外文翻譯--fdi和貿(mào)易對(duì)經(jīng)濟(jì)增長(zhǎng)的影響：來(lái)自發(fā)展中國(guó)家的實(shí)證分析

1、<p><b>　　中文3415字</b></p><p>　　本科畢業(yè)論文外文翻譯</p><p>　　外文題目： Impact of foreign direct investment and trade on economic growth:evidence from developing countries &

2、lt;/p><p>　　出 處： American Journal of Agricultural Economics, Vol. 86 </p><p>　　Issue 3 </p><p>　　作 者： Makki, Shiva S.,Somwa

3、ru, Agapi </p><p><b>　　原 文：</b></p><p>　　IMPACT OF FOREIGN DIRECT INVESTMENT AND</p><p>　　TRADE ON ECONOMIC GROWTH: EVIDENCE</p><p>

4、;　　FROM DEVELOPING COUNTRIES</p><p>　　Foreign direct investment (FDI) and trade are often seen as important catalysts for economic growth in the developing countries. FDI is an important vehicle of technolog

5、y transfer from developed countries to developing countries. FDI also stimulates domestic investment and facilitates improvements in human capital and institutions in the host countries. International trade is also known

6、 to be an instrument of economic growth (Romer). Trade facilitates more efficient production of goods and servi</p><p>　　This article analyzes the role of FDI and trade in promoting economic growth across se

7、lected developing countries and the interaction among FDI, trade, and economic growth. We examine data from sixty-six developing countries over the last three decades. Our results suggest that FDI, trade, human capital,

8、and domestic investment are important sources of economic growth for developing countries. We find a strong positive interaction between FDI and trade in advancing economic growth. Our results a</p><p>　　Met

9、hodology and Data</p><p>　　Our econometric model is derived from a production function in which the level of a country’s productivity depends on FDI, trade, domestic investment, human capital, and initial gr

10、oss domestic product (GDP) per capita. The model is based on endogenous growth theory, in the tradition of Balasubramanyam, Salisu, and Sapsford and Borensztein, Gregorio, and Lee, where FDI contributes to economic growt

11、h directly through new technologies and other inputs as well as indirectly through improving human c</p><p>　　g=a+b1FDI+b2TRD+b3HC+b4K+b5G0</p><p>　　+c1FDITRD+c2FDIHC+c3FDIK （1）</

12、p><p>　　+d1TRI+d2TX+d3GC+e</p><p>　　where g is the per capita GDP growth rate; TRD, the trade (exports plus imports) of goods and services; HC, the stock of human capital; K, the domestic capital i

13、nvestment; G0, the initial GDP(initial stock); IRT, the inflation rate; TX, the tax on income, profits, and capital gains in the host country expressed as percentage of current revenue; and GC is government consumption.

14、The variables FDI, TRD, K, GC are measured as ratios to GDP. We also account for interaction of FDI with trade and dome</p><p>　　The stock of human capital in a host country is critical for absorbing foreign

15、 knowledge and an important determinant of whether potential spillovers will be realized. We postulate not only a positive relationship between FDI and the GDP growth rate but also a positive interaction between FDI and

16、human capital in advancing economic growth. The application of advanced technologies embodied in FDI requires a sufficient level of human capital. That is, the higher the level of human capital in a hos</p><p&

17、gt;　　Data for our analysis are obtained from the World Development Indicators (WDI) database. However, we limit our analysis to 1971 through 2000 because the flow of FDI to most developing countries began in 1970s. All v

18、ariables represent the average over the following decades: 1971–1980, 1981–1990, and 1991–2000.</p><p>　　We estimate a system of three equations, where the dependent variables are the mean values of per capi

19、ta GDP growth rates in each decade. We estimate the system of equations using the seemingly unrelated regression (SUR) method as well as instrumental variable (three-stage least squares, TSLS) approach. The SUR estimatio

20、n allows for different error variances in each equation and for correlation of these errors across equations, while the instrumental variable technique allows us to overcome pote</p><p>　　Empirical Results&l

21、t;/p><p>　　The purpose of our empirical investigation is to analyze the effects of FDI and trade on economic growth and to examine how FDI interacts with trade, human capital, and domestic investment in advanci

22、ng economic growth in developing countries. We control for preexisting economic conditions by including initial GDP as one of the explanatory variables. We also account for differences in macroeconomic policies in the ho

23、st countries by including variables, such as inflation rate, tax burden, and gove</p><p>　　Table 1 presents the econometric results. Regressions 1.1, 1.2, and 1.3, different variants of equation (1) above ar

24、e estimated using the SUR method. Regression 1.1 is our basic specification with explanatory variables of FDI, trade, human capital, domestic investment, and initial GDP. Regression 1.2 extends 1.1 to include interaction

25、 of FDI with trade, human capital, and domestic investment. Regression 1.3 builds on regression 1.2 by controlling for inflation rate, tax burden, and government co</p><p>　　Regression 1.1 reveals that FDI a

26、nd trade have a positive impact on economic growth after controlling for human capital, domestic investment, and initial income. The estimated coefficient for FDI is positive and statistically significant while the estim

27、ated coefficient for trade is not statistically significant. Since the coefficient of FDI is larger than the coefficient of trade, it indicates the differential impact of FDI in the host country’s economic growth. The co

28、efficient for human capital</p><p>　　capture their interaction effects on economic growth.In regression 1.2, the interaction of FDI and trade yields a positive and statistically significant coefficient while

29、 the effects of FDI and trade, by themselves, are positive but not statistically significant. Regression 1.2 also reveals that the FDI interacts positively with domestic investment in advancing economic growth. The estim

30、ated coefficient for domestic investment is positive and statistically significant at a confidence level of 90</p><p>　　The diverse experiences from developing countries suggest that FDI and trade, by themse

31、lves, may not guarantee economic growth. A country’s economic growth is also affected by its macroeconomic policies and institutional stability. Sound macroeconomic policies and institutional stability are necessary prec

32、onditions for FDI-driven growth to materialize. The estimated coefficients for the three policy variables—inflation rate, government consumption, and tax on income, profits, and capital gains—a</p><p>　　FDI

33、“Crowds-In” Domestic Investment</p><p>　　One of the important questions raised in the literature is whether FDI augments a host country’s capital investment or crowds out domestic investment. Even though not

34、 statistically significant, the positive interaction between FDI and domestic investment in regression 1.3 implies that domestic investment is unlikely to be crowded out in developing countries. To further strengthen our

35、 argument, we estimate the contribution of FDI to domestic investment after controlling for trade, human capital, </p><p>　　Endogeneity Problems</p><p>　　The correlation between FDI and growth r

36、ate could arise from an endogenous determination of FDI. That is, FDI, itself, may be influenced by innovations in the stochastic process governing growth rates (Borensztein, Gregorio, and Lee). For example, market refor

37、ms in host countries could increase both GDP growth rates and the inflow of FDI simultaneously. In this case, the presence of correlation between FDI and the country-specific error term would bias the estimated coefficie

38、nts. The endogeneity</p><p>　　本科畢業(yè)論文外文翻譯</p><p>　　外文題目： Impact of foreign direct investment and trade on economic growth:evidence from developing countries </p><p&

39、gt;　　出 處： American Journal of Agricultural Economics, Vol. 86 </p><p>　　Issue 3 </p><p>　　作 者： Makki, Shiva S.,Somwaru, Agapi

40、 </p><p><b>　　譯 文：</b></p><p>　　FDI和貿(mào)易對(duì)經(jīng)濟(jì)增長(zhǎng)的影響：來(lái)自發(fā)展中國(guó)家的實(shí)證分析</p><p><b>　　引言</b></p><p>　　在發(fā)展中國(guó)家FDI和貿(mào)易常常被看作刺激經(jīng)濟(jì)增長(zhǎng)的重要因素。FDI是技術(shù)從發(fā)達(dá)國(guó)

41、家傳遞到發(fā)展中國(guó)家的載體。FDI還可以刺激東道國(guó)的國(guó)內(nèi)投資，提高人力資本和制度的便利。國(guó)際貿(mào)易也一直是促進(jìn)經(jīng)濟(jì)增長(zhǎng)的工具（羅默）。貿(mào)易使得具有比較優(yōu)勢(shì)的國(guó)家生產(chǎn)的產(chǎn)品和服務(wù)更便利的在國(guó)家間流通。盡管過(guò)去的研究證明FDI和對(duì)外貿(mào)易對(duì)經(jīng)濟(jì)增長(zhǎng)有積極正相關(guān)的影響，但是這種影響的大小從國(guó)家整體來(lái)說(shuō)取決于人力資本的水平、國(guó)內(nèi)投資、基礎(chǔ)設(shè)施、宏觀經(jīng)濟(jì)穩(wěn)定性和貿(mào)易政策。本文主要延續(xù)之前的討論：關(guān)于FDI和貿(mào)易對(duì)經(jīng)濟(jì)增長(zhǎng)的作用，經(jīng)濟(jì)和制度的發(fā)展在刺激F

42、DI與貿(mào)易上的重要性。</p><p>　　本文分析了FDI和貿(mào)易在經(jīng)濟(jì)增長(zhǎng)中扮演的角色，通過(guò)選擇66個(gè)發(fā)展中國(guó)家在最近三十年的數(shù)據(jù)分析FDI、貿(mào)易、經(jīng)濟(jì)增長(zhǎng)之間的關(guān)系。實(shí)證研究結(jié)果表明FDI、貿(mào)易、人力資本、國(guó)內(nèi)投資是發(fā)展中國(guó)家經(jīng)濟(jì)增長(zhǎng)的重要因素。結(jié)果還表明在推進(jìn)經(jīng)濟(jì)增長(zhǎng)方面FDI與投資之間存在積極正相關(guān)。同樣，F(xiàn)DI刺激國(guó)內(nèi)投資，但是FDI對(duì)經(jīng)濟(jì)增長(zhǎng)的貢獻(xiàn)可以由通FDI存在正相關(guān)性的人力資本、穩(wěn)定的宏觀經(jīng)濟(jì)政

43、策來(lái)提高。</p><p><b>　　研究方法和數(shù)據(jù)</b></p><p>　　本文采用的經(jīng)濟(jì)模型來(lái)源于生產(chǎn)函數(shù)，一個(gè)國(guó)家的生產(chǎn)力水平取決于FDI、貿(mào)易、國(guó)內(nèi)投資、人力資本、最初的人均GDP。這個(gè)模型建立在內(nèi)生增長(zhǎng)理論上，在傳統(tǒng)上Balasubramanyam, Salisu, and Sapsford and Borensztein, Gregorio, and

44、 Lee 的觀點(diǎn)，F(xiàn)DI對(duì)經(jīng)濟(jì)增長(zhǎng)的影響直接通過(guò)新技術(shù)和其他要素投入或者通過(guò)人力資本、基礎(chǔ)設(shè)施、制度的提高。為了實(shí)證分析FDI和貿(mào)易對(duì)經(jīng)濟(jì)增長(zhǎng)的作用，本文建立如下的方程：</p><p>　　g=a+b1FDI+b2TRD+b3HC+b4K+b5G0</p><p>　　+c1FDITRD+c2FDIHC+c3FDIK （1）</p><p>

45、;　　+d1TRI+d2TX+d3GC+e</p><p>　　g代表人均GDP增長(zhǎng)率；TRD代表貨物和服務(wù)的貿(mào)易（出口和進(jìn)口）；HC表示人力資本存貨；K表示國(guó)內(nèi)資本投資；G0表示初期GDP；IRT表示通貨膨脹率；TX是以當(dāng)前收入的百分?jǐn)?shù)，表示東道國(guó)的稅收收入、利潤(rùn)、資本收益。GC表示政府消費(fèi)。變量FDI、TRD、K、GC以比率表示GDP。本文還說(shuō)明了FDI對(duì)貿(mào)易、國(guó)內(nèi)投資、人力資本的關(guān)系。以前的實(shí)證分析證明國(guó)內(nèi)

46、投資對(duì)發(fā)展中國(guó)家的經(jīng)濟(jì)增長(zhǎng)有積極影響。我們預(yù)測(cè)這些變量的估計(jì)系數(shù)是正數(shù)，同樣預(yù)測(cè)FDI和貿(mào)易、FDI和國(guó)內(nèi)投資在促進(jìn)經(jīng)濟(jì)增長(zhǎng)方面有積極正相關(guān)作用。</p><p>　　東道國(guó)的人力資本存量對(duì)吸收國(guó)外知識(shí)和實(shí)現(xiàn)潛在的FDI技術(shù)外溢至關(guān)重要。我們假定不僅FDI和GDP增長(zhǎng)率存在積極正相關(guān)，而且在促進(jìn)經(jīng)濟(jì)增長(zhǎng)方面FDI和人力資本存在正互相關(guān)系。由FDI帶來(lái)的先進(jìn)技術(shù)的運(yùn)用需要足夠的人力資本，即東道國(guó)的人力資本水平越高，

47、FDI對(duì)經(jīng)濟(jì)增長(zhǎng)帶來(lái)的效應(yīng)越高。關(guān)于FDI和經(jīng)濟(jì)增長(zhǎng)之間的關(guān)系有一個(gè)重要問(wèn)題：FDI和國(guó)內(nèi)投資存在什么關(guān)系？一直以來(lái)，F(xiàn)DI被作為資本、技術(shù)、知識(shí)轉(zhuǎn)移到東道國(guó)的重要載體，因此帶來(lái)刺激經(jīng)濟(jì)快速增長(zhǎng)的機(jī)會(huì)。然而，實(shí)際上，F(xiàn)DI刺激增長(zhǎng)的作用取決于東道國(guó)吸收資本的能力，還存在FDI是否“擠出”國(guó)內(nèi)投資的問(wèn)題。因此，這個(gè)問(wèn)題變成了：FDI在多大程度上替代或補(bǔ)充國(guó)內(nèi)投資？在本文的實(shí)證模型中將FDI和國(guó)內(nèi)投資分開(kāi)，又將FDI和國(guó)內(nèi)投資作為術(shù)語(yǔ)交匯（

48、FDIK）來(lái)研究。FDIK的相關(guān)系數(shù)是正的話說(shuō)明FDI和國(guó)內(nèi)投資在促進(jìn)經(jīng)濟(jì)增長(zhǎng)時(shí)相互補(bǔ)充或加強(qiáng)。為控制東道國(guó)的現(xiàn)存的經(jīng)濟(jì)和制度條件以不變價(jià)格計(jì)量的初期GDP。我們預(yù)期以對(duì)數(shù)形式表示的初期GDP對(duì)GDP增長(zhǎng)率表現(xiàn)負(fù)相關(guān)。通貨膨脹率是一個(gè)國(guó)家財(cái)政和貨幣政策的重要指標(biāo)。低的通貨膨脹率意味著好的投資、貿(mào)易環(huán)境，因此有利</p><p>　　本文的數(shù)據(jù)來(lái)源于世界發(fā)展指標(biāo)數(shù)據(jù)庫(kù)（WDI）。本文之所以選擇66個(gè)發(fā)展中國(guó)家197

49、1年到2000的數(shù)據(jù)時(shí)因?yàn)榇蟛糠职l(fā)展中國(guó)家的FDI流入始于1970s。所有的變量的數(shù)據(jù)都是以下三個(gè)階段：1971–1980, 1981–1990, 1991–2000。本文運(yùn)用三個(gè)等式組成的系統(tǒng)，因變量是每十年的人均GDP增長(zhǎng)率。我們運(yùn)用無(wú)相關(guān)回歸（SUR）及工具變量（三步最小二乘法）來(lái)估計(jì)等式系統(tǒng)模型。SUR方法考慮到三個(gè)方程的不同誤差變量及誤差變量與等式的相關(guān)性，工具變量可以允許克服FDI和經(jīng)濟(jì)增長(zhǎng)之間的內(nèi)生性問(wèn)題引起的潛在偏差。&

50、lt;/p><p><b>　　三、實(shí)證檢驗(yàn)結(jié)果</b></p><p>　　本文的目的是實(shí)證分析FDI和貿(mào)易對(duì)經(jīng)濟(jì)增長(zhǎng)的效應(yīng)，檢驗(yàn)發(fā)展中國(guó)家FDI如何通過(guò)貿(mào)易、人力資本、國(guó)內(nèi)投資在經(jīng)濟(jì)增長(zhǎng)中發(fā)揮作用。把初期GDP作為其中一個(gè)解釋變量控制現(xiàn)存經(jīng)濟(jì)條件。通過(guò)加入通貨膨脹率、稅收負(fù)擔(dān)、政府消費(fèi)三個(gè)變量考慮東道國(guó)不同宏觀經(jīng)濟(jì)政策。</p><p>　　表

51、1表示實(shí)證分析的結(jié)果，回歸1.1,1.2,1.3是等式（1）用SUR方法估計(jì)的變形?；貧w1.1是解釋變量為FDI、貿(mào)易、人力資本、國(guó)內(nèi)投資、初期GDP的基本說(shuō)明。回歸1.2擴(kuò)充了回歸1.1，把FDI對(duì)貿(mào)易、人力資本、國(guó)內(nèi)投資、初期GDP的關(guān)系包括在內(nèi)?；貧w1.3通過(guò)控制通貨膨脹率、稅收負(fù)擔(dān)、政府消費(fèi)建立在回歸1.2的基礎(chǔ)上。實(shí)證分析的結(jié)果表明大多數(shù)回歸系數(shù)符合預(yù)期，特別是回歸1.3的說(shuō)明?？蓻Q系數(shù)R2雖比一般低但是對(duì)截面數(shù)據(jù)的分析是合理

52、的。</p><p>　　回歸1.1說(shuō)明了FDI和貿(mào)易在控制人力資本、國(guó)內(nèi)投資、初期收入后對(duì)經(jīng)濟(jì)增長(zhǎng)有正相關(guān)性。FDI的估計(jì)系數(shù)是正的并且在統(tǒng)計(jì)上顯著，但是貿(mào)易的估計(jì)系數(shù)在統(tǒng)計(jì)上不顯著。因?yàn)镕DI的系數(shù)比貿(mào)易的系數(shù)大，說(shuō)明兩者對(duì)經(jīng)濟(jì)增長(zhǎng)的不同作用。人力資本的系數(shù)是正的，說(shuō)明人力資本對(duì)經(jīng)濟(jì)增長(zhǎng)正相關(guān)（只在置信水平88%下）。國(guó)內(nèi)投資和初期收入的系數(shù)在統(tǒng)計(jì)上不顯著。把FDI和貿(mào)易、FDI和人力資本、FDI和國(guó)內(nèi)投資的

53、關(guān)系包括在內(nèi)不僅在整體上使估計(jì)偏高，還使得我們采集到他們對(duì)經(jīng)濟(jì)增長(zhǎng)的影響。在回歸2中，F(xiàn)DITRD的系數(shù)是正的而且統(tǒng)計(jì)上顯著，但是就FDI和貿(mào)易單獨(dú)來(lái)說(shuō)，系數(shù)雖是正數(shù)但統(tǒng)計(jì)上不顯著?；貧w2還說(shuō)明在促進(jìn)經(jīng)濟(jì)增長(zhǎng)過(guò)程中FDI對(duì)國(guó)內(nèi)投資正相關(guān)。在置信水平90%下，國(guó)內(nèi)投資的估計(jì)系數(shù)是正數(shù)，統(tǒng)計(jì)上顯著。這表明東道國(guó)不僅受益于FDI本身還受益于FDI對(duì)貿(mào)易、國(guó)內(nèi)投資的正相關(guān)作用。FDI和人力資本之間雖正相關(guān)但在統(tǒng)計(jì)上不顯著?；貧w3為了控制宏觀經(jīng)濟(jì)

54、政策和制度穩(wěn)定性加了另一些變量，這些變量對(duì)FDI、貿(mào)易甚至經(jīng)濟(jì)增長(zhǎng)都有重要影響。最近的研究表明東道國(guó)的貨幣政策、財(cái)政政策、開(kāi)放政策都對(duì)FDI影響很大。所以在回歸3中加入通</p><p>　　回歸3的結(jié)果表明FDI和貿(mào)易促進(jìn)經(jīng)濟(jì)增長(zhǎng)，但是估計(jì)系數(shù)統(tǒng)計(jì)上不顯著。另一方面，人力資本和國(guó)內(nèi)投資的存貨，有正向的統(tǒng)計(jì)上顯著的相關(guān)系數(shù)。這一結(jié)果也表明FDI對(duì)貿(mào)易，人力資本和國(guó)內(nèi)投資有正向的關(guān)系。但是只有FDI對(duì)貿(mào)易的相互作用

55、是具有顯著的統(tǒng)計(jì)意義的。這意味著FDI和貿(mào)易促進(jìn)在發(fā)展中國(guó)家收入快速增長(zhǎng)率中是互補(bǔ)的。這結(jié)果與FDI 流入帶來(lái)先進(jìn)技術(shù)并且與東道國(guó)貿(mào)易相互作用提高東道國(guó)的經(jīng)濟(jì)增長(zhǎng)率的觀點(diǎn)是相一致的。</p><p>　　來(lái)自發(fā)展中國(guó)家的多樣化的實(shí)證檢驗(yàn)表明外商直接投資和貿(mào)易自己本身并不能保證經(jīng)濟(jì)增長(zhǎng)。一國(guó)的經(jīng)濟(jì)增長(zhǎng)也受該國(guó)的宏觀經(jīng)濟(jì)政策和制度穩(wěn)定性的影響。良好的宏觀經(jīng)濟(jì)政策和制度的穩(wěn)定性是使以FDI驅(qū)動(dòng)的增長(zhǎng)能夠?qū)崿F(xiàn)的必要的先決

56、條件。關(guān)于三個(gè)政策變量的估計(jì)系數(shù)--通貨膨脹率，政府消費(fèi)和以收入、利潤(rùn)和資本獲得的稅收--都是負(fù)相關(guān)的而且在統(tǒng)計(jì)上顯著。這意味著降低通貨膨脹率、稅收負(fù)擔(dān)和政府消費(fèi)能夠促進(jìn)經(jīng)濟(jì)發(fā)展。低通貨膨脹率表明東道國(guó)的宏觀經(jīng)濟(jì)政策穩(wěn)定而有序。低稅負(fù)能夠使外國(guó)及國(guó)內(nèi)投資更具贏利性?？s減政府消費(fèi)能夠使更多的資金用以投資。</p><p>　　四、FDI“擠入”國(guó)內(nèi)投資</p><p>　　在文獻(xiàn)中提出了一個(gè)

57、重要問(wèn)題：FDI對(duì)東道國(guó)的國(guó)內(nèi)投資是擠入還是擠出。在回歸3中FDI對(duì)國(guó)內(nèi)投資的系數(shù)盡管是正的但不統(tǒng)計(jì)顯著，這不能退出在發(fā)展中國(guó)家國(guó)內(nèi)投資沒(méi)有被FDI擠出。為了深入實(shí)證分析，在控制貿(mào)易、人力資本、初期收入水平及表示各種宏觀經(jīng)濟(jì)政策的變量后，我們運(yùn)用SUR方法估計(jì)了FDI對(duì)國(guó)內(nèi)投資的貢獻(xiàn)。表1表示了回歸分析1.4的結(jié)果。因?yàn)楣烙?jì)系數(shù)是正數(shù)、統(tǒng)計(jì)顯著說(shuō)明FDI對(duì)國(guó)內(nèi)投資有正相關(guān)作用。正的相關(guān)性說(shuō)明FDI刺激或擠入國(guó)內(nèi)投資，這個(gè)結(jié)論與Boren

58、sztein, Gregorio, 和Lee的研究相符。盡管TRD的估計(jì)系數(shù)統(tǒng)計(jì)上不顯著，但是貿(mào)易正作用于FDI再影響國(guó)內(nèi)投資。在置信水平90%下，F(xiàn)DITrade的估計(jì)系數(shù)是正數(shù)且有意義。</p><p><b>　　五、內(nèi)生性問(wèn)題</b></p><p>　　FDI和增長(zhǎng)率的相關(guān)系數(shù)由FDI的內(nèi)生決定性，這就是說(shuō)FDI本身會(huì)受到在隨機(jī)控制增長(zhǎng)率的過(guò)程中產(chǎn)生的創(chuàng)新（

59、Borensztein, Gregorio, and Lee）。例如，東道國(guó)的市場(chǎng)改革可以同時(shí)提高GDP增長(zhǎng)率和FDI流入量。在這個(gè)例子中，外商直接投資和東道國(guó)相關(guān)性誤差項(xiàng)的存在會(huì)使得估計(jì)系數(shù)有偏差。內(nèi)生性問(wèn)題可以用工具性變量來(lái)闡述。關(guān)于工具變量估計(jì)方法的其中一個(gè)重要問(wèn)題是識(shí)別與FDI（或貿(mào)易）高度相關(guān)但是與誤差項(xiàng)無(wú)關(guān)的工具變量。本文在三步最小二乘法估計(jì)中用滯后的FDI值、貿(mào)易的滯后值、以對(duì)數(shù)形式表示的GDP總值作為工具變量。TSLS估