Government Budgets and Fiscal Policy

This is Unit Fourteen of Principles of Macroeconomics, part two of two: government budgets and fiscal policy. Part one built the budget quantities and named the fiscal actions. This part measures the two kinds of fiscal policy apart, and asks why the deliberate kind so often disappoints. Carry one idea through. The fiscal policy that requires a decision is the one that arrives late and arrives weak, and the fiscal policy that requires no decision works immediately and keeps working. Most of this part is the apparatus for telling those two apart, and the closing entry shows what a rule forbidding deficits would destroy. By the end you will classify a budget change as automatic or discretionary, separate the automatic part of a deficit from the chosen part, name which fiscal lag a delay is, assemble the reasons discretionary policy disappoints, and state the economic objection to an annual balanced budget rule.

Three entries on telling the two kinds apart, and on measuring them apart.

Sorting a budget change.

MAC 1322 Automatic or Discretionary

Classify a budget change as automatic or discretionary, then say which way it pushed, asking one question: did anyone have to decide? If the change followed from existing law as conditions changed, it is automatic; if a new measure was passed, it is discretionary. Unemployment rises and benefit payments rise with it under rules already in force: automatic, and pushing against the downturn. A legislature passes a new temporary tax rebate: discretionary, and expansionary. Corporate tax receipts fall because profits fell: automatic, and again cushioning. A boom raises income tax receipts as earnings rise: automatic, and leaning against the boom. Now the case worth the attention. A legislature raises the benefit payment per claimant: that is discretionary, even though it acts on an automatic programme, because someone decided. And note what the stabilisers do to the aggregate demand curve, which is nothing. Because they operate automatically, their effect is already built into the curve rather than shifting it.

Measuring the chosen part.

MAC 1323 The Standardised Employment Budget

The standardised employment budget asks what the deficit would be if the economy were at potential output. That strips the automatic stabilisers out and leaves the part of the balance that policy chose. In a recession it is smaller than the actual deficit; in a boom it is larger. The gap between the two measures is what the stabilisers did.

Reading the gap.

MAC 1324 Reading the Gap Between Two Budgets

Given an actual deficit and a standardised one, separate the automatic part from the chosen part and say what each tells you. Subtract the standardised figure from the actual: what remains is the stabilisers' contribution. Then read the standardised figure on its own as the stance policy deliberately took. An economy in recession runs an actual deficit of three hundred billion. The standardised estimate, computed as though output were at potential, is a hundred and twenty billion. So a hundred and eighty billion of that deficit is the automatic response to a weak economy, and a hundred and twenty is the deliberate stance. Now let the economy recover to potential with no new legislation. Tax receipts rise and benefit spending falls on their own, the automatic hundred and eighty disappears, and the actual deficit falls toward a hundred and twenty without anyone deciding anything. This is why a deficit narrowing during a recovery is weak evidence of a policy change: most of the movement can be the stabilisers unwinding.

Two entries on why the deliberate kind arrives late, and on the two delays the automatic kind never meets.

The three delays.

MAC 1325 The Three Lags

Discretionary fiscal policy runs on the same three lags as monetary policy. Recognition waits on data about where the economy is. Implementation waits on agreeing and enacting a measure, and the anchor calls that one the legislative lag. Impact waits on the economy's response. Fiscal policy's implementation lag is the long one, and its impact lag the shorter.

Naming a delay in practice.

MAC 1326 Name the Lag, Fiscal Version

Given a delay, name which lag it is and say what could shorten it, testing what is being waited on, exactly as at unit ten. What differs here is the weighting: agreeing a measure takes far longer than agreeing a rate change, so the middle lag dominates. A recession is later dated as having begun ten months before anyone announced it: recognition, shortened only by better and faster data. A tax cut is proposed as a remedy for a recession, recommended two years later, and enacted three years after that recession had already ended: implementation, and it is the one fiscal policy does worst. Money is appropriated but the agency approving the projects takes two more years: still implementation, because nothing has reached the economy. Spending finally begins and firms take a further year to hire and build: impact. Automatic stabilisers escape the first two lags entirely, which is their real advantage. They need no recognition, because they respond to the economy itself, and no implementation, because the law already exists.

Two entries on a second reason measures underperform, and on the two policies acting on one curve.

Why temporary is weaker.

MAC 1327 Temporary Against Permanent

People respond more to a change they expect to last than to one they know is temporary, because a temporary change barely moves lifetime income. This is unit nine's permanent income hypothesis arriving as a policy problem. Countercyclical policy is temporary by design, so the feature that makes a measure responsible also makes it weaker.

The two policies together.

MAC 1328 The Policy Mix

Fiscal and monetary policy act on the same aggregate demand curve, so the combination matters. Expansionary fiscal policy works better when the central bank holds short rates down, and monetary policy can keep a fiscal contraction from causing a recession. Four combinations exist, two pulling together and two against. Neither source names this arrangement: the term policy mix is the course's own.

One entry assembling the whole case against the deliberate kind, and saying what survives it.

Why it disappoints.

MAC 1329 Why Discretionary Fiscal Policy Disappoints

In the nineteen sixties it was widely believed the business cycle had been tamed by fiscal management, and the nineteen seventies ended that belief. The reasons are worth assembling carefully, because they are cumulative rather than alternative. The first is the lags: fiscal policy's middle lag is the worst of any policy instrument, so a measure aimed at a downturn can arrive after it and amplify the next expansion instead. The second is crowding out: government borrowing raises demand in financial markets, interest rates rise, and private investment and net exports fall, partly offsetting the expansion — unit fifteen owns that mechanism and works it. The third is that countercyclical measures are temporary by design, and people respond less to temporary changes. The fourth is structural: demand policy can raise total spending, but it cannot recreate jobs in an industry that has permanently contracted. The fifth is political asymmetry, since expansion is popular and restraint is not, so the symmetric policy the model requires is rarely what gets delivered. Take the lags alone. A downturn begins. Data arrive with a delay and are revised, so it is recognised nine months later. A measure is debated and enacted over the following year. The money is appropriated but must pass through agencies before it reaches contractors, taking many more months. Firms then take a year to hire and build. Add those and the bulk of the demand arrives roughly three years after the downturn started, by which time the economy may have recovered on its own, so the measure now pushes an economy near potential further past it and buys inflation rather than output. Nothing in that sequence required anyone to act foolishly. Three cautions. First, this is an argument about discretionary measures, not about fiscal policy as a whole: the automatic stabilisers escape the first two lags entirely and keep working throughout, and the honest conclusion is to lean on them and reserve discretionary action for downturns deep or long enough to outlast the delay. Second, the size of the fiscal multiplier is genuinely unsettled, and the range is wide. One careful review reports estimates near zero point eight to one point five, with a defensible range from zero point five to two. Another study of the same decade concludes the effect was close to zero. A third argues the available data cannot settle it, and a later survey calls the spread of published estimates embarrassingly large. Unit nine said the multiplier's size is contested; this is what that contest actually looks like. Third, none of this establishes that fiscal policy does not work, and the anchor's own summary is that it is a blunt instrument, more club than scalpel — which is a statement about precision, not about power.

One entry to close the unit: a proposal to remove the discretion entirely, and what the objection to it does not claim.

The balanced budget question.

MAC 1330 The Balanced Budget Question

Here the unit's governing idea pays off. A recurring proposal would require the federal budget to balance every year, and examining it pulls together most of this unit. The case for it is direct: deficits accumulate into debt, debt carries interest, interest crowds out other spending, and a rule removes the temptation to borrow for present consumption and bill the future. The economic objection is equally concrete, and it is the one this unit has built the apparatus to see. An annual balance requirement would disable the automatic stabilisers, because the stabilisers work precisely by letting the deficit widen in a downturn. Under such a rule, a recession that cut tax receipts would compel a tax rise or a spending cut in the same year, which contracts demand exactly when demand has already fallen. The rule would convert the budget from a shock absorber into an amplifier — and part one showed this is not hypothetical, because the state and local governments that operate under such rules do contract in recessions. Take two economies meeting the same recession. In the first, no balance rule applies. Output falls, tax receipts fall automatically, benefit spending rises automatically, the deficit widens by a hundred and eighty billion without any decision, and disposable income falls by less than output does, so consumption holds up and the downturn is shallower than it would have been. In the second, an annual balance rule applies. The same fall in output opens the same gap, and the government must close it within the year by raising taxes or cutting spending. Either move reduces demand further. The downturn deepens, receipts fall again, and the gap reopens, so the rule demands a second contraction. Three cautions, and the first is that the objection above is not an argument for deficits. It is an argument against an annual balance requirement, which is a different proposition: a rule framed over the cycle rather than the year would not have the same defect. Second, the household analogy that usually drives the proposal does not carry. Households do borrow for houses and education against future income, so even at the household level the rule is not how anyone behaves, and a government additionally has responsibilities for the whole economy that no household has. Third, none of this makes deficits safe. Borrowing that funds long-lived investment is different in kind from borrowing that funds present consumption, and very large sustained deficits can raise interest rates and crowd out private investment, which is unit fifteen's subject. The course holds both halves: a deficit is neither automatically prudent nor automatically reckless, and the question is always what the borrowing bought.

Look at what you can now do. You can classify a budget change as automatic or discretionary by asking whether anyone had to decide — including the case where a legislature changes the rules of an automatic programme, which is discretionary because someone chose it. You can separate the automatic part of a deficit from the chosen part, and say why a deficit narrowing during a recovery is weak evidence of a policy change. You can name which fiscal lag a delay is, and say which two the automatic stabilisers escape entirely. You can assemble the five reasons discretionary fiscal policy disappoints, keeping them cumulative rather than alternative, and still say what the criticism does not establish. And you can state the economic objection to an annual balanced budget rule without turning it into a defence of deficits. Unit fifteen asks what that borrowing does to everybody else.