• No results found

Immigrants and new firm formation in the service sector

N/A
N/A
Protected

Academic year: 2021

Share "Immigrants and new firm formation in the service sector"

Copied!
30
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

I m m i gr a nt s a nd

N ew Fi r m F o r m a t i o n

i n t he Se r vi c e Se c t or

Bachelor Thesis within Economics Author: Johanna Kullinger

Tutor: Professor Charlie Karlsson

Ph.D Candidate Mikaela Backman Jönköping June 2009

(2)

i

Bachelor Thesis within Economics

Title: Immigrants and new firm formation in the Service Sector. Author: Johanna Kullinger

Tutors: Charlie Karlsson, Professor

Mikaela Backman, Ph.D Candidate

Date: 2009-06-08

Keywords: Immigrants, new firm formation, Sweden, ordinary service sector, advanced service sector

JEL Classifications: J610, M130, L800

Abstract

This thesis is written with the purpose to analyze immigrants’ effect on new firm forma-tion. Immigrants can ease future sustenance problems of decreasing population growth and growing life expectancy. Immigrants start new firms to a higher extent than ethnic Swedes. Two reasons to that immigrants start new firms is to get out of unemployment and to satis-fy a demand of products preferred by certain groups.

This thesis point out cultural and structural factors that contribute to higher immigrant new firm formation rates and what in general diverges from ethnic Swedes’ new firm formation. Also, push and pull effects are discussed. The ordinary and the advanced service sector in Sweden, where 76 % of all new firms were started in 2005, are studied. The result that emerged shows that immigrants from non-Nordic countries have a positive effect on new firm formation. However, people from FIND-countries (Finland, Iceland, Norway and Denmark) did not have a positive effect on new firm formation.

(3)

ii

Kandidatuppsats inom Nationalekonomi

Titel: Invandrare och Nyföretagande i Servicesektorn Författare: Johanna Kullinger

Handledare: Charlie Karlsson, Professor

Mikaela Backman, Ph.D Candidate

Datum: 2009-06-08

Nyckelord: Invandrare, nyföretagande, Sverige, enkla tjänstesektorn, avancerade tjänstesektorn

JEL-koder: J610, M130, L800

Sammanfattning

Denna uppsats har skrivits i syfte att analysera invandrares effekt på nyföretagande. In-vandrare kan underlätta framtida försörjningsproblem som innebär minskad befolknings-tillväxt och ökad förväntad livslängd. Invandrare startar fler företag än etniska svenskar procentuellt sätt. Två orsaker till att invandrare startar nya företag är för att komma ut ur arbetslöshet och för att tillgodose en efterfrågan på produkter som föredras av särskilda grupper.

Uppsatsen pekar på de kulturella och strukturella faktorer som bidrar till ett högre tagande bland immigranter och vad som i huvudsak skiljer sig från etniska svenskar nyföre-tagande. Även push- och pull-effekter diskuteras. Den enkla och den avancerade tjänstesek-torn i Sverige, där 76 % av alla nya företag startades år 2005, studeras. Resultatet är att in-vandrare från utomnordiska länder påverkar nyförtagandet positivt. Däremot hade befolk-ningen från FIND-länderna (Finland, Island, Norge och Danmark) inte en positiv effekt på nyföretagande.

(4)

iii

Table of Contents

1

Introduction... 1

1.1 Research problem and purpose ... 3

1.2 Background ... 3

1.3 Outline ... 5

2

Immigrants and Entrepreneurship – a Theoretical

Framework ... 6

2.2 Structural factors ... 6

2.1 Cultural factors ... 7

2.2.1 Pull factors ... 7

2.2.2 Push factors ... 8

2.3 Models of determinants of entry in a region ... 8

2.4 Summary and hypotheses... 10

3 Immigrants and New Firm Formation – an Empirical

Framework ... 12

3.1 Data and variables ... 12

3.2 Method and limitations ... 13

3.3 Descriptive statistics ... 14

3.4 Regression models... 15

3.5 Regression results and analysis ... 17

4 Conclusions and suggestions for further research ... 21

5 References ... 22

Appendix ... 25

Tables Table 1.1 Comparison of new firm formation in Sweden and population. 2001-2004…..…1

Table 1.2 Previous studies conducted on similar topics……….…….…..5

Table 3.1 Definition of variables and the expected effect on the dependent variable……...12

Table 3.2 Descriptive statistics for dependent and independent variables………..…14

Table 3.3 Comparisons of models in ordinary service sector, OLS estimates……….17

Table 3.4 Comparison of models in advanced service sector, OLS estimates……….19

Table A.1 Variance inflation factors, VIF, Ordinary service sector………..………...…....25

Table A.2 Variance inflation factors, VIF, advanced service sector………...……...25

Table A.3 Correlation matrix………..…..26

Figures Figure 1.1 Gross new firm formations per capita in percentage terms, 2004.………...…..2

(5)

1

1 Introduction

In Sweden, the interest in the relationship between immigrants and new firm formation has grown since the 1990’s. The attention increased because the situation on the labor market became rougher for immigrants. At the same time, the interest of linking entrepreneurship to economic growth rose (Nutek, 2008). Today, immigrants in Sweden have a higher ten-dency than ethnic Swedes to start new firms. Potential immigrant entrepreneurs start a firm in order to get out of unemployment, realize dreams, earn money, work independently or due to push and pull effects, discussed further on. If new firms are prosperous it leads to economic growth, increase in export and higher turnover (Najib, 1999). New firm forma-tion plays an important role in the economic secforma-tion of the society.

Table 1.1 illustrates the total population of native-born and foreign-born people in Sweden, in 2001-2004. It also contains how many percent of people that have started a new firm. The statistics is based on population statistics from Statistics Sweden and a report from ITPS (2006)1.

Table 1.1 Comparison of new firm formation in Sweden and population. 2001-2004.

2001 2002 2003 2004

Total native-born population in Swe-den

7.881.154 7.887.325 7.897.595 7.911.130 Percent of native-born people that start a

new firm

0.36 % 0.38 % 0.38 % 0.43 % Total foreign-born population in

Swe-den

1.027.974 1.053.463 1.078.075 1.100.262 Percent of foreign-born people that

start a new firm

0.43 % 0.44 % 0.46 % 0.54 % Total population in Sweden 8.909.128 8.940.788 8.975.670 9.011.392 Total percent of people that start a

new firm

0.79 % 0.82 % 0.84 % 0.97 %

Source: Statistics from Statistics Sweden and ITPS. Own calculation

Table 1.1 shows that people born in a foreign country have a higher tendency to start new firms compared to people born in Sweden. 0.97 % of the total population started a new firm in 2004. It is also noticeable that both population groups are increasing in numbers. This thesis will explain and discuss what factors that influence immigrants to start new firms, in comparison to ethnic Swedes. Other studies shows that second generation immi-grants with European parents have a position in the labor market similar to ethnic Swedes. They use less unemployment benefits and social assistance (Ekberg & Hammarstedt, 2004, Ekberg & Rooth, 2003). It is therefore possible that several second-generation immigrant

1In the report from ITPS, information about the entrepreneurs’ country of birth has been partially missing in 2004.

(6)

2

groups will have another effect on new firm formation than the immigrant group in this thesis.

It is positive that new firms supply new jobs. Then the unemployment rate will decrease. Innovative new firms can generate entire new industries in the long run. Economic growth is stimulated. An empirical example from Sweden is a study by Braunerhjelm & Borgman (2004). They found that regional economic growth, especially in the service sector, had a positive relationship with entrepreneurship. In 2004, there were 41 792 new firm forma-tions in Sweden. A new firm formation is a firm that is completely new or has been resting for at least two years. Change of ownership or corporate form or similar changes are not included. The pattern of gross new firm formation in Sweden is illustrated in Figure 1.1 be-low.

Figure 1.1 Gross new firm formations per capita in percentage terms, 2004.

0.28 – 0.53 0.53 – 0.79 0.79 – 1.04 1.04 – 1.29 1.29 – 1.54

Gross new firm formation in the relation to the population is large in the mid-northwest of Sweden in general. It may be affected by the border to Norway. There are many new firms in the Stockholm region. Gross new firm formation per capita was low in municipalities such as Jönköping, Finspång and Smedjebacken. How many firms that got out of business in not included. Figure 1.1 gives an illustration of where entrepreneurs decide to start new firms. In 2004, 34 196 of all new firm formations was formed by ethnic Swedes. 1 378 firms started by people from the FIND-countries. 4 515 firms was started by people from non-Nordic countries (ITPS, 2006). People from FIND-countries are immigrants from Finland, Iceland, Norway and Denmark. The non-Nordic population includes all immi-grants that do not originate from a Nordic country. The regression results in this thesis show that people from non-Nordic countries have a positive effect on new firm formation rates. Entrepreneurs from the FIND-countries did not have a positive effect on new firm formation rates.

(7)

3

1.1 Research problem and purpose

Sweden faces two major future problems that it shares with many other European coun-tries. That is, declining birth rates and a raise of life expectations. Problems with shortage of labor and lack of financing pensions will be vast in the future (World Bank, 2007) unless the problem is solved. Entrepreneurial and labor immigrants can be a part of the solution. Together with new firm formation, they are key tools to be used to stimulate economic growth and support the welfare system. The fact that immigrants have a higher tendency to start new firms makes this problem possible to work out in the future. It is therefore of importance to ask: why do immigrants have a higher tendency to start new firms?

The purpose of this thesis is to analyze the effect immigrants have on new firm formation rates. The group of potential immigrant entrepreneurs is a relevant factor that contributes to new firm formation and, if successful, to economic growth. To start a new firm implies that there are laws and regulations to interpret, among other things. It is an advantage to have networks to get in contact with potential customers. For immigrants, these issues may be more difficult to handle. This depends on their language skills and the access to social networks. Immigrants are often denied taking bank loans needed to cover the start-up costs (Watson et al. 2000). Ethnic Swedes are in general not denied bank loans to the same ex-tent. These difficulties should make immigrants less able to start new firms. Still, immi-grants have a higher propensity to start new firms than ethnic Swedes. This opens up to a discussion. What factors separate immigrant entrepreneurs from ethnic Swedish entrepre-neurs? The answers to this question will support the purpose of this thesis.

With the regression results, it is interesting to see to what extent immigrant entrepreneurs can help to support the Swedish welfare system in the future. Furthermore, ethnic Swedish entrepreneurs and immigrant entrepreneurs can learn from each other and integrate to create a prosperous corporate environment.

1.2 Background

Throughout history, people have migrated to explore new continents, get access to new markets or seek for a better life when escaping war, poverty and/or oppression (Goldin & Reinert, 2007). In the EU-27 countries, 80 % of the population increase is due to migra-tion. Migration has been larger than the natural change in population since 1992 in the Eu-ropean Union (Lanzieri, 2008). Until the 1970’s, most immigrants came to Sweden as la-bor-force immigrants. This pattern came to change. Currently a large proportion of immi-grants are refugees or relatives to already settled immiimmi-grants (Ekberg, 2004). In 2004, about 12 % of the people living in Sweden were born in a foreign country. This is a large share compared to the other Nordic countries. For Denmark, the rate was 6.3%; Norway 7.8%; and Finland 3.2% (Lemaitre & Thoreau, 2006). They have restrictive immigrant policies in comparison to Sweden. In 2008, the largest group of immigrants arrived to Sweden came from Finland. The next largest groups were people from former Yugoslavia, Iraq, Iran and Poland. There were also many immigrants from Norway, Denmark, Germany and Turkey (Nutek, 2008).

In 2008, about 13 % of total firms in Sweden was owned or practiced by an immigrant (Nutek, 2008). The rate of immigrant new firm formation has fluctuated in the latest dec-ades. In the later 1980’s, the share of new firms formed by immigrants started to increase (Najib, 1999). This was due to the increase of immigrants in total. There was also an in-creasing trend in entrepreneurship in small-scale firms (Hjerm, 2004). The share of firms

(8)

4

formed by immigrants dropped in 1991 due to the economic crisis that occurred in Sweden (Najib, 1999). Three years later, it increased again. The large increase did not follow the state of the Swedish market. Najib (1999) concluded that the subsidy allowed to help creat-ing new firms became more extensive in this period. The subsidy was one factor that con-tributed to immigrant new firm formations. Another factor was the large unemployment that Sweden suffered during the recession in the mid 1990’s. This made people start new firms by necessity. In addition, a strong belief in better periods in the future made people start firms (Najib, 1999). Immigrant new firm formations continued to increase. In 1996, immigrants started 5 760 new firms. It has been steady since(Najib, 1999).

In 2004, 76 % of all new firm formations in Sweden were in the total service sector (ITPS, 2006). In 2005, 34 % of all new firm formations within repairing, wholesale and retail trade, hotels and restaurants were started by an immigrant (ITPS, 2006). According to Nutek (2008), both male and female immigrants often start new firms within merchandise trade, hotel and restaurants (ordinary service sector). Immigrants also start new firms within the advanced service sector within business services and letting. Figure 1.2 show a comparison of immigrant new firm formation and new firms started by ethnic Swedes in the service sectors for 2003 and 2004. The statistics are based on the report from ITPS (2006).

Figure 1.2 New firm formations in 2003 and 2004

Source: ITPS (2006)

In Figure 1.2, the two sectors to the left is the advanced service sector. The four other sec-tors is the ordinary service sector (Statistics Sweden, 2007). From Figure 1.2, the conclu-sion can be made that most firms started by ethnic Swedes is within the advanced service sector, while it is mixed for immigrants. They start most firms within merchandise trade, repairs, hotel and restaurants, which is included in the ordinary service sector. Immigrant new firm formation is also frequent within the advanced service sector.

As mentioned, the interest in immigrant entrepreneurs has been increasing during the last decades. Thus, there have been studies about what factors that influence immigrants to be-come entrepreneurs. Table 1.2 shows the key findings of a variety of studies on similar top-ics. 0 2000 4000 6000 8000 10000 12000 14000 Education, health care and

other services Company and financial services Transport and communication Merchandise trade, repairs, hotel and restaurants Construction Manufacturing Service sectors

(9)

5

Table 1.2 Previous studies conducted on similar topics

Earlier literature Key findings

Hammarstedt (2004) Discrimination and unemployment in the society make immi-grants start new firms by necessity.

Brune (1996) Ethnic entrepreneurship has cultural explanations. Immi-grants that have lived in a country where small-firm business is common have a higher tendency to start a firm in their new home country.

Bhachu & Light (1993)

Immigrant economies are created by the support from mi-grant networks. Immimi-grant entrepreneurs gain in being active in networks due to low costs of labor, access to information and assistance in any form.

Aldrich et al. (1990) Several factors explain immigrant enterprises: a group's posi-tion in the host society, the economic structure, certain situa-tional constraint and resource mobilization. The point in time of immigration is also of importance. Immigrants start new firms by necessity as they are discriminated or unemployed.

1.3 Outline

The next chapter is the theoretical framework. It is based on factors that influence immi-grants to start new firms. Furthermore, factors that determine the entry of firms, both into an industry sector and in a region, is explained and discussed. The hypothesis that follows is based on the theoretical framework. The third chapter defines the variables used in the regression model and includes descriptive statistics. An OLS (Ordinary Least Squares) re-gression is used in order to study the effect of the Nordic and non-Nordic population on new firm formation in Swedish municipalities. The regression result follows in Section 3.4. Finally, chapter 4 contains conclusions and suggestions for further research.

(10)

6

2 Immigrants and Entrepreneurship

– a Theoretical

Framework

As revealed above, there are a higher percentage of new firm formations among the immi-grant population compared to ethnic Swedes. This is also true for other Western countries. Clark and Drinkwater (1998) and Fairlie and Meyer (1996) have studied specific immigrants groups. They found that certain immigrant groups in United States and United Kingdom have self-employment rates that are above average. This thesis will point out that immi-grant groups are a relevant variable to consider when studying new firm formation. This thesis will also identify important aspects that influence immigrants to start new firms to a higher extent than ethnic Swedes.

2.2 Structural factors

A person will not start a new firm if his/hers current position expects to generate more in-come than the position as an entrepreneur will do. The current position is the situation that the individual is facing as either wage-employed or as unemployed, receiving unemploy-ment and/or social benefits. An individual is utility maximizing. He or she compares the expected profit from entering a market as an entrepreneur with the expected after-tax op-portunity profit. An entrepreneur only accepts low performance in the initial stage of the firm start. The potential entrepreneur needs to calculate expected income and costs for the current position as well as for the alternative as an entrepreneur for several periods. The following profit-maximizing model shows that an individual will enter a market and be-come an entrepreneur if:

1 − 𝐸(𝜏𝑒) [𝐸(𝑝0𝐿 𝑡𝑞𝑡)− 𝐸(𝑞𝑡𝑐𝑡)] 𝑒−𝛿𝑡𝑑𝑡 − 𝐸(𝐶𝑠𝑢) > 1 − 𝐸(𝜏𝑎) [𝐸(𝑌0𝐿 𝑡) 𝑒−𝛿𝑡𝑑𝑡 + 𝜌 (1.1)

where L represents the expected lifetime of the firm, 𝑒−𝛿𝑡 is the discount factor of income and cost in the future, where δ is the discount factor. 𝐸( 𝜏𝑒) is the expected tax rate on ex-pected profit. 𝐸(𝑝𝑡𝑞𝑡) is the expected income from selling the good or service. 𝐸(𝑞𝑡𝑐𝑡) is the expected costs from producing the good or service, where p is price per unit, q is quan-tity and c is cost per unit. 𝐸(𝐶𝑠𝑢) is the expected start up cost. The right hand-side equa-tion shows the current opportunity posiequa-tion. 𝐸(𝜏𝑎) is the expected tax rate on labor and capital, 𝐸 𝑌𝑡 is expected income at the current position as employed or unemployed. 𝜌 is the risk premium representing the risk with starting a new firm. The entrepreneur takes the expected value into account and is therefore assumed risk neutral (Nyström, 2006). As mentioned earlier, immigrants tend to be risk seeking. Therefore, several factors that influ-ence immigrants’ profit-maximizing decision.

First, studies show that immigrants in general are assumed to be risk seeking (Halek and Eisenhauer, 2001, Bonin et al. 2006 and Heitmuller, 2005). The risk they take to cover the start-up costs is therefore lower than for people who are not risk seeking, and 𝜌 is lower. Second, the capital needed to cover the start-up cost is assumed less for immigrants. It is three times more likely that they borrow capital informally from people in their network. Informal lenders often demand low interest rates/other additional costs. Thus, the proba-bility for immigrants to start new firms increases compared to native-born people that take bank loans more frequently. Note that immigrants are often denied loans by the banks (Na-jib, 1999). Thirdly, 𝑐𝑡 can be assumed to be lower for immigrants. Aldrich, Waldinger &

(11)

7

Ward (1990) pointed out that immigrants learn how to save and minimize their use of re-sources as they face situational constraints. They become efficient in profit maximizing and expect to make profit in a market due to lower expected costs. One example is that immi-grants ease organizational problems by employing their family or people from their net-work. Hereby, they mobilize resources and minimize costs. In addition, immigrants have different experiences from their circumstances of migration that influence their coming po-sition in the society (Aldrich et al. 1990).

2.1 Cultural factors

Several psychological factors influence immigrant new firm formation. First, one has to considerate what type of personality an individual probably possesses to be able to leave their home country and search for a new home in another country. People that are able to escape or leave must take initiatives. Second, escaping involves great risk, even if staying al-so may be risky. A study on risk aversion by Halek and Eisenhauer (2001) showed that immigrants were far more likely to accept a speculative income gamble than native-born Americans. This argument can be strengthened by Bonin et al. (2006) who concluded that earlier studies show that migrants are considered more risk seeking than native-born people. Furthermore, Heitmuller (2005) also concluded that migrants are less risk averse than the population in their home country.

Third, emigration is also associated with large costs. It is costly to move geographically and those with sufficient capital are more likely to move. People with sufficient capital and skills may use these tools to create a stable life in the future in a new country. These tools can be used to start a new firm. The fourth cultural factor is that immigrants from same cultures often are drawn to each other. This creates large networks in their host country. These networks create an opportunity to borrow capital. Informal lenders do not demand interest rates and installments to the same extent as banks do. Furthermore, networks can demand special products that immigrant entrepreneurs are able to sell. These products are similar to products from the immigrants’ original home country (Kloosterman & Rath, 2004) and ethnic Swedes may not have captured the demand. This is often related to a cul-ture and contains products such as magazines, books and recordings (Aldrich et al. 1990). In addition, migration networks let immigrant entrepreneurs get access to co-ethnic labor. In the migration networks, immigrant entrepreneurs also have access to information about pricing, technology and so on (Bhachu & Light, 1993).

2.2.1 Pull factors

Potential entrepreneur are affected by pull factors. The pull effect implies that there are positive factors that affect a potential entrepreneur’s decision-making. High economic growth in a market is a pull factor. It induces potential entrepreneurs to start firms. Norms, laws and regulations, and the organization and functioning of the labor market can also be pull factors. In example, if it is an easy task to handle administrative issues it will favor the entrepreneur’s decision-making. The functioning of the institutions affects the attitudes to entrepreneurship. If the approach is positive, it will act as a pull effect. Furthermore, subsi-dies are a pull factor. In Sweden, the subsidy “support to start of business operations” eas-es the burden of the start-up costs and the coming six-month period (Andersson, 2006). Najib (1999) show that there is no difference in the use of this subsidy between immigrants and ethnic Swedes.

(12)

8

Parker (2004) discussed that the enclave hypothesis holds for immigrants. It states that immigrant groups that live in clusters are creating enclaves that are managing the economy in an independent way. These clusters act as pull factors as they become attractive markets. A particular immigrant group are more likely to establish a business where there are many people living that originate from the same geographical area. Access to large networks or clusters is favorable to the potential immigrant entrepreneur. They have the ability to pro-vide goods and services that is demanded by people within certain networks. Native-born entrepreneurs do not offer these products, as they do not discover this demand (Klooster-man & Rath, 2004). The figuring of a role model also acts as a pull factor. A successful en-trepreneur from an immigrant group can create encouraging effects. It makes immigrants to identify themselves with the role model and push them towards new firm formation (Hamilton, 2000).

2.2.2 Push factors

Push factors are negative factors that affect individuals’ decision-making. Especially one push factor is important to mention in this discussion: the unemployment factor. Unem-ployment pushes the individual to decide to start a new firm by necessity. To start a new firm is a way out of unemployment (Andersson, 2006). As mentioned in the introduction, immigrant groups in Sweden have been facing a tougher labor market than ethnic Swedes. One reason is that immigrants that have an academic education from their home country may not be able to take account of this education as an academic degree in Sweden. This causes immigrants to be discriminated in the labor market. They do not have access to aca-demic jobs. This forces them to work with a profession that requires a lower educational level. Educated immigrants start new firms to get around this obstacle (Kloosterman and Rath, 2004). Moreover, immigrants face other obstacles, such as specific language skills and high specific qualifications that often are demanded by employers (Watson et al. 2000). Ram and Jones (1998) support this argument. They state that immigrant minorities have limited opportunities in the labor market. This is due to discrimination, language barriers, poverty and unemployment. By starting a new firm, immigrants overcome these obstacles. Additional factors that are not noted above are worth mentioning. Nutek (2007) show that the motive to start new firms among immigrants and ethnic Swedes is also due to that people want to realize their dreams, have an independent job and earn money or due to other, unidentified reasons. Reynolds (1997) found that people start new firms due to mul-tifaceted interactions of contextual factors, life course and personal factors. These factors do also affect the profit-maximizing model. If the potential entrepreneur considers these factors very important, it will make him/her start a new firm even if there are high costs and large risks.

To make clear affects new firm formation rates, the following section will give an idea about what set of variables that are commonly used in studies that study determinants of entry of firms. The variables showing the FIND-countries and non-Nordic population groups are integrated further on.

2.3 Models of determinants of entry in a region

Reynolds, Storey & Westhead (1994) did a cross-national study with several European countries and found that new firm formation varied in different regions (countries). The study focuses on the spatial variation of new firm births in all sectors across different coun-tries. Reynolds, Storey & Westhead (1994) based their study on seven underlying processes

(13)

9

that have the largest impact on new firm formation. They used the following variables to determine regional variation in new firm formations:

 Growth in regional GDP and population growth

 Change in unemployment and absolute level of unemployment

 Firm size and percentage of workers in the major economic sectors

 Household income, presence of owner-occupied housing, housing and land prices

 Population density, percentage of the population aged 25-44, percentage of work-force in managerial positions, percentage of population with post-high school de-grees of formal occupational training

 Local political ethos: extent of socialist voting patterns

 Government spending on infrastructure and local community infrastructure

Reynolds, Storey & Westhead (1994) found that same variables explained birth rates of new firms across the countries. For the linear models used, they stated that growth in demand was the most significant variable. In addition, urbanization and agglomeration variables were consistently positive on birth rates of firms. The variable showing change in unem-ployment was only significant in one model; the personal household wealth had a weak positive effect; local political ethos had diverse effects and local spending was not signifi-cant.

Based on the seven underlying processes used by Reynolds, Storey & Westhead, Aki Kan-gasharju (1997) did a study on the regional variation in firm formation in Finland 2. Kanga-sharju used local government spending as a variable in the study. It either increased new firm formation rate due to actions taken by the local government or decrease the rate as high spending possibly lead to higher taxes. Kangasharju’s cross-sectional results showed that high level of unemployment had a positive effect on firm formation. The panel data results showed that the presence of small firms particularly explained Finland’s firm formation, while

demand growth did not.

Nyström (2006) studied regional determinants of entry and exit of firms in industrial sec-tors. Nyström was splitting the regional determinants into three categories: supply of founders, local demand factors and agglomeration effects. The supply of founders concen-trates on what skills the potential entrepreneurs have and if they have other options than being an entrepreneur. Here, unemployment, education and firm size were used in order to see if entry rates are influenced by these factors. With the fixed effect model, Nyström found that the unemployment level had a significant coefficient. The importance of the effect was small, but entry rates increased if the unemployment rate increased. Furthermore, a higher educational level among the population increases entry rates. These people were expected to be good in taking advantage of entrepreneurial chances. However, they earn a higher wage as employed compared to self-employment. Due to the spin-off effect, small-scale firms often attract other new firms. The local demand factors indicate different signs in the market. By looking at population and income, Nyström (2006) could see if there was a poten-tial market with a sufficient demand. With a fixed effect model, the level of income was significant and negatively related to the entry rate. Wage is a cost. Therefore, regions with high wage levels are possibly not attracting potential entrepreneurs. However, high wage

(14)

10

levels attract some potential entrepreneurs as it indicates that demand is high. Nyström stated that the change in income makes potential entrepreneurs to consider entering a re-gion. Population dense and growing regions attract new firms. Agglomeration effects were divided in urbanization economies and localization economies. These factors were found to be of great importance in the study. Urbanization economies are the economies where firms are located close to each other even though they not necessarily operate in the same industry sector. Firms benefit from this by external effects such as reduced transportation costs. Lo-calization economies are economies that have clusters, meaning that many firms operate close together and benefit from spillover.

A study by Davidsson et al. (1994) found that structural characteristics, such as demand growth, were the most important factor that determined entry of firms in a study in Swe-dish regions. In addition, existing role models played a major role. The determinants of re-gional variation in new firm formations and the determinants of rere-gional entry of firms in industrial sectors will be used in the model in this thesis in order to give attention to factors that influence new firm formation rates.

2.4 Summary and hypotheses

This thesis have so far discussed that there are cultural and structural factors that contri-bute to immigrant new firm formation. Potential immigrant entrepreneurs decide to start a firm to overcome the problems aroused in the rough labor market. They start firms to get out of unemployment as well as to satisfy market demand. According to the previous stu-dies, change in demand, agglomeration and urbanization effects, change in unemployment and the presence of small firms have had significant influence on entry rates of firms. The studies did not match their results. None of these models considered the effect that immi-grants have on new firm formation rates. This is what thesis aims to analyze. By combining the models in the theoretical framework with the fact that immigrants in Sweden mostly start new firms within the service sector, makes it relevant to integrate immigrant groups as a variable in the model.

The analysis will be made with the dependent variable F for the ordinary and the advanced service sectors of the economy. The relationship between immigrants and new firm forma-tion is likely to be strong here, as most immigrants start new firms in these sectors (ITPS, 2006). The ordinary and the advanced service sectors are chosen to be able to compare if there are any different results depending on if it is the ordinary or the advanced service sec-tor that is studied.

The following hypotheses are stated:

Hypothesis 1: In the ordinary service sector, the population of the FIND-countries will have a positive effect on new firm formation.

The top-ten group of immigrants that arrive to Sweden comes from Finland, Denmark and Norway. Immigrants have a higher percentage of new firm formations compared to ethnic Swedes. Thus, the FIND-group will have a positive effect on new firm formation. The or-dinary service sector is studied as the relationship between immigrants and new firms is likely to be strong here. Most immigrant new firm formation is within the ordinary or ad-vanced service sector. It is likely that this will remain to be true in the nearest future. Look-ing at the result and find what effect immigrants have on new firm formation in the ordi-nary service sector can be useful for future economic growth.

(15)

11

Hypothesis 2: The non-Nordic population will have a positive effect on new firm forma-tion in the ordinary service sector.

The reasoning behind this hypothesis is that since statistics show that this group most fre-quent start firms within this sector, there will be a positive effect. This group includes people from countries with a culture different from the Swedish culture. The group is also discriminated in the labor market due to various reasons mentioned above. Thus, they start new firms by necessity. This population group has, together with the population of the FIND-countries been chosen to be studied, based on available statistics.

Hypothesis 3: In the advanced service sector, the population of the FIND-countries will have a positive effect on new firm formation.

This hypothesis is stated due to that immigrants from the FIND-countries easy adapt to the Swedish corporate culture. Their culture is similar to the Swedish culture. Highly edu-cated people in this group have the ability to capture entrepreneurial chances that appear. They enter the advanced service sector as they have an educational attainment that coin-cides with the market in the advanced service sector. It is likely that this group will have a stronger effect on the advanced service sector than for the ordinary service sector as people are assumed to the highly educated and come from a similar culture (Hypothesis 1).

Hypothesis 4: The non-Nordic population will have a positive effect on new firm forma-tion in the advanced service sector.

The reasoning behind this hypothesis is that immigrants that start new firms in the ad-vanced service sector are those who have attained a high educational degree in their home country. Still they are not integrated in the labor market. Therefore, this group starts new firms in this sector. They have the ability to capture entrepreneurial opportunities and the skills to manage a business. Based on Figure 1.2 it is likely that Hypothesis 2 will have a stronger effect. Assuming that people from FIND-countries start the majority of firm in the advanced service sector, it is people from the non-Nordic countries that start a majority of firm in the ordinary service sector.

(16)

12

3 Immigrants and New Firm Formation – an Empirical

Framework

3.1 Data and variables

The empirical analysis is based on data on two immigrant groups: people from FIND-countries and people from non-Nordic FIND-countries. The data is at the level of Sweden’s 290 municipalities. For the independent variables, data from 2003 is used. Data from year 2004 is used for the dependent variables. This is because of that it takes some time for people re-spond to changes in the economy. For example, if there is an increase in demand it takes some time for potential entrepreneurs to discover this and respond to it (here: start new firms). In 2003, the number of unemployed people rose to a higher extent in 2003 and 2004 compared to 2002-2003. From 1993-2001 unemployment decreased (Statistics Swe-den, 2005). During this time, people have been able to respond to the increasing unem-ployment and start new firms. The relationship between new firm formation and immi-grants can possibly be stronger in the model used in this thesis compared to other years. However, the main target here is to see if immigrants have a positive effect on new firm formation or not.

The data showing new firm formation rates have been collected from Raps (System for Regional Analysis and Forecasts). This is also valid for the data that involves number of small firms. Both the number of immigrants living in the municipalities, unemployment level and the number of firms have been collected from Raps. The other variables used have been collected from Statistics Sweden. Furthermore, a small firm is defined as a firm with maximum 50 employees, and a large public sector have a large share of employees in the public sector. The following dependent and independent variables are used:

Table 3.1 Definition of variables and the expected effect on the dependent variable

Variable Definition of variable Expected effect

ln(Fk,m) Log of number of new firm formations in

sec-tor k, municipality m.

ln(Nonm) Log of non-Nordic population in municipality

m, divided by population in municipality m. +

ln(Findm) Log of FIND-countries in municipality m,

di-vided by population in municipality m. +

ln(Urbm) Log of population density in municipality m. +

ln(Locm,k) Log of number of firms in sector k in

munici-pality m, divided by total firms in municimunici-pality

m.

+

ln(Demm) Log of GRP per capita in municipality m. +

ln(Incm) Log of average income from employment for

people aged 16+ in municipality m. +/-

ln(Unempm)

Log of number of unemployed people in nicipality m, divided by the population in mu-nicipality m.

(17)

13

ln(Edum) Log of number people with a university educa-tion of 3 years or more in municipality m, di-vided by population in municipality m.

+/-

ln(Smllk,m) Log of number of small firms in sector k in

municipality m, divided by total number of firms in municipality m.

+

Pub 1 if municipality has large public sector. 0 if not. +/-

The empirical analysis will show an expected positive or negative sign for each independent variable used. The expected signs are based on the theoretical framework and the assump-tion that immigrants have a positive effect on new firm formaassump-tion that has been presented in the thesis. A high average income make the potential entrepreneur to enter a market if a high average income is seen as the same as a high demand. A high average income also in-dicates a region with high wages. This may have negative effect on new firm formation. Having many people with higher education may cause either positive or negative effect on new firm formation. These people are more likely to possess high wage-employment and professional occupations compared to people without a higher education (Hammarstedt, 2004). They can also choose to start a new firm as they discover entrepreneurial chances. The dummy variable, Pub, have a positive effect on new firm formation if capital is used support new firm formations and generate additional factors that attract new potential en-trepreneurs, such as infrastructure. This variable reflects if there are many employees in the public sector, which have a negative effect on new firm formation. It then absorbs many employees. Unemployment will have a positive effect. People start new firms to try to sup-port themselves. A large prosup-portion of firms in a sector in relation to total firms in the mu-nicipality will have a positive effect because of benefits such as the spillover effect.

3.2 Method and limitations

The author has been calculating in order to create complete variables. The method used for the regression analysis is Ordinary Least Squares (OLS) and all variables have been trans-formed to natural logs, except the dummy variable. Natural log is useful when square of the conditional mean of the dependent variable is almost proportional to the variance of errors. Using natural log also reduces the heteroscedasticity in the data (Dougherty, 2002). The number of observations differs due to that the dummy variable data for four municipalities is missing.

Firms are assembled into different SNI-codes that classifies in what business the firm is operates. SNI-codes 2002 at a 2-digit level is used. The ordinary service sector in this analy-sis is firms with an SNI-code of 50-64. It corresponds to services such as wholesale and re-tail trade (repair of motor vehicles, motorcycles and personal and household goods), hotels and restaurants and transport, storage and communication. The advanced service sector is firms with an SNI-code of 65-99. Firms operate with financial intermediation, real estate, renting and business activities are included. Education, health and social work are counted in as well. Public administration and defense (compulsory social security) are counted in this group. Immigrants employed in municipalities and government may indirect have an effect on new firm formation. This by taking initiatives and decisions to start new firms by municipalities or government institutions. In addition, other community, social and

(18)

per-14

sonal service activities, activities of households and extra-territorial organizations are in-cluded (Statistics Sweden, 2007).

3.3 Descriptive statistics

Table 3.2 shows a comparison of the descriptive statistics for the dependent variables in year 2004 and independent variables in year 2003. The median is lower than the mean for both dependent variables. This shows that the distribution is not symmetric but positively skewed. Several of the independent variables are positively skewed. The variables showing the FIND-countries population, localization economies in both ordinary and advanced ser-vice sector, GRP per capita and unemployment have a mean higher than the median. For some variables, the distribution is nearly symmetric as the difference between the mean and median is very small. The mean for the dummy variable tells us that about 44 % of the municipalities have a large public sector.

Table 3.2 Descriptive Statistics for dependent and independent variables

Variable Mean Median Min Max Std. deviation

ln(Fk,m) ordinary 2.66 2.64 0.10 7.27 1.12 ln(Fk,m) advanced 2.58 2.40 -2.30 7.80 1.38 ln(Nonm) -3.12 -3.10 -4.51 -1.52 0.61 ln(Findm) -3.61 -3.68 -4.82 -0.96 0.59 ln(Urbm) 3.33 3.27 -1.20 8.31 1.61 ln(Locm,k)ordinary -1.78 -1.76 -2.40 -1.28 0.21 ln(Locm,k)advanced -2.29 -2.33 -3.58 -1.02 0.44 ln(Demm) -1.56 -1.59 -2.40 -0.08 0.31 ln(Incm) 5.24 5.22 5.08 5.85 0.11 ln(Unempm) -2.68 -2.62 -3.78 -1.78 0.35 ln(Edum) -2.72 -2.77 -3.36 -1.40 0.39 ln(Smllk,m)ordinary -1.52 -1.50 -4.20 -0.22 0.26 ln(Smllk,m)advanced -1.26 -1.27 -3.79 1.97 0.39 Pub 0.44 0.00 0.0 1.00 0.50 N=290

The difference between the minimum and maximum for the variable showing urbanization economies indicates that the population is unequally distributed in the Swedish municipali-ties. Both dependent variables have the highest standard deviation. This signifies that new firm formations are centered to certain areas in Sweden. Nevertheless, in some areas it is low. The standard deviation for the variable showing middle income shows that average in-come has a small standard deviation. This is because of that the total inin-come is statistically

(19)

15

distributed among the population. The distance between the minimum and maximum for the dependent variables show that new firm formation varies greatly in Sweden. The locali-zation economies in the ordinary sector do not have a large distance between its minimum and maximum. This indicates that firms in the ordinary service sector are in proportion to the total number of firms in the municipalities. The minimum and maximum is little higher for the advanced service sector.

3.4 Regression models

The ordinary and the advanced service sector are analyzed separately in the models below.

ε in all models is the error term. It shows other factors that influence new firm formation. The error term represents model misspecification and measurement error that may occur in the models. In this analysis, the adjusted R2 is used as a tool to see what variables that leads to the best-fitted model. The adjusted R2 is useful since adding the number of explanatory variables lead to an automatic upward shift. The adjusted R2 attempts to compensate for this shift (Dougherty, 2002). This is relevant here, as variables are added in the different models

The variance of inflation factors (VIF) is applied to see if there is multicollinearity among the variables. The estimated model for prediction will be less stable if there is multicolli-nearity. The independent variables’ VIF should be between 0.1 and 10. It illustrates the va-riables’ regression coefficient inflation of the variance relative to a regression where there are independent explanatory variables only (Everitt & Landau, 2004). See Appendix for ta-ble of VIF. Also, see Appendix for correlation matrix. It is discussed further on.

Model 1:

ln(Fk,m)= β01ln Nonm + β2ln Findm + ε (3.1)

The independent variables that is most relevant to this analysis: ln(Nonm) and ln(Findm) is

used in this model. It creates the ability to add other independent variables in following models. This is done in order to see how the adjusted R2 change when other variables are added. It makes it clear to compare how these variables changes when used in different models.

Model 2:

ln(Fk,m)= β01ln Nonm + β2ln Findm + β3ln Lock,m + β4ln Urbm + ε (3.2) In this model, the variables that fit in to the agglomeration effects are added. They have had strong significance in the mentioned previous studies. They are used to see if they are significant in this analysis. At the same time, it will become noticeable if ln(Nonm) and ln(Findm) are stable.

Model 3:

ln(Fk,m)= β01ln Nonm + β2ln Findm + β3ln Demm4ln Incm + ε (3.3)

Adding demand factors make it able to see changes in the regression model. Hereby, ana-lyze these changes. Demand was found to influence entry rates and new firm formation in previous studies. Income did not.

Model 4:

(20)

16

The supply of founders replaces the demand factors in this model. It is possible to com-pare the importance of these factors in relation to the variables in model 1. It makes it possible to compare the response of the variables in model 1 (due to adding supply of founders), contra the response in adding demand factors. The dummy variable is added. Model 5:

ln(Fk,m)= β01ln Nonm + β2ln Findm + β3ln Lock,m + β4ln Urbm + β5ln Demm +

β6ln Incm7ln Unempm8ln Edum + β9ln Smllk,m + β10Pub + ε (3.5)

In model 5, all independent variables are included in the model. This is to see the final ef-fect that all the independent variables have on new firm formations in the analyzed sectors. How do ln(Nonm) and ln(Findm) respond and what will the adjusted R2 show?

(21)

17

3.5 Regression results and analysis

Table 3.3 show that the critical values of F confirm that the independent variables relates to new firm formation in all models. The variable showing the population from FIND-countries has a negative effect on new firm formation in models 1-3. When the indepen-dent variable is increased with one unit, new firm formation decrease with 0.20-0.22 firms in the significant models, all else equal. The results in models 1-3 reject Hypothesis 1.The immigrants from FIND-countries will not have a positive effect on new firm formation. The non-Nordic population is significant at the 1 % level in all models. The variable has a positive effect on new firm formation between 0.35-1.06 for a one-unit increase in the in-dependent variable . Hypothesis 2 cannot be rejected. The non-Nordic immigrants tend to have a higher positive effect on new firm formation when the demand effects are added, compared to the model where the agglomeration effects are used.

Table 3.3 Comparisons of models in ordinary service sector, OLS estimates

Model 1 Model 2 Model 3 Model 4 Model 5

Constant (t-statistics) 5.25*** (12.95) 5.32*** (7.99) -8.40*** (-2.74) 13.20*** (19.29) 22.22*** (5.39) ln(Nonm) (t-statistics) 1.06*** (11.76) 0.35*** (2.89) 0.68*** (6.79) 0.69*** (8.17) 0.46*** (4.08) ln(Findm) (t-statistics) -0.20** (-2.12) -0.22** (-2.76) -0.22** (-2.52) 0.01 (0.10) 0.05 (0.61) ln(Urbm) (t-statistics) 0.31*** (6.78) 0.11** (2.21) ln(Locm,k) (t-statistics) 1.91*** (8.39) 0.99*** (3.99) ln(Demm) (t-statistics) 0.90*** (5.20) 0.23 (1.41) ln(Incm) (t-statistics) 2.63*** (4.79) -1.87** (-2.40) ln(Edum) (t-statistics) 1.96*** (15.20) 1.84*** (10.21) ln(Smllm,k) (t-statistics) 0.35** (2.25) 0.09 (0.57) ln(Unempm) (t-statistics) 0.93*** (6.56) 0.50*** (2.71) Pub (t-statistics) -0.07 (-0.75) 0.04 (0.44) Adjusted R2 0.32 0.50 0.42 0.65 0.68 F-statistics 69.15 73.33 52.70 88.72 60.75

(22)

18

Durbin-Watson 1.45 1.42 1.58 1.75 1.66

N 290 290 290 286 286

*** Significant at 1 % level ** Significant at 5 % level

Both the agglomeration factors and the demand factors are significant in the models where they are included. The exception is for the variable showing demand (GRP per capita). Av-erage income has a highly volatile parameter whereas in model three. It positively affects new firm formation. In model 5, it has negative effect. Looking at the VIF table, multicol-linearity is higher in model 5 than in model 3. This may be one reason to that the variable is unstable. Model 5 includes many variables and the variable may be affected by this. In addi-tion, the variable is correlated (see Appendix Table A.3) with several other variables. This may cause this volatility. As expected, unemployment has a positive effect on the depen-dent variable. When unemployment is increasing in a municipality, more people decide to start new firms. It is a way for people to get out of unemployment. The percentage of small firms has a weak positive effect on new firm formation. A high percentage of small firms in a municipality do attract potential entrepreneurs to enter the ordinary service sector. The dummy variable is not significant in any model.

The adjusted R2 show that 32 % of the variability in new firm formation is explained by FIND-countries immigrants and non-Nordic immigrants. We also see that the agglomera-tion factors tested in model 2 explain 50 % of the variability in new firm formaagglomera-tion. This is higher than for the demand factors tested in model 3. The model including all variables has the highest goodness of fit with 68 %.

Regarding the correlation matrix in Table A.3 (see Appendix), correlation is significant at the 1 % level or 5 % level in many cases. This is not surprising. For example, in a munici-pality with high population density, there are possibly many people from the FIND-countries and non-Nordic FIND-countries. They often settle down in large cities. In addition, if GRP per capita is high, it is likely that many inhabitants are employed. The industries de-mand many workers. Thus, there is correlation between these variables. Almost all inde-pendent variables correlate with each other. This probably affects the results in the regres-sion models. The correlation matrix show that variables showing immigrants from FIND-countries, small firms and the dummy variable are not correlated with any of the dependent variables.

Autocorrelation in OLS in general, have the same consequences as heteroscedasticity. The Durbin-Watson test tells us that there is positive autocorrelation if the statistic is less than 2 and vice versa(Dougherty, 2002). There is a higher positive autocorrelation in model 1 and 2 compared to model 4. There the Durbin-Watson test is closer to 2.

Table 3.4 below shows the comparison of models in the advanced service sector, OLS es-timates. The variable showing FIND-countries population has a negative effect on new firm formation in models 2 and 3. This tells us that hypothesis 3 is rejected. The variable representing the non-Nordic population is significant at the 1 % or 5 % level in all models. There is then a positive effect on new firm formation with 0.32-1.33 new firms when in-creasing with one unit, all else equal. Thus, hypothesis 4 cannot be rejected. Concerning model 2, where the agglomeration factors are included, the variable showing the localiza-tion economies is significant at the 1 % level. This is also true in model 5, and it has a

(23)

posi-19

tive effect on new firm formation. This is assumed to be due to the spillover effect. The variable representing urbanization economies positively effects new firm formation.

Table 3.4 Comparison of models in advanced service sector, OLS estimates

Model 1 Model 2 Model 3 Model 4 Model 5

Constant (t-statistics) 6.09*** (12.23) 6.73*** (10.01) -20.68*** (-5.80) 14.44*** (17.51) 22.36*** (4.48) ln(Nonm) (t-statistics) 1.33*** (12.02) 0.30** (2.46) 0.73*** (6.25) 0.77*** (7.50) 0.32** (2.25) ln(Findm) (t-statistics) -0.18 (-1.56) -0.25*** (-3.11) -0.23** (-2.31) 0.13 (1.48) 0.02 (0.22) ln(Urbm) (t-statistics) 0.13** (2.57) 0.09 (1.53) ln(Locm,k) (t-statistics) 1.99*** (14.43) 1.42*** (7.20) ln(Demm) (t-statistics) 1.04*** (5.24) 0.44*** (2.60) ln(Incm) (t-statistics) 5.02*** (7.89) -2.05** (-2.14) ln(Edum) (t-statistics) 2.52*** (15.84) 1.51*** (6.86) ln(Smllm,k) (t-statistics) -0.09 (-0.73) -0.33*** (-2.69) ln(Unempm) (t-statistics) 0.83*** (4.93) 0.24 (1.10) Pub (t-statistics) -0.05 (-0.51) 0.00 (0.02) Adjusted R2 0.33 0.67 0.49 0.67 0.73 F-statistics 72.3 144.62 69.33 95.12 78.65 Durbin-Watson 1.50 1.84 1.63 1.82 1.93 N 290 290 290 286 286

*** Significant at 1 % level ** Significant at 5 % level

Demand is significant at 1 % level in both model 3 and 5. A high GRP per capita attract potential entrepreneurs. In model 4, where the supply factors and the dummy variable are added, the educational level has a strong positive effect on new firm formation. This means that people with a university education of three or more years capture opportunities in the market. Possibly, the wage earned as employed is similar to the expected profit as an

(24)

entre-20

preneur. The income parameter varies a lot in the models in the quaternary sector as well. Again, it may be an effect of correlation with other variables or due to multicollinearity. The problem here is the differing in parameter value. The variable showing percentage of small firms has the opposite effect, compared to the results in ordinary service sector. A high percentage of small firms in the quaternary sector do not attract potential entrepre-neurs. A reason to this is that potential entrepreneurs in the quaternary sector have firms operating in businesses that are dominated by large public firms. In education and social work businesses for example. This makes potential entrepreneurs consider that the demand is satisfied if there are many additional small firms.

The variables showing the FIND-countries population and non-Nordic population togeth-er, explain 33 % of the variability in new firm formation. Model 2 and 4 equally have an ad-justed R2 of 0.67. Yet, the fifth model with all independent variables included, show that they explain 73 % of the variability in new firm formation. In this OLS analysis, the Dur-bin-Watson test shows that positive autocorrelation exist in all models. In model 5, it is close to two. There is almost no autocorrelation. The F-ratio shows that the independent variables relates to the dependent variable, for all models.

Comparing the ordinary and advanced service sector, the independent variables explain the variability in new firm formation is at a slightly higher percentage for the advanced service sector than for the ordinary service sector. Comparing the adjusted R2 in model 5 shows that there is a 5 % difference between the sectors. The independent variables in the ad-vanced service sector have a higher variability. While in the adad-vanced service sector, model 2 and 4 have the same adjusted R2 the ordinary service sector do not have the same expla-nation of the variability in new firm formation. Instead, the supply factors, the FIND-countries population and non-Nordic population variables and the dummy variable, explain about 15 % more of the variability in new firm formation compared to the model including the demand factors. The OLS regression for both the ordinary and the advanced service sectors show that people with higher education have a positive effect on new firm forma-tion. Again, this confirms the argument that people with higher education have the ability to ascertain and take advantage of entrepreneurial opportunities. This result is also true for the unemployment variable. Unemployed people get out of unemployment by starting a new firm. It is interesting to see that the variables showing small firms vary. In the ordinary service sector, it has a positive effect, while it has a negative effect in the advanced service sector.

(25)

21

4 Conclusions and suggestions for further research

The purpose of this thesis was to analyze the effect that immigrants have on new firm for-mation rates. Immigrant new firm forfor-mation has been argued to be dependent on discrimi-nation in the labor market, risk behavior, efficient cost minimizing and networks. Networks ease cost of loans and provide certain products in markets where demand has not been sa-tisfied before. This thesis has showed that the non-Nordic population has a positive effect on new firm formation. This was true for both the tertiary and the advanced service sector, for all used models. Despite what other variables that are used, the non-Nordic population remains to be significant. The conclusion is that this population group is considered being an essential variable when studying new firm formation.

This thesis shows that immigrants from FIND-countries have a negative effect on new firm formation. People from the FIND-countries countries often are wage-employed in Sweden. Neither do they face discrimination in the labor market. They have the skills re-quired by the employers and do not start firm by necessity. Immigrants from FIND-countries may start firms in sectors other than the ordinary and advanced service sector. However, this is not likely to be true since 76 % of all new firms are started in the whole service sector. Since the other economic sectors do not have such high new firm forma-tions rates it is highly unlikely that people from the FIND-countries start many firms in these sectors.

People from the FIND-countries may not contribute to new firm formation as they have other primary reasons for moving to Sweden than to start a new firm. On the other hand, neither people from non-Nordic countries do primarily arrive to Sweden in order to start new firms. Still they have an effect on new firm formation. Therefore, it can be concluded that there are factors that influence immigrants from the non-Nordic countries to start new firms that do not influence immigrants from the FIND-countries. It is possible that the unemployment factor is strongly related to non-Nordic peoples’ new firm formation. In the future, it is motivated to study how immigrant groups affect new firm formation with panel data. It would also be interesting to see how strong this relationship is in other economic sectors as well as in other countries. Furthermore, studies on the relationship be-tween other population groups and other economic sectors are an interesting perspective. For example, other population groups may be second and third generation immigrants. Future studies can investigate what effect a rapid and high GDP increase in a region will have on new firm formation.

(26)

22

5 References

Aldrich H., Waldinger, R. & Ward, R. (1990). Opportunities, Group Characteristics, and Strategies in Aldrich H., R., Waldinger & R., Ward (Eds.), Ethnic Entrepreneurs: immigrant

Business in Industrial Societies. United States: Sage Publications. 13-49.

Andersson, L-F. (2006). Företagsdynamik och Tillväxt. ITPS rapport A2006:016. Östersund.

Bhachu, P., Karageorgis S. & Light, I. (1993). Migration Networks and Immigrant Entre-preneurship, in Bhachu P. & Light I. (Eds.) Immigration and EntreEntre-preneurship, Culture, Capital

and Ethnic Networks. 25-49. New York: Transaction Publishers.

Bonin, H., Constant, A. Tatsiramos, K. & K.F. Zimmermann. (2006). Native-migrant Dif-ferences in Risk Attitudes. Discussion paper of DIW Berlin No 560. Berlin: Institute for the Study of Labor.

Braunerhjelm, P. & Borgman, B. (2004). Geographical Concentration, Entrepreneurship and Regional Growth. Evidence from Regional Data in Sweden, 1975-99. Regional Studies,

38(8), 929-947.

Brune, Y. (1996). Invandrare som Egenföretagare. Stockholm: Arbetsmarknadsstyrelsen.

Clark, K. & Drinkwater, S. (1998). Ethnicity and Self-employment in Britain. Oxford Bulletin

of Economics and Statistics, 60, 603-628.

Davidsson, P., Lindmark, L. & Olofsson, C. (1994). New Firm Formation and Regional Development in Sweden. Regional studies,36(4), 389-398.

Dougherty, D. (2002). Introduction to Econometrics. Bath: Oxford University Press.

Ekberg, J. (2004). Immigrants in the Welfare State in Södersten, B. (Ed.).Globalization and

the Welfare state. New York: Palgrave Macmillian. 96-127.

Ekberg, J. & Hammarstedt M. (2004). Unemployment compensation and the use of social assistance among second-generation immigrants .in Sweden. Int J Soc Welfare, 2004(13), 254-265.

Ekberg, J. & Rooth, D-O. (2003). Unemployment and earnings for second generation im-migrants in Sweden. Ethnic background and parent composition. Journal of Population

Eco-nomics,16(4). 787-814.

Everitt, B. & Landau, S. (2004). A Handbook of Statistical Analyses using SPSS. United States: Chapman & Hall/CRC.

Fairlie, R.W. & Meyer, B.D. (1996). Ethnic and Racial Self-employment: Differences and Possible Explanations. Journal of Human Resources, 31(4), 757-93.

Goldin, I. & Reinert, K. (2007). Globalization for Development: Trade, Finance, Aid, Migration,

and Policy. New York: Palgrave Macmillan & the World Bank. Retrieved February 10, 2009,

(27)

23

http://site.ebrary.com.bibl.proxy.hj.se/lib/jonhh/docDetail.action?docID=10173387&p00 =goldin

Gårdlund, T. (1942). Industrialismens Samhälle. Den Svenska Arbetarklassens Historia. Disserta-tion. Stockholm: Tiden förlag.

Halek, M. & Eisenhauer, J.G. (2001). Demography of Risk Aversion. Journal of Risk and

In-surance, 68, 1-24.

Hamilton, B.H. (2000). Does Entrepreneurship Pay? An Empirical Analysis of the Returns to Self-employment. Journal of Political Economy, 108, 604-31.

Hammarstedt, M. (2004). Self-employment among Immigrant Groups – an Analysis of In-tragroup Differences. Small Business Economics, 23, 115–126.

Heitmuller, A. (2005). Unemployment Benefits, Risk Aversion, and Migration Incentives,

Journal of population Economics, 18(1), 93-112.

Hjerm, M. (2004). Immigrant Entrepreneurship in the Swedish Welfare State.

BSA Publications Ltd, 38(4), 739–756.

ITPS. (2006). Nystartade företag - utlandsfödda företagare. Regleringsbrevsuppdrag, 6,

2006/0008 .

Kloosterman, R., & Rath, J. (2004). Immigrant Entrepreneurs : Venturing Abroad in the Age of

Globalization. New York: Berg Publishers. Retrieved February 10, 2009, from Jönköping

University Library website: http://site.ebrary.com/lib/jonhh/Doc?id=10060544&ppg=18 Lanzieri G. (2008). Population and Social conditions. Eurostat 81/2008. Retrieved April 18, 2009, from:

http://epp.eurostat.ec.europa.eu/cache/ITY_OFFPUB/KS-SF-08-081/EN/KS-SF-08-081-EN.PDF

Lemaitre G., & Thoreau, C. (2006). Estimating the Foreign-born Population on a Current Basis by. OECD. Retrieved March 19, 2009, from:

http://www.oecd.org/dataoecd/18/41/37835877.pdf

Najib, Ali. (1999). Myten om Invandrarföretaget. En Jämförelse mellan Invandrarföreta-gande och Övrigt FöretaInvandrarföreta-gande i Sverige. Nya jobb & företag, 9.

Nutek. (2008). Invandrares Företagande. En Statistisk Beskrivning av Utlandsföddas Företagande i

Sverige. Stockholm: Danagårds grafiska.

Nyström, K. (2006,) Entry and Exit in Swedish Industrial Sectors. Dissertation Series, 32, Swe-den: ARK Tryckaren AB.

Parker, C.S. (2004). The Economics of Self-Employment and Entrepreneurship. Cambridge: Cam-bridge University Press.

Ram, M., & Jones, T. (1998). Ethnic Minorities in Business. Milton Keynes: Small Business Re-search Trust.

(28)

24

Reynolds, P. (1997). Who Starts New Firms? – Preliminary Explorations of Firms-in-Gestation. Small Business Economics, 9(5), 449-462.

Reynolds, P., Storey, D.J, & Westhead, P. (1994). Cross-National Comparisons of the Vari-ation in New Firm FormVari-ation Rates in Blackburn, R. & Brush C. (Eds.), Small Business and

Entrepreneurship Volume 2. 191-213. London: Sage Publications.

Statistics Sweden. (2007). SNI Standard Industrial Classification 2002. Retrieved May 11, 2009, from: http://www.sni2007.scb.se/_pdf/040115lillagrona.pdf

Statistics Sweden. (2005). Sysselsättning och arbetslöshet 1976-2004. Retrieved June 11, 2009, from: www.scb.se/.../Sysselsattning_och_arbetsloshet_1975-2004.pdf

Tillväxtverket. (2009, March 27). Antal nyföretagare med utländsk härkomst, per län. Retrieved May 7, 2009 from:

http://www.tillvaxtverket.se/huvudmeny/faktaochstatistik/omforetagande/nyforetagande /fordjupningssidor/antalnyforetagaremedutlandskharkomstperlan.4.21099e4211fdba8c87b 800017887.html

Watson, R., Keasey K., & Baker, M. (2000). Small Firm Financial Contracting and Immi-grant Entrepreneurship, in J. Rath (Ed.), ImmiImmi-grant Businesses: The Economic, Political and Social

Environment. 70-90. Basingstoke/New York: Macmillan/St Martin’s.

World Bank. (2007). Migration and remittances: Eastern Europe and the former Soviet Union. Retrieved February 10, 2009 from World Bank website:

http://siteresources.worldbank.org/INTECA/Resources/257896- 1167856389505/Migration_FullReport.pdf

Figure

Table 1.1 illustrates the total population of native-born and foreign-born people in Sweden,  in 2001-2004
Figure 1.1 Gross new firm formations per capita in percentage terms, 2004.
Figure 1.2 New firm formations in 2003 and 2004
Table 1.2 Previous studies conducted on similar topics
+6

References

Related documents

Däremot är denna studie endast begränsat till direkta effekter av reformen, det vill säga vi tittar exempelvis inte närmare på andra indirekta effekter för de individer som

För att uppskatta den totala effekten av reformerna måste dock hänsyn tas till såväl samt- liga priseffekter som sammansättningseffekter, till följd av ökad försäljningsandel

Inom ramen för uppdraget att utforma ett utvärderingsupplägg har Tillväxtanalys också gett HUI Research i uppdrag att genomföra en kartläggning av vilka

Från den teoretiska modellen vet vi att när det finns två budgivare på marknaden, och marknadsandelen för månadens vara ökar, så leder detta till lägre

a) Inom den regionala utvecklingen betonas allt oftare betydelsen av de kvalitativa faktorerna och kunnandet. En kvalitativ faktor är samarbetet mellan de olika

• Utbildningsnivåerna i Sveriges FA-regioner varierar kraftigt. I Stockholm har 46 procent av de sysselsatta eftergymnasial utbildning, medan samma andel i Dorotea endast

Denna förenkling innebär att den nuvarande statistiken över nystartade företag inom ramen för den internationella rapporteringen till Eurostat även kan bilda underlag för

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