On Feb. 27, 1933, as Herbert Hoover struggled to combat the Great Depression that had plunged millions into poverty and caused a devastating wave of bank failures throughout his presidency, his press secretary Theodore Joslin came home after a long day and was worried about his money. “[A]lthough I felt unpatriotic in doing so, I drew out most of the money in my checking account,” Joslin wrote in his personal diary, noting he instructed his wife to do the same. “And I told the President what I had done. ‘Don’t hoard it, Ted,’ was his only comment … But I am ‘hoarding’ temporarily. No bank is really liquid today and won’t be until this panic is over.”

Of course, that was a time before deposit insurance, which is meant to preserve the funds of individuals who invest their savings with banks in the event of a failure, but the sudden collapse of Silicon Valley Bank (SVB) earlier this month brought a small taste of that era to panicked startup executives, nearly a century later.

The Federal Deposit Insurance Corporation (FDIC) was created almost as soon as Franklin Roosevelt swept Hoover out of the White House, as part of his New Deal in general and in particular, the game-changing Banking Act of 1933, also known as Glass-Steagall. Conceived as a way to restore stability to the nation’s ailing financial system, it passed a host of reforms including the establishment of the FDIC as an independent agency to insure deposits at commercial banks across the country.

But this regulation from the 1930s doesn’t quite fit the 2020s, when Bay Area startups park millions worth of company cash at the trendy bank that all their friends in venture capital also use: That’s right, Silicon Valley Bank. So when SVB went under on a Friday just a few weeks ago, the startup community panicked as it realized that over 90% of its deposits weren’t covered by FDIC insurance.

That led to fears of bank run “contagion” as tech startups warned they wouldn’t be able to make payroll and venture capitalists called for their clients to withdraw their funds from SVB and similar regional lenders. Experts warned of potential “systemic risk” for the financial system and within 48 hours, federal regulators agreed, invoking a “systemic risk exception” that allowed the FDIC, Federal Reserve and Treasury Department to backstop all deposits on an emergency basis. The three regulators noted that they conferred directly with President Joe Biden before making this dramatic change to the nearly 100-year-old concept of deposit insurance.

Now, in the wake of this banking instability, a number of politicians and business leaders believe the deposit insurance limit should be raised, including, earlier today, Senate Majority Leader Chuck Schumer. And some, including Massachusetts Sen. Elizabeth Warren, want to get rid of the $250,000 cap altogether.

But is it really a good idea for the federal government to give out an unlimited guarantee for all deposits in the banking system? Fortune talked to a range of industry experts who said the deposit insurance limit should be raised, but to infinity is another matter entirely. There might be another way, though—creating different “classes” of depositors.

A Silicon Valley Bank wake-up call

Will McDonough, CEO of EMG Advisors, a private equity firm that manages over $13 billion, told Fortune that SVB’s collapse was evidence that the $250,000 cap should either be updated or “more fluid.” He said that if regulators hadn’t stepped in to back uninsured depositors, he and many other high-net-worth individuals like him would have “gone into any bank account I had over $250,000” and started withdrawing, which harms institutions and actually spreads risk, on top of being unnecessary.

There is precedent for raising the deposit limit in the wake of financial instability. After the Savings and Loans (S&L) Crisis of the early 1980s—which has some uncanny parallels to today’s crisis—the deposit insurance limit was raised from $40,000 to $100,000, according to FDIC historians. And in 2008, after the Great Recession, the Emergency Economic Stabilization Act (EESA) raised the limit to its current $250,000.

However, both McDonough and Konrad Alt, co-founder of Klaros Group and a former Counsel to the Senate Banking Committee during the later stages S&L crisis, noted that raising the deposit limit won’t be enough to save depositors at many banks if they fail.

“Would it make a difference? Maybe a little bit, but not a lot,” Alt told Fortune. “Certainly not at a bank like Silicon Valley Bank where I would venture to guess … no matter where you draw that line, you’re gonna find that a large proportion of deposits would still have been uninsured.”

And the argument over raising the deposit insurance limit might be a moot point after the rescue of SVB, along with comments from Treasury Secretary Janet Yellen this week implying that uninsured deposits will be protected even at small banks in the event of another failure. McDonough argued that regulators’ words and actions signal that a form of quasi “unlimited deposit insurance” already exists in the U.S.

But Jan Hatzius, Goldman Sachs’ chief economist and head of Global Investment Research, said he believes regulators’ recent actions “still stop short of an explicit guarantee” for all depositors.

“In our view, an explicit guarantee of all deposits would take further intensification of bank stress, potentially involving another bank failure,” he wrote in a Wednesday research note.

Hatizus added that he believes there is a “fairly low” chance that Congress will increase the deposit insurance limit or create an unlimited deposit guarantee in the near term, but “the odds are higher” over the medium term. And like all the experts Fortune spoke with, the economist also warned that raising the deposit limit is “unlikely to address the more acute concerns currently facing financial markets.”

A potential solution

If raising the deposit insurance limit or even removing it altogether won’t fix banks’ problems, what will? Klaros’ Konrad Alt argued “differentiating more” between “different kinds of deposits” when it comes to insurance could do the trick. Alt knows whereof he speaks, as he was a part of the legislative staff that drafted the FDIC Improvement Act of 1991, which strengthened the power of the FDIC, allowed it to borrow from the Treasury, and forced failed banks to be resolved using “the least costly method available” to taxpayers.

“If you look at federal law right now, at least for deposit insurance purposes, it mostly distinguishes between retail deposits and broker deposits. There’s not a lot of other distinctions,” he noted.

Alt believes that consumer deposits should be treated differently—and given more FDIC protection—than other forms of commercial deposits from some large businesses. But he also made it clear that it’s not as easy as simply backing all consumer deposits and not backing businesses’ deposits.

The problem with SVB, and many banks these days, is that some depositors are actually fintech companies managing large FBO, or For Benefit Of, accounts for their users. FBO accounts are a type of umbrella fiduciary account that combines assets from multiple users under the control of one company.

Alt gave the example of a payments business that allows users to transfer money quickly via an app. A company like this would market its services, sign up customers, and then take their money and put it into a FBO account at a bank like SVB.

“But Silicon Valley Bank doesn’t even know who my customers are. And my customers aren’t necessarily very aware they are using Silicon Valley Bank,” he noted.

The problem with this model is it disrupts FDIC insurance, according to Alt. Users of fintech products are essentially individual depositors with small accounts at a bank, but their deposits aren’t protected because the FBO accounts that hold their funds could have millions or even billions of dollars in them, which the FDIC would subject to a $250,000 insurance limit.

This means that when a bank fails, the “effective insurance available to those depositors is likely to be near zero,” Alt said. Creating different classes of accounts, each with their own insurance status, could fix this issue. Most of these would be fully insured, others partially, and some not at all. Alt said this labeling process could enable regulators to isolate large accounts that are critical to consumers or businesses and protect them, without having to cover all of the less systemically important accounts.

“I think that that would be an interesting public policy debate to have, but it would require an act of Congress,” he said.