Opinion: India’s GDP puzzle — When statistics grow faster than the economy

If measurement itself becomes a subject of debate, the informational foundations of economic policymaking inevitably weaken 

By Pendyala Mangala Devi

Economic statistics occupy a central position in modern policy debates. Quarterly GDP figures, fiscal deficit ratios and inflation indices are often presented with mathematical precision, creating the impression that they represent objective measurements of economic reality. Yet economists have long recognised that national income statistics are not direct observations, but constructed estimates built upon assumptions, proxies and statistical conventions.

Measurement, Growth and the Limits of GDP

The intellectual foundations of modern national income accounting were laid by Simon Kuznets, whose pioneering work earned him the Nobel Prize. Kuznets himself warned against treating GDP as a complete indicator of economic welfare, famously observing: “The welfare of a nation can scarcely be inferred from a measurement of national income.”

This caution highlights a fundamental principle of macroeconomic analysis: economic measurement is always an approximation of reality rather than reality itself. The challenge becomes particularly complex in large developing economies where vast segments of production occur outside formal statistical systems.

These issues have gained renewed prominence following the publication of the working paper “India’s 20 Years of GDP Misestimation: New Evidence”, authored by Abhishek Anand, Josh Felman and Arvind Subramanian at the Peterson Institute for International Economics.

The authors argue that while India’s growth during the mid-2000s investment boom may have been underestimated, the period after 2012 may have experienced systematic overestimation of economic growth. Their recalibration suggests three striking implications:

• Average annual growth between 2012 and 2023 may have been closer to 4–4.5 per cent, rather than the approximately 6 per cent reported in official statistics.
• The cumulative effect could imply that India’s GDP level by 2025 may be overstated by roughly 22 per cent.
• Real consumption may be overstated by nearly 31 per cent, suggesting that household welfare could be weaker than implied by headline growth figures.

If even partially accurate, these findings raise important questions about how India’s economic performance over the past decade should be interpreted.

2015 GDP Revision, Statistical Puzzle

The immediate origins of the debate lie in the 2015 revision of India’s GDP methodology, when the base year for national accounts was shifted to 2011-12 and several statistical adjustments were introduced.
These included:

• Greater use of corporate financial data from the MCA-21 database.
• Alignment with international accounting standards under the United Nations System of National Accounts.
• Expanded coverage of the formal corporate sector.

While these reforms were intended to modernise India’s statistical system, they produced an unexpected puzzle. GDP growth appeared strong even when several macroeconomic indicators suggested a weaker economic environment. Among the indicators showing divergence were:

• Bank credit growth, which slowed sharply after 2012.
• Private investment, which declined due to corporate balance-sheet stress.
• Export performance, which stagnated during several years of the decade.
• Industrial capacity utilisation, which remained relatively moderate.

This divergence between robust statistical growth and subdued macroeconomic signals created what economists might describe as a measurement paradox.

The Nobel laureate Robert Solow once remarked that the computer revolution could be seen everywhere except in productivity statistics. In India’s case, the puzzle appeared reversed: rapid growth appeared clearly in the statistics but less visibly in the broader economic landscape.

Informal Economies, Constraints

A central challenge in measuring India’s economy arises from the scale of its informal sector. Millions of enterprises operate outside formal regulatory frameworks — including small retail shops, repair services, transport operators and household manufacturing units. These enterprises account for a large share of employment but often leave limited statistical records.

Nobel laureate Paul Samuelson emphasised that macroeconomic measurement requires constant methodological scrutiny, because statistical frameworks can generate misleading signals when applied to complex economies

Statistical agencies, therefore, estimate informal sector output indirectly by extrapolating trends observed in the formal corporate sector. Such methods may produce reasonable approximations under stable conditions. However, they become problematic when the two sectors experience different economic shocks.

Over the past decade, several shocks disproportionately affected informal enterprises:

• Demonetisation in 2016, which disrupted cash-dependent transactions.
• The introduction of the Goods and Services Tax in 2017, which imposed compliance costs on smaller firms.
• The Covid-19 pandemic, which severely disrupted informal services and labour markets.

Large corporations often possessed the financial and technological capacity to adapt. Informal enterprises did not. The Nobel laureate Angus Deaton has repeatedly emphasised that the greatest measurement challenges in developing economies arise precisely in sectors where statistical visibility is weakest.

Similarly, George Akerlof demonstrated through his theory of information asymmetry that imperfect information can distort economic outcomes. When statistical systems lack reliable information about informal sector dynamics, macroeconomic aggregates themselves can become distorted.

Deflators, Relative Prices

Another issue highlighted in the working paper concerns the construction of price deflators, which are used to convert nominal GDP into real GDP. Real growth estimates depend critically on the accuracy of these price indices. If the deflators used do not reflect the actual prices of final goods and services, the resulting estimates of real output may become distorted.

The study suggests that some deflators used in India’s national accounts rely heavily on input cost indices, including global commodity prices. During the mid-2010s, international oil prices fell sharply. Lower energy costs reduced production expenses and increased corporate profit margins.

However, when statistical deflators interpret falling input costs as falling output prices, part of the increase in profits is recorded statistically as higher real output.

The sequence can therefore unfold as follows:

• Falling global commodity prices reduce production costs.
• Corporate profits rise.
• Deflators interpret lower costs as falling output prices.
• Profit increases appear in national accounts as higher real GDP.

In effect, cost reductions can be statistically transformed into growth. The Nobel laureate Paul Samuelson frequently emphasised that macroeconomic measurement requires constant methodological scrutiny, because statistical frameworks can generate misleading signals when applied to complex economies.

Reconciling Macroeconomic Puzzles

If the adjustments proposed in the working paper are applied, several puzzles in India’s macroeconomic data become easier to interpret.

Investment Slowdown
Private investment declined sharply after the early 2010s as corporate leverage increased and the banking system accumulated non-performing assets. Growth rates closer to 4 per cent align more closely with the observed slowdown in capital formation. The Nobel laureate Edmund Phelps emphasised that long-term economic dynamism depends critically on sustained entrepreneurial investment and innovation.

Employment Challenges

India’s labour market has struggled to generate sufficient employment opportunities for its expanding workforce. Manufacturing employment has stagnated, while labour force participation rates have declined in several regions. These labour market trends appear inconsistent with a narrative of sustained rapid growth. The Nobel laureate Amartya Sen has long argued that development should be evaluated not merely through aggregate output but through the expansion of human capabilities.

As Sen wrote in Development as Freedom: “The success of an economy should be judged not merely by the growth of income but by the expansion of the freedoms that people enjoy.” If growth does not translate into employment opportunities and improvements in living standards, the strength of GDP figures alone becomes a weak indicator of development progress.

Capacity Utilisation

Industrial surveys consistently report moderate levels of capacity utilisation across manufacturing sectors. Factories operating below full capacity indicate subdued demand conditions — patterns that are difficult to reconcile with rapid growth but consistent with lower growth estimates.

Information, Expectations, and Statistical Credibility

Economic statistics play a critical role in shaping expectations within financial markets. The Nobel laureate Kenneth Arrow emphasised the importance of reliable information for rational decision-making, while Friedrich Hayek argued that economic systems depend upon the efficient transmission of information across markets.

When statistical credibility becomes contested, uncertainty spreads across both markets and policy debates. The Nobel laureate Joseph Stiglitz has been among the most prominent critics of excessive reliance on GDP as a measure of economic progress. He has argued that: “GDP tells us little about sustainability, inequality, or the well-being of citizens.”

Such critiques underline the broader concern that statistical aggregates can sometimes obscure rather than illuminate underlying economic realities.

Similarly, Milton Friedman emphasised the importance of credible economic data for effective policymaking. As Friedman observed: “The role of government statistics is not merely to measure the economy, but to provide reliable information upon which rational policy decisions can be made.” When measurement itself becomes a subject of debate, the informational foundations of economic policymaking inevitably weaken.

Growth Theory and the Importance of Measurement

Modern growth theory emphasises productivity, technological innovation and human capital accumulation as the principal drivers of long-term economic expansion. The Nobel laureate Robert Lucas famously argued that even small differences in growth rates can accumulate dramatically over time.

Similarly, the Nobel laureate William Nordhaus demonstrated how technological change and energy systems influence long-term growth trajectories. Yet understanding these processes requires reliable measurement. When measurement errors occur, they can distort both economic narratives and theoretical interpretations of development.

Rethinking India’s Growth Trajectory

If the estimates proposed in the working paper are broadly correct, India’s economic trajectory over the past two decades may require reinterpretation. Instead of a smooth and uninterrupted expansion, the economy may have experienced a more cyclical pattern:

• A strong boom during the mid-2000s investment surge, driven by infrastructure expansion and global capital flows.
• A structural slowdown after the global financial crisis, as corporate leverage and banking sector stress weakened investment momentum.
• Additional disruptions during the late 2010s, including demonetisation, GST transition costs and the economic shock triggered by the COVID-19 pandemic.

Such a trajectory would resemble the growth patterns observed in many emerging economies. In such contexts, the accuracy of statistical measurement becomes critically important. Growth statistics do not merely describe economic performance; they shape policy decisions, influence investor expectations and frame public debates about development.

Economic narratives built on flawed measurements can persist for years before being corrected. But when they are eventually revised, they often force policymakers, economists and the public alike to reconsider the assumptions that shaped earlier debates.

As economists have long understood, growth is not merely an economic phenomenon, it is also a statistical construct. And when the statistical foundations shift, the narrative of development must inevitably shift with them.