成长型市场视角：人力资本趋势1015.ppt

,R&Dspendingasa%ofGDP,2012 年 10 月 12 日成长型市场视角成长型市场视角：人力资本趋势证券研究报告,对成长型市场的人力资本趋势的调查作为推动经济中长期增长的重要（但并非唯一）因素，我们研究了成长型市场各经济体的人力资本趋势，调查内容涵盖从基础的识字比率、预期受教育年限、教,欧阳汉+65-6889-1199 高盛（新加坡）私人公司,育开支一直到专利数量和企业研发等指标。方超,基础教育结果不尽相同尽管韩国、巴西、南非和墨西哥的平均教育支出（在 GDP 中的占比）超过了经合组织国家，但从高等教育来看，仅有韩国属于发达国家水平。中国和俄罗斯的理科博士数量超越了其教育支出所隐含的水平。韩国和中国的研发和专利水平遥遥领先受私营部门推动，韩国的研发支出在 GDP 中的占比高于任何发达国家。中国和韩国的研发支出增长最为迅速，这可能推动了其高科技产品出口的增长。此外两国在美国和欧洲专利授予中的占比迅速增加，印度位居第三但差距较大；其它成长型市场几乎无力参与全球专利角逐。对研发支出和人均 GDP 的分析表明，韩国和台湾已经跳出了“中等收入陷阱”，在其它国家中，中国的情况相对较好，有望跳出“陷阱”。从股市来看，一国的大宗商品资源似乎与研发支出以及高科技行业的发展呈负相关关系。“中等收入陷阱”的另一解读,+65-6654-5395 高盛（新加坡）私人公司吕成,CFA+852-2978-6613 高盛（亚洲）有限责任公司Ramasubramanian Dharmaraj+91(80)6637-8592 高盛（印度）证券私人有限公司,4.0%,R=0.51,3.5%,Korea,Japan,3.0%,Taiwan,Germany,US,2.5%,2.0%1.5%,China,Russia,UK,FranceCanada,Australia,S.Africa,1.0%,Brazil,0.5%0.0%,India,Indonesia,TurkeyMexico,0,5,10,15,20,25,30,35,40,45,50,GDP per capita(US$),(thous),Note:Korea and Taiwan as of 2010;S.Africa as of 2008;India and Mexico as of 2007;others as of 2009资料来源:经合组织、世界银行、Haver、美国国家自然科学基金会、高盛全球经济、商品和策略研究高盛与其研究报告所分析的企业存在业务关系，并且继续寻求发展这些关系。因此，投资者应当考虑到本公司可能存在可能影响本报告客观性的利益冲突，不应视本报告为作出投资决策的唯一因素。有关分析师的申明和其他重要信息，见信息披露附录，或参阅 由非美国附属公司聘用的分析师不是美国 FINRA 的注册/合格研究分析师。高盛集团,Jan-12,Dec-11,Feb-12,Jun-12,Jul-12,May-12,Aug-12,Oct-11,Nov-11,Mar-12,Apr-12,Sep-12,Indonesia,China,Turkey,Mexico,Russia,Korea,Brazil,S.Africa,India,Oct-12,8,2,2012 年 10 月 12 日Growth Markets:economic and market dataExhibit 1:Market performance:equity,rates and currencyAs of Oct 10,2012,全球,Global marketsPrice return(US$,%),3000,GM performance chart*,Indices,1-month,Ytd,EMGSSTEM50GM*DMGSSTDM50,-0.4,-0.1,2.32.22.0,8.39.48.610.44.6,270024002100,-1,-0.5,0,0.5,1,1.5,2,2.5,Growth markets,Price return(US$,%),Currency(vs.USD),CY2012,10Y yield,MSCI IndicesIndia,1-month,9.8,Ytd21.4,Current53.06,1mo chg(%)4.5,PE(X)15.0,EPSgrowth(%)8.3,PB(X)2.6,ROE(%)16.9,Current(%)8.2,1 mochg(bp)-2,China offshore,5.4,6.3,9.8,3.1,1.5,15.7,MexicoIndonesiaKoreaChina domestic*TurkeyBrazil,-0.7,0.50.3,1.9,4.33.9,21.73.411.1-0.938.8-6.3,12.95961311156.281.822.04,1.1-0.31.30.9-0.8-0.9,18.815.19.510.911.211.6,35.511.017.87.99.0-11.7,2.83.51.21.61.71.4,15.424.711.715.415.810.7,5.35.92.93.58.49.6,-23-7-13316-7,RussiaS.Africa,-5.7-6,-4,-0.8-2,0,2,4,6,8,10,12,7.62.2,31.158.72,1.6-6.1,5.112.8,-7.314.7,0.62.2,13.918.2,3.07.2,-735,GSSTDM50=GS DM BRICs Nifty 50 basket;GSSTEM50=GS EM BRICs Nifty 50 basket*Growth market performance is based on aggregation of its corresponding MSCI country index market cap series;China domestic performance is based on CSI300 indexSource:MSCI,FactSet,Bloomberg,GS Global ECS Research,Exhibit 2:Macro economic indicators(GS est.)As of Oct 10,2012,Exhibit 3:MSCI country index return decomposition ofGMAs of Oct 10,2012,GDP(%yoy),CPI(%yoy),CA surplus/deficit(%GDP),(%)10,1-mo chg in NTM EPS1-mo FX return,1-mo chg in NTM P/E1-mo US$Price return,2012E,2013E,2012E,2013E,2012E,2013E,BrazilRussiaIndiaChinaMexicoIndonesiaKoreaTurkeyS.AfricaGMEMDM,1.64.15.77.63.76.02.63.22.65.65.61.4,4.04.37.08.04.06.43.55.53.46.46.31.5,5.35.17.22.84.24.52.49.25.74.44.72.1,4.96.45.73.03.75.72.89.05.04.34.62.0,-2.55.1-3.52.5-0.6-2.02.8-6.5-5.00.50.5-0.3,-2.64.1-3.72.3-1.20.51.4-8.0-4.50.20.3-0.6,6420-2-4-6-8,Note:Regional aggregations are based on nominal GDP PPP weighted,Source:GS Global ECS Research高盛全球经济、商品和策略研究,Source:MSCI,FactSet,GS Global ECS Research,3,2012 年 10 月 12 日,全球,Prologue:Trends in Growth Markets Human CapitalCurrent views:counting down to a potential 2013 takeoff for GM equitiesWeve just spent three weeks on the road,discussing the world and Growth Markets with a widevariety of clients in Europe,the US,and Latin America.For the last two years,the defensiveparts of GM have held up better than the cyclical(China,Russia,Brazil)parts of GM as emergingmarkets have largely disappointed investors as economic growth in GM has largely been reviseddown for the last 8 quarters(coincident with Chinas tightening cycle)and valuations havecompressed.Unsurprisingly,DM companies have been better about cutting costs and maintaining margins,while GM companies,presumably more focused on growth,have had weaker results as the topline has weakened.As we head towards 2013,and get past the Chinese leadership change,theUS election,and find some resolution to the US fiscal cliff,we are hopeful that GM growthestimates will stabilize and inflect,partly catalyzed by the expectation of more pro-active Chinesepolicy,thereby refocusing investors on top line growth,particularly in the cyclical part of GM.In this issue:a comparison of human capital in the growth marketsContinuing our comparative analyses across GM,in this issue we tackle the broad area of humancapital.We look at trends in education,intellectual property,patents,and royalties,and comparethem to the paths trodden by the developed economies.This topic supplements our previouswork on commodity drivers and linkages,because many of the primary balances in GM countriesare based on either commodity or human capital drivers.These two axes are naturally interlinked;the rise of China has largely been a human capital and manufacturing story but its growth hashad important effects on commodities markets,and countries with commodity stories havebecome important consumers for Chinese manufactured goods.From papers like Robert Lucas influential On the Mechanics of Economic Development(1988),human capital(H)is generally acknowledged to be one of the four inputs to economic growth,theother three being labor/demographics(L),financial capital(K),and total factor productivity(TFP,the residual).Since demographic trends are largely known(and hard to change),we examinehuman capital(the other side of the labor picture)since skill levels and the pace of improvementcan differ widely.The other subtext of our story is how quickly human capital is changing in many parts of thedeveloping world,and how competitive a place the world has become.Many espouse the ideathat it will take developing countries many decades to catch up to developed markets;in thisissue we explore some of those trends and see where growth markets stack up,both in currentcapabilities and rate of growth.In countries like China and Korea,where human capital is buildingrapidly especially at the college graduate level,we wonder whether the countries will be able tofully utilize this energy going forward.As we build some foundational pillars for addressing the question of the middle-income trap,something which is often cited as a reason that developing countries may never becomedeveloped,we begin to notice signs of Dutch disease and the resource curse,where adependence on commodities can crowd out the manufacturing sector via the exchange rate andterms of trade.高盛全球经济、商品和策略研究,4,2012 年 10 月 12 日,全球,Comparing education across GMIn Fortnightly Thoughts:Where countries succeed companies follow(Sept 20),human capital wasone of the factors examined.In this issue,we take a closer look at what we believe is one of thefundamental sources of economic energy for a country.We start with a montage of very basic education information:literary rates,school life expectancy,doctorate degrees awarded,and education spending.Although our analysis is quantitative,weunderstand there is an important qualitative dimension as well every dollar or year spent oneducation may produce very different outcomes because of teaching quality,area of focus(primary vs secondary vs tertiary),or income strata(as in some Latin American countries).Exhibit5,while imperfect,attempts to roughly show that expenditure and output may not be wellcorrelated.-Korea looks more like a DM than a growth market,with similar levels of educationalachievement as most of the OECD on a number of measures.Indeed,Korea may need to dealwith increasing education inflation,where the focus on educational achievement creates aworkforce overqualified for the jobs available.-In China,public spending on education is below the OECD average,but private spendingnarrows the gap considerably.China appears to punch well above its expenditure weight:Exhibit5 shows very high test scores(although only for wealthier Shanghai),China now mints more hardscience Ph.Ds annually than the US(Exhibit 6),and Chinese students in the US are the largestand fastest growing group(Exhibit 8).-Although the large fiscal deficit of India may hamper its ability to spend on public education,Indonesia spends about the same amount,Turkey spends even less.-Commodity wealthy countries like Brazil and Mexico spend more than the OECD average butthat doesnt appear to produce high average test scores,or many doctorates.Russia has arelatively high percentage of hard science doctorates,however(Exhibit 6).-Although there is a loose correlation between public education spending and per capita GDP(Exhibit 7),some countries exhibit greater mobility along the trend than others(China vs.SouthAfrica for instance).This graph does not take into account the effects of commodity wealth on percapita GDP.Related areas:international students in the US and brain drain/gainTo triangulate the quality dimension,we also looked at other related statistics,like the numberand proportion of students in the US(Exhibit 8).Again,Korea clearly punches well above itsweight,with over 8 times the propensity(0.17%vs.most other countries at 0.02%)to study in theUnited States,something which is clearly not shared by Koreas North Asian neighbors,evenChina(although Chinas CAGR is the fastest).And perhaps more importantly,the measure of brain drain/gain shown in Exhibit 9 suggests thatthe US is generally the big winner in attracting scientific talent(although this is for a few selectfields and does not include all of our countries).The point is an important one;if an Indianscientist is educated in the US and decides to stay there after getting his/her degree,then this willcreate GDP for the US,not India.So ultimately the ability to attract,absorb,and efficiently usehighly skilled people is also an important growth factor.高盛全球经济、商品和策略研究,MeanPISAscore,Indonesia,Japan,Taiwan,Germany,Turkey,France,Russia,Mexico,Korea,S.Africa,China,Brazil,India,US,UK,Educationspendingasa%ofGDP,Russia,China,Brazil,US,UK,Mexico,Germany,Taiwan,India*,Korea,Japan*,Turkey,Turkey,Russia,China,Korea,India,Indonesia,Germany,S.Africa,Mexico,France,UK,Japan,Brazil,UK(1.0),Russia(0.6),100,(thous),0,5,2012 年 10 月 12 日Exhibit 4:Literacy rates and school life expectancy,Exhibit 5:Education spending as a%of GDP,vsstandardized test scores,全球,18,School life expectancy(yrs),Literacy rate(rhs),100%,600,China(2.0),R=0.03,1614,90%80%,550,Japan(1.3),France(0.5),Korea(4.0),121086,70%60%50%40%30%,500450400,Germany(0.5)Turkey(0.9),US(1.6)OECD avg(0.9)Mexico(1.8)Indonesia(1.7),Brazil(2.9),4,20%,2,10%,350,0,0%,300,India*(1.4),(%),3,4,5,6,7,8,9,10,Education spending as a%of GDP,Note:For literacy rate:Brazil as of 2008;India as of 2007;S.Africa as of 2004;The rest as of2009;For school life expectancy:All data as of 2010Source:OECD,WorldBank,UNESCO,GS Global ECS ResearchExhibit 6:China produces more hard science Ph.Ds thanthe USEarned doctoral degrees,Note:Brackets=private education as a%of GDP;China as of 2011;Indonesia as of 2010;Turkey as of 2006;The rest as of 2009;PISA(OECD study)aims to evaluate educationsystems worldwide by assessing 15-year-olds competencies in the key subjects:reading,mathematics and science.China PISA score draw from students in Shanghai only.IndiaPISA score draw from students in two states.Source:OECD,WorldBank,UNESCO,Euromonitor,GS Global ECS ResearchExhibit 7:Public education spending as a%of GDP,vs.GDP per capita,(thous)7060,Hard scienceEngineeringOther doctoratesHard-sci.&Eng.doctorates per mn*,250200,(%)7.06.0,USBrazilMexico,JapanRussiaIndonesia,GermanyIndiaTurkey,KoreaChinaSouth Africa,50,403020100,150100500,5.04.03.02.01.00.0,(log scale),Note:India as of 2006;Japan as of 2007;others as of 2008;*based on urban populationSource:OECD,WorldBank,UNESCO,NSF,GS Global ECS ResearchExhibit 8:Korea has over 8 times students in the US interm of%of total urban populationInternational students in the US,1,000 10,000GDP per capita(US$)Source:OECD,WorldBank,UNESCO,GS Global ECS ResearchExhibit 9:Brain drain and gain,100,000,Total international students in the US(2010/11),Researchers who have left the country,Immigrants as a%of researchers,160,20.3%,%of total urban population(rhs),0.20%,40,20,0,20,40,60,0.18%,US,1208040,6.3%,4.4%-11.3%,0.16%0.14%0.12%0.10%0.08%0.06%0.04%0.02%0.00%,AustraliaSwedenSwitzerlandCanadaDenmarkUKFranceJapanNetherlandsGermanySpainBrazil,BelgiumItalyIndia,Note:numbers above the bars are the 5 year CAGR of international students in the USSource:Institute of International Education,GS Global ECS Research高盛全球经济、商品和策略研究,Note:ranked according to the difference between Immigrants as%of researchers andresearchers who have left the country;Results draw from a survey on 17,182 activeresearchers from four research areas,biology,chemistry,earth and environmental sciences,and materials across 16 countries during 2011Source:NBER,GS Global ECS Research,6,2012 年 10 月 12 日,全球,R itshows a reasonable correlation,and shows that high tech exports shares in DM are largelydeclining(which is also related to increases in DM consumption).From this admittedly basic chart,it would appear that Korea and China are very focused onmoving up the export value chain.Although this strategy has been largely successful,we wouldargue that other factors(terms of trade,government incentives,etc)must also play some part inthis development.R&D and per capital GDP signs of the middle-income trap?Exhibit 12 looks at the relationship over time between R&D spending and per capita GDP in bothEM and GM:a number of notable points emerge:-Higher R&D spending is associated with higher per capita GDP,although this is less true forcommodities countries,which tend to show both low sensitivities and R-2s(Exhibit 13)-The US doubled its relative R&D spending between 1953 and 1968,per capita GDP grewrapidly after that.Korea appears to have done something similar from 88 to 10.-Commodity countries(Russia,Brazil,Indonesia,South Africa)tend to have low levels of R&Dspending but per capita GDP has increased nonetheless.However,none of them has achievedper capita GDP over$15K/person so far.Interestingly,the two major DM commodity countries,Canada and Australia,spend more on R&D than the GM commodity countries,although both ofthem are at the low end of DM R&D spending.-Taiwan(in the last 20 years)has had high levels of R&D spending but relatively littleimprovement in per capita GDP,suggesting that R&D spending is not a panacea for growth.高盛全球经济、商品和策略研究,Bpchgofhigh-techexportsasa%of,GDP(2001-2010),Brazil*,India*,Russia,Indonesia,US,Korea*,France,UK,China,S.Africa*,Turkey,Japan,Germany,Mexico*,Taiwan*,R&Dspendingasa%ofGDP,R&Dspendingasa%ofGDP,Korea(2010),India(2005),US(2007),UK(2010),Turkey(2010),Russia(2010),Brazil(2008),Indonesia(2009),Germany(2010),S.Africa(2008),Mexico(2007),Taiwan(2009),France(2009),Ja