13 pages


of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  Energy consumption, output and trade in South America Perry Sadorsky ⁎ Schulich School of Business, York University, 4700 Keele Street, Toronto, Ontario, Canada M3J 1P3 a b s t r a c ta r t i c l e i n f o  Article history: Received 28 March 2011Received in revised form 15 December 2011Accepted 17 December 2011Available online 23 December 2011  JEL classi  󿬁 cation: Q43F14O54 Keywords: Energy consumptionSouth AmericaPanel cointegrationExport led growth This study uses panel cointegration regression techniques to examine the relationship between energy con-sumption, output and trade in a sample of 7 South American countries covering the period 1980 to 2007.Panel cointegration tests show a long-run relationship between 1) output, capital, labor, energy, and exportsand 2) output, capital, labor, energy, and imports. Short-run dynamics show a bi-directional feedback rela-tionship between energy consumption and exports, output and exports and output and imports. There is ev-idence of a one way short-run relationship from energy consumption to imports. In the long-run there isevidence of a causal relationship between trade (exports or imports) and energy consumption. These resultshave implications for energy policy and environmental policy. One important implication of these results isthat environmental policies designed to reduce energy use will reduce trade. This puts environmental policyaimed at reducing energy consumption at odds with trade policy.© 2011 Elsevier B.V. All rights reserved. 1. Introduction Economic output, trade and energy consumption tend to move to-gether across time and as countries around the world continue togrow and develop there is an interest in learning more about the dy-namic relationship between these variables. The relationship betweenenergy consumption and economic output is one of the most widelystudied areas in energy economics (e.g. Ozturk, 2010; Payne, 2010)while the relationship between exports and output is one of the mostwidely studied areas in international economics (e.g. Giles andWilliams, 2000a, 2000b). The relationship between energy consump-tion and trade is an important yet understudied area. Understandingtherelationship betweenenergyconsumption,tradeand outputis cru-cialtounderstandingcurrentenergyand environmentalpolicyandde-veloping new effective energy and environmental policy.The relationship between energy consumption and trade is an im-portant topic to study for several reasons. If energy consumption isfound to Granger cause trade, then any reductions in energy con-sumption, coming from say energy conservation polices designed toreduce greenhouse gas emissions, will reduce trade and lessen thebene 󿬁 ts of trade. Energy conservation policies which reduce energyconsumption will offset trade liberalization policies designed to pro-mote economic growth. This places energy reduction policies atodds with trade liberalization policies. If unidirectional Grangercausality is found to run from trade to energy consumption or noGranger causal relationship is found then energy reduction policieswill not affect trade liberalization policies designed to increase eco-nomic growth.At the time of writing, the papers by Lean and Smyth (2010a,2010b), Narayan and Smyth (2009), and Sadorsky (2011) appear to be the only published papers speci 󿬁 cally investigating the relation-ship between energy consumption and trade. In particular, Narayanand Smyth (2009)  󿬁 nd for a panel of six Middle Eastern countries(Iran,Israel,Kuwait,Oman,SaudiArabia,andSyria),short-runGrang-er causality running from electricity consumption to real GDP andfromincometoexports.Theyalso 󿬁 ndevidenceofalong-runGrangercausality relationship running from exports and electricity consump-tion to real income and from exports and real income to electricityconsumption. In two very similar papers, studying electricity genera-tion and consumption in Malaysia, Lean and Smyth (2010a)  󿬁 nd evi-dence of Granger causality running from electricity generation toexports while Lean and Smyth (2010b) do not  󿬁 nd any evidence of a Granger causal relationship between exports and electricity con-sumption. While Lean and Smyth (2010a, 2010b) and Narayan and Smyth(2009)focusonthe relationshipbetweenexports andelectric-ity, Sadorsky (2011) focuses on the more general relationship be-tween energy consumption and trade (measured using eitherexports or imports). Sadorsky (2011)  󿬁 nds that for a panel of MiddleEastern economies (Bahrain, Iran, Jordan, Oman, Qatar, Saudi Arabia,Syria, and United Arab Emirates) short-run dynamics show causalityrunningfromexportstoenergyconsumptionandafeedbackrelation-ship between imports and energy consumption. Energy Economics 34 (2012) 476 – 488 ⁎  Tel.: +1 416 736 5067; fax: +1 416 736 5687. E-mail address:$  –  see front matter © 2011 Elsevier B.V. All rights reserved.doi:10.1016/j.eneco.2011.12.008 Contents lists available at SciVerse ScienceDirect Energy Economics  journal homepage:  This present study contributes to the literature on energy con-sumption, output and trade in several ways. First, most studiesfocus on either the relationship between energy consumption andoutput or the relationship between output and trade. Energy con-sumption, output and trade tend to growth together across timeand it seems natural to include them together in a model. By usinga model that combines trade, output and energy consumption, a bet-ter understanding of the dynamic relationship between these vari-ables can be obtained. Whereas most previous papers only considerone trade variable, exports, this present paper includes productionfunction based models for both exports and imports. Second, it isalso the case that most previous studies focus on the relationshipbetween electricity and exports while this present study uses totalenergy consumption. This present paper, therefore, provides a morecomprehensive study on the relationship between energy consump-tion and trade. Third, the long-run relationship between output, cap-ital, labor, energy consumption and trade is estimated using panelcointegration techniques. Panel cointegration techniques have morepower over single equation techniques and should provide more reli-able estimates of the long-run relationship. Fourth, the short-run re-lationship between output, capital, labor, energy consumption andtrade is estimated using panel vector error correction models(VECMs). Panel approaches are desirable when the sample size of each cross-section is relatively short because combing differentcross-sections together in a panel increases the sample size and de-grees of freedom yielding more precise estimates than what wouldbe obtained for each cross-section individually. Causality tests frompanel VECMs are more reliable than causality tests from single equa-tion models. Fifth, this study analyzes the trade, energy consumptionrelationship for a panel of South American countries, a region of theworld experiencing rapid growth in trade and a region of the worldthat has recently created the Union of South American Nations. Thepurpose of the Union of South American Nations is to strengthen re-gional economic ties and to develop a major global trading blocklike the European Union. This is the  󿬁 rst panel study investigatingthe link between trade, output and energy consumption in SouthAmerica. The results from this paper are useful for understandingand developing energy and environmental policy in South America.The purpose of this paper is to investigate the dynamic relation-ship between energy consumption, output and trade for a panel of South American economies. Empirical models are estimated usingpanel cointegration regression techniques. The following sections of the paper set out the contextual material on trade and energy con-sumption in South America, the empirical model, data, empirical re-sults and discussion, policy analysis, and conclusions. 2. Energy consumption, economic growth and trade in South America South America is a major region of the world that over the past20 years has experienced rapid increases in trade, output, and energyconsumption butforwhichlittle is knownabout thedynamicinterac-tion between these variables. The importance of South America to theglobal economy is evident by its output growth and trade growth.Over the period 1990 – 2007, world real GDP grew at a compound av-erage annual rate of 3.3% (International Energy Agency, 2009, p 62).Over this same period of time, real GDP in the Middle East, Africa,and Latin America grew at average annual rates of 3.8%, 3.7%, and3.4% respectively. 1 Economic growth in the Latin America over thistime period was not as fast as in China (10.0%) or India (6.3%) but itwas higher than in OECD countries (2.5%). According to statisticsfrom the World Trade Organization (2009), world merchandiseexports grew at an average annual rate of 12% between 2000 and2008 (WTO, 2009, p. 8). Over this same period of time, merchandiseexports from South and Central America, Africa and the Middle Eastgrew at average annual rates of 15%, 18%, and 18% respectively indi-cating that exports in these regions grew, on average, faster thanthe world average. 2 While Narayan and Smyth (2009) and Sadorsky (2011) have investigated the relationship between trade and energyconsumption for Middle Eastern countries, and Lean and Smyth(2010a, 2010b) have investigated the relationship between tradeand energy consumption in Malaysia, there is very little knownabout the relationship between trade and energy consumption inSouth America.South American countries have recently created the Union of South American Nations. 3 The purpose of this Unionis to foster great-er political and economic integration between the member countriesand to also create a major global trading block to rival the EuropeanUnion. This Union is, in some ways being modeled after the EuropeanUnion. Economic development, energy and the environment are keyareas of interest to the Union but these initiatives are being pursuedseparately from each other. Whether this approach is likely to leadto undesirable or con 󿬂 icting outcomes from different policies canonly be answered empirically. If it is found, for example, that thereis no causal relationship between energy and exports, then energyconservation policies won't affect export promotion polices. If, on-the-other hand, it is found that there is a causal relationship from en-ergytoexports(asthispaper 󿬁 nds)thenenergyconservationpolicieswill affect trade policies. This puts trade promotion policies at oddswith energy conservation policies. This paper offers key insightsinto how energy consumption, trade and output interact and is there-fore useful for guiding economic, energy and environmental policywithin the Union. 3. Energy consumption, economic growth and trade This section presents the theory behind 1) the relationshipbetween energy consumption and economic growth, 2) the relation-ship between economic growth and trade and 3) the relationshipbetween energy consumption and trade. Empirical studies relatingto the South American context are summarized.  3.1. Energy consumption and economic growth Research looking into the relationship between energy consump-tion and GDP developed in response to the oil price shocks of the1970s and since then has grown into a huge literature (see for exam-ple the surveys by Ozturk (2010) and Payne (2010)). One popular approach to investigating this relationship is to test for Granger cau-sality between energy consumption and GDP and this leads to fourtestable hypotheses; 1) a Granger causal relationship from energyto GDP, 2) a Granger causal relationship from GDP to energy, 3) afeedback relationship between energy and GDP, and 4) no Grangercausal relationship between energy and GDP (neutrality). Assumingthat energy consumption and output are positively correlated, the 󿬁 nding of unidirectional causality from GDP to energy or the  󿬁 ndingof neutrality means that energy conservation policies can take placewithout harming economic growth. The 󿬁 nding of unidirectional cau-sality from energy to GDP or feedback between energy and GDPmeans that energy conservation policies that lower energy use willlower economic growth.A number of authors have studied the relationship between ener-gy consumption and GDP in South America. Nachane et al. (1988), in 1 According to the International Energy Agency (2009) Latin America includes coun-tries located in South America, Central America and the Caribbean. 2 Membership intheWTOisoftenconsideredessential foranycountryserious aboutincreasing their international trade possibilities. All major South American countriesare members of the WTO ( org6_e.htm). 3 ( P. Sadorsky / Energy Economics 34 (2012) 476  – 488  a 16 country study,  󿬁 nd unidirectional causality from commercialenergy consumption per capita to real GDP per capita for Argentinaand Chile and bidirectional causality between energy consumptionand output for Brazil, Colombia, and Venezuela. Murray and Nan(1996)  󿬁 nd evidence of unidirectional causality from real GDP toelectricity consumption for Colombia. Soytas and Sari (2003), in a12 country study of G7 and emerging markets,  󿬁 nd bidirectional cau-sality between energy consumption and GDP per capita in the case of Argentina. Cheng (1997)  󿬁 nds evidence of unidirectional causalityfrom energy consumption to real GDP for Brazil. Lee (2005), workingwith a panel of 18 developing countries which includes Argentina,Chile, Colombia, Peru, and Venezuela,  󿬁 nds unidirectional causalityfrom energy consumption to real GDP. Chontanawat et al. (2008) 󿬁 nd unidirectional causality from energy consumption per capita toreal GDP per capita for Chile, Colombia, and Uruguay; unidirectionalcausality from real GDP per capita to energy consumption per capitafor Bolivia, Paraguay, Peru, and Venezuela; bidirectional causality forArgentina and Brazil; and the absence of causality for Ecuador.MahadevanandAsafu-Adjaye(2007),ina studyofnetenergyexport-ing developing countries,  󿬁 nd evidence of bidirectional causality be-tween energy consumption per capita and real GDP per capita forArgentina and Venezuela. For a panel of 11 oil exporting countrieswhich include Ecuador and Venezuela, Mehrara (2007)  󿬁 nds unidi-rectional causality from real GDP per capita to commercial energyconsumption per capita. In a study of OPEC countries, Squalli (2007)provides evidence of unidirectional causality from electricity con-sumption per capital to real GDP per capita in Venezuela. In an 82country panel which includes Argentina, Bolivia, Brazil, Chile, Colom-bia, Ecuador, Paraguay, Peru, Uruguay, and Venezuela, Huang et al.(2008)  󿬁 nd for the low income panel the absence of causality be-tweenenergyconsumptionandrealGDPpercapitawhereasevidenceis found for causality from real GDP per capita to energy consumptionfor the middle and high income panels. Apergis and Payne (2010)study a panel of 9 South American countries (Argentina, Bolivia, Bra-zil, Chile, Ecuador, Paraguay, Peru, Uruguay, and Venezuela) and  󿬁 ndevidence of both short-run and long-run causality from energy con-sumption to economic growth.  3.2. Economic growth and trade There is an extensive literature looking into the relationship be-tween economic growth and trade (see for example the surveys byEdwards, 1998; Giles and Williams, 2000a, 2000b; Lewer and Vanden Berg, 2003). Much of this literature looks directly at the link be-tween exports and GDP. Two of the more interesting relationshipsto study are the export led growth (ELG) hypothesis and the growthled exports (GLE) hypothesis. There is also the possibility of a feed-back effect between economic growth and exports or the possibilityof no statistically signi 󿬁 cant relationship between exports andgrowth.There are several theoretical reasons to support the ELG hypothe-sis (Giles and Williams, 2000a). First, export growth may increase thedemand for a country's economic output and this generates an in-crease in real economic activity. Second, export expansion promotesspecialization in the production of exports which can lead to higherskilllevels,economiesofscaleandproductivitygainsfortheeconomyasa whole.Third,an increaseinexportscanprovideforeignexchangemakingiteasier toimportgoods,servicesandforeign 󿬁 nancialcapitalwhich can be used to help domestic capital formation. Fourth, exportgrowthisconsistentwithcomparativeadvantage — countriesspecial-izeinproducinggoodsthattheyhaveacomparativeadvantagein andthis increases economic growth.There are also theoretical reasons to support the GLE hypothesis(Giles and Williams, 2000a). Economic growth theory posits an ag-gregate production function that depends primarily on capital andlabor (like a Solow growth model). Over the long term, increasingeconomic growth depends upon increasing the capital to labor ratioand increasing technology to boost productivity (Weil, 2008). Eco-nomic growth leads to greater skills and technology which can leadto comparative advantage which facilitates the growth in exports.Neoclassical trade theory supports the idea that economic growthleads to greater exports.Granger causality between exports and GDP for Latin Americancountries has been examined by a number of authors. Bahmani-Oskooee et al. (1991) found that ELG hypothesis was supported forPeru while a bi-directional causal relationship was found for the Do-minican Republic and Paraguay. Van den Berg and Schmidt (1994)found a positive and statistically signi 󿬁 cant effect of exports on eco-nomic growth in Columbia and Peru but no signi 󿬁 cant effect wasfound for Argentina. Xu (1996) found support for the ELG hypothesisin Columbia but not in Argentina. Riezman et al. (1996) found sup-port for the ELG hypothesis in Costa Rica, Honduras, Suriname andUruguay. No signi 󿬁 cant relationship between exports and growthwere found for Argentina, Columbia and Peru. Awokuse (2008) 󿬁 nds little support for the ELG hypothesis in Argentina or Columbiabut does  󿬁 nd some evidence to support the ELG hypothesis in thecase of Peru.Importscan playanimportantrolein affectingdomesticeconomicactivity via the import led growth (ILG) hypothesis. Imported goodscanincreasedomesticoutputbyfacilitatingthetransferoftechnologyand factors of production into the domestic economy. Endogenousgrowth models show that imports can be a channel of economicgrowth because imports provide domestic  󿬁 rms with access to for-eigntechnologyandknowledge (Grossman andHelpman,1991). For-eign R&D knowledge can be an important source of productivitygrowth and foreign imports of technology intensive intermediate fac-tors of production implies that the imports can play a signi 󿬁 cant rolein economic growth. Imports can also affect economic growththroughtheef  󿬁 cienciesthatforeigncompetitionbringstothedomes-tic market place. Awokuse (2008) 󿬁 nds empirical evidence to supportthe ILG hypothesis for Argentina, Colombia and Peru.  3.3. Energy consumption and trade Theoretically, there are a number of reasons for how exports canaffect energy consumption. Export expansion increases the demandfor the factors of production (capital, labor, energy) used to makethe exports. Once exports are produced, machinery and equipmentmust be used to load and transport the exports to seaports, airportsor other docking stations where the exports are then of  󿬂 oaded andre-loaded for voyages abroad. The machinery and equipment usedin the production, processing and transportation of goods for exportrequire energy to operate. An increase in exports represents an in-crease in economic activity in export oriented sectors and this shouldincrease the demand for energy.It is also possible for changes in energy consumption to affectexports because energy is an important input into the production of goodsdestined for exports.The productionof exports requiresfactorsof production like capital, labor, and energy. A dramatic decrease inenergy consumption, resulting from say an energy conservation pro-gram, could affect the ability to produce goods for exports. It is alsopossible that a feedback relationship exists between energy and ex-ports whereby energy is important for explaining movements in ex-ports and exports are important for explaining movements inenergy demand. In this case, energy consumption and exports shareinterdependence and complementary effects. It is also possible forthe relationship between energy and exports to be neutral. In thiscase, the correlation between energy consumption and exports is sosmallthatitdoesnotshowupasa statisticallysigni 󿬁 cantrelationshipat conventional test levels.Imported goods can affect the demand for energy in two ways.First, imports are trade  󿬂 ows into a country and this require a 478  P. Sadorsky / Energy Economics 34 (2012) 476  – 488  well function transportation network to move goods around. Trans-portation requires energy and increases in trade  󿬂 ows are expectedto increase energy consumption. Second, the composition of im-ports can affect energy consumption especially if the imports areenergy intensive products like automobiles, dishwashers, air condi-tioners, etc. It is also possible that energy consumption can affectthe  󿬂 ow of imported goods especially if the imported goods aremachinery or equipment that requires energy to operate. Energyconservation policies or lack of accessible energy may reduce theusefulness and ef  󿬁 ciency of energy dependent imported goodsmaking it less likely that such good will be imported. There is alsothe possibility of a feedback relationship between imports andenergy or the possibility of no statistically signi 󿬁 cant relationshipbetween the two variables. 4. Empirical models Following Lean and Smyth (2010b), the relationship between en-ergy consumption, trade and output is modeled using a productionfunction. Whereas the model in Lean and Smyth (2010b) only in-cludes exports, the model in this present paper allows for exports orimports. 4,5 Output,  Y   can be written as a function of capital,  K  , labor, L , energy,  E  , trade openness,  O , and a country speci 󿬁 c variable,  V  . Y  it   ¼  f K  it  ; L it  ; E  it  O it  ; V  i ð Þ ð 1 Þ Eq. (1) can be parameterized as follows. Y  it   ¼  K  β  1 i it   L β  2 i it   E  β  3 i it   O β  4 i it   V  i  ð 2 Þ Taking naturallogarithms of Eq. (2), denotinglower case lettersasthe natural log of upper case letters and adding a random error termproduces the following equation.  y it   ¼  β  1 i k it   þ β  2 i l it   þ β  3 i e it   þ β  4 i o it   þ ν  i  þ ε  it   ð 3 Þ In Eq. (3), countries are denoted by the subscript  i  ( i =1, … ,N) andthe subscript  t   denotes the time period (t=1, … , T  ). Eq. (3) is a fairlygeneral speci 󿬁 cation which allows for individual  󿬁 xed country effects( υ ) and a stochastic error term ( ε  ).This paper uses panel cointegration techniques to investigate therelationship between energy consumption and trade in a sample of South American economies. Panel estimation techniques are attrac-tive because models estimated from cross-sections of time serieshave more degrees of freedom and ef  󿬁 ciency than models estimatedfrom individual time series. This is particularly useful if the time se-ries dimension of each cross-section is short. Panel cointegrationtechniques have recently been used by a number of authors to inves-tigate the relationship between energy consumption and output (e.g.Apergis and Payne, 2009, 2010; Chen et al., 2007; Lee, 2005; Lee andChang, 2008; Lee et al., 2008; Mahadevan and Asafu-Adjaye, 2007;Mehrara, 2007; Narayan and Smyth, 2008, 2009; Narayan et al.,2007; Sadorsky, 2009a, 2009b, 2011). Panel cointegration approacheshave drawbacks including which approach (within-dimension orbetween- dimension) to use in estimating the long-run cointegrationrelationship and how much heterogeneity to include in the panelVECM. 6 Estimation of Eq. (3) will provide estimates of long run elasticitiesbut in the context of panel estimation the ordinary least squares(OLS) estimator of Eq. (3) is asymptotically biasedand its distributiondepends upon nuisance parameters. The nuisance parameters areregressors that are not part of the true data generating process butcould introduce unwanted endogeneity and serial correlation(Pedroni, 2000, 2001). There are several possible approaches toaddressing the bias, including fully modi 󿬁 ed OLS (FMOLS) and dy-namic OLS (DOLS). FMOLS uses a non-parametric approach to correctfor endogeneity and serial correlation. DOLS uses a parametric ap-proach (adding leads and lags of the differences of the right handside variables) to correct for endogeneity and serial correlation. Insmall samples, the DOLS approach can use up a lot of degrees of freedom. 7 In addition, there is also a choice to be made regarding how thedata is pooled (Harris and Sollis, 2003). Pooling can be either alongthe within-dimension or between-dimension (sometimes called thegroup mean estimator). Within-dimension estimators assume acommon regression coef  󿬁 cient vector across all cross sections.Between-dimension (or group means) estimators assume the regres-sion coef  󿬁 cients vary across cross sections and this allows for greaterheterogeneity. Eq. (3) is written allowing for country speci 󿬁 c regres-sion coef  󿬁 cients and this is consistent with between-dimensionpooling.If evidence of cointegration is found, then the approach of  Engleand Granger (1987) can be used to estimate an error correctionmodel (ECM). The  󿬁 nding of cointegration between a set of variablesimplies that there exists causality, in the Granger (1969) sense, in atleast one direction. In this approach, the  󿬁 nding of cointegration be-tween a group of variables is very important because it ensures thatthere exists an error correction mechanism by which changes in thedependent variable are modeled as a function of the level of the dis-equilibrium in the cointegration relationship and changes in theother explanatory variables. The VECM for Eq. (3) can be written asfollows. Δ  y it   ¼  c  1 i  þ X q j ¼ 1 γ  11 ij Δ  y it  −  j  þ X q j ¼ 1 γ  12 ij Δ k it  −  j  þ X q j ¼ 1 γ  13 ij Δ l it  −  j þ X q j ¼ 1 γ  14 ij Δ e it  −  j  þ X q j ¼ 1 γ  15 ij Δ o it  −  j  þ γ  16 i ε  it  − 1  þ υ 1 it   ð 4a Þ Δ k it   ¼  c  2 i  þ X q j ¼ 1 γ  21 ij Δ  y it  −  j  þ X q j ¼ 1 γ  22 ij Δ k it  −  j  þ X q j ¼ 1 γ  23 ij Δ l it  −  j þ X q j ¼ 1 γ  24 ij Δ e it  −  j  þ X q j ¼ 1 γ  25 ij Δ o it  −  j  þ γ  26 i ε  it  − 1  þ υ 2 it   ð 4b Þ 4 Incorporating trade variables like exports and imports into the production functionhas a long history of usage (see for example the early papers by Balassa (1978) andSheehey (1992)). 5 In this model, exports or imports are included in the production function and twoseparate empirical speci 󿬁 cations estimated (one with exports, one with imports). In-cluding exports or imports separately reduces the multicollinearity from includingboth variables together. It is also possible to include a trade openness variable likenet exports divided by GDP but this was not tried. 6 The main drawbacks to panel cointegration are with respect to the choices that theresearcher needs to make. The long-run cointegration relationship can be estimatedassuming either a common regression coef  󿬁 cient vector across all cross sections orallowing the regression coef  󿬁 cients to vary across cross sections and this allows forgreater heterogeneity. Similarly, the VECM can be estimated with a common regres-sion coef  󿬁 cient vector across all cross sections or allowing the regression coef  󿬁 cientsto vary across cross sections. There are four possible ways to combine the long-runcointegration vector with the VECM. 7 This paper reports estimates from group-means OLS and FMOLS. Asymptotically,DOLS and FMOLS should produce similar parameter estimates although in small sam-ples it is not clear which approach works better. In practice, many researchers preferto use the FMOLS approach because it is easier to  󿬁 t on data sets with short time pe-riods and does not reduce the degrees of freedom the way parametric approaches likeDOLS do.479 P. Sadorsky / Energy Economics 34 (2012) 476  – 488
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!