On June 30, 1908, Akulina, a reindeer herder in Siberia, left her tent to greet the day. It was about 7 a.m., and all she could see for miles around her was the vast forest, standing tall against the clear blue sky. A heartbeat later, that forest vanished. A preternatural screech issued from above, and the world flushed crimson. “All around, we saw a miracle, a terrible miracle,” Akulina later recalled. An estimated 80 million trees across an 800-square-mile patch of forest—larger than the city of Houston—had been felled instantaneously.

This devastation, known as the Tunguska event, resulted from the force of a 10-to-20-megaton blast. It was so tremendous that its glow lit up the dark, moonless night sky in Northern Ireland, 3,500 miles away. And it was caused by the midair explosion of a space rock just 180 feet long.

As best as anyone can tell, if any hazardously sized asteroid were to hit Earth in the relatively near future, it would most likely be 2024 YR4, a 130-to-300-foot rock that’s essentially the same size as the asteroid that exploded over Siberia. Not long after it was discovered, just two days after it zipped right past Earth, NASA and the European Space Agency’s sky watchers calculated that 2024 YR4 has a nonzero chance of hitting Earth on December 22, 2032. Right now, the asteroid is 68 million kilometers away from the planet, but it’ll come back around. The odds of an impact in December 2032 have been fluctuating, but—based on the latest telescopic observations—they stand at just over 2 percent.

These odds will likely drop close to zero when more observations come in. But the current odds are still higher than anyone would prefer. A strike in a remote part of the world wouldn’t be a problem. But if the asteroid directly hits a city, millions could die.

However apocalyptic that might sound, the fact that we can imagine 2024 YR4’s impact is by design: Space agencies—particularly NASA—have become rather good at spotting asteroids. And two revolutionary observatories coming online in the next few years will only continue to strengthen those skills, so that the planet will have an even better shot at blocking any real threats. That people are effectively taking bets on Earth’s chance of being hit is a reason to be genuinely optimistic about the future of the world: Asteroid strikes are a rare but very real type of natural disaster, and we have never been safer from them.

Right now, NASA’s Near-Earth Object Observations Program funds several observatories whose only directive is to spot and track near-Earth asteroids and comets, just in case one of them may be bound for Earth. 2024 YR4, a relatively small asteroid by space standards, was spotted by one of those groups, the Asteroid Terrestrial-impact Last Alert System, which has found almost 1,200 near-Earth asteroids to date. Another group of telescopes, the Catalina Sky Survey, in Arizona, has found 16,500 near-Earth asteroids since 1995. In total, the program’s observatories have collectively found close to 40,000.

Earth can’t defend itself against an asteroid if no one sees it coming. 2024 YR4 was spotted eight years in advance of its potential impact, which means that space agencies can decide to do something about it—whether that’s trying to ram into it with an uncrewed spacecraft, using a nuclear weapon to deflect or vaporize it, or evacuating the future impact site. The best chance to act may be in 2028, the asteroid’s next Earth flyby. That doesn’t leave a lot of time to plan an anti-asteroid defense mission, but the planet is certainly in a better position than it would be not knowing 2024 YR4 existed.

Asteroid spotters were still somewhat fortunate to have noticed it though. Near-Earth objects are found when they reflect starlight, and huge asteroids, the sort that could end civilization, are essentially giant spherical mirrors floating about in space—very easy to spot. As they get smaller and smaller, though, asteroids look like specks of light. An asteroid the size of 2024 YR4 is stealthy enough that astronomers may not have seen it until 2028, or even just prior to its possible impact in 2032.

This will soon be an uncommon problem. The U.S. has invested heavily in two next-generation observatories that will be able to spot alarming asteroids with ruthless efficiency. One of them is the Vera C. Rubin Observatory, a nearly complete facility atop a Chilean mountain, which was funded by both the U.S. National Science Foundation and the Department of Energy. Rubin has an ambitious goal: to document everything that shimmers, explodes, or zips by in the night sky. It is a multipurpose, polymathic telescope that will look for exploding stars and distant galaxies. But it will also find an abundance of asteroids. After the Italian priest and astronomer Giuseppe Piazzi first discovered an asteroid in 1801, astronomers needed two centuries to tally 1 million of them. Thanks to an extremely wide-angle lens and a colossal nest of light-collecting mirrors, Rubin will likely double this number just a few months after it starts its survey—and plenty of the asteroids it finds may be those that linger disquietingly close to Earth’s orbit.

But Rubin, for all its strengths, will still rely on reflected starlight, which can be deceptive. If an asteroid has a dusty coating, it reflects less light than one with a shiny shell. That means that a tiny, shiny asteroid looks the same as a large, dusty space rock—and astronomers cannot tell how large it is. (This is the case with 2024 YR4, hence the size range given.)

The Near-Earth Object Surveyor, funded by NASA’s Planetary Defense Coordination Office, works differently. After surviving a painful gantlet of funding cuts, the observatory is scheduled to launch into space before the decade’s end and make its way to a lonely spot far from Earth. Because of its sunshade and its very dark paint job, it will be an extremely cold object—which will allow its infrared, heat-seeking eyes to operate with unparalleled precision. And when viewed in infrared, a large asteroid glows brighter than a smaller one, with no exceptions.

NEO Surveyor will also be unimpeded by Earth’s obfuscating atmosphere and will even be able to spot small asteroids hidden by the bright, thermonuclear glare of the sun, where several near-Earth objects are thought to be hiding. As the name suggests, this observatory will be solely dedicated to looking for near-Earth objects, and within a decade of its operations, it should find at least 90 percent of near-Earth asteroids that are at least 460 feet long.

Such asteroids are called “city killers” because if one hits a city, it’s virtually guaranteed to destroy it. There are estimated to be 25,000 city-killer-size asteroids in near-Earth orbits, and just under half have been found. 2024 YR4–size asteroids are far more plentiful. There are 230,000 of them in near-Earth orbits, and only about 7 percent of them have been found. Some of them will have a greater than 2 percent chance of hitting us, but humanity has never been less reliant on luck to dodge this kind of space-borne catastrophe.

Achieving the next level of certainty, though, requires both Rubin and NEO Surveyor to go forward as planned. Planetary defense is an international-security issue, and NASA’s partners, particularly the European Space Agency and the Japan Aerospace Exploration Agency, are developing their own asteroid-studying space missions and expanding their Earth-based telescopic coverage. The U.S., though, is clearly leading the way. Anti-asteroid research efforts are celebrated by the American public, and have garnered support in Congress from both Democrats and Republicans for decades.

Still, the second Trump administration has not yet outlined its space priorities, and deep funding cuts are expected for a variety of scientific programs. NASA declined to comment on the future of planetary-defense programs; a spokesperson noted in an email that they are “looking forward to hearing more about the Trump Administration’s plans for the agency.” And planetary defense could remain a priority: Elon Musk, who has so far been central to Trump’s drive to shrink the federal government, has well-documented interests in space and existential risk (albeit with more of a focus on getting humans to Mars than defending Earth). But if the cuts at the level experts fear go through, “we would face severe program disruptions at NASA, even for widely supported activities like planetary defense,” Casey Dreier, the chief of space policy at The Planetary Society, told me.

The odds of this impact happening are arguably higher than 2024 YR4’s chances of hitting Earth, but Patrick Michel, the principal investigator of Europe’s asteroid-chasing Hera mission, told me that, “at least for now, NASA keeps very active in planetary defense. And I don’t have any indication that it would change.” Even so, he notes, developing more redundancy in technology that can spot, examine, or deflect asteroids would keep Earth safer in the long run. 2024 YR4 will probably turn out to be harmless. But if it is heading Earth’s way—or the next asteroid is—the world will look to America to prevent a potentially catastrophic impact.

The most important intervention in the United States labor market is not unionization or the minimum wage. It is professional licensing—government-required permission to work in a particular profession, earned after significant education and testing—that covers twice as many workers as unionization and federal wage laws combined. And the system that oversees it is broken.

Researchers have known for decades that professional licensing is a bad deal for consumers and workers. High-profile critiques of licensing go back to at least 1945, when Milton Friedman’s Ph.D. thesis presented some of the earliest evidence that licensing costs consumers dearly. In the decades since, economists and journalists have developed a body of evidence supporting these critics’ views. The idea that licensing raises barriers to professions that are far higher than necessary to protect the public has remained a focus of “libertarian” and “liberaltarian” causes alike, giving rise to a bipartisan reform movement that aimed at reducing barriers to work for people with criminal records, lowering the price for health care, and making starting a new business easier.

But despite these efforts—and despite the clarity of the problem—very little has been done to meaningfully roll back licensing. In fact, the institution of professional licensing has only grown in its reach and outlandishness. More and more new professions are becoming licensed, such as art therapists and, most recently and most absurdly, fortune tellers.

[Jerusalem Demsas: Permission-slip culture is hurting America]

Reform efforts haven’t worked because none of them addresses the center of the problem: the regulatory boards that control professional licensing. When a state makes a licensing law—a rule that only practitioners who have jumped through certain hoops can practice—it usually also creates a board to interpret and implement the law. Each state has dozens of these boards; almost 1,800 have been established nationwide. They are powerful engines of professional regulation, deciding who is in and who is out, setting the terms of what you can do as a provider and, ostensibly, disciplining professionals for misbehavior.

Importantly, most statutes require that most board seats go to part-time volunteers working in the very profession they are supposed to regulate. The seats on these boards can be hard to fill, because serving can be a big time commitment and offers no pay; often, only those already involved in advocacy through professional associations are willing to sign up.

For anyone interested in licensing reform, ignoring boards is akin to someone interested in criminal-justice reform ignoring the role of courts and judges. And in this case, the boards have all the wrong incentives for public protection. Licensing works to protect consumers only if it doesn’t go too far. If getting into a profession is too hard, or the rules are too strict about what professionals can and can’t do, professional service will be expensive and scarce. But for those already licensed, more is more. The harder that entering and practicing are, the less competition those professionals face, which can mean better pay, a better lifestyle, and more prestige.

As an antitrust professor who has studied how companies act when they have control over who competes with them and how, I had a guess about how boards stacked with advocates for their profession would behave when given control over licensing. They would act like a cartel—keeping competition down and profits high. I thought board members would struggle to “change hats” from professional to regulator. When I decided to write a book about professional licensing, I started attending licensing-board meetings in my home state to see whether I was right.

[Read: The onerous, arbitrary, unaccountable world of occupational licensing]

Some of what I saw confirmed this hypothesis. For example, I watched the Tennessee Board of Alcohol and Drug Abuse Counselors nix a proposed reform that would relax a requirement that applicants need to have majored in a behavior-health field, a rule that all but foreclosed the profession to anyone who hadn’t decided to be a counselor when they were a teenager. The reasons they gave had nothing to do with the sort of public protection licensing ostensibly serves. Rather, they wanted “to protect what we’ve got.” Another said anything less than a “fully robust” license would mean less pay and prestige for counselors. One put it simply: “It’s our responsibility to make sure we are looking out for this profession.”

That has it exactly backwards. A licensing board’s responsibility is to look out for the public and to implement decisions made by legislatures, including efforts to dial down licensing. But time and again, I saw licensing reform initiated by a state government die in these board meetings. For example, in 2019, Arizona passed a “universal recognition” law that purportedly allowed anyone licensed in any state in most professions to practice in Arizona. The law was touted as promoting interstate mobility and cutting the red tape of a state-by-state licensing system. But a member of Arizona’s psychology board later told me her board had functionally killed it, at least with respect to psychologists, by interpreting it narrowly. Similarly, in Tennessee, the legislature responded to the crisis of too few physicians by streamlining the licensure process for applicants from foreign medical schools. The licensing board flat-out refused to implement it.

Boards not only resist efforts to reduce licensing barriers; they actively work to increase them. They do this by lobbying the very legislatures that are supposed to oversee them, even using their licensing fees to fund their efforts. In these efforts, licensing restrictions are often portrayed as a win-win for the profession that lobbies for them and for the consumers who get more public protection.

But this ratchet isn’t always good for consumers, because professional services can become scarce and expensive as a result. Returning to the example of alcohol- and drug-abuse counselors, one used to need 1,500 hours of practical experience to be a counselor; then that doubled to 3,000. Today one needs 6,000 hours—as much as a medical residency—to qualify for a license in Tennessee. That and the college-major requirement have made this an exceedingly difficult profession to enter. Only about 400 counselors practice in Tennessee, a state where about 70,000 people deal with opioid addiction.

Some of what I saw, however, seemed to refute my theory that what amounted to industry self-regulation at the licensing boards would work to keep down the number of professionals.

An example: The medical-board disciplinary case of an ob-gyn who had lost his license the year before. He had had sexual relationships with a number of his patients. He had written some of them (and others) off-book prescriptions for controlled substances, in at least one case prescribing a quantity so large that he later said he had come to believe the patient was selling it on the street. He also admitted to occasionally having done drugs at work with his patients. Only six months after his license was revoked, he asked the board for it back, as a changed man with a new commitment to be a better physician. The board voted to grant him a new license the day of his hearing, a fresh start for a physician who sees mostly Medicaid patients in inner-city Memphis. Much of the board discussion focused on whether a chaperone requirement could be imposed on the newly relicensed physician without raising an alarm among his patients.

[Read: The disappearing right to earn a living]

My theory that boards would keep out unwanted competitors could not explain why the board didn’t bar this doctor from the profession for good. There seemed to be more to the story of self-regulation than cartel-like behavior. When it came to dealing out disciplinary measures, board members’ professional identity and years of advocacy created blind spots where they could not see the worst of their profession. Their professional associations, too, encouraged board members to give their peers the benefit of every doubt and to believe a fellow professional who promised to do better next time. This generosity of spirit was particularly notable in the healing professions, where doctors and nurses were dispositionally inclined to see practically every provider before them as capable of redemption. The effects of this dynamic have been devastating: For example, these impulses contributed to the opioid crisis, as prescribing practices went unchecked by professional licensing boards until too late.

The diagnosis is old: Professional licensing needs to be rolled back, to be used only where necessary to protect the public and where lighter regulatory touches—that don’t so severely impact consumers and workers—aren’t effective. And where we need professional licensing, such as in many health-care professions and in law, a lighter regulatory touch will keep professional services affordable and accessible.

But the prescription is new: States need to overhaul their licensing-board systems to eliminate the self-regulation that has made licensing a lose-lose for workers and consumers alike.