Elsevier

Energy Policy

Volume 69, June 2014, Pages 555-565
Energy Policy

The potentials of a reverse auction in allocating subsidies for cost-effective roof-top photovoltaic system deployment

https://doi.org/10.1016/j.enpol.2014.01.029Get rights and content

Highlights

  • Return-on-investment of PV varies by roof suitability, system size and subsidy level.

  • A reverse auction for subsidies is a cost-effective mechanism for PV system deployment.

  • Simulating a reverse auction for a case study region using a detailed solar cadaster and historical subsidy data.

  • Results indicate electricity generation increases by up to 18% and reductions of public funding by up to 41%.

Abstract

Photovoltaic (PV) has developed to one of the most promising technologies for renewable electricity generation. The Austrian government currently provides subsidies for roof-top PV systems through a constant, administratively determined feed-in tariff or an investment co-funding. In both subsidy schemes, applications are approved on a first-come, first-served basis. There are concerns about (i) the selection of suitable roofs for PV systems, and (ii) allocating subsidies among applicants to deploy roof-top PV systems cost-effectively. Thus we analyze the potentials of a simple discriminative first-price reverse auction application scheme. Applicants define individually the required level of subsidy and those with the lowest request for subsidies are selected. In an ex-post analysis, we evaluate the potentials of such a scheme in increasing power output and saving public spending for the federal state of Vorarlberg in Austria. Results indicate a potential increase of cumulated produced electricity between 15% and 18% in comparison to the current policy. In addition, a reverse auction-based system would lead to savings of public spending per kWh between 20% and 41%.

Introduction

The European renewable energy directive 2009/28/EC requires member states to comply with the national targets (European Commission, 2009, Kettner et al., 2010). In Austria, solar photovoltaic (PV) is expected to achieve a cumulated installed capacity of 1200 MW by 2020 (Baumann and Lang, 2013). Over the last decade, PV has become an increasingly feasible and promising source of renewable energy due to technical progress and steadily decreasing installation costs. However, even though prices for PV modules are declining and market diffusion is increasing, the investment in PV systems without subsidization still remains unprofitable in Austria. This is evident from the fact that around 98% of all installed PV systems are subject to some kind of public co-funding in Austria (Biermayr et al., 2012).

A feed-in tariff (FIT) for PV electricity generation was first implemented in Austria in 2002. It has a similar design as the ‘traditional’ FIT in Germany: a constant, periodically updated, administratively defined tariff for a certain duration and system size. As concluded by Rickerson et al. (2007), the advantage of a FIT is that it enables a rapid and substantial growth in the renewable electricity markets. Furthermore, a FIT policy is also expected to have a positive impact on the creation of jobs and economic growth by promoting manufacturing industries. Consequently, more than 80 countries and jurisdictions around the world have adopted a FIT policy to promote PV (Wang and Cheng, 2012). Apart from the FIT, investment co-funding (ICF) has been implemented as a second support scheme for PV deployment in Austria. In this case, investors in PV systems receive initial financial support for the construction and installation of a PV system. A FIT subsidized PV system usually feeds all generated electricity into the grid, because the FIT tariff is higher than the end-user electricity price (see Table 3). Owners of PV receiving ICF, however, are incentivized to self-consume the generated electricity and sell the excess power to an electricity retailer as market prices for selling electricity are usually much lower than for buying it (KLIEN, 2012). Until 2012, ICF and FIT were available for investments in PV in a mutually exclusive manner.

Three major concerns have arisen in the context of the Austrian subsidy policy. Firstly, as discussed in Lesser and Su (2008), it is difficult for policy-makers to define FIT attributes administratively, such as the level of the FIT tariff and its duration. Policy-makers are required to anticipate future market development and technological progress. Misjudgment could result in a cost-ineffective deployment of PV systems, while disproportional high subsidies can lead to avoidable windfall profits for investors, low subsidies might deter potential investors from investing in PV systems (Del Río, 2012). Similar information problems are relevant in the case of ICF subsidies as well. Secondly, there are concerns about the economic efficiency of subsidy allocation, if there is no competition. In traditionally more market oriented countries such as the UK, Australia or the USA, it is often argued that more competitive subsidy schemes, such as the Renewable Portfolio Standard in the USA or the Renewable Obligations Scheme in the UK, achieve a more cost-effective subsidization of Renewable Energy Technologies (RET) in comparison to the German-style FIT (Dong, 2012). Thirdly, there are serious concerns about the granting procedure for PV subsidies, such as in Austria. There, subsidies are currently approved via a web-based ‘first-come, first-served’ application procedure. Even though both subsidy schemes (FIT and ICF) require certain regulations with respect to size, material or installation of PV systems, there are no regulations governing the decision which roof is eligible for subsidization or whether a roof is even suitable for PV electricity generation. Furthermore, PV subsidies are usually awarded within minutes after opening the application procedure, thus implying that there is a high demand for PV subsidies (KLIEN, 2012, OeMAG, 2012a).

Therefore we propose the use of a reverse auction to allocate subsidies for a cost-effective deployment of roof-top PV system. Reverse or procurement auctions are common for support of public procurement (De Silva et al., 2008, Nakabayashi, 2013) and environmental services (Greenhalgh et al., 2007, Jindal et al., 2013). In such an auctioning procedure, one buyer faces many sellers, who are bidding their services at the lowest possible price (Giebe et al., 2006). Laffont and Tirole (1993) argue that, in a reverse auction, competitive bids can be elicited, when several applicants are possible candidates to realize a project. Thurston et al. (2010) have analyzed a reverse auction for the implementation of distributed stormwater management practices in Shepherd Creek, USA. There, home owners were invited to participate in a reverse auction to receive public funding for the installation of either rain barrels or rain gardens. The auction procedure is assumed to achieve efficiency, objectivity, transparency and flexibility in the allocation of public funds, as those investors who are situated to make the best use of funding, are selected. Furthermore, prices are determined by the market and not by a governmental agency. The rules in bidding for and granting subsidies are known by all participants. Flexibility can be achieved, in that the mechanism can be altered by introducing, for instance, a ceiling of subsidies being allocated or by predetermining a lower and upper limit for the bids in the auction.

Reverse auction-based subsidy schemes for PV are also in line with the recent international trends to revise the traditional FIT policies. Lesser and Su (2008) recommend the introduction of an auction to distribute FIT subsidies in order to avoid problems of insufficient information for the determination of FIT attributes (such as the duration or structure of payments). An auction represents an efficient price discovery mechanism, which thus prevents prohibitively low or inefficiently high subsidy levels (Lusk and Shogren, 2008). Since subsidies for photovoltaic represents a monopson where the government faces a large number of potential PV electricity suppliers, an auction leads to price discrimination and thus limits windfall profits of producers (Bofinger, 2013). Auctioning of subsidies for renewable energies has not only been discussed in the literature (Becker and Fischer, 2013, Kreycik et al., 2011), but has already been implemented in some regions. Wang and Cheng (2012), for instance, present Taiwan׳s approach of introducing a bidding procedure for the determination of FIT levels for large-scale PV projects. In particular, the Taiwanese government has faced problems in determining an appropriate FIT support scheme due to rapidly declining PV system prices. In addition, auctions for PV subsidies have also been implemented in other countries such as China (Grau et al., 2012), India, Peru, South Africa and California (Bazilian et al., 2013).

However, we are not aware of any published research study attempting to quantitatively compare different subsidy policies to promote PV systems. In order to bridge this gap, we thus conducted this study in which we simulated a reverse auction for subsidies and quantified potential electricity generation increases and savings of public spending for the federal state of Vorarlberg in Austria.

This article is structured as follows: in Section 2 data and methods are described and a reverse auction procedure for subsidies is developed. Section 3 presents the results of the analysis and outlines the implications of the different scenarios. Furthermore, a sensitivity analysis is conducted in order to investigate the effects of parameter variations on outcomes. Finally, we discuss the methodology used and the results of our analysis in Section 4 and present conclusions in Section 5.

Section snippets

Data and methods

We propose a simple discriminative first-price reverse auction for PV subsidies, as illustrated in Fig. 1. In a call for bids, several issues are specified, such as the total capacity to be funded, the technical criteria or the payment structure. In addition, quotas to support particular types of installation (e.g. small-scale installations on dwellings) and a lower and upper limit for the bids can be considered similar to the South African auctioning procedure (Becker and Fischer, 2013).

The ROI in the REF scenario

Fig. 3 shows the mean ROI of four systems sizes that received the FIT funding as well as the mean ROI of ICF funded PV systems in the REF scenario. It can be seen that not all system sizes are installed in each year. Since the investment analysis for a large number of buildings involves several assumptions, the exact magnitude of the ROI might not be sufficiently meaningful for interpretation. Moreover, due to relatively conservative assumptions and the uncertainties associated with the data,

Discussion

The ex-post simulation of a reverse auction-based allocation of PV subsidies shows that electricity generation outputs can be increased while reducing public funding. A reduction in the subsidy per kWh between 20% and 41% can be achieved in comparison to the current first-come, first-served subsidy scheme. Our findings are in line with those of Lesser and Su (2008) and Wang and Cheng (2012), who endorse an auction-based allocation for PV subsidies in order to achieve a more cost-effective

Conclusions and policy implications

The findings show that buildings can differ substantially with respect to their roof-top PV power potential and return-on-investment. Thus a uniform, constant subsidy (whether FIT or ICF), determined by a governmental agency and applied to all PV systems, might be highly inefficient. In contrast, a simple discriminative first-price reverse auction has several advantages. Firstly, in comparison with administratively price setting, an auction ensures market prices and prevents inefficiencies due

Acknowledgments

This article has been supported by the Austrian Climate and Energy Fund, which also provided data for this project. The federal state government of Vorarlberg has additionally supported this article with detailed street map data and permitted the use of the Solar Potential Cadaster for the analysis. Appreciation is also expressed towards the research institute Laser Data GmbH, which has provided solar cadaster data and detailed information about the calculation of the solar cadaster. Further

References (54)

  • T. Giebe et al.

    How to allocate R&D (and other) subsidies: an experimentally tested policy recommendation

    Res. Policy

    (2006)
  • J.K. Goeree et al.

    Winner׳s curse without overbidding

    Eur. Econ. Rev.

    (2003)
  • T. Grau et al.

    Survey of photovoltaic industry and policy in Germany and China

    Energy Policy

    (2012)
  • R. Haas et al.

    Socio-economic aspects of the austrian 200 kwp-photovoltaic-rooftop programme

    Sol. Energy

    (1999)
  • R. Jindal et al.

    Social dimensions of procurement auctions for environmental service contracts: evaluating tradeoffs between cost-effectiveness and participation by the poor in rural Tanzania

    Land Use Policy

    (2013)
  • R. Kannan et al.

    Life cycle assessment study of solar PV systems: an example of a 2.7 kWp distributed solar PV system in Singapore

    Sol.r Energy

    (2006)
  • M. La Gennusa et al.

    A model for predicting the potential diffusion of solar energy systems in complex urban environments

    Energy Policy

    (2011)
  • J.A. Lesser et al.

    Design of an economically efficient feed-in tariff structure for renewable energy development

    Energy Policy

    (2008)
  • F. Muhammad-Sukki et al.

    Revised feed-in tariff for solar photovoltaic in the United Kingdom: a cloudy future ahead?

    Energy Policy

    (2013)
  • J. Nakabayashi

    Small business set-asides in procurement auctions: an empirical analysis

    J. Public Econ.

    (2013)
  • W.H. Rickerson et al.

    If the shoe FITs: using feed-in tariffs to meet U.S. renewable electricity targets

    Electric. J.

    (2007)
  • I.H. Rowlands

    Envisaging feed-in tariffs for solar photovoltaic electricity: European lessons for Canada

    Renew. Sustain. Energy Rev.

    (2005)
  • A. Strzalka et al.

    Large scale integration of photovoltaics in cities

    Appl. Energy

    (2012)
  • H.W. Thurston et al.

    Using a reverse auction to promote household level stormwater control

    Environ. Sci. Policy

    (2010)
  • K.M. Wang et al.

    The evolution of feed-in tariff policy in Taiwan

    Energy Strategy Rev.

    (2012)
  • L.K. Wiginton et al.

    Quantifying rooftop solar photovoltaic potential for regional renewable energy policy

    Comput., Environ. Urban Syst.

    (2010)
  • M. Baumann et al.

    Entwicklung energiewirtschaftlicher Inputdaten und Szenarien für das Klimaschutzgesetz und zur Erfüllung der Österreichischen Berichtspflichten des EU Monitoring Mechanismus 2013 (Klima und Energiefonds Endbericht)

    (2013)
  • Cited by (35)

    • Renewable auctions: Bidding for real options

      2021, European Journal of Operational Research
    • The different types of renewable energy finance: A Bibliometric analysis

      2021, Energy Economics
      Citation Excerpt :

      Non-market instruments include investments in research and development (R&D) (Park et al., 2013), grants and subsidies (Cansino et al., 2011) – which generally intervene upstream of the innovation cycle in the solar sector – as well as the regulations governing the activities of solar firms (Lazzeroni et al., 2017). Market instruments include FITs (AlYahya and Irfan, 2016; Bakhtyar et al., 2015; Queiroz et al., 2017), solar REC (Borenstein, 2017; Lee et al., 2017), Green Technology Financing Schemes (GTFS) (Muhammad-Sukki et al., 2014), and co-funding reverse auctions (Mayr et al., 2014). Cluster 6, with 31 articles, focuses on the assessment of RE investment opportunities by private investors in emerging and developed countries.

    View all citing articles on Scopus
    View full text