China’s OFDI Finance in Comparative Perspective Pt 1

June    2014   Paper  6   Exporting  National  Champions:   China’s  OFDI  Finance  in  Comparative  Perspective   Abstract   Scholars  have  compared  China’s  liberalization,  inward  FDI  attraction,  and  export   promotion  policies  to  those  of  its  “Asian  Miracle”  predecessors  to  assess  China  as  a   ‘developmental  state.’  We  build  on  that  literature  by  drawing  a  new  but  similar   comparison:  the  extent  to  which  Chinese  development  banks  have  financed  the   globalization  of  China’s  ‘national  champion’  firms.    We  focus  on  the  role  of  state   finance  in  promoting  China’s  outward  foreign  direct  investment  (OFDI)  in   comparative  perspective.  In  order  to  answer  this  research  question,  we  created  a   database  of  Chinese  finance  for  OFDI  and  compared  our  results  to  the  existing   literature  on  previous  developmental  states.    We  estimate  the  total  value  of  China’s   OFDI  finance  from  2002m 2012  at  $140  billion.  As  a  percentage  of  total  OFDI,   China’s  lending  is  roughly  three  times  55%  higher  than  Japan,  the  previous  global   leader  in  OFDI  finance.  Like  Japan  and  South  Korea  at  earlier  developmental  stages,   China’s  lending  also  goes  overwhelmingly  toward  natural  resource  acquisition,   though  to  a  much  greater  degree.  Unlike  Japan  or  Korea,  we  find  that  China’s   market  entry  has  more  to  do  with  developing  project  expertise  and  supporting   exports  than  it  does  with  tariffm hopping  or  outsourcing  industries  that  are  fading   on  the  mainland.  We  identify  two  major  reasons  for  China’s  high  (31%)  ratio  of   OFDI  lending  to  total  OFDI.  First,  China  has  a  greater  incentive  to  give  OFDI  loans   than  Japan  or  Korea  ever  did  because  its  borrowers  are  statem owned  so  it  can   more  easily  dictate  how  they  use  the  money.  Second,  China  has  a  greater  capacity  to   give  OFDI  loans  because  it  has  significantly  higher  savings  and  foreign  exchange   reserves  than  Japan  and  Korea,  both  today  and  especially  during  equivalent   developmental  stages.     Introduction   Numerous  studies  compare  the  policies  behind  China’s  rise  to  those  of  its  “Asian   Miracle”  predecessors,  including  liberalization,  inward  FDI  attraction  and  export   promotion.  In  this  paper,  we  focus  on  the  less  researched  tool  of  outward  foreign   direct  investment  (OFDI)  promotion  by  state  development  banks.  Starting  in  the   1950s,  Korean  and  Japanese  policy  banks  gave  billions  of  dollars  in  loans  to   “national  champion”  companies  in  order  to  encourage  overseas  investment.  In  2003,   Solís  correctly  labeled  Japan  the  “undisputed  leader”  of  OFDI  lending  (Solís  2003b   153).  Today,  Chinese  state-­‐owned  banks  are  loaning  billions  to  state-­‐owned   companies  to  invest  abroad  as  part  of  the  government’s  ‘Go  Global’  policy.  This   paper  compares  China’s  OFDI  lending  to  that  of  Japan  and  Korea  today  and  during   their  ‘developmental’  periods.     We  also  consider  the  motives  for  the  Chinese  loans  and  compare  them  to  Japan  and   Korea.  Scholars  have  suggested  that  in  the  case  of  Japan,  the  loans  were  designed  to   acquire  primary  resources,  gain  access  to  advanced  technology,  enter  foreign   markets,  and  outsource  declining  industries.  The  Chinese  government  and   numerous  scholars  agree  that  China  encourages  OFDI  mainly  for  the  three  former   motives.  These  scholars  base  their  arguments  about  the  government’s  motives  on   OFDI  figures,  which  represent  the  priorities  of  China’s  diverse  companies,  not  the   government’s  intentions.  We  consider  the  state’s  own  OFDI  lending  to  be  a  better   indicator  of  its  motives.  Using  our  OFDI  loan  database,  we  evaluate  both  the  sectors   and  motives  behind  China’s  OFDI  lending.     In  the  sixty  years  since  Japan  began  a  period  of  unprecedented  economic  expansion,   Korea  and  now  China  have  used  similar  tools  to  achieve  record  economic  growth.   Scholars  have  documented  the  policies  of  these  “Asian  Tigers”  and  discussed  the   lessons  their  example  sets  for  other  developing  countries  (Wade  1990;  World  Bank   1993;  Krugman  1994;  Stiglitz  1996;  Amsden  2001).  While  other  East  Asian   countries  have  achieved  high  growth,  the  state-­‐led  industrialization  strategies  of   Japan,  Korea  and  Taiwan  bear  the  strongest  similarities  to  China’s  strategy  today   (Perkins  1994).  Yet  today,  as  China  shatters  records  for  sustained  growth,  much  of   the  literature  fails  to  draw  parallels  back  to  these  examples.     Scholars  have  documented  how  the  Asian  Tigers  used  the  state  to  coordinate  a   structural  shift  toward  industrialization.  They  show  that  Japan,  Korea,  and  Taiwan   gradually  liberalized  trade,  investment,  and  finally  capital  markets  after  using  state   policy  to  achieve  a  certain  level  of  industrialization  and  per  capita  income  (Johnson   1982;  Wade  1990;  Amsden  1992;  Amsden  2007).  Alongside  industrial  policy  and   gradual  liberalization,  the  Asian  Tigers  gave  strong  state  financial  support  to   “national  champion”  companies  to  groom  them  for  international  competition.  Since   infant  Asian  industries  could  not  compete  with  Western  companies,  the   governments  of  Japan,  Korea  and  Taiwan  launched  two-­‐pronged  strategies  of   protection  and  promotion.  They  protected  domestic  exporters  in  capital-­‐intensive   industries  through  import  tariffs,  foreign  exchange  controls,  and  in  Japan’s  case   direct  blocks  on  foreign  investment  (Johnson  1982).  The  governments  also   purchased  technology  licenses  to  learn  from  more  advanced  economies  (Komiya  et   al  1988).  In  exchange  for  the  loans,  they  required  and  incentivized  borrowers  to   export  in  order  to  force  them  to  become  competitive  (Amsden  1989;  Lawrence  and   Weinstein  1999).  Scholars  disagree  on  the  extent  to  which  the  Asian  Tigers’  active   efforts  to  support  national  champions  contributed  to  the  success  of  these  industries   but  it  is  clear  such  policy  played  some  role  (Galenson  1979;  Amsden  1989;  Wade   1990;  Lawrence  and  Weinstein  1999;  Ozawa  1999).     Perhaps  the  most  undisputedly  important  feature  of  state  promotion  was  subsidized   finance  for  these  national  champions.  Indeed,  one  of  the  most  critical  studies  on  the   role  of  industrial  policy  in  the  East  Asian  Miracle  singled  out  subsidized  credit  as   being  the  only  successful  policy  in  state-­‐led  industrialization  in  the  region  (World   Bank  1993).    Japan  offered  its  industrial  exporters  funding  through  the  Japan   Development  Bank  (JDB)  and  Export-­‐Import  Bank  of  Japan  (JEXIM).  JDB  and  JEXIM   focused  their  financing  on  “rationalization,”  or  merging  domestic  companies  to   create  national  champions  that  would  gain  efficiency  through  reorganization  and   economies  of  scale  (Johnson  1982).  Korea  went  further,  nationalizing  all  banks.  It   created  the  Korea  Development  Bank  (KDB)  and  Export-­‐Import  Bank  of  Korea   (KEXIM)  to  support  the  textile,  steel  and  shipbuilding  industries.  Since  Korea  could   not  afford  as  much  subsidized  finance  as  Japan,  KDB  and  KEXIM  arranged  and   guaranteed  foreign  loans  (Amsden  1989;  Stern  1995).  In  Taiwan,  the  government   also  offered  substantial  amounts  of  concessional  credit  to  exporters  (Wade  1990).   Western  countries  recognized  the  attractiveness  of  these  national  champion  export   subsidies  for  developing  nations  and  thus  sought  to  ban  them.  Under  the   Washington  Consensus,  Western  nations  argued  that  eliminating  restrictions  on  free   trade  would  benefit  the  world  economy.  Once  the  Asian  Tigers  began  taking  market   share  away  from  the  West  in  industries  like  steel  and  shipbuilding,  Western  leaders   began  pressuring  them  to  end  the  export  subsidies.  By  the  1970s,  Japan  and  Taiwan   were  under  substantial  pressure  to  end  these  subsidies  (Komiya  1988  317;  Wade   1990  96).  Since  Korea  emerged  later,  the  West  was  even  quicker  to  challenge  its   subsidies  (Perkins  1994;  Amsden  2007).  Beginning  in  1975,  developed  countries   agreed  to  ban  subsidized  export  loans  through  the  Organization  of  Economic   Cooperation  and  Development  (OECD)  (Moravcsik  1989).   At  the  same  time,  the  leading  economic  powers  have  largely  ignored  another  tool  of   the  Asian  Miracle:  government  funding  for  outward  foreign  direct  investment   (OFDI).  Europe  and  the  U.S.  have  occasionally  toyed  with  public  financial  support  for   companies  that  invest  abroad.  But  in  addition  to  the  problem  of  identifying  the  right   companies  to  support,  scholars  have  shown  that  overseas  investment  support   suffers  from  moral  hazard,  since  governments  are  unable  to  force  companies  to  use   the  funds  to  follow  government  priorities  (Safarian  1993,  Solís  2003b  156).   Governments  worry  that  companies  will  use  this  support  to  move  jobs  and  profits   overseas  (Ahn  2005).  Why  pay  a  company  to  remove  value  from  your  economy?   Japan  became  a  pioneer  in  government-­‐subsidized  OFDI  in  the  1950s,  with  little   competition.  Even  before  the  government  liberalized  OFDI  in  the  1960s,  the   government  began  to  offer  subsidized  loans  to  companies  investing  abroad.  In  1953,   it  started  a  branch  of  JEXIM  focused  on  OFDI,  which  gave  almost  $70  billion  by  1999.   Solís  argued  in  2003  that  “No  other  country  in  the  world  is  as  active  as  Japan  in   financing  its  corporations’  foreign  investment”  (2003a  103).  Indeed,  while  Japan’s   OFDI  lending  continues  today,  other  nations  have  displayed  little  interest  in  this   strategy.  While  Japan’s  public  FDI  financing  comprised  10.3%  of  its  total  OFDI  from   1953  to  1999,  the  German  equivalent  of  JEXIM  has  contributed  OFDI  financing   equivalent  to  only  0.53%  of  German  OFDI.  In  the  U.S.,  the  Overseas  Private   Investment  Corporation  (OPIC)  is  explicitly  prohibited  from  lending  in  support  of   OFDI  that  might  hurt  American  employment  or  exports.  As  a  result,  OPIC’s  OFDI   financing  comprises  0.08%  of  American  OFDI  (Solís  2003b  156).  There  are  no  WTO   or  OECD  rules  limiting  public  OFDI  financing.  While  Japan  leads,  most  of  the  world   wonders  why  the  Japanese  would  possibly  want  to  pay  companies  to  outsource  jobs.   With  time,  Japan’s  leadership  in  this  area  has  only  grown.  Japan  Bank  of   International  Cooperation  (JBIC),  the  successor  to  JDB,  has  evolved  from  almost   entirely  giving  export  credits  in  1950  to  giving  74%  OFDI  loans  in  2012  (JBIC  2013   12).  When  Japan  met  its  “lost  decade”  in  the  1990s,  many  critics  in  Japan  railed   against  the  hollowing-­‐out  of  Japanese  industry  (kudoka)  that  resulted  from   outsourcing  labor-­‐intensive  production  (Solís  2003a  106).  But  rather  than  backing   off  of  FDI  funding,  the  government  broadened  FDI  loans  to  small  and  medium   enterprises  (Solís  2003a  116).   The  Korean  state  has  also  backed  OFDI,  though  less  enthusiastically  than  Japan.   From  1976  to  1999,  KEXIM’s  OFDI  loans  comprised  9%  of  Korea’s  total  OFDI,  only   1.3%  short  of  Japan  (Solís  2004  16).  However,  the  Korean  figure  is  high  largely   because  total  OFDI  is  low.  In  1995,  nine  years  after  the  state  relaxed  its  grip  on  OFDI   controls,  FDI  as  a  percent  of  GDP  was  only  2.24%,  versus  6%  in  Japan  and  9.8%  in   the  U.S.  (Solís  2004  214).  The  9%  ratio  is  due  mainly  to  the  fact  that  Korea  had  little   OFDI  in  the  denominator.  Most  of  the  OFDI  lending  in  the  numerator  was  for  scarce   natural  resources.  By  the  1990s,  OFDI  lending  in  both  Korea  and  Taiwan  grew   substantially,  but  it  still  lagged  OFDI  rather  than  leading  it.  Most  OFDI  came  from   Korea’s  chaebol  groups,  since  they  were  able  to  borrow  enough  funding   internationally  to  invest  abroad  without  help  from  the  government.     Enter  China   It  is  well  known  that  China  has  followed  Japan,  Taiwan  and  Korea  in  subsidizing   capital-­‐intensive  domestic  industries  and  supporting  national  champions  with   export  subsidies  (Amsden  2001  281).  Scholars  have  traced  China’s  state-­‐led   industrial  development  and  export  strategy  back  to  the  legacy  of  Japan,  Taiwan  and   Korea  (Buckley  2007).  Scholars  began  using  the  phrase  “China,  Inc.”  to  describe   China’s  state-­‐led  capitalism,  just  as  they  had  used  “Japan,  Inc.”  in  the  1960s  and   1970s  and  “Korea,  Inc.”  in  the  1980s  (Downs  2011;  Pempel  1987;  Lee  and  Han   2006).  Just  like  in  Japan  and  Korea,  the  Chinese  government  created  China   Development  Bank  (CDB)  and  China  Export-­‐Import  Bank  (CHEXIM)  in  1993  as  two   new  policy  banks  that  would  lend  according  to  government  priorities  rather  than   commercial  success.     CDB  and  CHEXIM  have  become  the  major  forces  behind  China’s  “Go  Global  Policy.”   In  1998,  then  President  Jiang  Zemin  championed  the  internationalization  of  Chinese   investment  and  lending.  He  argued  that  “Regions  like  Africa,  the  Middle  East,  Central   Asia,  and  South  America  with  large  developing  countries  [have]  very  big  markets   and  abundant  resources;  we  should  take  advantage  of  the  opportunity  to  get  in”   (Chen  2009).  Scholars  have  shown  that  CDB  is  the  main  bank  funding  the  overseas   expansion  of  Chinese  companies  (Downs  2011;  Gallagher  et  al  2012;  Sanderson  and   Forsythe  2012;  Shambaugh  2013).  CHEXIM  has  also  been  a  key  player  in  China’s   global  financial  reach  (Bräutigam  2009;  Gallagher  et  al  2012).   Recently,  China  has  also  emerged  as  a  major  OFDI  lender.  As  with  Japan  and  Korea,   China  initially  blocked  OFDI  to  conserve  foreign  exchange.  Through  the  1990s,  the   policy  banks  prioritized  domestic  lending.  However,  beginning  in  2001,  Jiang   Zemin’s  Go  Global  policy  expanded  to  encourage  Chinese  OFDI.  The  state  created  tax   incentives  and  signed  double  taxation  treaties.  CDB  and  CHEXIM  began  backing  this   policy  with  large  loans  and  lines  of  credit.  In  2004,  Chinese  authorities  relaxed   oversight  restrictions  on  OFDI  and  OFDI  exploded.  Since  2007,  Chinese  OFDI  has   averaged  over  30  billion  dollars  a  year,  easily  exceeding  Korean  OFDI  during  the   same  period.   Scholars  applying  the  lessons  of  the  East  Asian  Miracle  to  China  have  paid  little   attention  to  public  OFDI  lending.  First,  this  is  because  China’s  OFDI  lending  is   relatively  new.  But  perhaps  more  significantly,  public  OFDI  lending  has  received   little  attention  and  certainly  little  consensus  as  a  “lesson”  from  the  East  Asian   Miracle  in  general.  It  is  not  mentioned  as  a  key  strategy,  and  barely  mentioned  at  all,   in  the  studies  of  Japan’s  economic  miracle  (Johnson  1990;  Ozawa  1999).  We  begin   by  situating  Chinese  public  OFDI  lending  in  the  context  of  the  East  Asian  Miracle.   Estimating  Chinese  OFDI  Finance   In  this  section,  we  estimate  the  size,  composition,  and  terms  of  Chinese  loans  and   lines  of  credit  to  support  OFDI  by  domestic  firms  and  we  compare  those  estimates   to  figures  in  the  literature  on  Japan  and  Korea.     China  does  not  publish  disaggregated  data  on  CDB  or  CHEXIM  financing.  Thus,  we   constructed  a  database  by  combing  through  English-­‐  and  Chinese-­‐language  news   articles  like  the  Wall  Street  Journal  and  the  People’s  Daily,  company  filings  with  the   U.S.  Securities  and  Exchange  Commission  (SEC),  government  reports  from  both   China  and  the  host  countries,  and  bank  reports  from  CDB  and  CHEXIM—all   documented  in  Appendix  1.  We  include  only  loans  that  were  confirmed  by  multiple   reliable  sources.  We  compare  our  Chinese  lending  estimates  with  Japanese  OFDI   loans  from  the  Japan  Bank  for  International  Cooperation  (JBIC)  and  Korean  OFDI   from  KEXIM,  both  today  and  in  historical  context.   Our  China  estimates  should  only  be  considered  estimates.  Neither  the  Chinese   government  nor  the  lending  institutions  publish  official  data  on  OFDI  lending.   Unlike  Japan  and  Korea,  its  lending  is  spread  across  multiple  banks,  making  it  even   more  difficult  to  track  down.  It  is  also  difficult  to  distinguish  between  OFDI  and   trade  finance.  CDB  and  CHEXIM  provide  three  main  types  of  foreign  lending:  loans   to  support  foreign  governments  and  companies,  trade  finance  loans  to  support   China’s  exports  and  imports,  and  loans  that  support  Chinese  companies’  physical   operations  abroad.  The  two  former  types  of  loans  have  been  catalogued  for  Latin   America  and  Africa  in  previous  research  (Gallagher  et  al  2012;  Gallagher  and   Bräutigam  2014).  Only  the  latter  qualifies  as  OFDI  lending.    Still,  there  is  a  significant   gray  area  in  distinguishing  between  FDI  loans  and  export  finance.  Chinese  banks   have  lent  billions  to  Chinese  construction  companies  and  their  customers  to  build   dams,  ports,  and  telecom  infrastructure  abroad.  We  only  count  these  loans  if  the   Chinese  firm  is  the  owner  rather  than  the  contractor,  since  contractors  are  trading   in  goods  and  services  rather  than  investing.  However,  we  do  consider  loans   supporting  Build-­‐Own-­‐Operate-­‐Transfer  (BOOT)  projects  to  be  OFDI  loans,  since  the   contractor  will  own  the  project  for  a  significant  length  of  time.2   We  report  a  lower  and  upper  bound  in  addition  to  our  main  estimate  because  many   of  these  loans  were  split  between  OFDI  and  other  purposes.  In  many  cases,  the   purpose  of  a  loan  was  simply  described  as  “to  support  the  company’s  Going  Global   strategy.”  Since  Jiang  Zemin  spearheaded  China’s  Go  Global  strategy  in  1998,  global   expansion  has  become  a  buzzword  that  every  company  tacks  on  in  order  to  receive   funding.  Upon  closer  inspection,  “Going  Global”  finance  includes  not  only  OFDI   lending,  but  also  the  financing  for  contractors’  customers  described  above  and   traditional  export  financing  (as  well  as  industrial  restructuring  and  any  domestic   venture  to  become  more  internationally  competitive).  Often  one  loan  will  claim  to   cover  all  of  these  areas.  We  address  these  “all  of  the  above  loans”  by  giving  a  lower   and  upper  bound  to  our  estimates.  The  lower  bound  comprises  all  the  lending  that   included  OFDI  as  a  possible  use,  while  the  upper  bound  comprises  only  the  lending   that  explicitly  named  OFDI  as  its  single  purpose.  Thus,  the  lower  bound  includes  all   the  lending  that  Chinese  companies  could  have  used  for  OFDI,  while  the  upper   bound  includes  only  the  lending  that  Chinese  companies  did  use  for  OFDI.     2  One  could  complicate  the  boundary  between  loans  and  investment  even  further.  Many  loans   promote  investment  on  the  surface  but  support  trade  on  a  deeper  level.  If  a  car  manufacturer   gets  a  “tariff-­‐jumping”  loan  to  build  an  assembly  plant  overseas,  the  loan  will  directly  support   overseas  investment,  but  it  also  supports  the  export  of  car  designs,  assembly  technology  and   unassembled  car  parts.  If  a  wind  turbine  manufacturer  gets  a  loan  to  set  up  a  wind  farm   overseas,  it  will  use  the  loan  largely  to  buy  its  own  exports.  We  attempt  to  draw  a  line  that   mirrors  the  definition  of  overseas  investment  finance  used  by  JBIC  and  KEXIM—finance  with  a   stated  main  purpose  of  supporting  domestic  companies’  overseas  investment.     Bank    CDB    Ex-­‐Im    BoC    ICBC    CCB    Multiple    Total   Amount   $92,860   $34,151   $12,310   $770   $475   $3,800   $144,366   %  of  total   64.32%   23.66%   8.53%   0.53%   0.33%   2.63%   100.00%   Table  1.  Chinese  OFDI  Finance  by  bank,  in  millions  of  $   As  shown  in  Table  1,  since  2002,  we  constrain  China’s  OFDI  lending  between  a  lower   bound  of  $88  billion  and  an  upper  bound  of  $192  billion.  Using  a  simple  average,  we   estimate  China’s  OFDI  lending  at  $140  billion.  As  with  China’s  lending  to  foreign   governments  and  companies,  and  despite  CHEXIM’s  public  notices  in  support  of   OFDI  lending,  CDB  provided  the  lion’s  share.  Gallagher  et  al  (2012)  find  that  CDB   provided  82%  of  Chinese  lending  to  Latin  American  governments  and  companies,   with  CHEXIM  adding  12%.  Our  database  of  China’s  OFDI  lending  revealed  64%  from   CDB,  24%  from  CHEXIM,  and  9%  from  the  Bank  of  China.  CHEXIM  is  better   represented  in  OFDI  lending  than  foreign  lending.  This  is  not  surprising  since   CHEXIM  issued  a  joint  notice  with  China’s  main  economic  planning  agency,  NDRC,   that  both  bodies  “will  jointly  set  up  a  credit  support  mechanism  for  overseas   investments”  (NDRC  2004).  Still,  while  CDB  has  not  issued  notices  on  its  OFDI   lending  policies,  it  provides  2.5  times  more  OFDI  lending  than  CHEXIM.     A  recent  company  survey  in  China  suggests  that  despite  these  large  loans,   companies  primarily  finance  their  overseas  investments  through  retained  earnings.   China  Council  for  the  Promotion  of  International  Trade’s  2013  China  Outbound   Foreign  Direct  Investment  Survey  reports  that  only  21%  of  foreign-­‐investing  firms   rely  primarily  on  bank  loans,  while  52%  rely  on  retained  earnings  (Zhang  2013).   Thirty  percent  report  using  bank  loans  as  a  major  source  of  funding,  compared  to   70%  for  retained  earnings.  However,  this  survey  polls  thousands  of  private  and   state-­‐owned  companies  in  China,  mostly  with  revenues  below  $1  billion.  As  will  be   shown  later,  the  vast  majority  of  China’s  OFDI  funding  goes  to  state-­‐owned   enterprises  (SOEs)  with  revenues  over  $10  billion.  Since  the  survey  reports  the   results  by  number  of  firms,  rather  than  by  total  revenue,  its  results  cannot  be  taken   as  representative  for  our  discussion  of  national  champions.  Still,  it  is  useful  to   remember  that  companies  are  using  retained  earnings  to  fund  OFDI  in  addition  to   these  loans.    China    Japan    Korea   Amount   139.9   140.7   37.1   %  of  GDP   0.31%   0.26%   0.38%   %  of  OFDI   58.0%   15.6%   20.3%   Table  2.  Chinese,  Japanese  and  Korean  OFDI  finance,  2003-­‐2011,  in  billions  of  $   In  Table  2,  we  compare  the  quantity  of  OFDI  lending  today  in  China,  Japan  and   Korea.  We  find  that  they  give  in  roughly  equal  proportions  to  GDP,  but  China  gives   far  more  as  a  percentage  of  OFDI.  Since  2002,  we  constrain  China’s  OFDI  lending   between  a  lower  bound  of  $88  billion  and  an  upper  bound  of  $192  billion.  Using  a   simple  average,  we  estimate  China’s  OFDI  lending  at  $140  billion.  Japan’s  OFDI   lending  through  JBIC  since  2002  sums  to  $145.2  billion.  Korea’s  OFDI  loans  plus   guarantees  through  KEXIM  since  2002  total  $37  billion.  Since  one  would  expect  a   larger,  wealthier  country  to  give  more  OFDI  loans,  we  also  compare  these  figures  as   a  percentage  of  GDP.  China’s  OFDI  lending  comprises  0.31%  of  GDP,  while  Japan’s   comprise  0.26%  and  Korea’s  loans  and  guarantees  make  up  0.38%.  Loans  as  a   percentage  of  GDP  appear  fairly  constant  across  the  three  countries.  Finally,  we   compare  these  figures  as  a  percentage  of  total  OFDI,  since  one  would  expect  a  nation   with  more  OFDI  to  invest  more  in  OFDI  loans.  China’s  loans  come  to  31%  of  total   OFDI,  Japan’s  to  16%,  and  Korea’s  loans  and  guarantees  to  20%.  China  lends  roughly   three  times  more  than  Japan  or  Korea  as  a  percentage  of  total  OFDI.  This  looks  very   different  from  the  GDP  comparison,  since  China’s  total  OFDI  comprises  only  0.6%  of   GDP,  compared  to  1.4%  for  Japan  and  1.8%  for  Korea.   It  should  be  noted  that  our  Korean  OFDI  figures  are  also  imprecise.  KEXIM’s  Annual   Report  lumps  together  OFDI  loans  and  guarantees.  As  a  result,  all  of  the  KEXIM  OFDI   figures  are  overestimates  because  they  include  guarantees  in  addition  to  loans.3  Our   Japan  data  includes  only  OFDI  loans,  as  does  the  China  data.  KEXIM  also  differs  by   recording  disbursements  rather  than  commitments.     It  appears  that  China  is  encouraging  OFDI  more  proactively  than  Japan  or  Korea.  We   compared  OFDI  finance  to  total  OFDI,  as  reported  by  Japan  and  Korea  to  the  OECD.   China’s  OFDI  statistics  are  notoriously  difficult  to  acquire,  since  the  Ministry  of   Commerce’s  official  statistics  show  most  OFDI  flowing  to  tax  havens  rather  than   3  KEXIM  also  introduced  a  new  OFDI  loan  category  in  its  2012  Annual  Report.  In  addition  to  the  reported   $7.2  billion  in  Overseas  Investment  Finance,  it  also  gave  $1.9  billion  in  Natural  Resources  Finance,  which   includes  loans  and  guarantees  to  resourceg related,  Koreang held  companies  operating  abroad   (KEXIM  2012).   actual  OFDI  destinations  (MOFCOM  2011).  We  used  publicly  available  OFDI  data   from  the  Heritage  Foundation’s  China  Global  Investment  Tracker,  created  by  Derek   Scissors.  This  database  most  likely  overestimates  Chinese  OFDI  since  it  includes   portfolio  investment  and  reports  publicly  announced  commitments  rather  than   actual  disbursements  (Bräutigam  2013).  Though  Japan  was  the  “undisputed  leader”   in  OFDI  loans  in  2003  (Solís  2003b  153),  today  China  gives  at  least  55%  more  OFDI   finance  per  dollar  of  OFDI.  From  2002  to  2012,  China’s  OFDI  lending  comprised  31%   of  total  OFDI,  in  contrast  to  16%  for  Japan  and  20%  for  Korea.  In  addition,  since  our   OFDI  total  is  an  overestimate,  this  31%  ratio  should  be  considered  an   underestimate.  China’s  high  loan-­‐to-­‐OFDI  ratio  suggests  that  the  state  banks  are   more  invested  in  encouraging  OFDI  than  the  companies  themselves.  This  makes   sense  in  today’s  world,  where  South  Korea  and  Japan’s  firms  have  grown  out  of  their   ‘infancy’  and  can  finance  their  expansion  without  government  support  through   international  capital  markets.   It  seems  more  appropriate  to  compare  China’s  OFDI  lending  today  with  that  of  Japan   and  Korea  when  they  were  at  equivalent  stages  of  development.  China’s  OFDI   encouragement  makes  sense,  since  Chinese  companies  have  relatively  little   experience  investing  abroad  and  need  subsidized  loans  to  make  the  leap.  Japanese   and  Korean  companies,  by  contrast,  do  not  need  the  help.  They  have  already  built   global  supply  chains,  acquiring  natural  resources  and  shipping  to  cheap-­‐labor   assembly  plants.  Indeed,  JBIC’s  predecessor  Japan  Development  Bank  gave   enormous  subsidies  to  the  infant  shipbuilding  and  computer  manufacturing   industries.  Today,  we  find  little  evidence  of  state  support  for  these  industries   because  the  state  removed  the  subsidies  as  they  became  globally  competitive   (Shinjo  1988;  Yonezawa  1988).  If  Japan  and  Korea  ever  mirrored  China’s  heavy   OFDI  push,  it  would  have  been  decades  ago.    China    Japan,  1971-­‐1984   Amount   139.9   8.4   %  of  GDP   0.31%   0.08%   %  of  OFDI   58.0%   12.2%   Table  3.  Chinese  and  Japanese  OFDI  finance  during  industrialization.   When  comparing  China’s  OFDI  today  to  that  of  Japan  in  the  1960s  and  Korea  in  the   1980s,  we  find  that  China  still  stands  out.  First,  we  compare  China’s  2002-­‐2012   period  with  Japan  in  the  late  1960s  and  1970s,  when  it  had  equivalent  per  capita   GDP.  Japan’s  OFDI  lending  as  a  percent  of  GDP  was  0.08%,  or  roughly  a  fourth  of   China’s  0.31%.  As  Solís  (2004)  discusses,  Japan  engaged  in  little  OFDI  before  1972.   China  started  lending  in  earnest  by  2003,  a  few  years  shy  of  reaching  the  same  per   capita  GDP  as  Japan  in  1972.  Japan’s  OFDI  in  1972-­‐1973  easily  exceeded  total  OFDI   from  the  previous  two  decades  (Solís  2004  42).  But  even  in  1973,  when  Japan  gave   its  largest  OFDI  finance  push  of  the  century  in  response  to  the  collapse  of  the   Bretton  Woods  fixed  exchange  rate  system  and  the  oil  crisis,  OFDI  lending  only   reached  0.20%  of  GDP.5  The  ratio  of  OFDI  lending  to  OFDI  increased  from  10%  in   1972  to  18%  in  1973  and  20%  in  1974.  For  the  periods  1951-­‐1970  and  1971-­‐1984,   Japanese  OFDI  lending  reached  12%  of  total  OFDI.  This  ratio  decreased  to  9%  from   1985-­‐1999.6  Korean  OFDI  lending  similarly  comprised  9%  of  total  OFDI  from  1976   to  1999  (Solís  2004  16).  All  of  these  percentages  are  dwarfed  by  modern-­‐day   China’s  31%  average.     As  more  companies  follow  the  Chinese  state’s  enthusiastic  lead,  we  expect  Chinese   OFDI  to  greatly  increase.  Japan’s  lending  history  suggests  that  OFDI  lending  pushes   do  increase  OFDI.  Solís  (2003a)  demonstrates  that  Japan’s  manufacturing  OFDI   loans  in  the  1960s  through  the  1980s  resulted  in  corresponding  increases  in   manufacturing  OFDI.  If  OFDI  lending  translates  to  OFDI  in  China  as  well,  it  will  cause   OFDI  to  grow.  We  expect  that  it  will  happen,  since  Reform  and  Opening  has  required   Chinese  companies  to  act  boldly  upon  perceived  shifts  in  state  policy.  Just  as  city   governments  used  Deng  Xiaoping’s  Southern  Tour  to  justify  economic  reforms,   companies  endlessly  cite  “Going  Global”  as  the  foundation  of  their  international   ambitions.  With  the  state  pushing  OFDI  so  strongly,  it  seems  inevitable  that   companies  will  recognize  the  push  and  join  the  bandwagon.   Country    Japan    China    Japan    Japan    China    Japan    China    China   Borrower    Yen  rate    Chinalco    $  rate    $  rate    TCL    Yen  rate    Huawei    CNOOC   Rate   0.88%   L+0.01%   L+0.25%   L+0.44%   L+0.6%   2.50%   L+2%   4.05%   Year   2011   2008   2010   2004   2004   2005   2009   2006   Table  4.  Chinese  and  Japanese  interest  rates  on  OFDI  loans  (L=LIBOR)   5 6  Calculated  with  data  in  Solís  2004  (40)  and  other  data  compiled  by  Solís.    Calculated  with  data  in  Solís  2004  (40)  and  other  data  compiled  by  Solís.   We  also  compared  the  interest  rates  on  Chinese  and  Japanese  OFDI  loans,  finding   that  while  both  Japan  and  China  subsidize  their  loans,  Japan’s  rates  may  be  lower  on   average.  We  were  able  to  find  a  few  Chinese  OFDI  loan  interest  rates  through   interviews  and  newspaper  articles.  At  the  high  end,  China  National  Overseas  Oil   Corporation  (CNOOC)  reportedly  took  out  a  4.05%  fixed-­‐rate  loan  in  2006.  At  the   low  end,  Chinalco  paid  0.01%  over  LIBOR,  the  rate  that  banks  charge  each  other,  for   a  loan  in  2008.  CNOOC’s  loan  is  not  particularly  cheap,  while  Chinalco  is  essentially   paying  the  lowest  possible  rate.  We  found  interest  rates  on  similar  Chinese  loans  as   well—export  loans  to  telecom  and  infrastructure  companies.  Huawei  and  TCL  paid   spreads  over  LIBOR  of  2%  and  0.6%  in  2009  and  2004,  respectively.  The  former  is  a   fairly  typical  commercial  rate,  while  the  latter  is  clearly  subsidized.  By  contrast,   JBIC’s  reported  OFDI  loan  interest  rates  all  appear  subsidized.7  Over  the  past   decade,  its  rates  range  from  0.25%  to  0.5%  over  for  foreign  currency  loans  and  from   0.875%  to  2.5%  (fixed-­‐rate)  for  yen-­‐denominated  loans.  All  of  these  rates  were   reportedly  lowered  further  for  highly  encouraged  projects  (JBIC  2013).  While  China   has  a  reputation  for  “cutthroat”  finance,  its  OFDI  loan  rates  vary  widely,  and  many   exceed  Japan’s  rates.     Comparison  of  Motives   In  this  section,  we  break  down  China’s  OFDI  lending  by  sector  and  motive  to   compare  it  on  a  more  detailed  level  with  Japanese  lending.  Inter-­‐sectoral   comparisons  are  difficult  because  of  China’s  lack  of  data.  Some  sectors  were  more   likely  to  receive  clear-­‐cut  OFDI  lending  tied  to  specific  projects  (e.g.  mining  and  oil),   while  manufacturing  usually  received  “all  of  the  above”  loans  for  export  credits,   overseas  investment  and  overseas  contracting.  Including  these  loans  would   overestimate  OFDI  lending  to  manufacturing  and  infrastructure  sectors,  while   excluding  them  would  underestimate  these  sectors.  For  the  purposes  of  comparison   with  Japan,  we  applied  a  strict  test  that  seemed  closest  to  the  definition  of  OFDI   lending  used  by  JEXIM—finance  intended  primarily  to  support  OFDI—that   disqualified  the  “all  of  the  above”  manufacturing  loans.  As  a  result,  our  sectoral   analysis  underestimates  the  amount  of  Chinese  manufacturing  lending.   Solís  (2003a)  categorized  Japan’s  OFDI  lending  into  three  stages;  the  literature   agrees  that  the  first  stage  focused  on  natural  resource  acquisition  (Farrell  et  al  2004   164;  Komiya  and  Wakasugi  1991  51;  Solís  2003a).  From  1953  to  1970,  JEXIM  gave   7  While  the  OECD  has  banned  subsidized  export  finance  for  fear  of  competition,  it  does  not  regulate  OFDI   loans  because  there  is  no  need—only  Japan  and  Korea  really  use  them.   35%  of  its  loans  for  natural  resource  acquisition.8  In  addition,  another  34%  went  to   the  lumber  and  steel  industries,  largely  to  acquire  raw  materials  (Solís   correspondence;  Solís  2003a  104).  It  was  during  this  period  that  OFDI  loans  in  the   iron,  steel  and  nonferrous  metals  sector  reached  55%  of  total  manufacturing  OFDI   (Solís  2003a  106).  Still,  it  is  important  to  note  that  Japan’s  OFDI  and  OFDI  lending  in   this  period  were  miniscule  compared  to  future  stages.  Japan  lent  twice  as  much  in   1973  as  it  did  in  this  entire  period.   Japan’s  counterintuitive  second  stage  began  in  the  mid-­‐1960s,  when  the   government  began  paying  to  outsource  industries.  It  gave  loans  to  support   “industrial  adjustment,”  or  getting  rid  of  industries  that  were  on  the  way  out   anyways  (Mah  and  Noh  2012  310).  By  the  mid-­‐1960s,  the  government  recognized   that  Japan  would  not  be  competitive  in  the  labor-­‐intensive  textile  industry.   Preferring  outsourcing  to  bankruptcy,  JEXIM  gave  subsidized  loans  to  help  Japanese   textile  firms  move  to  China  and  Southeast  Asia,  (Solís  2003a  105;  Ogawa  and  Lee   1996;  Mah  and  Noh  2012  310).  After  the  end  of  the  Bretton  Woods  fixed  exchange   rate  system  in  1972  and  the  1973  oil  crisis,  Japan’s  metal  and  oil-­‐refining  industries   suddenly  lost  competitiveness.  So  from  1971  to  1984,  the  majority  of  Japan’s   manufacturing  OFDI  loans  went  to  outsourcing  unprofitable  heavy  industry.  The   chemicals  and  iron,  steel  and  nonferrous  metals  sectors  alone  received  over  two   thirds  of  OFDI  lending  for  manufacturing,  up  from  24.9%  in  the  first  period  (Solís   2003a  105).    In  the  third  stage,  from  1985  to  1999,  the  Japanese  government  continued  to  use   OFDI  lending  to  outsource  other  “sunset  industries.”  As  wages  rose,  low-­‐margin,   low-­‐wage  sectors  began  losing  money,  and  the  government  pushed  them  overseas.     OFDI  loans  for  vehicle  and  electronics  assembly  rose  from  13.8%  of  manufacturing   OFDI  lending  in  the  second  stage  to  44.7%  in  the  third  (Solís  2003a  104).  Heavy   industry  lategoers  in  the  chemicals  and  iron,  steel  and  nonferrous  metals  sectors   added  another  41.0%.  According  to  the  literature,  despite  fears  of  “hollowing-­‐out,”   Japan’s  state  support  for  outsourcing  did  not  hurt  its  domestic  economy  (Solís   2004).     Beginning  in  the  third  stage,  Japan  also  engaged  in  OFDI  to  tap  into  overseas   markets  and  acquire  technology,  though  it  is  unclear  how  much  OFDI  lending  went   to  support  these  efforts.  Japan  had  large  incentives  to  engage  in  market-­‐access  FDI   beginning  in  1977,  when  the  U.S.  banned  TV  imports  from  Japan.  Similarly,  in  1981   the  U.S.  placed  a  quota  on  Japanese  car  imports  (Solís  2004  42).  In  the  early  1990s,   Japanese  companies  used  OFDI  to  “tariff-­‐hop”  into  Europe  after  the  1992  European   8  Calculated  with  data  in  Solís  2004  (40)  and  other  data  compiled  by  Solís.   Community  integration  and  into  the  U.S.  after  NAFTA  in  1994  (Farrell  et  al  2004   172,  Komiya  and  Wakasugi  1991  55).  Even  in  China,  much  of  Japan’s  OFDI  in  the   2000s  went  to  open  the  Chinese  market  through  joint  ventures  (Mah  and  Noh  2012   314).  Recently,  Japanese  OFDI  has  also  used  mergers  and  acquisitions  (M&A)  to   acquire  foreign  technology  as  opposed  to  the  traditional  licensing  method  (Komiya   and  Wakasugi  1991  56;  Farrell  et  al  2004  174).     Korean  OFDI  lending  began  with  the  same  first  stage  of  acquiring  natural  resources.   In  the  1960s  and  1970s,  almost  all  OFDI  went  to  acquire  primary  resources  (Kim   and  Mah  2006  887).  In  1975,  the  government  finally  began  granting  OFDI  permits   for  companies  that  needed  to  outsource  to  “regain  international  competitiveness”   (UNCTAD  2006  208).  Still,  even  into  the  mid-­‐1980s  it  remained  more  concerned   about  preventing  capital  flight  than  promoting  OFDI  (Kim  2000  109;  Kumar  and   Kim  1984  52).  Over  60%  of  OFDI  lending  from  1976  to  1984  still  went  to  secure   natural  resources  (Solís  2004  210).     Like  Japan,  Korea  then  entered  a  stage  of  industrial  adjustment,  which  began  in  the   mid-­‐1980s.  In  1985,  Korea  Export-­‐Import  Bank  (KEXIM)  performed  an  about-­‐face   and  began  giving  FDI  loans  mostly  to  the  manufacturing  sector.  Between  1985  and   1997,  manufacturing  received  89.8%  of  all  FDI  lending  (Solís  2004  210).  Once  the   Asian  Financial  Crisis  began  in  1997,  KEXIM  largely  abandoned  FDI  lending  in  favor   of  its  traditional  export  financing.  Since  KEXIM  Annual  Reports  only  disaggregate   the  sectors  of  export  loans,  not  OFDI  loans,  we  are  unable  to  reliably  compare  the   sectoral  breakdown  of  China  and  Korea’s  OFDI  lending  today.   While  Korean  and  Taiwanese  companies  have  outsourced  labor-­‐intensive  industries   en  masse  and  acquired  foreign  technology  through  M&A,  they  have  done  it  mostly   without  government  assistance.  In  the  last  decade,  Korean  companies  have   outsourced  most  labor-­‐intensive  production  to  China,  which  currently  absorbs  most   Korean  OFDI  (Athukorala  and  Hill  2002;  Kim  and  Mah  2006  883).  Korean  OFDI  also   focuses  on  M&A  to  gain  foreign  technology  and  open  up  new  markets  by  jumping   tariff  barriers  (Kim  2000  111).  But  despite  the  growth  in  outsourcing,  tariff-­‐hopping   and  technology-­‐seeking  OFDI,  the  state  did  not  proactively  push  these  motives  with   OFDI  lending  (Kim  2000).   The  Chinese  government  and  numerous  scholars  agree  that  China  encourages  OFDI   to  acquire  primary  resources,  gain  access  to  advanced  technology,  and  enter  foreign   markets,  but  not  to  outsource  dying  industries.  The  existing  studies  on  China’s  OFDI   draw  from  the  literature  on  OFDI  in  general  and  in  emerging  economies.  The   international  business  literature  on  firm  internationalization  highlights  four  main   motivations:  outsourcing  for  cost  reduction,  resource  acquisition,  technological   learning,  and  market  entry  (Buckley  1976;  Dunning  1977).  China  scholars  agree  that   outsourcing  is  not  a  significant  factor  in  China’s  OFDI  lending.  Although  the   government’s  OFDI  data  is  unusable  due  to  tax  havens,  scholars  have  begun  to  use   other  OFDI  sources  to  analyze  the  reasons  for  Chinese  OFDI.  They  have  published   case  studies,  descriptions  of  overall  trends,  and  quantitative  analyses  (Child  and   Rodrigues  2005;  Buckley  2007;  Deng  2009).  These  studies  conclude  that  China’s   OFDI  follows  the  latter  three  drivers  above  while  they  reject  outsourcing,  citing   China’s  relatively  low  labor  costs  (Zhang  and  Daly  2011;  Ye  forthcoming).The   Chinese  government  has  confirmed  the  latter  three  motives  in  both  its  Five-­‐Year   Plan  for  2010-­‐2015  and  in  a  joint  notice  by  CHEXIM  and  China’s  main  economic   planning  agency,  NDRC  (NDRC  and  CHEXIM  2004;  Luo  2010  76).   These  scholars  base  their  arguments  about  the  government’s  motives  on  OFDI   figures,  which  represent  the  amalgamated  priorities  of  China’s  diverse  companies,   not  the  government’s  intentions.  In  the  interest  of  using  available  data,  most  studies   rely  on  databases  of  China’s  OFDI.  Yet  China’s  OFDI  comes  from  many  different   types  of  companies,  including  small  and  large  private  companies  as  well  as  local,   provincial  and  national  SOEs.  Even  the  national  SOEs  do  not  act  according  to  the   central  state’s  priorities.9  Thus,  a  study  on  how  much  companies  are  investing  in   particular  countries  and  industries  can  only  suggest  the  motives  of  the  various   companies,  rather  than  the  motives  of  the  government.  Existing  studies  do  report  on   the  government’s  OFDI  policy  statements  and  policies,  which  include  restrictions   abolished,  tax  incentives  created,  loans  encouraged  and  treaties  signed.  But  since   the  government  does  not  publish  data  on  its  OFDI  lending,  the  literature  on  the   government’s  motives  has  not  moved  beyond  policy  descriptions  and  case  studies.   We  assess  the  state’s  actual  motives  by  cataloguing  its  OFDI  lending  by  sector  and   purpose  and  comparing  them  to  Japan’s  JBIC  lending  today  and  in  the  past.  Today,   JBIC  provides  a  detailed  sectoral  breakdown  of  manufacturing,  natural  resource  and   other  loans  in  its  annual  reports,  and  it  was  relatively  straightforward  to  assign  the   Chinese  loans  to  the  same  categories.  The  main  difference  lay  in  sectors  like  steel,   since  China’s  steel  companies  are  investing  abroad  in  raw  materials  (iron),  while   Japan’s  are  more  likely  to  be  building  steel  mills  abroad.  We  categorized  iron   investments  by  China’s  steel  companies  as  Mining  and  Oil  rather  than   Manufacturing.  In  Japan’s  case,  we  had  no  details  on  the  Japanese  projects,  and  the   reported  totals  for  natural  resource  and  manufacturing  loans  often  overlapped.  We   had  to  make  some  educated  guesses  when  simplifying  JBIC’s  sectoral  breakdown.   9  To  complicate  matters  further,  the  central  state  itself  does  not  speak  with  one  voice.  The  major  relevant   actors  here  include  the  Politburo  Standing  Committee,  National  Development  and  Reform  Council   (NDRC),  SASAC,  banking  sector  leaders,  and  the  Ministry  of  Commerce  (MOFCOM).   As  Table  5  shows,  the  main  takeaway  from  China’s  sectoral  breakdown  since  2002  is   that  the  mining  and  oil  sector  remains  completely  dominant.  The  mining  and  oil   sector  absorbed  roughly  80%  of  China’s  OFDI  loans  from  2002  to  2012.  Another   12%  went  to  infrastructure  and  4%  to  manufacturing.  Services  received  just  over   1%,  and  agriculture  and  textiles  received  less  than  1%.  As  discussed  earlier,  the   manufacturing  share  would  have  been  roughly  three  times  greater  if  we  had   counted  “all  of  the  above”  loans  designed  primarily  as  export  credits.   Japan   By  sector   Mining  and  Oil   Agriculture   Textiles   Manufacturing   Infrastructure   Services   Other   Total   Amount   $63,797   $180   $421   $19,540   $19,050   $8,307   $29,393   $140,689   China   Percent   45.3%   0.1%   0.3%   13.9%   13.5%   5.9%   20.9%   100.00%   Amount   $73,742   $1,560   $40   $3,378   $10,928   $2,600   $0   $92,248   Percent   79.94%   1.69%   0.04%   3.66%   11.85%   2.82%   0.00%   100.00%   Table  5.  Chinese  and  Japanese  OFDI  Lending  By  Sector,  2002-­‐2012,  in  $  millions   Japan,  1971-­‐1984   By  sector   Mining  and  Oil   Agriculture   Textiles   Manufacturing   Infrastructure   Services   Other   Total   Amount   $2,693   $118   $278   $5,184   Percent   32.0%   1.4%   3.3%   61.6%   $143   $8,415   1.7%   100.00%   China   Amount   $73,742   $1,560   $40   $3,378   $10,928   $2,600   $0   $92,248   Percent   79.94%   1.69%   0.04%   3.66%   11.85%   2.82%   0.00%   100.00%   Table  6.  Chinese  and  Japanese  OFDI  Lending  By  Sector  during  Industrialization  in  $  millions   As  seen  in  Tables  5  and  6,  Japan  spread  its  OFDI  lending  out  between  sectors  much   more  evenly  than  China,  both  historically  and  today.  In  the  first  stage  from  1953-­‐ 1970,  mining  and  oil  received  the  largest  share  at  32%,  followed  by  lumber  at  19%,   steel  at  15%,  and  textiles  at  14%.  From  1971-­‐1984,  steel  moved  into  the  top  slot   with  33%  as  Japan  outsourced  its  heavy  industry.  Mining  and  oil  followed  with  32%   and  chemicals  with  13%.  From  1985-­‐1999,  transport  and  electrical  machinery   received  30%,  steel  14%,  chemicals  13%,  and  mining  and  oil  11%.  From  2002-­‐2012,   45%  of  JBIC’s  loans  went  to  mining  and  oil,  14%  each  to  manufacturing  and   infrastructure,  6%  to  services,  and  21%  to  other  sectors.  It  is  interesting  that  today,   mining  and  oil  has  become  even  more  dominant  than  it  was  in  the  first  stage.  Even   so,  its  45%  share  of  total  OFDI  lending  falls  well  short  of  China’s  80%.   Sector    Mining    Oil    Agriculture    Manufacturing    Infrastructure    Services    Total   OFDI   Loans   $57,760   $45,850   $1,560   $22,962   $13,593   $2,600   $144,366   Total   OFDI   $101,640   $148,530   $12,310   $34,990   $18,220   $77,000   $399,780   Ratio   56.8%   30.9%   12.7%   65.6%   74.6%   3.4%   36.1%   Table  7.  Chinese  OFDI  Lending  as  a  Percent  of  OFDI  By  Sector,  2002-­‐2012,  in  $  millions   From  Table  7,  it  is  evident  that  the  Chinese  government  is  pushing  OFDI  harder  in   some  sectors  than  in  others.  The  ratio  of  OFDI  lending  to  total  OFDI  varies  from  3%   in  services  to  75%  in  infrastructure.  Chinese  companies  investing  overseas  in  real   estate,  the  financial  sector,  and  agriculture  are  receiving  little  state  support.  At  the   other  end,  the  state  is  heavily  involved  in  supporting  overseas  manufacturing  and   infrastructure  projects.  Interestingly,  mining  receives  almost  twice  as  much  support   as  oil.  When  compared  to  developing  Japan,  China  is  pushing  most  sectors  quite   vigorously.  Japan’s  average  ratio  was  10%,  and  its  “most  extreme”  support  went  to   the  mining  sector  from  1953-­‐1970,  with  a  loans-­‐to-­‐OFDI  ratio  of  55%  (Solís  2003a   106).  In  China,  all  sectors  except  services  exceed  Japan’s  10%  average.  Three  sectors   exceed  Japan’s  all-­‐time  high  of  55%.   We  also  catalogued  China’s  stated  motives  for  each  of  its  individual  OFDI  loans  in   order  to  compare  them  to  JEXIM’s  three  stages  of  lending.  In  our  China  search,  loan   descriptions  included  enough  information  to  reliably  determine  whether  the  loan   addressed  natural  resource  acquisition,  market  access,  technological  learning   through  M&A,  or  industrial  adjustment.  While  a  bank  may  have  had  multiple   reasons  for  providing  an  individual  loan,  we  maintain  that  each  loan  had  one  of   these  four  motives  as  its  primary  raison  d’être.    Unfortunately,  the  same  data  was   not  available  for  Japan  or  Korea  today.  While  JBIC  offers  data  disaggregated  by   sector,  neither  JBIC  nor  KEXIM  reports  data  on  its  motives  or  provides  information   on  enough  individual  loans  for  us  to  make  estimates.   As  with  Japan  and  Korea’s  first  stages,  China’s  leading  OFDI  loan  motive  is  natural   resource  acquisition.  It  is  responsible  for  81%  of  China’s  OFDI  loans  since  2002.  In   the  mining  and  oil  sector,  all  but  one  loan  was  aimed  at  acquiring  resources.  The   outlier,  a  $230  million  loan  to  Sinochem,  went  to  acquire  advanced  oil  drilling   technology  from  Norway.  While  there  were  other  loans  funding  acquisitions  of   developed  country  mining  and  oil  operations,  those  operations  were  valuable  for   their  resources  rather  than  their  technology.  China’s  81%  exceeds  the  over  60%  of   Korean  lending  that  went  to  natural  resource  acquisition  from  1976  to  1984  (Solís   2004  210),  as  well  as  Japan’s  47%  share  for  mining  and  oil  loans  in  its  modern-­‐day   stage.  The  main  purpose  of  OFDI  lending  in  China  is  still  the  basic  acquisition  of   resources.   China’s  focus  on  OFDI  lending  for  natural  resource  acquisition  suggests  that  China’s   overseas  investments  in  natural  resources  are  far  from  slowing  down.    As  Solís   (2003a)  showed  for  Japan,  OFDI  lending  leads  OFDI.  This  suggests  that  the  Chinese   government  is  mounting  a  strong  push  for  OFDI,  and  that  Chinese  companies  are   still  in  the  process  of  reacting.  It  may  come  as  a  surprise  to  those  who  raise  alarm   over  China’s  growing  natural  resource  OFDI  that  in  fact  this  OFDI  is  likely  to  grow   considerably.   It  is  also  apparent  that  China  is  not  using  its  OFDI  loans  to  fund  “industrial   adjustment.”  In  the  entire  textile  sector,  we  found  only  one  loan  for  $40  million,  to   build  a  textile  factory  in  Mauritius.  CHEXIM  justifies  this  loan  by  stating  that  it  will   improve  overseas  business  management  and  aid  economic  development  in   Mauritius,  with  no  hint  of  industrial  adjustment.  All  other  manufacturing  loans   aimed  to  gain  access  to  either  new  markets  or  advanced  technology.  It  could  be  that,   as  was  mostly  the  case  for  Korea,  the  state  feels  that  sunset  industries  are  already   outsourcing  smoothly  enough  on  their  own.  Either  that,  or  China  is  not  so  keen  on   seeing  them  go.  This  would  be  understandable  given  China’s  regional  inequalities.   Far  away  from  the  East  Coast,  where  rising  wages  are  dragging  down  labor-­‐ intensive  industries,  China’s  West  holds  hundreds  of  millions  of  workers  willing  to   work  for  a  fraction  of  East  Coast  wages.  Rather  than  following  Japan’s  lead  and   helping  East  Coast  companies  move  abroad,  the  Chinese  government  is  attempting   to  coax  the  factories  west  through  tax  incentives  and  new  infrastructure.     China’s  second-­‐largest  OFDI  lending  motive  is  increasing  market  share,  surprisingly   dominated  by  power  companies.  This  motive  comprised  15%  of  total  OFDI  lending.   Infrastructure  projects  accounted  for  81%  of  these,  including  ports,  dams  and   power  plants.  Other  than  Three  Gorges  Group  and  China  Merchants  Group,  the   recipients  were  all  major  power  companies:  Huadian,  Huaneng,  and  State  Grid   Corporation.  Exporting  power  projects  have  received  no  attention  in  the  literature   thus  far,  though  they  do  not  fit  neatly  into  the  conventional  explanation  of  tariff-­‐ hopping.  While  Japan  and  Korea  began  OFDI  lending  to  support  tariff-­‐hopping  in  the   1980s  and  1990s  in  response  to  new  quotas  and  trade  blocs,  (Mah  and  Noh  2012   310,  Kim  2000  109),  it  still  has  not  become  a  major  motive  in  the  case  of  China.   Perhaps  more  importantly,  the  power  company  loans  double  as  export  loans  for   Chinese  goods  and  services.  Of  the  investments  with  known  destinations,  only  a  few   are  in  the  U.S.  and  Europe.  Others  seek  to  build  experience  in  developing  markets.   Infrastructure  projects  in  Sri  Lanka  and  Indonesia  involve  a  Build-­‐Own-­‐Operate-­‐ Transfer  (BOOT)  model,  requiring  the  company  to  own  and  operate  the  facility  for  a   specified  number  of  years  before  it  sells  back  to  the  government.  This  could  be   beneficial  from  the  perspective  of  learning  overseas  business  management,  though   that  is  not  one  of  China’s  sanctioned  reasons  for  OFDI  lending.  Instead,  we  suggest   that  the  banks  are  supporting  these  investments  for  the  dual  purpose  of  market   access  and  export  subsidy.  Most  of  the  loans  are  for  solar,  wind,  hydropower  and   grid  projects,  which  the  parent  companies  will  equip  with  Chinese  solar  panels  and   turbines.  It  is  well  known  that  China’s  wind  and  solar  manufacturing  is  in   oversupply  and  needs  added  demand.  From  this  point  of  view,  an  OFDI  loan  for  a   solar  farm  looks  very  similar  to  an  export  subsidy  for  solar  panels.   The  third  OFDI  lending  motive  is  acquiring  advanced  technology  through  M&A,  an   approach  rarely  employed  by  industrializing  Japan  or  Korea.  Chinese  companies  in   the  manufacturing,  oil,  auto  and  chemical  sectors  received  $3.6  billion  in  OFDI  loans   to  acquire  foreign  companies  with  desirable  technology.  This  comprises  only  4%  of   China’s  total  OFDI  lending,  but  even  so  it  represents  a  departure  from  its  Asian   counterparts.  Japan  and  Korea  industrialized  by  licensing  advanced  technologies   from  Western  industry  leaders,  who  in  order  to  protect  their  lead  offered  overly   expensive,  second-­‐rate  technologies  (Amsden  1989  150;  Shinjo  1988  342).  For  a   developing  country  drowning  in  foreign  currency  surpluses  during  a  global   economic  downturn,  acquiring  Western  companies  at  a  discount  does  not  sound  like   a  bad  alternative.  One  disadvantage  of  this  strategy  is  stirring  up  anti-­‐Chinese   sentiment  in  the  host  country.  The  U.S.  Congress  has  blocked  China’s  attempts  to   purchase  American  oil,  telecommunications  and  appliance  companies,  citing   national security threats. OFDI loans for technology acquisition would be twice as high if these loans had gone through.