Reading the SMK3 Numbers

In May 2026 Indonesia's Minister of Manpower said the occupational safety and health management system known as SMK3 was applied by only about 18,000 of 450,000 companies, and named "real and measurable" SMK3 implementation as one of three priorities to pursue with the social security agency BPJS Ketenagakerjaan (Zhafira, 2026). Three months later the ministry's Director of Labour Inspection Development and OSH Testing put the figure at about 19,000 (Republika, 2026). The ministry's own open data portal, Satudata, gives a different number again: 3,416 companies applying SMK3 in September 2026 (Kementerian Ketenagakerjaan [Kemnaker], 2026d).

The gap matters because the obligation now sits in statute. Law No. 27 of 2026 on Labour Protection requires companies with at least 100 workers or a high risk level to apply an SMK3 integrated with the company's management system (Article 130(1) and (3)), and a company that does not may face administrative sanctions ranging from a written warning to revocation of its business licence (Article 132) (Republik Indonesia, 2026). This piece reads the official data that exist: what they show, how to read them, and what they cannot yet tell us.

Three Sources, Three Numbers

The first source is ministerial statements. Our arithmetic puts 18,000 of 450,000 at 4 percent (Zhafira, 2026), but the statement does not say whether 450,000 is every company or only those obliged to apply SMK3. We found no official count of companies above the statutory threshold, so true compliance coverage cannot yet be calculated.

The second source is Satudata. Its metadata define a company applying SMK3 as one that "has an SMK3 and has been monitored by a labour inspector", and group these companies by audit criteria category (Kemnaker, 2026d). The categories come from Appendix II of Government Regulation No. 50 of 2012 on the Application of OSH Management Systems (PP 50/2012): initial level with 64 criteria, transition level with 122, and advanced level with 166, scored as poor (0 to 59 percent), good (60 to 84 percent) or satisfactory (85 to 100 percent) (Pemerintah Republik Indonesia, 2012).

The third source is BPJS Ketenagakerjaan work accident insurance claims, also published on Satudata: 319,382 cases in 2025, counted from claims approved and paid (Kemnaker, 2026f). The 2024 figure was 462,241 (Kemnaker, 2025b).

What the Series Shows

In September 2026, of the 3,416 companies, 2,090 were assessed at the initial level, 117 at the transition level and 1,209 at the advanced level. By score, 958 were rated good and 2,458 satisfactory, and the poor category reads zero in every province (Kemnaker, 2026d). So 61 percent were assessed against 64 criteria, and 72 percent scored satisfactory.

The geography is lopsided. Jakarta recorded 657 companies, West Java 523, East Java 456 and Banten 323; these four provinces hold 1,959 companies, or 57 percent of the total, and the six provinces of Java hold 64 percent. Gorontalo recorded 3, and West Papua and Highland Papua 2 each (Kemnaker, 2026d). Central Sulawesi stands out: 36 of its 41 companies were assessed at the advanced level, yet 29 of them scored good rather than satisfactory.

The monthly series does not trend upward. January 2026 recorded 4,129 companies (Kemnaker, 2026a), June jumped to 5,511 (Kemnaker, 2026b), July fell to 3,293 (Kemnaker, 2026c), and September reached 3,416. From January to September the total fell 17 percent, and advanced-level companies fell from 1,406 to 1,209.

Reading the Numbers

In our reading, the gap between 18,000 and 3,416 is less about which figure is right than about what each one counts. The quarterly 2025 series rose through the year and closed at 1,965 companies in the fourth quarter (Kemnaker, 2026e), the usual shape of a year-to-date cumulative count. The 2026 monthly series lacks that shape. Read as monthly flows, the four months quoted above already add up to 16,349 companies, which is hard to reconcile with 18,000 to 19,000. Annual SMK3 awards also run in the low thousands, for example 2,456 companies in 2024, alongside 1,756 companies receiving zero accident awards (Kemnaker, 2025c). Our interpretation is that the 2026 monthly figures more likely reflect companies with active status in a given month, while 18,000 to 19,000 may be a multi-year cumulative count. No source confirms this; the metadata state neither the reference period nor how expired certificates are treated.

The category mix also shifts sharply between series. In the fourth quarter of 2025 the transition level accounted for 410 of 1,965 companies, or 21 percent (Kemnaker, 2026e); by September 2026 it was 3 percent. The June 2026 spike that vanished a month later, from 5,511 to 3,293, a 40 percent drop, looks more like a change in how the data were compiled than a change on the ground. In our view these two features are reason enough not to read the monthly trend as SMK3 adoption rising or falling.

The zero in the poor category also needs care. PP 50/2012 states that a company with critical or major findings is judged not yet to have applied SMK3 successfully and is not scored on the achievement table (Pemerintah Republik Indonesia, 2012). The Satudata figures are therefore figures for companies that passed, not the distribution of all audit results. A high share of satisfactory scores says nothing about how many companies failed or were never audited.

On the accident side, BPJS claims are a lagging indicator shaped by enrolment and claims processing. A 31 percent fall from 2024 to 2025 does not by itself mean safer workplaces without data on insured workers and hours worked. The minister cited 319,224 claims for 2025, with 9,834 deaths, 4,133 disability cases and 158 occupational disease cases (Zhafira, 2026); the difference from the Satudata figure is exactly 158, but neither source says whether the two are related.

Gaps in the Data

No public data link a company's SMK3 status to its accident claims, so whether SMK3 reduces accidents in Indonesia cannot be answered from national data. Data integration was one of the steps BPJS Ketenagakerjaan named after the May 2026 meeting (Zhafira, 2026), but no result is visible on the portal yet.

Evidence from abroad is mixed. The systematic review by Robson et al. (2007) concluded there was insufficient evidence to recommend for or against specific OHS management system interventions. A study of 5,147 Spanish firms found OHSAS 18001 certification only loosely related to accident rates and found a negative selection effect, with certification more common in sectors with worse accident records (Heras-Saizarbitoria et al., 2019). A United States study linking certification data from 1995 to 2016 with establishment-level injury data found selection in the opposite direction, with safer establishments more likely to certify, and estimated that certification reduced injury and illness cases by 20 percent (Viswanathan et al., 2024). Both studies concern OHSAS 18001 in other countries, not SMK3 audits in Indonesia, so the size of the effect does not transfer directly.

The realist review by Madsen et al. (2020) suggests why results vary: management systems work through integration, learning, motivation, translation and attention, and those mechanisms appear only in certain contexts. A certificate records that criteria were met on audit day; it does not measure whether those mechanisms keep running after the auditors leave. The official message in August 2026 that companies should not stop at certification (Republika, 2026) is consistent with that finding.

Indicators Worth Tracking

Since national data cannot yet judge performance, the useful measures sit at company level. Article 6 of PP 50/2012 and Article 130(5) of Law 27/2026 both set out five stages: policy, planning, implementation, monitoring and evaluation of performance, and review and improvement (Pemerintah Republik Indonesia, 2012; Republik Indonesia, 2026). ISO 45001 asks for the same through clause 9.1 on monitoring and performance evaluation and clause 9.3 on management review (International Organization for Standardization [ISO], 2018). The set below is, in our view, realistic:

• Leading indicators drawn from the new duties: coverage of periodic health checks at least once a year, delivery of routine OSH training, and hazard identification and risk assessment kept up to date (Law 27/2026, Article 130(6)) (Republik Indonesia, 2026).
• Dangerous situation reports: number received, time to follow-up and share closed. Article 130(6)(g) now refers to reporting dangerous situations to the relevant authorities, so an internal register is worth setting up now.
• Audit findings by PP 50/2012 element: not just the final percentage, but which criteria were missed and how long they took to fix, since reporting and correcting deficiencies is itself an assessed element (Pemerintah Republik Indonesia, 2012).
• Lagging indicators: accidents, occupational disease and lost days, expressed per hours worked so years can be compared.

One belief worth correcting is that a satisfactory score means a safe site. The score shows the criteria were met at audit; the evidence above shows its link with accident rates depends on whether the system is actually used.

The K3 Committee

Article 131 of Law 27/2026 requires companies to form a K3 committee of employer and worker representatives, tasked with formulating policy, supervising, coordinating education and training, and receiving and following up reports on potential OSH hazards and risks. Article 307 states that existing P2K3 safety committees are to be called K3 committees (Republik Indonesia, 2026). For companies that already have a P2K3 the change of name is automatic; what remains is making sure the committee actually works with data.

Current regulation already gives the framework. Minister of Manpower Regulation No. 13 of 2025 tasks the P2K3 with compiling and processing OSH data (Article 9(2)(m)), including accident and occupational disease analysis and statistics in its programme (Article 10(2)(d)), and reporting electronically every 6 months (Article 11) (Kemnaker, 2025a). The indicators above can form the content of that report. Note that Article 131 of Law 27/2026 contains no worker-number threshold of the kind in Article 3 of Permenaker 13/2025, while implementing rules on OSH and SMK3 are to come in a Government Regulation (Article 133) due within 1 year of 28 October 2026 (Article 312) (Republik Indonesia, 2026). Until then, existing implementing regulations remain in force where they do not conflict (Article 310); how the threshold applies to smaller companies is a question for legal counsel.

How Nusa Safety Helps

Nusa Safety can support preparation and monitoring. The Nusa Safety team carries out SMK3 internal audits against the PP 50/2012 criteria with findings traceable by element, supports SMK3 and ISO 45001 implementation including performance indicators and management review inputs, and conducts ISO gap assessments and internal audits. For K3 committees that want to record hazard reports and indicators digitally, HSE digitalization is an option.

Nusa Safety does not issue SMK3 certificates and does not carry out external SMK3 audits; the official assessment is made by an independent audit body appointed by the Minister, and our internal audit results do not guarantee an external audit outcome. This article is general information, not legal or professional advice for any specific case.

References

1. Heras-Saizarbitoria, I., Boiral, O., Arana, G., & Allur, E. (2019). OHSAS 18001 certification and work accidents: Shedding light on the connection. Journal of Safety Research, 68, 33-40. https://doi.org/10.1016/j.jsr.2018.11.003
2. International Organization for Standardization. (2018). ISO 45001:2018 Occupational health and safety management systems: Requirements with guidance for use. https://www.iso.org/standard/63787.html
3. Kementerian Ketenagakerjaan Republik Indonesia. (2025a). Peraturan Menteri Ketenagakerjaan Nomor 13 Tahun 2025 tentang Panitia Pembina Keselamatan dan Kesehatan Kerja [Minister of Manpower Regulation No. 13 of 2025 on Occupational Safety and Health Committees]. Berita Negara Republik Indonesia Tahun 2025 Nomor 954. https://jdih.kemnaker.go.id/asset/data_puu/2025pmnaker013.pdf
4. Kementerian Ketenagakerjaan Republik Indonesia. (2025b). Jumlah klaim kasus kecelakaan kerja, Januari-Desember tahun 2004 [sic; 2024 data] [Number of work accident claims, January to December 2024] [Data set]. Satudata Kemnaker. https://satudata.kemnaker.go.id/data/kumpulan-data/2447
5. Kementerian Ketenagakerjaan Republik Indonesia. (2025c). Perusahaan penerima penghargaan SMK3 (zero accident) tahun 2024 [Companies receiving SMK3 (zero accident) awards in 2024] [Data set]. Satudata Kemnaker. https://satudata.kemnaker.go.id/data/kumpulan-data/2778
6. Kementerian Ketenagakerjaan Republik Indonesia. (2026a). Perusahaan yang menerapkan SMK3, periode Januari tahun 2026 [Companies applying SMK3, January 2026] [Data set]. Satudata Kemnaker. https://satudata.kemnaker.go.id/data/kumpulan-data/3651
7. Kementerian Ketenagakerjaan Republik Indonesia. (2026b). Perusahaan yang menerapkan SMK3, periode Juni tahun 2026 [Companies applying SMK3, June 2026] [Data set]. Satudata Kemnaker. https://satudata.kemnaker.go.id/data/kumpulan-data/3656
8. Kementerian Ketenagakerjaan Republik Indonesia. (2026c). Perusahaan yang menerapkan SMK3, periode Juli tahun 2026 [Companies applying SMK3, July 2026] [Data set]. Satudata Kemnaker. https://satudata.kemnaker.go.id/data/kumpulan-data/4069
9. Kementerian Ketenagakerjaan Republik Indonesia. (2026d). Perusahaan yang menerapkan SMK3, periode September tahun 2026 [Companies applying SMK3, September 2026] [Data set and metadata]. Satudata Kemnaker. https://satudata.kemnaker.go.id/data/kumpulan-data/4071
10. Kementerian Ketenagakerjaan Republik Indonesia. (2026e). Perusahaan yang menerapkan SMK3, Tw. IV tahun 2025 [Companies applying SMK3, Q4 2025] [Data set]. Satudata Kemnaker. https://satudata.kemnaker.go.id/data/kumpulan-data/3224
11. Kementerian Ketenagakerjaan Republik Indonesia. (2026f). Jumlah klaim kasus kecelakaan kerja tahun 2025 [Number of work accident claims, 2025] [Data set]. Satudata Kemnaker. https://satudata.kemnaker.go.id/data/kumpulan-data/3246
12. Madsen, C. U., Kirkegaard, M. L., Dyreborg, J., & Hasle, P. (2020). Making occupational health and safety management systems 'work': A realist review of the OHSAS 18001 standard. Safety Science, 129, 104843. https://doi.org/10.1016/j.ssci.2020.104843
13. Pemerintah Republik Indonesia. (2012). Peraturan Pemerintah Nomor 50 Tahun 2012 tentang Penerapan Sistem Manajemen Keselamatan dan Kesehatan Kerja [Government Regulation No. 50 of 2012 on the Application of Occupational Safety and Health Management Systems]. Lembaran Negara Republik Indonesia Tahun 2012 Nomor 100. https://peraturan.go.id/files/pp50-2012pjl.pdf
14. Republika. (2026, August 30). Kemnaker dorong perusahaan tak berhenti pada sertifikasi K3 [Ministry urges companies not to stop at OSH certification]. Republika Online. https://ekonomi.republika.co.id/berita/tkkh9e349/kemnaker-dorong-perusahaan-tak-berhenti-pada-sertifikasi-k3
15. Republik Indonesia. (2026). Undang-Undang Nomor 27 Tahun 2026 tentang Pelindungan Ketenagakerjaan [Law No. 27 of 2026 on Labour Protection]. Lembaran Negara Republik Indonesia Tahun 2026 Nomor 179. https://jdih.setneg.go.id/detailperaturan?jns=UU&no=27&thn=2026
16. Robson, L. S., Clarke, J. A., Cullen, K., Bielecky, A., Severin, C., Bigelow, P. L., Irvin, E., Culyer, A., & Mahood, Q. (2007). The effectiveness of occupational health and safety management system interventions: A systematic review. Safety Science, 45(3), 329-353. https://doi.org/10.1016/j.ssci.2006.07.003
17. Viswanathan, K., Johnson, M. S., & Toffel, M. W. (2024). Do safety management system standards indicate safer operations? Evidence from the OHSAS 18001 occupational health and safety standard. Safety Science, 171, 106383. https://doi.org/10.1016/j.ssci.2023.106383
18. Zhafira, A. N. (2026, May 22). Menaker tegaskan BPJS Ketenagakerjaan harus jadi motor penggerak K3 [Minister says BPJS Ketenagakerjaan must drive OSH]. ANTARA News Jawa Timur. https://jatim.antaranews.com/berita/1067747/menaker-tegaskan-bpjs-ketenagakerjaan-harus-jadi-motor-penggerak-k3