GDP Figures: The Politics of Economics

Above: Unemployed youth at a job fair in Assam. Unemployment has increased to an all-time high of 6.1 percent in 2018/Photo: UNI

An exclusive research paper for India Legal questions the legitimacy of the Modi government’s GDP figures. This is reflected in decreased economic activity in other spheres such as steel, cement, coal, electricity and oil. Loss of faith in government data can have serious repercussions for future growth

By Sanjiv Bhatia

“A single lie discovered is enough to create doubt in every truth expressed.”

Recently there has been much debate about the Indian economy. Surveys conducted by government agencies such as the National Sample Survey Office (NSSO) and data collected by reputed private agencies like the Centre for Monitoring the India Economy (CMIE), show a dramatic uptick in unemployment over the last four years.

The NSSO survey shows that unemployment increased from 2.2 percent in 2011-12 to an all-time high of 6.1 percent in 2018, with the highest unemployment among youth between the ages of 15 and 29. The CMIE data shows that the number of people employed dropped by 11 million—from 408 million in 2017 to 397 million in 2018. Additionally, the number of youth who are “not in education, employment or training (NEET)” increased from 70 million in 2004-5 to 116 million in 2017-18 with 70 percent of this increase coming in the past two years. Rural distress and massive disruption in the informal economy after demonetisation accentuated the unemployment problem.

Faced with the evidence that there are more people unemployed and underemployed now than when he took over as PM, Narendra Modi came up with the argument that the labour data was incorrect and posed the following question: How can an economy growing at 7 percent not produce jobs? That seems like a reasonable argument except when flipped on its head: How can an economy that has lost almost 20 million jobs be growing at 7 percent? Is the Indian economy really growing at 7 percent or is it a misrepresentation? Do the high unemployment numbers cast doubt on India’s economic growth?

This issue becomes all the more relevant in light of recent accusations that the government is suppressing weak employment data and manipulating GDP numbers to make the economy appear better. In a bizarre sequence of events, the government first revised 2016-17 GDP numbers downwards to make the GDP for 2017-18 look better by comparison (GDP growth is calculated relative to the previous year), and then six months later, in a complete turnaround, revised the 2016-17 GDP upwards to 8.2 percent. This made the demonetisation year the best year for the economy in a decade. Within a period of six months, the GDP for 2016-17 mysteriously changed by almost 2 percent (200 basis points) first down and then upwards. Believe it or not.

Recently, Niti Aayog, which is run by government-appointed bureaucrats with no expertise in national income accounting, rejiggered the GDP numbers going all the way back to 2005. They lowered the average annual GDP growth for the Singh government (2005-2014), from 8.2 percent to 6.8 percent, and simultaneously raised the average growth rate of the Modi years from 6.2 percent to 7.3 percent. This juggling allowed the Modi government to crow that it was a better steward of the economy than the Singh government.

This attempt to rearrange the data was so blatant that a group of 108 pro­minent Indian and foreign economists wrote to Modi in March and raised concerns over “political interference” in the reporting of statistical data and cast doubt on the government’s achievements because of the revision and suppression of data. The economists warned that tampering with data could have serious consequences not only for internal policymaking but for external investments.

An economy’s growth, as measured by GDP, comes from the production of goods and services. There are many problems with measuring GDP, especially in an economy like India’s where 85 percent of the people work in what is known as the “informal” sector. As record keeping is lax in this sector, it makes it difficult to generate reliable estimates on the output generated. But given the number of people that work in the informal sector, it can be safely assumed that a substantial portion of the country’s GDP comes from this sector.

Empirical estimates are that the informal economy contributes roughly one-third to one-half of the nation’s GDP. A recent study by WIEGO estimates the contribution of the informal sector to the economy to be about 46 percent of non-agricultural GVA (gross value added). Since calculating production from the informal sector is hard, GDP statisticians impute to it a value equal to roughly half of the formal sector. So, if the formal economy produces Rs 100 worth of goods and services, the total GDP is estimated to be Rs 150—where Rs 50 is the imputed output of the informal sector.

It is easy to see how the real GDP could be higher or lower depending on the variance of the actual and the imputed contributions of the informal sector. An economy growing at 7 percent could in reality be growing much slower if the performance of the informal sector is lower than the value imputed in the GDP calculation. We know that demonetisation disproportionately affected the informal economy, so a failure to lower the imputed value of this sector could well lead to an overestimation of GDP growth.

Errors in estimating GDP are not limited to informal economies such as India. Countries ruled by an authoritarian regime often misreport economic data to appear more successful and stable than they are. In the ’50s and ’60s, estimates of the Russian economy were so inflated that many economists started to believe that socialist economic systems were better at generating growth than capitalist systems. A good contemporary example of data manipulation is China. The government officially reported 10 percent plus GDP growth for 14 consecutive years, and everyone believed those numbers till recently when a group of analysts and a Chicago professor, using entirely different methodology, concluded that China’s GDP had been overestimated by between 30 percent to 35 percent.

This confirms what many analysts, economists and even some high-level Chinese officials have said for years—China’s economic data are cooked and can’t be trusted. In a secret diplomatic cable (released by Wikileaks), Chinese Premier Li Keqiang called China’s GDP numbers “man-made” and “unreliable”. He admitted that he ignores those numbers—focusing instead on more tangibly measured things such as bank loans, rail freight and electricity output.

In this research, I use a similar approach and focus on other proxies of economic activity for the Indian economy. A growing economy will need increasing amounts of inputs like steel, cement, coal, electricity, oil. It will also require higher amounts of capital and will involve a greater movement of goods. GDP growth should, therefore, be reflected in a proportional growth in these input factors. Any discrepancy between the two suggests a misreporting of GDP.

I examine 21 different indicators of economic activity to answer two questions: (1) does the data support the Modi government’s assertions on GDP growth and (2) what does the underlying data reveal about economic growth under the Singh (2004-2014) and the Modi (2014-present) governments. Let’s examine the data.

As an economy grows, there is an increase in demand for goods and factories start using a higher percentage of their installed production capacity to produce more goods. Capacity utilisation is a key indicator of growth in the economy. Data from the Reserve Bank of India shows that aggregate capacity utilisation declined from a peak of 82 percent in 2011 to 73 percent in 2018 (Figure 1). This decline is not consistent with a growing economy, especially one purported to be the “fastest growing economy in the world” growing at 7.3 percent.

Further evidence of declining demand in the post-2014 period is the reduction in capacity utilisation in two key sectors—steel (Figure 2) and cement (Figure 3). Clearly, this data is incompatible with a growing economy.

Table 1 gives the annual growth rate (CAGR) of key economic variables for two different periods, the 2004-2014 representing the Singh government and the 2014-2018 period representing the Modi government. First, some observations about growth in the post-2014 period (data in column 3 of Table 1).

• None of the input variables in the post-2014 period grew at a rate that would be supportive of an economy growing at 7.3 percent. Electricity consumption grew at only 4 percent, ce­ment consumption at a mere 3.6 percent, steel by 5.6 percent, coal by 5.4 percent and mining activity at 2.6 percent. If it takes two yards of cloth to make one shirt and if 10 yards of cloth were used one has to conclude that only five shirts were made and not 20.
• Infrastructure output which consists of eight core sectors—coal, cement, fertilisers, steel, natural gas, refinery, electricity and crude oil—grew at only 4.1 percent per year from 2014 to 2018. This too is incompatible with an economy growing at 7.3 percent as infrastructure output comprises about 40 percent of the country’s industrial production.
• If goods are produced, they have to be moved. During the post-2014 period, rail freight grew at an annual rate of only 2.9 percent and road freight by 5.7 percent—both indicating slower economic growth than claimed.
• Capital is a critical factor in economic growth. Bank credit to the private sector grew at a measly 4.8 percent during the 2014-18 period. This was the slowest rate of credit growth in 20 years. It is virtually impossible for an economy to grow at 7.3 percent with such poor growth in bank credit.

One can see a clear pattern emerging here. The numbers just don’t add up. And data on manufacturing activity supports that contention.

• From 2014-2018, manufacturing activity grew by only 3.9 percent annually and production of motor vehicles by a mere 4.6 percent—the slowest growth in 15 years (Figure 4, facing page). These numbers do not reflect an economy growth of 7.3 percent.
• Modi travelled to 93 countries and spent $500 million on those travels, yet between 2014 and 2018, India’s exports fell at an annual rate of 2.9 percent making this the first government in history under whose watch exports dropped. That this happened despite the rupee dropping almost 12 percent (a lower rupee helps exports) is indeed alarming. Exports make up about 11 percent of the country’s GDP, and it is hard to comprehend how the economy could have grown at 7.3 percent with declining exports.

Let’s now compare the state of the economy under the Singh and Modi governments by comparing the numbers in columns 2 and 3 of Table 1. The following observations stand out:

• Each of the 21 proxies of economic activity show a significantly higher growth rate during the Singh administration (shaded area in the Table 1) compared to the Modi government. Niti Aayog can rejig the numbers any way it wants, but the underlying data tells the real story—the economy grew much faster during the Singh administration than the Modi one.
• All the inputs required to produce goods and services—steel, cement, coal, electricity, oil, mining, capital—grew at higher rates during the Singh administration. More goods were produced and moved from 2004-2014 than any time in the country’s history. Rail freight grew at an annual rate of 8 percent from 2004-2014 compared to only 2.9 percent in the post-2014 period, and road freight grew at twice the rate under Singh than under Modi (11 percent versus 5.7 percent). If more was produced and shipped, then GDP would have been higher during the Singh administration.
• Foreign Direct Investment (FDI), which many have touted as a significant achievement of the Modi government, averaged only 1.9 percent of GDP. In contrast, under the Singh government, FDI inflows were twice as much at 4.1 percent of GDP.
• Gross Fixed Capital Formation, which is an indicator of investment in fixed assets, grew at almost twice the rate during the Singh administration compared to Modi—14 percent versus 8.6 percent. Figure 5 (next page) shows that investment as a per cent of GDP averaged 33.5 percent during the Singh administration compared to 28.9 percent during the Modi years.
• Credit growth averaged 14.4 percent during the Singh administration, more than twice the 6.2 percent growth under the Modi government (Figure 6).
• Tourism, which contributes about nine per cent to GDP grew more than twice as fast (18.2 percent increase annually) during the Singh government than under Modi (7.8 percent).
• Exports increased by 14.1 percent every year during the 10 years of the Singh government. In contrast, under Modi, exports declined at around 3 percent a year.
• The Business Expectations Index reflects how businesses feel about the economy. The average value of the index was 20 percent higher under the Singh government (despite the 2008 global economic meltdown), compared to the Modi administration (Figure 7).
• Indian companies made record profits during the Singh administration. Earnings per share (EPS) of the Nifty 50 companies increased at an annual rate of 12.1 percent during the Singh years, almost five times higher than the meagre 2.6 percent per year increase under Modi. Corporate profits as a percent of GDP dropped to a 15-year low of 3 percent in 2018.
• The debt-service ratio shows the ability of companies to pay back their current debt from earnings. During good economic times, corporate earnings increase and businesses find it easier to service their debt. From 2004-2014, the debt-service ratio doubled from 4.5 to 9, but by 2018, it had dropped to 7 (Figure 8). This partly explains why bad loans have increased in the past four years.
• The Indian stock market (Nifty 500) increased by almost 500 percent under the Singh government, an annual percentage increase of 15.5 percent. In contrast, the stock market rose by only 7.7 percent a year during the Modi years. A rising stock market is a good indicator of an expanding economy and the results indicate that investors felt twice as good about the economy under Singh than under Modi.
• The performance of the Singh government is even more remarkable given that there was a global economic crisis in 2008 and the average crude oil price between 2004 and 2014 was $81 per barrel. In contrast, the Modi government enjoyed significant tailwinds: the best global economy in 30 years, and an average crude oil price of $60 per barrel.
• There is no evidence of productivity gains to explain the difference in input factor usage between the two regimes.

In a recent article (http://www.indialegallive.com/constitutional-law-news/economy-news/gdp-figures-gaming-the-system-58141), I compared the performance of the Singh and Modi governments using the national income accounting identity in which GDP growth is the sum of consumption, investment, government spending and net exports. I document that each of the components grew at a faster pace during the 2004-2014 period compared to the post-2014 period.

The evidence is quite clear. Fewer raw materials were used, fewer goods and services produced, fewer items moved around and even fewer goods exported in the Modi years compared to the Singh period. Yet, mysteriously, GDP grew at a faster pace under Modi than under the Singh government.

Based on all this evidence, it is impossible to conclude anything but the obvious: The Singh government did a better job handling the economy than the Modi government. The facts are irrefutable and Niti Aayog’s attempt to rearrange GDP numbers is a crude attempt to distort the underlying truth.

The real barrier to India’s economic growth is the small group of people that sit in Delhi and make decisions affecting the entire economy. That’s bad enough, but that they live in fool’s paradise about the actual growth because of fudged data is indeed scary. It’s hard to provide meaningful solutions if you don’t accept there is a problem.

In a recent interview, ex-RBI governor Raghuram Rajan was asked how it was possible for an economy growing at 7 percent to produce no jobs. His response: “One possibility is that the economy is not growing at 7 percent.” The evidence presented here supports Rajan’s assertion.

In a recent Reuter’s poll of leading global economists, more than half the respondents admitted that they had lost faith in the official statistics put out by the Indian government. The data shows that this fear is not unfounded. The GDP numbers have clearly been misreported. Fabricated data not only hampers policy making, but also foreign investment. Foreign investors are sceptical about investing in countries where governments cannot be trusted. The loss of faith in the Indian government’s ability to provide reliable economic data could have serious repercussions for India’s future growth.

It is unclear who misrepresented the data, but it should be investigated as a serious crime because it leads to questions about the integrity of an entire nation. A nation’s economic growth comes from the toil and sweat of its citizens, not the government. Distorting economic data is a slap on the hard work of India’s millions. Corruption pertains not just to money, but also to the abuse of entrusted power to misuse and misstate facts.

—The writer is a Financial Economist and Founder, contractwithindia.com