중소기업 정책자금 고용효과의 지속성 분석 = The Employment Effects of Policy Loan on the SMEs in Korea; Sustainability Analysis

다국어 초록 (Multilingual Abstract)

Since the Korean economy is considered as being trapped in a jobless economic growth stage, it is important to create jobs via implementing various economic policies. Among them, this study investigates whether the policy loan on the SMEs in Korea is being operated in employment-friendly ways. Using a micro-econometrics analysis, we compare the performance between treated and controlled establishments to calculate the employment impact. While the employment effects using an input-output table of a central bank in the perspective of macro-level public expenditure have an indirect impact on the establishments which have used policy loans, micro analysis could generate a direct impact on employment in the participating establishments relative to the non-participants. This study focuses only on the loans associated with the ``Small and medium Business Corporation`` (SBC), which is a non-profit public organization established to implement such government policies for the development of SMEs. Regarding this program, it is well known that commercial banks are reluctant to serve SMEs. This is mainly due to lenders` incomplete information about SMEs and the firms` lack of collateral to secure commercial loans. Policy loans for SMEs, therefore, are implemented to compensate commercial banks` losses and take inherent risks, while keeping borrowing rates low relative to the commercial rates. In this study, we evaluate the job-creation effects of the policy loans in the following ways: First, we use an establishment-level employment insurance database of the Korea Employment Information Service for the employment effects evaluation. Second, to cure a potential bias from treatment selection conditional on observed variables due to the effects of unobserved variables, we first introduce a simple two-period, two-group comparison associated with the employment effect before and after implementing government programs, which is well known as the ``difference in difference`` (DID) method. Third, since we do not know outcomes for untreated when it is under treatment, and for treated when it is not under treatment, individual performance cannot be observed simultaneously. Thus, we estimate the employment effect by comparing the performance of the treatment group that participated in the relevant policy and that of a similar control group. Forth, we propose strategies for increasing jobs with the support of policy loans for SMEs. Analyzing the effect of SME policy funds on employment, we extracted 25,613 establishments out of the establishments that applied for policy-loans between 2005~2010. To control for double support, the sample is restricted to the establishments that applied only once to the policy loan. The employment effect of the beneficiary establishments of the SBC policy funds is analyzed to see the T+1 year, T+2 year and T+3 year,starting from target year (T). After the policy fund support, within the three years of the research period, the result clearly shows that the number of average employees of participant establishments increased compared to that of non-participants. According to a simple DID analysis, the average number of employees of the participants increased by 1.84 persons after three years from receiving the policy loan compared with non-participants. This effect, however, seems to be a short-run phenomena as we have difficulty finding evidence of employment sustainability associated with policy loans. This result is robust with the Heckman (1976) type of adoption and usage models. It is also investigated that the employment effects are higher in the industries such as electronic components and precision instruments than other industries. The short-run effects are higher in policy loans associated with the management stabilization, while the long-run effects observed in loans associated with the commercialization of R&D results, foundations of new economic growth, and venture business start-ups. It was also found that in the short run, working funds are an effective way of maintaining and increasing employment rather than plant and equipment investment, and vice versa. The effect of approved policy loan on employment is negatively associated with firm age and its size, while it is positively associated with the return on net sales. It also found that the collateral and debt ration of establishments do not significantly affect establishments` employment. SME financing from the public sector seems to be essential. It has been shown that, at the minimum, there exists a significantly positive relationship between the policy fund and its effectiveness on employment in the short run. Our results also imply that when providing policy loans to increase the employees, it is more desirable to target young prospective SMEs rather than matured firms in terms of firm size and age. Lending to SMEs via commercial banks is decreased in a recession phase, while increased in the expansion phase, causing excess investment. Policy loans can play a pivotal role in alleviating this kind of co-movement between commercial loans to SMEs and the business cycle. In particular, for maintaining and increasing employment of SMEs, policy loans in the recession phase should be concentrated on short-run working funds, while those in the expansion phase should be concentrated on long-run plant and equipment investment. This paper is divided into four sections, including the introduction. In section 2, we introduce the methodology and the description of data used in the paper. In section 3, we empirically analyze the effect of policy loans on the establishments` employment using micro level pooled data. The last section concludes the paper by providing policy recommendations to reinforce the employment effect of loans from the public sector.