Can Direct Regulations Spur Innovations in Environmental Technologies? A Study on Firm‐Level Patenting

• DOI: http://doi.org/10.1111/sjoe.12201
• Published date: 01 April 2018
• Date: 01 April 2018

©The editors of The Scandinavian Journal of Economics 2016.

Scand. J. of Economics 120(2), 338–371, 2018

DOI: 10.1111/sjoe.12201

Can Direct Regulations Spur Innovations

in Environmental Technologies?A Study on

Firm-Level Patenting*

Marit E. Klemetsen

Statistics Norway, NO-0033 Oslo, Norway

mkl@ssb.no

Brita Bye

Statistics Norway, NO-0033 Oslo, Norway

bby@ssb.no

Arvid Raknerud

Statistics Norway, NO-0033 Oslo, Norway

rak@ssb.no

Abstract

Using a rich Norwegian panel dataset that includes information about the type and

number of patent applications, direct environmental regulations, and a large number

of control variables, we analyze the effects of direct regulations on environmental

patenting. We use inspection violation status as a measure of regulatory stringency,

while controlling for risk class. Violation status captures the probability that a firm

might be sanctioned for violating its emission permit. Controlling for risk class captures

firm heterogeneity related to dirtiness and inspection frequency. We empirically identify

strong and significant effects on innovations resulting from the implicit regulatory costs

of direct regulations.

Keywords: Command-and-control regulations; environmental technologies; innovation;

patents; Poisson count model

JEL classification:C23; O34; Q52; Q53; Q55; Q58

I. Introduction

In this paper, we investigate the relation between direct (command-

and-control) environmental regulations and innovations in environmental

*This paper has benefited from numerous comments and suggestions from two anonymous

referees, and during presentations at the University of Oslo, Statistics Norway, EAERE

2013, and 5th ZEW/MaCCI. We thank the Norwegian Environment Agency for providing

data and information and acknowledge financial support from the Research Council of

Norway’s RENERGI-programme and the EC7 Framework Programme, ENTRACTE (project

No. 308481). Klemetsen and Bye have been associated with CREE – Oslo Centre for Research

on Environmentally friendly Energy; the Centre acknowledges financial support from the

Research Council of Norway, the University of Oslo, and user partners.

M. E. Klemetsen, B. Bye, and A. Raknerud 339

technologies. There are several real-life examples showing that direct

environmental regulations can spur innovations. In 1998, an international

agreement – the Oslo and Paris Convention (OSPAR) – legally required

the EU and other European countries to reduce emissions of polycyclic

aromatic hydrocarbons. As a consequence, the Norwegian Environment

Agency (NEA) banned the use of a specific polluting technology, called the

Søderberg technology, in Norwegian aluminum plants with effect from a

certain date.1Firms responded differently to the coming prohibition. Some

plants (e.g., Årdal, Sunndal and Karmøy, owned by Norsk Hydro ASA)

purchased the alternative pre-bake technology with substantial emission

reductions per production unit, while others (e.g., Alcoa-Lista) started to

develop new technologies based on the old technology framework, which

also led to considerable emission reductions per production unit. This

technology was later patented and commercialized on the international

market.2

To investigate the relationship between direct regulations and innovation,

we use a rich firm-level panel dataset containing information about

environmental regulations, patent applications, granted patents, and several

other key economic variables for the total population of Norwegian

incorporated firms. The Norwegian patent data contain organizational

numbers allowing us to match patents to datasets on regulations and

control variables from several other sources.3We use the number of

patent applications as a measure of innovative efforts, where environmental

technologies are identified by International Patent Classification (IPC)

codes.

With our data, we contribute to the existing body of literature in three

ways. First, we measure environmental regulations at the firm level, which

is appropriate because regulations generally vary greatly across firms.

1In 2000, the NEA introduced a prohibition on the Søderberg technology with effect from

2007. Some plants were granted extensions because of the consequences for employment in the

local community and the need for time to adapt and be able to offer alternative employment.

All production based on the old Søderberg technology was eradicated by 2009.

2Patent numbers: NO20055096; US2004195091; US6805777; ZA200507999;

WO2004094697; RU2005133706; RU2299276; EP1618231; CN1768164; CA2519170;

BRPI0408980; AU2004233150. Other examples are the EU standards for NOxemissions

(e.g., the Euro 5 standard), which have been made increasingly stringent in order to reduce

health problems in bigger cities. In response to the increased demand for technology with

lower emission intensity, new catalysts and particle filters for diesel cars were developed,

patented, and commercialized.

3In most countries, there is no unique identifier allowing researchers to link intellectual

property information directly to other firm-level data (Helmers et al., 2011). Researchers

using data from other countries use names indicated on patent documents, including assignee

and inventor names, and the firm names contained in firm-level databases to merge datasets.

Matching firm names across datasets is challenging and prone to errors (Helmers et al., 2011).

©The editors of The Scandinavian Journal of Economics 2016.

340 A study on firm-level patenting

Second, we use inspection violation status – the regulator’s assessment

of the severity of any violation (an ordinal variable) – to measure the

firms’ regulatory costs relating to technology standards and non-tradable

emission quotas. We argue that this variable captures the risk that a firm

might be sanctioned for violating its emission permit, and that this is a

more appropriate measure of regulatory costs than used in the previous

body of literature. For example, Brunnermeir and Cohen (2003) use a

measure of monitoring or inspection frequency. Third, our dataset allows

us to control for observed firm heterogeneity through a wealth of control

variables, including risk class, which captures firm heterogeneity with

regard to dirtiness and inspection frequency (see Cohen, 2010, for an

overview of the body of literature on the determinants of innovation).

To estimate the parameters of the regulatory costs and other control

variables on the number of patent applications, we follow the most common

approach in the body of literature and base our statistical inference on a

Poisson count model. We identify a strong positive effect of the implicit

costs of non-tradable emission quotas and technology restrictions measured

by inspection violation status, on the number of environmental patent

applications. This main conclusion is strengthened further when we restrict

the analysis to granted patents. Our findings indicate that implicit costs

associated with the threat of being sanctioned for violating emission

permits provide significant incentives for innovation. Our results contradict

conventional economic theory, supported by several empirical studies,

which suggests that direct regulations such as technology standards or

non-tradable emission quotas provide little or no incentive to innovate

(Zerbe, 1970; Wenders, 1975; Downing and White, 1986; Milliman and

Prince, 1989; Jung et al., 1996).

The rest of the paper is organized as follows. In Section II, we present

the relevant background and body of literature. Section III contains a

description of the data and variables used in the empirical analysis.

The econometric model and results are presented in Section IV. Finally,

in Section V, we present our conclusions and suggest some policy

implications.

II. Background

During the last two to three decades, there has been a wide range

of governmental policies aimed at reducing polluting emissions and

stimulating innovation in environmental technologies.4To date, direct

4See, for example, The European EnvironmentalAgency (2005), Bruvoll and Bye (2009), and

Ministry of Finance (2007, 2015) for descriptions of environmental policies in use. Downing

and White (1986) and Porter (1991) are early contributions to the body of literature on the

effects of environmental policies on technological change, innovation, and productivity.

©The editors of The Scandinavian Journal of Economics 2016.