Productivity of Research Scientists Jinyoung Kim, Sangjoon John Lee, Gerald Marschke

1
Productivity of Research ScientistsJinyoung
Kim, Sangjoon John Lee, Gerald Marschke

• Thoughts from
• Alex BrysonPolicy Studies Institute
• SEWP Research Conference, NBER
• 19TH-20TH October 2005

2
Patents as Productivity

• To patent or not to patent
• Not all innovations are patented
• Depends on rents to be earned (James Bessen)
• All patents are not the same
• Weighting by value
• typically assigned
• WHOSE patent? If workers then raises market
  value. If firms raises their market value.
  Important to use assignee type and explain
  institutional context
• of semi-conductor and pharmaceutical
  inventors patenting outside of these two
  industries

3
Patent-Inventor Ratio by Foreign-Experience

• Gives 2 reasons on page 19
• Causal impact of foreign experience
• Migration of more productive inventors to the US
• 3rd possibility
• Foreign experience proxies some other
  productivity enhancing factor, eg. Age which is
  associated with accumulated human capital
• Test in multivariate framework
• Increase in relative productivity of current US
  residents with foreign experience since 1993
  coincides with decline in their relative numbers
• Suggests possible selection effect
• Expected to see Figure V-1 for the two industries
  that are central to the rest of the paper

4
Determinants of citation to foreign-assigned
patents

• Clear effect of inventor foreign experience
  (knowledge spillover)
• Split this dummy by whether current foreign
  resident or ex-foreign resident
• Direct test of whether increase in use of foreign
  resident inventors is having positive spillovers
  for firms
• If extended analysis by pooling years interact
  these two dummies with time to see if these
  returns have risen
• Might explain why we see increase in foreign
  resident inventors
• Some puzzling results that need further
  investigation
• Difference in firm age effects across 2 sectors

5
Data issues

• Meticulous data construction exemplary
• Addresses and animal rights
• Degree/higher degree rates seem low
• How complete are the 1k universities
• Foreign qualifications beyond US/Europe
• Multiple patent assignments each year
• Whose employee?
• Industry dynamics
• Semicon firms smaller throughout but increase in
  N employees per firm whereas fall in N employees
  per firm in phara
• Increase in sales and RD in both industries but
  faster in semiconductors
• Implications if any?
• Four pharmaceutical firms missing from Table
  II_6a

6
Implications of Compact as Firm Universe

• Dont know all patents covered
• 5m assets, 500 shareholders
• SIC only in Compact data?
• Lots of innovation below this
• Some firms with very low N employees (Table II-6)
• How so? Subsidiaries of larger firms?
• Fall in of firms reporting employment in both
  sectors over time (Table II-1). Not true of
  sales/rd.
• Why?
• Any bias introduced?

7
Why Do Foreign Inventors Remain Home-based?

• No need to move
• Telecom improvements (authors suggest)
• Implies codification of tacit knowledge
• Cheap flights
• Demand outstripping supply
• May explain why trends so pronounced in
  industries identified.
• Could do more to elaborate on market context
• Laboratories abroad (authors suggest)
• But why? Labour cost reduction? Tax
  breaks/burdens
• Changes in nature of collaboration
• N inventors per patent
• Increased specialisation in contribution of
  inventors
• Increased ability to codify