For the best experience, open
on your mobile browser or Download our App.

Do GDP Growth Figures Reflect Robustness or a Statistical Overstatement?

Given the contrastingly frail household income situation, the continued slackness in private capex and deceleration in trade, the divergent core and headline real GDP growth are not trivial. 
Photo: Rawpixel. Public domain.

The razzmatazz around the robust 7.6% real second quarter (Q2) GDP growth pumped up an adrenaline rush last week, prompting many to up their projections for fiscal year 2024 (FY24) closer to 7%, higher than the RBI’s projection of 6.5%. 

But what is obscured is that consumer companies are facing a supply gridlock as demand falls, inventories pile up, and the festive season demand remained weak. 

Separately, there is extensive evidence of flatness in real incomes of households (76% of the GDP), derived from the RBI’s data on the contraction of financial savings and the Periodic Labour Force Survey (PLFS) report. CMIE says that the unemployment rate has spiked again, to an average of 9.5%. 

Is there a disconnect between the stupendous GDP data and the lugubrious household situation? We think not.

The answer lies in the way GDP is calculated, which is confounded with systemic statistical aberrations. While the headline real GDP growth numbers are disconnected from the ground reality, the removal of the discrepancies shows the true picture. 

Despite the robust headline growth situation as reflected in the 7.4% year-on-year (YoY) gross value added (GVA) growth, employment-intensive agriculture and services – which together contribute 72% of the GVA – saw a major deceleration. 

Robust contribution from manufacturing and construction, which grew by 13.9% and 13.3% YoY respectively, translating into Industry GVA growth of 13.2%, essentially derives from the imputed higher value addition. This was due to low inflation related to the WPI driving up margins and profits of large firms.

On the expenditure side, real GDP growth at 7.6%, however, encloses weak core demand. 

Putting the identified components of the expenditure GDP explained by the components of household and government consumption, capital formation, and net exports, i.e. GDP excluding the unexplained discrepancy portion, the real core GDP stands at 3.0% in Q2 of FY24, following 1.4% in Q1. This is only 40% of the reported GDP of 7.6% in Q2!

Given the contrastingly frail household income situation (a la PLFS 2022-23, RBI’s savings data FY23), the continued slackness in private capex and deceleration in trade, the divergent core (2.3% YoY in the first half of FY24) and headline real GDP growth (7.7%) are not trivial. 

The substantial 71% unexplained portion of the real GDP growth in the first half of FY24 is because the estimated real GVA side, representing the value added from production across all sectors, is 2.7% higher than the real core expenditure GDP. But for the discrepancies, the real GVA plus net indirect taxes should match with the real GDP. 

Why the discrepancy is cause for worry

Normally, discrepancies should not be worrisome if it is a random error, where frequent offsetting positive and negative deviations average at zero. But that isn’t the case. 

The average discrepancy during Q1 FY12 to Q2 FY24 is +0.6% of GDP with a maximum of 5.4% and a minimum of -3.9%; 32 out of 50 quarters saw a positive discrepancy, indicating that real GVA exceeded core GDP in 64% of quarters. And more importantly, it has an inverse correlation of (-)0.63 with the GDP/GVA deflator, which rose to (-)0.76 in the post-pandemic era. 

So, the overestimation of GVA occurred during times of falling inflation or deflation, mainly driven by WPI. Thus, indexed to Q2 FY12, the real GDP in Q2 FY24 is 3.8% (or Rs 1.6 trillion) higher than the core real GDP.

Therefore, the discrepancies capturing a large portion of unaccounted GDP growth are not random and relate to the usage of WPI as a deflator to derive real GVA. Such systemic discrepancies, therefore, pose a challenge to the credibility of the real GDP estimation process.

GVA is the value of output excluding net indirect tax less cost of input consumption valued at purchasers’ prices. Hence, the measured value added would rise sharply with a fall in commodity or raw material prices, as has been the case recently, even with slowing consumption demand, particularly in a scenario where large firms have gained considerable operating margins due to elevated monopolistic power since mid-2016 at the cost of smaller firms. 

Conversely, given the sharp deceleration in estimated household real disposable incomes to 2% YoY in FY23 due to the decline in the quality of jobs, aligning with the decline in savings and slowing consumption and flat real wage earnings in four years (PLFS 2022-23), the real core GDP growth or the demand side does not share the robustness of the output side or estimated GVA growth. 

With the PLFS data also indicating a structural rise in dependence of households on less productive rural (73.3% of the population) and agri (45.8% of workers, 2022-23), the resultant decline in labour productivity and the household income which contributes 76% of GDP is misaligned with the sharp gains in reported productivity of the producing firms represented by real GVA growth.

This feebleness of the household consumption demand combined with the lack of private investment and the recent deceleration in exports has overpowered robust government capex. Hence, the aggregate demand remains significantly weak.

Also Read | Full Text: Pronab Sen Explains Why Data on Which GDP Is Calculated Is A Major Concern

A crucial factor is that the deflator used for deriving the real GVA is highly correlated with the WPI inflation. However, given the dominant contribution of household consumption, core expenditure GDP is also sensitive to CPI inflation. Hence, the sharp decline in WPI inflation (-1.8% in the first half of FY24) and higher CPI inflation (5.5%) also contributed to the rise in divergence between low real core GDP and strong GVA growth. 

Thus, all put together, the GVA side of the national income is susceptible to over-identification of large surviving companies that have gained market share from the non-corporate smaller firms in the manufacturing and services industries. As a corollary, it underrepresents the falling productivity arising from gaining ruralisation. Secondly, it is also susceptible to higher projections of gross value added due to the sharp decline in commodity prices (or underestimation in times of sharp rise in commodity prices).

Contrastingly, growth in real GDP ex discrepancies is a more relatable measure to depict the actual situation of the households’ livelihood, employment and productivity. Also, unlike the GVA estimates, the problem of exclusion and volatility induced by WPI inflation is much lesser in real core GDP.  

Thus, the endogenous growth, depicting the situation of households, in terms of income and consumption capability remains very weak. Behind the robust real headline GDP growth, padded up by overstated GVA is a flattening trend in the core real GDP growth, which is worrisome. 

Whereas the enfeebled demand reflects the structural issues, as the divergence between CPI and WPI inflation eventually narrows it would imply that the reported GDP will also converge towards the lower structural trend in core GDP growth.

Dhananjay Sinha is co-head of Equities and head of Research of Strategy and Economics at Systematix Group.

Make a contribution to Independent Journalism
facebook twitter