China’s Influence on Deforestation in Brazilian cont.

Figure  13.  Chinese  Investment  projects  by  State  (number  of  projects).   Source:  CBBC  (2013,  p.15).   The  official  website  of  the  state  of  Mato  Grosso  reports  that  the  China  Development   Bank  Corporation  (CDBC)  intends  to  finance  the  China  Railway  Engineering   Corporation  (CREC),  which,  together  with  the  Asian  Trade  &  Investments  (ATI)  (a   holding  in  Hong  Kong),  is  interested  in  building  and  operating  a  1800-­‐km  railroad   between  Cuiabá  (Mato  Grosso)  and  Santarém  (a  port  city  on  the  Amazon  River);  the   railway  (EF-­‐170)  would  cut  through  Amazonia  beside  the  BR-­‐163  Highway  (Mato   Grosso,  2012;  Bland,  2013;  CBBC,  2013).  The  China  National  Machinery  Import  &   Export  Corporation  (CMC)  would  be  among  the  partners,  according  to  Business  News   Americas  (2011).  CREC  staff  also  met  with  the  governor  of  Mato  Grosso,  who  led  a   delegation  to  China  to  negotiate  support  for  the  planned  railway  (Lucatelli,  2012).     The  Cuiabá-­‐  Santarém  railway  (Ferronorte)  has  long  figured  in  Brazilian   development  plans  (see  Laurance  et  al.,  2001),  but  the  high  cost  has  kept  it  from   being  built  until  now.  This  railway  has  appeared  in  Brazilian  government  plans   since  the  1990s  (See:  Fearnside,  2002a),  but  the  project  did  not  rise  to  the  top  of  the   priority  list  until  China’s  interest  in  providing  a  US$  10  billion  loan  to  finance  it   25   became  clear  in  2012.  Chinese  financing  could  remove  this  barrier  (Maisonnave,   2012).     Another  project,  the  “Transcontinental  Railway”  (EF-­‐354)  is  planned  to  link  to  the   North-­‐South  Railway  at  Anapolis,  Goiás,  cutting  across  the  entire  state  of  Mato   Grosso  from  east  to  west.  It  would  connect  Lucas  do  Rio  Verde  (a  major  soy   production  area  in  central  Mato  Grosso)  to  Porto  Velho,  Rondônia,  which  is  already   connected  to  a  deepwater  port  on  the  Amazon  River  via  the  Madeira  waterway.  The   railway  project,  which  is  led  by  VALEC  (a  Brazilian  government-­‐owned  company   under  the  Ministry  of  Transportation),  would  subsequently  continue  on  from  Porto   Velho  to  Peruvian  ports  on  the  Pacific  (VALEC,  2014).  Just  the  Brazilian  portion  of   the  route  totals  4400  km,  including  a  connection  of  this  railway  from  Mato  Grosso  to   the  Atlantic.  The  president  of  the  Brazil-­‐China  Business  Council  has  suggested  that   Chinese  firms  have  “funds  to  finance  partnerships  with  Brazilian  construction   companies”  that  would  be  “strategic”  in  completing  the  rail  connection  from  Lucas   do  Rio  Verde  to  the  Atlantic  (Amaral,  2014).   These  major  infrastructure  development  projects  are  intended  to  facilitate  trade   and  save  on  logistical  costs,  improving  and  having  more  efficient  export  corridors  in   Brazil.  The  railway  can  be  expected  to  stimulate  substantial  soy  expansion  in  Mato   Grosso.  Soybean  and  Corn  Advisor  Inc.  (2012)  reports  “The  [Cuiabá-­‐Santarém]   railroad  alone  could  save  soybean  producers  in  the  state  R$2  billion  annually  in   reduced  transportation  costs.”     It  should  be  remembered  that  major  construction  projects  such  as  these  very   frequently  involve  corruption  with  significant  effects  on  decision-­‐making  in  the   Brazilian  government.  An  example  is  provided  by  revelations  regarding  the  ongoing   construction  of  Brazil’s  North-­‐South  Railway,  with  recent  revelations  being  just  the   latest  in  a  series  of  scandals  since  construction  began  in  1986  (Mello  &  Amora,   2012).  Options  for  improved  regulation  of  this  process  include  the  establishment  of   26   a  system  of  railway  use  permits  that  would  be  sold  through  auctions  organized  by   the  Brazilian  government.   7. Global  investments  for  Chinese  demand Chinese  investments  in  Brazil,  such  as  the  planned  soy  railway  from  Mato  Grosso  to   the  Amazon  River,  are  not  the  only  way  that  China’s  market  influences   infrastructure  and  deforestation  in  Amazonia.  Multinational  (as  well  as  Brazilian)   corporations  are  also  investing  with  the  intent  of  supplying  Chinese  markets.  For   example,  Bunge,  a  multinational  soy  company  currently  responsible  for  25%  of   Brazil’s  production,  opened  a  US$700  million  soy  port  in  Barcarena,  at  the  mouth  of   the  Amazon  River,  in  April  2014.  The  company  expects  its  exports  from  Brazil  to   double  in  the  next  10  years,  mainly  as  exports  to  China,  and  considers  Brazil  to  be   the  only  capable  of  responding  to  China’s  expected  increase  in  demand  in  the   coming  years  (Freitas,  2014).  In  the  future,  the  soy  to  be  exported  from  Barcarena  is   expected  to  arrive  from  Mato  Grosso  by  barge  via  the  planned  Tapajós  Waterway.   This  waterway  would  convert  the  Tapajós  River  in  Pará,  and  its  tributaries  in  Mato   Grosso  (the  Teles  Pires  and  Juruena  Rivers)  into  navigable  “hidrovias”  to  bring  soy   to  the  Amazon  River  from  the  northern  part  of  the  state  of  Mato  Grosso.  The   waterway  is  a  high  priority  in  the  “transport  axis”  of  Brazil’s  current  five-­‐year   development  plan,  the  second  “Program  for  the  Acceleration  of  Growth”  (PAC-­‐2).   Land  use  in  the  northern  part  of  Mato  Grosso  is  currently  dominated  by  cattle   pasture,  but  the  reduced  cost  of  transportation  would  lead  to  the  area  being   converted  to  soy.  The  Tapajós  waterway  is  controversial  because  it  depends  on  a   series  of  hydroelectric  dams  and  locks  being  built  to  allow  barges  to  pass  a  number   of  formidable  rapids.  Part  of  Amazonia  National  Park  has  already  been  degazetted  to   make  way  for  the  São  Luiz  de  Tapajós  reservoir  (e.g.,  WWF  Brasil,  2012).  The  São   Luiz  de  Tapajós  and  Jatobá  reservoirs  would  flood  land  of  the  Munduruku  tribe  that   has  not  yet  been  officially  designated  as  an  “indigenous  land”  (Lourenço,  2014).  The   government  plans  to  remove  part  of  the  Juruena  National  Park  to  make  way  for  the   27   São  Simão  Alto  and  Salto  Augusto  Baixo  Dams  on  the  Juruena  River  (WWF  Brasil,   2014).  Most  controversial  is  the  Chacorão  Dam,  which  would  flood  18,721  ha  of  the   Munduruku  Indigenous  Land.  This  dam  does  not  appear  in  Brazil’s  current  ten-­‐year   plan  for  energy  expansion  (Brazil,  MME,  2013)  and  in  the  “energy  axis”  of  PAC-­‐2,  but   it  is  a  key  part  of  the  waterway  plan  (Brazil,  MT,  2010)  and  it  appears  in  the  viability   study  for  the  Tapajós  Dams  (CNEC,  2014).   The  branch  of  the  Tapajós  waterway  on  the  Juruena  River  would  connect  roads  to   bring  soy  from  the  central-­‐western  portion  of  the  state,  including  Sapezal  -­‐-­‐  location   of  the  44,500-­‐ha  property  that  serves  as  headquarters  of  the  A.  Maggi  Group,  which   has  20  properties  spread  throughout  Mato  Grosso  (e.g.,  Ondei,  2012).  Blairo  Maggi,   known  as  the  “soy  king”  in  Brazil,  is  an  influential  senator  and  past  governor  of  Mato   Grosso;  in  2005  he  was  presented  with  the  Greenpeace  “golden  chainsaw  award”   (Greenpeace,  2005).  The  Juruena  River  branch  of  the  waterway  would  begin  at  a   new  port  in  Juina  on  the  Juruena  River  and  at  Porto  dos  Gaúchos  on  the  Arinos   River,  a  Juruena  tributary.  Soy  would  reach  these  ports  via  roads  from  the  south,   including  a  new  road  (MT-­‐319)  to  connect  Juina  to  Vilhena,  in  eastern  Rondônia,   bisecting  two  indigenous  areas  (Macrologística,  2011).   Blairo  Maggi  also  has  an  80,800  ha  property  in  Querência,  in  northwestern  Mato   Grosso.  This  would  export  soy  via  the  BR-­‐163  (Cuiabá-­‐Santarém)  Highway  that  is   planned  for  reconstruction  under  PAC-­‐2.  This  soy  export  corridor  is  expected  to   have  strong  impacts  on  deforestation  other  than  that  caused  by  expansion  of  soy   plantations  (Fearnside,  2007).  The  Maggi  plantation  in  Querência  would  be  linked  to   the  BR-­‐163  via  the  MT-­‐322  (former  BR-­‐080)  Highway  Reconstruction  of  this  east-­‐ west  road  bisecting  the  Xingu  Indigenous  Park  is  resisted  by  indigenous  peoples;  the   Mato  Grosso  state  government  released  a  statement  that  an  agreement  had  been   reached  to  allow  construction  (Martins,  2014),  but  the  indigenous  groups  involved   are  emphatic  that  no  such  agreement  was  made  (Mayalu  Kokometi  Waurá   Txucarramãe,  pers.  com.,  2014).  A.  Maggi  is  Brazil’s  largest  soy  company.  The   28   multinationals  Cargill,  Bunge  and  ADM  (Archer  Daniels  Midland)  are  also  present  in   the  portion  of  Mato  Grosso  to  be  served  by  the  BR-­‐163  connection  through  Pará.   Another  branch  of  the  Tapajós  waterway  would  extend  up  the  Teles  Pires  River  to   Sorriso;  this  requires  a  series  of  five  dams,  two  of  which  (Sinop  and  São  Manoel)  are   already  under  construction.  The  São  Manoel  Dam  is  located  adjacent  to  an   indigenous  area  and  has  already  provoked  conflicts  with  the  Kayabi  tribe  (ISA,   2013).       Yet  another  area  of  Mato  Grosso  that  is  expected  to  be  converted  from  pasture  to   soy  is  the  southwestern  corner  of  the  state.  This  area  would  be  opened  for  soy   export  by  the  planned  Guaporé  waterway  that  would  connect  with  the  Madeira   River  waterway  once  locks  are  installed  in  the  recently  built  Santo  Antônio  and  Jirau   Dams,  plus  one  additional  planned  dam  (Guajará  Mirim,  also  known  as  “Cachoeira   Riberão”)  (Fearnside,  2014).  Soy  would  be  transported  by  barge  to  Maggi’s  soy   terminal  and  deepwater  port  at  Itacoatiara,  on  the  Amazon  River  near  the   confluence  with  the  Madeira.  As  with  all  of  Brazil’s  major  soy  facilities,  China  is  the   main  destination  for  exports.   An  intriguing  possible  change  in  China’s  future  soy  imports  has  been  raised  by   reported  discussions  within  China  of  banning  genetically  modified  organisms   (GMOs).  Were  this  to  occur,  Brazil  would  experience  the  negative  consequences  of   excessive  dependence  on  a  single  trading  partner.  However,  China  might  find  an   immediate  switch  to  non-­‐GMO  imports  difficult  to  implement  quickly,  given  the   scale  of  the  country’s  demand  for  imported  soy.  Were  Brazil  to  attempt  to  convert   its  soy  back  to  non-­‐GMO  varieties,  the  effort  and  expense  needed  to  obtain   uncontaminated  harvests  would  be  substantial.  Ironically,  Brazil  was  one  of  the  last   major  soy-­‐producing  countries  to  make  the  switch  to  GMOs.  Even  Maggi  himself   opposed  the  switch  when  Europe  was  the  major  importer  of  Brazilian  soy  and  the   price  of  non-­‐GMO  soy  was  higher  than  that  for  GMO  soy  by  an  amount  that  made   non-­‐GMO  soy  the  more  profitable  option  (see  Fearnside,  2001a).  The  price   29   differential  subsequently  diminished,  and  GMOs  were  made  legal  in  Brazil  in  2003   (Decree  4680)  over  the  objections  of  the  Ministry  of  the  Environment.  Since  concern   over  risks  of  GMO  technology  is  greatest  among  European  consumers,  Brazil’s   switch  to  GMO  soy  undoubtedly  contributed  to  the  replacement  of  Europe  by  China   as  the  main  destination  for  Brazilian  exports.   8. Land  purchases  by  China Currently  land  purchased  directly  by  foreigners  is  limited  to  a  maximum  of  50  rural   modules  (making  the  limit  5000  ha  in  most  of  the  Amazon  region).  The  Brazilian   government  is  planning  to  lower  this  limit  with  the  express  purpose  of  inhibiting   land  purchases  by  China  (Reuters,  2011).  Among  other  effects,  the  Brazilian   government  believes  that  a  spate  of  recent  Chinese  land  purchases  is  an  important   factor  in  a  sharp  rise  in  land  prices  in  the  country  (Latin  American  Herald  Tribune,   2012).  However,  the  rising  price  of  soy  is  also  a  factor  (Agrimoney.com,  2011).   Chinese  land  purchases  in  Brazil  in  progress  in  January  2012  are  shown  in  Table  1.     Table  1:  Chinese  land  purchases  in  Brazil  in  progress  in  January  2012   Company   Area   Investment   Purpose   1. Chongquin  Grain 100,000  ha,  with  option  to   US$879  million,  much   Soybeans   Group  China expand  to  200,000  ha.   of  this  from  the  CDB   2. Pengxin  Group China 200,000  ha   Cotton,   soybeans   Source:  GRAIN  (2012).   The  China  National  Agricultural  Development  Group  Corporation,  the  Pengxin   Group  China  and  the  Chongqing  Grain  Group  have  announced  acquisitions  of  land   (GRAIN,  2012;  Raimundo  &  Azevedo,  2011).  This  is  part  of  the  strategy  connected   with  the  railroad  project  cited  above,  with  a  joint  venture  including  the  Chinese   Chongqing  Huapont  Pharm.  Co.  Ltd.  (a  Chinese  pesticide  industry),  the  Brazilian   Agricultural  Cooperative  Consortium  (which  includes  16  cooperatives  of  grain   producers  from  different  states)  and  Chinatex Corporation,  which  is  a  large  Chinese   30   state-­‐owned  company  engaged  in  producing,  trading  and  integrated  service  of   textiles  and  edible  oils  (Cintra,  2013).   This  is  undoubtedly  very  incomplete,  since  the  Brazilian  government  stopped   tracking  foreign  land  purchases  in  1994  and  only  resumed  collection  of  this   information  in  April  2012.  Restrictions  on  outright  purchases  by  foreigners  are  not   likely  to  halt  the  trend  to  increasing  control  of  land  from  abroad  because  Brazil’s   1988  Constitution  changed  the  definition  of  “Brazilian”  companies:  rather  than   requiring  a  majority  of  the  capital  to  be  Brazilian,  companies  can  be  classified  as   “Brazilian”  merely  by  having  a  headquarters  in  Brazil.  Moreover,  the  lower  house  of   the  National  Congress  is  currently  debating  the  question  of  land  acquisition  by   foreigners  and  a  number  of  influential  deputies  have  proposed  changes  in  the  law  in   order  to  relax  existing  restrictions  (Brazil,  Agência  Câmara  de  Notícias,  2012).     9. China’s  impact  on  “sustainable  development” The  changes  driven  by  China’s  commodity  purchases  from  and  investments  in   Brazil’s  Amazon  region,  including  the  state  of  Mato  Grosso,  have  significant  impact   on  the  set  of  concerns  grouped  under  the  rubric  of  “sustainable  development.”  The   term  “sustainable  development”  implies  that  what  is  happening  is  “development,”   meaning  a  change  in  a  direction  seen  as  improving  human  wellbeing  (not  to  be   confused  with  “growth,”  or  increase  in  the  throughput  of  matter  and  energy  –  or   their  proxy  in  terms  of  money  (e.g.,  Daly  &  Cobb,  1989).  The  adjective  “sustainable”   implies  that  these  benefits  will  last  for  a  very  long  time,  theoretically  indefinitely.   The  widely  used  description  of  sustainable  development  from  the  World  Council  on   Environment  and  Development’s  1987  “Bruntland  Report”  allows  the  current   generation’s  actions  to  be  considered  “sustainable”  if  future  generations,  which  are   presumably  wealthier  and  more  technologically  advanced,  are  capable  of  coping   with  the  environmental  and  social  losses  provoked  by  the  current  generation’s   activities  (WCED,  1987).  This  allows  the  current  generation’s  activities  to  be   31   considered  “sustainable”  even  if  these  activities  destroy  their  bases  of  support,  as  by   exhausting  a  non-­‐renewable  resource  such  as  a  mineral  deposit  or  by  destroying  a   potentially  renewable  one  like  a  forest  of  a  fishery.   Deforestation  destroys  a  potentially  renewable  resource,  eliminates  indigenous  and   traditional  cultures  that  predominated  in  the  area  (clearing  only  takes  place  at  some   time  after  expelling  the  traditional  inhabitants),  and  provokes  losses  of   environmental  services  such  as  maintaining  biodiversity,  carbon  storage  and  water   cycling  (Fearnside,  2008b).  Throughout  most  of  Brazilian  Amazonia,  deforestation   has  been  for  low-­‐value  cattle  pastures  that  are  ephemeral  and  that  support  only  a   few  people  after  the  initial  clearing  activity  has  passed  (Fearnside,  1986,  2002b).   Soil  and  other  limits  constrain  this  activity  (Fearnside,  1997b,  1998),  but  land   speculation,  land-­‐tenure  establishment,  money  laundering,  fiscal  incentives  and  a   variety  other  “ulterior”  motives  have  driven  clearing  for  pasture  even  in   agronomically  unpromising  locations  (Fearnside,  2005a;  Carrero  &  Fearnside,   2011).  Recent  dramatic  increases  in  beef  prices  and  the  opening  of  export  options   by  elimination  of  foot-­‐and-­‐mouth  disease  have  added  a  significant  economic  force  to   the  previous  drivers  for  pasture:  beef  that  a  decade  ago  could  only  be  exported  in   canned  form  is  now  shipped  abroad  frozen  or  even  as  live  cattle.  Is  the  conversion  of   Amazon  forest  to  cattle  pasture  “sustainable  development”?  Various  indicators   suggest  that  it  is  neither  sustainable  nor  development  (Fearnside,  1979,  1989c).   Particularly  in  the  state  of  Mato  Grosso,  soybeans  have  replaced  tropical  forest  with   a  land  use  that  can  generate  much  more  financial  return  than  does  either  cattle   pasture  or  management  of  the  original  forest.  Soy  cultivation  is  entirely  mechanized   and  generates  relatively  little  employment  per  hectare.  Those  who  own  or  are   employed  in  the  soy  plantations  enjoy  substantially  higher  incomes  than  most   Brazilians.  Soy  production  centers  in  Mato  Grosso  have  some  of  the  highest  human-­‐ development  indices  in  Brazil,  the  municipality  of  Sorriso  having  become  famous  by   being  ranked  number  one  among  Brazil’s  5570  municipalities  (Folha  de  São  Paulo,   32   2005).  Most  of  the  individuals  involved  in  soy  cultivation  are  relatively  recent   arrivals  from  other  parts  of  Brazil,  especially  the  states  of  Rio  Grande  do  Sul,  Santa   Catarina  and  Paraná,  rather  than  the  descendants  of  the  population  that  inhabited   Mato  Grosso  before  these  areas  were  converted  to  soy.   The  physical  sustainability  of  soy  cultivation  depends  on  the  purchase  (in  some   cases  including  importing  from  other  countries)  of  inputs  such  as  fertilizer,  lime,   pesticides  and  herbicides.  The  financial  viability  of  supplying,  for  example,  soil   nutrients  from  distant  sources  once  the  initial  stocks  are  depleted  depends  on  the   costs  relative  to  those  in  other  potential  production  areas.  An  example  of  potential   loss  of  competitive  viability  is  provided  by  soy  in  Bolivian  Amazonia,  where  older   plantations  in  the  Santa  Cruz  area  have  progressively  been  abandoned  in  favor  of   moving  farther  north  into  the  rainforest  portion  of  the  country  (Barber  et  al.,  1996;   Fearnside,  2001a).  Phosphorus  is  a  non-­‐renewable  resource  that  is  limiting  in   Amazonian  soils  and  for  which  global  resources  are  expected  to  be  depleted  well   before  the  end  of  the  current  century  (United  States,  CEQ  &  Department  of  State,   1980).  Brazil  is  not  particularly  well-­‐endowed  with  phosphate  deposits  (de  Lima  et   al.,  1976;  Beisiegel  &  de  Souza,  1986).  The  supply  of  phosphorous  could  be   significantly  altered  by  plans  of  the  Peruvian  government  to  build  a  railway   connecting  the  phosphate-­‐rich  area  of  Piura,  located  on  the  Pacific  coast  in  the  north   of  Peru,  with  Cruzeiro  do  Sul  in  the  Brazilian  state  of  Acre;  this  is  one  of  several   plans  for  a  rail  connection  from  Brazil  to  a  Pacific  port  in  Peru,  with  soy  export  to   China  being  the  main  justification  (Dourojeanni,  2013).  Brazil’s  planned   “Transcontinental  Railway”  would  connect  Mato  Grosso  to  this  railhead  (Marquina,   2013).   So,  is  converting  tropical  forest  (or  former  tropical  forest  that  first  is  converted  to   cattle  pasture)  into  soy  plantations  “sustainable  development”?  Different  parties   would  respond  differently  to  this  question,  depending  on  whether  they  are   economic  winners  or  losers  as  a  result  of  the  transformation  to  soy.  Those   33   concerned  with  the  environment  are  likely  to  conclude  that  encouraging  soybeans  is   not  a  wise  development  path  (Fearnside,  2001a,  2008c).   Expansion  of  soybeans  is  different  from  many  types  of  development  in  terms  of  “net   benefits”  sensu  Zarsky  and  Stanley  (2013).  Projects  such  as  establishment  of  a  mine   have  major  environmental  and  social  impacts  in  surrounding  communities,  plus   gains  from  employment  opportunities  and  from  the  monetary  flows  derived  from   wages  and  procurement.  In  such  cases,  interviews  with  community  members  and   meetings  held  in  the  area  can  reveal  strong  dissatisfaction  among  affected  people,   indicating  a  net  impact.  The  Marlin  mine  case  in  Guatemala  studied  by  Zarsky  and   Stanley  (2013)  offers  a  clear  example.  In  the  case  of  soy  plantations  in  Mato  Grosso,   however,  only  the  winners  remain  in  the  area.  The  losers  have  sold  their  land  or   been  otherwise  expelled,  and  are  now  spread  across  frontiers  elsewhere  in   Amazonia.     Successive  waves  of  displacement  have  taken  place  before  reaching  the  soy   plantation  stage.  First  the  indigenous  peoples  were  displaced  (or  killed),  followed  by   occupation  either  by  small  farmers  or  by  ranchers.  The  ranchers  may  either  follow  a   first  wave  of  small  farmers,  or,  alternatively,  may  obtain  the  land  directly  without  its   passing  through  the  small-­‐farmer  phase.  Logging  is  also  a  major  activity,  which  may   occur  in  forested  land  held  by  actors  of  any  size  from  small  farmers  to  large   ranchers.  Timber  may  be  bought  or  stolen,  including  logs  taken  from  unclaimed  land   (or  from  indigenous  areas  and  conservation  units).  Logging  is  a  temporary  and   unsustainable  phase,  and  most  loggers  move  on  to  other  frontiers  after  exhausting   the  resource  (Lentini  et  al.,  2011).  Finally,  the  land  is  bought  by  soy  farmers.   Extreme  income  concentration  has  repeatedly  been  the  result  of  conversion  of  land   to  soybeans  throughout  Latin  America  (Kaimowitz  et  al.,  1999).  The  net  balance  and   the  apportionment  of  blame  for  the  various  environmental  and  social  impacts  is   therefore  complicated.  Once  established,  soy  farmers  continue  to  clear  the   remaining  forest  despite  the  widely  believed  myth  that  they  only  plant  in  degraded   34   cattle  pasture  and  thus  provide  an  economic  boom  at  no  environmental  cost  (e.g.,   Macedo  et  al.,  2012).     10. Conclusions The  estimates  confirmed  the  effect  of  soybean  planted  area  in  increasing  exports  to   China  and  in  increasing  deforestation,  even  though  deforestation  rates  were  lower   in  2010  than  in  2000  and  the  general  downward  trend  continued  through  2012.   Exports  to  China  from  the  Brazilian  Legal  Amazon  were  also  significant  in  explaining   the  increase  in  deforestation.     The  cattle  herd  size  also  had  a  significant  relation  to  increases  in  deforestation,  but   the  relationship  of  cattle  to  soybean  planted  area  is  strong  and  positive.  Therefore,   the  change  in  land  use  from  pasture  to  soybean  cropping  (unless  it  occurs  in   degraded  areas)  may  lead  to  more  deforestation.  The  increase  in  cattle  would,   however,  lead  to  more  pasture  and  deforestation  associated  with  opening  new   areas.  The  recent  authorization  of  Chinese  beef  imports  may  lead  to  additional   deforestation.   Chinese  purchases  of  agricultural  and  forestland  and  Chinese  imports  of   commodities  such  as  timber  and  aluminum  also  cause  environmental  impacts  in   Amazonia.  Chinese  financing  and  investment  in  Amazonian  infrastructure  such  as   railways  and  mineral  processing  facilities  have  additional  impacts.   Brazil’s  ability  and  willingness  to  mitigate  the  risks  of  soy-­‐led  expansion  has  been   very  limited.  This  is  in  part  due  to  a  newly  emboldened  ruralist  class  that  has   benefited  from  the  boom.   35   12. Technical  Appendix The  system  uses  the  following  variables,  for  the  Brazilian  Legal  Amazon:   endogenous  variables  of  deforested  area  (Defor)  and  the  FOB  value  of   Brazilian  exports  to  China  (Export) ;  and  exogenous  variables  of  soybean   planted  area  (SoyArea), cattle herd (Herd), beef price (BeefPrice), soybean price (SoyPrice)5, and a trend variable (Trend). Previous---year deforestation, SoyArea, previous---year herd, beef and soybean prices and previous---year exports as instruments (the variables used in the first step of the three---stage linear system). All variables refer to the Legal Amazon from 2000 to 2010 with values expressed as logarithms. The system can then be expressed as (1) for year “t”. Dlog(Defor) t = β0 + β1Dlog(Herd)t---1 + β2Dlog(SoyArea)t + β3Dlog(Export)t + β4Dlog(SoyPrice)t + β5× Dlog(BeefPrice)t + β6 × Trend + ε1t Log(Export)  t  =  β7  +  β8Log(Herd)t-­‐1  +  β9Log(SoyArea)t  +  β10Log(Export)t-­‐1  +   ε2t (1)   where  the  variables  are  as  described  above  and  the  Dlog  operator  denotes  the  first   difference  between  the  logarithms,  or  Dlog(Xt)=log(Xt)-­‐log(Xt-­‐1).  This  is  done  to   account  for  the  increase  in  deforested  area,  as  well  as  the  size  of  the  cattle  herd,  the   area  in  soybeans  and  prices.   The  estimation  follows  the  three-­‐stage  least-­‐squares  method,  where  a  generalized   least-­‐squares  estimator  is  applied  to  a  system  of  equations  (in  this  case  two   equations)  with  a  variance-­‐covariance  parameter  matrix  estimated  in  a  previous   step  (in  the  first  and  second  stages,  the  endogenous  variables  are  regressed  against   instrumental  variables  and  forecasts  of  endogenous  variables  are  then  used  to   calculate  the  variance-­‐covariance  parameter  matrix).     Unfortunately,  at  this  time  we  do  not  have  data  on  exports  by  municipality.   Municipal-­‐level  data  would  increase  the  degrees  of  freedom  in  this  combination  of   cross-­‐sectional  and  time-­‐series  data,  allowing  spatial  regression  techniques  to  be   applied.   The  results  of  the  system  estimation,  as  in  Equation  1,  are  presented  in  Table  2.  The   system  residual  Portmanteau  tests  for  autocorrelations  showed  no  autocorrelation   at  the  90%  confidence  level.  There  were  satisfactory  fits  in  both  equations,  with   most  of  the  parameters  significant  at  the  99%  confidence  level,  except  the  intercept   in  the  second  equation.     Table  2.  System  estimation  for  the  change  in  the  Brazilian  Amazon   deforestation  and  its  exports  to  China,  2002-­‐2010   Variable   Coefficient   Standard   t-­‐Statistic   Prob.   Error   Dependent:  Change  in  Deforestation  (Dlog(Deforest))   Intercept   3.375   0.188   17.932   0.000   -­‐3 Dlog(Herd)t-­‐1 0.142   4.59*10 30.949   0.000   Dlog(SoyArea)t 0.035   1.84*10-­‐3   19.050   0.000   Dlog(Export)t 0.014   1.02*10-­‐3   13.614   0.000   Dlog(SoyPrice)t 0.043   1.24*10-­‐3   34.304   0.000   Dlog(BeefPrice)t -­‐0.020   1.11*10-­‐3   -­‐17.830   0.000   Trend   -­‐0.002   9.36*10-­‐3   Dependent:  Exports  (Dlog(Export))   Intercept   9.174   9.096   Log(Herd)t-­‐1 -­‐2.105   0.867   Log(SoyArea)t 1.987   0.557   Log(Export)t-­‐1 0.910   0.057   Increase  in  Deforestation   R2 0.9997   2 Adjusted  R 0.9986   -­‐17.982   0.000   1.009   -­‐2.427   3.596   16.007   Exports   0.347NS 0.046*   0.009   0.000   0.9912   0.9859   Source:  Research  data.  Note:  Portmanteau  tests  confirmed  the  hypothesis  of  no  residual   autocorrelations,  as  did  system  residual  normality  tests.  NS=  statistically  not  significant;  *  =   statistically  significant  at  95%  confidence;  all  other  coefficients  were  significant  at  the  99%   confidence  level.   In  the  equation  for  the  increase  in  deforestation,  the  difference  in  soybean  planted   area  as  well  as  the  previous  year  herd  exhibited  a  positive  relationship,  meaning   that  the  expansion  in  soybean  planted  area  or  in  the  herd  may  increase   deforestation.  Additionally,  increases  in  exports  to  China  and  the  world  price  of   soybean  have  a  statistical  association  with  the  increase  in  deforestation.  The  beef   price,  in  the  other  direction,  exhibited  a  negative  relationship  with  the  change  in   deforestation.   In  the  export  equation,  the  cattle  herd  size  in  the  previous  year  reduces  exports  to   China.  The  current  exports  are  positively  related  both  to  soybean  planted  area  and   to  exports  in  the  previous  year.  All  of  these  results  may  be  explained  by  the  fact  that   most  of  soybeans  in  the  Legal  Amazon  are,  in  fact,  exported,  and  that  China  is   expanding  its  imports.  This  functions  like  a  learning-­‐by-­‐doing  process,   strengthening  the  trade  relationship.