Getting A Measure Of GDP Numbers

Doubts are often raised by some economists over the veracity of India’s GDP estimates. This typically happens when GDP growth is high. But no such doubts are expressed when growth is slowing, as after the Q2 GDP print. Such asymmetric questioning of the veracity of GDP estimates — non-existent when growth is low and vociferous when growth is high — is typically reserved for India’s data.Now take into account quarterly estimates of Real GDP and Real Gross Domestic Income (GDI) for US over the last decade. GDI measures income that individuals and businesses make through wages and profits while GDP measures the value produced in the economy. To understand the key takeaway, note a fundamental tenet of economics: people’s income must equal the amount they spend or save. Similarly, for any economy, GDI and GDP should be identical with minor differences attributed to statistical discrepancies.Yet, this has not been the case in US for a couple of years. What’s happening in US is that the statistical discrepancy between Real GDP and Real GDI has widened over the last eight quarters and is now over $800bn. At 2.7% of GDP, this discrepancy is incredibly large as US’s GDP growth itself is only 3-4%. Such a large discrepancy, which is more than the entire GDP of about 150 countries, has never prevailed over the last 80 years.Going by the logic employed by Indian economists who use discrepancies in India to attribute India’s GDP growth as ‘fake’, would they label GDP growth in US as also ‘fake’? One working paper some years ago drew a conclusion that India’s GDP growth is overestimated by over 2.5%. To set up the necessary strawman, such papers introduce other countries as ‘purported’ apple-to-apple comparisons to implement a common technique in economic research called a difference-in-difference. In this method, economists compare data before and after a certain event, like the year 2022 when the statistical discrepancy started manifesting in US, with similar changes in other countries that did not experience the same event. The critical assumption underlying this method is that changes in other countries represent the counter-factual, ie, the pattern that would manifest without the event. Using this method, economists can hit the headlines reporting a finding that US GDP growth is “overestimated” by 2.7%.But a key nuance is glossed over. Countries differ substantially in measurable and, critically, unmeasurable dimensions. So, when comparing across countries, rigorous econometrics is required to ensure that the comparison is indeed apples-to-apples. Top academic journals set high standards for the quality of econometrics employed in any cross-country study. If such a comparison is not apples-to-apples, methodological flaws would only get exposed if the flaky research were presented in an academic seminar or underwent peer review in a high-quality journal.What is then the point of such inferences from India’s data? They are trumpeted by former finance ministers and other politicians to serve their political agenda. Such narrative inflicts material damage on the assiduously developed reputations of statistical institutions, one soaked in the blood and sweat of several generations of patriotic and capable statisticians. In the opportunistic play made under the full knowledge that the media grabs negative headlines, the larger national interest can be damned.In sum, note three key takeaways. First, large statistical discrepancies in GDP numbers can manifest even for a mature economy like US. Therefore, when we jump to attribute political colour to such statistical exercises in an emerging economy like ours, the larger national interest gets significantly damaged. Rather than painting politically motivated narratives, our efforts should be to improve the quality of economic statistics.Second, when GDP growth will be high next quarter and the same compulsive pessimists will be back questioning the veracity of GDP estimates, remember to separate the messenger from the message. To use a cricket parallel, play the ball and not the bowler because even a Jasprit Bumrah can bowl bad deliveries. Do not forget that the politically coloured narratives peddled on India’s economic statistics were rank full tosses to be hit out of the park.By remaining vigilant to fake narratives, the truth does prevail. Even a ray of light is enough to dispel all the darkness in a room. Similarly, the darkness created by such ‘storytelling’ on ‘fake’ GDP growth cannot withstand the light brought by collective vigilance.The author is Executive Director (India) at IMF. Views are personal.