Border security is at the forefront of the American political psyche. With the on-going crisis at the southern border and the endless efforts by the Trump administration to address it, it’s a recurring subject of heated debate. However, a fundamental question is often overlooked: What does a secure border even look like?

This is a much harder question than it appears to be. To answer it, numerous metrics have been implemented through the years to measure illegal activity at the border. Unfortunately, while some of these metrics provide critical data on immigration enforcement, others had to be discarded or modified for failing to provide reliable information.

Accordingly, there are still major knowledge gaps regarding border security. This is because gathering such information is difficult.

The border is remarkably vast and diverse: 1,933 miles long, with varying climates and terrains, stretching from the Pacific coast to the estuary of the Rio Grande River. This makes monitoring, enforcing laws, and measuring activity extremely demanding.

The clandestine nature of illegal immigration is another obstacle. While the increased use and reliability of technology have made monitoring the border more efficient, it is still far from perfect. The lack of assets in remote areas and sophisticated evasion techniques allow many foreign nationals to enter the country undetected.

Given that it is currently impossible to physically detect all illegal border crossings, other methods are used to gauge border security. Conducting surveys and using statistical models are two of them. While both are useful tools and provide great insight, neither is immune to defects. Their primary flaw is that they fail to incorporate major groups of the border crossing community by focusing mainly on adult Mexican nationals — who now make up a decreasing percentage of illegal border crossers.

The failure to capture the true composition of border crossers is actually a feature of the whole border security metrics’ framework. This is especially true with the absence of asylum claims — the leading cause of the current crisis — in their measurements.

Asylum is intended to provide safe haven for those fleeing persecution abroad. However, the majority of asylum claims are now ultimately denied, as loopholes in the system have attracted people who want to come to the United States for family reunification or employment opportunities, not safety. With the backlog in immigration court, the full adjudication of an asylum claim can take years, allowing many to reside in the United States and eventually get work authorization.

Put simply, the system is being abused as people are encouraged to make frivolous asylum claims. Experts understand this but have not yet incorporated asylum measurements into the border security metrics’ framework. Without accounting for asylum when addressing border security, policymakers are working with limited information.

Besides that, a fixation on apprehension data has also clouded border security as a whole. Don’t be mistaken, having the ability to apprehend border crossers is extremely important for border security. However, it is not the be-all and end-all. Turning back those attempting unlawful entry and deterring them from attempting reentry prevents immigration enforcement agencies from bearing the responsibility of housing, moving, and caring for crossers. While there are metrics in place intended to measure these key aspects of border security, all of them have deficiencies that need to be overcome.

Even if there was perfect data, however, directly crediting improved border security figures to increased performance of immigration enforcement is problematic since factors outside the control of the American government — such as the weather and the economic health of neighboring countries — may impact migration.

Highlighting the limitations of border security metrics does not mean that they are obsolete. To the contrary. Without the information they provide, there would be nothing to guide strategic planning or decision-making on how to secure the border. Pointing out their flaws is not intended to dismiss their importance, but to demonstrate that they must be improved. Luckily, several things can be done to do just that.

More frequent reporting of certain metrics would aid in addressing the shifting realities on the ground effectively and expediently and provide more oversight of the performance of immigration enforcement agencies.

Expanding the scale of surveys and statistical models beyond adult Mexican nationals and incorporating asylum into the network of border security metrics would provide a more accurate assessment of the true composition of the border crossing community.

Introducing red-teaming — the practice of recruiting individuals to attempt crossing the border to establish the likelihood that they will be detected, or remain undetected — can highlight deficiencies in border security and the reliability of measurement capabilities.

As a sovereign nation, the United States has the right and the duty to know who resides within its borders. Without reliable and consistent quantitative data on border activity and immigration enforcement capabilities, policies aimed at preserving American sovereignty and national security will be made arbitrarily.