Go for Gigabit? First Evidence on Economic Benefits of High‐speed Broadband Technologies in Europe

• Published date: 01 September 2019
• Author: Wolfgang Briglauer,Klaus Gugler
• Date: 01 September 2019
• DOI: http://doi.org/10.1111/jcms.12872

Go for Gigabit? First Evidence on Economic Benef‌its of High-

speed Broadband Technologies in Europe*

WOLFGANG BRIGLAUER

1

and KLAUS GUGLER

2

1

ZEW Centre for European Economic Research, Mannheim

2

Department of Economics and Research Institute for Regulatory Economics,

Vienna University of Economics and Business, Vienna

Abstract

The literature on the effects of investment in broadband infrastructure f‌inds positive macroeco-

nomic effects. However, it is severely constrained because it could hitherto only analyse invest-

ment or adoption up to basic broadband, but not up to the newer generations of hybrid f‌ibre and

end-to-end f‌ibre-based broadband. Utilizing a comprehensive panel dataset of EU27 member states

for the period from 2003 to 2015, we estimate a small but signif‌icant effect of end-to-end f‌ibre-

based broadband adoption over and above the effects of basic broadband on GDP; specif‌ically,

we f‌ind that a 1 per cent increase in adoption leads to an incremental increase of 0.002–0.005

per cent in GDP. The incremental effect of hybrid f‌ibre broadband adoption over basic broadband

is slightly lower (0.002–0.003%). Our cost-benef‌it analysis implies that policy intervention –as

foreseen by the European Commission –is only justif‌ied for coverage and adoption levels of

around 50 per cent, whereas for a 100 per cent coverage level net losses are likely.

Keywords: Cost-benef‌it analysis; EU27 panel data; High-speed broadband adoption; EU broadband;

policy targets

Introduction

In the late 1990s broadband markets experienced their f‌irst disruptive innovation when in-

ternet technology transitioned from narrowband to basic broadband via digital subscriber

line (DSL) and cable modems based on copper wire and coaxial cable access infrastruc-

tures, respectively. On the grounds of their alleged huge positive externalities, economists

analysed the effects of these infrastructure investments on macroeconomic growth, and

found a broadly positive effect. Currently, broadband markets are subject to major disrup-

tive innovations in the forms of the transition to modern broadband internet, which is

based in part or entirely on a f‌ibre-optic (wireline) infrastructure in the f‌inal segments

of access networks. Deployment costs, however, increase disproportionately when

switching from hybrid to end-to-end f‌ibre technologies. Whereas in the former case f‌ibre

is deployed only at distribution points as far as several hundred metres away from the cus-

tomer’s premises, the latter involves deploying f‌ibre access networks directly into the

building or apartment of a customer. Given substantial differences in deployment costs,

it is important to quantify the economic benef‌its and costs when deciding on the optimal

evolution of broadband deployment and associated policies.

* The authors are very grateful for the insightful and valuable comments of three anonymous referees and to Frank

Verboven, Thomas Niebel, Günter Knieps, Volker Stocker and conference participants in Trento at the 29

th

ITS

European Regional Conference for many useful comments and suggestions. Financial support by the State Government

of Baden-Württemberg, Germany, through the research programme ‘Strengthening Eff‌iciency and Competitiveness in the

European Knowledge Economies’(SEEK) is gratefully acknowledged.

JCMS 2019 Volume 57. Number 5. pp. 1071–1090DOI: 10.1111/jcms.12872

© 2019 University Association for Contemporary European Studies and John Wiley & Sons Ltd

The deployment of hybrid and end-to-end f‌ibre-based broadband technologies is cur-

rently a hot policy issue. In its current major framework review, the European Commis-

sion (EC) published its proposal for a directive establishing the European Electronic

Communications Code (EC, 2016a), with one key aim being the provision of suff‌icient

incentives for investment in the infrastructure of f‌ibre-based communications networks.

Following the Digital Agenda Europe (DAE) objectives for 2020 (EC, 2010), the EC

further specif‌ied ambitious and specif‌ic long-term objectives for 2025 with a strong em-

phasis on the promotion of very high-capacity networks that enable gigabit connectivity

(EC, 2016b).

1

Although these objectives appear to be neutral regarding the preferred type

of technology, the EC implicitly favours end-to-end f‌ibre-based network technologies

(see EC, 2016a, recital 13 and Art. 2 (2)).

From a dynamic perspective the core question is to what extent welfare is lost through

slower rather than faster migration to end-to-end gigabit networks and whether dirigiste

market interventions, such as those favouring certain broadband technologies in public

policy targets instead of market-driven technology choices, are justif‌ied in view of consid-

erably higher deployment costs. The latter is referred to as the principle of technological

neutrality (EC, 2009, recital 18). The answer depends on the extent of incremental eco-

nomic benef‌its related to end-to-end f‌ibre-based broadband and on the further evolution

of hybrid broadband. As shown in Section I, the empirical literature on economic effects

related to specif‌icf‌ibre technologies is extremely scant, and the evolution of future tech-

nology standards and demand is subject to a high degree of uncertainty. This implies that

rigorous empirical evidence on the causal impact of modern broadband internet is much

needed. Accordingly, we provide evidence on the following research question: What

are the incremental economic benef‌its of different f‌ibre technologies vis-à-vis basic

broadband and do incremental benef‌its justify favouring high-cost investment scenarios?

We utilize a recent panel dataset on 27 EU member states from 2003 to 2015, covering

almost the entire f‌ibre-deployment period. The present analysis provides econometric ev-

idence based on EU data with unique data for all relevant wireline broadband technolo-

gies.

2

We employed a twofold research strategy. First, in order to identify causal effects

of individual broadband technologies on economic benef‌its, which we measured in terms

of gross domestic product (GDP), we explicitly accounted for potential endogeneity uti-

lizing panel econometric estimators with instrumental variables. Second, we related the

estimated benef‌its of broadband adoption to available cost estimates from external indus-

try studies in order to provide some rudimentary cost-benef‌it analysis. We found that the

incremental benef‌its of end-to-end f‌ibre-based broadband are statistically signif‌icant

and slightly larger than those of hybrid-f‌ibre broadband, however, we found the largest

estimated growth effects for basic broadband adoption. Our cost-benef‌it analysis reveals

that partial but not full end-to-end f‌ibre-based broadband coverage entails the largest

net benef‌its.

The article is organized as follows. Section I reviews the recent and most relevant

empirical literature. Section II then presents the empirical baseline specif‌ication and our

1

In its gigabit society objectives the EC foresees ‘[a]ll European households, rural or urban, as having access to internet con-

nectivity offering a downlink of at least 100 Mbps, upgradable to gigabit speed’(European Commission, 2016b, pp. 35–36).

2

Although mobile broadband will assume an increasingly important role in view of current 4G and upcoming 5G technol-

ogies (as acknowledged in the EC’s gigabit society objectives), the EC as well as most national regulators have not to date

def‌ined common f‌ixed and mobile broadband markets due to the lower average bandwidth and quality levels in the past.

Wolfgang Briglauer and Klaus Gugler1072

© 2019 University Association for Contemporary European Studies and John Wiley & Sons Ltd