The National League of Cities’ 2026 Municipal Infrastructure Conditions Report finds that municipalities are increasingly funding capital projects out of current revenue rather than borrowing — a shift the NLC attributes to “both fiscal caution and limited financing capacity.” Property taxes, which account for roughly 60 percent of municipal tax revenue and are relied on by nearly 90 percent of cities, increasingly cover both operating costs and capital cash needs.

Nationally, that looks like a contradiction: Major municipal bond underwriters expect the aggregate market to set another record year, near $600 billion. But that masks a separate trend inside typical city budgets: governments leaning harder on cash than a few years ago, against a backdrop where the stakes keep rising. While the American Society of Civil Engineers’ 2025 Report Card gave U.S. infrastructure its best grade ever, a C, that still reflects a widening $3.7 trillion investment gap.

The caution is understandable. Rates remain well above where they sat for most of the 2010s, and officials who lived through the post-pandemic spending surge are wary of committing future revenue while so much else — federal aid, tax conformity, reserve capacity — is in flux. I’ve written before about how that volatility is reshaping budget assumptions generally; avoiding new debt keeps options open.

But there’s a difference between using debt sparingly and treating debt avoidance as a virtue in itself. The research on this question is older than the current caution cycle, and it doesn’t support a blanket answer either way.

Public finance scholars have tested this question against actual spending data for decades, and the findings cut both ways. Wen Wang and Yilin Hou’s influential study of state capital outlay found that while pay-as-you-go reduces long-run volatility, it increases short-run “lumpiness” as projects wait for cash, then bunch together. Debt smooths spending and stabilizes tax rates, at the cost of interest and committed future revenue.

A Michigan Department of Transportation-commissioned review reached a similarly split verdict: Bonding adds interest costs but speeds construction; accumulating cash for five or 10 years can raise costs through inflation and deterioration. Its conclusion was appropriately blunt: “There is no one-size fits all answer.”

For some governments, the shift away from borrowing isn't a preference; legal debt capacity is the constraint. State debt limits vary widely by jurisdiction and government type, but many cap general obligation borrowing relative to assessed valuation, and some impose additional voter-approval requirements. The state of Washington, for example, generally limits non-voted general-purpose debt for cities and counties to 1.5 percent of assessed value and total general-purpose debt to 2.5 percent.

Those limits aren’t arbitrary. Many of today's municipal debt restrictions trace to the 19th century, when a wave of railroad-bond defaults convinced states that local officials needed a legal backstop against overcommitting future taxpayers. They also explain a shift buried inside the cash-versus-debt numbers: a move from general obligation debt toward revenue bonds secured by a specific income stream — water and sewer charges, tolls, utility fees — rather than general taxing power. Revenue bonds are often exempt from those limits because repayment is pledged from that income source, preserving general obligation capacity while tying repayment more closely to the people who use the system.

“Cash versus debt,” in other words, is often really three choices — pay-as-you-go, general obligation debt and revenue debt — each matching payer to beneficiary differently and constrained differently by law. Treating it as a binary can obscure that the best-matched option, for a project with identifiable users, is often the one reached for last.

That no-one-size-fits-all framing understates what’s really at stake, because the varying methods don’t just distribute cost differently — they distribute it across different people. This is the benefits-received principle underneath most public finance textbooks’ treatment of capital debt: Infrastructure built to last decades will serve residents who move to town later, businesses not yet located and children not yet born. Debt is a matching mechanism. It lines up who pays with who benefits, spreading the bill across the years the asset is in service.

Pay-as-you-go inverts that. Fund a $30 million water plant entirely from five years of accumulated cash, and today's ratepayers can bear a disproportionate share of the construction cost for an asset that will serve residents for decades. In that sense, pay-as-you-go can create an intergenerational transfer of its own, from current taxpayers to future users.

The Government Finance Officers Association’s (GFOA) own guidance reflects this logic rather than a preference for either method: Link funding strategy to the asset’s useful life and treat debt as “a valuable strategy for governments to spread the cost of significant long-term assets over their useful life” — the benefits-received principle almost verbatim. The implicit rule: Pay-as-you-go suits short-lived, recurring needs, such as resurfacing a stretch of sidewalk, where today's payer and user are close enough to the same person that matching hardly matters. Debt suits large, long-lived assets, such as the street that sidewalk runs along, because it aligns payments with the years of public benefits.

None of this argues for reflexive borrowing. The research points toward an adaptive mix, cash when available and debt capacity preserved for when it isn’t, not a permanent posture either way. A cash-only default, adopted for good short-term reasons and never revisited, carries its own hidden cost: Deferred projects don’t disappear, they compound, and the eventual bill — higher construction costs, worse asset condition — still lands on someone. It’s just less visible than a line item for debt service, and not obviously on the people best positioned to pay it.

For every major capital item, the financing decision should start with a question, not a standing policy: Whose benefit does this project deliver, and over what time horizon? That answer — not comfort with reserve levels or discomfort with rates — should drive the choice.

Worth revisiting: Do capital policies still link financing method to asset useful life, as GFOA recommends, rather than defaulting to cash because it feels safer? Is the cost of pay-as-you-go — projects delayed or re-bid at higher prices while cash accumulates — tracked the way debt service is? And has debt capacity actually been recalculated against current debt limits lately, or is it being assumed away? A project with a clear, dedicated revenue stream deserves a look at a revenue bond, not just a choice between cash and general obligation debt.

Debt is not indiscipline. It’s a tool for matching who pays with who benefits, and for a long-lived asset it can be the more equitable choice, not the more reckless one. The question isn’t whether to avoid debt. It’s whether the financing choice for each project reflects who is actually going to use it.

Governing's opinion columns reflect the views of their authors and not necessarily those of Governing's editors or management.