A government does not spend ninety days explaining a law that everybody already understands.
On 28 September 2026, Union Labour Minister Dr Mansukh Mandaviya launched Shram Samvad, an awareness campaign on the four Labour Codes that will cover industrial clusters across 316 districts in all 36 States and Union Territories over the next 90 days. The first district programme ran the same day at the Regional Labour Institute, Faridabad — five sessions, about 75 participants, mostly employers, state officials and advocates. The launch came at the National Conference on Industrial Relations and Labour & Awards 2026, organised by the Confederation of Indian Industry (CII) and the Employers' Federation of India (EFI) on the theme "Reengineering Industrial Relations & Workplace Participation in the AI Age".
The Codes have been in force since 21 November 2025. Ten months later the Ministry is sending teams to 316 districts to explain what they say. That gap is the most informative fact in the story, and it is worth understanding before anything else.
What the four Codes are
India's labour statute book was, until recently, 29 central laws, most of them framed between the 1930s and 1950s. They were consolidated into four Codes:
| Code | Year | Subsumes, broadly |
|---|---|---|
| Code on Wages | 2019 | Minimum wages, payment of wages, bonus, equal remuneration |
| Industrial Relations Code | 2020 | Trade unions, standing orders, industrial disputes |
| Occupational Safety, Health and Working Conditions Code | 2020 | Factories, mines, dock workers, contract labour, inter-state migrant workers, plantations |
| Code on Social Security | 2020 | Provident fund, ESI, gratuity, maternity benefit, unorganised workers |
The Code on Wages received presidential assent on 8 August 2019. The other three received assent on 28 September 2020 — six years to the day before Shram Samvad was launched. There is no indication the date was chosen deliberately, and it is a coincidence rather than an anniversary, but it is a useful peg for remembering that three of the four Codes are 2020 statutes and only one is from 2019.
Note the five-year gap between assent and commencement. A Code that has received assent is law on paper; it operates only when the government notifies it into force and frames rules under it. Labour is in the Concurrent List (List III, Entry 22-24), which means the Centre frames the Codes and the states frame most of the operational rules. That division is exactly why commencement took so long, and it is also why an awareness campaign aimed at 316 districts makes sense: a great deal of what an employer must actually do is determined at state level.
What actually changed
Strip away the slogans and the substantive changes fall into four groups.
Formalisation. Every worker must now be issued an appointment letter stating designation, wages and social security entitlements. This sounds procedural and is not. A worker without written proof of employment cannot establish continuity of service, cannot claim gratuity, and in a dispute has no document to put before a tribunal. Mandatory appointment letters convert an oral arrangement into an enforceable one.
Coverage. The Code on Social Security defines 'gig work', 'platform work' and 'aggregator' in statute for the first time. Aggregators must contribute 1-2% of annual turnover, capped at 5% of the amount paid or payable to gig and platform workers. Read that ceiling carefully — it is a cap on the contribution, not a floor, and it binds the contribution to payouts rather than to headcount. ESI coverage is extended pan-India: voluntary for establishments with fewer than ten employees, mandatory for any establishment with even one worker in a hazardous process. Employers must provide a free annual health check-up to every worker above 40.
Employment forms. Fixed-term employees must receive the same wages, leave and social security as permanent staff doing the same work, and become eligible for gratuity after one year of continuous service instead of five. This is the single most consequential change for anyone studying labour economics: it makes direct fixed-term hiring cheaper relative to hiring through a contractor, which is precisely the substitution the government wants.
Flexibility for employers. Women may work night shifts and in all categories of work, including underground mining and hazardous processes, subject to consent and safety measures. Compliance moves to single registration, single licence and single return. Enforcement shifts from inspection to an Inspector-cum-Facilitator model. Disputes go to two-member industrial tribunals, with direct access after conciliation fails. Establishments with 500 or more workers must constitute safety committees, and a National OSH Board sets harmonised standards.
There is a genuine trade in here, and it is worth naming plainly rather than pretending the Codes are all gain. Workers get portability, statutory minimum wages, written contracts and a wider social security net. Employers get lower compliance burden, easier use of fixed-term contracts, and higher thresholds before the most restrictive provisions bite. Whether that trade favours labour or capital depends almost entirely on how vigorously the new rights are enforced — which brings us back to why a campaign is being run in 316 districts.
The number that is not a percentage
Dr Mandaviya told the conference that India's growth has become far more employment-intensive: that every 1% of GDP growth translated into 1.11% employment growth during 2017-23, "compared to just 0.008% earlier".
This claim is doing a lot of work, and three things about it are worth a student's attention.
First, it is an elasticity, not a percentage. Employment elasticity is the ratio of the percentage change in employment to the percentage change in output. It is a pure number. An elasticity of 1.11 means employment grew 1.11% for every 1% of value added — the "%" attached to it in the release is a category error, repeated often enough that it has become standard in Indian policy discourse. Writing "1.11" is correct; writing "1.11%" is not.
Second, the comparator is unnamed. The figure of roughly 0.01 comes from the period 2011-16. The release simply says "earlier". Whether growth has become more job-rich depends entirely on which earlier period you pick, and 2011-16 was an unusually weak stretch for employment. Naming the comparator is not a technicality; it is the whole claim.
Third, the underlying data matters. Both figures derive from the RBI's KLEMS database (Capital, Labour, Energy, Materials, Services), which put employment at 64.33 crore in 2023-24 against 47.5 crore in 2017-18. KLEMS counts all workers, including the self-employed and unpaid family workers. A large part of the increase over that period was a rise in self-employment and unpaid family labour, particularly in agriculture — a pattern visible in the Periodic Labour Force Survey. An elasticity above 1 sustained over six years would be extraordinary by international standards. It is not impossible; it does require that you know what is being counted as employment.
The honest formulation is this: on the KLEMS measure, employment growth relative to output growth in 2017-23 was far higher than in 2011-16. That is a defensible statement. "Every 1% of growth creates 1.11% more jobs" is a looser one.
Two different denominators
The same speech reported that social security coverage has risen from 19% in 2015 to 101 crore people, or 68%, today, as recognised by the International Labour Organization.
Compare that with the Ministry's own release of 21 November 2025, which described the identical trajectory as rising "from about 19% of the workforce in 2015 to more than 64% in 2025". One release says percentage of workforce; the other says percentage of population. These are not the same denominator, and the gap between them is roughly a factor of two.
The ILO's measure — used in its World Social Protection Report — is effective coverage by at least one social protection benefit, expressed as a share of population. On that definition 101 crore out of a population of about 146 crore gives roughly 69%, so the 2026 framing is internally consistent. The 2015 baseline of 19% almost certainly refers to the same population-based measure, which means the November 2025 release's "of the workforce" was the loose phrasing.
This matters for a reason beyond pedantry. Coverage on the ILO metric counts anyone receiving any one benefit — a ration entitlement, a pension, an insurance cover, a cash transfer. It is a real achievement and it is what India's social security net crossing 101 crore measures. It is not a measure of formal employment, and it does not mean two-thirds of Indians have a provident fund. The register of unorganised workers built by e-Shram and the raising of the EPFO wage ceiling to ₹25,000 are the instruments that move the formal number, and they move it far more slowly.
Why the campaign exists
The four Codes replaced 29 laws whose provisions were familiar to every labour officer, every HR department and every trade union in the country. Familiarity is an asset a statute book accumulates over decades: people know where to look, which clause governs, what a phrase has been held to mean. Consolidation destroys that asset at a stroke and takes years to rebuild.
Add to this that the Codes' transitional provision keeps the old Acts, rules, notifications and schemes in force during the changeover, that most operational rules are made by states, and that the single largest group affected — small employers in industrial clusters — has neither in-house counsel nor a compliance department. A campaign covering 316 districts, running sessions on employer obligations alongside sessions on PM-SYM and PM-VBRY, is the reasonable response to that problem. It is not a publicity exercise. It is the cost of consolidation, being paid in the tenth month.
🔑 Revision block
- Shram Samvad launched 28 September 2026: 316 districts, 36 States and UTs, 90 days. First programme at Regional Labour Institute, Faridabad.
- Launched at the National Conference on Industrial Relations and Labour & Awards 2026 (CII + EFI), theme "Reengineering Industrial Relations & Workplace Participation in the AI Age".
- 4 Labour Codes, in force 21 November 2025, rationalising 29 central labour laws. Assent: Code on Wages 8 August 2019; the other three 28 September 2020.
- Labour is in the Concurrent List — Centre frames the Codes, states frame most rules.
- Gig/platform workers: defined in statute for the first time; aggregator contribution 1-2% of annual turnover, capped at 5% of payouts to such workers.
- Fixed-term employees: parity with permanent staff; gratuity after 1 year instead of 5.
- Free annual health check-up for workers above 40. ESI pan-India; mandatory for even 1 worker in a hazardous process.
- Single registration, single licence, single return; Inspector-cum-Facilitator; 2-member industrial tribunals; safety committees at 500+ workers; National OSH Board.
- Employment elasticity 1.11 for 2017-23 vs about 0.01 for 2011-16 (RBI KLEMS). It is a ratio, not a percentage.
- KLEMS employment: 47.5 crore (2017-18) to 64.33 crore (2023-24).
- Social security coverage: 19% (2015) to 101 crore / 68% now, on the ILO population-based measure of coverage by at least one benefit.
🎯 Practice MCQs
Q1. The four Labour Codes were brought into force with effect from which date? (a) 1 April 2025 (b) 15 August 2025 (c) 21 November 2025 (d) 1 January 2026
→ (c) The Codes were notified into force on 21 November 2025, rationalising 29 central labour laws. The Code on Wages had received assent as far back as August 2019.
Q2. Under the Code on Social Security, 2020, the contribution required from an aggregator is: (a) 1-2% of annual turnover, capped at 5% of the amount paid or payable to gig and platform workers (b) 5% of annual turnover, with no ceiling (c) 2% of the wages of each registered gig worker (d) 1% of profit after tax
→ (a) The contribution is pegged to turnover (1-2%) but ceilinged by reference to payouts to gig and platform workers (5%). Tying it to turnover rather than headcount was deliberate, since aggregators do not employ gig workers in the conventional sense.
Q3. Labour, as a subject of legislation under the Constitution of India, falls in the: (a) Union List (b) State List (c) Residuary powers of Parliament (d) Concurrent List
→ (d) Labour is in the Concurrent List, which is why the Centre enacted the Codes while most operational rules under them are framed by state governments — a division that explains both the delay in commencement and the district-level focus of Shram Samvad.
Q4. Which Code subsumes the law on trade unions, standing orders and industrial disputes? (a) Code on Wages, 2019 (b) Industrial Relations Code, 2020 (c) Code on Social Security, 2020 (d) Occupational Safety, Health and Working Conditions Code, 2020
→ (b) The Industrial Relations Code, 2020 consolidates the Trade Unions Act, the Industrial Employment (Standing Orders) Act and the Industrial Disputes Act. Wages, bonus and equal remuneration sit in the 2019 Code.
Q5. Under the Codes, a fixed-term employee becomes eligible for gratuity after: (a) 1 year of continuous service (b) 3 years of continuous service (c) 5 years of continuous service (d) 7 years of continuous service
→ (a) The five-year threshold that applies to permanent employees is reduced to one year for fixed-term employees. This is the provision most likely to change hiring behaviour, because it narrows the cost gap between direct fixed-term hiring and contractor-mediated hiring.
Q6. "Employment elasticity of growth" is best described as: (a) The number of jobs created per crore rupees of public expenditure (b) The share of the workforce in formal employment (c) The ratio of the percentage change in employment to the percentage change in output (d) The percentage of GDP accounted for by wages
→ (c) It is a ratio of two growth rates and therefore a pure number. An elasticity of 1.11 means employment grew 1.11% for every 1% of output growth. Expressing it with a percentage sign, as official statements often do, is a category error.
Q7. The employment figures of 47.5 crore for 2017-18 and 64.33 crore for 2023-24 are drawn from: (a) The Periodic Labour Force Survey (b) The RBI's KLEMS database (c) The Economic Census (d) The Annual Survey of Industries
→ (b) These are KLEMS figures — the RBI's productivity database covering Capital, Labour, Energy, Materials and Services. KLEMS counts the self-employed and unpaid family workers, which is why its employment totals run well above formal-sector counts.
Q8. Under the Occupational Safety, Health and Working Conditions Code, a safety committee is mandatory in establishments employing at least: (a) 100 workers (b) 250 workers (c) 300 workers (d) 500 workers
→ (d) The threshold for a mandatory safety committee is 500 workers. Students routinely confuse this with the 300-worker threshold that governs standing orders and prior permission for retrenchment and closure.
Q9. The ILO measure under which India's social security coverage is reported as having reached about 68% counts: (a) Population covered by at least one social protection benefit (b) Workers enrolled with the EPFO (c) Households below the poverty line receiving cash transfers (d) Workers with written employment contracts
→ (a) The ILO's World Social Protection Report measures effective coverage by at least one social protection benefit, as a share of population. It is therefore not a measure of formal employment, and does not imply that two-thirds of Indians hold a provident fund account.
Q10. Which of the following is NOT among the changes introduced by the four Labour Codes? (a) Mandatory appointment letters for all workers (b) Free annual health check-up for workers above 40 (c) A statutory bar on women working night shifts in hazardous processes (d) Single registration, single licence and single return
→ (c) The Codes do the opposite: women are permitted to work at night and in all categories of work, including underground mining and hazardous processes, subject to consent and prescribed safety measures. The other three are all introduced by the Codes.
📋 How this gets asked (PYQ pattern)
Labour questions in CDS and other UPSC papers have historically clustered in three places, and the Codes have not changed the pattern so much as renumbered it.
The first is constitutional placement — labour, trade unions, industrial disputes and social security are all Concurrent List entries, and the examiner tests whether you know that before testing anything else. Expect this as a straight one-liner.
The second is which law does what. Under the old scheme this meant matching the Factories Act, the Industrial Disputes Act and the Minimum Wages Act to their subjects. Under the new scheme it means matching the four Codes to what they subsume. The mapping is the question; learn it as a table, not as prose.
The third is threshold numbers, and this is where candidates lose marks. The Codes contain several similar-looking figures — 300 workers for standing orders and prior permission, 500 for safety committees, 40 years for health check-ups, 1 year for fixed-term gratuity, 1-2% and 5% for aggregator contributions, 10 workers for ESI voluntariness. Statement-based questions are built precisely by swapping one of these for another.
A fourth pattern is emerging and is worth anticipating: data-source questions. When a paper cites an employment figure, it increasingly names the source — KLEMS, PLFS, EPFO payroll, e-Shram registrations — and asks which one it is. These four measure different things and disagree with one another by design. Knowing what each counts is now part of the syllabus in practice, whatever the syllabus says on paper.
Preparing for CDS/OTA? Labour law rewards table-learning over reading: fix the four Codes against what they subsume, then fix the thresholds against each other, and the statement-based questions answer themselves. Build the base with our CDS/OTA economy notes, follow the daily CDS current affairs, and prepare with our faculty in the upcoming Cavalier courses in Delhi.
✍️ Written by Hitendra Deswal — Economy and polity faculty at The Cavalier. Reviewed by the Cavalier Faculty Desk.