2.  It is often asserted that consumers purchasing automobiles and other durable goods under-weigh gasoline or other future add-on costs.  When gas prices rise, demand for high fuel economy vehicles increases, pushing up their relative prices.

3.  There is a growing body of evidence that consumers choosing between products may underweight, relative to purchase prices, product costs that are less salient or accrue in the future.  Consumers on eBay, for example, are less elastic to shipping and handling charges than to the listed purchase price (Hossain and Morgan 2006).  Mutual fund investors appear to be less attentive to ongoing management fees than to upfront payments (Barber, Odean, and Zheng 2005).  Shoppers are less elastic to sales taxes than to prices (Chetty, Looney, and Kroft 2009). Consumers’ tradeoffs between the purchase price and future energy costs of air conditioners imply relatively high discount rates (Hausman 1979).

4.  It is also often asserted that vehicles’ gasoline costs are not relevant to automobile consumers at the time of purchase  In 2007, the median-income American family spent on average $2400 on gasoline, and American households spent $286 billion in total (U.S. Bureau of Labor Statistics 2007).  Jaffe and Stavins (1994) call the "Energy Paradox“ this; that consumers and firms have been remarkably slow to adopt to apparently high-return energy efficient technologies.

5.  Hypothesis: An increase in gasoline prices over the past decade should increase the relative prices and market shares of high- vs. low-fuel economy vehicles.  They used a model with datasets on vehicle prices, quantities, and attributes between 1999 and the present, which includes nearly 55 million vehicle transactions at both auto dealerships and auctions.  They then used the monthly average prices for all vehicles to observe the national market shares of each of these vehicles and match these to the price data, fuel economy, and other variables to complete a 25,000-household survey.  They conservatively estimated that between 1999 and March 2008, American auto consumers were willing to pay only thirty-seven cents to reduce expected discounted gas expenditures by $1.

6.  The demand function is: q = α – η p – γ G  q is the vehicle’s quantity  α is an intercept  p is the vehicle’s purchase price  G is the discounted gasoline costs over the vehicle’s lifetime, which is assumed to be the same for all consumers.  The variable G will depend on a discount rate, future gas prices, fuel economy and the vehicle’s usage and scrappage probability over time

7. u ijat = η (w - p jat ) - γ G jat + Ψ jat + ε ijat  In this equation, w is the consumer’s wealth, p jat is the purchase price, and G jat is the discounted In reality, the consumer’s true problem is dynamic: at any point in time, she has the opportunity to re-sell the vehicle and purchase a new one.  If consumers value purchase price and future gas costs equally, then η = γ. If they undervalue future gas costs relative to purchase price, then γ < η  G jat depends on the discount rate, expected future gasoline prices, and expected usage of the vehicle. Note that G jat will be constructed using expectations, and we assume that consumers are risk neutral.  The variable Ψ jat is the present discounted value of the flow utility that vehicle ja will provide to the average consumer over the rest of its lifetime from year t forward.  Ε ijat is unobserved taste shocks from the changes in prices of substitutes They try to eliminate biases by adding a bunch of variables which is confusing to follow!

8.  Manheim, the largest automobile auctioneer in the United States  Power Information Network, a network of dealerships managed by JD Power and Associates.  National Vehicle Population Profile  U.S. Environmental Protection Agency (fuel economy only since 1974)  National Household Travel Survey for 2001  U.S. City Average Motor Gasoline Retail Prices from the US Energy Information Administration.  And another 15ish sources!

9.  r = ß -1 -1  If the vehicle is purchased with cash, the discount rate is the opportunity cost of expected rate of return for alternative investments. If the vehicle is financed with a loan, it’s the annual interest rate of the loan  People discount expected future vehicle-miles traveled, survival probability, discount rates, and expected gasoline costs  So…. People undervalue mostly everything in the future. We think we won’t put that many miles on a car, we expect that gas prices won’t increase that much in the future, etc. etc.

10.  Auto makers produce fewer low fuel economy vehicles when gas price expectations are higher  A one percent increase in the market share of a vehicle causes a $18-19 drop in its price  Even with a short time horizon, consumers appear to account for only 59% of gas costs. (AKA they still undervalue in the SR)  The private costs of gasoline are ten times the carbon costs, so any misoptimization over future fuel costs reduces the misoptimizing consumer’s welfare more than it damages the climate by an order of magnitude.  Perhaps the most important takeaway is that behavioral misoptimization can be a more powerful justification for fuel economy policies than internalizing environmental externalities.  In a theoretical sense, if consumers did not misvalue future gasoline costs, there is no economic argument for adding a fuel economy standard to the optimal gasoline tax.  This simple analysis suggests that understanding consumer choice is much more important.

11.  This paper tests whether vehicle prices and market shares respond to changes in gasoline price expectations in a way that is consistent with rational, perfectly informed, and forward-looking consumers who value $1 in purchase price the same as they value future fuel costs with present discount value of $1.  The price and quantity changes strongly suggest that the market substantially under-adjusts to changes in future gas costs: depending on the specification, vehicle prices and market shares move as if consumers are willing to pay only $0.37 up front to reduce expected discounted future gasoline expenditures by $1!!

12.  Consumers underweight all future gasoline costs in making decisions (not optimal decisions)  A gasoline tax increase or a carbon tax intended to offset negative externalities would have less than the efficient level of effects on consumer’s vehicle choice  Provides support for why there are fuel economy standards and the carbon cap-n-trade