THE “cash for clunkers” program, introduced during the depths of the downturn, was supposed to be a policy two-for-one. By paying people to trade in their old fuel-guzzling cars for new efficient ones, the scheme was supposed to jump start consumption of durable goods, boost the hard-hit car industry, and improve the fuel efficiency of America’s fleet. The eight-week programme generated an immediate spike in car sales, with almost $3 billion spent on the popular rebates. However, a new paper suggests that the programme may have actually decreased total spending on new motor vehicles creating a net drag on the economy.

The study, by Mark Hoekstra, Steven Puller and Jeremy West from Texas A&M University, examined the difference between consumers who were on the border-line of being eligible for the scheme. To qualify as a ‘clunker’, a vehicle needed to achieve less than 18 miles per gallon. Thus by comparing households whose cars were just under this threshold, with those who were just over it, the authors were able to measure the impact of the scheme on consumption patterns. Unsurprisingly, they found that the scheme dramatically increased the car purchases for eligible households over the 8 weeks. But this was largely due to households changing the timing of their purchase. Instead of stimulating new car sales, the spike was mostly attributable to sales that would otherwise have occurred over the subsequent 11 months. The authors show that after one year the proportion of households that had bought a new car was the same, regardless of whether they qualified for the subsidy.

In addition to altering the timing of purchases the programme also changed what type of cars people bought. Since the size of the subsidy increased with the efficiency of the newly purchased car, consumers had a strong incentive to buy cars thatr guzzled less fuel. While some consumers bought expensive hybrids, most chose to increase efficiency by buying smaller, and thus cheaper, cars. In order to meet the fuel efficiency requirements consumers ending up spending on average $4,600 less per vehicle compared with what they otherwise would have spent without the scheme.

Since the scheme only induced a change in the timing of purchases, not an increase in the aggregate number, and caused households to substitute towards cheaper cars, the gross impact was to decrease total spending on new vehicles. The authors estimate that the scheme decreased total industry revenues by around $3 billion over the subsequent year. In addition, estimates of the environmental benefits from the scheme suggest that they were small and relatively expensive. In trying to hit two birds with one stone, it seems, the “cash for clunkers” managed to miss both.

The programme was not the only underwhelming policy response to the recession. Most policymakers dramatically underestimated the size of the ensuing recession and as a result the majority of policy responses were not nearly ambitious enough. It is however a cautionary tale about the dangers of trying to do too much with one policy lever. Tax credits during a downturn and incentives for more fuel efficient cars might be good ideas in their own-right, but mixing them together is a recipe for both policies stalling.