Why Arvind Subramanian thinks GST 2.0 won’t fix deep structural risks

In an exclusive conversation with The Federal, former Chief Economic Adviser Arvind Subramanian said India’s latest GST reforms, announced by Finance Minister Nirmala Sitharaman last week as a major stimulus measure, could move the country closer to a simplified tax system, but they are unlikely to deliver lasting economic gains.

Sitharaman, unveiling what she described as "GST 2.0", cut rates across multiple slabs, eliminated the top 28 per cent bracket, trimmed items in the 12 per cent and 18 per cent slabs and projected that the measures would spur consumption by lowering prices for households, ahead of the festive season.

Watch | 'GST 2.0 is a correction, not a reform; earlier model was exploitative'

Subramanian, who helped shape the original GST rollout under Arun Jaitley, cautioned that while rate rationalisation is a welcome step, structural impediments, enforcement risks and weakened trust between states and the Centre limit the reforms’ ability to permanently boost growth. The economist also flagged that revenue losses could be far larger than the government’s estimate of ₹48,000 crore, potentially affecting fiscal targets for FY26.

Edited excerpts:

India had multiple rates during the introduction of GST. Now, these have collapsed into two slab structures. How close are we to the "ideal" GST that the founding team had envisioned? What do you think is still missing?

Firstly, I think it's a little bit misleading to call it a two-slab GST. Because even before these reforms were done—excellent reforms, we’ll come to that in a second—we wrote a piece in the Business Standard which showed that there were actually about 45 slabs, if you include the cesses and things. So, it's actually a very complicated system.

There are also all kinds of duty inversions, no ITC, and then these pizza–pappadam classifications—whether it’s salted, sweet, or something else. Then you have end-use-based classification. So, there’s a lot of complexity.

There’s a bit of irony here. If you believe the official numbers, India is growing at 8%. In that case, the economy doesn’t need any boost—because if you're growing at 8%, this is the world’s best-performing economy by far.

What this reform does is eliminate the 28 per cent slab, which is an excellent step. When I wrote the report, I said there should be nothing more than 18 per cent. So we’ve gone back to that, and that’s very good. They’ve also eliminated many items in the 12 per cent slab. So those two slabs have been removed.

I think some of the cess items have been brought to 40 per cent and hence some simplification has taken place. Rate cuts have happened as well. I would say we’ve moved one step closer to a good and simple tax, but we’re still quite far away from an ideal GST, because there’s still a lot of complexity in the system. And then, there’s the whole question of enforcement, which we can talk about later.

Overall, it’s a very desirable and excellent step, involving mostly rate cuts and some simplification.

Also read: GST reform shows how indirect taxes can be made even more indirect

The objective of this exercise, as the Finance Minister has been saying, is to spur the economy by increasing consumption. The idea is that by lowering the rates, more people will buy goods, which in turn will result in better consumption, and therefore better tax mobilisation. This, in turn, is expected to help the economy pick up. Do you agree with this strategy?

See, I think you have to look at it in two ways. You're cutting a tax—this is a consumption tax cut. Then the question is: will it revive or boost the economy? And you have to separate that into two different questions: will it boost the economy in the short run, in terms of the cycle, and will it boost it permanently?

Now, there’s a bit of irony here. If you believe the official numbers, India is growing at 8 per cent. In that case, the economy doesn’t need any boost—because if you're growing at 8 per cent, this is the world’s best-performing economy by far. But of course, you can take a view on how many people actually believe the economy is growing at 8 per cent. Assuming it’s not, and that the economy does need some stimulus, then yes—a tax cut can help.

Remember, the boost to consumption will depend on the size of the tax cut, how much prices actually come down, and how consumers respond.

The rule of thumb is that when you do a tax cut—say, about half a percentage point of GDP—then GDP growth might increase by roughly half a percentage point. So if you have 7 per cent growth, it may become 7.5 per cent. These are estimates, of course—it could be a little less or a little more—but it’s not going to be huge. In the short run, the impact will be around that half a percentage point.

Is it going to permanently boost the economy? That’s very unlikely, because the problems are much deeper. Private investment is weak, the risks of doing business are high, and now there’s the Trump tariff shock. We also know there are many structural impediments. So in terms of permanent potential growth, this move will probably have only a marginal, maybe negligible, impact. In the short run, it will be something small.

There are economists who disagree. For instance, Swaminathan Aiyar has accused you of being a pessimist. He says you’re not believing the numbers. He points to 2023, when growth was 9.5 per cent in the September–December quarter. After that, the next three quarters came down—to 8.4 per cent, 6.5 per cent, and 5.6 per cent. But then the following three quarters rose again—from 6.4 per cent to 7.4 per cent to 7.8 per cent. His argument is that we are now in the zone of a “miracle economy”, which means growth at 7 per cent and above.

I want to ask you this: if you go to the common man and say we are in a “miracle economy”, how many will actually agree?

See, firstly, I think when looking at these very short-term numbers, one always has to be careful. Even with the latest figures—you don’t have to listen to me—lots of people are questioning the numbers and doubting them. The point is: does it feel like an economy that’s growing at 7–8 per cent?

It’s not that GST is bad; the overall atmosphere has changed. As Keynes said, when facts change, opinions must adapt. The GST negotiations of 2016–17 cannot be repeated under today’s conditions.

If you look at sales growth, for example, it’s very weak. Many high-frequency indicators are also not performing strongly. So, the question is not whether I’m optimistic or pessimistic—it’s whether we’re looking at the numbers over a broader period of time.

It is true that after COVID, for a year or two, the economy rebounded well. But look at private investment: it has been completely weak and stagnant for a long time. Just yesterday, Harish Damodaran wrote a piece on how capital inflows have collapsed. FDI is weak.

So I don’t want to get into a contest about which number is the “right” number. What I’m saying is that this is an economy that clearly needs a lot of reform and a boost through good policy measures.

Now, related to that—if GDP is growing strongly, then its constituents should also reflect that. And I’ve been saying this for a long time: consumption and investment are both relatively weak. Over the long run, the components have never really added up to the headline GDP growth.

Also read: Will the GST rejig revive consumption and boost growth?

Are you implying that the data itself is not reliable? That the figures being put out by the government of India cannot be fully trusted?

You know, I don’t want to get into that debate now—I’ve written on this extensively, and people should read that. It’s just that, as you yourself said, the headline numbers have never really tallied with the constituents or with a variety of other indicators.

Everyone also knows there are two or three obvious issues, which the government itself has admitted. For instance, there’s the double-deflation problem and inadequate deflators. These are expected to be corrected going forward. So I would put it this way: MoSPI (Ministry of Statistics and Program Implementation) is doing a commendable job in terms of future reform and in working to fix the data framework. Let’s focus on that.

Meanwhile, I think we should be concentrating on what the economy actually needs—structural improvements and reforms.

The government has estimated a revenue loss of about ₹48,000 crore based on FY25 consumption numbers, but you and Josh Felman wrote in The Economic Times that it could be closer to ₹2 lakh crore. Given that earlier attempts—through Budget tax breaks and RBI’s CRR move—failed to lift urban demand, if GST rate cuts don’t generate the additional revenue the Centre is banking on, won’t the FY26 fiscal deficit target of 4.4 per cent be at risk?

I think there’s a lot of confusion about the revenue loss numbers. For this year, the impact will only be for about five months—Navratri is in October—so the full-year effect won’t be felt. Also, some estimates focus only on GST loss, but the correct calculation should include both GST and the compensation cess. On an annualised basis—not just for this year—that loss will be around ₹1.5–2 lakh crore, roughly half a percentage point of GDP.

I want to ask you this: if you go to the common man and say we are in a “miracle economy”, how many will actually agree?

Now, remember, the boost to consumption comes precisely because there’s a revenue loss: taxes are cut, so prices are lower for consumers. But that means about half a percentage point of GDP loss for the overall kitty—GST plus cess—so the budget target this year will likely slip somewhat.

Here’s the tricky part: neither the Centre nor the states currently get revenue from the compensation cess, because it is being used to repay past borrowings. That’s why some argue there will be no loss to government revenues. But in aggregate, by definition, cutting taxes reduces the total kitty, and that has implications not only for this year but also more permanently for both Centre and states.

Can you list top five measures that, in your view, the government should take—either immediately or at least in the medium term?

I think the fundamental question is: why is private investment weak? The government has done a good job on public investment, building up infrastructure, and expanding welfare—whether it’s cooking gas, electricity, or other essential services. The social welfare state has broadly improved. But the heart of the matter is this: if everyone says growth is 8–9 per cent, why hasn’t private investment picked up?

Now, there’s never one single solution to a problem like this. And it’s not just about the Central government—state governments also have to act. For instance, the electricity sector badly needs reform, and that falls under the states. Manufacturing power tariffs are almost double what they should be, hurting international competitiveness. Unless states address that, private investment will remain constrained.

Also read: GST 2.0: Will 2 slabs simplify India's tax maze? | Talking Sense with Srini

At the Centre’s level, yes, a Deregulation Commission has been set up, and it may well come up with useful recommendations. But I’d like to stress a slightly different perspective: the “don’t do” list is just as important as the “do” list.

Josh Felman and I argued in a recent piece that the biggest impediment to private investment is not returns. Returns are actually quite good in India: the corporate tax rate has been cut, banks are ready to lend, and schemes like PLI have been rolled out. The real problem is the high risk of doing business. This affects both domestic and foreign investors.

Why are the risks so high? We identified a few factors. First, some groups are favoured over others—what we called the “national champion” risk. Second, there is arbitrary and weaponized enforcement, what many still call “tax terrorism”, especially with GST. Third, India has become more closed, with a proliferation of quality control orders that have undermined export competitiveness.

So the key questions are: is the playing field level between all investors? Is it level across states, or are some states favoured over others? These risks are what need to be tackled. Yes, issues like “Inspector Raj” remain important, but ultimately reducing the risks of doing business—and avoiding new distortions—is as critical as ticking off items on a reform “to-do” list.

Since you were part of the founding team along with Mr Jaitley, I remember listening to your lecture last year at the Madras Institute of Development Studies where you called GST “Arun Jaitley’s breakfast”.

Former Finance Minister P Chidambaram had said these GST reforms are eight years overdue—eight years late—even though he welcomed the reforms the finance minister has put forth. You were at the forefront of crafting the Jaitley policy and shaping the GST rollout in 2017. Why couldn’t India deliver a truly “good and simple tax”? Were there obstacles—political, economic, or both? What’s the inside story?

Well, there’s no “inside story” as such—I’ve written about all this in my book. As you know, I argued in my report that there should be only three rates, just three. Unfortunately, we couldn’t achieve that. And in India, whenever policy goals aren’t achieved, the reasons are almost always political.

It’s not just about the Central government—state governments also have to act. For instance, the electricity sector badly needs reform, and that falls under the states.

In this case, it was a combination of politics and bureaucratic pressures. The simple truth is that there just wasn’t enough commitment—either political or bureaucratic—for simplicity at that stage. The overriding concern was simply to get GST done, because many states were opposing it. To make it happen, compromises had to be made, and those compromises outweighed the commitment to careful design.

As a result, we ended up with the system we have today. But there’s no point crying over spilt milk. The government deserves credit for at least having implemented GST, and now the focus should be on moving forward—getting closer and closer to the ideal of a truly good and simple tax.

Also read: How Centre has bullied states over GST compensation over the years

When GST was introduced, states were assured 14 per cent annual revenue growth for five years and a share of service tax revenues. But many feel squeezed now, having given up much of their tax sovereignty. States like Kerala have protested, pointing not just to GST but also to central assistance and wider disparities among states.

How do you see this evolving, and what should states focus on going forward?

See, first of all, two or three things. One, because of the compensation guarantee that was given for the duration—five or six years—the states did very well. They got about, on average, half a percentage point of GDP, which was more than they were getting before. Of course, there are differences among states, which I won’t go into here, because it gets complicated. But overall, states benefited substantially during the compensation period.

Now, because the overall kitty is going to decline due to these tax cuts, both the Centre and the states are going to face a revenue shortfall. That’s point number one.

Point number two—and this is the bigger point—is that the original GST in 2016 was done in a true spirit of cooperative federalism under Mr Jaitley, to get the states on board. Many states, even Gujarat, Tamil Nadu, and Maharashtra, were initially unhappy with GST because they feared they would lose out. That’s why, in the spirit of cooperative federalism, we provided compensation, which the states benefited from.

But now, if rates are cut, revenues will decline for everyone. Both Centre and states are equally responsible for reducing the size of the overall kitty, and both will face the consequences.

Point number three is that cooperative federalism depends on a broader sense of shared objectives and, above all, trust between the Centre and the states. This trust is not just about GST; it’s about what’s happening more broadly. Some southern states, and others as well, feel that the spirit of cooperative federalism has weakened.

So when people ask what should be done or not done—raising or cutting rates—that’s only part of the picture. There has to be this broader spirit of cooperative federalism for GST to move forward. For example, I’ve been accused of being anti-GST because I don’t want petroleum included in it. But I am pro-GST. What has changed is the spirit of cooperative federalism. When that trust is weakened, asking states to give up more sovereignty becomes harder to accept.

It’s not that GST is bad; the overall atmosphere has changed. As (John Maynard) Keynes said, when facts change, opinions must adapt. The GST negotiations of 2016–17 cannot be repeated under today’s conditions. The key point is that reviving the broader spirit of cooperative federalism is essential if further reforms are to succeed.

I think what you’re saying is that the time isn’t right right now to bring petrol, electricity, and alcohol—particularly for states like Tamil Nadu—under the GST umbrella. It’s been eight years and we still don’t have a unified national market under GST.

What I mean is this: we certainly have a more common market than we did pre-GST. Borders are smoother, there’s less friction, and movement of goods has improved. So yes, it’s not a perfectly unified national market, but it is better than before.

The real question now is: to move further and make it even more of a common market, is there the underlying trust and spirit of cooperative federalism? You cannot force this progress if that foundational spirit isn’t there.