Fiscal Policy: Spending, Taxing and the Cycle

Fiscal policy is what a government does to the economy through its own spending and its own taxes. Spending adds to demand directly, and taxing takes away from it. A tax on income collects less when incomes fall, so part of the lever moves on its own as the cycle turns. The other part has to be decided, drafted and passed. The two parts work at very different speeds.

Three things stand behind everything below. An economy runs hot and runs cold rather than sitting still, and the cycle is the name for that movement. The output measure is the total an economy produces in a year, and every ratio below is taken against it. The policy rate is the first of the two levers, and a lever can be pulled today and land months later. A government, though, is not only a rule setter standing outside the economy. A government is a buyer inside it, and a very large one. The government hires, it builds, it pays pensions, and it takes money off every earner and every sale to do so. The moment a government is treated as a participant in the economy rather than an observer of it, fiscal policy stops being a political word and becomes an arithmetic one.

What does a government actually hold, once the word fiscal is stripped back to its parts?

Two things. A government spends, and a government takes. Spending and taking are the whole toolbox, and everything else written about government finance is one of the two or the gap between them. Borrowing is not a third instrument, whatever the headlines suggest. Borrowing is what happens when the first is larger than the second, and it is a consequence rather than a choice made on its own.

Take a household for a moment. The shape is identical and the scale is the only thing that differs. A household takes money in each month and lets money out each month, and the difference either builds a balance or draws on a loan. Nobody in that household would call the loan a source of income. The loan is the arithmetic result of the first two lines. A government works the same way, except that a government can change either of the first two lines deliberately. A household usually cannot, and that difference is what makes fiscal policy a lever at all. A household cannot decide to be paid more. A government can decide to tax less, and it can decide to spend more, and either decision lands on aggregate demandThe total spending on goods and services in an economy over a period, taking households, businesses, government and the outside world together. It is the demand side of the economy considered as one number. inside the same year.

Fiscal policy is the second lever. The first one, the policy rate, works by changing what borrowing costs and letting everybody else decide what to do about it. The government does the buying itself, so fiscal policy skips that step entirely on the spending side.

What is a direct tax, and what is an indirect tax?

The taking side of the lever splits into two kinds, and the split is not about size or about which is fairer. The split is about where the tax is attached. Get that one distinction and the vocabulary of every budget document opens up.

A direct tax is levied on the person who bears it

A direct tax is attached to a person or a company, and it is calculated from something that person or company has: an income, a profit, a holding. The salaried worker whose employer deducts tax from a monthly pay slip is meeting a direct tax. So is the firm that computes a tax on the year's profit. The defining feature is that the person the tax is calculated on is also the person whose pocket it leaves from, so there is one hop and no passing on. Nothing sits in between.

An indirect tax is levied on a transaction and reaches the buyer in the price

An indirect tax is attached to an event rather than a person. Something is sold, something crosses a border, something is manufactured, and a tax attaches to that act. The seller is the one who collects it and hands it over, but the seller has put it into the price, so the buyer at the counter is the one out of pocket. A packet of biscuits bought at a tea stall shows the whole route. The shopkeeper hands money to the government; the customer handed money to the shopkeeper; the tax rode inside the price of the biscuits the whole way.

The difference between the two is who a tax is collected from against who ends up paying it, and those are two separate questions that a direct tax happens to answer with the same name. The full contrast between the two kinds, how far an indirect tax really travels into the price, what each does to people at different incomes, and what each costs to administer, is worked through separately under Direct Tax vs Indirect Tax.

Why is a rupee of spending not the same as a rupee of tax cut?

Both instruments reach demand, but they do not reach it with the same certainty or by the same route, and confusing the two is how a package gets oversold before it has done anything.

When a government spends, the government is the buyer. The government orders the cement, it pays the salary, it signs the contract for the culvert. The demand is created by the act of spending itself, and there is no second party who has to agree to anything. The full amount arrives.

When a government cuts a tax, it hands money back to households and firms and then waits to see what they do with it. Some of it is spent, and spending is demand. Some of it pays down a loan or is simply held, and neither of those is demand in that period. Picture a household that runs a small tailoring unit and finds an extra Rs 4,000/- in the month because a tax fell. Some of it goes on cloth and electricity; some of it goes into the tin behind the sewing machine because last year was frightening. The first rupee is demand by definition, and the second is demand only to the extent somebody chooses to spend it. A rupee of spending and a rupee of tax cut are not the same rupee.

How much of a tax cut gets spent is not a fixed number, and nobody can know it in advance. The share spent moves with who received the money, how confident they feel, and how much debt they are carrying. The direction is certain even when the size is not: the reach of a tax cut is smaller and less certain than its face value, and the reach of spending is not.

Which part of a deficit did nobody decide?

Part of any deficit moves without anybody deciding it, and that part changes how a deficit number should be read.

Watch what happens when output falls and nothing at all is decided. Incomes fall, so a tax calculated on income collects less. Sales fall, so a tax attached to transactions collects less. Nobody cut a rate; the same rates simply have less to bite on. At the same time, payments that depend on need go up. More people are in the state the support was written for, so more people qualify for it. The gap between what the government takes and what it spends widens on its own, and not one decision was taken anywhere in that sentence.

Run it the other way and it works in reverse. When output rises, the same tax rates collect more and need-based payments fall away, so the gap narrows without a single announcement. Built that way, the lever tightens in a boom and loosens in a slump, automatically. A transfer paymentMoney a government hands to a person or a household without receiving a good or a service in return, such as a pension or a support payment. It is spending, but nothing is bought with it. that is written around need is doing this every single day, quietly, with nobody at a desk deciding anything.

Everything else is the decided part: a new road programme, a change in a tax rate, a support package announced from a podium. The decided part needs somebody to choose it. Any change in a deficit is therefore the sum of two very different things, and reading every widening as a choice credits governments with acting when the cycle acted, and hides genuine choices inside movements that look automatic. The misreading runs in both directions.

Why does fiscal policy act with a lag, and where does that lag sit?

A lever can land late, as the policy rate does. The fiscal lag sits at the opposite end of the process, and the difference in position matters more than the length of either lag.

The policy rate can be changed at a meeting and it is changed the same day. The waiting comes afterwards. Borrowers reprice, firms reconsider a project, and households take a few months to feel it. The monetary lag sits after the lever has moved, and the waiting is for behaviour to respond.

Fiscal policy is the mirror image. Deciding to spend more is not one act; it is a sequence. Somebody has to draft the proposal. The proposal has to be debated and passed. Then land has to be acquired, tenders have to be issued, contracts have to be awarded and somebody has to actually turn up with a lorry. Only at the end of all that does a single rupee reach anybody. The fiscal lag sits before the money moves, and the waiting is for a decision to become a payment. Once the payment happens, the government has already bought the thing, so demand responds immediately.

Where the lag sits is why the two levers feel so different in practice. One acts instantly and works slowly. The other takes a long time to act and then works at once. How the policy rate transmits through the economy, and how long each stage of it takes, is set out under the policy rate.

What can fiscal policy not do?

Three limits, and each one is the reason a later part of this subject exists at all.

The first limit is physical. Fiscal policy acts on demand, and demand is not production. If an economy is already running at what it can produce, with its workers employed and its plant running, then adding more demand does not add more output. The extra demand adds to prices instead. The ceiling is potential outputThe level of production an economy can sustain with the workers, skills and capital it actually has, without pushing prices up. It is an estimate rather than a figure anyone can read off a meter., and the lever cannot lift it. Building the capacity that raises the ceiling is a different and much slower project than adding to the demand that presses against it.

The second limit is speed, and it has just been shown. The decision has to be drafted, passed and executed before anything reaches anybody, so the decided part of the lever cannot be pulled quickly. By the time it lands, the phase of the cycle it was aimed at may have passed.

The third limit is the one that carries the rest of this subject. Every rupee a government spends has to be taken from somebody or borrowed from somebody. There is no third source. Every fiscal action has a funding side, and the funding side is not optional. Borrowing more puts the government into competition for the same pool of savings as everyone else. The competition pushes on what lenders demand, and the pressure shows up in the bond yieldThe return a lender actually earns on a government bond, expressed as a per cent a year. It moves with what lenders demand, and how it is worked out belongs to fixed income rather than to this subject.. The mechanism from borrowing to yields is set out under How Government Borrowing Pushes Bond Yields, and how a bond itself is priced belongs to fixed income.

What does a real government account look like, read straight through?

The Republic of Sankhya published this account for the year its output measured Rs 17,47,200 crore. The account reads as a statement of money in and money out, before anybody attaches a name to the difference.

Every figure above belongs to the Republic of Sankhya. The output of Rs 17,47,200 crore is the same published Sankhya figure used elsewhere in this subject and is a nominalMeasured in the prices of the year in question, with no adjustment for inflation. A nominal figure grows when prices grow even if nothing more was produced. figure, measured in that year's own prices.

Read it top to bottom and the arithmetic is plain. Rs 3,00,000 crore came in and Rs 4,00,000 crore went out, so Rs 1,00,000 crore had to be borrowed. As a share of that year's output the gap is 5.72 per cent. The single gap of Rs 1,00,000 crore has more than one name and more than one useful reading, and pulling it apart into the several deficit measures is set out under the deficit measures.

What does an analyst actually watch when a package is announced?

Not the headline. The headline is a single number, usually chosen to be large, and it says almost nothing about what will reach demand or when. An analyst pulls apart the composition instead.

The first question is how much of the package is spending the government does itself. Spending of that kind becomes demand at the moment it is paid, at close to its full size. The second question is how much is a payment handed to households. A payment becomes demand only to the extent it is spent, and how much gets spent depends heavily on who received it. The third question is how much is a tax cut, the least certain of the three. The fourth is how much of the total was already in the previous year's plans and has simply been re-announced. Two packages of identical headline size can reach demand at completely different speeds and completely different amounts, and the split between those four buckets is what decides which of the two is on the table.

A lender running a credit book asks a narrower version of the same question. The lender needs to know whether the money lands on the borrowers in the book, and in which quarter. Only one of those two lines buys equipment, so an investor looking at an infrastructure supplier asks whether the capital line moved or only the running line. In every case the useful reading comes from the split, and the split is in the document rather than in the headline.

The Republic of Sankhya is invented.
Educational material. Not advice on any investment, tax, budget or market position.