The Optimal Reset-Hour of a Once-Daily Petrol Price Increase Limit

Christoph Siemroth · 2026

Read The Optimal Reset-Hour of a Once-Daily Petrol Price Increase Limit by Christoph Siemroth (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-18920.txt.

Introduction

Siemroth asks when a daily petrol price reset should occur under a proposed German rule. The rule would let stations raise prices at most once a day. He infers how price-sensitive drivers are at each hour from station price data. Prices peak at seven in the morning and bottom out at seven in the evening. He tests every reset hour in his model. An eleven o'clock reset gives the lowest traffic-weighted price, with flat prices all day in that case.

Catalog record

CREATORChristoph Siemroth
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE44370
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.18920

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The Optimal Reset-Hour of a Once-Daily Petrol Price Increase Limit Christoph Siemroth Abstract A German ministry recently proposed a limit of at most one price increase per day for petrol stations. At what time should the price reset be allowed in order to lower price levels the most throughout the day? To answer this question, I infer the share of price-sensitive consumers for every hour of the day from German petrol station price data, based on a simple spatial-competition model. I focus on weekdays, which are the relevant target because commuter demand is less flexible than weekend demand. Hourly petrol station prices peak at 07:00 and bottom out at 19:00. Given the inferred composition of price-sensitivity throughout the day and hourly passenger-car traffic frequencies as a proxy for quantity, I evaluate every possible reset-hour of the new policy. The lowest traffic-weighted average price is achieved by an 11:00 reset. With this reset-hour, the resulting equilibrium price throughout the day is constant. This would lead to lower prices in the morning but higher prices in the evening, harming price-sensitive consumers but benefiting morning commuters and firms. Keywords : retail gasoline, dynamic pricing, price discrimination, regulation, spatial competition JEL Classification : L13, L41, L81 1 Introduction In March 2026, the Iran war pushed crude oil prices and petrol-station prices up throughout the world. Consumers in many countries called for mitigation in the form of price controls or petrol tax decreases. Going a slightly different route, on 11 March 2026, the German Federal Ministry of Economic Affairs proposed that petrol stations would only be allowed to increase prices once per day ( 15 ) , whereas the number of price decreases would remain unlimited. No constraints on the magnitude of price changes would be imposed. Austria has been using this kind of regulation since 2011 ( 20 ) . A possible reason why petrol station prices currently change so much and often is that stations price-discriminate across consumers at different times of the day. In the morning, commuters and firms are more likely to be forced to buy petrol for work, so they are relatively price-insensitive. In the evening, there is little work-related traffic, so consumers are more price sensitive. This pattern explains why petrol station prices tend to be highest in the morning and lowest in the evening ( 12 ) . Recent ADAC evidence based on more than 14,000 stations shows that this pattern remained very pronounced in 2025, with an average Super E10 spread of 12.5 cents per liter between the most expensive and cheapest moments of the day ( 1 ) . A January 2026 ADAC release still reports an evening-morning spread of roughly 13 cents, so the pattern does not appear to have vanished in more recent data ( 3 ) . Given this setting, the timing of the price reset in the new proposed policy matters for the overall price level. Stations will still want to have high prices in the morning to extract surplus from commuters and firms, while wanting to compete more fiercely in the evening on price in order to attract price-sensitive consumers. Hence, if the price reset were allowed only at the time when prices peak without restrictions, it might change price levels relatively little. On the other hand, allowing a price increase only in the early afternoon—when prices were decreasing absent constraints—would change the pricing among petrol stations considerably. For the regulator, this induces an optimization problem to impose constraints under which petrol station equilibrium behavior yields the lowest price level. In order to determine which price reset-hour is optimal in reducing the price level throughout the day, I use a simple spatial-competition model to infer the share of price-sensitive consumers \lambda_{t} for every hour t of the day from observed hourly prices during weekdays. Then, given the inferred sequence of the price-sensitivity share, I use the spatial-equilibrium price paths—constrained by the fact that prices after the reset cannot rise for 24 hours—to determine which reset-hour would lead to the lowest prices across the day. Because petrol sale quantity data are not publicly available, I weight the different hours of the day by a proxy, the weekday passenger-car traffic frequencies from BASt (German Federal Highway and Transport Research Institute) federal-road counting stations. Under that traffic-weighted objective, the optimal reset-hour of the new policy is 11:00. This optimal reset-hour is close to the reset-hour used in the Austrian regulation, which allowed price increases only at noon. The optimal reset-hour at 11:00 induces an equilibrium price path that is constant, i.e., prices do not change during the day. This is because for the reset at 11:00, stations cannot lower prices in the evening to compete for price-sensitive consumers without being locked into a low price in the morning, hurting their profits from price-insensitive commuters. Thus, the optimal price response to an 11:00 reset is a flat price path. Other reset-hours induce non-flat price paths, i.e., they induce intertemporal price discrimination. A theoretical result shows that, in this model, any reset-hour that induces a flat equilibrium price path minimizes the weighted average price, given stations’ equilibrium best responses. The intuition is straightforward. Because marginal cost is constant throughout the day, firms’ objective is aligned with raising the weighted average price, whereas the regulator’s objective is to lower it. The regulator therefore prefers exactly those reset-hours that firms dislike most. A flat equilibrium price path means that, under the reset rule, stations cannot profit from intertemporal price discrimination. They are unable to improve on a single common price by charging relatively high prices in hours with less price-sensitive demand and lower prices in hours with more price-sensitive demand. By contrast, whenever the reset-hour-induced equilibrium path is non-flat, stations are still able to segment the day and exploit differences in demand composition across hours. That same flexibility raises the weighted average price. Hence, within the model, reset-hours that induce flat equilibrium prices deliver the lowest attainable average price level. This means that a secondary regulator objective—reducing price volatility—is perfectly consistent with lower price levels in this model framework. One limitation of this analysis is that no hourly quantity data are readily available. While high frequency price data for every minute of every day are available, we do not know exactly how many units are sold at different hours. But this information naturally matters for welfare considerations and counterfactual pricing scenarios. The hourly passenger-car traffic frequencies as a transparent proxy to weight the different hours of the day are better than no weighting, but they are not perfect. Petrol is storable, and someone who always commutes in the morning may never buy in the morning. An additional limitation is that we do not observe how “purchase-time-flexible” consumers are. Even if we had perfect quantity data, these would be based on current daily price paths. If the new policy induces considerably different price paths, then consumers might switch to buying petrol at different hours of the day given these new price paths. Since we do not observe this counterfactual, this time-of-purchase substitution is hard to quantify and model. This article is meant as a policy contribution on the specific question of when to set the “reset-hour” of the new

Read the complete text