New Evidence on Price Effects of Transparency Regulations in European Fuel Markets

doi:10.21203/rs.3.rs-4856675/v1

Download PDF

Research Article

New Evidence on Price Effects of Transparency Regulations in European Fuel Markets

https://doi.org/10.21203/rs.3.rs-4856675/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

This paper analyses the impact of transparency regulations on Super 95 and Diesel prices in Austria, France, and Germany to shed new light on the price effects of regulatory measures. To identify the effects of regulation, we use publicly available data from January 2005 to the end of December 2019 and estimate synthetic difference-in-difference models. Overall, we find little evidence of any significant price effects due to transparency regulations. In France, prices did not change significantly. In Austria, transparency regulations had a small but significant positive effect merely on Super 95 prices in the short run. In Germany, the Market Transparency Unit (MTU) had negative but small long-term effects on Diesel prices. Therefore, we conclude that the implemented transparency regulations in all three countries did not result in long-lasting lower prices, contrary to the desired outcome of the policy treatments.

fuel prices

regulation

market transparency

synthetic difference-in-differences

oligopoly

competition

In recent decades, fuel prices in Europe have experienced numerous fluctuations due to events such as the oil price crisis, the global financial crisis, the COVID-19 pandemic, and the Russian war on Ukraine. Beyond these event-driven fluctuations, fuel prices have been steadily increasing (Le et al., 2021; Sokhanvar et al., 2023). The volatility and upward trend in fuel prices have significant implications for consumers, businesses, and policymakers. In response to these challenges, several European countries have implemented transparency regulations aimed at stabilising and reducing fuel prices by increasing market transparency. Countries like France, Austria, Germany, Belgium, and Luxembourg have tried to counteract rising fuel prices by introducing various measures to increase market transparency in fuel markets, thereby intensifying competition between gas stations. In Germany, a market transparency unit was established as part of the Federal Cartel Office on August 31, 2013 (Bundeskartellamt, 2024). This transparency unit provides real-time data on current gasoline (Super 95) and Diesel prices from gas stations across the country. However, the effects of such price disclosures are not easily determined, as transparency regulations increase both vertical and horizontal market transparency among gas stations. On the one hand, greater vertical market transparency is expected to lead to more intense competition, as consumers gain information about current prices at the pumps. With reduced search costs, consumers can quickly react to price reductions and choose the cheapest gas stations in their vicinity. On the other hand, horizontal price transparency ensures that gas stations have more information about competitors’ prices, allowing them to adjust their pricing decisions accordingly. Thus, greater transparency can lead not only to increased competition but also to coordinated effects (Kühn & Vives, 1995; Møllgaard & Overgaard, 2001; Stiglitz, 1989; Vives, 1990). Especially when automated pricing systems are used, gas stations can adjust prices more rapidly. Automated price setting can also facilitate coordinated behaviour between stations, particularly if the market structure is characterized by a tight oligopoly of large oil companies complemented by a few independent stations (Varian, 2019; OECD, 2017; Assad et al., 2020).

In this study, we empirically analyse the effects of market transparency treatments and regulations on fuel prices in Germany, Austria, and France. We use a synthetic difference-in-differences method, introduced by Arkhangelsky et al. (2021), and apply it to a publicly available dataset for European fuel prices. To the best of our knowledge, this study is the first to implement this approach to examine the effects of introducing market transparency regulations.

Our results vary across the different countries studied. In France, we observed no statistically significant price change for fuels. In contrast, the effect of the MTU in Germany is negative, resulting in a reduction of one to two cents per litre. For Super 95, the effects are statistically significant only in the short term, whereas for Diesel, they are significant only in the long term. In Austria, the Price Transparency Enactment has a small positive effect of less than 1 cent per litre, which is significant only in the short term for Super 95 prices.

In contrast, the results for the Price Labelling Act Revised are consistently negative. In the short term, there is a significant negative effect of 3 cents per litre on Super 95 prices and a significant negative effect of 2 cents per litre on Diesel prices. Since the results indicate that transparency regulations have little influence on gas station pricing and can even lead to price increases in some cases, the question arises as to whether the low economic benefits offset the economic costs associated with the regulations.

Previous research has yielded contrasting results regarding the impact of transparency regulations. Dewenter et al. (2017) found that the MTU did not result in a decrease in fuel prices. Using a difference-in-differences approach to analyse the effects of the MTU introduction in Germany, they discovered evidence suggesting an increase in gasoline prices by 1.2 to 3.4 cents per litre and Diesel prices by about 2 cents per litre. In contrast, Haucap et al. (2017) observed a price-cutting effect when examining German fuel prices. Comparing the average gasoline prices of gas stations in four German cities between 2012 and 2015, they noted that average prices in 2015 were lower than those in 2012.

In more recent studies, Horvath (2019) explored the relationship between the MTU and fuel prices in Germany, employing fixed effects regression and a difference-in-differences approach with a control group of 18 other European countries. The findings revealed a price decrease of one to two cents following the implementation of the MTU. Similarly, Becker et al. (2021) utilized a synthetic control method to examine the effects of the amendment of the Price Labelling Act in 2009 on fuel prices in Austria. Their base model included a control group consisting of 11 other European countries. The results indicated that shortly after the implementation of the regulation, the net prices of gasoline decreased by 23.4%, whereas Diesel net prices decreased by only 6.6%. Furthermore, gasoline prices dropped by 7.5%, and Diesel prices by 5.4% in the long run.

This paper is organized as follows: Section 2 considers the market structure of the German, Austrian, and French fuel markets. Section 3 provides an overview of the regulations in Germany, Austria, and France. Section 4 introduces the dataset and provides descriptive statistics. Section 5 presents the methodology used. Section 6 discusses the results of the empirical analysis. Finally, Section 7 concludes.

In May 2008 the German Federal Cartel Office initiated a sector investigation into the fuel market. Triggered by indications for antitrust concerns in the fuel sector as well as by complaints of independent gas station operators and consumers. The German fuel market is divided into two markets: federal highway gas stations and road gas stations. These segments differ in their competitive conditions. Road gas stations are in more or less strong competition with other gas stations located nearby. However, the operation of a federal highway gas station by a mineral oil company depends on the allocation of delivery and distribution rights by Tank & Rast GmbH, which is also responsible for the construction, operation, and leasing of federal highway gas stations (Bundeskartellamt, 2011). In this way, the gas station operators acquire a quasi-monopoly, as competitive relationships with other gas stations are scarcely present. Consequently, gasoline and diesel prices at highway gas stations are higher than at roadside gas stations.

Overall, there are around 350 gas stations on German highways. Fuel sales via federal highway gas stations are particularly lucrative for mineral oil companies, as high prices can be charged and around twice as much fuel is sold per federal highway gas station compared toa road gas station. Since the 1960s, Tank & Rast GmbH has awarded the delivery and distribution rights for fuel in line with the market share of road gas stations. Since January 1, 2013, the rights are no longer awarded exclusively according to market share, but also via auctions (Bundeskartellamt, 2011, pp. 7ff.).

The number of German roadside gas stations was around 14,700 in 2012, but has fallen to around 14,450 in recent years. Although sales per federal highway gas station are higher compared to a road gas station, the total sales of road gas stations are higher than that of federal highway gas stations, simply because of the significantly higher number of road gas stations. In addition, the sales ratio of Diesel and gasoline is balanced at road gas stations, whereas more Diesel fuel is sold at highway gas stations. This can be explained by the fact that most trucks and company cars are Diesel vehicles, which cover long distances and fill up more frequently at federal highway gas stations than at road gas stations (Bundeskartellamt, 2011, pp. 10ff.).

From an antitrust perspective, the road gas station market is segmented into Diesel and gasoline markets, each with regional limitations. BP (Aral), ConocoPhillips (Jet), ExxonMobil (Esso), Shell, and Total collectively form a dominant oligopoly in these regional gas station markets. The big five oil companies have significant market power, being the sole entities with access to refinery capacities in Germany and maintaining a nationwide network of gas stations. The market shares of these oil companies vary from one regional market to another. A nationwide sales share analysis reveals that the Big Five account for approximately 65% of annual fuel sales in Germany: BP (Aral) and Shell hold the largest shares, each around 20%, followed by ConocoPhillips (Jet) and Total, each with around 10%, and ExxonMobil (Esso) with over 5%. The remaining market share is distributed among a few larger companies such as Orlen, Hem-Tamoil, OMV, Eni, and Avia and the respective independent gas stations (Bundeskartellamt, 2011, pp. 10ff.; Doetsch, 2023).

With the introduction of the transparency unit, market transparency has significantly increased. Even before the implementation of the MTS, oil companies had a comprehensive overview of their competitors' pricing through the prices displayed on the price boards. High transparency facilitates possible parallel behaviour in pricing strategies. Furthermore, coordination among the oligopolists can also be facilitated by product homogeneity, as competition parameters are primarily narrowed down to fuel prices. Market transparency also implies that deviant behaviour by the oligopolists can be promptly identified and penalized accordingly. In addition to examining the market structure, a price analysis revealed that the oligopolistic market structure results in higher sales prices compared to those in a competitive market (Bundeskartellamt, 2011, pp. 12ff.).

The Austrian fuel market is similarly dominated by several major oil companies, including OMV, BP (Aral), Shell, Eni (Agip), and ConocoPhillips (Jet). The country has approximately 3,000 gas stations, although some of them exclusively sell diesel fuel. Both the number of gas stations and the market shares of these groups have remained relatively stable over the past few years. This stability can be attributed to various factors. On the supply side, there are barriers to market entry and exit, as companies must obtain official approval for gas station construction and have their own refineries or contracts with refineries. On the demand side, the market is relatively saturated, requiring new gas stations to compete regionally with well-established and strategically positioned counterparts (Bundeswettbewerbsbehörde, 2022, pp. 25ff.).

In Austria, Eni and Jet petrol stations primarily source their fuel from refineries in which they have no ownership stake. Conversely, OMV primarily obtains its fuel from its own refineries. BP sources approximately a third of its diesel fuel from its own production. Shell acquires roughly equal amounts from both its internal sources and external suppliers. Gas station operators outside the five main companies lack their own refineries and thus rely on the mineral oil companies. Therefore, it can be inferred that Austria also has an oligopolistic market structure (Bundeswettbewerbsbehörde, 2022, pp. 27f.).

In France, an oligopolistic market structure is also assumed. The market is dominated by a few major gas station operators such as TotalEnergies, BP (Aral), Shell, Eni (Agip), Avia, and ExxonMobil (Esso). TotalEnergies stands out as one of the largest traditional players in the fuel market in France (Si et al., 2023). In addition to the major gas stations also supermarket chains such as Carrefour, Leclerc, Auchan and Intermarché operate fuel stations located adjacent to their stores play an important role in French fuel markets. Moreover, also smaller, independently owned gas stations are scattered throughout France, particularly in rural and less densely populated areas.

As of 2024, there are about 11,000 gas stations in France, with a significant number located in urban areas and along major highways. Traditional stations, which include those operated by oil companies as well as independent stations, account for approximately 52% of the stations. Meanwhile, supermarket stations make up around 48%. Supermarket gas stations, known for generally lower prices compared to traditional stations, hold a significant market share, accounting for nearly 60% of fuels sold. This leaves traditional stations with about 40% of the market share. Interestingly, these market shares have remained relatively stable over the past decade (Statista, 2024).

France

The first regulation on market transparency was introduced in France. On December 12, 2006, the law on "Informing Consumers about the Retail Prices of Fuels" (JORF, 2006, No. 291, Text No. 17) was enacted. According to this law, every fuel retailer is required to publish its sales prices on the website operated by the Ministry of Economics, Finance and Industrial and Digital Sovereignty⁴. Changes in the sales price must be reported immediately, either through the Internet or voice services. Retailers can also indicate price changes in advance, provided they specify the date and time when the price change will take effect.

Austria

In Austria, on July 1, 2009, the "Price Labelling Act" (BGBl, 1992, No. 146; BGBl, 1992, No. 813) was extended by the enactment of the "Code of Conduct for Petrol Station Operators on the Timing of Fuel Price Indication at Petrol Stations" (BGBl, 2009, II, No. 190). This regulation stipulated that gas stations were only permitted to increase their prices once a day in the morning. For gas stations operating 24 hours a day, this meant setting their maximum price at 12:00 a.m. Gas stations with limited opening hours were required to set their maximum price at opening, while self-service stations had to set their maximum price by 8:30 a.m. Price reductions were allowed at any time.

In January 2011, the enactment was revised (BGBl, 2010, II, No. 484), specifying that gasoline and diesel prices could only be increased once a day at 12:00 p.m. Price reductions remained permissible at any time. Additionally, in August 2011, the "Price Transparency Enactment" (BGBl, 2011, I, No. 54) came into force, amending the "Price Transparency Act." According to this amendment, gas station operators in Austria are obligated to report current prices to the price transparency database of E-Control within half an hour of any price change. The revision made at the beginning of 2011 remained unaffected (Bundesministerium für Arbeit und Wirtschaft, 2024).

Furthermore, in 2012, gas station operators were prohibited from changing their prices between 6 June 2012, 11:00 a.m., and 10 June 2012,12:00 p.m.; between 28 June 2012, 11:00 a.m., and 1 July 2012, 12:00 p.m.; as well as between 5 July 2012, 11:00 a.m., and 8 July 2012, 12:00 p.m. (BGBl, 2012, II, No. 186). Since 2012, the Price Labelling Act, as well as the Price Transparency Ordinance, have been consistently renewed without major changes.

Germany

Since August 31, 2013, operators of German gas stations have been required to promptly report fuel price changes to the relevant agency of the German Federal Competition Authority (Bundeskartellamt) in real-time. This enables consumers to access the most up-to-date prices online or through apps at any given time. On December 1, 2013, the MTU for fuels in Germany (Markttransparenzstelle für Kraftstoffe) officially commenced operation following a three-month trial period. The MTU is jointly operated by the German Federal Competition Authority and the Federal Network Agency (Bundeskartellamt, 2024).

While precise information on actual costs of such transparency regulations is not publicly available, some general considerations can be made. It can be assumed that establishing a market transparency unit, as seen in Germany, involves substantial initial investments. This encompasses the development, setup, and implementation of technical infrastructure (such as software, hardware, databases, network security, interfaces with gas station operators and the Federal Cartel Office, etc.). Ongoing costs would include operational and administrative expenses (such as data preparation, analysis, maintenance, building rental, supervision, etc.). Additionally, gas station operators may incur costs related to the real-time transmission of prices.

The Federal Ministry for Economic Affairs and Technology estimated that the total conversion effort would amount to 3.96 million euros for large and medium-sized operators and 2.23 million euros for small operators. Additionally, ongoing costs were expected to be between 152,000 euros and 1.67 million euros. There would also be annual costs of 887,900 euros for updated data. For administration, one-time costs were estimated at 525,000 euros, with ongoing costs of approximately 2 million euros.⁵

The dataset consists of the publicly available fuel price database Weekly Oil Bulletin provided by the European Commission and covers the period from January 2005 to the end of December 2019 (European Commission, 2024). The observation period varies for each country due to changes in EU membership over time. In our analysis, we use the fuel prices of Euro-Super 95 and Automotive Gas Oil (Diesel), which are both pump prices. Member states communicate the prices that are most frequently charged based on a weighted average. Prices are reported as weekly fuel prices per 1000 litres in Euro.

Fuel price data are available with and without taxes. For comparability, we refer to prices excluding taxes, as taxes differ greatly between countries. Besides the data of Germany, Austria and France, we include information of 19 other EU-countries for our control group⁶. The prices from Bulgaria, Croatia, Malta, and Romania are not included because of missing data. In addition to Germany, Austria, and France, Belgium and Luxembourg have also introduced regulations concerning their fuel prices. However, a MTU has not been installed. Therefore, these countries are also removed from the sample to obtain suitable comparative markets.

Table 1 gives an overview of the descriptive statistics of the fuel prices (Euros per 1000 litre) without taxes for a period of 4 years for France, Austria and Germany. While the dataset of France includes 196 weekly observations over the period from January 2005 to December 2008, the data for Austria comprises 197 weekly observations over the period from January 2009 to December 2012. For Germany, the dataset consists of 196 weekly observations from January 2012 to December 2015.

For France, the average price for Super 95 is 457 euros, while Diesel averages 513 euros. Austria has higher average prices, with Super 95 at 544 euros and Diesel at 591 euros. Germany shows the highest average prices, at 647 euros for Super 95 and 677 euros for Diesel. This indicates that the average pricesof Super 95 are consistently lower than those of Diesel across all countries.

To understand these price differences, it is important to consider the specific factors influencing fuel demand and pricing in each country. In Germany and Austria, most commercial vehicles such as trucks, company cars, and agricultural vehicles are equipped with Diesel engines, resulting in relatively high demand. Additionally, Diesel prices are more affected by seasonal fluctuations. For instance, in winter, the demand for heating oil—which is refined similarly to Diesel—increases, driving up prices. These factors contribute to higher Diesel prices, explaining why the average value for Diesel is higher than that for gasoline (Sterner, 2012).

The differences in mean values among the countries are primarily due to the varying observation periods and the associated fluctuations in oil prices. Consequently, the lowest minimum value for Super 95 is 260 euros in France, while the highest maximum value is 848 euros for Diesel in Germany. Additionally, the standard deviation for Diesel is higher than for gasoline, ranging between 8 and 12 cents per litre.

Table 1 Descriptive statistics of fuel prices (Euros per 1000l) excluding taxes over a four-year period for France, Austria, and Germany.

Country	Period	Fuel	N	Mean	Std. Dev.	Min	1^st Qu.	Median	3^rd Qu.	Max
FR	01/2005 – 12/2008	Super 95	196	457	83	260	387	461	510	646
FR	01/2005 – 12/2008	Diesel	196	513	98	348	450	488	538	788
AT	01/2009 – 12/2013	Super 95	246	566	116	245	500	611	647	763
AT	01/2009 – 12/2013	Diesel	246	615	123	361	529	673	715	785
DE	01/2012 – 12/2015	Super 95	196	647	93	420	599	670	705	835
DE	01/2012 – 12/2015	Diesel	196	677	105	417	593	699	751	848

We analyse the effects of transparency regulations using the synthetic difference-in-differences (SDID) approach introduced by Arkhangelsky et al. (2021). It combines different aspects of the Synthetic Control (SC) and difference-in-differences (DID) method.

In the model, a balanced panel with \(\:T\) time periods (weekly from January 2005 to the end of December 2019) and \(\:N\) units (countries) is assumed. The outcome for country \(\:i\) in week \(\:t\) is denoted by \(\:{Y}_{it}\) and \(\:{W}_{it}\:\in\:\{0,\:1\}\) denotes the exposure to the binary treatment. Furthermore, it will be assumed that the control countries \(\:{N}_{co}\) (countries without regulations), are never treated, while the treated countries \(\:{N}_{tr}=N-{N}_{co}\) (countries with regulations), are treated after time \(\:{T}_{pre}\). Weights which align the pre-treatment trends in the outcome of untreated countries with those for the treated countries, denoted by \(\:{\widehat{\omega\:}}^{sdid}\), are also included, as well as time weights \(\:{\widehat{\lambda\:}}_{t}^{sdid}\) which balance the pre-treatment with the post-treatment time periods. This results in the following two-way fixed effects regression \(\:\tau\:\), to estimate the average causal effect of the treatment:

(1) \(\:\left({\widehat{\tau\:}}^{sdid},\:\widehat{\mu\:},\:\widehat{\alpha\:},\:\widehat{\beta\:}\right)=\begin{array}{c}\text{arg}\text{min}\\\:\tau\:,\:\mu\:,\:\alpha\:,\:\beta\:\end{array}\left\{\sum\:_{i=1}^{N}\sum\:_{t=1}^{T}{\left({Y}_{it}-\:\mu\:-\:{\alpha\:}_{i}-{\beta\:}_{t}-{W}_{it}\tau\:\right)}^{2}{\widehat{\omega\:}}_{i}^{sdid}{\widehat{\lambda\:}}_{t}^{sdid}\right\}\).

Comparatively, the DID method does not include country or time period weights.

(2) \(\:\left({\widehat{\tau\:}}^{did},\:\widehat{\mu\:},\:\widehat{\alpha\:},\:\widehat{\beta\:}\right)=\begin{array}{c}\text{arg}\text{min}\\\:\tau\:,\:\mu\:,\:\alpha\:,\:\beta\:\end{array}\left\{\sum\:_{i=1}^{N}\sum\:_{t=1}^{T}{\left({Y}_{it}-\:\mu\:-\:{\alpha\:}_{i}-{\beta\:}_{t}-{W}_{it}\tau\:\right)}^{2}\right\}\).

For the SDID estimator \(\:{\widehat{\tau\:}}^{sdid}\), time and country weights are used, giving more weight to time periods and countries that are on average similar to the treated periods to reduce potential violations of the parallel trend assumption. Furthermore, the localisation of the regression increases the robustness of the estimator by including similar countries and time periods, compared to the DID method. By removing the systematic parts of the outcome, weights may improve the precision of the estimator.

In this section, we analyse the effects of transparency regulations (see chapter 3) for Germany, Austria, and France. First, we build a base model for each country over a period of around two years (one year before and after the implementation). After that we run different robustness checks with changing time periods. On the one hand, we restrict the sample to a total of one year (six months before and six months after the implementation of the regulation). On the other hand, we extend the sample to a total of four years (two years before and two years after the implementation of the regulation). Due to the overlapping regulations in Austria, the base models as well as the robustness checks deviate from the periods mentioned above.

6.1 Base Model

Table 2 gives an overview of the observation periods of the base models for the different countries and regulations. First, we assume a period from January 2006 to the end of 2007 for France (F.I) and from September 2012 to the end of August 2014 for Germany (G.I). To analyse the effect of the Price Labelling Act Extension on July 2009 in Austria we assume a period from July 2008 until June 2010 (A.I). For the Price Labelling Act Revised on January 2011 we assume a period from January 2010 until July 2011 (A.II). Last, for the Price Transparency Enactment on August 2011 we observe a period from January 2011 until July 2012 (A.III).

Table 2 Overview Base Model Periods

Country	Model	Regulation	Period
France	F.I	12/12/2006: Law on Informing Consumers about the Retail Prices of Fuels	01/01/2006 – 31/12/2007
Austria	A.I	01/07/2009: Price Labelling Act Extension	01/07/2008 – 30/06/2010
	A.II	01/01/2011: Price Labelling Act Revised	01/01/2010 – 31/07/2011
	A.III	01/08/2011: Price Transparency Enactment	01/01/2011 – 31/07/2012
Germany	G.I	31/08/2013: Market Transparency Unit	01/09/2012 – 31/08/2014

6.1.1 France

We begin our analysis with France, which has already introduced a transparency regulation in 2006 and thus has the longest experience with such a measure. The base model covers a two-year period from January 2006 to December 2007. As shown in Table 3, all countries are included in the control group for Diesel and Super 95. Ireland was excluded from the Super 95 sample. In particular, Super 95 prices from Estonia (10.2%), Lithuania (9.7%), the Netherlands (8.3%), and Sweden (10.5%) are weighted to a greater extent, whereas Czechia (0.9%), Poland (1.1%), and Slovenia (0.5%) play only minor roles in the control group. Considering the Diesel prices, the control group is particularly built on Denmark (9%), the Netherlands (8.8%), Spain (8.4%), and Sweden (7.8%). In contrast, Estonia (0.5%) and Ireland (1.3%) are barely included.

Our results in Table 4 show hardly any significant average treatment effect (ATT). While the price of Super 95 has increased by about 0.064 cents, the price of Diesel has increased by about 0.04 cents. However, both price increases are statistically insignificant. In addition, from the graphical analysis (see Figure 1), no price reduction can be derived for either Diesel or Super 95. At first glance, transparency regulations have no obvious effects on fuel prices in France.

Table 3 Model F.I: Overview of the assigned weights for each country

Country	Super 95	Diesel
Cyprus	0.041	0.044
Czechia	0.009	0.044
Denmark	0.077	0.090
Estonia	0.102	0.005
Finland	0.018	0.044
Greece	0.078	0.061
Hungary	0.041	0.051
Ireland	0.000	0.013
Italy	0.063	0.072
Latvia	0.041	0.048
Lithuania	0.097	0.032
Netherlands	0.083	0.088
Poland	0.011	0.042
Portugal	0.032	0.066
Slovakia	0.060	0.056
Slovenia	0.005	0.037
Spain	0.068	0.084
Sweden	0.105	0.078
United Kingdom	0.069	0.047
^{Note: Period: 01/2006 to 12/2007.}

Table 4 Model F.I: Average treatment effects for France

	ATT	Std. Err.	t	P > \|t\|	[95% Conf. Interval]
Super 95	6.406	8.704	0.740	0.462	-10.654	23.465
Diesel	3.910	7.229	0.540	0.592	-10.396	18.217
^{Note: The control group consists of a weighted average of all available countries. Period: 01/2006 until 12/2007.}

6.1.2 Austria

Next, we analyse the effects of the different regulations and treatments in Austria: The extension of the “Price Labelling Act” by the “Code of Conduct for Petrol Station Operators on the Timing of Fuel Price Indication at Petrol Stations” in July 2009, the adjustment of this code of conduct in January 2011 as well as the “Price Transparency Enactment” in August 2011. Because some of the laws and related measures overlap, it is difficult to separate the potential resulting effects. However, we attempt to account for the overlap in determining the observation periods.

6.1.2.1 Model A.I: Price Labelling Act Extension

First, we analyse the effect of the extension of the “Price Labelling Act” in July 2009. We assume an observation period of two years (July 2008 to June 2010).

Table 5 lists the weights assigned to the control group. For Super 95, the method particularly involves Denmark (8.6%), Sweden (7.8%) and Estonia (6.6%), whereas for Diesel Denmark (9.2%), Sweden (9%) and Italy (8.1%) are weighted the most. United Kingdom (3.1%), Poland (3.4%) and Czechia (3.5%) are weighted least for the prices of Super 95. For Diesel, Ireland is not included, Finland (3.4%) and Slovenia (3.6%) are weighted the least.

Table 6 shows the average treatment effects of this extension on the Super 95 and Diesel prices in Austria. It is noticeable that the average treatment effect on the Super 95 prices is positive with 0.7 cent per litre, whereas the effect on the Diesel prices is negative with -1 cent per litre. However, both effects are not statistically significant. The graphical analysis of the average treatment effects over time also shows that no significant effect can be identified (see Figure 2).

Table 5 Model A.I: Overview of the assigned weights for each country

Country	Super 95	Diesel
Cyprus	0.054	0.046
Czechia	0.035	0.043
Denmark	0.086	0.092
Estonia	0.066	0.058
Finland	0.055	0.034
Greece	0.051	0.052
Hungary	0.045	0.060
Ireland	0.040	0.000
Italy	0.063	0.081
Latvia	0.045	0.040
Lithuania	0.062	0.069
Netherlands	0.050	0.061
Poland	0.034	0.047
Portugal	0.052	0.057
Slovakia	0.045	0.018
Slovenia	0.042	0.036
Spain	0.055	0.077
Sweden	0.078	0.090
United Kingdom	0.031	0.039
^{Note: Period: 07/2008 until 06/2010.}

Table 6 Model A.I: Average Treatment Effects of the Price Labelling Act Extension

	ATT	Std. Err.	t	P > \|t\|	[95% Conf. Interval]
Super 95	6.652	14.115	0.47	0.637	-21.012	34.317
Diesel	-9.720	9.528	-1.02	0.308	-28.394	8.955
^{Note: The control group consists of a weighted average of all available countries. Period: 07/2008 until 06/2010.}

6.1.2.2 Model A.II: Price Labelling Act Revised

Now, we analyse the effect of the revised version of the “Price Labelling Act” in January 2011. The Super 95 and Diesel prices may only be increased once a day at 12:00 p.m. We assume an observation period of one and a half years (January 2010 to July 2011) to exclude the effect of the “Price Transparency Enactment”.

Table 7 summarizes the weights of the countries for the synthetic control groups. In the synthetic control group for the Super 95 prices, Slovakia (8.6%), Finland (8.1%), Czechia (7.7%) and Spain (7.7%) are included. Estonia (0.7%), Cyprus (1.1%) and Slovenia (1.7%) are the least represented in the control group. In contrast, Denmark (9.6%), the Netherlands (9.4%), and Spain (9%) are the most represented in the Diesel control group. Estonia is not included, and Ireland is included the least (0.7%).

The average effect of the revised “Price Labelling Act” on the Super 95 prices is negative with -1.68 cent per litre, but the effect is not significant. The average effect on the Diesel prices is negative (-2.14 cent per litre) and statistically significant at 5%-level (see table 8). This may indicate that the regulation that gas stations are only allowed to increase their prices once a day was accompanied by a price reduction of around 2 cents per litre during this period, at least for Diesel. Figure 3 compares the prices during the period between the treated country Austria and the control group. Because of the small ATTs it is difficult to point out a graphical difference after the implementation. However, it is striking that just before the revision of the Price Labelling Act the prices of Austria fall quite sharply only to rise again after implementation. This price drop is not recognizable for the control group. The Super 95 prices do not increase as much as the Diesel prices after implementation. After around 1 month of the implementation the prices of Austria decrease again.

Table 7 Model A.II: Overview of the assigned weights for each country

Country	Super 95	Diesel
Cyprus	0.011	0.042
Czechia	0.077	0.044
Denmark	0.071	0.096
Estonia	0.007	0.000
Finland	0.081	0.041
Greece	0.023	0.063
Hungary	0.062	0.049
Ireland	0.033	0.007
Italy	0.057	0.074
Latvia	0.063	0.047
Lithuania	0.027	0.026
Netherlands	0.073	0.094
Poland	0.046	0.040
Portugal	0.050	0.068
Slovakia	0.086	0.059
Slovenia	0.017	0.034
Spain	0.077	0.090
Sweden	0.065	0.080
United Kingdom	0.067	0.045
^{Note: Period: 01/2010 until 07/2011.}

Table 8 Model A.II: Average Treatment Effects of the Price Labelling Act Revised

	ATT	Std. Err.	t	P > \|t\|	[95% Conf. Interval]
Super 95	-16.814	13.039	-1.29	0.197	-42.370	8.743
Diesel	-21.356	9.748	-2.19	0.028	-40.463	-2.250
^{Note: The control group consists of a weighted average of all available countries. Period: 01/2010 until 07/2011.}

3.1.2.3 Model A.III: Price Transparency Enactment

Next, we analyse the effects of transparency regulation in Austria in August 2011. Again, we assume an observation period of one and a half years (January 2011 to July 2012). The period starts with the revised “Price Labelling Act”. Since this act continues to run during the introduction of the “Price Transparency Enactment”, we only assume an observation period in which the act already came into force.

Table 9 lists the weights assigned to the control group. All countries are included in the analysis for both Super 95 and Diesel. For Super 95, Denmark (9.7%), Sweden (7.9%), the Netherlands (7.4%), Ireland (7.2%) and Portugal (7.1%) are weighted more heavily. For Diesel, Denmark (9.3%), Spain (7.5%), the Netherlands (7.2%), and Portugal (7%) have particularly high weights. The Super 95 prices of Cyprus (1.7%) and Estonia (0.2%) are included with weights of less than 2%. For the Diesel control group, the prices of Ireland (1.3%) and United Kingdom (2.4%) have the lowest weights..

Table 9 Model A.III: Overview of the assigned weights for each country

Country	Super 95	Diesel
Cyprus	0.017	0.053
Czechia	0.040	0.030
Denmark	0.097	0.093
Estonia	0.002	0.056
Finland	0.060	0.050
Greece	0.052	0.044
Hungary	0.047	0.046
Ireland	0.072	0.014
Italy	0.045	0.067
Latvia	0.058	0.064
Lithuania	0.058	0.067
Netherlands	0.074	0.072
Poland	0.045	0.040
Portugal	0.071	0.070
Slovakia	0.037	0.030
Slovenia	0.043	0.037
Spain	0.055	0.075
Sweden	0.079	0.069
United Kingdom	0.049	0.024
Note: Period: 01/2011 until 07/2012.

Table 10 summarizes the average effects of transparency enactment on fuel prices in Austria. Overall, relatively small positive effects for Super 95 (1.5 cent per litre) and for Diesel (0.3 cent per litre) can be observed. However, neither of the effects are statistically significant. Moreover, Figure 4 shows no significant effect. Approximately two weeks before the implementation Austrian prices decreased by about 5 euros per 1000 litre compared to the control group for either Diesel or Super 95. After the implementation the prices rise briefly before they fall again as the control group.

Table 10 Model A.III: Average treatment effects of the Price Transparency Enactment

	ATT	Std. Err.	t	P > \|t\|	[95% Conf. Interval]
Super 95	15.387	9.948	1.55	0.122	-4.110	34.885
Diesel	2.911	11.259	0.26	0.796	-19.156	24.977
^{Note: The control group consists of a weighted average of all available countries. Period: 01/2011 until 07/2012.}

6.1.3 Germany

Turning to Germany, even though the MTU only went into regular operation on December 1, 2013, we assume the start of the trial run as implementation date, August 31, 2013. We use a symmetric sample (September 2012 to August 2014) around the time of introduction (August 31, 2013).

Table 11 summarizes the weights for each European country in our sample. For Super 95, Poland and Slovenia are not included. For Diesel, Poland and Finland are skipped from the control group. Denmark (15.2%) and Sweden (11.7%) contribute the most to the calculation of the Super 95 control group. Prices from Denmark (15.3%), Hungary (11.6%), and Sweden (15.5%) have greater weight in the Diesel control group. Conversely, Greece (1.9%) has minimal contribution to the Super 95 control group. Similarly, Greece (1.6%), Slovenia (1.6%), and Spain (0.2%) are barely included in the Diesel control group.

The average treatment effects for Super 95 and Diesel are both negative but rather small (1-2 cents per litre). Both effects are not statistically significant (see Table 12). The graphical analysis also shows no statistically significant effect (see Figure 5).

Table 11 Model G.I: Overview of the assigned weights for each country

Country	Super 95	Diesel
Cyprus	0.028	0.029
Czechia	0.071	0.072
Denmark	0.152	0.153
Estonia	0.023	0.064
Finland	0.050	0.000
Greece	0.019	0.016
Hungary	0.038	0.116
Ireland	0.067	0.022
Italy	0.023	0.067
Latvia	0.067	0.022
Lithuania	0.081	0.044
Netherlands	0.054	0.072
Poland	0.000	0.000
Portugal	0.067	0.085
Slovakia	0.040	0.041
Slovenia	0.000	0.016
Spain	0.064	0.002
Sweden	0.117	0.155
United Kingdom	0.039	0.024
^{Note: Period: 09/2012 until 08/2014.}

Table 12 Model G.I: Average treatment effect for Germany

	ATT	Std. Err.	t	P > \|t\|	[95% Conf. Interval]
Super 95	-12.432	8.367	-1.490	0.137	-28.830	3.967
Diesel	-18.972	12.443	-1.520	0.127	-43.360	5.417
^{Note: The control group consists of a weighted average of all available countries. Period: 09/2012 to 08/2014.}

Overall, there is only little evidence for price-reducing effects of transparency regulations in France, Austria and Germany. If any, there are only a few and minor price reductions after the introduction of the regulations.

6.2 Robustness checks

A possible determining factor for the measurement of significant effects is the underlying sample period. A significant effect may only be detectable in the long term or may disappear after just a few months. Therefore, we now vary the sample periods regarding all countries under scrutiny, analysing periods of about 4 years and one year, respectively. Table 13 gives an overview of the observed periods.

Table 13 Overview Robustness Checks Periods

Country	Regulation	Model	Period
France	12/12/2006: Law on Informing Consumers about the Retail Prices of Fuels	F.II	01/07/2006 – 30/06/2007
France		F.III	01/01/2005 – 31/12/2008
Austria	01/07/2009: Price Labelling Act Extension	A.IV	01/01/2009 – 31/12/2009
	01/07/2009: Price Labelling Act Extension	A.V	01/07/2007 – 31/12/2010
	01/01/2011: Price Labelling Act Revised	A.VI	01/02/2011 – 31/01/2012
	01/01/2011: Price Labelling Act Revised	A.VII	01/01/2011 – 31/07/2013
	01/08/2011: Price Transparency Enactment	A.VIII	01/02/2011 – 31/01/2012
	01/08/2011: Price Transparency Enactment	A.IX	01/01/2011 – 31/07/2013
Germany	31/08/2013: Market Transparency Unit	G.II	01/03/2013 – 28/02/2014
Germany	31/08/2013: Market Transparency Unit	G:III	01/09/2011 – 31/08/2015

3.2.1 France

First, we analyse the fuel prices of France for the periods of January 2005 until the end of December 2008, as well as from July 2006 to the end of June 2007.

For both periods, the effects are less than 1 cent per litre and not statistically significant.⁷ Therefore, a similar pattern can be observed after the implementation of the treatment independently from the sample period. The prices increase in the beginning of 2008, followed by a sudden drop in the end of 2008, which may be an effect of the global crisis in 2008/2009 (Mau & Ulyukaev, 2015). Overall, none statistically and economically significant effect can be determined.

3.2.2 Austria

We now analyse further different periods of Austria. Due to the overlapping treatments the periods are not symmetric around the implementations. Compared to the base models of Austria, we analyse both a shorter and a longer period.

To analyse the “Price Labelling Act Extension” in July 2009, we consider the periods July 2007 until December 2010 as well as January 2009 until December 2009.⁸As well as in the base model, we find no significant results for both Super 95 and Diesel prices.⁹ When analysing the average treatment effects for the “Price Labelling Act Revised” in January 2011, the observation periods range from July 2010 until June 2011 and July 2009 until July 2011. Overall, results show that except of Super 95 prices from July 2009 until July 2011, there are significant negative effects of 2 to 3 cents per litre.¹⁰ Analysing the “Price Transparency Enactment” in August 2011 by using periods from January 2011 until July 2013 and February 2011 until January 2012, there is a significant positive effect of 1.5 cent per litre on Super 95 prices for the period from February 2011 to January 2012. However, all other average treatment effects are not significant. ¹¹

3.2.3 Germany

Finally, we analyse the periods from March 2013 to February 2014 and from September 2011 to August 2015 for Germany. A statistically significant result only emerges in model G.II for Super 95 and in model G.III for Diesel. In both cases, the indicates significance at the 10% level, with both effects being relatively small, less than 2 cents per litre.¹²

6.3 Comparison of results

Table 14 provides an overview of the average effects of the transparency regulations for each model to compare the results of the different periods for Austria, Germany and France. In case of France, positive effects amount to less than one cent per litre. Moreover, the results are statistically insignificant for both Super 95 and Diesel in all model specifications. Also, for Austria the effects are consistently positive. The effects on Diesel are less than one cent per little, but not statistically significant, as for France. The effect on Diesel is higher with around 1.5 cents per litre. However, the effect is only significant for the short period from February 2011 until January 2012 (A.VIII). In contrast, the average treatment effects of Germany are negative for alle model specifications. For, Super 95 only in the short period (G.II) a significant effect of -1.7 cents per litre can be obtained. In comparison, for Diesel only in the long period (G.III) with -1.8 cents per litre.

Therefore, it can be assumed that the transparency regulations of individual countries have either no effect or only a minimal effect on fuel prices. In case of France and Austria, the results could indicate that the laws have increased not only vertical but also horizontal market transparency. In fact, it seems like the effect of the horizontal transparency may be greater than the vertical one, whereas the results of Germany indicate the opposite.

Table 14 Overview of the obtained ATT for Transparency Treatments

Fuel	Country	Model	Period	ATT	Std. Err.	t	P > \|t\|
Super 95	France	F.I	01/2006 – 12/2007	6.406	8.704	0.74	0.462
		F.II	07/2006 – 06/2007	8.704	11.197	0.78	0.437
		F.III	01/2005 – 12/2008	5.916	11.713	0.51	0.613
	Austria	A.III	01/2011 – 07/2012	15.387	9.948	1.55	0.122
		A.VIII	02/2011 – 01/2012	15.298	7.543	2.03	0.043
		A.IX	01/2011 – 07/2013	14.878	13.021	1.14	0.253
	Germany	G.I	09/2012 – 08/2014	-12.432	8.367	-1.49	0.137
		G.II	03/2013 – 02/2014	-16.651	9.091	-1.83	0.067
		G.III	09/2011 – 08/2015	-16.255	12.189	-1.33	0.182
Diesel	France	F.I	01/2006 – 12/2007	3.910	7.229	0.54	0.592
		F.II	07/2006 – 06/2007	3.936	6.293	0.63	0.532
		F.III	01/2005 – 12/2008	3.327	10.735	0.31	0.757
	Austria	A.III	01/2011 – 07/2012	2.911	11.259	0.26	0.796
		A.VIII	02/2011 – 01/2012	5.141	11.301	0.45	0.649
		A.IX	01/2011 – 07/2013	3.302	13.568	0.24	0.808
	Germany	G.I	09/2012 – 08/2014	-18.972	12.443	-1.52	0.127
		G.II	03/2013 – 02/2014	-6.223	7.192	-0.87	0.387
		G.III	09/2011 – 08/2015	-17.728	9.685	-1.83	0.067
^{Note: The control group consists of a weighted average of all available countries.}

Table 15 compares the different regulations in Austria. The effects of the Price Labelling Act Extension are for Super 95 positive and for Diesel in the short period (A.IV) positive and for the longer periods (A.I, A.V) negative. The effects of the Price Labelling Act Extension are not significant. In contrast, the effects of the Price Labelling Act Revised are continuously negative. For Super 95 the effects amount -2 to -3 cents per litre, whereas the effect in the short period (A.VI) is significant. For Diesel the effects are all significant and amount around -2 cents per litre. As already shown in table 19, all effects of the Price Transparency Enactment are positive and not significant except for the short period of Super 95 (A.VIII). This may indicate that transparency regulations can tend to increase prices, whereas regulations which affect the pricing of gas stations may reduce prices. It is striking that there are different results for the Price Labelling Act Extension and for the Price Labelling Act Revised, although only the reporting time of the maximum price has changed (see chapter 3). This could be related to the fact that a maximum price set at 12:00 p.m. is noticed more by consumers than a change at 8:30 a.m. or 12:00 a.m. As a result, gas stations may not be able to raise their maximum price as much.

Table 15 Overview of the obtained ATT for the treatments in Austria

Fuel	Treatment	Model	Period	ATT	Std. Err.	t	P > \|t\|
Super 95	Price Labelling Act Extension July 2009	A.I	07/2008 – 06/2010	6.652	14.115	0.47	0.637
		A.IV	01/2009 – 12/2009	10.368	13.737	0.75	0.450
		A.V	07/2007 – 12/2010	5.737	13.177	-0.44	0.663
	Price Labelling Act Revised January 2011	A.II	01/2010 – 07/2011	-16.814	13.039	-1.29	0.197
		A.VI	07/2010 – 06/2011	-29.807	11.425	-2.61	0.009
		A.VII	07/2009 – 07/2011	-22.743	14.060	-1.62	0.106
	Price Transparency Enactment August 2011	A.III	01/2011 – 07/2012	15.387	9.948	1.55	0.122
		A.VIII	02/2011 – 01/2012	15.298	7.543	2.03	0.043
		A.IX	01/2011 – 07/2013	14.878	13.021	1.14	0.253
Diesel	Price Labelling Act Extension July 2009	A.I	07/2008 – 06/2010	-9.720	9.528	-1.02	0.308
		A.IV	01/2009 – 12/2009	1.841	9.722	0.19	0.850
		A.V	07/2007 – 12/2010	-8.576	12.558	-0.68	0.495
	Price Labelling Act Revised January 2011	A.II	01/2010 – 07/2011	-21.356	9.748	-2.19	0.028
		A.VI	07/2010 – 06/2011	-23.727	8.980	-2.64	0.008
		A.VII	07/2009 – 07/2011	-23.324	12.200	-1.91	0.056
	Price Transparency Enactment August 2011	A.III	01/2011 – 07/2012	2.911	11.259	0.26	0.796
		A.VIII	02/2011 – 01/2012	5.141	11.301	0.45	0.649
		A.IX	01/2011 – 07/2013	3.302	13.568	0.24	0.808
^{Note: The control group consists of a weighted average of all available countries.}

In summary, it can be assumed that the effect of individual countries' regulation has no or only a very small effect on fuel prices. In Austria it is noticeable that the transparency regulation has a positive effect, whereas the Price Labelling Act Revised has a negative effect on the fuel prices. Also, for France there are positive effects due to their transparency regulation, but not significant. By contrast, Germany’s MTU shows negative effects for both Super 95 and Diesel.

This paper analyses the impact of transparency regulations on fuel prices in Austria, Germany and France, utilizing a synthetic difference-in-differences approach (SDID) as an identification strategy.

Using data from the Weekly Oil Bulletin which provides information on fuel prices for all member states of the European Union, we find only little evidence for a price-reducing effect of transparency regulation. While results for France indicate positive albeit tiny effects, results for Germany suggest that prices have fallen only slightly due to the introduction of the transparency regulation. However, price effects in France are statistically insignificant and even the results of Germany are only significant for Super 95 in the short-term and for Diesel in the long run. In addition to the transparency regulation, Austria also introduced two other regulations that specifically affected prices. Gas station operators are only allowed to increase prices once a day at a fixed time. While the 2009 regulation, which required gas station operators to set their maximum price at 12:00 a.m., had no significant price effect, the 2011 regulation, which required them to set their maximum price at 12:00 p.m., had a short-term negative price effect of 2 to 3 cents per litre. For Diesel, even for the observation period of one year there is a negative statistically significant effect of 2 cents per litre. Like for France, the transparency regulation in Austria had slightly positive effects on the fuel prices (effect on Super 95 higher than on Diesel). Except of the short observation period for Super 95, all results were statistically insignificant.

In summary, significant and long-lasting effects of transparency regulations on fuel prices are largely absent. Short-term decreases in prices typically dissipate within a year or even earlier. The findings suggest that transparency regulations do not yield the intended price-reducing effects on fuel markets, with the influence of horizontal transparency prevailing. Consequently, the costs associated with implementing and maintaining transparency regulations in fuel markets should be carefully weighed against the minimal or nearly non-existent benefits for consumers. Although the results from Austria indicate that price regulation may lower fuel prices, our findings are not sufficient to recommend further regulatory interventions in fuel markets.

Funding information:

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Contribution

While Lea Bernhardt have equally contributed, Xenia Breiderhoff has the largest share in the data analysis of the paper.

Arkhangelsky, D., Athey, S., Hirshberg, D. A., Imbens, G. W., & Wager, S. (2021). Synthetic Difference-in-Differences. American Economic Review, 111(12), 4088–4118. https://doi.org/10.1257/aer.20190159
Assad, S., Clark, R., Ershov, D., & Xu, L. (2020). Algorithmic Pricing and Competition: Empirical Evidence from the German Retail Gasoline Market (No. 8521; CESifo). https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3682021
Becker, M., Pfeifer, G., & Schweikert, K. (2021). Price Effects of the Austrian Fuel Price Fixing Act: A Synthetic Control Study. Energy Economics, 97. https://www.sciencedirect.com/science/article/abs/pii/S0140988321001122
BGBl (1992). Preisauszeichnung für bestimmte Leistungen und für Treibstoffe bei Tankstellen, Pub. L. No. 813, Bundesgesetzblatt, https://rdb.manz.at/document/ris.c.BGBL_OS_19921223_0_0813++?source=726462233230323330393135237269732e632e4247426c5f5f49495f4e725f5f3139305f3230303923534c2334303534323032373936
BGBl (1992). Preisauszeichnungsgesetz (PrAG), Pub. L. No. 146, Bundesgesetzblatt, https://rdb.manz.at/document/ris.c.BGBL_OS_19920319_0_0146
BGBl (2009). Standesregeln für Tankstellenbetreiber über den Zeitpunkt der Preisauszeichnung für Treibstoffe bei Tankstellen, Pub. L. No. II 190, Bundesgesetzblatt, https://www.ris.bka.gv.at/eli/bgbl/II/2009/190
BGBl (2010). Standesregeln für Tankstellenbetreiber über den Zeitpunkt der Preisauszeichnung für Treibstoffe bei Tankstellen, Pub. L. No. II 484, Bundesgesetzblatt, https://www.ris.bka.gv.at/eli/bgbl/II/2010/484
BGBl (2011). Änderung des Preistransparenzgesetzes, Pub. L. No. I 54, Bundesgesetzblatt , https://www.ris.bka.gv.at/Dokumente/BgblAuth/BGBLA_2011_I_54/BGBLA_2011_I_54.pdfsig
BGBl (2012). Änderungen der Verordnung betreffend Standesregeln für Tankstellenbetreiber über den Zeitpunkt der Preisauszeichnung für Treibstoffe bei Tankstellen, Pub. L. No. II 186, Bundesgesetzblatt, https://www.ris.bka.gv.at/Dokumente/BgblAuth/BGBLA_2012_II_186/BGBLA_2012_II_186.pdfsig
Bundeskartellamt (2011). Sektoruntersuchung Kraftstoffe - Abschlussbericht Mai 2011 - Zusammenfassung. https://www.bundeskartellamt.de/SharedDocs/Publikation/DE/Sektoruntersuchungen/Sektoruntersuchung%20Kraftstoffe%20-%20Zusammenfassung.pdf?__blob=publicationFile&v=1
Bundeskartellamt (2024). Markttransparenzstelle für Kraftstoffe. https://www.bundeskartellamt.de/DE/Aufgaben/MarkttransparenzstelleFuerKraftstoffe/markttransparenzstellefuerkraftstoffe_node.html
Bundesministerium für Arbeit und Wirtschaft (2024, May 23). Preispolitik. https://www.bmaw.gv.at/Themen/Europa/Wettbewerbspolitik/Preispolitik.html
Bundeswettbewerbsbehörde. (2022). Branchenuntersuchung Kraftstoffmarkt. https://www.bwb.gv.at/branchenuntersuchungen/
Dewenter, R., Heimeshoff, U., & Lüth, H. (2017). The impact of the market transparency unit for fuels on gasoline prices in Germany. Applied Economics Letters, 24(5), 302–305. https://doi.org/10.1080/13504851.2016.1184371
Doetsch, E., & Energie Informationsdienst. (2023). Marktanteile der führenden Tankstellenmarken am gesamten Kraftstoffabsatz in Deutschland im Jahr 2022. Statista. https://de.statista.com/statistik/daten/studie/3444/umfrage/marktanteile-auf-dem-tankstellenmarkt-in-deutschland/
European Commission. (2024, May 23). Weekly Oil Bulletin. European Commission. https://energy.ec.europa.eu/data-and-analysis/weekly-oil-bulletin_en
Horvath, M. (2019). Germany’s Market Transparency Unit for Fuels: Fostering Collusion or Competition? (No. 836; Ruhr Economic Papers). https://www.econstor.eu/bitstream/10419/213494/1/168912864X.pdf
JORF (2006). Arrêté du 12 décembre 2006 relatif à l’information du consommateur sur les prix de vente des carburants, Pub. L. No. 291, Text No. 17, JORF (JOURNAL OFFICIEL DE LA RÉPUBLIQUE FRANÇAISE) (2006). https://www.legifrance.gouv.fr/eli/arrete/2006/12/12/ECOC0600171A/jo/texte
Kühn, K.-U., & Vives, X. (1995). Information Exchanges Among Firms and their Impact on Competition. https://blog.iese.edu/xvives/files/2011/09/Information-Exchanges-and-their-Impact-on-Competition.pdf
Le, T.-H., Le, A. T., & Le, H.-C. (2021). The historic oil price fluctuation during the Covid-19 pandemic: What are the causes? Research in International Business and Finance, 58. https://doi.org/10.1016/j.ribaf.2021.101489
Møllgaard, P., & Overgaard, P. B. (2001). Market Transparency and Competition Policy. https://www.researchgate.net/publication/46524724
Pailanir, D. (2022), sdid – Sythetic Difference-in-Differences for Stata, https://github.com/Daniel-Pailanir/sdid
OECD (2017). Algorithms and Collusion: Competition Policy in the Digital Age. https://www.oecd.org/competition/algorithms-collusion-competition-policy-in-the-digital-age.htm
Si, Y., Desai, D., Bozhilova, D., Puffer, S., & Stephens, J. C. (2023). Fossil fuel companies’ climate communication strategies: Industry messaging on renewables and natural gas. Energy Research and Social Science, 98. https://doi.org/10.1016/j.erss.2023.103028
Sokhanvar, A., Çiftçioğlu, S., & Lee, C.-C. (2023). The effect of energy price shocks on commodity currencies during the war in Ukraine. Resources Policy, 82. https://doi.org/10.1016/j.resourpol.2023.103571
Statista (2024). Parts de marché des stations-service dans la vente des carburants en France de 2005 à 2022, par type de reseau, https://fr.statista.com/statistiques/503607/ventes-carburants-routiers-reseau-france/
Sterner, T. (2012). Distributional effects of taxing transport fuel. Energy Policy, 41, 75–83. https://doi.org/10.1016/j.enpol.2010.03.012
Stiglitz, J. E. (1989). Imperfect Information in the Product Market. In Handbook of Industrial Organization (Vol. 1, pp. 769–847). https://www.sciencedirect.com/science/article/pii/S1573448X89010162
Varian, H. (2019). Artifical Intelligence, Economics, and Industrial Organization. In A. Agrawal, J. Gans, & A. Goldfarb (Eds.), The Economics of Artificial Intelligence (pp. 399–422). University of Chicago Press. https://www.degruyter.com/document/doi/10.7208/9780226613475-018/pdf
Vives, X. (1990). Trade association disclosure rules, incentives to share information, and welfare. RAND Journal of Economics, 21(3), 409–430. https://blog.iese.edu/xvives/files/2011/09/79.pdf

Ministère de l'économie, des finances et des la souveraineté industrielle et numérique. Available at: www.prix-carburants.gouv.fr (June 7, 2024).
See Deutscher Bundestag, Verordnung zur Markttransparenzstelle für Kraftstoffe (MTS-Kraftstoff-Verordnung) https://dserver.bundestag.de/btd/17/123/1712390.pdf (June 1, 2024).
The control group consists of Cyprus, Czechia, Denmark, Estonia, Finland, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Netherlands, Poland, Portugal, Slovakia, Slovenia, Spain, Sweden, United Kingdom.
Results are not shown here and can be provided on request.
Results can be provided on request. See Table 14 for a summary of the results.
Results can be provided on request. See Table 14 for a summary of the results.
Results can be provided on request. See Table 14 for a summary of the results.
Results can be provided on request. See Table 14 for a summary of the results.
Results can be provided on request. See Table 14 for a summary of the results.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

New Evidence on Price Effects of Transparency Regulations in European Fuel Markets

Status:

Version 1

Abstract

Figures

1. Introduction

2. Fuel Market Structure

3. Regulations in European fuel markets

4. Data

5. Method

6. Results and Discussion

6.1 Base Model

6.1.1 France

6.1.2 Austria

6.1.2.1 Model A.I: Price Labelling Act Extension

6.1.2.2 Model A.II: Price Labelling Act Revised

3.1.2.3 Model A.III: Price Transparency Enactment

6.1.3 Germany

6.2 Robustness checks

3.2.1 France

3.2.2 Austria

3.2.3 Germany

6.3 Comparison of results

7. Conclusion and Policy Implications

Declarations

Funding information:

Author Contribution

References

Footnotes

Additional Declarations

Status:

Version 1