It’s a scene right out of the movie Groundhog Day. Every year around this time, the administration submits its annual request to Congress to appropriate billions of dollars for America’s allies and partners in the Middle East to finance their purchase of U.S. military training and equipment.

Congress rubber stamps these requests with little regard for whether this assistance achieves U.S. foreign policy objectives. It does the same when the executive branch requests congressional approval of arms sales for cold hard cash. Such docility might be good industrial policy—after all, it creates jobs in key congressional districts, provides corporate welfare for America’s defense companies, and helps maintain the defense industrial base. But it makes for lousy foreign policy. The United States will continue to pour money down a rat hole until Congress and the executive branch better understand why these problems keep recurring and muster the political will to fix them. Based on our experience in the State Department, here is our diagnosis of the problem and some remedies for what ails U.S. military assistance in the Middle East.

In the U.S. foreign policy toolkit, security assistance and arms transfers have become the instruments of choice for American diplomats and soldiers. Grant assistance and weapons sales are treated as Swiss Army Knives, all-purpose tools appropriate for use in virtually any scenario. According to the prevailing view in the U.S. government, security assistance works wonders: it builds the capabilities of partner countries, provides influence over their policies, and guarantees access to influential institutions and personalities in capitals across the globe. If true, this would seem to more than justify the $48.7 billion the U.S. has spent on security assistance to the Middle East over the past decade.

In reality, U.S. military assistance promises more than it delivers. There is scant evidence outside of a few isolated cases that U.S. material support to Middle Eastern countries has fulfilled any of these purposes. Recipients of U.S. funding and weapons have largely failed to make major strides in their capabilities and, in some instances, may have even regressed.

Despite $47 billion in U.S. military assistance over 40 years, the Egyptian military has struggled mightily to contain an ISIS-affiliate numbering no more than 1,200 militants. The Saudis barely used their American-made advanced combat aircraft in the U.S.-led anti-ISIS operation in Syria, and $89 billion in arms sales to the kingdom over the last 10 years has not prevented Riyadh from getting bogged down in an increasingly costly quagmire in Yemen with U.S.-supplied weapons. The U.S. has sold hundreds of billions of dollars in military hardware to Persian Gulf countries and yet collectively they are not capable of defending the free flow of oil from the Gulf against a militarily weaker Iran without U.S. assistance.

Likewise, the track record of using security assistance to increase U.S. influence in the region is no more encouraging. While recipient countries are happy to utter platitudes about increased cooperation, they generally—and successfully—resist Washington’s requests to modify their policies in exchange for assistance. Ongoing U.S. assistance to Egypt did not leave Cairo open to American pleas to desist from forcibly dispersing two largely non-violent sit-ins in the capital, in which over 800 people were massacred.

Meanwhile, U.S. attempts to explicitly link military assistance and arms sales to a recipient country’s domestic political behavior have not borne much fruit. For instance, the Obama administration’s suspension of some types of military assistance to Egypt in 2013 did not lead to “credible progress” toward democratic reforms. Nor did putting a $4 billion arms package for Bahrain on hold yield an improvement in that country’s human rights environment. Importantly, these failures have more to do with a lack of political will in Washington, in which the U.S. capitulated before its coercive measures could have the desired effect, than any inherent limitation in what withholding weapons shipments can accomplish. But the frequency with which the United States folds in these standoffs suggests a structural problem in U.S. assistance mechanisms that undermines its efficacy as a tool of influence.

U.S. officials have excellent access in Middle Eastern capitals, but it is hard to attribute this to military assistance and arms sales. The United States remains a predominant international player and most countries do not have the luxury of ignoring Washington for long. Pentagon officials argue that the provision of material support increases their contacts with foreign militaries, creating opportunities to learn more about partner armed forces. In practice, however, recipient countries take great precautions to limit and regulate U.S. access to their troops. As an example, most Egyptian military personnel are prohibited from interacting with U.S. officials, while a small core of vetted senior officers are entrusted with managing Egypt’s military relationship with the United States.

Under existing conditions, U.S. interests and taxpayers are not the primary beneficiaries of military assistance and arms sales. Instead, it is U.S. defense contractors and regional militaries that often prioritize domestic political influence over operational capabilities. In recent years, the U.S. arms industry has registered record profits, a pattern likely to continue given President Trump’s initiative to expedite government approval of weapons sales. Indeed, the State Department cleared a record number of arms sales in Fiscal Year 2017 ($75.9 billion). While champions of the U.S. arms industry defend it as an engine of job growth, economists have found that investments in other industries are more efficient job generators.

Meanwhile, Middle Eastern militaries have exploited arms sales to buttress their prestige and to support local patronage networks, both of which help to sustain their dominant position in domestic politics. Egypt’s procurement of over 1,000 M1A1 Abrams tanks, for instance, has less to do with their military value than with the Egyptian jobs supported by a co-production plant in country. The Egyptian Armed Forces have so far opted not to deploy M1A1s in combat in the restive Sinai Peninsula.

In his State of the Union address, President Trump endorsed “legislation to help ensure American foreign-assistance dollars always serve American interests, and only go to America’s friends.” But even when this assistance goes to America’s friends, it rarely serves American interests. Decisions to sell weapons to allies, friends, and partners often involve hand-waving and intellectual laziness. It is unusual when clear objectives with measurable benchmarks of progress are identified for weapons sales and assistance levels. Instead, proponents of these sales invoke tired bromides about how assistance will provide access to the recipient’s military leadership or further cement the bilateral relationship.

But access should not be confused with influence—and “relationship maintenance” should not be treated as an end in itself. Washington has become so fixated on doling out billions of dollars for this purpose that it often forgets what this assistance is for in the first place: securing U.S. interests. More often than not, our allies and client states take the money and use their weapons in pursuit of policies inimical to U.S. interests or kvetch about American unreliability. Saudi Arabia, which has used American-supplied weapons to visit ruin on Yemen and strengthen Jihadist groups there, is a poster child for this phenomenon. So, too, is the UAE, which is an accomplice in Riyadh’s immoral and strategically disastrous campaign in Yemen and used American-supplied weapons in Libya in support of a renegade general.

A second and related problem is that the U.S. government does a poor job of holding allies and clients to account for behavior that runs counter to American interests. There is no systematic review of what U.S. military assistance accomplishes. The key questions that rarely get asked, let alone answered, are what does the U.S. want and expect from the assistance we provide and how does this aid help or hurt America’s ability to achieve these goals?  If the U.S. cannot identify actions that the recipient would not have otherwise taken as a result of this assistance, then it is nothing more than a welfare program, and has two pernicious effects. First, it encourages “moral hazard”—recipients to do whatever they want with the assistance without having to fear the consequences of their actions. Second, it creates “reverse leverage”— Washington bends over backwards to keep relations smooth and the assistance flowing, rather than leverage the recipient’s dependence on U.S. military support and political commitments.

Both of these pathologies are, in part, a legacy of the Middle East peace process. The Arab countries at peace with Israel (Egypt and Jordan) view their U.S. military aid packages as an entitlement owed to them in exchange for burying the hatchet with the Jewish state, which supersedes U.S. concerns about military capacity or influence. What Washington gets for providing this aid is confidence that two of Israel’s neighbors will remain at peace with it. But this is a misrepresentation of history; Washington did not make a formal aid commitment to either country when they signed peace treaties with Israel. And, most importantly, the future of the peace treaties is no longer dependent on U.S. assistance. Egypt and Jordan find it in their own interests to preserve peace with Israel, with or without American support.

U.S. military assistance in the Middle East (and more broadly) is in need of serious reform. Here are four major innovations to help fix it:

  • Before allocating security assistance, recipients should first have to demonstrate their commitment to a set of norms, standards, and rules of good security-sector governance.
  • The State Department and Congress should create new mechanisms to rigorously assess, monitor, and evaluate security assistance against performance benchmarks that are linked to U.S. foreign policy objectives.
  • The sale of weapons should be linked to new training commitments that the host country would have to fulfill in advance before taking delivery of the weapons and equipment.
  • To demonstrate its commitment to performance and results, all contracts for the supply of weapons and training should include “sunset” provisions based on mutually agreed upon performance milestones.

Most in the American defense establishment would agree in principle that introducing more accountability for U.S. military assistance and arms sales is a worthy goal. But they would also express concern that the measures we recommend would prompt Middle Eastern countries to turn to Russia or China for easier terms. However, for many of our security partners, especially Saudi Arabia and Egypt, integrating Russian and Chinese weapons into their force structure would create serious operational and logistical problems. Moreover, neither Moscow nor Beijing offers grant assistance, which means they are not viable substitutes for countries that depend on U.S. financial help to buy equipment. Even for countries that use their own money, their strong preference for U.S. equipment, which they view as both superior to the alternatives and a sign of American support, suggests they would be willing to submit to more rigorous oversight if that is the price of obtaining American weapons.

U.S. military assistance in the Middle East could magnify our influence and help build local military forces that reduce the burden on U.S. forces. Unfortunately, billions in American taxpayer money have been spent for unclear purposes without sufficient oversight. By learning from past mistakes and implementing our recommended reforms, the U.S. government can begin to break this cycle of waste and missed opportunities.

Andrew Miller is a former director for Egypt on the National Security Council and current deputy director for policy at the Project on Middle East Democracy and a nonresident scholar at the Carnegie Endowment for International Peace.

Richard Sokolsky is currently a nonresident senior fellow at the Carnegie Endowment for International Peace. He served in the State Department for 37 years and was a member of the Secretary of State’s Office of Policy Planning from 2005-2015.