Though implementation was faulty in this case, there is a strong economic rationale for subsidy removal that goes well beyond their high budgetary cost. Subsidies distort prices and hence, too much of the subsidized good is used by consumers and producers because it is relatively cheap. Despite being perceived as equalizing, subsidies actually accrue largely to the wealthy, who spend a greater absolute amount on most goods. In addition, there is often leakage through smuggling to neighboring nations. Fuel subsidies are especially egregious because they encourage excessive fuel consumption, which is bad for the environment and distorts the critical energy market. It also pushes resources into production of capital-intensive goods, a major distortion in labor-rich countries.
The consensus view is that they should be replaced with targeted cash transfers. But, somewhat surprisingly, people receiving transfers often prefer subsidies, or at least they say they do. One reason is to insure against price volatility: subsidies do more to smooth things in difficult times if they are ad valorem or in the form of price controls because they expand when prices rise. A second reason is that they may not believe that alternative benefits will be forthcoming.
An important additional component is loss aversion. It is well known that people dislike losses more than they like similarly sized gains. I was a subject in one of the early experiments in the now vast literature. Some of my classmates were given mugs, while others were not. Sometime later, after a host of irrelevant exercises, those with mugs were offered to exchange them for $2. Those of us without mugs were given the choice of receiving the mugs or $2. The result: mug owners were reluctant to sell them, while most with the option chose the money (I took the money). In sum, ownership made people place greater value on the mugs.
Another example is from a factory in China where a group of workers were told they had received a bonus that would come in their next paycheck, but that they would lose it if their work was below average. In contrast, another group was told they would receive a bonus in their next paycheck if they produced above average productivity. In sum, both groups were offered the same amount to have high productivity, but it was framed in terms of losses for one group and gains for another. Interestingly, members of the punishment group were far more likely to have productivity above average. Fear of losses stimulated harder work than promise of gains.
Over and over again, experiments show that people dislike losses far more than similar-size gains.
Such aversion to losses makes reform especially difficult. The problem is that the removal of subsidies—which in many cases have been internalized over decades—is associated with a much bigger loss than the gain from a similarly sized cash transfer. This appears to be true even though the cash transfer is far more versatile, as it can be used for the previously subsidized commodity, other goods, saved or invested.
If loss aversion is in fact the problem there is a clear way to move forward. To limit losses it is optimal to provide the transfer before eliminating the subsidy, but with a clear name, such as “subsidy-replacement transfer.” This way the transfer gets internalized as something owned before the subsidy is removed. This also alleviates potential fears of not receiving alternate benefits when subsidies are removed.
Iran’s subsidy transformation policy did just this, putting money into people’s bank accounts before prices rose. Iranians could see the deposit but could not touch it until subsidies were actually removed. An interesting study by Djavad Salehi-Isfahani examines Iran’s Economic Transformation Program. Iran had among the largest subsidies on energy and food in the world, costing the government nearly 20 percent in GDP. Yet, the plan, which led price to rise by multiple amounts, was met with relative calm. The innovative approach of early cash transfers led people to focus on how to spend the cash transfer rather than on the loss of the subsidy. Some feared bank runs as soon as the cash became available, but that did not happen. Instead, those who had failed to register for the cash subsidy rushed to do so. Importantly, the price adjustment had the expected effects, with use of various fuels falling by up to 36% percent and income equality expanding significantly.
A key problem with subsidy removal is that loss is experienced by the whole population, making reform extraordinarily unpopular. And, the larger is the subsidy, the larger is the associated loss. In order to eliminate subsidies with less pain it is important to offset the loss with a gain before they are removed, especially to low-income people that value them the most. Early cash transfers help.