財富分配與分配不均的三篇論文

此論文由三篇研究財富分配的文章組成。第一篇文章為推估台灣個人財富分配之研究; 第二篇與第三篇文章探討遺產繼承與遺產稅對財富分配不均度的影響; 其中, 第一篇為中文文章, 第二篇與第三篇則使用英文撰寫。第一篇文章使用台灣 2004–2014 年財政資訊中心的個人財產登錄資料與所得登錄資料, 建構台灣詳細的個人財富資料, 並以此計算吉尼係數、富有者財富占比等不均度指標。我們的結果顯示, 多數個人的財富主要由房地產組成, 而房地產亦是台灣 2004–2014 年間成長最為快速的資產。然而, 極端富有的財富階級資產以股票為主, 但股票的組成比例在 2008 年金融風暴後逐漸由房地產取代。我們發現台灣在此11 年間的財富分配不均度並沒有太大的惡化, 2008 年金融風暴在股票市場的衝擊造成富有者有較大的財富減損, 因此減緩了財富不均度的惡化。世代分配不均度的差異也可能是整體不均度並未嚴重惡化的原因。我們發現行政院國富統計採用的房地產價值計算方法會嚴重低估房地產價值。除此之外, 透過公司持有的股票價值(間接持股) 對於公司股東的個人財富有相當大的影響, 特別是對於非常富有的股東。第二篇文章使用第一篇文章所建構的個人財富資料並結合財政資訊中心的遺產稅資料, 研究繼承遺產對於財富分配不均度的影響。透過差異中之差異法 (differenceindifferences) 估計的結果顯示, 繼承遺產會導致繼承者相較於非繼承者具有離散程度較大的財富分配。不過繼承遺產卻會顯著降低財富分配的吉尼係數與前端富有者的財富占比。我們的發現與先前文獻中所提到的平均效果 (equalizing effect) 一致。然而, 我們也發現遺產的平均效果並非持續性的效果。該效果會在繼承後的二至三年後消失, 繼承者的財富分配不均度在長期反而有所惡化。遺產價格表現的異質性是造成平均效果長期被反轉的可能原因。第三篇文章使用與第二篇文章相似的資料, 透過台灣 2009 年突然的遺產稅減稅政策研究遺產稅對於財富分配不均度的影響。透過三重差異法 (triple difference) 估計的結果顯示, 無論是以分配的離散程度、吉尼係數或是極端富有者的財富占比來看, 遺產稅都具有顯著減緩不均度惡化的效果。我們的結果也顯示, 2009 年的遺產稅減稅政策實際上只有影響極富有的繼承者, 大多數的繼承者並沒有因減稅而增加財富。

關鍵字

財富分配；財富不均；國富統計；吉尼係數；繼承；遺產稅

並列摘要

This dissertation consists of three essays on wealth distribution and wealth inequality. The first essay provides a study on individual wealth distribution of Taiwan. The second and third essays analyze the impacts of inheritance and the estate tax on wealth inequality. The first essay attempts to construct individual wealth distribution of Taiwan by the new released administrative data provided by the Fiscal Information Agency (FIA), Ministry of Finance of Taiwan. We find that real estate accounts for the most part of individaul wealth and it also has fastest growth among all of assets between 2004 and 2014. Besides, there is a slight deterioration of wealth inequality between 2004 and 2014. We show that the value of houses and land in Statistics on National Wealth are seriously underestimated, and indirect holding of stocks affects wealth of the richest individuals a lot. The second essay utilizes the constructed wealth data in the first essay and the records of the estate tax to examine the effect of inheritance on wealth distribution both in the short and long run. Consistent with the findings in previous studies, we find that inheritance increases absolute wealth inequality by enlarging the width of wealth distribution, but reduces relative wealth inequality (top wealth shares or Gini coefficient) in the short run (i.e. 12 years after the inheritance event). However, we provide the first evidence that inheritance can increase relative wealth inequality in the long run (i.e. 35 years after the inheritance event). The heterogeneity in the quality of bequests between the rich and the poor could be a possible reason for our results. The third essay utilizes the similar data in the second essay and provides the distributional consequences of the estate tax in empirical economics. We use the triple difference estimation to investigate the effect of a sudden estate tax reduction on wealth inequality in Taiwan. Our results show that wealth distribution are longer right tailed with higher Gini coefficient and larger top wealth shares. Besides, we find that the impact is not significant in the upper middle wealth class. This article provides the evidence to support the argument that the estate tax reduction only benefits those who are in the top of wealth ditribution.