables. There is correlation between the investment freedom and the property right variables at a 5% significant level. This correlation can be due to the variable included in the mea-surement of the two variables. For example investment freedom is measuring the right for an individual to keep their own profit earned from investments. The property right also measures these variable and the two variables have some more common measure-ment variables which would explain the correlation. One can also see a correlation be-tween property rights and cost of starting a business. This correlation is a bit more com-plicated to explain. Although it could arise due to the fact that property rights measure-ment might include the cost of starting a business due to the probability of keeping it.

OLS POP Model 1 OLS population Model 2.1 OLS population Model 2.2 Fixed ef-fect time Intercept (t-value) 0.925 (6.016)* 0.189 0.474 0.816 1.941 0.633 1.32 -1.163 -1.431 -1.447 (-1.681) -2.798 (-1.879) -2.789 (-1.842) REM (t-value) 2.72E-07 (0.443) 3.71E-07 0.586 4.43E-07 0.74 4.09E-07 0.654 -6.67E-07 -0.542 -9.17E-07 (-0.73) -9.73E-07 (-0.776) -9.8E-07 (-0.762) PROP (t-value) 0.018 (2.008)* 0.012 1.333 0.013 1.439 0.004 0.457 0.0001 0.011 0.002 (0.178) 0.002 (0.176) STRT (t-value) -0.007 (-3.362)* -0.007 (-3.213)* -0.007 (-2.819)* -0.007 (-2.774)* -0.006 (-2.624)* -0.0063 (-2.605)* EXIT (t-value) 0.012 0.831 0.012 0.803 0.013 (0.862) 0.012 (0.797) 0.0122 (0.79) CRED (t-value) 0.302 (2.838)* 0.298 (2.789)* 0.285 (2.667)* 0.284 (2.552)* INFE (t-value) 0.001 (0.999) 0.009 (0.929) 0.009 (0.874) FIFE (t-value) 0.019 (1.111) 0.019 (1.09) R-square -0.009 0.023 0.128 0.114 0.245 0.191 0.194 0.181 F-value 0.196 2.049 5.298* 3.641* 4.537* 3.947* 3.571* 3.079* DW-test 1.052 1.104 1.11 0.872 0.945 0.954 0.953 0.246 observa-tions 90 89 89 83 76 76 76 76

(24)

19

Since there is correlation present between some of the variables these variables will be excluded from the regressions in different combinations and the regression will be tested again to see if more reliably results can be presented.

Table 3: Correlation matrix

*. Correlation is significant at the 0.05 level (2-tailed). **. Correlation is significant at the 0.01 level (2-tailed).

Remit-tances GDP Remit-tances Pop New Bus GDP New Bus Pop Cost of Starting Cost of Clos-ing Property Rights Credit rights Invest Free Fiscal Free Remit-tances GDP Person Cor. Sig. (2-talid) N 1.000 90 .675** .000 90 -.146 .171 90 -.018 .863 90 .049 .646 90 -.015 .984 84 -.243* .022 89 .215 .050 83 -.118 .271 89 .275** .009 89 Remit-tances Pop Person Cor. Sig. (2-talid) N .657** .000 90 1.000 90 -.060 .575 90 -.047 .659 90 .049 .645 90 .086 .435 84 -.166 .121 89 .333** .002 83 -.005 .960 89 .303** .004 89 New Bus GDP Person Cor. Sig. (2-talid) N -.146 .171 90 -.060 .575 90 1.000 90 .342** .001 90 -.221* .036 90 .057 .604 84 .068 .527 89 .018 .874 83 -.026 .809 89 .136 .204 89 New Bus Pop Person Cor. Sig. (2-talid) N -.018 .863 90 -.047 .659 90 .342** .001 90 1.000 90 -.375** .000 90 .016 .883 84 -.228* .031 89 .374** .000 83 .219* .039 89 .172 .106 89 Cost of Start-ing Person Cor. Sig. (2-talid) N .049 .646 90 .049 .645 90 -.221* .036 90 -.375** .000 90 1.000 90 .201 .067 84 -.228* .031 89 -.156 .159 83 -.149 .164 89 -.047 .663 89 Cost of Clos-ing Person Cor. Sig. (2-talid) N .015 .646 84 .086 .435 84 .057 .604 84 .016 .883 84 .201 .067 84 1.000 84 -.074 .506 83 .070 .547 77 -.182 .099 83 .061 .583 83 Prop-erty Rights Person Cor. Sig. (2-talid) N -.243* .022 89 -.166 .121 89 .068 .527 89 .204 .055 89 -.228* .031 89 -.074 .506 83 1.000 89 .179 .107 82 .488** .000 89 -.082 .446 89 Credit rights Person Cor. Sig. (2-talid) N .215 .050 83 .333** .002 83 .018 .874 83 .347** .000 83 -.156 .159 83 .070 .547 77 .179 .107 82 1.000 83 .170 .126 82 .125 .262 82 Invest Free-dom Person Cor. Sig. (2-talid) N -.118 .271 89 -.005 .960 89 -.026 .809 89 .219* .039 89 -.149 .164 89 -.182 .099 83 -.448* .000 89 .170 .126 82 1.000 89 .067 .553 89 Fiscal Free-dom Person Cor. Sig. (2-talid) N .275** .009 89 .303** .004 89 .136 .204 89 .172 .106 89 -.047 .663 89 .061 .583 83 -.082 .466 83 .125 .262 82 .067 .533 89 1.000 89

(25)

20

4.2.4 Model 3

In model 3 property rights and credit rights indexes are excluded as explanatory va-riables. This is done to reduce the correlations between these vava-riables. Since one can find correlations between property rights and investment freedom. According to the theory credit right index should also be excluded since this variable already can be in-cluded in investment freedom and fiscal freedom, this is done to reduce that likelihood of having multicollinearity present in the model. The following regression will be used;

NBSt = β1+β2REMt+β3STRTt+β4EXITt+ β5INFEt+β6EFIFEt +ut

Even though we are excluding what seems to be two of the correlated variables it still results in that remittances is not increasing the level of new SME`s started. The only significant value at the 5% significant level is the cost of starting a business which shows the expected sign and similar numbers as before. These results can be seen in ta-ble 2. There might be positive correlations present in the regression.

4.2.5 Model 4

Since model 3 still is suffering from autocorrelation model 4 will exclude the variables that are causing the autocorrelation including property rights and credit market index in-stead of fiscal freedom and investment freedom. It also excludes the costs of starting a new SME and closing it, this is due to that this might be measured already in the index of property rights.

NBSt = β1+β2REMt+β3PROPt+β4CREDt+ ut

Again we receive a significant F-value but only one significant explanatory variable a low r-square adjusted and low values of the Durbin-Watson test. The only significant explanatory coefficient is credit rights that showed the results in model 2.3.

(26)

21 Table 4: Regression results 2

Model 3 Model 4 Intercept (t-value) -1.399 (-1.095) -1.822 (0.018) REM (t-value) 5.7E-10 (0.999) -3.4E-07 (0.772) PROP (t-value) - 0.01 (3.172) STRT (t-value) -0.007 (-3.177)* - EXIT (t-value) 0.0143 (0.996) - CRED (t-value) - 0.327 (0.002)* INFE (t-value) 0.0157 (1.918) - FIFE (t-value) 0.0235 (1.457) - R-square 0.146 0.118 F-value 3.813* 4.615* DW-test 0.241 0.178 Numbers of observations

*t-value is significant at the 0.05 level (2-tailed).

4.2.6 Remaining results

All the above regressions have been used to run other regressions with a different nor-malize for the remittances and new SME`s registered. The values of remittances and new SME`s registered where normalized with respect to GDP. These results are in-cluded in appendix 3 in the thesis since it showed almost the same results as the regres-sion which is normalized with respect to population.

There are four more regression result preformed but all of them are showing similar re-sults as the one above, as remittances is increasing the level of new SME start ups are not affected. One of the regressions that were performed was to exclude the linear rela-tionship with using the following formula;

NBSt = β1+β2REMt+β3REM2+ut

The regression results from this model showed no significant value in the explanatory coefficients. A cross-sectional regression was preformed in order to eliminate the chances of model misspecification in the former regression although the results still showed no change in the level of new SME start ups as remittances is increasing.

4.2.7 Low and middle income countries

The regressions in this section are preformed to test if there is a stronger relationship be-tween remittances and new SME in the middle income countries than in the low income countries. These two variables are including all the six business environment variables

(27)

22

this is done even though there seems to be correlations between the variables. The vari-ables are included since there is theoretical base to include them in the regressions. One could expect that this relationship should be stronger in the middle income countries since they would have a relative higher level of income and the middle income countries also received in total amount more remittances. So there should be more money left over for investments in new SME`s after the daily consumption. The results from the both regressions are presented in table 3.

The results from the first regression based on the low income countries shows a signifi-cant relationship with the level of new SME`s and with a positive sign but the increase is so small that it is marginal. From this model‟s results we might actually state that there is a very small but still significant relationship so when remittances increase so does the level of new SME`s start-ups. The R-square adjusted level is higher in the for-mer regression. When finding this result one could conclude that remittances are in-creasing the level of SME`s. So for a poorer that has an increase in the remittances we will expect to see an increase in SME. This shows that remittances can be used to help a family to overcome poverty and create a better life for themselves and for the future family members. This would in the long term lead to a wealthier society and country; the nation could in a long term drag themselves out of poverty and into a developed country.

The results from the second regression are based on the middle income countries. This regression is not significant and has only one significant value which is cost of starting a new firm SME. With this model it can be stated that as remittances is increasing the level of new SME`s are not increasing or decreasing although this relationship should be present and it might be the case that the data is unable to find if this relationship eventually is existing.

When analysing both data from the two regressions we can see that the result is not in line with the theory. We would expect a poorer country to use there remittance for daily basic consumption while we would expect a middle income country to have more capi-tal over and be able to invest this in SME`s. Since this is not the case we could draw the conclusion that a smaller amount of capital in a low income country can create more opportunities than a small amount of capital in middle income countries. It could also be the case that SME`s in middle income countries are already more frequent and that the start up of SME`s in middle income countries thereby is not increasing as rapidly as in poorer countries but part of the remittances is already invested in the already existing SME.

(28)

23

Table 5: Regression results for low and middle income countries Low

income

Middle income

Rempop^2 Remgdp^2 Remgdp gdp/pop Rempop Gdp/pop NBS (t-value) -1.642 (-1.83) -3.595 (-1.409) 0.181 (0.903) 0.639 (3.083) 0.290 (0.249) -3.26 (-2.307) REM (t-value) 3.5E-06 (3.571)* -5.01E-07 (-0.244) 1.19E-06 (0.896) -1.97E-06 (-0.437) -9.32E-08 (-0.047) -9.14E-07 (-0.77) PROP (t-value) 0.012 (1.608) -0.008 (-0.512) 0.004 (0.515) -0.002 (-0.182) STRT (t-value) 0.001 (0.832) -0.023 (-2.495)* -0.005 (-2.437) -0.003 (-1.255) EXIT (t-value) -0.007 (-0.811) 0.042 (1.373) 0.018 (1.507) 0.012 (0.811) CRED (t-value) 0.064 (0.955) 0.156 (0.811) 0.038 (0.421) 0.167 (1.544) INFE (t-value) -0.003 (-0.585) 0.017 (0.917) -0.007 (-0.894) 0.006 (0.68) FIFE (t-value) 0.013 (1.541) 0.048 (1.521) 0.006 (0.431) 0.03 (1.85) GDP per cap (t-value) 0.0002 (5.110)* -5.08E-05 (-1.07) -0.001 (-1.655) 0.0002 (3.026)* Rem^2 (t-value) -1.64E-12 (-1.3) 1.44E-11 (0.807) R-square 0.3318 0.1611 0.22 0.006 0.029 0.28 F-value 3.696* 1.988 9.344 1.174 1.285 4.644* DW-test 0.097 1.397 1.204 1.074 1.401 1.122

Figure

Table 1: Hypotheses on variable fluctuation
Figure  1:  Remittances  in  US$  per  capita      Figure  2:  SME  start-ups  per  1000  inhabitants
Figure 3: Remittances in US$ per capita in low       Figure 4: SME`s start-ups per 1000 inhabitants in
Table 3: Correlation matrix
+3

References

Related documents

The increasing availability of data and attention to services has increased the understanding of the contribution of services to innovation and productivity in

This is the concluding international report of IPREG (The Innovative Policy Research for Economic Growth) The IPREG, project deals with two main issues: first the estimation of

Generella styrmedel kan ha varit mindre verksamma än man har trott De generella styrmedlen, till skillnad från de specifika styrmedlen, har kommit att användas i större

Närmare 90 procent av de statliga medlen (intäkter och utgifter) för näringslivets klimatomställning går till generella styrmedel, det vill säga styrmedel som påverkar

Den förbättrade tillgängligheten berör framför allt boende i områden med en mycket hög eller hög tillgänglighet till tätorter, men även antalet personer med längre än

På många små orter i gles- och landsbygder, där varken några nya apotek eller försälj- ningsställen för receptfria läkemedel har tillkommit, är nätet av

Det har inte varit möjligt att skapa en tydlig överblick över hur FoI-verksamheten på Energimyndigheten bidrar till målet, det vill säga hur målen påverkar resursprioriteringar

Detta projekt utvecklar policymixen för strategin Smart industri (Näringsdepartementet, 2016a). En av anledningarna till en stark avgränsning är att analysen bygger på djupa