A comparison of the use and value of patents and trade marks in large and small firms

1
A comparison of the use and value of patents and
trade marks in large and small firms

• by Mark Rogers, Christian Helmers and
  Christine Greenhalgh
• Harris Manchester College, Oxford University
• Wolfson College, Oxford University
• St Peters College, Oxford University
• and all associated with the
• Oxford Intellectual Property Research Centre
• Seminar at Melbourne Institute for Applied
• Economic and Social Research

The research was supported by the UK IP Office
and UK Trade and Investment.
2
Intellectual Property and firm size

• Well known that IP is widely used by large firms
• Many studies of these large firms show benefits
  of both patents and trade marks
• One hypothesis concerning SMEs is that they may
  do less innovation, so obtain less IP
• SMEs may have a lower propensity to protect their
  innovation, so again leads to less IP
• Rather few studies have comprehensive data on IP
  in SMEs
• Notable exception is Jensen and Webster for
  Australia see Economic Record, March 2006

3
Some reasons for less IP in SMEs(full survey of
factors in Jensen and Webster)

• SMEs do less innovation investment as face higher
  risk, e.g. bankruptcy, cannot diversify risk as
  in large firm with many product lines for RD
• Liquidity constraints affect ability to undertake
  RD investment with long pay-back period
• SME applies for less IP per innovation, as
  lacking in information about procedures and
  cannot afford an IP specialist in management team
• Cost of IP application high in relation to
  turnover and potential litigation costs also
  high, so see less potential return to IP
• All these factors would lead to lower IP
  intensity (IP relative to size of firm) in SME

4
Our objectives in this study

• Document for the first time for the UK the actual
  IP use by SMEs and compare this with larger firms
  and micro firms
• Explore the determinants of IP intensity
  (measured relative to assets) focusing on the
  role of firm size
• Analyse the comparative performance of IP active
  and inactive firms to see if IP is good for SMEs
• Performance to be investigated by analysis of
  firm survival rates does IP help or hinder?
• Also explore firm asset growth in years following
  acquisition of new IP

5
Definitions

• Small and medium enterprises (SMEs)
• 2m Euros lt total assets lt 43 million Euros
• 10 lt employment lt 250
• Euro 10m lt turnover lt 50m Euros
• Subsidiaries of large UK firms are not classed as
  SMEs
• Problem in that some SMEs have foreign parents of
  unknown size we can exclude later
• Micro firms have assets lt 2m or missing, but if
  owned by an SME reclassified as an SME
• Large firms have assets gt43m Euros
• MR calls this the Oxford Firm Level IP (OFLIP)
  data

6
UK companies in Fame database (actively trading
in 2005)
The FAME Dec 2006 data also contains 926,503
firms that are now classified as inactive. The
FAME database only removes inactive firms after
five years from dissolution so the record during
2001-2006 is complete.
7
Matching these firms IP acquisitions

• Matching of patents and trade marks registered in
  either (or both) the UK or in Europe
• Matching conducted by company names recorded in
  the IP application
• Also matched IP for firms that existed during
  2001 to 2005, but listed as inactive by 2006
• Inactive includes live but not trading, in
  receivership, dissolved, and liquidated firms
• Total sample approx. 3m firms of which 2.1m live,
  but bulk of population is micro firms

8
All SMEs and IP active SMEs 2001-2005
 For comparison, during 2001-2005, 5.4 of large
firms and 0.8 of micro firms were IP active in
one or more IP types
9
Benchmarking the matching outcome (IP
publications in 2003)
Not expecting 100 as official data is for UK
residents individual, corporate, university, or
government agency
10
Numbers of IP assets by year and type
11
Numbers of patents by year and class of firm
12
Numbers of trade marks by year and class of firm
13
The facts so far

• UK has a large number of SMEs and even larger
  number of micro firms
• A small but significant proportion of both
  classes were actively seeking patents and trade
  marks during 2001-05
• The absolute number of patents by SME plus micro
  firms stands comparison in scale with the total
  for all large firms over the period
• By 2005 the within-year total for SME micro
  patents exceeded that for large firms
• The absolute number of trade mark applications by
  SME plus micro firms exceeded that of large firms
  in each year of the study

14
IP active SMEs and IP assets per firm (by
sector 2001-2005)
15
Median IP intensity of firms without foreign
parents by size class(IP intensity per 1m
assets)
16
Regressions of IP intensity (trimmed at 95th
ile)(include year, region and industry dummies)
17
Conclusions so far

• SME and micro firms are more IP intensive
  relative to their asset base than are large firms
• Declining IP intensity with size is confirmed in
  multivariate regressions controlling for many
  firm characteristics (including industry and
  region)
• Older firms are more IP intensive
• Firms with foreign parents are more IP intensive
• Co-location with a university increases EPO
  intensity (not strongly significant)
• Regional dummies are insignificant once industry
  dummies are included
• No trends exist in IP intensity over 5 years

18
What outcomes for SMEs in 2001 by 2004? Are they
related to IP activity?
19
Survival outcomes depend on many factors

• Firm-level
• IP use, experience, strategy, human capital, etc
• Industry level
• stage of product cycle, competition, growth
  rates, innovation, spillovers, etc
• Regional and macro level
• Initially use probit analysis exit20041, 0
• Explanatory variables all from 2001
• IP dummies (0,1) e.g. did SME TM in 2001?
• Industry IP intensities

20
Using IP to create industry variables
21
Probit Exit 1, Survive 0
22
IP and survival - interpretation

• Own IP activity in 2001 has some impact
• UK trade mark(s) reduces prob(exit) by 0.015
• Community mark(s) have impact for older firms
• Industry variables
• More trade marking by large firms tends to reduce
  SME exit (1 sd ? ?0.004 prob)
• More trade marking by SMEs increases exit (1 sd ?
  ? 0.003 prob)
• SME industry patenting may imply spillovers (but
  none from large firms) but crude method used
  here (no proximity weighting)

23
Asset growth of 2001 SMEs
Total assets available for 82 of initial
sample. All cross-tabs statistically
significantly different from non-IP active firms
24
Turnover growth of 2001 SMEs
Turnover available for 28 of SMEs. Cor(growA,
growT)0.44 0.52 in Manu UK patent distribution
not statistically different from non-IP SMEs,
other IP types are different
25
Modelling firm growth

• More intuitive and revealing to re-write

Asset growth mean55, median1,max193,000 Tur
nover growth mean75 , median3.7 IP active
firms have similar distributions . an example of
results
26
(No Transcript)
27
Growth summary

• OLS regressions sensitive to sample
• Robust regression and median least squares as
  checks
• Main results
• SME UK trade marking associated with increased
  growth (3 to 6 p.a.)
• No. with UKTM 1670 SMEs in 2001 (1.58 of
  sample)
• Some evidence of positive association with
  Community trade marks
• Patents mostly have little association with
  growth, sometimes negative
• Of course, only looking at growth in next three
  years (and patents published), so maybe growth
  long term
• But, not encouraging for SME patent use (UK or
  EPO)

28
Conclusions about innovation and IP in SMEs

• Our evidence firmly refutes view that SMEs
  innovate less than larger firms (pro rata)
• (unless prepared to argue that large firms are
  innovating more, but have lower patent and
  trademark propensity!)
• Findings on IP intensity support the view that
  SMEs do perceive value of IP protection
• Traditional view that SMEs are so disadvantaged
  that they cannot use IP is firmly rejected
• Cannot yet positively establish level playing
  field in costs and returns compared with larger
  firms

29
Conclusions about IP and performance

• Exit probability of SMEs within three years of IP
  activity is reduced in case of UK trade marks
• Exit of older firms is reduced by Community trade
  marks
• Growth of assets and turnover in all SME firms
  are enhanced by UK trade marks
• Other types of IP use are not significant for
  exit or growth over this short period
• Seems likely that the returns to patents take
  longer given that we observe similarly high
  intensity of patents and trade marks in SMEs