The South Asian nation’s gross domestic product grew at more than 8% this fiscal year compared with the previous year, driven by public spending on infrastructure, services growth, and an uptick in manufacturing. That would put India well ahead of China, which is growing at about 5%, and on track to hit Prime Minister Narendra Modi’s goal of becoming a developed nation by 2047.

But the way India calculates its gross domestic product can at times overstate the strength of growth, in part by underestimating the weakness in its massive informal economy. There are also other indicators, such as private consumption and investment, that are pointing to soft spots. Despite cuts to corporate taxes, companies don’t appear to be spending on expansions.

“If people were optimistic about the economy, they would invest more and consume more, neither of which is really happening,” said Arvind Subramanian, a senior fellow at the Peterson Institute for International Economics and former chief economic adviser to the Modi government.

Private consumption, the biggest contributor to GDP, grew at 4% this fiscal year, still slower than pre-pandemic levels. What’s more, economists say, it could have been even weaker if the government hadn’t continued its extensive food-subsidy program that began during the pandemic.

The problem is driven in part by how India emerged from the pandemic. Big businesses and people who are employed in India’s formal economy are generally doing well, but most Indians are in the informal sector or agriculture, and many of them lost work.

While India’s official data last year put unemployment at around 3%, economists also closely track data from private economic research firm, the Centre for Monitoring Indian Economy. It put unemployment at 8% for the year ended March.

At a small tea-and-cigarette stall in the southern city of Bengaluru, 55-year-old Ratnamma said many of her customers in the neighborhood, which once bustled with tech professionals and blue-collar workers, have moved out of the city and returned to rural villages. Some have come back, but she has fewer customers than she once did.

“Where did everyone go?” she said.

She makes about $12 in sales a day, she said, compared with as much as $100 on a good day in the past. It isn’t enough to cover her living expenses or repay a business loan she took out six months ago.

Economists say that the informal sector has been through three shocks in a decade—a 2016 policy aimed at tax evasion called “demonetization” that wiped out 90% of the value of India’s paper currency, a tax overhaul the following year that created more paperwork and expenses for small businesses, and the pandemic.

That has hurt employment in cities and contributed to millions of people going back to farms in recent years.

Dhiraj Nim, an economist with ANZ Research, said that a weak recovery in the number of monthly train journeys, an indicator of migration levels, shows that hasn’t fully reversed. “My guess is that overcrowding of farms persists even now,” he said.

India’s GDP measurement doesn’t fully capture the lingering weakness in the informal sector of the economy because it uses an index of firms in the formal sector to estimate informal activity. That could be a problem if it leads to government policies that aren’t adequately responding to what is actually happening in the economy.

“There is no doubt that a wide range of Indian economic indicators are pointing in a somewhat different direction to that suggested by the headline GDP numbers. But India is hardly an outlier in that regard,” said Udith Sikand, senior emerging markets analyst at Gavekal Research. “The far more important question is whether the inaccuracy of statistics is so systemic and pervasive that it is clouding the judgment of economic policymakers.”

“To put it differently,” he said, “the set of policies required to achieve and maintain trend growth of 8-10% is very different from the policies required when growth is trending at 4-6%.”

Another dynamic in India’s GDP-measurement process is also likely to be at play. In 2015, India switched to calculating GDP at market prices—the norm for most countries and international institutions—and then adjusting for changes in prices to arrive at the real GDP. The new methodology, a long-planned revision, was overall seen as an improvement in the measurement of the economy.

But because of the way India adjusts for prices, the methodology could overestimate GDP growth when prices of commodities decline but other prices remain high, according to Subramanian. The economist says that is what happened in the last financial year and believes growth could be overstated by about 2 percentage points.

“In the GDP calculation, because they are underestimating inflation they are overestimating real GDP growth,” said Subramanian.

The statistics ministry didn’t immediately reply to a request for comment. Economists say this issue self-corrects when commodity prices rise.

Still, there are signs that some of those hurt the most by the pandemic are recovering ground. A March report by Morgan Stanley suggested two-wheeler sales, an indicator of middle-class consumption, had returned to 2019 levels.

Nim, of ANZ Research, says that improved rural wage growth and increased availability of construction work means “the lower rungs of the economy are now doing better than a couple of years ago.”

And the infrastructure-building spree, along with improved law and order, has put larger northern Indian states, such as Uttar Pradesh, in a better position to participate in India’s efforts to attract manufacturers as they move out of China. More economic activity and jobs there would make a huge difference to India’s economy—and Modi’s goals—given a projected population of 240 million in Uttar Pradesh.

“Just the arithmetic weight is so huge, it would be a game-changer in many respects,” said Subramanian.