影響台灣儲蓄率總體變數的探討

Macroeconomic Determinants of Saving Ratio in Taiwan

賴姿萍(Tzu-Ping Lai)

指導教授 ： 謝德宗

國立臺灣大學/社會科學院/經濟學研究所/碩士(2009年)

https://doi.org/10.6342/NTU.2009.02243

摘要

中文摘要
本文主要目的在探討儲蓄率與總體變數之間的關係；經濟持續的成長與發展是決策當局追求的目標，而儲蓄是推動經濟發展的重要因素，是以探討影響儲蓄率變化的關鍵因素是值得我們關心的議題。本文針對儲蓄率與各總體變數間的互動進行驗證，同時也探討彼此間的領先落後關係為何，以及影響台灣儲蓄率的總體變數因素為何。
研究架構上，先對儲蓄率與各種總體變數進行因果關係檢定，並以VAR模型來驗證各種的互動關係，然後再就儲蓄率與其他總體變數做迴歸分析。本研究利用單根檢定、相關性分析、Granger因果檢定、向量自我迴歸模型(VAR)等計量方法建立實證模型。
綜合本文實證結果可以歸納出下列結論：
一、總體變數與儲蓄率的互動關係
(1)在台灣的重要總體經濟變數中，與儲蓄率存在雙向因果關係者包括：領先指標、M1B、M2、CPI、GDP與失業率；另外，儲蓄率和匯率有單向因果關係，亦即匯率領先儲蓄率。

(2)在儲蓄率與各種總體變數間的領先落後關係中，以M2為變數的VAR模型結果，儲蓄率單向領先M2；GDP、失業率單向領先儲蓄率；至於儲蓄率與領先指標、CPI兩者間無明顯因果關係；而儲蓄率與匯率、股價指數兩者呈雙向因果關係；以M1B為變數的VAR模型結果，儲蓄率單向領先領先指標；M1B、CPI、匯率、失業率單向領先儲蓄率；至於儲蓄率與股價指數兩者間無明顯因果關係；而儲蓄率與GDP兩者呈雙向因果關係。發現以M1B與M2分別做總體變數間的領先落後分析，儲蓄率與其他變數領先落後關係不盡相同，惟兩者做出相同結果為失業率單向領先儲蓄率。

二、影響儲蓄率的總體變數因素
(1)儲蓄率受領先指標、M1B、M2、GDP、失業率以及季節性因素第二季與第三季的顯著正向影響；其中，M1B上升與M2上升將會促使儲蓄率會隨之上升，此是反映金融體系寬鬆將有助於提昇儲蓄率。至於GDP成長將促使儲蓄率隨之上升，此種現象反映經濟好轉，人們可支配所得提高或財富增加，消費習慣未立即改變，從而有助於提升儲蓄率。另外，失業率上升，儲蓄率會隨之上升，此種現象反映失業率上升人們會因為未來不確定性升高、缺乏安全感而增加儲蓄；儲蓄率會受季節性因素第二季與第三季的顯著正向影響，此種現象反映春節期間金融市場對貨幣需求增加，而於第二季再將未用完的資金回存。
(2) 儲蓄率受自我落後1期顯著負向影響，此種現象反映儲蓄率有落後性，人們落後1期才反映在儲蓄率上。
(3) 儲蓄率受季節性因素第一季顯著負向影響，此種現象反映春節期間金融市場對貨幣需求增加，導致儲蓄率下降。
(4) 儲蓄率和消費者物價指數、股價指數與匯率無顯著關係

關鍵字

儲蓄率 ； 經濟成長 ； 總體變數 ； 消費理論 ； VAR模型

並列摘要

Abstract
The main purpose of this paper is to explore the correlation between Saving Ratio and Macroeconomic Variables. The government aims to achieve sustainable economics growth and development. Savings is the main factor to stimulate economic growth, and the factor to affect saving ratio is also worthy of our concern. We attempt to examine the coefficient of savings ratio and all macroeconomic variables, investigate the reason of lead-lag relation and the reason to affect saving ratio of Taiwan and macroeconomic variables.

On the basis of research mechanism, first, examine the causation of saving ratio and macroeconomic variables. Second, verify the interaction by using VAR module. Lastly, examine affects on saving ratio and other macroeconomic variables by using regression analysis. The empirical model is derived from quantitative method such as Unit Root Test, Correlation Analysis, Granger Causality Test and Vector Auto-regression Model.

Following conclusions can be drawn from empirical result:
1.Interaction between macroeconomic variables and saving rate
(1)In Taiwan, the two-way causality of saving ratio among the macroeconomic variables includes: leading indicator, M1B, M2, CPI, GDP and unemployment rate; In addition, saving ratio is only one-way causality to exchange rates, namely, exchange rate takes the lead in saving ratio.
(2)In the lead-lag relation among saving ratio and other macroeconomic variables, the results of VAR model based on M2 variable are:
-Saving ratio one-way leads to M2
-GDP and unemployment rate one-way lead to saving ratio
-There is no apparent causality between saving ratio and leading indicator and CPI
-Saving ratio is two-way causality to exchange rates and stock index
The results of VAR model based on M1B variable are:
-Saving ratio one-way leads to leading indicator
-M1B, CPI, Exchange Rates and unemployment rate one-way leads to saving ratio
-There is no apparent causality between saving ratio and stock index
-Saving ratio is two-way causality to GDP
The results of lead-lag analysis of saving ratio and other variables based on M1B and M2 variable are not all the same. The only identical result is unemployment rate one-way leads to saving ratio.
2.Factors of Macroeconomics Variables affect Saving Ratio
(1)Saving ratio has a positive impact on leading indicator, M1B, M2, GDP,
unemployment rate and Q2 & Q3 seasonal factor. Among these factors,
-The growth of M1B and M2 will cause a rise in saving ratio. It reflects that loose monetary policy boosts saving ratio.
-the growth of GDP will also boost the higher saving rate. The phenomenon reflects the economic turnaround. More personal disposal income and consumption attitude remained will also boost saving ratio.
-Additionally, higher unemployment rate will cause higher saving rate. It shows the rise of uncertainty in future, the rise of saving while the unemployment rate rose.
-Saving ratio has a positive impact on Q2 and Q3 seasonal factor. It reflects higher demand on New Year period and the residual funds will deposit back in Q2.
(2)Saving ratio has a negative impact on self-delay. It shows that saving ratio
has time lags and one period of time delay reflects on saving ratio.
(3)Saving ratio has a negative impact on Q1 seasonal factor. It reflects higher
demand on New Year period and decrease in saving rate.
(4)There is no correlation between saving rate to CPI, Stock Index and exchange
Rate.

並列關鍵字

Saving Ratio ； Economic Growth ； Macroeconomic Variables ； The theory of Consumptio ； VAR Model

參考文獻

3.Carroll, C. D. & Samwick A, A., The nature of precautionary wealth.(NBER Working Paper, No.5193). Cambridge: National Bureau of Economic Research, 1995.

8.Dardanoni, V. Precautionary savings under income uncertainty: A cross-sectional analysis. Applied Economics, 23(1), 153-160, 1991.

9.Deaton, A. Understanding consumption. New York: Oxford university press, 1992.

10.Fershtman, C. and Weiss, Y.,”Social Status culture and Economic Performance” Economic Journal 103(419), pp.946-959, 1993.

12.Kazarosian, M. Precautionary saving-A panel study. Review of economics and statistics, 79(2), 241-247, 1994.