The Economist December 4th 2021 Business 69
lineadrevenuesinChinahit767bnyuan.
Howwillthefirmsrespond?Themost
outrageousinternetstunts,attractingmil
lionsofviewersandgeneratingstronge
commerce sales, have become increas
inglyimportantforgroupssuchasByte
DanceandKuaishou.Nowtheywillproba
blyblockmuchproblematiccontent,says
anexecutive.Internettrafficwillfall.
A purgeonprogrammingisalreadyun
derway.Forexampleiqiyi, China’sNetflix,
saidinAugustthatit willnolongerfeature
talentshowsorvenueswherefanscanvote
forstars.ItsNewYorklistedshareshave
tumbledbyalmost60% sincemidyear.
Douyin,Weibo,Kuaishouandotherplat
formshavealreadyshutdowntheirceleb
rityrankinglists,venueswherefansoften
paidtobuyproductsinordertosupport
theirfavouritestars.
One internet executive says that the
government’smovesdocommandpublic
support.ManyparentsinChinaagreewith
theparty’sviewofonlineentertainmentas
vacuous andeven dangerous for young
people.Regulatorsaroundtheworld are
grapplingwithhowtodealwithpotential
ly harmfulinternetcontent. ButMrXi’s
driveisextreme.MrHuo’smillionsoffans
willfindfewopportunitiestovoicetheir
oppositiontowatchingtheirfavouritestar
disappearintoa censoriousmist.n
Valuingpatents
Billion-dollar blueprints
I
t istestamenttohumaninventiveness
that 50m patents are estimated to have
been granted globally. But in aggregate
much of the collection resembles an intel
lectual junkyard. Included are plausible
ideas that no firm ever wanted to pay for,
plausible ideas that fell short, and absurdi
ties. A patent on the crustless peanutbut
terandjelly sandwich, for example, failed
to be renewed in 2007.
Pare the list to those that are both sensi
ble and in force legally, meaning a fee is
paid to a patent office to keep them alive,
and there are 16m patents that count. Last
year, 1.6m were granted.
Most are the property of companies, but
balancesheets and conventional account
ing are illsuited to capturing their worth.
Using acquisition cost, then depreciating
it, does not work. Instead, lawyers provide
subjective numbers based on factors such
as a patent’s likely validity, royalties and
litigation history. Many firms reckon it is
not worth paying the tens of thousands of
dollars that it costs for a valuation.
In 2008 an intellectualproperty ex
change opened in Chicago to do for patents
what other bourses did for stocks, bonds
and commodities. Its backers were blue
chip firms like Hewlett Packard and Sony,
but it closed in 2015. Patents cannot be
treated like commodities, said the Cornell
Law Review. A subsequent effort to value
them used software to read and evaluate
the documents. So far, however, not even
machinelearning techniques have al
lowed code to penetrate the opaque legal
language in which patents are couched.
Now a startup called PatentVector,
founded by a law professor, an informa
tionscienceprofessoranda softwareengi
neer, is trying something new. It uses a va
riation of a method started in the 1960s
which evolved into tallying up how fre
quently individual patents are cited (a sim
ilar process based on citations is used to
evaluate academic research).
Rather than attempting to understand
the patent, PatentVector employs artificial
intelligence to comb through 132m patent
documents kept by the European Patent
Office in Munich (the world’s biggest col
lection). Then it evaluates, first, how fre
quently a patent is cited and, second, how
frequently it is cited by patents that are
themselves cited frequently. That provides
an indication of importance which is then
multiplied by a mean value of patents
based on an estimate by James Bessen, an
economist at Boston University, which has
become a reference point. A number of
companies, legal firms and institutions
(including the Canadian Patent Office) are
buying PatentVector’s product.
The results contain interesting insights
into inventing. Frederick Shelton iv(pic
tured) does not feature among prominent
innovators of the 20th century but he prob
ably should. He works at Ethicon, a medi
caldevices subsidiary of Johnson & John
son, and PatentVector values his inven
tions at a cool $14bn, placing him aeons
ahead of anyone else. His top three are for a
mechanical surgical instrument, surgical
staples and the cartridge for the staples; in
short, tools to cut tissue and bind it up.
Ethicon itself, a medicaldevice maker,
holds 95 of the world’s 200 most valuable
patents, PatentVector finds. The firm also
employs Jerome Morgan, who is listed in
second place with $5bn worth of patents
(many overlapping with Mr Shelton’s). On
ly one other person is in the $5bn club:
Shunpei Yamazaki, president of Semicon
ductor Energy Laboratory, a Japanese re
searchanddevelopment firm. Mr Yama
zaki’s most important patent covers the
displays on computers, cameras and other
semiconductor devices.
PatentVector found 65 other people
each responsible for patents worth in ex
cess of $1bn. Only 14 of the top 650 tinker
ers are women. The highest ranked is Mar
ta Karczewicz, who works for Qualcomm,
an American chip designer, and played a
vital role in inventing the videocompres
sion technology that makes Zoom and oth
er video services function.
Almost all valuable patents can be
found in a few broad industry groups: bio
pharma, software, computer hardware,
medical devices and mechanical equip
ment. Over the past 40 years the impor
tance of specific categories has marginally
expanded and contracted, but biopharma
and information technology (it) have
dominated and their significance has
grown. The companies with the largest ag
gregate value of patents are in it,topped by
ibm, Samsung and Microsoft.
PatentVector’s figures on the patent
holdings of countries are revealing, too.
America has the most active patents of any
country, at 3.3m, followed closely by China
with 3.1m. But there is a world of difference
in how frequently they are cited and their
imputed value. America’s library is calcu
lated to be worth $2.9trn, compared with
China’s collection at $392bn.
Of course, PatentVector’s methodology
will face scrutiny. Naturally, the startup
has patented its own technique. Informa
tion about patents, which are critical com
ponents of invention, has never been more
important. Perhaps it was inevitablethat
innovation would be applied notonlyby
means of patents, but to them as well.n
N EW YORK
A new way of understanding the high but elusive worth of intellectual property
Fred and his 1701th invention