Issues

Taxes and Public Finance

Brass balance scales and a ledger book on a stone windowsill

Public finance is where economics is least intuitive, because the person who hands over the money is frequently not the person who ends up worse off. Almost every recurring confusion in tax coverage traces back to that one gap between the legal and the economic account.

Incidence: who actually pays

The statutory incidence of a tax is who is legally required to remit it. The economic incidence is who ends up bearing the cost after prices, wages and quantities have adjusted. They are different, sometimes completely.

The mechanism is straightforward once seen. A tax drives a wedge between what a buyer pays and what a seller receives; the wedge is shared between them in proportion to how little each can adjust. Whichever side is less able to walk away bears more of it. A tax on a good that people buy regardless of price falls mainly on buyers. A tax on a good that buyers can easily substitute falls mainly on sellers. This is why the economics literature treats the corporate income tax as partly borne by workers and consumers rather than wholly by shareholders — the degree is genuinely disputed and estimates vary widely, but that some of it shifts is not.

The practical rule for reading tax coverage: "a tax on X" describes a remittance obligation, not a conclusion about who is worse off.

Marginal and average rates

A progressive income tax applies rising rates to successive bands of income. A marginal rate is the rate on the next unit earned; an average rate is total tax divided by total income. Because only the top slice of income faces the top rate, the average is always lower than the marginal, often by a great deal.

This is the source of the persistent belief that a raise can leave someone worse off by pushing them into a higher band. Within a pure banded income tax it cannot. It can happen where a benefit or credit withdraws sharply at a threshold, and those cliff edges are real and worth reporting — but they are a feature of the benefit design, not of the tax bands.

The marginal rate is the one that matters for behaviour, since it prices the next hour of work or the next unit of investment. The average rate is the one that matters for the household budget. Both are legitimate; they answer different questions, and coverage often quotes one while discussing the other.

Deficits, debt and the ratio

A deficit is a flow: the gap between spending and revenue over one year. Debt is a stock: the accumulated total of past deficits less surpluses. A falling deficit still adds to the debt. Only a surplus reduces it.

Almost all serious analysis uses the ratio of debt to national income rather than the absolute figure, because the ratio is what determines whether the burden is manageable. That ratio has a denominator, which is why nominal growth reduces it without a single fiscal decision being taken, and why comparing debt across countries or decades without adjusting is close to meaningless.

The sustainability question turns on a comparison economists write as r minus g: the average interest rate on the debt against the growth rate of the economy. When growth exceeds the interest rate, a country can run modest primary deficits indefinitely without the ratio rising. When the relationship reverses, arithmetic starts working against it. This is why the same debt level can be alarming in one decade and unremarkable in another, and why "the debt is now larger than ever" is, on its own, an almost content-free statement.

Projections and their assumptions

Most fiscal coverage is built on ten-year projections, and these have a specific and often unstated character: they are usually current-law baselines, showing what would happen if every existing provision ran exactly as written, including scheduled expirations that everybody expects to be changed. A baseline is a measuring stick, not a forecast.

Projections also compound their assumptions. Small differences in assumed growth, interest rates or health-cost inflation produce very large differences by year ten. Official scorekeepers publish sensitivity analyses showing exactly this, and those tables are usually the most informative page in the document. In the United States the Congressional Budget Office publishes both the baselines and the sensitivity ranges; the Government Accountability Office publishes long-run fiscal outlooks on a different methodology, and the comparison between them is instructive.

Spending: where the money actually goes