本文以台灣1981年至2024年之資料為基礎，探討失業率變動對景氣波動之關聯是否存在性別差異，並建構性別化之歐肯法則分析架構。研究同時採用年度與季度資料，建立多階段實證設計：首先，透過性別虛擬變數與交互項檢驗性別差異；其次，納入景氣衰退虛擬變數分析不對稱關聯；再者，引入產業結構變數探討其與失業率變動之關聯；最後，針對2001年網路泡沫、2008至2009年全球金融海嘯及2020至2021年COVID-19疫情進行事件分析，以辨識不同衝擊類型下失業率變動與景氣波動關聯之性別異質性。 實證結果顯示，以1981年至2024年資料進行估算，台灣失業率變動與經濟成長率之間存在穩定的負向關係，女性失業率變動對GDP成長率之估計關聯係數約為−0.018，男性則約為−0.029，惟性別差異未達統計顯著。進一步納入產業結構後，衰退期交互項係數方向符合景氣不對稱之理論預期，顯示產業結構為不對稱關聯之間具有密切關係效果的重要傳導機制。事件分析結果則指出，以工業部門為主之經濟衝擊期間，男性失業率變動與景氣波動之關聯程度較女性明顯；其中金融海嘯期間男性歐肯係數之偏移幅度高於女性，而COVID-19期間之相關估計結果則未達統計顯著。此外，台灣於1990年代中期出現男性失業率高於女性之結構性轉折，反映產業轉型與人力資本變化對性別失業動態之關聯。 關鍵詞：歐肯法則、性別失業率、經濟成長率、景氣不對稱、產業結構;This study investigates whether there are gender differences in the relationship between unemployment fluctuations and business cycle fluctuations in Taiwan， based on data from 1981 to 2024. It constructs a gender-specific framework of Okun’s Law and employs both annual and quarterly data within a multi-stage empirical design. Specifically， gender differences are first examined using a gender dummy variable and interaction terms. Asymmetric relationships are then analyzed by incorporating a recession dummy variable. Next， industry structure variables are introduced to examine their association with unemployment fluctuations. Finally， an event analysis is conducted for the 2001 dot-com bubble， the 2008–2009 global financial crisis， and the 2020–2021 COVID-19 pandemic to identify gender heterogeneity under different types of economic shocks. The empirical results show a stable negative relationship between unemployment and economic growth in Taiwan. The estimated coefficient for female unemployment is approximately −0.018， while the corresponding coefficient for male unemployment is around −0.029; however， gender differences are not statistically significant. After controlling for industry structure， the recession interaction term becomes more negative， suggesting that industry composition is closely associated with asymmetric patterns in Okun’s Law. The event analysis further reveals that economic shocks concentrated in industrial sectors are more strongly associated with male unemployment， with a larger shift in the Okun coefficient for males during the global financial crisis. In contrast， the estimated effects associated with the COVID-19 pandemic are not statistically significant. Additionally， a structural turning point is identified in the mid-1990s， when male unemployment began to exceed female unemployment， reflecting the association between industrial restructuring， human capital changes， and gender-specific unemployment dynamics. Keywords: Okun’s Law; Gender Differences in Unemployment; Economic Growth; Business Cycle Asymmetry; Industry Structure
