Context

• In recent debates surrounding job creation in India, the KLEMS database has become a frequently cited source.
• Developed as part of an international project, this database has been curated by scholars from the Delhi School of Economics and the Indian Council for Research on International Economic Relations (ICRIER) since 2009, and is now housed at the Reserve Bank of India (RBI) since 2022.
• Given its prominent use in economic discussions, particularly in countering claims of poor job creation, it is imperative to examine the methodology behind this data collection, its sources, and its sectoral breakdowns.

The KLEMS Database: Composition and Sources

• The KLEMS database encompasses data on capital (K), labour (L), energy (E), material (M), and services (S) for the period from 1980 to 2024.
• It is designed as a measurement tool to monitor and evaluate productivity growth both at the industry level and across the broader economy.
• The data sources include the employment-unemployment surveys (EUS), periodic labour force surveys (PLFS), National Account Statistics, and the Annual Survey of Industries.
• In the absence of annual data from the National Statistical Office, these data points serve as benchmarks, with interpolations used for years where direct data is unavailable.

Methodological Limitations Employed in Employment Figures within the KLEMS and its Implications

• Population Estimation and Worker-Population Ratios (WPR)
  • The first step in the KLEMS methodology involves estimating the population for different demographic groups—rural male, rural female, urban male, and urban female.
  • The employment estimates are derived by multiplying the Worker-Population Ratios (WPR) for these groups by the total population.
  • The WPR, which indicates the proportion of the population that is employed, is drawn from employment-unemployment surveys (EUS) and periodic labour force surveys (PLFS).
  • However, the surveys themselves do not provide the absolute number of workers; instead, they offer ratios that are then applied to population estimates.
  • The population figures used for these calculations can be interpolated using Census data or taken from the projections provided by the National Population Commission under the Ministry of Health and Family Welfare (MoHFW).
  • The choice of data source and the method of interpolation can significantly impact the final employment figures.
• Use of Projections and Potential Overestimation
  • One of the critical issues with the KLEMS methodology is the reliance on population projections, particularly those provided by the MoHFW.
  • These projections have been criticised for being on the higher side, primarily due to an overestimation of population growth following a significant decline in fertility rates between 2010 and 2020.
  • This overestimation has far-reaching consequences and when these inflated population figures are multiplied by the WPR, the resulting estimates for the labour force and total workforce tend to be exaggerated.
  • Moreover, the methodology assumes uniform growth rates for rural and urban populations.
  • This assumption is problematic because, in reality, rural areas typically experience slower population growth than urban areas.
• Sectoral Employment Distribution
  • Another aspect of the KLEMS methodology involves the distribution of the total estimated number of workers across various industry sectors.
  • This distribution is based on the shares of employment in these sectors as reported in the PLFS.
  • However, this approach does not account for changes in industry dynamics, such as shifts in the economic landscape or the emergence of new sectors, which could alter the distribution of employment over time.
  • For instance, if a sector experiences a decline in employment opportunities but the methodology assumes a constant or increasing share based on outdated PLFS data, the employment figures for that sector may be misleading.
• Temporal Comparability and WPR Variability
  • A significant methodological challenge in the KLEMS database is ensuring temporal comparability of data, particularly when transitioning from EUS to PLFS.
  • The WPR derived from EUS data for the period 2011-12 is not directly comparable with the WPR from PLFS data for 2017-18, yet the KLEMS methodology assumes continuity without addressing potential discrepancies.
  • This shift could introduce biases in the employment estimates, especially if the methodologies or survey designs of EUS and PLFS differ significantly.
• Inclusion of Subsidiary Employment and its Impact on Employment Figures
  • The KLEMS database includes individuals engaged in subsidiary employment—those who have secondary or marginal work engagements—in its employment figures.
  • This inclusion can be misleading, as it encompasses individuals with tenuous connections to the labour market, such as unpaid family workers.
  • These workers, though counted as employed, may not have secure or full-time jobs, thus distorting the perception of employment quality and stability.
• Implications for Policy and Economic Analysis
  • If employment figures are overestimated due to inflated population projections, incorrect growth rate assumptions, or the inclusion of subsidiary employment, policymakers might develop a misguided understanding of the labour market.
  • This could lead to inappropriate policy responses, such as underestimating the need for job creation initiatives or overlooking the challenges of underemployment and poor job quality.
  • Moreover, using the KLEMS data to claim rapid employment growth, especially in the absence of a critical examination of the methodology, could result in complacency among policymakers and stakeholders, who might believe that the employment situation is better than it actually is.

An Analysis of Emerging Employment Trends According to KLEMS and Contrasting Data between ASUSE and KLEMS

• Employment Trends According to KLEMS
  • The data trends within the KLEMS database suggest a significant rise in employment across sectors post-2018.
  • For instance, agricultural employment reportedly increased from 20 crore to 25 crore between 2018-19 and 2022-23.
  • Similarly, service sector employment rose from 17.2 crore to 20.2 crore, while manufacturing employment grew from 5.5 crore to 6.3 crore.
• Contrasting Data between ASUSE and KLEMS
  • A study by economists at the State Bank of India (SBI) highlights discrepancies between employment figures from different sources.
  • The Annual Survey of Unincorporated Sector Enterprises (ASUSE) estimated employment in unorganised enterprises to be 10.96 crore.
  • Yet, this figure has been inflated to claim that total employment in 2022-23 reached 56.8 crore, aligning closely with KLEMS data.
  • This discrepancy suggests that the KLEMS data may not fully account for the nuances of employment, particularly within unorganized sectors.
  • The employment figures from enterprise surveys, which indicate positions in enterprises, do not easily correlate with individual-level data collected through household surveys, which are generally considered more reliable.

Conclusion

• The methodological limitations of the KLEMS database raise significant concerns about its use in evaluating job creation in India.
• While the database is valuable for tracking productivity and employment trends, its reliance on interpolated data, assumptions about population growth, and the inclusion of tenuous employment connections could lead to inflated or misleading employment figures.
• It is crucial to critically evaluate the data and consider the broader context before drawing conclusions about the state of employment in the country.