Fudging of data by government authorities

The central government of recent has come under sharp criticism by the opposition and economists for suppressing and fiddling with key economic indicators. In a combined statement, 108 senior economists and social scientists from across the world have voiced their concerns against central government’s attempt at fudging important statistical numbers. Their statement reads, “This is the time for all professional economists, statisticians, independent researchers in policy – regardless of their political and ideological leanings – to come together to raise their voice against the tendency to suppress uncomfortable data…”.

Economists quote the discrepancy in GDP numbers and withholding of Periodic Labour Force Survey (PLFS) released by NSSO as two important instances where there is an outright interference by the central government. Interestingly, this statement met with a reply from 131 Chartered Accounts who came in support of the government, calling economists concerns as “bogus” and “choreographed”[1]. This has driven the public discourse into a blame-game between “economists” and “CA” and a question of who is siding with whom.

Instead, what we should have been focussing on is the impact manipulating the data in a certain way could have on India’s economic standing. Does changing these numbers hide critical information which might negatively affect India’s policy making abilities? This article is written with the aim of providing the reader with enough background on what this debate is actually about and what do the numbers presented by economists and CAs mean for the country.

In 2015, the NDA government changed the base year for calculation of real GDP (GDP at constant prices) from 2004-05 to 2011-12. The change of base-year was a welcome move, among other benefits, it allowed us to compare the growth rates and data of newer sectors in economy based on recent data. However, a change in base year also meant that we do not have growth numbers for previous years calculated with the new base year of 2011-12 which rendered comparisons between the current year growth rates and previous year growth rates impossible.

The government was expected to release back-series data with previous year growth rates calculated with the new base year, however, the numbers only came out as late as 2018. Central Statistics Office (CSO) reportedly did calculate back-series data but the report was allegedly rejected by Niti Aayog because it showed an upward revision of growth rates for UPA[2].

Finally, in 2018 two back-series data were released. One was prepared by National Statistical Commission (NSC) and another by CSO, which had now trimmed the growth under UPA substantially. The NSC report was rejected by the Niti Aayog since it also showed improved growth rates during UPA’s tenure but the revised data by CSO was accepted by Niti Aayog and released in November 2018 (Fig.1).

The data raised many eyebrows since it showed a very high growth rate of GDP in 2016-17 at 7.1% which was the year of demonetisation.  However, as it turned out, the average growth rate under UPA still remained above the average growth rate under NDA. This once again necessitated the government to make a change in the GDP growth data. On 31st January 2019, a day before the final budget was presented by NDA, CSO revised its data again, further increasing the growth numbers under NDA. Fig.1 plots the results of both the November and the January report.

Growth rate in the year of demonetisation was projected to be 8.2%, the highest since 2011. According to IMF, India was expected to grow at a rate of around 6.6% after the announcement of demonetisation, however, government released data showed a growth of 8.2%. The difference was so astounding that it made this data seem absurd and confusing. The result was that more experts were now sceptical of government released data.

Source: Created using CSO reports released by MoSPI[3] [Note that November data has been removed from MoSPI site but the report can be found at http://pibphoto.nic.in/documents/rlink/2018/nov/p2018112801.pdf]

Keeping aside the political motivations behind squashing certain reports and tampering with others, even if we were to assume this data to be correct, the GDP numbers were not in tandem with other key economic indicators. The investment rate fell from around 34.7% in UPA’s tenure to 28% during NDA’s tenure. The agrarian distress is at its worst. Industrial production slowdown which began in 2007-08 continues and the banking credit in the non-food sector has also shrunk. If our GDP was growing at such an impressive rate, it would have translated into on-ground results. However, as Prasanna Mohanty demonstrates, this clearly hasn’t happened[4].

This discrepancy means that that our policy decisions might actually be based on incorrect statistics. A prudent government would have considered a complete picture and based their policy decisions accordingly, but a skewed and unrepresentative picture is likely to further deepen our economic troubles.

Consider the case of unemployment. According to the leaked Periodic Labour Force Survey (PLFS) conducted by NSSO– official release of which has been withheld by NDA– the unemployment rate of India is the worst it has ever been in 45 years at 6.1%[5]. The NDA government is so determined to prevent unemployment numbers from releasing that they have essentially just set aside an important figure which could have helped the government in better policy decisions.

Instead of the PLFS report, government is now releasing irrelevant numbers to show falling unemployment. The government released numbers showing increase in Employees’ Provident Fund Organisation (EPFO) enrolment to claim that these reflect an increase in number of jobs created[6]. This is not true. EPFO numbers could increase even with the same size of the labour market if more workers are choosing to go for EPFO option than not. While, EPFO can be regarded as a good proxy to check formalisation of an economy, it does not say anything about the job creation. Even the chief statistician of India, Pravin Srivastava has himself criticised government’s move to use EPFO numbers as proxy for job creation.

The government may succeed in hiding uncomfortable data but there would inevitably be a disastrous impact of ignoring the unemployment crisis that has presented itself in form of the PLFS survey and rising number of youth protests around the country today. To see what I mean, let’s compare the labour force participation numbers to unemployment numbers.

Data released by CMIE, a reputed private think-tank, shows a rise in unemployment rate from 5% in February 2017 to 7.2% in February 2019 and at the same time a fall in labour force participation rate (given by proportion of people willing and able to work out of the total working age population) from 44.6% in February 2017 to 42.8% in February 2019 (Fig.2 & Fig.3)[7]. Note that unemployment rate is calculated as the proportion of people employed out of the labour force and not the total working age population. So, people who are not in the labour force are not counted as unemployed. Still, even with the fall in labour force participation, unemployment continues to increase.

This reflects on a deeper ailment with the Indian Labour markets. Less and less people are making up the labour force, yet a greater proportion out of the labour force is not finding jobs. The unemployment crisis is far worse than what 7.2% unemployment rate tells us. Consider employment numbers. CMIE estimates employed population in February 2019 to be 400 million which is 7.5 million less than the workforce in February 2019. The employed labour has actually fell by 7.5 million in 2 years. Not only are no jobs being created but also the existing labour is now moving out of employment. A 7-8% growth rate in GDP cannot be sustained along with such a drastic fall in employment numbers.

Source: Created using CMIE data

In a growing economy like India, less and less people willing to enter the job markets and a greater proportion out of these not getting jobs is extremely worrisome. If policy experts do not respond to these numbers, then the demographic dividend that India currently possesses will soon go to waste and a jobless growth would eventually stagnate the economy.

The unemployment numbers are only a small part of what hiding important statistical numbers could mean for country’s policy making abilities. The debate around data manipulation is far more important than the political battle between UPA and the NDA. It is about sustaining the growth of our country while keeping a check on all important economic indicators. Politically motivated data, without a doubt, hides critical details and impacts our ability to take better policy decisions. It is therefore best if political parties maintain their distance from important central statistical institutes of the country to keep their integrity intact and ensure that subsequent policies are not misinformed by biased data.

BIBLIOGRAPHY

“131 Chartered Accountants Respond To 108 Economists Who Raised Data Integrity Concerns.” Bloomberg Quint, 18 Mar. 2019, www.bloombergquint.com/politics/131-chartered-accountants-question-motive-of-108-economists-who-raised-data-integrity-concerns#gs.31bjq0.

“EPFO, NPS Data Show 2.2 Million Formal Jobs Added in 6 Months.” The Economic Times, 26 Apr. 2018, economictimes.indiatimes.com/news/economy/indicators/india-shows-jobs-growth-as-3-11-million-join-social-security-fund-in-six-months/articleshow/63912079.cms.

Jha, Somesh. “Unemployment Rate at Four-Decade High of 6.1% in 2017-18: NSSO Survey.” Business Standard, Business-Standard, 30 Jan. 2019, www.business-standard.com/article/economy-policy/unemployment-rate-at-five-decade-high-of-6-1-in-2017-18-nsso-survey-119013100053_1.html.

Mohanty, Prasanna. “’Jobs – Reality Check’: No Data on Unemployment Means No Attention, No Remedy.” Business Today, 29 Mar. 2019, www.businesstoday.in/current/economy-politics/data-fudging-no-data-on-unemployment-means-no-attention-no-remedy/story/331893.html.

Mohanty, Prassana. “Data Fudging: Dressing up GDP and Budget Numbers Does No Good to Economy.” Business Today, 28 Mar. 2019, www.businesstoday.in/current/economy-politics/data-fudging-dressing-up-gdp-and-budget-numbers-does-no-good-to-economy/story/331960.html.

Vyas, Mahesh. “Employment Rate Falls in February 2019.” CMIE, Feb. 2019, www.cmie.com/kommon/bin/sr.php?kall=warticle&dt=2019-03-05%2B09%3A32%3A41&msec=496

[1] “131 Chartered Accountants Respond To 108 Economists Who Raised Data Integrity Concerns.” Bloomberg Quint, 18 Mar. 2019, www.bloombergquint.com/politics/131-chartered-accountants-question-motive-of-108-economists-who-raised-data-integrity-concerns#gs.31bjq0.

[2] Mohanty, Prasanna. “Data Fudging: Dressing up GDP and Budget Numbers Does No Good to Economy.” Business Today, 28 Mar. 2019, www.businesstoday.in/current/economy-politics/data-fudging-dressing-up-gdp-and-budget-numbers-does-no-good-to-economy/story/331960.html.

[3] “Annual and Quarterly Estimates of GDP at constant prices, 2011-12 series.” Ministry of Statistics Programme Implementation, 31 January 2019, http://mospi.nic.in/data

[4] Mohanty, Prasanna. “Data Fudging: Dressing up GDP and Budget Numbers Does No Good to Economy.” Business Today, 28 Mar. 2019, www.businesstoday.in/current/economy-politics/data-fudging-dressing-up-gdp-and-budget-numbers-does-no-good-to-economy/story/331960.html.

[5] Jha, Somesh. “Unemployment Rate at Four-Decade High of 6.1% in 2017-18: NSSO Survey.” Business Standard, Business-Standard, 30 Jan. 2019, www.business-standard.com/article/economy-policy/unemployment-rate-at-five-decade-high-of-6-1-in-2017-18-nsso-survey-119013100053_1.html.

[6] “EPFO, NPS Data Show 2.2 Million Formal Jobs Added in 6 Months.” The Economic Times, 26 Apr. 2018, economictimes.indiatimes.com/news/economy/indicators/india-shows-jobs-growth-as-3-11-million-join-social-security-fund-in-six-months/articleshow/63912079.cms.

[7] Vyas, Mahesh. “Employment Rate Falls in February 2019.” CMIE, Feb. 2019, www.cmie.com/kommon/bin/sr.php?kall=warticle&dt=2019-03-05%2B09%3A32%3A41&msec=496.

Share

Leave a Reply

Your email address will not be published. Required fields are marked *

You don't have permission to register