The German government provides subsidies to foster investments in broadband infrastructure – a key element in reaping economic and social benefits of digitalisation. From the Public
Management perspective on service provision models, this paper finds that the subsidies rely on flawed assumptions about the motivation and behaviour of subsidy-receiving telecom companies
(free-riding fear, low demand, red tape) and suffers from ill-designed targets in its execution (time gaming, cream skimming, administrative rivalry). The paper encourages subsidy providers to
increase per-project funds, lower eligibility criteria, and increase demand-side voice. The potential losses in efficiency are justified by increased broadband coverage rates and higher speeds –
being socio-economic imperatives at this juncture of time.
Since the advent of the Fourth Industrial Revolution, the internet has become imperative for an advanced economy to progress. Studies for Germany show that high-quality broadband connectivity spurs growth and productivity in almost all firms (Bertschek, Cerquera, & Klein, 2013), leads to increased labour demand (Arntz, Gregory, & Zierahn, 2016; Stockinger, 2017); more effective labour allocation (Gürtzgen, Nolte, Pohlan, & Berg, 2018); and enhanced political participation (Czernich, 2012). Despite available subsidies, Germany has lagged in keeping up with the pace of network infrastructure needs (Girard, Mattes, & Michelsen, 2018). In the following, this paper analyses why the provision of subsidies failed to adequately improve broadband connectivity by explaining the provision model, highlighting flawed assumptions and ill-designed targets, and finally recommending amendments.
Subsidy provision framework
In public service delivery terms, the Federal Ministry of Transport and Digital Infrastructure (BMVI) acts as the subsidy provision agency and telecommunication companies receive funds as service recipients. The policy objectives, i.e. more and better broadband, are determined by the Federal Government; executives of the ministry design specific targets via regulation; and local administrations manage the admission process. After having met eligibility criteria, the subsidy model allows max. 15 million Euros per project for initial cost-benefit analysis, publicly run network construction with private operation (“Betreibermodell”), and for full privately-run construction and operation of new broadband networks (“Wirtschaftlichkeitslückenförderung”) (BMVI, 2018).
Research for Germany shows that subsidies can increase broadband coverage by 19 % to 27 % – if provided and deployed successfully (Duso & Nardotto, forthcoming). However, the provisions model has proven unsuccessful as latest data shows: Out of the 3.5 billion Euros available for broadband expansion since 2015, only 26.6 million Euros have been retrieved and only approx. 3 million Euros have been utilized (Deutscher Bundestag, 2018). As elaborated hereafter, this is due to two major dysfunction of the subsidy regime: (1) flawed assumptions on telecommunication companies’ incentive structure, and (2) wrong target setting in the implementation phase.
Subsidies for fibre-optics: A failed service provision
Flawed assumptions: Subsidies ill-designed for recipients’ behaviour
Deutsche Telekom is Germany’s biggest network operator, partially owned by the German government (32% shareholding) and holds approx. 80% of all existing fibre-optic networks (Blank, 2017; Deutsche Telekom, 2018). To counter monopolisation of the network market, “Open Access” regulation enables other companies to rent-out space in the network grid for a state-determined price. Though beneficial for competition, the regulated access undermines rents from infrastructure investments in a classical “first mover” disadvantage scenario as Deutsche Telekom fears free-riding competitors that claim legal access through the Open Access regulation. Assuming the potential recipients’ behaviour, the subsidy does not consider the “rationally inactive” motivation of Deutsche Telekom and other operators as limited subsidy amounts cannot cover the potential loss of investment rents (Heymann & Körner, 2018).
Apart from the initial cost-benefit analysis, the subsidy aims at lowering the initial investment costs to lay fibre-optic cables – a very significant factor in the expansion of broadband connectivity. Moreover, as the subsidy does not aim at supporting the business models of operating a network, sufficient demand for broadband connection is essential. Especially in rural/not-yet-digitalised areas, this leads to a predicament: Lacking connectivity reduces demand for fast internet access, and a lack of demand lowers pay-off for broadband expansions. Without proper “voice of demand”, telecommunication companies cannot assess the later pay-off of their – although subsidised – investment (Girard et al., 2018).
Lastly, as German federalism requires the local administration to execute subsidy procedures, their overcharged and inefficient administrations create long waiting times and red tape (Klehm, 2018). Firms’ ignorance of applicability and high information costs reduces the number of applications for subsidies.
Misguided targets: Gaming, Cream Skimming, and Rivalry
Intended to increase flexibility, subsidy recipients can delay the start of the expansion project by up to three years. This is highly problematic as it allows for gaming among the recipients. For example, Deutsche Telekom has been found to apply for subsidies in order to pre-empt its competitors, then they either cancel the project and repay the subsidies or start the project with a maximum delay (Bündnis 90/Die Grünen, 2018).
Complex calculations and limited means to reconstruct cost-benefit analysis of applicant companies lead to the provision of subsidies for areas that would have been profitable without the subsidies (Heymann & Körner, 2018). As the problem lies in the misplaced provision, not general scarcity, of subsidies, the “cream skimming” behaviour of the companies seems to be not as detrimental as in competitive circumstances. However, as reallocation between cities, regions, or states is difficult, the misuse can become problematic when local administrations reach their provision limit (Deutscher Bundestag, 2018).
For the broadband connection of the future, existing copper cables need to be substituted through fibre-optic cables. However, the current broadband policy objective (transfer rate of 50 megabits per second) can still be reached by improving existing copper connections through so-called “Vectoring”. This adds to the goal in the short-run and avoids high transaction costs between cable types. As the Federal Network Agency (BNetzA) encourages firms to vector existing connections (it still improves overall internet connectivity), this rivals the subsidy regime aimed at substituting copper cables. Even though both initiatives further the overall objective, vectoring misses the point of incentivising long-term investments (DIHK, 2016).
Conclusion: Less efficiency, better coverage
Despite opposition parties calling for radical solutions like complete privatisation or nationalisation, the paper proposes four path-dependent amendments to improve subsidy provision and boost broadband expansion immediately: First, rural households and companies require more voice to signal their demand for broadband services – maybe by even paying upfront money (Girard et al., 2018). Secondly, regulative loopholes, such as the postponement of construction date by up to three years or the competing support for copper vectoring, must be closed (Bündnis 90/Die Grünen, 2018). Thirdly, red tape and administrative requirements on the local level must be lowered to increase general information levels and hasten processes (Heymann & Körner, 2018). Fourth, to address more companies, per-project funding must be increased and the eligibility criteria for subsidies reduced.
In combination, higher subsidies, lower admission hurdles and less space for gaming will cushion “free riding” detriments, bridge the demand predicament, and avoid gaming. Though funding projects that are a little profitable, the Federal Government should approve this loss of efficiency to increase the overall number of subsidies provided. At the current level of broadband coverage in Germany and with the political pressure to substantively digitalise the economy, one Euro employed at a 50 % efficiency rate is better than no Euro employed at an 80 % rate.
This policy briefing is result of a group project for the course "Public Management" at Hertie School of Governance, lectured by Salvador Parrado, Director of Governance International.
- Arntz, M., Gregory, T., & Zierahn, U. (2016). The risk of automation for jobs in OECD countries, (189). https://doi.org/10.1787/5jlz9h56dvq7-en
- Bertschek, I., Cerquera, D., & Klein, G. J. (2013). More bits - more bucks? Measuring the impact of broadband internet on firm performance. Information Economics and Policy, 25(3), 190–203. https://doi.org/10.1016/j.infoecopol.2012.11.002
- BMVI. (2018). Richtlinie - Förderung zur Unterstützung des Breitbandausbaus in der Bundesrepublik Deutschland. Berlin. Retrieved from https://www.bmvi.de/SharedDocs/DE/Anlage/Digitales/foerderrichtlinie-breitbandausbau.pdf
- Bündnis 90/Die Grünen. (2018). Breitbandausbau ist ein Trauerspiel. Retrieved November 26, 2018, from https://www.gruene-bundestag.de/medien/breitbandausbau-ist-ein-trauerspiel-12-06-2018.html
- Czernich, N. (2012). Broadband internet and political participation: Evidence for Germany. Kyklos, 65(1), 31–52. https://doi.org/10.1111/j.1467-6435.2011.00526.x
- Deutscher Bundestag. (2018). Abfluss der Mittel aus dem Bundesförderprogramm Breitbandausbau: Antwort der Bundesregierung auf Kleine Anfrage der Abgeordneten Margit Stumpp, Oliver Krischer, Matthias Gastel, weiterer Abgeordneter und der Fraktion BÜNDNIS 90/DIE GRÜNEN – Drucksache 19/. Berlin.
- DIHK. (2016). DIHK-Stellungnahme zum Entwurf des Nationalen Reformprogramms 2016 der Bundesregierung. Retrieved from https://www.dihk.de/themenfelder/aus-und-weiterbildung/ausbildung/ausbildungspolitik/umfragen-und-prognosen/dihk-ausbildungsumfrage
- Duso, T., & Nardotto, M. (n.d.). A retrospective Study of State Aid Control in the German Broadband Market.
- Girard, Y., Mattes, A., & Michelsen, C. (2018). Gigabit access: Germany lags behind in international comparison but demand is low. DIW Weekly Report, (25), 220–229.
- Gürtzgen, N., Nolte, A., Pohlan, L., & Berg, G. J. van den. (2018). Does the Internet Help Unemployed Job Seekers Find a Job? Evidence from the Broadband Internet Expansion in Germany. 71st European Summer Meeting of the Econometric Society (ESEM), Cologne, 27-31 August, 2018. Retrieved from http://conference.iza.org/conference_files/MacroEcon_2018/8579.pdf
- Heymann, E., & Körner, K. (2018). Digital Infrastructure: Bottlenecks hamper Europe’s progress. Frankfurt a. Main.
- Klehm, G. (2018). Bürokratie bremst Breitbandausbau. Retrieved November 30, 2018, from https://www.sächsische.de/buerokratie-bremst-breitbandausbau-3902066.html
- Stockinger, B. (2017). The effect of broadband internet on establishments’ employment growth: evidence from Germany. Retrieved from http://doku.iab.de/discussionpapers/2017/dp1917.pdf