影響政策學習效果因素之研究：臺灣地方政府成立騎警隊之分析

A Study on the Factors Affecting the Effect of policy learning: An Analysis of the Establishment of Police Mounted Unit in Taiwan Local Government

自 1990 年代起，各國及各政府間為解決政策問題或改善績 效而相互學習，經驗學習已成為政府機關政策規劃的重要基礎。 2003 年我國台北縣（現升格為新北市）仿效美、加等國成立了 騎警隊，其後高雄市、彰化縣及臺中市政府警察局相繼引進，然 而近一步檢視四個地方政府在成立騎警隊後，歷經時間的推移，卻是分別產生不同的政策結果。此狀況點出本文主要研究問題：影響地方政府政策學習的因素為何？又是哪些原因造成政策學習後的效果可以持續與否？藉由政策學習的理論架構，以及透過國內、外學者專家質性資料的蒐集與分析與討論，本文發現政府縣市首長及機關首長的意識型態在政策傳遞中占了極重要因素，而政策學習平臺之建立以儲存、分享及公開政策知識，有助於讓政策不因隨著選舉、調動、發展及轉任等因素而改變，並降低凌駕專業的意識型態因素。此外，彌補公務員所欠缺的行銷專業，用不同角度思維及經驗發展出的專業跨領域多元學習，也可以提供平臺機關首長作為決策基礎之資訊，進而產生政策適應性而能繼續延續。最後藉由吸取來自各領域不同之專業知識，以社會監督賦予課責，將有助於學習而來的政策有機會延續下去。

Since the 1990s, policy learning between governments to solve policy problems or improve performance has become a critical topic. In 2003, Taipei County (renamed New Taipei City in 2010) followed the policy of the mounted police in the United States and Canada to establish a mounted police force , extending the reach of police services to local tourism. The police departments in Kaohsiung City, Changhua County, and Taichung City also established mounted units to enhance their competitiveness. However, a closer look at the four local governments that established mounted police forces produced different policy outcomes over time. To understand this situation, this study first constructs an analytical framework from the discussion of policy learning theory and collects data using in-depth interviews with decision-makers and executives. To expand the breadth of the research, it refers to the success of the New York and Royal Canadian Mounted Police to identify key factors that trigger policy learning and affect the continuation of the policy. This article finds that in terms of the effect of policy learning, the influence of the knowledge of mayors of local government and commissioners of police departments play critical roles in the transmission of policies. Therefore, a national policy knowledge and learning platform can be established to store and share policy knowledge and continuous learning effects. This makes policy learning will not be changed due to such as elections, developments and transfers, and reduce practice ideological factors. In addition, due to the lack of marketing for civil servants, professional cross disciplinary and multi-disciplinary learning and the use of learning outcomes from different perspectives and experiences can also help provide information for the decision-makers, and enable policies to adapt to changing contexts such as the environment and politics. Finally, encouraging self-learning, ensuring accountability, with social supervision, and through positive competition to improve the effectiveness of policy learning from various fields, will increase the opportunity of the succession of learned policies.