초록 열기/닫기 버튼
본 연구에서는 NPO(공익법인) 감시기관이 부여한 제3자 등급(별점 평가)을 조사하여, 그러한 등급의 증가 또는 감소가NPO의 기부금 수입에 독립적인 영향을 미치는지를 회귀불연속설계(Regression Discontinuity Design: RDD) 접근법을 사용하여 분석하였다. 분석결과를 요약하면 다음과 같다. 첫째 NPO에 대한 제3자 등급(평점)이 기부금 의사결정에 유용한 정보가 된다는 가설에서는, 모든 등급의 증가가 기부금에 우호적인 영향을 미친 것이 아니라, 일부의 범위(별 등급 2개에서 3개로 증가)에서만 지지되었다. 둘째 규모에 따른 등급과 기부금 크기의 관계는, 대규모 NPO의 경우에 별 등급 2개에서 3개로 등급의상승이 임계점(90점)에서 유의한 기부금 수입의 증가(jump)가 나타난 것을 제외하고는 소규모NPO의 경우에는 등급의증가가 모두 임계점에서 기부금의 증가 즉 회귀불연속이 발생하지 않았다. 셋째 개인기부자와 기업기부자의 등급에 대한반응이 상이할 것이라는 가설에서는, 개인기부가 임계점 90에서 등급의 처리효과가 기업 및 단체기부에 비해 두 배 이상강하게 나타났다. 넷째 NPO에 대한 제3자 등급 정보와 웹 사이트 공시 정보가 언론보도에 비해 상대적으로 기부금 크기와 강한 관련성을 갖는다는 것이 확인되었다. 또한 추가분석에서는 매년 연속적으로 별 3개 등급을 받은 NPO가 그렇지않은 NPO에 비해 기부금 수입에 매우 강한 양의 영향을 미친다는 것을 보여주었다. 본 연구의 시사점은 제3자 등급평가의 향상이 NPO의 기부금 수입에 양의 영향을 미치는 것은 분명하지만, 무엇보다 실제 평점의 절단점에서 변화하게 되는별 등급 정보가 기부 의사결정에 핵심적 역할을 담당한다는 것이다. 향후 연구에서는 최근 신용평가 제도가 등급에서 점수제로 전환되는 것처럼 별 등급을 보완할 수 있는 다양한 평가방안이 제시되어야 할 것이다.
In this study, we investigate the role of third-party ratings (star ratings) given by watchdog organizations and empirically examines whether the zero- to three star ratings provided by GuideStar KOREA have additional information content for donors using the Regression Discontinuity Design (RDD) approach. To test the relationship between charity ratings and donations, we examine 315 non-profits rated by GuideStar in Korea for the period 2015-2018. The results are as follows: First, the overall ratings have a statistically significant effect on the he amount of contributions receives. Both star ratings and rating score(overall ratings) have a significant positive impact on the contributions, especially star ratings have shown a strong correlation with donations rather than rating score. That's because stars are relatively easy to interpret than overall ratings. Second, before doing RDD analysis, we also verified that assumptions for applying RDD are properly established in our model. The results show that there is no evidence of a significant change in control variables at the threshold values of the overall ratings, the forcing variable (a function of overall rating) is not manipulated, and McCrary’s density test yields a nonrandom sorting of charities across cutoff ratings. This finding implies that the assumption of applying RDD is found to be satisfied. Third, for NPOs that are rated as 3 stars, a one point increase in overall ratings is associated with a 1.4 times increase in charitable contributions received, while for NPOs that are rated as 2 and 3 stars, the estimation results show that an increase in overall ratings does not have a significant impact on charitable contributions received. Fourth, this study shows that for large NPOs that are slightly below the 3 star threshold rating, a one star increase has a significant impact on the amount of contributions that the NPO would receive. However, large NPOs that are slightly below the 2 star threshold rating and small NPOs that are below 2, 3 star threshold rating, a one star increase has no significant impact on the amount of contributions an organization receives. Fifth, we find that the responses to the ratings of individual donors and corporate donors are all significant positive, but contrary to expectations, the treatment effect of the ratings at 90 cut-points is more than twice as strong as that of corporate donations. Lastly, the study indicate that third-party rating information and Web site disclosure information on NPOs are relatively relevant to the contributions received compared to media coverage. It also show that NPOs received the first grade for the third consecutive year have higher contributions than those who do not. Overall, we find that a change in GuideStar ratings is associated with a directionally consistent change in donations. Above all, the study suggests that star rating informations at the cut-point (assignment point) of the actual rating, plays an important role in the decisionmaking for charity giving. However, the effect of these star ratings is limited to changes in the relatively high star ratings, from two star ratings to three. Therefore, in order to address complaints from NPOs that do not receive three star ratings due to the difference of actual marginal rating, various measures should be worked out to supplement star ratings system, just as the latest credit rating system shifts from star rating to scoring system. The results of this study will be useful information to improve the rating system and activation of donations by NPO CEOs and the government.
키워드열기/닫기 버튼
Donation, Regression Discontinuity Design (RDD), NPO (non-profit organization), star ratings, rating score
