Many people likely remember the March 2023 failures of Silicon Valley Bank and Signature Bank, both of which had more than $100 billion in total assets. But there were actually three high profile failures that month, the third being Silvergate Bank, one with “only” $12 billion in assets.

Last week, Silvergate’s Chief Administrative Officer, Elaine Hetrick, filed a declaration with the U.S. Bankruptcy Court for the District of Delaware. In technical terms, Silvergate Capital Corp., the parent holding company of Silvergate Bank, filed for Chapter 11 bankruptcy protection.

While these filings are not usually very exciting, this one is at the center of what Nic Carter refers to as Operation Choke Point 2.0, the federal government’s effort to choke off the cryptocurrency industry from traditional financial services.

In the filing, Hetrick explicitly blames the collapse of the bank “on regulators who soured on its crypto-friendly business model.” While her claim seems plausible, and Carter is right to be upset about federal regulators pressuring banks to stop doing business with crypto companies, I see the problem a bit differently: There are too many federal regulators and they have too much discretion to influence how banks operate.

They have a ton of discretion, for instance, to judge whether a bank has engaged in “unsafe or unsound practices,” as well as whether a bank’s activities pose a “reputational risk.” In fact, in 2003 the Federal Deposit Insurance Corporation issued guidance warning banks of “dealing directly with payday lenders and even with third-party firms that deal with payday lenders” because such banks “face increased reputation risks when they enter into certain arrangements with payday lenders.”

In 2013, this discretionary pressure gained widespread notoriety through something called Operation Chokepoint.

After numerous public reports, several members of Congress expressed concern over the FDIC working with the Department of Justice to pressure banks into denying accounts to customers. After an investigation, the Inspector General absolved the FDIC of any major wrongdoing in Operation Chokepoint, and its report underscores just how much discretion the FDIC has in regulatory matters.

To be clear: I do not think regulators should have this much discretion, and I do believe Congress should pare it back. So, while nothing in Hetrick’s filing really surprises me, I do think her declaration—and what it implies—is worth delving into.

First, it is easy to dehumanize these agencies, but that would be a grave mistake. It is incorrect to say, for example, the FDIC wrongfully pressured a bank to stay away from crypto firms. Similarly, it is wrong to say that the Fed prevented a bank from expanding its business. None of these actions can occur unless someone at one of these agencies makes a conscious decision to do something.

And it is dangerous to give someone this kind of power over others. It can allow, for example, someone to harm thousands of people working in the small dollar lending industry, people who are helping millions of other people engage in mutually beneficial exchanges, all on a whim, based on nothing more than his or her own subjective views.

Such an abuse of power is why the U.S. founders set up a limited government, and it is one of the main reasons federal agencies should not have so much discretion.

Still, even in the absence of abuse of power, this kind of broad discretionary authority allows just one individual to hold sway over the economic decisions of everyone else. That kind of system will undoubtedly result in a more unfair and fragile economic system than would otherwise exist. And this fact holds even when Congress sets up the discretionary framework in the name of improving banks’ safety and soundness.

A market system lets the masses “vote” on what they think is best. There will surely be mistakes, but those would occur in both systems. A market system has an advantage because it lets people move beyond those mistakes more easily. Over time, markets can become more resilient as most people learn from those mistakes and stay away from the riskiest decisions.

The idea that allowing members of Congress or the regulators—the so-called experts—to determine the best way for people to invest and make loans might sound reasonable, but history has proven it does not work. And that should hardly be a surprise because nobody is infallible.

The same goes for the so-called “systemic risk” regulations implemented after the 2008 financial crisis. People should not expect them to work as advertised. And, as Hetrick’s filing shows, the same goes for the idea that government backing ensures safety in financial markets—even though Silvergate’s customers had FDIC deposit insurance, there was still a massive run on the bank, something that isn’t supposed to happen with a federal guarantee in place.

Still, in the wake of the 2023 bank failures, many people used such reasoning—government backing will stop panics—to argue for expanding FDIC coverage. Aside from the indisputable fact that FDIC coverage has not stopped bank runs even by insured customers, it is entirely plausible that FDIC insurance, which historically has had caps much higher than needed to protect the typical American, has prevented private insurance solutions from arising, leaving markets more fragile.

Similarly, it is also plausible that excessive regulatory discretion has kept financial markets less competitive, and less resilient, than they would be otherwise. But Congress created this system, and members of Congress are the only ones who can change it.

In the meantime, there is simply no reason to expect that someone else at a regulatory agency, at some point in the future, won’t decide to go after some other industry. And that’s why Congress should change the system, paring back regulators’ discretion.