Don't lose sleep over job dip data

Recently, NSO released the first Quarterly Bulletin of Unincorporated Sector Enterprises (QBUSE), covering Jan-Mar 2025 and Apr-Jun 2025. Until now, only annual estimates of Annual Survey of Unincorporated Sector Enterprises (ASUSE) were available for 2021-22, 2022-23 and 2023- 24. The survey has now been remodelled to provide a high-frequency indicator for India's vast unincorporated sector, the only other one being PLFS.

Fresh quarterly estimates from QBUSE reveal that employment surged to 131.3 mn in Jan-Mar 2025, and held at 128.6 mn in Apr-Jun 2025 - significantly higher than 120.6 mn workers counted in the latest ASUSE 2023-24 (Oct 2023-Sep 2024). The number of establishments also increased sharply, from 73.4 mn in ASUSE 2023-24 to 78.5 mn and 79.4 mn in the first two quarters of 2025, respectively. Gains were evident in metrics like rural workforce, women's employment and digital adoption. MoSPI linked the fall in employment in Apr- Jun quarter to the 'moderate' performance of the manufacturing sector and a reduction in the number of hired worker establishments in the quarter.

Yet, headlines zoomed in on one point: fall in employment in Apr-Jun quarter in comparison to Jan-Mar quarter, especially in the backdrop of a phenomenal GDP growth in the first quarter (Apr-Jun) of 2025-26. At first glance, it appears to be a paradox. In reality, it isn't.


Every three months, when India's national accounts are released, headlines ask whether the economy has accelerated or slowed. But the quarterly real GDP growth rate is always measured against the same quarter of the previous year (suitably adjusted for price changes), not the immediately preceding quarter. Therefore, the 7.8% growth in Apr-Jun 2025-26 indicates that real GDP expanded by   7.8% over Q1 of 2024-25, not over Jan-Mar 2025. There is good reason for this.

The Apr-Jun quarter almost always looks weaker than Jan-Mar because economic activity peaks at the end of FY. Companies push output to close annual targets, governments front-load spending, and investments crest before Mar. Once that surge eases, Apr-Jun naturally shows a dip.

Add to this, extreme weather condition makes manual outdoor work especially difficult. Similar inter-quarter differences are seen throughout the year, which is why NSO reports GDP growth relative to the same quarter of the previous year or IIP growth over same period last year.

The apparent paradox arises because frames of comparison are different.

In QBUSE, change in employment is shown against the preceding (Jan- Mar 2025) quarter, whereas GDP growth is reported against the same quarter (Apr-June) of the previous year 2024-25. Reading the two side by side, therefore, is like comparing apples with oranges.

But what if we applied the same method to GDP or IIP and compared them with the immediately preceding quarter? We would then see dips after March in most years, because the year-end surge fades away in Apr-Jun. Thus, when a similar quarter-on-quarter comparison is applied to NAS or IIP data, the pattern looks almost identical, explaining the apparent paradox.

As India's data ecosystem becomes richer, with new surveys and more frequent releases, the risk of misinterpretation grows. High-frequency data are invaluable, but only when read with care. Policymakers, businesses and the public must recognise the nuances of each dataset and avoid drawing alarmist conclusions using mismatched yardsticks.

Considering the inherent seasonality, it is only proper to compare quarterly estimates with figures of the same quarter in the previous year. Until QBUSE builds up a longer time series that allows such comparisons, its findings should be interpreted with due caution.