BARRIERS TO BANK LOANS AND SMEs EARNINGS: EVIDENCE FROM KOSOVO

The purpose of this paper is to investigate the impact of bank loans on financing the business activities of manufacturing SMEs (Small and Medium-sized Enterprises) in Kosovo. Increasing the level of production of these SMEs is vital for a transition economy such as Kosovo. Representatives of Kosovo's 200 SMEs were interviewed regarding loan barriers and financial support from local and EU (European Union) institutions. Kosovo has a limited number of manufacturing SMEs. The random sampling method was used as a sampling method. The data are cross-section. The collected data were analyzed through the probit model. The SPSS (Statistical Product and Service Solutions) program has also been used to derive statistical results. According to the achieved findings, Not consulting an expert and obtaining loans twice within the analyzed period are significant variables. The results confirm the hypotheses that manufacturing SMEs have obstacles in financing their activities through bank lending and collateral is important for facilitating bank lending.


INTRODUCTION
Entrepreneurship and small businesses are the backbone of economic development all over the world. They play an important role for employment, income and societal changes, particularly in transition economies (BSC, 2011;Qorraj and Jusufi, 2019). Therefore, the development and growth of SMEs is vital for these countries (Watson and Newby 2009;Hashi and Krasniqi, 2011). The Western Balkans consist of six countries: Albania, Bosnia and Herzegovina, Kosovo, Montenegro, North Macedonia and Serbia. The term "Western Balkans" is used for these countries that are not yet part of the European Union (Sanfey and Milatovic 2018;Lubeniqi 2019, Jusufi andBellaqa 2019). The region has a great potential for growth and development of SMEs and their overall economy especially in the service sector, which holds the highest percentage of these firms followed by manufacturing (Rehman et al. 2019). Access to finance is a direct factor that determines the growth and performance of SMEs (Zeqiraj et al., 2020).
The financial system in Western Balkans is composed 90% from the banking sectors, while the other sectors such as non-banking sector and capital market play only a marginal role (Alishani, 2012). Despite the new regulations that these countries have implemented, their banking system continually faces new challenges in a dynamic change of financial system. This arrangement has enabled these countries to be protected from the effects of the global crisis and the Greek crisis (Hafizi and Bushi, 2014). Banking sectors are dominated by foreign banks. Banks from EU countries dominate the region (World Bank Group, 2016). This can be explained by strong economic ties, especially with Germany and Austria Badivuku-Pantina and Alishani, 2012).
The most important positive sides of the foreign banks entry into the Western Balkan countries market are: Innovations of products and services and rapid improvement in banking business, economies of scale and scope, making a competitive environment, attracting FDIs (Foreign Direct Investment), providing stability of host countries' financial systems and bringing a fresh capital (Filimonović and Radovanović, 2011). But despite these impacts, bank lending to the private sector is relatively low in transition countries including the Western Balkans, about 40% of GDP, compared to the Eurozone loan, about 93% of GDP (Košak and Čok, 2008;Kastrati et al. 2018).
This paper aims to analyze the impact of bank loans on financing the business activity of manufacturing SMEs in Kosovo. Therefore, this issue will be explored through theoretical review and empirical analysis. This will be the research problem of the paper.

BANKING SECTOR IN KOSOVO
At the end of war of 1999, financial intermediation in Kosovo was nonexistent and virtually all transactions were settled in cash. In November 1999 the Banking and Payments Authority of Kosovo (BPK) was established, which was an entity successfully regulating the banking and insurance sectors. Kosovo replaced Yugoslav dinar with the German mark in 1999. Banking activity started in earnest in late 2001 and was spurred by the replacement of the deutsche mark by euros in 2002(Demekas et al. 2002. With the use of the stable currency as an official means of payment, the prerequisites for the development of an efficient banking sector and stable macroeconomic environment were in place. Since 2002, the Kosovar banking sector has been profitable. The reasons were lowering interest rates and shifting from short-term to long-term financing (Kubiszewska 2012, 85). Until 2002, the cost of bank lending in Kosovo was high because, as a post-war country, there was great uncertainty, lack of bank liquidity, etc. It can be said that the profitability of the banking sector of Kosovo has significantly improved since 2002. Also the Central Bank of Kosovo has made general reforms of the banking system (writing non-performing loans, privatization of state-owned banks, introduction of modern banking techniques, credit expansion). Also, the high prevalence of intermediation in Kosovo has affected the stability of the bank lending process (Kubiszewska 2012, 91).
This sector started functioning after the end of the war, specifically in 2000. There are ten commercial banks in Kosovo. Only two of these banks have local capital. During the operation of this sector, there is a rapid growth of deposits and loans. The Central Bank of Kosovo licenses banks. Commercial banks operating in Kosovo are ProCredit Bank (the first bank in post-war Kosovo), Economic Bank, NLB Bank (Nova Ljublanska Banka), Business Bank, Raiffeisen Bank Kosovo, National Commercial Bank of Kosovo, TEB (Turkish Economy Bank), Iş bank (Commercial bank).
Commercial banks offer the following types of loans to their clients: Different types of consumer loans, Loans for improving housing conditions, Mortgage loans, Investment loans, Business development loans, etc. (Hasanaj et al. 2017,225-228). It should be noted that there is also AD Belgrade Commercial Bank with its activity in North Mitrovica and Gracanica. It is licensed by Central Bank of Kosovo and operates under its supervision (Kukaj et al. 2020, 88). Financial system assets at the end of December 2020 amounted to 7.91 billion euros. Compared to last year, the financial system in the country is expanding at a slower pace. The banking sector had the main contribution to the growth of financial system assets. This sector has marked an annual growth of 12.4% in 2020. Its growth in 2019 was 13.7% (Central Bank of the Republic of Kosovo, 2021). According to the Kosovo Agency of Statistics (2021) the size of the banking sector compared to GDP is 4.1%. The graph 1 below shows the loans by economic activity.
6 Graph 1 Source: Central Bank of Kosovo (2021), pp.40 The figures are in millions of euros. As can be seen from the graph, more loans are provided to the trade sector. The value of active loans at the end of 2020 reached 3.25 billion euros, representing an annual increase of 7.1% (Central Bank of the Republic of Kosovo, 2021). For the first time in the last five years, lending activity was characterized by a slowdown in growth dynamics from double-digit to single-digit growth, which was affected by the crisis caused by pandemic condition. In terms of economic activities, only lending to the agriculture and manufacturing sector marked an annual decline. The trade, construction and financial services sectors were characterized by a more pronounced increase in lending. The deposit stock at the end of 2020 amounted to 4.36 billion euros. Deposits marked an annual increase of 11.5%.The banking sector at the end of 2020 realized a profit of 22.7 million euros, which is 2.1 million euros higher compared to 2019. The highest profit realized in 2020 was mainly result of higher revenue growth (by euro 6.6 million) compared to expenditure growth (by euro 4.5 million). The level of non-performing loans to total loans continues to be low. The non-performing loan ratio to total loans at the end of 2020 stood at 2.7%, while at the end of 2019 it was 2.0%. Non-performing loans continue to be wellprovisioned, at a rate of 141.4%. In 2019 they were 163.5%. At the end of December 2020, as a result of the higher growth of deposits, compared to the growth rate of loans, the loan-to-deposit ratio decreased to 74.5% from 77.6% in December 2019. The volume of loans in 2019 was 3,031.9 Euros, while in 2020 it was 3,246.4 Euros (Central Bank of the Republic of Kosovo, 2021).

LITERATURE REVIEW
SME financing is of great importance for the development of business activities. Therefore, the impact of the banking sector on SME activity is very significant and financial constrain ranked as one of most constraining feature of the business environment by SMEs (Law and Singh, 2014;Rahman et al., 2017;Moscalu et al. 2020;Beck 2007, pp.3). According to Ahmeti et al. (2014); Horvatova (2018) stability, liquidity and efficiency of a bank among others depends largely on credit. The credit must be real, in the sense that it can maintain the stability and liquidity of the entire banking system.
High real GDP growth levels improve credit quality. The quality of this portfolio is significantly improved by the depreciation of the national currency. While a very important factor that affects the increase of the quality of the portfolio is the protection of property rights. The quality of banks 'loan portfolio is largely determined by factors related to the operating environment, which may be exogenous to banks' actions (Horvatova, 2018). Gherghina et al. (2020) have explored the various barriers that are encountered during the process of obtaining loan by companies. In countries where GDP per capita is high, the legal system is developed, corruption is low, SMEs face fewer restrictions or barriers on access to finance or loan. Beck et al. (2005) studied the financial constraints of SMEs growth. According to them difficulties in dealing with banks, such as bank paperwork and bureaucracies, and the need to have special connections with banks, do constrain SME earnings growth. Grmanová and Ivanová (2018), Radojicic et al. (2018), Dudchenko et al., 2020) claim that collateral requirements and certain access issues, such as financing for leasing equipment also turn out to be significantly constraining for SMEs. Therefore, the efficiency of the banking sector regarding pricing, communication, policy, products, management, must be increased in order to eliminate the barriers to SMEs lending.
Also Athanasoglou (2006); Nekrep (2013) highlight some barriers in the SME lending process. According to them, banks need to be very creative, to offer new services in order for SMEs to be satisfied with lending services. On the other hand SME managers need to be more creative and persuasive in this process, in order to eliminate all possible barriers to the borrowing process. Successful business performance will be difficult to achieve without encouraging creativity and innovation in banks. Fatoki and Asah (2011); Kozarević et al., (2017) emphasise that training and communication on the requirements of banks and trade creditors can help to make SME owners to get investment ready and thus improve access to debt. So this would eliminate all possible barriers during the loan process.
Like other countries, the countries of the Western Balkans face different barriers in the lending process, which affects the growth and development of SMEs in this region. 2019) in her papers concludes that SMEs in the Western Balkan region face many barriers to getting loans from commercial banks. She emphasizes that similar experiences from the past of the Western Balkan countries do not constitute a sufficient basis for the development of banking stability under the influence of the same factors. Market concentration and market competition are the most important factors that influence loan and banking stability in this region. According to Bošnjak et al. (2017) the household sector is the largest consumer of loans in the banking sector of Western Balkan region, after 2010 the number of non-performing loans in this region has increased, while the number of loans to SMEs has not increased. So despite the fact that the governments of the countries of this region are making efforts to ensure a stable banking environment without barriers to business financing, there are still barriers of various natures for business financing.
According to the research of Deakins et al. (2008) the categories of SMEs of Western Balkans that face difficulties in financing their business activities are: SMEs with strong growth in rural areas, new SMEs, young entrepreneurs seeking initial financing and SMEs of the manufacturing sector seeking to diversify and finance the development of new products. While Peci (2015) emphasizes that especially in Kosovo, exporting SMEs are more in need for the bank loans so that they satisfy their increased investment needs in business for expansion and export in other countries. Hashi and Toçi (2010) conclude that SMEs rely more on internal funds and less on bank loans to finance investment, are less likely to apply for a loan, are more likely to be refused a loan, are more likely to fall within the group of discouraged borrowers, and face greater difficulties in accessing both short-and long-term loans. High collateral requirements, limited movement of money, funding, and high interest rates pose a greater obstacle to the operation and growth of SMEs than larger enterprises. Maziku (2012) claims that SMEs' access to loan is problematic due to the uncertainty of operations and their inadequate collateral. Their information is limited to creditors. These SMEs need to demonstrate convincing and realistic business plans that show possible future returns and sustainability. Poor transparency is a weakness of SMEs. If the availability of information through credit registers and other notification systems increases, the cost of reviewing credit applications by banks will decrease. Based on these sources of literature, the first hypothesis of this research can be defined: H1: The differences in Kosovar manufacturing SMEs' barriers to obtaining loans affect the probability of financial growth 9 The liquidity of commercial banks is of significant importance in the realization of their business activity (Devjak 2018, p. 3). In order not to jeopardize their liquidity, banks take various measures. According to Haron et al. (2013) collateral refers to the guarantee for the loan taken from the bank. Collateral acts as an indicator that enables the bank to mitigate or eliminate the problem of unfavorable selection caused by the existence of information asymmetries between the bank and the enterprise or entrepreneur at the time of the loan decision. Despite the fact that the commercial bank recognizes the quality of the clients' credit, the collateral helps to alleviate the problems of moral risk after the loan has been granted to the enterprise. Aghion and Bolton (1992) argue that collateral can be seen as an instrument that ensures good behavior on the part of borrowing enterprises. There is always a threat that the loan taken may not be repaid. Therefore, from the point of view of the commercial bank, the collateral is very important, while from the point of view of the enterprise, it is a burden. The presence of collateral despite being seen as an obstacle by enterprises, it can also be seen as a facility which enables easier obtaining of credit by SMEs. In this way it would also accelerate its business activity as a result of secured credit.
Hasnah et al. (2012) claim that managerial skills influence more than the presence of collateral in obtaining credit by SMEs. According to them, as long as SME managers have convincing skills regarding the financial and material capacities of their SMEs, the presence of collateral in obtaining credit from SMEs would not play a major role. According to Boot el al (1991) Collateral is more required for young SMEs and which do not have financial viability, while SMEs who have long business experience, proven business activity the presence of collateral is not required much. Hyz (2011) investigated the barriers to the use of banking services by Greek banks. According to the results achieved, banks avoid providing loans to start-up, innovative and very new SMEs because these SMEs do not have sufficient collateral. Providing financial products to these SMEs is high risk. Voordeckers and Steijvers (2006) in their research have concluded that household SMEs require the highest level of collateral, while competition between banks reduces the demand for collateral regardless of the degree of risk. Similar results have been achieved by Bukvic and Bartlett (2003), where according to them the demand for collateral from SMEs by commercial banks is a barrier for SMEs. Therefore only the entry of new banks into the market, ensuring competition between commercial banks would reduce the demand for collateral by SMEs. The discussion so far orients us to the second and third hypothesis of this paper: H2: The differences in Kosovar manufacturing SMEs' frequency of obtaining loans affect the probability of financial growth H3: Collateral for obtaining loans is a significant predictor of Kosovar manufacturing SMEs' financial growth After reviewing the literature on the issue of financing of manufacturing SMEs, barriers to obtaining credit, the effect of collateral and their impact on profits or growth of SMEs, the hypotheses arising from the literature review were presented.

METHODOLOGY
The research sample consists of 200 SMEs which are from the manufacturing sector, specifically from sectors such as Food industry, Beverage industry, Processing of metals, Plastic industry, Furniture, Chemical industry, Wood industry, Textile industry, shoe industry, Packaging industry, sheet metal industry, Stone processing, Production of mechanical parts, Production and processing of mushroom, ect. These SMEs are exporters to the EU market, and to the Western Balkans market. Data on these SMEs are provided by the authors of the paper from the Ministry of Trade and Industry of Kosovo. So all the data related to these SMEs are provided by the officials of this ministry. According to the Kosovo Agency of Statistics (2020), there are 354 manufacturing SMEs in Kosovo.
The reason why only manufacturing SMEs have been selected and not the service and trade ones, is that the production base in Kosovo is low, as in other countries of the Western Balkans. For a sustainable economic development of this region, the production capacities must be increased and the entrepreneurs who establish manufacturing enterprises must be supported. They face many obstacles, especially in getting loans from commercial banks. Therefore, the analysis of these issues will enable the better recognition of this issue and the drawing of conclusions regarding the solution of these difficulties in access to finance of manufacturing SMEs. Also a better rationale for including manufacturing SMEs in the sample is that they require material which incurs costs higher than trade or service sector.
The questionnaire was used to collect data. The data were collected by interviewing CEOs or even the financial managers of these SMEs by the authors of the paper. The CEOs and financial managers of these SMEs were interviewed during the period July-September 2020.Their interview was conducted face to face. The interview lasted about 40 minutes. Meanwhile, the interview took place in the locations of these SMEs, where the authors went to interview these managers during the aforementioned period. The collected data is processed through the SPSS Program. The Probit model was used to find the influence of independent variables in increasing the earnings of these SMEs.
Because the dependent variable consists of 2 categories, we decided to use the probit model rather than SEM model, Correlation, Regression or Factor Analysis. Starting from the principles of statistics and econometrics, in cases when the dependent variable has only 2 categories (Yes or No, Male or Female) then the logic or probit model is used. It should be noted that all these SMEs have taken loans once or even several times from commercial banks. Some of them even have financial support from various institutions.
The econometric model used to measure the impact of these variables on turnover growth is presented below: P (Yi= 1) = Φ (β0 + β1 xi1 + β2 xi2+ β3 xi3 + β4 xi4 +ἐ) The following table shows the coding, descriptions and measurements of the variables which are used in the econometric model. As can be seen from the table, the variables with more than 2 categories, have reference categories. The reference category does not enter the analysis because the values of the other categories are compared with the reference category. All variables are nominal, except for how_many_times_have_you_got a_loan_during_this_period? which is ordinal variable. These values represent the average amount of loan received by SMEs within the analyzed period.

EMPIRICAL RESULTS AND DISCUSSION
In this section of the paper the results achieved by empirical research will be presented. Initially, the results obtained from the descriptive statistics will be presented. Regarding the first variable, 123 SMEs have declared that their earnings has increased during the period 2015-2019. While 77 others have declared that their earnings have not increased significantly. So the managers of these 77 SMEs claim that they are not satisfied with the increase of earnings of their SMEs, so they had the opportunity to have significant growth or earnings but various objective and subjective reasons have influenced not to have growth.

Source: Authors own calculations.
It is interesting that a large number of SMEs have declared that not consulting an expert in the field of credit or finance is the main obstacle in obtaining loan from commercial banks. Based on these results, it can be said that manufacturing SMEs lack the necessary knowledge and expertise regarding shortterm and long-term lending. Most of the surveyed SMEs declared that they only took bank loans twice. Various obstacles have made it difficult for manufacturing SMEs to access bank loans. It is worrying because a non-signature number of SMEs have received financial support from various local and international institutions. In particular, EU funds such as IPA funds. This shows that these SMEs are not sufficiently prepared to meet the various conditions. Commercial banks should have appropriate financing policies that would greatly help the SMEs' manufacturing activities.
In order for the results of our model to be relevant, we must use model diagnostics. For this we need to use the specification test, which shows how specific our model is. Then to check the statistical significance of the model Goodness-of-fit test is applied. And finally to check if the regressors are orthogonal we used the Multicollinearity test. Specification test results for the probit model are presented below. Source: own calculations.
Based on these results our model is rightly specified because it is statistically significant. As part of Goodness of Fit Test are: a) Likelihood ratio (LR) test; b) Hosmer and Lemeshow's Goodnessof-fit test In most cases, the log of the likelihood is used more often than the likelihood because it is easier to elaborate (Salisu 2016). The log of the likelihood is always negative, with higher values (close to zero). This indicates a more convenient model. In our model the log of the likelihoodis -14.87382. Table 4 Hosmer and Lemeshow's Goodnessof-fit test
Our model fits the data well if the Hosmer and Lemeshow's test statistic is not statistically significant. Multicollinearity test is performed to verify the heavy correlations between the regressors in the given model. Both the tolerance and variance inflation factor (VIF) are 1, which shows that they are completely uncorrelated. The tolerance goes to 0, and the VIF gets very large if a variable is very closely related to another variable. Below are the test results for our model.

Source: Authors own calculations.
In variables Barriers_to_bank_lending, all categories have a positive relationship with the reference category, with the exception of the Poor loan history (4) category which has a negative relationship with the reference category. While the first category Not consulting with an expert (1) is significant. Manufacturing SMEs were asked whether not consulting an expert who would provide information on the lending process is a barrier to bank lending. There are firms in Kosovo that provide consultancy to SMEs that intend to apply for bank loans. Consulting with the experts of these firms will undoubtedly create more favorable conditions for obtaining loans by SMEs. The positive correlation of the third category (2,126) is greater with the reference category than the other categories. So the third category has a greater negative impact than other categories as a barrier to SME financing.
From these results it can be concluded that these SMEs do not receive consultancy by financial or banking experts regarding obtaining the bank loans. This is a big barrier for these SMEs because they do not have enough knowledge of the lending process. So it can be said that this is the biggest barrier of these SMEs which negatively affects the lending process. It should be noted that in Tmava et al. (2013) research, the business plan presented statistical significance in the process of obtaining loan by SMEs in Kosovo. Meanwhile, in this research, Not consulting with an expert is presented as the biggest problem during the lending process by SMEs.
How_many_times_have_you_get a_loan_during_this_period? variable highlights as the first reference category One time which with most other categories has negative relationships. Only the Odd Ratio of the Three time category 1.006 show positive relationship with reference category. Two time (1) category is significant. From the data obtained, it can be understood that these SMEs receive very little loan within the five-year period. This is also evidence that due to various constraints, these SMEs have stalled in obtaining loans for development and increasing production capacity.
The pillars of the European SME Assistance Scheme include the following elements: Structural funds, Thematic funding opportunities, Financial instruments, and Funds aimed support for internationalization of SMEs. Adina-Simona (2013) claims that EU funds increase the performance of the SMEs in terms of developing production and circulation capacity and acquiring new equipment. Also Vujicic et al. (2012) state that EU financial assistance mechanisms or IPA funds are of great importance in the development of business activities, specifically in the development and growth of SMEs.
The research of Jusufi et al. (2020) shows that IPA (Instrument for Pre-Accession Assistance) funds are of great importance in the business activity of manufacturing SMEs in Kosovo. They are even of great importance in modifying and developing new products, internationalizing these SMEs, as well as expanding the market. Despite this, Kosovar SMEs do not have the skills and potential to absorb these funds. The Ministry of Trade and Industry, the Ministry of Finance and various government agencies should organize continuous training aimed at increasing the absorption capacity of Kosovo SMEs. Currently very few trainings and workshops are organized with the aim of absorbing IPA funds from Kosovo SMEs.
In terms of Kosovo government lending programs, Kosovo ranks very low compared to the Western Balkan countries for SME support services and SME start-ups. SME support programs include modernization initiatives, capacity building, and quality improvement, primarily come from donors such as the World Bank, EBRD, and GIZ, government initiatives in this area are lacking (European Investment Bank 2016, 15).In this research, the effect of receiving financial support from local and international institutions in increasing the earnings of manufacturing SMEs was also analyzed.
The coefficients or Odd ratios of the other 2 categories show a positive relationship with the reference category which is IPA funds. Meanwhile, the last category has a negative relationship with the reference category. As financial support for these SMEs increases, so do their earnings. Due to low absorption capacity, a small number of SMEs in the Western Balkans region have benefited from EU financial support. All manufacturing SMEs need to be trained in capacity building, in order to benefit more from EU funds and various local institutions.
Another important issue is the presence of collateral and its impact on improving the likelihood of obtaining loan from Kosovo's manufacturing SMEs. It should be noted that the banking system and SME financing is inadequate in relation to the real needs of SMEs in Kosovo. Commercial banks in Kosovo, through high interest rates on loans and strict criteria for getting loans, have created a system to protect against the risk of financial crisis. Bajçinca (2015) in his study claims that consumer loans can increase poverty by pushing families to face major financial difficulties. Many SMEs take out loans to buy cars and increase luxury in their offices. However, they do not receive loans for the professionalization of the workforce or even the expansion of production lines and bases.
Many SMEs do not have a problem with mortgages, but they do have problems with liquidity and the possibility of repaying the loan installment. Commercial banks have strict rules that make it difficult for businesses to repay their loans. Among them is the collateral. This variable which consists of 2 categories is significant in our model.Although SMEs do not like collateral, they think that collateral facilitates the process of obtaining and approving a loan from a commercial bank. Therefore, the second hypothesis is correct because the collateral facilitates the lending process. Rudhani and Balaj (2019) claim that despite the fact that the banking system of Kosovo has financial stability, credit officials are avoiding liquidity risk. For this reason, they reject the loan applications of many SMEs. Also according to IMF (2017) non-performing loans are still high in the Western Balkans. To avoid the growth of these loans, commercial banks have set strict rules for both consumer loans and SME loans. So despite the fact that collateral is a hassle and an obstacle for SMEs, it still facilitates the lending process.

CONCLUSION
This research is of particular importance because it aims to analyze the impact of bank loans on financing the business activities of manufacturing SMEs in Kosovo. So, are there differences in the effects of banking barriers, in the frequency of obtaining credit and the presence of collateral are the research questions of this paper which are elaborated in the sections of this paper. Regarding the first hypothesis, from the achieved results, we can prove it as correct because not consulting an expert in the field of credit or finance is the biggest barrier, the effects of which are greater compared to the effects of other barriers. So the first hypothesis is supported.
From the research results, taking bank loans twice is significant. Therefore, there is a difference regarding the frequency of obtaining loans by Kosovo manufacturing enterprises. Thus the second hypothesis is also supported. These figures show that manufacturing enterprises do not get enough loans in order to grow their business. Most of them take loans twice. Therefore, commercial banks should ease their criteria when it comes to obtaining loans in order to increase the production capacity of Kosovo manufacturing enterprises.
The third hypothesis of this research can also be supported because the collateral is significant in the obtained statistics. Collateral is an important predictor of financial growth of Kosovo manufacturing enterprises. The results obtained from this research highlight some managerial and policy implications. The main policy implication is that Kosovo manufacturing SMEs must be supported by both state institutions and commercial banks. The state through appropriate fiscal policies should support these SMEs, while commercial banks through loan restructuring should facilitate the lending process. Despite the fact that the risk policy for radical change is complicated and very often it is subject to approval by the central offices of parent banks from the countries of origin of the banks.
If we rely on the results achieved by the probit model, it can be argued that not all bank lending barriers are the same for SMEs. Not consulting with an expert is a significant category in our model. From the perspective of SMEs this is a major barrier which affects the failure of the lending process. There are very few consulting offices that provide financial advice on the lending process, so another policy implication is that in Kosovo more credit consulting offices will have to be opened in order to inform SMEs about the lending process. Banks need legible, reliable, transparent and comparable financial documents. Therefore, there should be consulting offices with financially experienced experts, so that SMEs get the right information about the lending process.
The category of SMEs that have twice taken out a loan is significant in this model. Except for the third category which has a positive relationship with the reference category, all other categories have negative relations with the reference category. Due to the very strict rules of the banks and the inability of most SMEs, the number of loans taken for investment purposes during the five-year period is extremely low. With the increase in the number of quality loans, the possibility of increasing the turnover of these SMEs also increases. This is the third important managerial implication achieved by this paper.
Interest rates on loans in Kosovo continue to remain high and even high compared to those in the region. The average interest rate on loans is 6.8%. These current rates have prevented the development of businesses and some of them have even gone bankrupt as a result of these astronomical rates. Even the processed data of this research prove this fact. Categories that show great satisfaction with interest rates are negatively related to categories that show low satisfaction with these rates. Therefore, the Central Bank and other supervisory institutions should monitor commercial banks in order to have reasonable interest rates. Also, as a managerial implications it can be said that SME managers should, through various organizations, express their dissatisfaction with interest rates. This paper has some limitations. The analysis was conducted only with Kosovo manufacturing SMEs. Also, the larger the number of variables, the higher the level of research. Therefore, future studies should eliminate these limitations so that research is more comprehensive. Regarding the current gaps and limitations of this research, it can be said that future research should include more variables and indicators to gain more empirical evidence regarding loan barriers by manufacturing SMEs, as well as the impact of these barriers on growth and profits of these SMEs in this sector. This is one of the few studies that thematically has the impact of credit barriers on the performance of manufacturing SMEs. About 70% of SMEs in Kosovo are commercial (trade) and service SMEs. Even for these SMEs there is not enough literature regarding the impact of loan barriers on their business performance. Future research should therefore expand the research sample to include all SMEs to achieve more general results on this issue. Hence where research can be termed as a pilot study to serve later in other larger and more variable studies where more generalized empirical conclusions and evidence would be drawn.