보충급여방식에 의거하여 급여가 주어지는 국민기초생활보장제도는 수급자의 근로동기를 훼손할 것이 우려되고 있다. 수급자의 근로동기 훼손을 방지하고, 적극적으로 노동시장 진입을 촉진하는 유력한 수단의 하나가 근로소득공제제도이다. 수급자의 근로동기를 강화할 수 있는 적정한 소득공제율과 공제방식은 사회경제적 여건과 수급자 집단의 특성에 따라 달라지므로, 우리나라에 적합한 공제방식을 설계하기 위해서는 이론적 탐구만으로는 부족하며 현실검증이 선행되어야 한다. 본 연구는 다양한 소득공제모형을 적용한 근로소득공제시범사업에서 생산된 여러 종류의 실증자료를 결합하여 수급자의 근로동기를 유발하는 요인을 분석하고 있다. 근로동기 강화여부는 수급자의 근로소득과 근로시간의 증가 여부로 지표화 하였다. 분석기법은 Panal Data 분석에 적합한 Random Effect Multiple Regression Model이 사용되었다. 분석결과 다양한 근로소득공제 모형 중 30%의 정율공제가 주어지는 모형이 비교집단에 비해 수급자의 근로소득을 유의미하게 증가시키는 것으로 나타났다. 또한 근로소득공제 모형 외에 수급자의 거주지역, 배우자와의 관계, 고용형태, 소득의 객관적 파악정도 등이 근로소득증감에 유의미한 영향을 미쳤다. 근로시간에 대해서는 채택된 어떤 근로소득공제모형도 유의미한 증가를 유발하지 못했다. 그러나 근로소득과 마찬가지로 근로시간에 있어서도 가구주와의 관계, 건강상태, 고용형태, 소득의 객관적 파악정도에 따라 유의미한 증감이 나타났고, 이에 더하여 연령과 성별도 근로시간에 영향을 미쳤다.

This paper examined the impacts of the Earned Income Disregard Demonstration Project (EIDDP) implemented from July 2002 to December 2004. The EIDDP was designed to move the recipients of the National Basic Livelihood Security Program into the labor market by using the tool of earned income disregard. The EIDDP examined six different models of earned income disregard as work incentive tool for the welfare recipients. This study used a subset of 24 months panal data created by the EIDDP. The subset is composed of 5,269 recipients selected from 137 local communities. Each case of the subset has information on work activity, socio-demographic features, and administrative treatments. To examine the effect of the six different earned income disregard tools, a random effect multiple regression model was used in this study. The results of statistical analysis indicated that Model C, 30% proportional earned income disregard model, significantly increased the earned income of the recipients, while the working hours of the recipients did not increased by any earned income disregard model. The analysis also revealed that the earned income and working hours of the recipients could be significantly affected by other factors including age, health status, transparency of income, and job status, which implies that the Earned Income Disregard Program should be implemented not by nationwide at once but by step-by-step incrementally.

This paper examined the impacts of the Earned Income Disregard Demonstration Project (EIDDP) implemented from July 2002 to December 2004. The EIDDP was designed to move the recipients of the National Basic Livelihood Security Program into the labor market by using the tool of earned income disregard. The EIDDP examined six different models of earned income disregard as work incentive tool for the welfare recipients. This study used a subset of 24 months panal data created by the EIDDP. The subset is composed of 5,269 recipients selected from 137 local communities. Each case of the subset has information on work activity, socio-demographic features, and administrative treatments. To examine the effect of the six different earned income disregard tools, a random effect multiple regression model was used in this study. The results of statistical analysis indicated that Model C, 30% proportional earned income disregard model, significantly increased the earned income of the recipients, while the working hours of the recipients did not increased by any earned income disregard model. The analysis also revealed that the earned income and working hours of the recipients could be significantly affected by other factors including age, health status, transparency of income, and job status, which implies that the Earned Income Disregard Program should be implemented not by nationwide at once but by step-by-step incrementally.

Earned Income Disregard, Work Incentive, the National Basic Livelihood Security Program