Current Policy Debates: Patents

Một phần của tài liệu The Economics of Trade Secrets: Evidence from the Economic Espionage Act (Trang 37 - 44)

2.3 Policy Issues and Current Debates

2.3.1 Current Policy Debates: Patents

Schumpeter  (1934)  defined  our  modern  interpretation  of  innovation  as  an   applied  invention.    Since  then,  researchers  have  sought  to  determine  how   innovation  occurs,  how  it  can  be  encouraged,  how  it  affects  economies  and  how   firms  interact  in  innovation.    There  are  four  main  areas  of  research  within  patent   policy,  all  of  which  have  in  mind  the  ultimate  goal  of  maximizing  social  surplus.    

These  areas  are:  cumulative  innovations,  knowledge  diffusion,  patent  scope  and   breadth,  and  harmonization.  Cumulative  innovation  and  knowledge  diffusion  are   aspects  of  encouraging  innovation,  which  need  to  be  accounted  for  in  IP  policy.    

Patent  scope  and  breadth,  along  with  international  patent  system  harmonization,   examine  the  design  and  implementation  of  policies.  

Patents  and  Knowledge  Diffusion    

An  important  aspect  of  patent  policy  design  is  the  disclosure  incentive  it  

incorporates.    In  return  for  being  granted  a  legal  monopoly,  the  innovator  must   disclose  information  related  to  the  innovation  (as  in  Scotchmer  and  Green,   1990.)    The  goal  of  this  disclosure  policy  is  to  increase  the  spread  of  knowledge   and  encourage  cumulative  innovation,  which  subsequently  increases  social   surplus.    As  Scotchmer  and  Green,  (1990)  argue,  the  “social  goal”  of  disclosure   allows  for  “shared  technical  information  [that]  would  help  other  innovators  in   their  own  research,  reduce  redundancy,  and  hasten  the  time  to  subsequent  

innovation.”24    From  the  firm’s  perspective,  disclosure  may  not  be  desirable  as  it   can  benefit  competitors,  as  noted  in  Scotchmer  (2005.)    The  disclosure  

surrounding  patents  reveals  technological  advances  and  can  indicate  the   strategic  plans  of  a  company  (Baker  and  Mezzetti,  2005.)    With  cumulative   innovation,  disclosure  by  one  firm  may  allow  a  competitor  to  leapfrog  (as  in   Fudenberg  1983.)    This  is  one  reason  why  trade  secrets  may  be  employed,  as  will   be  discussed  later.  

 

Breschi  and  Lissoni  (2007)  use  patent  data  to  study  knowledge  diffusion   between  researchers.    Using  patent  data,  they  map  knowledge  networks  and   highlight  the  concept  of  technological  gatekeepers,  which  are  figures  that  sit  at   the  centre  of  and  link  important  research  areas.      Their  findings,  while  

inconclusive,  point  to  the  concept  that  social  and  physical  distance  may  have   greater  influence  on  knowledge  diffusion  than  the  disclosure  of  knowledge   through  patenting.    This  suggests  limitations  to  the  effectiveness  of  patents  as  a   mechanism  for  disclosure.  

Patent  Scope  and  Breadth  

 In  addition  to  examining  patent  policies  as  a  whole,  examining  patent  duration   and  breadth  was  also  a  popular  research  topic  in  the  1990s.    Patent  duration  has   obvious  research  appeal  (Gold  and  Gruben,  1996)  as  it  is  merely  the  length  of  the   patent  (typically  20  years).    Additionally,  patent  duration  is  dependent  on  the   owner  of  the  patent  paying  renewal  fees,  which  can  indicate  the  value  of  the   patent  (Shankerman  and  Pakes,  1984;  Shankerman,  1998;  and  Lévêque  and   Ménière,  2006.)    However,  this  represents  a  simplistic  view  as  patent  duration   may  not  be  the  same  as  the  market  duration  of  an  innovation.    Thus,  the  focus  of   research  shifted  to  effective  patent  duration,  as  in  Denicolo  (1996),  which  is   more  difficult  to  measure.    The  concept  of  effective  patent  duration  highlights  the   fact  that  a  multitude  of  factors,  including  market  conditions,  type  of  innovation,   industry  and  cumulative  innovation,  can  affect  the  optimal  policy.      

 

                                                                                                               

24  Scotchmer  and  Green  (1990),  p.  132.  

A  second  focus  of  patent  policy  research  is  that  of  patent  scope  or  breadth.    This   line  of  research  examines  how  much  a  patent  covered  in  terms  of  determining   what  similar  innovations  could  be  patented,  what  is  considered  infringement  and   how  this  is  interpreted  in  the  courts.    From  an  economic  perspective,  breadth  is   the  closeness  of  non-­‐infringing  substitutes;  or,  as  Gallini  (1992)  identifies,  the   cost  of  inventing  around  a  patent.    This,  however,  presents  an  empirically   challenging  model  as  patent  scope  is  a  less  tangible  measurement  and  is  often   determined  in  court  cases  and  practice  rather  than  written  policies  (Chang,   1995.)      

 

Patent  length  and  breadth  are  two  policy  tools  that  counterbalance  each  other.    

Longer  patents  are  typically  balanced  by  having  narrower  patent  breadth.    This   argument  is  developed  by  Klemperer  (1990),  who  examines  both  infinitely  lived,   narrow  patents  and  short-­‐lived,  broad  patents  looking  at  product  differentiation.    

He  argues  that  the  market  conditions  of  the  demand  side  (consumer  

preferences)  determine  the  ideal  policy.    However,  in  the  same  issue  of  the  same   journal,  Gilbert  and  Shapiro  (1990)  develop  a  similar  argument,  assuming  

homogeneous  products,  which  calls  for  infinitely  lived  patents  offset  by  adjusting   patent  breadth.    Yet  both  of  these  papers  acknowledge  the  fact  that  they  do  not   consider  cumulative  innovations,  which  could  dramatically  alter  the  results.      

Gallini  (1992)  furthers  the  Klemperer  and  Gilbert  and  Shapiro  models  by   allowing  for  the  possibility  of  costly  imitation.    Contrary  to  their  papers,  she   argues  that  the  ideal  patent  length  is  shorter  because  longer  patent  lengths   encourage  others  to  invent  around  patents  and/or  infringe.    Furthermore,  the   consequence  of  a  longer  patent  length  may  actually  be  decreased  R&D  

expenditures  as  innovators  face  the  increased  likelihood  of  losing  their   monopoly  as  their  competitors  invent  around  the  patent  or  infringe.    Gallini   concludes  that  the  optimal  policy  is  broad  patents  that  limit  imitation  and   adjustable  patent  lengths,  which  are  adjusted  to  achieve  the  desired  reward  to   the  innovator.  

 

Reid  and  Roberts  (1996)  focus  on  the  scope  of  patenting  as  opposed  to  its  

duration  by  examining  the  patent  width,  defined  as  the  technological  coverage  of  

the  patent,  and  breadth,  which  they  define  as  the  scope  for  exploiting  the  market   power  which  the  patent  rights  confer  on  the  innovator.    They  use  this  to  analyse   the  patenting  activity  of  the  scientific  instruments  industry  in  the  UK  and  

conclude  that  patent-­‐active  firms  are  typically  larger  than  non-­‐patent  active   firms  and  suggest  that  the  patent  activity  has  a  non-­‐linear  relationship  with  firm   size  characterized  by  diminishing  returns.      

 

Scotchmer  (2005)  refines  the  breadth  argument  further  by  making  an  important   distinction  when  analysing  patent  policy.    She  separates  patent  breadth  from  the   inventive  step,  whereas  other  authors  treat  them  singly.    Scotchmer  notes  that   the  breadth  determines  how  different  another  innovation  must  be  in  order  to   avoid  infringement.    The  inventive  step  determines  what  is  patentable.    The   product  space  of  these  concepts  is  represented  in  Figure  2-­‐1,  where  policy   determines  the  infringement  and  patentability  of  products.    For  product  A,  the   inner  circle  is  infringing  and  unpatentable  substitutes  due  to  the  narrow  patent.    

The  outer  circle  are  substitutes  that  are  not  considered  infringing,  but  do  not   meet  the  inventive  step  required.    For  product  B,  the  inner  circle  is  infringing  due   to  the  small  inventive  step,  while  the  outer  circle  is  considered  infringing  due  to   the  broad  patent.25  

                                                                                                               

25  Scotchmer  develops  an  example  of  this  concept  using  the  laser.    The  maser,  which  had  a  broad   patent,  was  a  blocking  patent  to  the  laser.    The  maser  would  be  product  B  in  Figure  1,  while  the   laser  would  fall  in  the  outer  circle.    See  Chapter  3  of  Scotchmer  (2004)  for  further  information.  

Figure  2-­1:  Scotchmer’s  Patent  Breadth  and  Inventive  Step  

 

The  concepts  of  patent  length,  breadth,  scope  and  the  inventive  step  are  

important  policy  tools  available  to  IP  regimes.    However,  as  this  section  notes,  no   consensus  has  been  reached  on  the  ideal  design  of  IP  policy.    Therefore,  IP  

regimes  often  employ  a  mix  of  these  tools  and  vary  significantly  across  regimes.  

Patent  Law  Harmonization    

The  movement  towards  harmonization  of  patent  laws  has  also  proven  to  be  a   fertile  research  ground.    Harmonization  seeks  to  standardize  IP  protection   internationally,  which  is  hoped  to  encourage  investment  in  developing  countries,   promote  innovation  and  reduce  the  costs  of  achieving  IP  protection  (Pitkethly,   1999.)    The  signing  of  the  Agreement  on  Trade  Related  Aspects  of  Intellectual   Property  Rights  (TRIPS)  in  1994  created  a  natural  experiment,  which  allows  for   researchers  to  observe  the  effects  of  harmonization.    Authors  such  as  Beath   (1990)  model  the  effects  of  TRIPS  on  developing  countries,  international  trade,   etc.  While  the  effects  may  not  be  seen  for  some  time,  they  call  attention  to  the   conflicts  surrounding  harmonization.    The  main  points  of  contention  are  the   enforcement  of  IP  laws  and  the  level  of  protection  provided.    On  a  more  subtle  

Patent  A:  Narrow,  large  required   inventive  step  

Patent  B:  Broad,  small  required   inventive  step  

A   B  

Infringing,   unpatentable   Non-­‐infringing,   unpatentable   Infringing,   patentable  

Patent  Policy  product  space  

note,  the  first-­‐to-­‐invent-­‐rule26  of  the  US  versus  the  international  standard  of   first-­‐to-­‐file  and  the  patentability  of  business  methods  and  organisms,  provide   examples  of  underlying  ideological  conflicts,  as  examined  in  Kotabe  (1992)  and   Kawaura  (2005.)  

 

An  example  of  the  efforts  to  harmonize  national  systems  and  develop  an   international  standard  is  the  European  Patent  Convention.    This  agreement   created  an  international  filing  system  that  reduces  foreign  filing  fees  and  is   developing  a  unified  opposition  system  (Caldirini  and  Scellato,  2003.)    Unlike   trademarks  and  geographical  indicators,  which  are  internationally  unified  within   the  EU  and  fall  under  the  jurisdiction  of  the  Office  for  the  Harmonization  of  the   Internal  Market  (OHIM),  patents  enjoy  no  such  consistency.    However,  the  British   Patent  Office’s  Gowers  Review  (2006)  of  IP  notes  the  high  cost  of  translation  fees   associated  with  filing  in  multiple  countries.    Despite  these  costs,  the  fact  remains   that  patent  offices  generate  significant  revenue  for  governments  at  the  national   level.    The  vested  financial  interests,  national  pride  and  ideological  differences   have  created  challenges  to  a  unified  system  (Pitkethly,  1999.)    

Patents  and  Cumulative  Innovations    

If  I  have  seen  further,  it  is  only  because  I  am  standing  on  the  shoulders  of   giants.      

Newton  (1676)    

In  his  now  clichéd  phrase,  Newton  typifies  the  concept  of  cumulative  innovation   by  stating  that  his  advances  were  only  possible  due  to  the  advances  of  other   scientists  before  him.    Innovations  can  thus  be  divided  into  two  categories:  

independent  (i.e.  one-­‐off  or  isolated)  and  cumulative  (i.e.  sequential),  as  in   Scotchmer  (2005.)    Cumulative  innovations  are  a  series  of  innovations  that  build   on  each  other,  whereas  independent  innovations  do  not  spawn  subsequent   innovations.      

 

                                                                                                               

26  The  first-­‐to-­‐invent  rule  in  the  U.S.  dictates  that  the  first  entity  to  invent  the  innovation  should   have  the  right  to  the  patent.    This  differs  from  the  first-­‐to-­‐file  rule,  in  which  the  first  entity  to  file   the  patent  has  the  right  to  the  patent.  

There  are  numerous  variations  of  cumulative  innovations,  as  Scotchmer  (1991)   defines  in  her  seminal  paper  on  cumulative  innovation.    The  first  innovation  may   be  a  research  tool  without  a  commercial  application.    It  may  be  a  small  or  large   innovation.    The  second  innovation  may  be  the  commercially  successful  

application  of  the  tool,  a  great  leap  forward  (a  radical  innovation),  or  merely  a   minor  improvement  (an  incremental  innovation.)    For  example,  Percy  Spencer   invented  the  microwave  oven  when  he  realized  that  the  magnetron  he  was   working  with  melted  the  chocolate  bar  in  his  pocket;  Spencer  developed  a  hugely   successful  commercial  application  for  the  magnetron  that  had  been  used  

primarily  for  ship  navigation.27    However,  the  first  innovation  is  a  necessary   predecessor  to  the  second.    Therefore,  the  first  innovator  is  only  appropriately   rewarded  once  the  profit  and  social  surplus  of  the  second  is  taken  into  account   (Scotchmer  1991;  Encaoua  and  Lefoulli,  2006.)      

 

Scotchmer  also  notes  the  difficulty  in  rewarding  each  generation  appropriately.  

In  order  to  give  the  second  innovator  an  incentive  to  invent,  they  must  receive   some  of  the  surplus.    However,  the  first  innovator  must  also  be  rewarded  with   some  of  this  surplus.    This  inherent  conflict,  referred  to  as  a  double  

marginalization,28  makes  the  one-­‐size-­‐fits-­‐all  approach  to  patenting  inefficient,   as  the  surplus  generated  by  the  second  innovation  cannot  be  given  to  both  

innovators.    One  of  her  suggested  solutions,  also  addressed  by  others  (Cugno  and   Ottoz,  1991;  Choi,  2004;  and  Anand  and  Kahnna,  2000),  is  to  allow  more  

collusion  between  firms  through  licensing  and  other  means.    This,  however,   poses  anti-­‐competition  problems,  as  Chang  (1995)  also  notes,  due  to  the  anti-­‐

competitive  nature  of  collusion  agreements.    Chang  examines  collusion  in  the   context  of  cumulative  innovation  and  licensing  and  concludes  that  allowing   collusion  may  attract  inefficient  entry  by  imitators  who  invent  around  the   original.    Even  Thomas  Edison  recognized  the  process  of  cumulative  innovation   in  his  quote,  “I  start  where  the  last  men  left  off.”  

 

                                                                                                               

27  See  Lemelson-­‐MIT  Program,  1996.  

28  Scotchmer  (1991),  p.  34.  

Một phần của tài liệu The Economics of Trade Secrets: Evidence from the Economic Espionage Act (Trang 37 - 44)

Tải bản đầy đủ (PDF)

(328 trang)