| 摘要: | 本研究使用雙重差分法(difference-in-difference, DID)分析基本工資調整時，對於不同族群實質薪資成長率的差異。主要探討不同性別之間的差異，另外也討論城鄉差距、公司規模、是否接受高等教育、不同年齡層以及不同行業別等群體間之薪資成長率。本文主要以行政院主計處1998至2016年的「人力運用擬-追蹤調查資料庫」(Manpower Utilization Quasi-Longitudinal Survey Database, MUS-QL)進行實證分析。與文獻較不同之處在於，本研究以分量迴歸的方式選取DID模型所需的控制組。文獻中大多採取薪資水準高於基本工資20%(或30%)的觀察樣本作為分析的控制組，似較為隨性(arbitrary)。基於基本工資調整可能對其他工資率的勞工薪資造成外溢的效果，因此，本文採用分量迴歸的方式選定薪資不受基本工資調整影響的百分位族群，作為DID分析的控制組。將薪資介於兩次基本工資之間者作為實驗組，即薪資直接受到基本工資影響的低薪族群。實證結果發現，
實證結果發現，相對於控制組，基本工資調高對於男性低薪族群薪資成長的影響效果介於-7.85%至-16.8%，女性則為-11%至0.88%。雖然基本工資對女性低薪族群的薪資成長無嚴重負面效果，但女性高薪族群在基本工資調整後，可能因為企業雇用成本提高而影響其就業機率，故基本工資調整的效果有一部分可能是反應在女性控制組就業機會的降低。此外，對於城市及鄉村的受雇勞工薪資成長率都有負向影響，但兩者估計係數不具顯著差異。對於公司規模而言，不論大小公司，基本工資政策對於薪資成長率雖然都有負面影響，二者亦無顯著差異。部分基本工資調高的年份，其對具高學歷但屬低薪的族群薪資成長的衝擊相對較大。由於受過高等教育的低薪族群有63%是未滿30歲的青少年族群，意味著青少年人即使是高學歷，仍可能存在頗明顯的低薪問題。基本工資的調高雖然提高了他們的薪資，但對後續的薪資成長速度反而是不利的。基本工資對15至29歲青少年以及30歲以上受雇者二群體薪資成長率的影響差異雖不顯著，但其調高對於不同年齡層的薪資成長幾乎都有顯著負向的影響，且近年來不斷提高基本工資所造成的負向影響有越來越嚴重的趨勢。最後，本研究發現不論是服務業或製造業，近年連續調整的基本工資顯著不利低薪族群的薪資成長率。金融海嘯之後，基本工資的調漲對於服務業在薪資成長率上的負向影響比製造業更嚴重些。綜合上述各種不同條件的低薪族群實證結果，本研究認為連續調高的基本工資對於台灣薪資成長的速度已經造成明顯負向的影響，且負向效果有逐年加重的趨勢。;In this paper we use an individual-level quasi-longitudinal database to study the impact of changes in legal minimum wages on labor market outcomes. We use difference-in-difference (DID) model to estimate the impact of minimum wage laws on wage growth in Taiwan. This study differs from the literature in that the control group for the DID model is determined by quantile regression. We analyze the minimum wage rate effect on wage growth for both gender and emphasize in particular on urban-rural disparity, company size, educational levels, age groups, and different industrial groups.
Compared to the control group, male low-income group is found to have a larger negative impact on their wage growth than the female low-wage groups. Although there is no serious negative effect on the wage growth rate for female low-wage groups, female high-wage groups may be affected by their employment opportunities after the national minimum wage (NMW) adjustment. Therefore, part of the effect of the NMW adjustment may be reflected in the reduction of employment opportunities for the female control group. Moreover, there is no difference between the impact on urban and rural areas, and among companies of different sizes; however, all of them got negative impacts by the rising NMW. For some years of NMW adjustment, highly educated low-income groups are of relatively larger negatively being shocked by the NMW adjustment on the salary growth than their less educated counterparts. Since about 63% of the low-income groups with higher education are people younger than 30 years old, it means that, obviously, even young people with highly educated may still influenced by the adjustment of NMW negatively. However, there is no difference in the wage growth for workers at various age groups, although all of them are negatively affected by the rising NMW. Overall the continuous adjustment of the NMW is significantly unfavorable for the wage growth rate of the low-income group, no matter in service or manufacturing sectors. After the financial crisis, the negative impact on service sector’s wage growth rate is more serious than that of the manufacturing sector. To sum up, we believe that the continuous increase in the NMW had a significant negative impact on Taiwan′s wage growth rate, and the negative effect has been increasing year by year. |
