Are India's falling consumer spending and strong GDP growth in sync?

At the macro level, a similar dichotomy is visible in two sets of data released by the same institution, the Ministry of Statistics and Programme Implementation

The government on Friday cited a discrepancy between the national accounts numbers and findings of the survey report of 2017-18 on consumer expenditure as one of the reasons for scrapping the latter.

However, the consumer expenditure estimates (derived from the gross domestic product data) produced by the Central Statistics Office (CSO), historically had divergence with the survey reports on consumer expenditure, conducted by the National Statistical Office (NSO). 

The two bodies are arms of the Ministry of Statistics and Programme Implementations (MoSPI).

Consider this double dichotomy. The NSO’s consumer expenditure survey of 2017-18 showed a contraction in consumer spending on most food items, in comparison to 2011-12. But that on milk has increased.

Contrary to this, the National Accounts Statistics (NAS), which is the annual series on GDP estimation, showed a sharp decline of 10 per cent in consumer spending on milk between 2016-17 and 2017-18.

The CSO data pointed to a 7.5 per cent rise in consumer spending, or private final consumption expenditure, while the overall economy grew by 7.2 per cent. However, the NSO survey presented a different narrative: Consumer spending declined by 3.7 per cent between 2011-12 and 2017-18, a first in over four decades.

One would think that these divergent findings do not square up at all and how did the GDP grow by 7.2 per cent when consumption spending fell according to the NSO estimates? 

Consumption expenditure forms more than half of the country’s GDP.

The answer to this lies in findings of an expert committee which had submitted its study to the MoSPI in 2015. It pointed to a growing divergence between the consumption data produced by national accounts and the NSO surveys over four decades. It has grown from 5 per cent in 1972-73 to 45 per cent in 2011-12.

There are various reasons for the divergence between the two estimates, including its coverage, according to the committee. For instance, the private final consumption expenditure captures non-profit institutions serving households (NPISH) like temples and gurdwaras while the household consumption expenditure is based on survey of households and doesn’t capture the former.

In addition, while the NSS is estimated from a primary survey (interactions with members of a household), the NAS is derived using the “commodity flow method”.

Moreover, several experts said that the divergence visible is a global phenomenon.

Former chief statistician Pronab Sen said that the key difference between the two is the way the two datasets capture the unorganised sector.

“The NAS is usually based on the organised sector data, sourced from the index of industrial production, annual survey of industries and the ministry of corporate affairs, among others,” he said. The NSS surveys, wherein 70 per cent of the sample size is in rural areas, captures the unorganised sector better.

Sen added that the demonetisation exercise (November 2016), adversely affected the unorganised sector the most, and divergence in the national accounts and the consumer expenditure survey could be more than what was seen in the previous years.

Economics professor N R Bhanumurthy, who teaches at the National Institute of Public Finance and Policy, said as 2017-18 was not a “normal year”, results of the surveys would deviate from the normal.

“That particular year would not be a suitable base year for the new national accounts series,” he said. The government is planning to change the base year of national accounts from the current 2011-12 to 2017-18.

Economists, who have worked at top positions in the Indian statistical system, but who did not wish to be named, also said that a particular survey on consumer spending (NSS) forms the basis for the national accounts series (NAS) that begins in that particular year. 

For example, the survey of 2011-12 formed the basis for the calculation of consumption estimates in the GDP series of 2011-12.

So, while one (NSS) is the starting point for estimating the other (NAS), the two tend to diverge over time, they said. This is the very reason for change in base year at regular intervals of less than a decade, so that the changes in the economy are captured, they said.

“The national accounts capture the unorganised sector better in its base year,” Sen added.