# Tax Cuts Are Context

Every tax cut makes a trade. The state gives up a claim on resources. Someone outside the state keeps those resources. The cut helps the system only if the private activity released is more productive than the public capacity surrendered.

That is the whole test, and almost no tax politics wants to hold it still long enough to use it.

The company analogy is real at one level. A firm can cut costs and become stronger. The good version is not starvation. It is discrimination. Remove the projects, meetings, managers, software seats, and internal rituals that absorb attention without producing learning. Preserve the capabilities that make the firm worth owning. Convert the freed margin into search, speed, or price. In that context, spending less is not anti-growth. It is how the firm stops paying for its own confusion.

The state version is stricter. Taxes are not just overhead imposed on an otherwise complete market. [The Tax Floor](the-tax-floor) names one reason: taxation is also fiat's demand engine, the recurring obligation that forces economic actors to acquire the state's currency. But demand for the state's money is not the same question as growth. The next question is whether the marginal tax dollar creates more productive capacity than it consumes.

The state supplies part of the market's operating layer: courts, property rights, roads, ports, police, defense, schools, money, tax administration, public health, bankruptcy law, statistics, permitting, and macro stability. These are not sentimental public goods. They are production conditions. A tax cut that preserves them can release productive energy. A tax cut that weakens them can make private actors richer in cash and poorer in usable environment.

So the analogy survives, but only after translation. A state is efficient when it raises the social return per public dollar claimed. Sometimes that means taxing less. Sometimes it means taxing differently. Sometimes it means collecting more reliably. Sometimes it means cutting programs. Sometimes it means spending more on the shared capacities without which private growth slows down.

The shared mechanism is not less. It is better allocation while the system continues to function.

## Why The Record Looks Contradictory

The historical evidence looks confused because the phrase *tax cut* names several mechanisms at once.

One mechanism is bottleneck relief. Christina and David Romer's postwar U.S. study found that exogenous tax increases were strongly contractionary; their baseline estimate says a tax increase equal to one percent of GDP lowered real GDP by almost three percent over the next three years. That is not a small effect. It means taxes can bind hard when they hit margins that matter.

Another mechanism is ordinary demand stimulus. In a slack economy, a tax cut can put purchasing power into hands that spend it. Output rises because idle resources return to use. That is real, but the evidence points to an economy below capacity, not to lower taxes permanently raising the economy's speed limit.

A third mechanism is tax-mix efficiency. OECD work ranks tax instruments by likely growth cost: corporate taxes tend to be most harmful, then personal income taxes, then consumption taxes, with recurrent property taxes least harmful. The policy implication is not "tax less" in the abstract. It is "tax with less distortion." Lower the rate that most changes investment or labor behavior; broaden the base; reduce avoidance; preserve the revenue needed for state capacity. The total tax take might not move much. The system can still become more pro-growth.

A fourth mechanism is fiscal leakage. A tax cut can become asset appreciation, profit shifting, shareholder return, avoidance, or deficit-financed consumption. GDP may rise a little. The state balance sheet may worsen a lot. The Tax Cuts and Jobs Act lives near this case. Original projections expected modest GDP gains rather than a self-funding boom, and later CRS review found that the post-2017 empirical literature as a whole did not demonstrate significant economic effects from the law. The cut may have moved some output. It did not prove the doctrine.

Then there is the hard reversal: state-capacity formation. IMF work on the tax-capacity threshold finds that countries crossing roughly ten percent tax-to-GDP, when the increase is sustained and accompanied by broader institutional progress, see meaningfully faster cumulative growth over the following decade. The growth comes not from extraction as such, but from escaping a low-capacity equilibrium. Below a certain floor, the state cannot buy enough administration, public investment, law, or credibility for private compounding to become durable.

Put these together and the contradiction dissolves. Tax cuts raise GDP when they relieve a real bottleneck, stimulate idle demand, or shift the tax mix toward lower-damage collection. Tax cuts fail or backfire when they mostly transfer cash, worsen debt, subsidize avoidance, or cut below the public-capacity floor.

The sign was never in the cut. The sign was in the context.

## The State Is Not A Bad Department

The mistake in the crude analogy is treating government as if it were a bloated department inside the economy.

Some of it is. Some state activity is rent. Some is obsolete. Some exists because its beneficiaries are organized and its victims are diffuse. Some compliance regimes convert real labor into box-checking. Some agencies defend process because process is easier to measure than value. Cutting that can help.

But the state is also the thing that decides whether contracts mean anything, whether violence is privately priced, whether roads connect, whether children learn, whether banks can fail without freezing the payment system, whether property is legible, whether food and medicine can be trusted, whether the currency keeps score, whether a new firm can be formed without asking a local patron for permission. That is not a department. That is the condition layer for many departments.

This is why "government should run like a business" fails at the exact point where it becomes emotionally satisfying. A business can abandon customers it cannot serve profitably. A state cannot abandon regions, legal order, disease control, defense, courts, or the money system without changing the game everyone else is playing. A business can narrow its scope to what it is best at. A state has to maintain the shared floor on which specialized excellence becomes possible.

The state-capacity defense has its own failure mode. Every incumbent program can describe itself as capacity. Every agency can call its survival public value. That cannot be the answer either. The burden has to be symmetric. The defender of the tax must name the capacity bought and the return it plausibly produces. The defender of the tax cut must name the bottleneck relieved and the capacity not lost.

Cutting a useless subsidy and cutting court capacity both reduce spending. They do opposite things to growth.

## Four Tax-Cut Contexts

The first context is **bottleneck relief**. A tax is high, narrow, complex, or marginally placed in a way that changes productive behavior. The cut releases work, investment, formation, or risk-taking. This is the clean case for the analogy.

The second is **tax-mix repair**. The state does not mainly reduce its claim; it changes the shape of the claim. It lowers high-damage rates, broadens the base, closes avoidance paths, simplifies compliance, and leans toward less distortionary revenue. This is the strongest form of state efficiency because it preserves capacity while reducing drag.

The third is **cash transfer**. The state gives up revenue, but the released resources do not become much real production. They become consumption that fades, asset prices, buybacks, avoidance, or larger deficits. The private sector has more cash; the productive system has not become much more capable.

The fourth is **capacity destruction**. The cut weakens public goods or administration that private actors rely on. It can feel pro-growth because cash visibly returns to taxpayers while the lost capacity degrades slowly. The bridge is not repaired. The court backlog grows. The school system worsens. The tax agency loses enforcement ability. The statistical base thins. Years later the country discovers that it did not cut fat. It sold coordination capacity.

Spending cuts have the same four-context structure. They can remove waste, repair incentives, transfer pain, or destroy capacity. Fiscal consolidation based on spending cuts has sometimes been less costly than tax-based consolidation, especially when paired with credible reforms. But austerity during depressed conditions can impose large output losses. Again, the word *cut* is not enough information.

## The Real Growth Question

GDP is a useful aggregate and a dangerous oracle.

It can rise because people are producing more valuable things with the same inputs. It can rise because deficit-financed demand pulled activity forward. It can rise because profit was booked in a low-tax jurisdiction. It can rise while household security falls, public infrastructure decays, or state capacity gets consumed. The number is not fake. It is incomplete.

The better question is whether the policy increases the system's capacity to compound.

Does the tax cut cause new investment that would not otherwise have happened? Does it increase work, formation, or risk-taking on a real margin? Does it simplify the code enough that effort moves from avoidance to production? Does it preserve the public inputs that private actors need? Does it keep the debt path credible? Does it make the state more capable per dollar, or merely smaller per headline?

Those questions make the analogy usable.

The firm version says: cut the spending that prevents the firm from learning, but keep the capability that lets it search.

The state version says: reduce the public claim where it damages production more than it funds capacity, but keep or improve the capacity that lets private production compound.

That is why "taxing less leads to higher GDP" is sometimes right and usually underspecified. Taxing less can help when the tax was the bottleneck. Taxing less can hurt when the revenue bought the floor. Taxing less can do little when the released money flows into claims on existing production rather than new production.

Do not ask whether lower taxes are pro-growth. Ask what margin the tax touches, what capacity the revenue funds, what behavior the cut releases, and whether the released behavior compounds.

Tax cuts are context. State efficiency is context. The goal is not a smaller state or a larger state. The goal is a state whose claims on the economy produce more future capacity than they consume.

---

*Source trail: Christina and David Romer, ["The Macroeconomic Effects of Tax Changes"](https://www.aeaweb.org/articles?id=10.1257%2Faer.100.3.763) and the Berkeley-hosted PDF; Congressional Research Service, ["Tax Rates and Economic Growth"](https://www.congress.gov/crs-product/R42111) and ["Economic Effects of the Tax Cuts and Jobs Act"](https://www.congress.gov/crs-product/R48485); OECD, ["Tax Policy Reform and Economic Growth"](https://www.oecd.org/en/publications/tax-policy-reform-and-economic-growth_9789264091085-en.html); World Bank, ["Fiscal Policy"](https://www.worldbank.org/ext/en/topic/fiscal-policy-and-growth/fiscal-policy); IMF working paper, ["State Capacity, Institutions, and Growth: Taxing for Takeoff"](https://www.imf.org/-/media/files/publications/wp/2025/english/wpiea2025203-source-pdf.pdf); NBER working papers by Alesina, Favero, and Giavazzi on austerity composition and by Jorda and Taylor on state-contingent austerity costs; OECD, ["Assessing government spending in OECD countries and searching for savings"](https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/10/assessing-government-spending-in-oecd-countries-and-searching-for-savings_6724b307/0697f1d7-en.pdf).*

provenance · first_seen 2026-05-14T14:27:06Z · drafted 2026-05-14T14:27:06Z · published 2026-05-14T14:55:43Z · edited 2026-05-24T16:30:57Z