A non-regulatory environmental program has become increasingly common as a tool for public organization especially at local level, such as residential house weatherization programs, incentives for car pool programs, and LEED certificates for new buildings. However, due to its voluntary nature, participating public organizations nominally adopt the program without considering how to design the program for better quality. Low quality of the non-regulatory environmental program is represented in the piecemeal, uncoordinated approach producing a non-committal response to environmental protection. For a quality of the program, many scholars argue that the non-regulatory programs should take comprehensiveness into consideration when designing them.

Much scholarly effort went into non-regulatory programs in business, meanwhile few studies addressed non-regulatory environmental programs in public organizations. In addition, virtually no attempt was made to investigate the effect of non-regulatory programs on their environmental performance in public organizations. To bridge the gap, this study examines whether better quality of non-regulatory environmental programs in local government really matter for environmental improvement in the locality. Through literature review, a quality of non-regulatory programs in public organizations is defined as more comprehensive, well-integrated approach in this study. Therefore, this study will investigate the relationship between comprehensiveness of non-regulatory environmental programs taken by local governments and their environmental performance.

In this analysis, I create a dataset to test the relationship between non-regulatory environmental programs and environmental performance at local level. The dataset is constructed by integrating the 2010 Local Government Sustainability Policies and Programs survey of ICMA, the U.S. Census Bureau County Government Finance, and EPA Air quality statistics by county and city. The unit of analysis for this study is 2,176 county and city governments that responded to the ICMA survey. In examining the effects of more comprehensive non-regulatory programs, it raises a concern of the endogeneity issue where some of the variables accounting for comprehensiveness of non-regulatory programs are likely to influence environmental performance as well. To avoid the endogeneity bias, this study employs a two-stage least-squares regression model (2SLS). Basic instrumental variables (IV) are political leadership and financial condition of local government variables.

In the first stage, this study estimates the relationship between determinants and comprehensiveness of non-regulatory environmental programs by using IV. Comprehensiveness is measured as index score that is the sum of programs adopted by local governments in 2010 ICMA survey. In the second stage, I estimate the effect of comprehensiveness of the programs on environmental performance that is calculated as increase or decrease in the level of Ozone (O₃) from 2006 to 2010. Results from preliminary analysis imply that the comprehensiveness of non-regulatory environmental programs contributes to environmental performance. The result gives implication on how to design non-regulatory environmental programs that are closely related to the actual policy goal of improvement in environmental quality.