Page 75 - 臺大管理論叢第33卷第1期

NTU Management Review Vol. 33 No. 1 Apr. 2023

modularity below the mean minus one standard deviation. Given the estimation of low-
cost talent on the likelihood of location choices (in terms of odds ratio), we then draw the
interaction graph to illustrate the high- project-modularity and low- project-modularity
regressions (Figure 3). Figure 3 shows that the slope of the high-project-modularity

regression is steeper than the slope of the low-project-modularity regression. While
the slope of the high-project-modularity regression is significantly positive (β = 1.95,
p-value < 0.01), the slope of the low-project-modularity regression is also significantly

positive (β = 1.57, p-value < 0.01). After the Hausman test, nonetheless, the result shows
2
that both are significantly different (χ = 12.27, p-value < 0.01). We also evaluate the
joint impact of low-cost talent and project modularity for the three country groups (very
strong IPR protection countries, strong IPR protection countries, and weak IPR protection
countries) in Figure 4. Specifically, adding project modularity to the availability of low-

cost talent could lead to a 1.60% drop in the odds of outsourcing an innovation activity to
the very strong IPR protection countries and a 1.8% increase in the odds of outsourcing
an innovation activity to the weak IPR protection countries. These are consistent with

Hypothesis 3.
Lastly, the mean of the Variance Inflation Factors (VIFs) for each model in Table 3
ranges from 4.032 to 4.918, while none of the single VIF is excessively greater than 10
(O’Brien, 2007). That is, the multicollinearity is relatively mild. Overall, these results
corroborate the expectations formulated in the previous hypotheses.

5. Discussions and Conclusions

5.1 Main Findings

Conducting the empirical test on the ORN, IEF and WEF data, we validate our
major premises and show several intriguing findings. As firms strategically make their
decisions on outsourcing innovation activities to the countries with weak IPR protection,
we observe systematic differences across different countries and their impacts on the

choice of an outsourcing location. Specifically, we adopt the KBV perspective to examine
human capital, in terms of low-cost and high-skilled talent, as one of location differences
in developing countries. We then explore the effect of human capital on the location
choice when firms decide to outsource innovation activities. As a result, we find that

67