長期以來各國政府非常重視地下經濟 (shadow economy) 的發展，由於地下經濟具有隱蔽性及訊息揭露不充分等特性，不僅扭曲了經濟指標，更重要的是地下經濟的存在導致政府流失了大量的稅基及租稅收入，已影響一國長期的財政穩定與經濟發展。因此，探討地下經濟形成的原因及其影響以及地下經濟在經濟發展過程中所可能扮演的居中角色，是各國政府所必須重視的課題。在第二章，基於現存文獻在探討金融發展與經濟成長的互動關係中完全忽略了地下經濟所扮演的角色，因此我們利用共整合模型 (cointegration model) 試圖探討兩者之間的長、短期因果關係、相互解釋的能力以及模型配適度與預測能力是否因考量地下經濟而有所影響。第三章則基於既有文獻鮮少考量地下經濟規模的變動對於金融市場波動的影響，因此我們採用Chou (2005) 提出的自我迴歸條件變幅 (conditional autoregressive range, CARR) 模型來檢視地下經濟的變動如何影響我國股市波動率，並進一步探討我國金融發展的變動是否對股市波動率的影響產生不對稱效果。於第四章中我們嘗試運用非線性分析方法中的門檻模型 (threshold model) 對通貨膨脹與地下經濟的門檻效果進行實證研究，主要是探討租稅負擔、銀行發展、股市發展、政府支出、經濟成長、工資與薪資以及通貨膨脹等七個主要影響因素對於臺灣地下經濟規模變動的影響效果，是否在高、低通膨率的狀態下會發生結構性改變，據以驗證過去基於線性模型下所分析的結果。
Governments have been paying greater attention to the development of the shadow economy (SE) for a long period of time. Owing to SE’s hidden nature and its inefficient information disclosure, SE’s continual growth is distorting various economic indicators and most importantly causing erosion to a country’s tax base through loss of government revenue. This type of an economy can affect the financial stability of a country over the long term. Therefore, governments are strengthening their investigation of the determinants and effects of SE and exploring the roles played by SE during the process of economic development. Based on the existing literature that mostly explore the finance-growth relationship while completely ignoring the roles played by SE, in chapter two we employ a cointegration model to investigate whether or not the presence of this type of economy significantly alters Taiwan’s finance-growth nexus. Because the existing literature rarely takes into account the impact from changes to SE on volatility in the financial market, in chapter three we utilize the conditional autoregressive range (CARR) model proposed by Chou (2005) to examine how such changes influence volatility in Taiwan’s domestic stock market and further study the asymmetric response of volatility to changes in financial development. In chapter four we adopt Hansen’s (1996) threshold model to empirically investigate the existence of inflation threshold effects in the relationship between SE and the chosen causes of SE such as tax burden, banking development, stock market development, government expenditure, economic growth, wages and salary, and inflation. We also compare the empirical results using a non-linear model with the empirical results using a linear model.
