UC Santa Barbara

This dissertation consists of three works that estimate the distributional effects of changes to market conditions. Each essay utilizes different techniques to identify behavioral responses from consumers and firms, which then plays a key role in the distribution of consumer and producer surplus.

In the first essay, I study how a carbon tax differentially affects the welfare of electricity producers and consumers, also known as incidence. In so doing, I develop a new framework to estimate the incidence of input taxes that accounts for incomplete pass-through to retail prices, imperfect competition, and heterogeneity in marginal costs of producers. Leveraging exogenous variation in the level of the Australian carbon tax, I then apply this framework to the context of the National Electricity Market in Australia. I find that 95 to 97 percent of the welfare cost from changes in the carbon tax falls on electricity consumers. The large burden borne by consumers is driven by full pass-through of increases in wholesale electricity prices to retail electricity prices in combination with negligible decreases in aggregate profits for electricity producers. Simulations show that incorporating heterogeneous firms is particularly important when market demand is elastic.

The second essay, joint with Guangli Zhang, asks: Do customers respond differently to tip suggestions based on whether or not the suggested tip amount is an integer and, if so, what does this reveal about human behavior? With the advent of cashless payment systems, customers are increasingly being presented convenient, low-stake tip suggestions following purchases. Despite the rising frequency of these interactions, we still know little about the preferences underlying tipping behavior. Previous research, for example, has documented that customers tend to tip integer amounts, but it is unclear if this is due to smaller cognitive costs associated with tipping an integer amount or direct utility benefits from integer tips. Utilizing plausibly exogenous variation in the occurrence of integer tip suggestion in New York City taxi rides, we find that passengers presented with an integer tip suggestion are 21 percentage points more likely to tip that amount. In our setting where the suggested tip percent is larger than the average passengers give, this leads to a 0.6 percentage point increase in tip rates and an approximately 2.38 million dollars transfer from riders to drivers as a result of a 2012 rate fare change that increased the probability of integer tip suggestions. Using a theoretical model for customer tipping behavior, we find that these results are consistent with direct utility benefits from tipping integer amounts.

In the third essay, I return to the context of the Australian carbon tax to estimate if wholesale electricity prices respond differently to the implementation of the carbon tax compared to when it is repealed. I find that when the carbon tax is implemented prices rise by more than double the amount they decrease when the carbon tax is repealed. I then utilize one of the key advantages of wholesale electricity markets -- detailed data on the offer (supply) curve that is used to set prices. I identify that that the pattern evident in prices directly reflects asymmetric changes in how plants respond to the implementation and repeal of the carbon tax. When the carbon tax is implemented, plants that could potentially set prices increase their willingness-to-accept by significantly more than they decrease them when the carbon tax is repealed. These findings provide the first direct evidence that asymmetric price changes are driven by asymmetric supply-side responses.