Why Public Expenditure Grows

2. The inspection effect. A crisis makes the public and the government look at problems they had previously tolerated, such as poor health, illiteracy and bad housing, often because the crisis exposes them. Once inspected, they generate demands that persist.

3. The concentration effect. During a crisis, activity and revenue move to the central government, because only the centre can manage a war, and the shift is not fully reversed afterwards.

Test it against India and the answer improves. The pandemic year, 2020-21, produced exactly this pattern: expenditure rose sharply for emergency relief and free foodgrain, and did not fall back to its earlier share of national income afterwards; a free foodgrain programme begun as emergency relief was continued; and both the response and the resources were concentrated at the Union. A student who can name a domestic instance of the displacement effect is doing more than reciting the hypothesis.

Criticisms.

• Expenditure has also grown in long periods without any crisis.
• "Crisis" and "tolerable level of taxation" are not measurable, so the hypothesis is hard to falsify.
• It explains the timing of growth but not its direction: it does not say why the new expenditure goes to welfare rather than anything else.

Baumol's cost disease

William Baumol's explanation is the one students rarely produce, and it answers a question the other two cannot: why public expenditure rises even when the State does exactly the same things.

The argument. Divide the economy into two sectors. In the progressive sector, output per worker rises steadily with technology: manufacturing, agriculture, telecommunications. In the stagnant sector, output per worker barely rises at all, because the labour is the product: teaching a class, nursing a patient, hearing a case, policing a street.

Wages, however, tend to move together across the economy, because workers can move between sectors. So wages in the stagnant sector rise with productivity in the progressive sector, while its own productivity does not rise. Its costs therefore rise relative to everything else, year after year.

The consequence. Government is concentrated in exactly those stagnant activities: education, health, justice, policing, administration. So the real cost of the same quantity of public service rises continuously, and public expenditure grows as a share of national income even if the State's functions never expand at all.

Why this matters for a law student in particular. Baumol's stagnant sector is the courtroom. A judge can hear only so many matters a day, and no technology has changed that as it has changed the number of shirts a factory produces in an hour. It follows that the cost of justice, measured against the price of manufactured goods, will rise indefinitely, and that judicial expenditure will absorb an increasing share of the budget merely to maintain the same service.

The Indian causes

Beyond the general theories, an examiner expects the specifically Indian reasons.

1. Population and its changing composition. More people need more schools, hospitals, police and roads. India's population growth rate has fallen, but its composition is now the source of pressure: an ageing population increases pension and health expenditure, and the Union's pension liability is one of the fastest growing heads.

2. Urbanisation. As [India's Population: Size and Composition] shows, urban infrastructure costs far more per head than rural: water, sewerage, transport, drainage and policing in dense settlements are expensive.

3. The constitutional commitment to a welfare State. The Directive Principles are not merely aspirational statements: article 38 requires the State to secure a social order for the promotion of welfare; article 39 directs policy towards an adequate means of livelihood and the distribution of material resources to subserve the common good; article 41 requires effective provision for education and public assistance in unemployment, old age, sickness and disablement; article 47 makes the raising of nutrition and public health a primary duty. Each generates permanent expenditure.

4. Defence and internal security. Large standing forces, long land borders and a persistent internal security requirement.

5. Subsidies. Food, fertiliser, cooking gas, electricity and interest subventions. These are politically very difficult to withdraw once granted, so they ratchet upward.

6. Interest on accumulated debt. The most mechanical cause of all, and the largest: 14,03,972 crore rupees on the Budget Estimates for 2026-27, about a quarter of total Union expenditure. Every year's deficit adds to the stock of debt and therefore to next year's interest, so past deficits force present expenditure upwards without any decision being taken.

7. Inflation. Part of the growth in nominal expenditure is simply prices. Comparisons must be made in real terms or as a share of gross domestic product, and an answer that quotes only nominal growth has not made the point.

8. Development expenditure and public enterprise. The post independence choice of a mixed economy described in [Industrial Policy Before 1991] placed the responsibility for heavy industry and infrastructure on the State.

9. Democracy and political competition. Governments facing periodic elections have a strong incentive to announce new expenditure and a weak one to withdraw old.

A worked example: which explanation fits which fact

What beginners get wrong

"Wagner's law says public expenditure increases." It says it increases faster than national income, so its share rises. Expenditure rising alongside a growing economy is not Wagner's law.

"The displacement effect means expenditure rises during a war." Anyone can see that. The hypothesis is that it does not come back down, because the tolerable level of taxation has been permanently displaced.

"Rising expenditure means the government is doing more." Not necessarily. Baumol's argument is that the same service costs more every year, and inflation adds another layer of nominal growth on top.

"These are alternative theories and one must be chosen." They answer different questions. Wagner explains the trend, Wiseman and Peacock the path, Baumol the cost of an unchanged service.

"Growing public expenditure is bad." The question is what it buys. Expenditure that builds infrastructure or educates a workforce differs from expenditure that services past borrowing.

Limits

Wagner's law is empirical and has exceptions. Several countries have reduced the share of public expenditure for extended periods without ceasing to industrialise.

Wiseman and Peacock is hard to falsify because neither the tolerable level of taxation nor the threshold of a crisis can be measured.

Baumol's cost disease assumes wages move together across sectors, which holds less well where labour markets are segmented, as much of India's is.

Union figures understate the growth, since a large part of welfare expenditure is by the States.

Correlation is not cause. That expenditure and income both rose does not establish which caused which, and the direction of causation between them is contested in the literature.

Quick revision

1. Wagner's law of increasing State activity: as per capita income rises, public expenditure rises faster than national income, so its share increases. Three reasons: expansion of administrative and protective functions; growth of cultural and welfare expenditure; and the need for large indivisible investments in infrastructure.
2. Wiseman and Peacock: expenditure grows in steps, not smoothly. Displacement effect, a crisis raises the tolerable level of taxation permanently; inspection effect, the crisis exposes problems that then demand attention; concentration effect, activity and revenue move to the centre and stay there.
3. Baumol's cost disease: wages rise together across sectors, but productivity rises only in the progressive sector, so the stagnant services in which government is concentrated, teaching, nursing, policing, judging, cost more every year for the same output.
4. Indian causes: population and its ageing composition; urbanisation; the constitutional welfare commitment in articles 38, 39, 41 and 47; defence and internal security; subsidies, which ratchet; interest on accumulated debt at 14,03,972 crore rupees, about a quarter of Union expenditure; inflation; development expenditure and public enterprise; and political competition.
5. Always compare as a share of gross domestic product, never in nominal rupees alone.
6. The three theories are complements, not rivals: Wagner explains the trend, Wiseman and Peacock the timing, Baumol the rising real cost of an unchanged service.

Test yourself

1. State and explain Wagner's law of increasing State activity, and criticise it. Adolph Wagner, writing in the 1880s from the experience of industrialising Europe, observed that public expenditure was everywhere growing faster than national income, so that its share in national income rose as per capita income rose. He gave three reasons. First, the administrative and protective functions of the State expand, because an industrial and urban society is far more complex than an agricultural one and requires more regulation, more police, more courts and more inspection, so that factory law, company law, banking regulation and environmental control each bring an apparatus with them. Second, cultural and welfare expenditure grows, because education and health carry large external benefits, are undersupplied by the market, and are demanded collectively as society becomes richer.

Third, industrialisation requires large indivisible investments in railways, ports, power and communications which the private sector may be unwilling or unable to make. The law is criticised on the ground that it is an empirical observation rather than a law, since Wagner supplied no mechanism that compels the result; that it says nothing about when the growth occurs, which was the point of Wiseman and Peacock; that it was drawn from a particular historical experience and several countries have since reduced the share of public expenditure for long periods; and that it does not distinguish expenditure that purchases goods and services from expenditure that merely transfers income, although the two have quite different economic effects.

2. Explain the Wiseman and Peacock hypothesis with an Indian illustration. Jack Wiseman and Alan Peacock studied British public expenditure from 1890 to 1955 and found that it did not grow smoothly but in steps, remaining flat for long periods and then jumping, with the jumps coinciding with the two World Wars. They explained this by three effects. The displacement effect is that people ordinarily tolerate only a certain level of taxation which governments do not exceed, but a great disturbance forces expenditure up and makes higher taxation acceptable, and when the crisis passes neither expenditure nor taxation returns to the earlier level, because the tolerable level has been displaced upwards and the government finds new uses for the revenue. The inspection effect is that a crisis exposes problems previously tolerated, such as poor health or bad housing, and once they have been inspected they generate demands that persist. The concentration effect is that a crisis shifts activity and revenue to the central government, because only the centre can manage it, and the shift is not fully reversed. India furnishes a clear illustration in 2020-21, when the pandemic drove Union expenditure up sharply for relief and free foodgrain, the share of expenditure in national income did not return to its earlier level afterwards, a scheme introduced as emergency relief was continued, and both the response and the resources were concentrated at the Union.

3. What is Baumol's cost disease, and why is it important for a student of law? Baumol divided the economy into a progressive sector, in which output per worker rises steadily with technology, such as manufacturing and telecommunications, and a stagnant sector, in which output per worker barely rises because the labour is itself the product, such as teaching a class, nursing a patient, policing a street or hearing a case. Wages tend to move together across the economy because workers can move between sectors, so wages in the stagnant sector rise in step with productivity growth in the progressive sector while its own productivity does not rise, and its costs therefore increase relative to everything else year after year. Since government activity is concentrated precisely in those stagnant services, the real cost of providing the same quantity of public service rises continuously, and public expenditure grows as a share of national income even if the functions of the State never expand. For a law student the point is direct: the courtroom is the stagnant sector in its purest form, since a judge can hear only so many matters in a day and no technology has altered that as it has altered the output of a factory, so the cost of justice relative to the price of manufactured goods must rise indefinitely and judicial expenditure must absorb an increasing share of the budget merely to maintain the existing service.

4. What are the principal causes of the growth of public expenditure in India? The general causes are those given by Wagner, Wiseman and Peacock and Baumol. To them must be added the specifically Indian. Population growth, and now more importantly its changing composition, since an ageing population increases pension and health expenditure. Urbanisation, since providing water, sewerage, transport and policing in dense settlements costs far more per head than in villages. The constitutional commitment to a welfare State, for article 38 requires the State to secure a social order promoting welfare, article 39 directs policy towards an adequate livelihood and a distribution of material resources subserving the common good, article 41 requires effective provision for education and for public assistance in unemployment, old age, sickness and disablement, and article 47 makes the raising of nutrition and public health a primary duty, each of which generates permanent expenditure.

Defence and internal security, given long land borders and a standing internal requirement. Subsidies on food, fertiliser, cooking gas and electricity, which are politically almost impossible to withdraw and therefore ratchet upwards. Interest on accumulated debt, which on the Budget Estimates for 2026-27 is 14,03,972 crore rupees, about a quarter of total Union expenditure, and which rises automatically as past deficits add to the stock of debt. Price inflation, which inflates nominal expenditure without any real increase. The mixed economy inherited from the industrial policy of the 1950s, which placed heavy industry and infrastructure on the State. And political competition, which rewards the announcement of new expenditure and penalises the withdrawal of old.

5. Do the three theories compete with one another? They do not, because each answers a different question about the same phenomenon. Wagner explains the long run trend, namely why the share of public expenditure in national income rises as an economy industrialises and grows richer. Wiseman and Peacock explain the path by which that trend is realised, namely that the rise occurs in steps produced by crises rather than smoothly, because the level of taxation the public will tolerate is displaced upwards by upheaval and does not fall back. Baumol explains something neither of the others addresses, namely why expenditure rises even when the functions of the State are unchanged, since the services government provides are those in which productivity cannot easily rise while wages must nevertheless keep pace with the rest of the economy. A complete answer uses all three, and adds the mechanical Indian cause of interest on accumulated debt, which requires no theory at all.

6. Why must public expenditure be compared as a share of gross domestic product rather than in rupees? Because a nominal figure confounds three quite different things. Part of any increase is price inflation, which raises the rupee cost of the same real quantity of service and reflects no expansion of government at all. Part is the growth of the economy itself, since an economy twice as large will support a government twice as large without the State having taken any larger place in national life. Only what remains after both are allowed for represents a genuine increase in the relative size of the public sector, and that is what Wagner's law and every serious discussion of the growth of public expenditure are about. Expressing expenditure as a share of gross domestic product removes the first two effects at once, which is why budget documents and comparisons between countries are always framed in those terms, and an answer that reports only that expenditure has doubled in rupees over a decade has not established that public expenditure has grown in any meaningful sense.