Redistributive innovation policy, inequality, and efficiency

• Published date: 01 June 2020
• Author: Parantap Basu,Yoseph Getachew
• DOI: http://doi.org/10.1111/jpet.12386
• Date: 01 June 2020

J Public Econ Theory. 2020;22:532–554.wileyonlinelibrary.com/journal/jpet532

|

© 2019 Wiley Periodicals, Inc.

Received: 8 January 2019

|

Accepted: 6 June 2019

DOI: 10.1111/jpet.12386

ORIGINAL ARTICLE

Redistributive innovation policy, inequality,

and efficiency

Parantap Basu

1

|

Yoseph Getachew

2

1

Department of Economics and Finance,

Durham University Business School

Durham University, Durham, UK

2

Department of Economics, University

of Pretoria, Pretoria, South Africa

Correspondence

Yoseph Getachew, Department of

Economics, University of Pretoria,

Pretoria 0028, South Africa.

Email: yoseph.getachew@up.ac.za

Abstract

We examine the efficiency and distributional effects of

regressive and progressive public R&D policies that

target high‐tech and low‐tech sectors using a hetero-

genous‐agent growth model with in‐house R&D and

incomplete capital markets. We find that such policies

have important implications for efficiency and inequal-

ity. A regressive public R&D investment financed by

income tax could boost growth and welfare via a

positive effect on individual savings and effort. It could,

however, also lower growth and welfare via its effect on

the efficiency–inequality trade‐off. Thus, the relation-

ship between public R&D spending and welfare is

hump‐shaped, admitting an optimal degree of regressiv-

ity in public R&D spending. Using our baseline model,

and the US state‐level GDP data, we derive the degree of

regressiveness of public R&D investment in US states.

We find that US states are more regressive in their R&D

investment than the optimal regressiveness implied by

our growth model.

KEYWORDS

growth, inequality dynamics, public R&D investment, welfare

JEL CLASSIFICATION

D31; E13; H4; O41

1

|

INTRODUCTION

The role of public and private research and development (R&D) investment

1

in economic

growth is a widely debated topic.

2

However, the distributional effect of public R&D investment

has received little attention. In the extant literature, the focus is more on public education (e.g.,

Benabou, 2002; Glomm & Ravikumar, 1992, 2003), globalization (e.g., Bontems & Gozlan, 2018)

and infrastructure and taxes (e.g., Alesina & Rodrik, 1994; Chatterjee & Turnovsky, 2012;

Garcia‐Penalosa & Turnovsky, 2007; Getachew, 2010; Getachew & Turnovsky, 2015). R&D

investment could have uneven impacts on the economy and, through this channel, could

impact rich and poor differently. In general, most public R&D investment in developed

countries is concentrated on high‐tech industries such as information technology, biotechnol-

ogy, communication, and environment industries. In the United States, for instance, public

R&D investment in equipment and software has increased from 20% in 1980 to 50% in 2001,

which contributed to rising inequality in the United States in recent decades (Cozzi &

Impullitti, 2010). R&D investment in South Korea is concentrated more on high‐tech sectors

(Kim, Chun, & Kim, 2013). In contrast, in most of the developing world, a significant amount of

public R&D investments are made in agriculture, a low‐tech sector dominated by small‐scale

farmers. Beintema, Stads, Fuglie, and Heisey (2012) report an accelerated public investment in

agricultural R&D in developing countries during the period 2000 and 2008. Using provincial

data in China spanning more than four decades, Zhang and Fan (2004) argue that government

spending on agricultural R&D contributed to a reduction in regional inequality.

Our own calculations suggest a contrasting relationship between inequality and public R&D

spending, consistent with previous studies. Figures 1 and 2 show the relationships between the

GINI index and R&D intensity in the United States and public R&D spending in sub‐Saharan

Africa (SSA), respectively.

3

In both regions, R&D spending has sharply increased during the last

two decades, but the inequality experience is the opposite.

4

While there is a positive correlation

between GINI and R&D investment across US states, for SSA countries; however, the

correlation is negative. Given that R&D spending in the United States is more geared to

the high‐tech sector while in SSA it is primarily focused on subsistence farming, this reversal of

the sign of the correlation, between R&D intensity and the GINI, between Figures 1 and 2 is

intriguing. Inequality could result from public R&D investment due to its destination.

The key research question in this paper is therefore whether the public R&D spending is

likely to be regressive or progressive. The effects of public R&D investment on efficiency and

inequality are therefore examined using the lens of a heterogenous‐agent growth model, where

agents are heterogenous in their initial endowments of knowledge and their ability to generate

knowledge. The model includes in‐house R&D which yields monopolistic profit for the firms.

Both inequality and growth are endogenously determined. The source of endogenous growth is

in‐house R&D investment using private and public resources. Endogenous inequality is

1

R&D expenditure includes a broad range of activities: “Research and development (R&D) comprise creative work undertaken on a systematic basis in order to

increase the stock of knowledge (including knowledge of man, culture and society) and the use of this knowledge to devise new applications.”(http://

www.oecd.org/sdd/08_Science_and_technology.pdf).

2

Particularly, in early 1990s, there was an influx of R&D‐based growth theories, following the seminal works by Romer (1990), Grossman and Helpman (1991),

and Aghion and Howitt (1992) that emphasize the role of R&D in economic growth, through influencing technological progress. R&D policies are also widely

debated as to whether public R&D investment complements private R&D investment or crowds it out (e.g., David, Hall, & Toole, 2000). Early work in public

R&D investment includes Shell (1967).

3

The R&D intensity is measured by the ratio of total R&D spending (including private and public GDP) to state GDP (gross domestic product). The breakdown

of public and private R&D for each state is not available. The National Science Foundations sources suggest that the Federal share of total R&D spendingis

about 11.6% over the period 2008–2015. All the GINI data came from US Census Bureau. The R&D intensity data came from the Science, Technology,

Innovation and Entrepreneurship (SSTI) database of the United States. The correlation coefficient between R&D Intensity and State GINI index is 0.18. For SSA

countries, the GINI and public R&D spending data came from World Bank (2015). Due to sparse nature of the data, we take the average of GINI index from

2000 onward whatever data are available. Same is done for the public R&D spending ratio. Details of all these data are available from the authors upon request.

4

The average annual agricultural R&D spending growth in SSA countries, for instance, increased from 0.3% during 1981–1990 to 2.8% during 2000–2008 except

for a small dip of 0.01% during 1990–2000, which is indicative of the bulk of recent R&D innovations in SSA being progressive in nature (Beintema et al., 2012).

BASU AND GETACHEW

|

533