“The less there is to justify a traditional custom, the harder it is to get rid of it,” remarked Mark Twain. Subsidies in the GCC are a tradition that is particularly difficult to justify, and Twain’s maxim has likely underlain some of the challenges faced by the Gulf governments as they have tried to scale back subsidies. Biases in the way that humans think present additional obstacles to reform.

Economists have been advising the GCC countries to dispense with subsidies long before the 2014 oil price crash, and the associated budgetary pressure. Subsidies were originally introduced as a way of raising the living standards of lower-income households, by lowering the price of some of the key commodities that they purchase.

Today, subsidies have come to resemble acquiring a Ferrari to travel from your living room to your kitchen: extremely expensive, and highly ineffective.

To see why, note that an effective poverty-alleviating subsidy should lower the prices of commodities that are consumed by low-income households, while leaving the commodities that are consumed by high-income households unchanged. A good example is public transport—a service used heavily by the poor and rarely by the rich.

GCC subsidies satisfy the first condition, but they spectacularly fail to satisfy the second one: goods such as electricity, water, fuel, and flour are consumed by low-income households, but high-income households consume them in much larger quantities, with their large homes, powerful cars, and extravagant parties.

Thus, compared to a system of direct income support for low-income households, or to an updated selection of commodities to subsidize, the prevailing GCC subsidy system redistributes income from the poor to the rich, rendering it fundamentally counterproductive.

In fairness to their creators, the GCC subsidy systems were a good choice at the time of conception: most households lived modestly, and the tools required to efficiently administer means-tested income support, such as electronic databases and detailed censuses, were absent. The 21st century GCC economy, however, is an unrecognizable descendent of its 20th century progenitor.

Setting aside their effectiveness, subsidies are also highly expensive because they engender wasteful consumption; the GCC countries are some of the highest per capita energy consumers in the world, and this is mostly the result of artificially low energy prices. A negative corollary has been needless environmental damage.

However, even if the GCC governments craft compensatory support schemes for low-income households, they will still struggle to overcome a behavioral bias known as “loss-aversion”.

As a precursor to a famous experiment, when people are given the chance to select between a free mug and a free chocolate bar, they pick the mug 50% of the time, and the chocolate bar 50% of the time, confirming that the two goods have a similar value. Nothing strange so far.

Human behavior becomes anomalous when, instead of giving people a choice between the mug and chocolate bar, the researcher randomly allocates each individual one of the two goods based on the outcome of a coin flip. After receiving the mug or chocolate bar, and holding it for a couple of minutes, the researcher then offers the individual the opportunity to trade the good that they were assigned for the other one at no cost. What happens?

In principle, they should be willing to trade 50% of the time, because the coin ensures that people end up with the good that they prefer half of the time. In fact, people end up trading less than 30% of the time, whichever good they receive.

This anomaly, also known as the “endowment effect,” is the result of people developing an irrational attachment to objects in their possession. That explains why most people refuse to sell their old pen for $1, even though they can buy a new one for $0.10. It also explains why humans vociferously oppose threats to eliminate existing public parks, yet exhibit scant support for proposals to create new ones.

More generally, loss aversion poses a problem for policymakers who want to replace an existing, inefficient benefit with a new, efficient one, because for most people, the pain of losing something that they already have is stronger than the joy of receiving something that they did not previously have, even if the two benefits are objectively equal.

In the GCC, governments are now convinced of the ineffectiveness of subsidies, and are considering various alternatives to compensate citizens for their removal, such as direct, means-tested transfers or subsidy cards. Despite the moral and economic wisdom of such actions, low-income households have reacted negatively, including hand-wringing on social media, and calls for boycotts of previously subsidized commodities.

The opposition is not merely an artifact of loss-aversion, as the subsidy reforms coincide with a broader decline in the economic capabilities of the governments, caused by declining oil prices. Thus, for example, while GCC citizens were previously virtually guaranteed public sector jobs with generous salaries, all six governments have described such luxuries as no longer tenable. Yet, humans’ irrational attachment to what is already tangibly in their possession undoubtedly has a role to play in the current subsidy-reform backlash.

What can the GCC governments do to circumvent loss-aversion? Research has yielded practical advice on how to diminish the incidence of loss-aversion, most notably giving people a lot of experience in owning and surrendering the good in question, but such strategies are of no use to policymakers considering infrequent reforms to policies that have been in place for decades.

Whatever the impediments to subsidy reform, the need for their implementation constitutes a rare congruence of views among economists, environmentalists, and ethicists, and so the GCC governments must persevere. “I am a slow walker,” remarked Abraham Lincoln, “but I never walk back.”