India’s Defence Budget: How Much and Where It Goes

Every February, one number does the rounds: India’s defence budget crossed Rs 6.8 lakh crore. It gets quoted in headlines, pasted into notes, and repeated in interviews. And almost everyone who repeats it believes it means India spends 6.8 lakh crore rupees on weapons and soldiers. It doesn’t. Close to a quarter of that money goes to people who left the forces years ago, and only about a rupee in four is available to buy anything new. The headline number is real. It’s the reading of it that goes wrong, and reading it correctly is a small skill that separates an answer that scores from an answer that recites.

How big India’s defence budget actually is

For 2025-26, the Ministry of Defence was allocated Rs 6,81,210 crore, roughly 79 billion dollars, and that makes it the single largest allocation to any ministry in the Union Budget. Three anchors turn that number into meaning.

First, as a share of what the Union government spends. At about 13.4% of total central government expenditure (out of roughly Rs 50.65 lakh crore), defence takes about one rupee in every seven the Centre spends. Nothing else comes close as a single ministry.

Second, as a share of the economy. That same allocation is about 1.9% of GDP. Hold that against the 1980s, when India routinely spent close to 3% of GDP on defence, and you can see the direction of travel. The economy has grown much faster than the defence budget, so the absolute number keeps setting records while the share of GDP keeps drifting down. Both facts are true at once, and that’s the trap. A statement like “India’s defence spending is at an all-time high” and a statement like “India’s defence spending is historically low” can both be defended, depending on which denominator you pick. The Parliamentary Standing Committee on Defence has repeatedly asked for the figure to be lifted toward 3% of GDP, and it has repeatedly not happened.

Third, as a global rank. In the Stockholm International Peace Research Institute’s estimates for 2024, India was the fifth-largest military spender in the world at about 86 billion dollars, behind the United States (about 997 billion), China (about 314 billion), Russia (about 149 billion), and Germany (about 88.5 billion). China alone spends roughly three and a half times what India does. India has hovered between third and fifth in these tables for a decade, so treat the rank as a band, not a fixed fact.

You probably noticed the mismatch: the MoD budget is about 79 billion dollars, but SIPRI puts India at 86 billion. That gap isn’t an error. SIPRI uses its own definition of military expenditure, which folds in things the MoD budget line leaves out, such as paramilitary spending on forces with a military role. So when a question says “India’s defence spending,” check which basket is meant. If the source is a Union Budget document, use the MoD allocation. If the source is SIPRI, expect a bigger number. You can see the fuller picture in the SIPRI data on India’s defence spending and arms imports.

One caveat before we go further. Every figure here moves on the first of February, when the next budget lands. Learn the structure, the proportions, and the reasons. Refresh the digits once a year from the Union Budget documents themselves. The proportions barely shift; the digits always do.

The split that decides everything: revenue versus capital

Here’s the distinction that carries most of the marks on this topic. Revenue expenditure is what it costs to run the armed forces as they exist today: salaries, rations, aviation fuel, spare parts, repairs, training, electricity in the cantonment. Capital expenditure is what it costs to make the armed forces different tomorrow: a new fighter, a submarine, an artillery gun, a missile regiment, the land and buildings that go with them.

The analogy that works, and its limit. Think of a household earning Rs 1 lakh a month. Rent, groceries, school fees, and EMIs on things already bought take Rs 74,000. Rs 26,000 is left over to actually buy anything new. That family isn’t poor, but it can’t re-equip its life quickly, and if the grocery bill rises, the new-purchase money is the first thing squeezed. That’s the defence budget in one image. Where the analogy breaks: a family can cut the grocery bill. A defence ministry can’t cut soldiers’ pay or their pensions, because both are legal entitlements. So when the money gets tight, the only flexible line is the one that buys tomorrow’s equipment. That’s the whole reason a high revenue share is treated as a problem rather than a preference.

Here’s the actual split.

Read the table once more and the sentence writes itself: about 74 paise of every rupee keeps the existing force running or pays for the force that already retired, and about 26 paise builds the force of the future. That single ratio is the most quotable line on this topic, and it’s the one most notes never state plainly.

Where the money actually goes: the four components

Each of those four lines behaves differently, and the differences are testable.

Defence Services Revenue (Rs 3,11,732 crore) is dominated by pay and allowances. It follows the shape of the force, and the Indian force is shaped like an army. The Army has roughly 12 lakh personnel against about 1.4 lakh in the Air Force and around 75,000 in the Navy, so the Army swallows the largest share of revenue spending simply by existing. This is also why service-wise budget shares look lopsided in a way that misleads. The Army looks “expensive” on revenue and modest on capital; the Navy and Air Force look modest on revenue and hungry on capital. Neither is a verdict on importance. It’s arithmetic about people versus platforms.

Capital Outlay on Defence Services (Rs 1,80,000 crore) is the modernisation line, and within it, capital acquisition sits at about Rs 1,48,723 crore. The rest goes to land, works, and infrastructure. Capital acquisition is the money that actually signs contracts for equipment.

Defence Pensions (Rs 1,60,795 crore) gets its own section below, because it deserves one.

Ministry of Defence (Civil), Rs 28,682 crore, is the quiet fourth line that trips people up in objective questions. It isn’t the armed forces. It holds the Indian Coast Guard, the Border Roads Organisation, Defence Estates, the ministry’s own secretariat, and the canteen system. Useful, and not what most people picture when they hear “defence budget.”

Two organisations worth naming separately. DRDO was allocated about Rs 26,816 crore, of which roughly Rs 14,924 crore is capital spending on research and development. That’s under 4% of the MoD budget going into the country’s main defence research body, and it’s the number to reach for when a question asks whether India funds defence R&D seriously. For a concrete sense of what that money is trying to solve, and how long it takes, look at the Kaveri engine programme. The second is iDEX, the scheme that routes small grants to startups and MSMEs working on defence problems. Its allocation is tiny next to the acquisition budget, which is exactly the point: it’s seed money for a supplier base, not a procurement line.

The pension bill: why a quarter of the budget goes to people who already left

Defence pensions take Rs 1,60,795 crore, about 24% of the MoD budget, which is nearly as much as the entire modernisation budget. No other ministry carries anything like this, and there’s a structural reason most notes skip: civil pensions for the rest of the central government sit outside the concerned ministry’s budget, while defence pensions sit inside the defence budget. So a comparison of “ministry budgets” is comparing unlike things unless you strip pensions out.

Then there’s the reason the bill is genuinely large, not just oddly placed. Roughly 32 lakh defence pensioners are drawing against about 14 to 15 lakh serving personnel, so more than two pensioners for every serving member. That ratio exists because a soldier’s career is deliberately short. A sepoy retires around 35 to 37 so that the infantry stays young, which means the state pays that pension for forty years or more. Keeping the fighting force young and paying pensions for four decades are the same decision seen from two ends.

One Rank One Pension made the bill structurally bigger. The principle is simple enough to say in one line: the same rank, retiring after the same length of service, gets the same pension, regardless of the year of retirement. Before OROP, a havildar who retired in 1990 could draw substantially less than a havildar who retired in 2010 with an identical record, because pensions were pegged to the pay scales of the retirement year. OROP fixes that inequity, and equity here costs money. It was implemented from 1 July 2014, with arrears of roughly Rs 10,795 crore at the time. The first revision, OROP-2, took effect from 1 July 2019 and carried arrears of about Rs 28,000 crore. Since OROP promises periodic revision, every revision resets the base upward, permanently.

OROP was the right call, and the pension bill is not a scandal. It’s the priced consequence of an early-retirement model that the country wants, applied honestly. The problem isn’t the pension. The problem is that nothing was restructured to pay for it, so it lands on the same budget line that has to buy fighter jets. The Agnipath scheme, introduced in June 2022, is the one structural answer attempted so far: Agniveers serve four years, receive a Seva Nidhi package of about Rs 11.71 lakh on exit, and draw no pension. Whether that works is a genuinely open question, and the timing is the part the debate keeps missing. Agnipath’s savings arrive in the 2040s and 2050s, because the pensioners of the next twenty years have already earned their entitlement. Nothing in the next decade’s budget gets easier because of it. If you want the wider argument about defined-benefit pensions and the fiscal load they create, it runs parallel to the Old Pension Scheme debate.

Capital acquisition and the push to buy Indian

About Rs 1,48,723 crore is earmarked for capital acquisition, and roughly 75% of that, close to Rs 1,11,544 crore, is reserved for procurement from domestic industry. That reservation is the sharp end of Make in India in defence: not a slogan, a ring-fenced share of a real budget line that foreign vendors cannot bid for.

The second instrument is the positive indigenisation list. The name sounds bureaucratic and the idea is blunt: the ministry publishes a list of items with a date after which they can only be bought from Indian sources. The Department of Military Affairs has notified five such lists covering 509 major systems and platforms, and the Department of Defence Production has notified further lists covering over 5,000 sub-systems, assemblies and components. Note what a positive list really does. It isn’t an import ban for its own sake. It’s a demand guarantee, telling Indian industry that if you build this, there’s a buyer, which is the one assurance a defence manufacturer needs before sinking capital into a production line. The wider architecture of policy around this sits in the note on defence manufacturing in India.

The results show up best in exports. Defence exports reached about Rs 23,622 crore in 2024-25, against Rs 686 crore in 2013-14, roughly a thirty-fold rise in about a decade, with a stated target of Rs 50,000 crore by 2029. That’s a real trend line and a small base. India remains one of the world’s largest arms importers even while its exports climb, and both sentences belong in the same paragraph if you want to be accurate rather than triumphal.

Now the part that gets left out. A large share of the capital acquisition budget each year goes not to new contracts but to committed liabilities, the installments due on deals signed in earlier years. The Standing Committee on Defence has flagged this repeatedly. It means the headline modernisation figure overstates the freedom to buy new things, because much of it is already spoken for before the year begins. When you read “Rs 1.48 lakh crore for acquisition,” read it as mostly paying off yesterday’s decisions, with a smaller slice available for tomorrow’s.

The tension you can’t budget away

Strip the numbers back and one trade-off runs through the whole document: manpower or modernisation. India runs a large, young, manpower-heavy force along two contested land borders, and that force is expensive to pay now and expensive to pension later. Those two commitments take three rupees in four. What’s left buys the equipment.

You can attack that from three directions, and India is trying all three at once, with different odds. Shrink future pension liability, which is Agnipath, with a payoff decades out. Cut the cost of what you buy by buying at home, which is indigenisation, with real progress and a long tail on complex systems like aero-engines. Or use the existing force more efficiently by reorganising it, which is what theaterisation is about, and which costs political capital rather than money.

What isn’t on the table, realistically, is a fourth option: a large jump in the defence share of GDP. With the fiscal deficit under a consolidation glide path and health, education, and welfare competing for the same rupee, the 3% of GDP that the Standing Committee keeps recommending has stayed a recommendation for thirty years. Any answer that solves the problem by simply demanding more money hasn’t engaged with the problem.

How to study and apply this

Carry four numbers, not forty. The total (about Rs 6.8 lakh crore), the share of central expenditure (about 13.4%), the share of GDP (about 1.9%), and the 74:26 split between running-plus-pension costs and modernisation. Everything else is elaboration you can rebuild from those four.

Then attach three arguments as branches. One, the pension bill is structural, not wasteful, and Agnipath’s relief arrives in the 2040s. Two, indigenisation is a demand guarantee, evidenced by the 75% domestic earmark and the positive lists, with exports up from Rs 686 crore to about Rs 23,622 crore as the proof of concept. Three, the capital budget is smaller than it looks because committed liabilities eat into it first.

For updating, go to the source rather than to summaries. The Union Budget’s Demand for Grants for the Ministry of Defence carries the four-way split on a single page, and the annual reports of the Ministry of Defence and the reports of the Standing Committee on Defence carry the arguments. Read one Standing Committee report end to end once. It will give you sharper criticism of the budget than any coaching handout, because it’s the criticism the government itself has to answer.

One tip on framing. When a question asks about the defence budget, the weak answer lists allocations. The strong answer states the ratio, names the constraint, and then discusses the three levers. Numbers are the evidence. The trade-off is the argument.

Frequently Asked Questions

What is India’s defence budget and what does it include?

For 2025-26 the Ministry of Defence was allocated Rs 6,81,210 crore, about 79 billion dollars. It includes four things: revenue spending on the armed forces (Rs 3,11,732 crore), capital outlay for modernisation (Rs 1,80,000 crore), defence pensions (Rs 1,60,795 crore), and the MoD’s civil budget covering the Coast Guard, Border Roads Organisation and the secretariat (Rs 28,682 crore).

What percentage of GDP does India spend on defence?

About 1.9% of GDP, down from close to 3% in the 1980s. The absolute figure keeps rising while the GDP share falls, because the economy has grown faster than the budget. The Parliamentary Standing Committee on Defence has repeatedly recommended raising it toward 3% of GDP.

Where does India rank in global military spending?

India is consistently among the top five. In SIPRI’s estimates for 2024, India ranked fifth at about 86 billion dollars, behind the United States, China, Russia and Germany. China spends roughly three and a half times as much. SIPRI’s figure is higher than the MoD allocation because SIPRI’s definition of military expenditure covers items the MoD budget line excludes.

What is the difference between revenue and capital expenditure in the defence budget?

Revenue expenditure runs the force that exists: salaries, rations, fuel, spares, repairs, training. Capital expenditure changes the force: new aircraft, ships, submarines, guns, missiles, land and works. Revenue costs are legal entitlements and can’t easily be cut, so when money is tight, the capital line is the one that gets squeezed.

Why are defence pensions so large a share of the budget?

Because about 32 lakh pensioners draw against roughly 14 to 15 lakh serving personnel, and because a soldier’s career is deliberately short. A sepoy retires around 35 to 37 to keep the fighting force young, so the state pays that pension for forty years or more. One Rank One Pension, effective from 1 July 2014 and revised from 1 July 2019, raised the bill further by equalising pensions across retirement years.

What is the 75% domestic procurement rule in the defence budget?

Roughly 75% of the capital acquisition budget, about Rs 1,11,544 crore of Rs 1,48,723 crore in 2025-26, is reserved for procurement from Indian industry. Foreign vendors cannot compete for that share. It works alongside the positive indigenisation lists, which bar imports of listed items after a notified date.

How much have India’s defence exports grown?

From about Rs 686 crore in 2013-14 to about Rs 23,622 crore in 2024-25, roughly a thirty-fold rise, with a target of Rs 50,000 crore by 2029. The growth is real and the base was very small. India remains one of the world’s largest arms importers at the same time.

What are committed liabilities in the defence budget?

They’re the installments payable in the current year on contracts signed in earlier years. They consume a large share of the capital acquisition budget before it can be spent on anything new, which is why the modernisation figure overstates the room available for fresh purchases. The Standing Committee on Defence has flagged this repeatedly.

Practice Questions

1. In the Union Budget for 2025-26, the Ministry of Defence allocation as a share of total central government expenditure was closest to:

a) About 5%
b) About 13%
c) About 25%
d) About 33%

Answer: b) About 13%

2. Which of the following is NOT part of the Ministry of Defence (Civil) budget?

a) Indian Coast Guard
b) Border Roads Organisation
c) Capital acquisition of fighter aircraft
d) Defence Estates Organisation

Answer: c) Capital acquisition of fighter aircraft

3. The principle of One Rank One Pension means that:

a) Every retired soldier receives the same pension regardless of rank
b) Pension is linked only to the last salary drawn at retirement
c) The same rank retiring with the same length of service receives the same pension, irrespective of the date of retirement
d) Pension is paid only to those who complete a full 20 years of service

Answer: c) The same rank retiring with the same length of service receives the same pension, irrespective of the date of retirement

4. A positive indigenisation list notified by the Ministry of Defence primarily serves to:

a) Ban the export of listed defence items
b) Restrict the procurement of listed items to Indian sources after a notified date
c) List items on which customs duty is waived for imports
d) Identify items to be transferred to DRDO for research

Answer: b) Restrict the procurement of listed items to Indian sources after a notified date

5. “Committed liabilities” in the context of the defence budget refer to:

a) Pension payments to retired armed forces personnel
b) Installments due in the current year on contracts signed in previous years
c) The salary bill of serving personnel guaranteed by statute
d) Guarantees extended by the government to defence public sector undertakings

Answer: b) Installments due in the current year on contracts signed in previous years

Mains-style questions

1. “India’s defence budget is at a record high in absolute terms and near a historic low as a share of GDP.” Examine this apparent contradiction and discuss what it means for military modernisation.

2. Analyse the structure of India’s defence budget in terms of the revenue-capital split, and explain why a high revenue share constrains the acquisition of new capability.

3. Defence pensions absorb close to a quarter of India’s defence budget. Discuss the structural reasons behind this, and critically evaluate the Agnipath scheme as a response.

4. “Indigenisation in defence is a demand guarantee before it is a technology policy.” Discuss with reference to the domestic procurement earmark, positive indigenisation lists, and India’s defence export performance.

5. Evaluate the trade-off between manpower and modernisation in Indian defence planning, and assess the relative merits of the options available to resolve it.

The defence budget is not a scoreboard, and treating it as one is why so many answers on it read the same. It’s a set of constraints wearing the costume of a large number. India has chosen a young, manpower-heavy army on two hard borders, and having chosen it, the country pays for it twice, once in salaries and once in pensions, which leaves about a quarter of the budget to build anything new. Everything interesting in Indian defence policy right now, from Agnipath to the indigenisation lists to theaterisation, is an attempt to loosen that grip from a different direction. So when you read next February’s number, don’t ask how big it is. Ask what share of it is free, and you’ll be reading the budget the way the people who write it do.

PDF