Geography Reference
In-Depth Information
Table 16
.
2
Inventors' activity across assignees, MSAs*
No. of assignees joined
No. of inventors
(% of all inv.)
% of inventors
active in >1 MSA
1
46458 (73.52)
2.3
>1
16730 (26.48)
28.4
of which
2
10645 (16
.
85)
22.8
3
3679 (5
.
82)
34.1
4
1439 (2
.
28)
40.3
5
562 (0
.
89)
46.4
6-10
392 (0
.
62)
46.7
>10
13 (0
.
02)
53.8
All inventors
63188 (100)
9.2
Note:
* The table reports the distribution of all inventors in the sample, according to the number of
dif erent assignees for which they have recorded patents over the period examined. The calculation includes
patents registered by individual inventors, i.e. not assigned to organizations. The calculation includes also
patents that have been co-assigned to dif erent organizations.
inventive career (which, in many cases, is limited to only one patent). Most 'cross-i rm'
inventors sign patents for just two dif erent assignees (17 per cent of all inventors; 64
per cent of cross-i rm inventors); only a very few of them sign patents for more than i ve
assignees (0.6 per cent of all inventors; 4 per cent of cross-i rm inventors; Table 16.2,
second column).
If inventors' cross-i rm activity looks limited, mobility in space looks even more
limited. Only 28.4 per cent of all cross-i rm inventors (9.2 per cent of all inventors) have
been active in more than one MSA. Figures grow with the number of assignees the inven-
tors have signed patents for, and go over 40 per cent for cross-i rm inventors with four
assignees or more. The latter, however, are very few. Almost no inventor has been active
in more than two MSAs.
11
Table 16.3 reports some descriptive statistics for the one-mode network of inventors.
By construction, the size of the network changes over time (see the social networks sub-
section above), with a tendency to expand because of the 'patent explosion' of the 1990s
(Hall, 2004). Table 16.3 shows how the network of inventors exhibits a characteristic
typical of 'small world' network of scientists, namely the asymmetry in size between the
principal component (which in 1999 includes 46 per cent of the inventors) and all the
other components (in 1999, the second largest component collects only 0.9 per cent of
the inventors).
12
Inventors from dif erent companies are linked one to another by cross-i rm inven-
tors. To the extent that the latter do not move much across cities the resulting social
network will be concentrated in space. Table 16.4 coni rms this intuition for the princi-
pal component: C5 and C4 concentration indexes always result above 50 per cent, when
calculated for active inventors (namely, those inventors in the network who sign patents
in the current year). These i gures are relatively stable over time, despite an increase in
the number of MSAs reached by the expanding network, which suggests that peripheral
areas host just a very few inventors.
