Page 73 - IJPS-9-2

International Journal of
Population Studies Relationship between population aging and innovativeness

A B

C D

Figure 2. Patent applications per 100,000 residents by quartile, 1985 – 2019
Note: Country ISOCODES: AUT: Austria; BGD: Bangladesh; BEL: Belgium; BGR: Bulgaria; BRA: Brazil; CAN: Canada; CHL: Chile; CHN: China; HKG:
Hong Kong; CRI: Costa Rica; DNK: Denmark; FIN: Finland; FRA: France; DEU: Germany; GRC: Greece; GTM: Guatemala; IND: India; ISL: Iceland; ISR:
Israel; JPN: Japan; KOR: Republic of Korea; MYS: Malaysia; MEX: Mexico; NLD: Netherlands; NZL: New Zealand; NOR: Norway; PHL: the Philippines;
POL: Poland; PRT: Portugal; ROU: Romania: ZAF: South Africa; ESP: Spain; SWE: Sweden; CHE: Switzerland; THA: Thailand; TUR: Turkey; GBR: United
Kingdom; USA: United States. Low, medium, medium-high, and high refer to each quartile of the patent applications for inhabitants (PAI) in 2019.

Jones (2009) introduced the concept of the “burden of forms of heteroskedasticity), the two-step estimator
knowledge.” In particular, Jones argued that the higher with robust standard errors, and the same model as in
the level of knowledge that is reached, the more difficult it Column 1 but with the number of trademark applications
becomes to produce a newer one. This implies, for instance, incorporated for analysis. We also report the Arellano-
that in the academic and/or the scientific ambit, researchers Bond’s test for all the models, whose null hypothesis is that
need to spend more time on the path to discoveries and autocorrelation in first-differenced errors is zero. Then, a
innovations, and as a result, they are no longer young by rejection of the null hypothesis indicates that the model is
the time they attain the first achievements; thus, the life misspecified.
cycle of a researcher is inevitably, by the natural course,
reducing their overall scientific productivity (see also 4. Discussion
Schweitzer & Brendel, 2021). In our opinion, this is another First of all, it should be remarked that the coefficient
reason, we need to explore the relationship between aging associated with the first lag of the dependent variable is
and innovativeness. positive, smaller than one (ensuring the dynamic stability
Table 2 shows the results of the estimation of Equation I. of the model) and highly statistically significant. This result
In particular, Columns 1, 2 and 3 report, respectively, the supports the notion that innovation is a path-dependent
estimates derived from the one-step GMM estimator (with process, whereby a country’s current efforts in innovation
robust standard errors that take into account arbitrary positively affect its future efforts.

Volume 9 Issue 2 (2023) 67 https://doi.org/10.36922/ijps.0429