To hide the price shock from big banks inevitably passing the costs on to their customers, the bill would phase in higher deposit-insurance fees and increased required reserves over the course of a decade. Consequently, the FDIC’s ratio of guaranteed deposits to reserves—a critical indicator of the fund’s ability to protect taxpayers—would be dangerously distorted for 10 years.

We’ve seen this before. A similar lack of reserves prevented the shutdown of troubled savings-and-loan associations in the 1980s. Regulators lacked the resources in their insurance fund to close bankrupt S&Ls, forcing an era of “forbearance” when thrifts stayed open despite insolvency. It dramatically drove up the resolution cost of the S&L crisis from an estimated $25 billion had the problem been addressed in 1983 to an actual cost of $160 billion by the 1990s.

The newly proposed FDIC deposit hike resembles Congress’s 1980 increase of the insured deposit limit from $40,000 to $100,000. In its review of the S&L crisis, the FDIC said the increase “added substantially to the potential costs of resolving failed financial institutions” and worsened the moral hazard problem. The increase Congress is now considering would be more than 160 times the size of the 1980 hike.

The 2008 financial crisis represents another infamous example of expanding liabilities without necessary capital. After Congress authorized the Department of Housing and Urban Development in 1992 to set affordable-housing goals, Fannie Mae and Freddie Mac repeatedly lowered credit standards. The mortgage assets held or guaranteed by Fannie and Freddie then grew from roughly $1.1 trillion in 1992 to $5.2 trillion by 2008. Despite skyrocketing portfolio risk, they never raised guarantee fees. When widespread subprime mortgage defaults began in 2007, Fannie and Freddie only held roughly 1.6% in capital and sustained combined losses of over $100 billion in 2008 alone. They have remained functionally insolvent and under government conservatorship ever since.

Proponents of raising deposit insurance from $250,000 maintain that “too big to fail” banks have a competitive advantage over smaller financial institutions, where government bailouts are unlikely. But smaller banks have their own advantages. Under the Dodd-Frank Act, capital requirements for the largest banks are often 10% or more, compared with roughly 7% for smaller banks. Larger banks also face stress tests and other burdensome regulations that smaller banks don’t.

With this legislation, regional and community banks will have deposit guarantees comparable to large banks but without similar capital and regulatory burdens. Does anyone believe Democrats won’t eventually push similarly crushing regulations onto these smaller institutions? After the 2023 collapse of Silicon Valley Bank, Sen. Elizabeth Warren (D., Mass.) supported higher deposit-guarantee limits but only in return for more regulation.

The government is full of programs guaranteeing assets without imposing fees or premiums commensurate with risk. Flood insurance rebuilds the same house five times, and taxpayers are stuck with 60% of the premiums for federal crop insurance. When flooding happens or the crop fails, insufficient funds exist to pay claims. Government then passes emergency funding legislation to make up the shortfall. Federal regulatory power inevitably expands.

The same principles apply to federal deposit insurance. Increasing the level of insured deposits without imposing the actuarially sound insurance cost on the banks and businesses that benefit encourages reckless behavior. Federal and state laws would never permit any private insurer to operate as federal insurance programs regularly do.

Public insurance has become one of the most sustained sources of systemic risk in our economy. Congress shouldn’t make it worse.

Mr. Hensarling is an economics fellow at the Cato Institute. He served as chairman of the House Financial Services Committee, 2013-19. Mr. Solon is a senior fellow at the Hudson Institute

Yet even as our economy has experienced broad-based growth, real federal welfare spending has soared by 765%, more than twice as fast as total federal spending, and now costs $1.4 trillion annually. Were that money simply doled out evenly to the 19.8 million families the government defines as poor, each household would receive more than $70,000 a year.

The source of this dramatic mismatch is a fraud built into how various programs determine welfare eligibility: The government doesn’t count any refundable tax credits or benefits that aren’t paid in cash as income to the recipients.

Some claim this is appropriate because the beneficiaries aren’t free to spend noncash benefits on whatever they like. But that is a specious argument, because money is fungible. Receiving Medicaid, for example, frees up cash that would otherwise be spent on healthcare, allowing the recipients to spend the newly freed cash on other things. Noncash benefits aren’t in the end that different from income—except that salaries are taxed while government benefits aren’t. And individual welfare programs often don’t even count benefits paid in cash as income for the purpose of gauging eligibility.

The government’s failure to count its largess as recipients’ income allows welfare households to blow past the income level above which a working family no longer qualifies for government help. Take a single parent with two school-age children who earns $11,000 annually from part-time work. The government considers this household in poverty because its income is below $25,273. But this family would qualify for benefits worth $53,128. It would receive Treasury checks of $3,400 in refundable child tax credits and $4,400 in refundable earned-income tax credits. The family would also receive Food Stamp debit cards worth $9,216 a year, $9,476 in housing subsidies, $877 of government payments for utility bills, $16,033 to fund Medicaid, $3,102 in free meals at school and $6,624 in Temporary Assistance for Needy Families. All this puts the family’s income at $64,128, or 254% of the poverty level.

A hardworking family earning anything like $64,128 in salary wouldn’t be eligible for any of these welfare benefits in four-fifths of the states. Meanwhile, the welfare family would be eligible for another 90 small federal benefits and sundry state and local welfare programs.

According to the Congressional Budget Office and other independent researchers, when all means-tested payments are counted as income, most welfare recipients have incomes that put them in the middle class, and the proportion of poor people in the U.S. falls from more than 10% to less than 1%.

This unjust system also penalizes work. Unsurprisingly, the percentage of work-age persons in the bottom 20% of income who in fact work has in the last 50 years fallen from 68% to 36%.

The budgetary effects of these inaccurate income calculations are enormous. Look at what government programs cost minus any dedicated revenue they collect and interest on the debt, which government is obligated to pay. Payroll taxes fund 87% of Social Security spending, requiring an additional $188 billion, or 4% of unobligated spending. Medicare is 45% funded by payroll taxes and uses $478 billion of unobligated spending, or 11%. Defense spending of $851 billion is 20% of unobligated spending. Means-tested welfare programs absorb $1.4 trillion, 34% of unobligated spending, and the rest of the federal government spends $1.3 trillion, or 30% of unobligated spending.

If the government simply gave every poor family in America enough money to raise its income above the official poverty level, it would cost only $240 billion. That would reduce the annual deficit by two-thirds.

In light of the mounting evidence of rampant benefits fraud, Congress should institute a comprehensive audit of all means-tested programs. But it should start with removing the largest fraud in welfare—the government’s gross overstatement of poverty. President Trump should issue an executive order requiring the Census Bureau to count all welfare benefits received from the government as income to the recipients. Then Congress should codify the executive order and require that all means-tested programs use the corrected Census income definition to determine eligibility for welfare payments.

At a minimum, the resulting debate would inform the public about the bias in how the government measures income and how that bias has promoted welfare benefits that give recipients a standard of living that most middle-income families struggle to enjoy. The debate would force spending advocates to defend a wasteful and unjust system. Welfare reform would not only help the nation begin to deal with its budget problems, but could be a powerful issue in American politics headed into the 2026 elections for those willing to champion it.

Mr. Gramm is a former chairman of the Senate Banking Committee and a nonresident senior fellow at the American Enterprise Institute. Mr. Early served twice as assistant commissioner at the Bureau of Labor Statistics and is an adjunct scholar to the Cato Institute. Mike Solon contributed to this article.