Peak concentrations of ground-level ozone pose health risks to millions of U.S. citizens across the U.S. In order to reduce peak ozone concentrations, nitrogen oxide (NOx) emissions from the power sector, among others, have been regulated with technology-based standards or, more commonly in recent years, market-based instruments such as cap-and-trade programs. However, the lack of temporal flexibility in current designs of these market-based instruments limits their cost-effectiveness on days forecasted to have the highest levels of pollution, including ozone precursors such as NOx, and as further emission reductions are sought, the marginal cost of these approaches increases dramatically. In this paper, we compare three regulatory schemes for reducing NOx emissions on high-ozone days: time-differentiated pricing, which prices NOx emissions only on days with high-ozone concentrations; undifferentiated pricing, which represents current NOx emission regulations; and technology-based standards. We develop a novel model that captures for the first time both the short- and long-term response of generators, through redispatching and control technology adoption, to a dynamic pricing scheme such as time-differentiated pricing. Unlike prior studies on time-differentiated pricing, the heart of our model, a unit commitment model, accounts for inter-temporal constraints on power generation that may be crucial to accurately capturing the response of generators to a transient price signal. We apply this model to the Texas power system and find that while control technology adoption (specifically selective catalytic reduction) does occur at very high time-differentiated prices, time-differentiated pricing mainly affects emissions and costs through redispatching of gas- for coal-fired generation. Furthermore, we show that time-differentiated pricing, due to its targeted pricing mechanism, provides a more cost-effective approach than undifferentiated pricing or technology-based standards for reducing NOx emissions on high-ozone days, but is not cost-effective at reducing summer-wide NOx emissions. Our results illustrate the trade-offs between these regulatory approaches and suggest that states should consider dynamic pricing schemes such as time-differentiated pricing for achieving further reductions in peak ozone concentrations.