Estimation of the spillover effect between patents and innovation using the GARCH–BEKK model

• Published date: 01 December 2017
• Date: 01 December 2017
• DOI: http://doi.org/10.1111/1468-0106.12244
• Author: Muhammad Z. Mumtaz,Zachary A. Smith

ORIGINAL MANUSCRIPT

Estimation of the spillover effect between patents

and innovation using the GARCH–BEKK model

Muhammad Z. Mumtaz

1

| Zachary A. Smith

2

1

National University of Sciences & Technology,

Islamabad, Pakistan

2

Saint Leo University, Saint Leo, FL, USA

Correspondence

Muhammad Z Mumtaz, School of Social

Sciences & Humanities, National University of

Sciences & Technology, Sector H-12, Islamabad,

Pakistan.

Email: zubair@s3h.nust.edu.pk

Abstract

This study examines the mean and variance spillovers

between patents filed and total factor productivity in

Pakistan during the period lasting from 1964 to 2015.

Using the multivariate generalized autoregressive condi-

tionally heteroskedastic–Baba, Engle, Kraft and Kroner

(GARCH–BEKK) model, we find no evidence of mean

spillover effects between patents filed and productivity

growth. However, the results support the existence of the

variance spillover effect, which illustrates that the innova-

tions may affect productivity in the long-run. Moreover,

bidirectional causality in the mean and variance between

innovation and productivity growth is non-existent. We

find that R&D/GDP, skilled labour force and economic

freedom are the significant factors affecting technological

advancements in the country. We also determine that

R&D/GDP and economic freedom are the important fac-

tors affecting patent activities in Pakistan. Our findings

suggest that the government needs to emphasize the regu-

lations of the patent system to safeguard the interests of

investors, which ultimately promotes economic growth.

1|INTRODUCTION

Patents lead to innovations by devising new processes, methods and products, which have a strong

influence on promoting economic development. The study of the relationship between patents and

innovation is considered an important area of research in the field of economics. According to neo-

classical growth theorists (e.g. Solow, 1956; Swan, 1956) the rate of technological progress is inde-

pendent of economic growth but the new endogenous growth theory (Barro & Sala-i-Martin, 2003;

Romer, 1990) argues that the technological progress is ascertained through endogenous variables

occurring through innovations, which create economic activity. With regard to endogenous growth

theory, investment in innovation, human capital and knowledge are important components of

Received: 24 May 2017 Revised: 7 September 2017 Accepted: 11 September 2017

DOI: 10.1111/1468-0106.12244

772 © 2017 John Wiley & Sons Australia, Ltd wileyonlinelibrary.com/journal/paer Pac Econ Rev. 2017;22:772–791.

economic growth production. Hence, Romer (1990) suggests that the market power is created by the

protected process of intellectual property and the rights associated with the protection of this intel-

lectual property (e.g. patents generate a strong impact on the economic growth).

The World Intellectual Property Organization (WIPO, 2005) states that ‘technological innovation

is a key determinant of total factor productivity (TFP) growth which illustrates unlimited effects

exist on factor accumulation’(WIPO, 2005, p. 27). This is because TFP growth captures the residual

growth experienced after accounting for the influence of production factors. However, technological

innovation has a complex relationship with the behaviour of firms and the structure of economies.

Article 27 of the Agreement on Trade-Related aspects of Intellectual Property Rights stipulates that

patents are any inventions related to products or processes including all fields of technology.

We consider the patents filed as an index to examine the productivity of the country. Guo

(2015) find a strong positive relationship between patents and innovation in Japan. We treat TFP as

a productivity index, which formulates the combinations of inputs and outputs in all the production

factors, contrary to productivity relating to capital or labour. Generally, an increase in TFP leads to

an increase in productivity growth; therefore, TFP may be used as a variable for gauging

innovation.

The main hypothesis of this paper is that there is a relationship between patents and TFP growth

in an emerging market economy. We evaluate this hypothesis by applying a multivariate generalized

autoregressive conditionally heteroskedastic–Baba, Engle, Kraft and Kroner (GARCH–BEKK)

model to examine the relationship between patents and TFP in Pakistan. The second set of hypothe-

sis tests is focused on evaluating whether the determinants of TFP would change based on whether

the patents filed were foreign or domestic patents. Finally, the third set of hypotheses that we evalu-

ate is focused on determining if there are differences in the relationship between patents and TFP in

developed and emerging market economies.

This is the first attempt to determine whether patent applications and TFP growth (hereafter

referred to as ΔlnTFP) has any relationship in the case of Pakistan. The purpose of the present

study is to: (i) test the hypothesis of whether the relationship between patents filed and ΔlnTFP

exist; (ii) test the mean and variance spillovers between patent applications and ΔlnTFP,

(iii) explore the existence of bidirectional causality between the two variables; (iv) in relation to

the second hypothesis, compare the relationship between patent applications and TFP in emerging

and developed economies; and (v) in relation to the second and third hypotheses, identify the

determinants of TFP and patents and examine whether the determinants change when we focus

solely on domestic or foreign patents filed in an emerging market economy. Considering the

period from 1964 and 2015, this study finds no mean spillover effects between patents filed and

ΔlnTFP, but explores evidence of the presence of a variance spillover effect using multivariate

GARCH–BEKK methodology. This indicates that innovations may influence productivity in the

long-term. To promote and encourage patent activities in a country, the government should

strengthen the patent system to protect the interests of investors, which consequently fosters eco-

nomic growth in the country.

This present paper is structured as follows. The Introduction provides a brief description of the

study. Section 2 presents an evaluation of the innovation concept and examines the association

between patents and innovation, including the relevant literature associated with theoretical and

empirical evidence, and discusses the patent process in Pakistan. Section 3 discusses patenting in

Pakistan. Section 4 describes the empirical methodology, focusing specifically on the GARCH–

BEKK model. Section 5 describes the empirical findings and provides a discussion, including a

comparison of the results of current and prior studies, and Section 6 concludes.

MUMTAZ AND SMITH 773