縣市長選舉對股市之影響

本文採用事件分析法，針對2005年起地方選舉制變革後，至2018年間之的六次縣市長、直轄市首長選舉結果，利用選前最後一次民調顯示執政黨席位的變化呈現有利與否，將樣本區分為高變化和低變化兩組，探討選舉結果對股市的影響。本文發現，不論選前最後民調對執政當局呈現有利或不利衝擊，選舉事件對投資人所產生之不確定性，終因選舉結果揭曉、不確定性降低，則選後事件期帶來顯著的正報酬，且整體事件期的報酬，高變化組明顯大於低變化組。此外，考慮選舉席位、選舉層級、選舉轄區範圍不同的選舉效應，結果發現僅有全國九合一大選的報酬明顯大於區域性的一般縣市長或直轄市選舉。最終檢驗九大產業的選舉報酬，除金融、水泥產業外，其餘七大產業股，在高變化組的選舉事件中，均產生顯著較高的累積異常報酬；而電子科技產業的報酬更超越大盤整體報酬率。此外，不論市值高低在高變化組下，均有較低變化組為高之報酬率。本文結果支持不確定訊息假說。

關鍵字

縣市長選舉；不確定訊息假說；累積平均異常報酬

並列摘要

We use the event study approach to examine the impact of six direct mayoral elections from 2005 to 2018 on the Taiwan stock market. Based on whether the pre-election polls show an increase on seats or not for the ruling party, we separate all sample into high-variation and low-variation subsamples. We find a positive abnormal return during the post-election windows for the whole subsamples. However, the cumulative abnormal returns (CAR) are greater for the high-uncertainty subsamples than for the low-uncertainty subsamples as the prediction of the Uncertain Information Hypothesis. Moreover, based on jurisdictional area of mayoral elections, we divide them into the national election and the regional election and find that the national-election subsamples have the greater CAR during the during the whole 30-day window than the regional-election subsamples. Additionally, according to various industries, high-tech stocks have a greater CAR during the pre-to-post-election period than the market. We also find that the high-uncertainty subsamples have greater CAR not only for low-market-value-firms but for high-market-value firms. These findings are consistent with the Uncertain Information Hypothesis, suggesting that investors in Taiwan stock market have stronger reaction to high uncertainty than to low uncertainty.