The next municipal budget crisis will not arrive because mayors overspent; it will arrive because federal fiscal architecture creates state obligations, withdraws the money, and then calls the resulting hole local irresponsibility. That is the heart of Richard Morrison and Thomas Savidge’s “A Series of Unforced Errors: The Next Municipal Budget Crises” in National Review. They want Congress to rehearse “absolutely not” to any state or city bailout, on the theory that officials made a string of bad fiscal choices and now want Indiana or Tennessee taxpayers to underwrite their experiments. The theory sounds clean. It will deliver the opposite.
Start with the analogy the column leans on: the donor who maxed out his credit cards on good causes. A household that overspends on charity must pay its own bill. So, the argument goes, must a state.
A state is not a household.
A household can stop buying things. A state cannot stop paying teachers, firefighters, or pensioners without triggering federal consequences. A household pays its debts in a currency it did not issue. A state’s bond market is denominated in federal currency and is held in part by federal institutions. And a household is a stranger to other households in a way Indiana is not a stranger to Sacramento. They share a labor market, a federal tax base, a constitution, and a union designed in part to absorb regional shocks.
When Sacramento’s bond market cracks, unemployment-insurance claims, Medicaid enrollment, and lost income-tax revenue land in Washington. They land on the same Indiana family’s tax bill within months. Refusing to share fiscal risk across the union does not restore discipline. It moves the cost from one federal line item to a worse one.
Here is the mechanism, plain. State budgets run on three big taxes: income, sales, and property. All three weaken in a recession. Income tax falls because layoffs and capital gains collapse. Sales tax falls because consumer spending falls. Property tax falls more slowly, but assessed values drop and appeals rise. At the same time, demand for state-funded services rises: Medicaid, unemployment assistance, food assistance, homeless shelters.
States are generally required to balance their budgets every year. So when revenues fall and obligations rise, they have two choices: cut services or raise taxes. Both make the downturn worse.
The op-ed notices that national growth, inflation, and unemployment numbers may look decent while local balance sheets deteriorate. It does not follow the thread. The national average includes federal spending, which expands automatically in downturns. States do not have that lever. They cannot print money, run an ordinary deficit, or adjust payroll-tax rates on the fly. They can cut or tax.
That is not a character flaw. It is a balance sheet.
Consider Medicaid. It is one of the largest lines in most state budgets, and it is jointly funded, often through a roughly 50-50 or 60-40 federal-state match. In a recession, enrollment rises because people lose employer coverage. The federal share does not automatically rise with the need. The state share of a swelling program therefore shoots up precisely when state revenues fall.
That is not a mysterious failure of local virtue. It is the design. Congress wrote the matching rule. Congress can rewrite it.
The same problem appears in federal mandates. When Washington requires states to administer programs or meet standards — IDEA special education, criminal-justice reforms, environmental rules — without fully funding the obligation, the cost moves to property taxpayers and sales-tax payers who never voted on that particular bill. The federal government gets the policy credit. The local budget gets the invoice.
The column wants to draw a clean line between “genuine disasters” and “operating deficits.” That line is harder to defend than it looks. A state whose Medicaid rolls double because a factory closes is facing a disaster by any reasonable definition. The disaster is simply slow-moving and statistical rather than visible and meteorological.
The federal government already knows how to recognize slow-moving disasters. It has flood insurance, crop insurance, pandemic unemployment assistance, and countercyclical transfers. Refusing to treat a state fiscal crisis as a legitimate consequence of a recession is not a fiscal principle. It is a category error with a tie on.
The 2008–09 stimulus makes the point. The American Recovery and Reinvestment Act sent roughly $140 billion in fiscal aid to states. It helped prevent hundreds of thousands of teachers, firefighters, and police officers from being laid off in the deepest recession since the Great Depression. The Congressional Budget Office later assessed the law as one of the more effective countercyclical measures in modern history.
Did some of that money get wasted? Probably. Did some support state pension systems that were already underwater? Likely. Was a wave of state layoffs in the middle of that recession a better alternative? No.
A recession does not become less real because its victims are spread across payrolls instead of standing under a collapsed roof.
The 1830s precedent the column invokes is real. States borrowed heavily to build canals and railroads. Many defaulted when recession hit. Congress refused to bail them out. Several states then rewrote their constitutions to require balanced budgets. That was a genuine reform, and Morrison and Savidge are right to treat it as one.
But the aftermath was not clean fiscal learning. It included depressed credit, bank failures, and the recession that helped produce the modern federal fiscal system. The lesson was not “discipline alone.” It was discipline paired with architecture that absorbs cross-state shocks.
The states of the 1830s were also genuinely independent fiscal actors in a way no state has been for a century. There was no federal income tax, no Medicaid, no federal highway system, and no unemployment insurance. Today’s states are administrative partners in federal programs as much as they are sovereign governments. Treating their budgets like a household budget that merely needs more discipline is not conservative. It is nostalgic.
Modern federal systems with explicit no-bailout clauses — Australia, Germany, and Canada — pair those clauses with fiscal-equalization transfers and automatic stabilizers. They do not pair “absolutely not” with nothing. That is the difference between a rule and a trap.
California’s case is more pointed than the column admits. Covid-era surpluses did not simply “become” recurring commitments because Sacramento was profligate. The federal government designed the American Rescue Plan Act as one-time money while the underlying problems — housing scarcity, child-care infrastructure, and climate-transition costs — were structural. When federal funds stop, state obligations do not.
That is a federal design failure transmitted through state balance sheets.
Pensions deserve a separate answer. Morrison and Savidge call underfunded public-sector pensions “politicized investing gimmicks.” The label is wrong. A pension promise is deferred compensation. Workers accepted lower wages today on the explicit promise of defined benefits tomorrow. When those promises go underfunded, the cost does not disappear. It shifts to current workers through stagnant wages and to future taxpayers through catch-up contributions plus interest.
Calling the promise a gimmick does not cancel the contract. It only disguises the bill.
Some pension promises were structurally too generous and should be reformed. Some investment assumptions were foolish. New York’s $5 billion pension deferral was bad fiscal management. Chicago splitting its supplemental payment into two installments was a budget trick. Sacramento built recurring programs on one-time money. Those are real failures, and the officials responsible should be held accountable.
But reform through default is reform through theft. And the federal government has a stake in preventing that theft because pension failures do not stay neatly inside the jurisdiction that caused them. The federal government already bears the consequences of pension collapse through federal backstops such as the Pension Benefit Guaranty Corporation, and through the broader public systems that catch people when private or local promises fail.
The disagreement is not about whether bad choices exist. They do. It is about whether a categorical “no bailout ever” produces accountability.
It does not. It produces unprepared crises.
When the next state or city hits the wall, it will arrive at midnight with no institutional structure for the response. Congress will then pass an emergency package under duress, after the bond market has seized and services have been cut. That package will cost more than a planned one would have. The Indiana family loses either way, and loses more under the duress scenario.
So build the architecture.
First, create automatic stabilizers that flow to states during downturns without an ad hoc congressional vote. When unemployment rises above a threshold, the federal Medicaid match should rise with it. Unemployment insurance and food assistance should be fully federally funded during severe downturns. The countercyclical valve should open on its own. No press conference. No special session. No begging.
Second, establish fiscal-equalization transfers that lift poorer states toward a common level of service capacity. The point is not to reward waste. It is to prevent a recession from forcing a state to choose between taxing poor residents and cutting the services they need. Australia and Canada use versions of this architecture. They also retain explicit no-bailout rules. Discipline and solidarity are not opposites.
Third, create conditional liquidity facilities for genuinely illiquid but solvent states. These are not blank checks. They are bridge loans with transparent conditions, independent audits, pension reforms where necessary, and structural-reform covenants. A state that needs a bridge should not receive a vacation from accounting. It should receive time to cross the river.
Fourth, stop writing federal mandates onto state ledgers without paying for them. If Congress wants states to administer a program, Congress should fund the administrative and service costs. If it wants a national standard, it should own the national price.
This is not compassion replacing arithmetic. It is arithmetic completed.
Morrison and Savidge are right that compassion does not repeal arithmetic. They are wrong about whose arithmetic applies. The relevant arithmetic is national: one labor market, one currency, one bond market, one constitutional union. The arithmetic of Medicaid does not stop growing when wages fall. The arithmetic of pensions does not improve because we refuse to look at it. The arithmetic of a recession does not become local merely because Congress has arranged for states to carry the first bill.
The economy is a set of choices, not the weather. Federal fiscal rules are choices, too.
Refusing to plan for the next crisis is not fiscal discipline. It is the most expensive kind of delay. Build the stabilizers, equalization transfers, funded mandates, and conditional liquidity facilities before the next state budget breaks. Then let elected officials answer for the choices they actually made.
That is not a bailout. It is a federation that finally owns its balance sheet.