Although the marginal cost of producing electricity may vary significantly over time, electricity retailers have historically charged simple tariffs subject to infrequent updates. However, as smart metering technologies are being rolled out, more complex tariff structures are becoming implementable. Peak-Time Rebate (PTR) pricing is one of the most popular of these new tariff structures. It consists in rewarding consumers financially if, during specific hours when wholesale prices are high, they decrease their consumption relative to a counterfactual called baseline. Because customers are likely to be better informed than their retailer about (some aspects of) their future consumption, computing this baseline raises a problem of asymmetric information. This work explicitly takes asymmetric information into account to study Peak-Time-Rebate contracts in day-ahead electricity markets. We start by highlighting a structural flaw of these mechanisms: embedded arbitrage opportunities. Consumers are allowed to buy their baseline power at a constant (state-independent) price while this power is worth more by construction. Under asymmetric information, PTR tariffs thus incentivize strategic consumers to inflate their baseline. We then show that making PTR contracts incentive compatible is equivalent to implementing a variable Critical-Peak-Pricing mechanism (vCPP), in which customers have to purchase their peak consumption at the spot price. Whenever asymmetric information is a significant concern, vCPP mechanisms should thus be preferred. The relevant economic issue then becomes to design vCPP contracts optimally in order to achieve high enrollment rates under voluntary opt-in. We argue that this problem has different solutions depending on the industry structure on the one hand, and on whether or not policy-makers choose to maintain historical cross-subsidies on the other hand. We suggest there may exist complementarities between both aspects. Indeed, if subsidies to non-switchers are not to be maintained, retailers under perfect competition will offer Real Time Pricing contracts, all consumers will enroll, and the most efficient outcome will be reached. If on the contrary subsidies are to be maintained, a monopoly retailer may be in a better position to reach the second-best outcome, due to its ability to monitor the level of public spending. Other combinations of retail industry structures and political choices regarding historical cross-subsidies face difficulties than may induce inferior outcomes.

中文翻译：

---

需求响应：智能消费者的智能市场设计

尽管发电的边际成本可能会随着时间的推移而显着变化，但电力零售商历来收取简单的电价，但不经常更新。然而，随着智能计量技术的推出，更复杂的电价结构变得可行。高峰时段回扣 (PTR) 定价是这些新关税结构中最受欢迎的一种。它包括在经济上奖励消费者，如果在批发价格高的特定时间，他们相对于称为基线的反事实减少了消费。由于客户可能比他们的零售商更了解他们未来消费的（某些方面），因此计算此基线会引发信息不对称问题。这项工作明确考虑了不对称信息来研究日前电力市场中的高峰时间回扣合同。我们首先强调这些机制的结构性缺陷：嵌入式套利机会。消费者被允许以固定（独立于国家）的价格购买他们的基准电力，而这种电力在建设中的价值更高。在信息不对称的情况下，PTR 关税会激励战略消费者夸大他们的基线。然后我们表明，使 PTR 合约激励兼容等效于实施可变临界峰值定价机制 (vCPP)，其中客户必须以现货价格购买其峰值消费。当不对称信息是一个重要问题时，vCPP 机制应该是首选。然后，相关的经济问题就变成了优化设计 vCPP 合同，以便在自愿选择加入的情况下实现高入学率。我们认为，这个问题一方面取决于行业结构，另一方面取决于政策制定者是否选择维持历史交叉补贴，有不同的解决方案。我们认为这两个方面之间可能存在互补性。事实上，如果不维持对非转换者的补贴，完全竞争下的零售商将提供实时定价合同，所有消费者都将参与进来，并且将达到最有效的结果。如果要维持相反的补贴，垄断零售商可能更有能力达到次佳结果，因为它有能力监控公共支出水平。