Rana Plaza — factory-ownership decision to reoccupy a condemned building
2013 · Catastrophic Failure · scored under OTA methodology v4
Scoring
Attribution weights under OTA methodology v4. Percentages express how much of the episode’s outcome each phase and modality accounts for — not a performance grade.
Phase attribution
Observe Easy-Correct · Think Easy-Wrong · Act Easy-Wrong
Modality weights
Modalities scored at zero weight are omitted; the case narrative records why an evidenced modality carries no independent weight.
- Primary modality
- Culture
- Reliability band
- High
- Fraud-related
- No
1. Episode summary
Rana Plaza was an eight-storey commercial building in Savar, on the outskirts of Dhaka, Bangladesh, housing five ready-made-garment factories that manufactured apparel for international brands including Primark, Benetton, Mango, Joe Fresh, The Children's Place, Walmart, Matalan and others. The building had been approved as a five-storey structure for shops and offices; three upper floors and heavy industrial machinery, including rooftop diesel generators, had been added in violation of the approved permit. [Storey count standardized 2026-07-25: the case previously gave inconsistent figures (eight-storey overall but "six-storey" approved + three added = nine) across different sections; four independent sources confirm the original permit was for five storeys, consistent with the eight-storey total already stated above.] On the morning of 23 April 2013, visible cracks — some reported at two inches deep — appeared in structural columns on the upper floors. Workers evacuated, a local television crew filmed the cracks, and a civil engineer called to the site recommended that the building remain closed pending a professional structural assessment. Shops and a bank located on the lower floors remained closed on 24 April. The factory owners on the upper floors, together with the building's owner Sohel Rana, ordered the approximately 3,000–5,000 garment workers back to work on the morning of 24 April 2013. Within hours of the morning shift starting, the building collapsed. Official investigations recorded 1,134 deaths and more than 2,500 injured, making it the deadliest garment-industry disaster on record. The strategic question the episode turned on was whether to accept the engineer's condemnation of the structure and forgo one day's garment production, or to override the warning and compel workers to re-enter an identified hazard.
2. Sources
Primary:
- Anne Trebilcock, "The Rana Plaza disaster seven years on: Transnational experiments and perhaps a new treaty?," International Labour Review, Vol. 159 (2020), No. 4.
- International Labour Organization (ILO), "The Rana Plaza building collapse… 100 days on," ILO news article, 2013; and "The Rana Plaza disaster ten years on: What has changed?" ILO InfoStories, 2023.
- Bangladesh Anti-Corruption Commission case filing against Sohel Rana and 17 others, 15 June 2014, reporting illegal construction of upper floors beyond permitted five-storey approval; Dhaka court indictment of Sohel Rana and 17 others for building-code violations, 14 June 2016, reported in Dhaka Tribune and Business & Human Rights Resource Centre. [Defendant count corrected 2026-07-25: previously "14 individuals" — four independent sources (Daily Star, Dhaka Tribune, OCCRP, Wikipedia) confirm 17 others (18 total with Rana).]
- Accord on Fire and Building Safety in Bangladesh, founding agreement text (15 May 2013) and subsequent inspection reports covering 2,000+ ready-made-garment factories.
Secondary (with justification):
- Quelch, John A., and Margaret L. Rodriguez, "Rana Plaza: Workplace Safety in Bangladesh (A)," Harvard Business School Case 514-034, September 2013 (revised June 2014). Synthesises documentary and interview evidence from the immediate aftermath of the collapse into a teaching account.
- Human Rights Watch, "Decade After Rana Plaza, Safety Flaws Persist," April 2023. Investigative synthesis of post-collapse reform trajectory drawing on Accord inspection data and worker interviews. 2a. Human Rights Watch, "'Whoever Raises their Head Suffers the Most': Workers' Rights in Bangladesh's Garment Factories," 22 April 2015 — documents, verbatim, that none of the five factories operating in Rana Plaza had a trade union and that workers were powerless to resist management pressure to enter the building; the direct source for the Structure "no trade union" finding and the Culture speaking-up/blacklisting norm.
- Organized Crime and Corruption Reporting Project (OCCRP), "Bangladesh: Rana Plaza Collapse Result of Corruption," investigative report summarising anti-corruption findings on the approval chain, permit violations, and politically connected ownership.
- Clean Clothes Campaign, "Rana Plaza" campaign file and "Justice still outstanding: an update of legal cases related to Rana Plaza eight years on" (2021). Aggregates labour-rights investigative reporting, court filings, and compensation-fund tracking.
- EHS Today, "Bangladesh: The Catastrophic Failure of the Apparel Industry's Factory-Inspection Regimes and the Birth of a New Model," EHS Today, 2013. https://www.ehstoday.com/safety/article/21915947/bangladesh-the-catastrophic-failure-of-the-apparel-industrys-factoryinspection-regimes-and-the-birth-of-a-new-model — Documents the pre-collapse failure of the BSCI social-audit regime to cover structural building safety; confirms TÜV Rheinland audit of Phantom Apparel Ltd. failed to flag structural deficiencies despite "obvious flaws in construction and building safety." Added during §4 research to anchor Processes modality evidence. [Added §4 research, 2026-06-04]
Tertiary (flagged):
- Wikipedia, "Rana Plaza collapse" article, used for cross-checking dates, casualty counts, and buyer-brand lists against the primary and secondary sources above; not relied on for load-bearing factual claims.
3. OTA narrative
Observe. The observation apparatus produced the right signal, clearly and in time. Cracks in structural columns on the upper floors were visible to workers, to factory management, to the building's owner, and to a local television crew on 23 April 2013. A civil engineer was called to the site and made an unambiguous recommendation that the building be closed pending professional structural assessment. The shops and the bank in the lower floors acted on the signal and did not reopen. The observation that Rana Plaza was structurally unsafe on the morning of 24 April 2013 was not a subtle or contested signal requiring expert reading; it was the plainest possible form of the observation problem, and it had already been resolved into an explicit expert verdict the preceding afternoon. Observe was not a root cause of this failure. It was the transmission step that delivered an accurate, timely, expert-validated warning into the hands of the decision-makers, who then acted against it.
Think. The reasoning step was the operative root cause. The question in the hands of the factory owners and the building owner on the evening of 23 April and the morning of 24 April 2013 was whether the cost of one day's lost garment production and the risk of missed shipment deadlines outweighed an engineer's explicit condemnation of the structure housing several thousand workers. Under any defensible framework available to reasonably-resourced industrial-building operators in 2013 — including Bangladesh's own National Building Code, standard industrial safety practice, and the operators' own prior experience of the 2012 Tazreen Fashions fire — the correct answer was to keep the building closed. The framework existed, was accessible, and had been made concrete in a specific engineer's recommendation the previous day. It was not applied. The reasoning failure is therefore Easy-Wrong Think: the correct interpretation of the signal was routine for the peer group of industrial building operators, and the decision to override the engineer's verdict was the interpretive failure that converted the warning into a mass-casualty outcome. Think is the root-cause phase in this episode.
Act. Act carries root-cause weight alongside Think. Implementation of the wrong decision was not a passive follow-on: factory managers actively ordered workers to return, some reportedly threatened to withhold a month's pay from workers who refused, and contemporaneous accounts describe coercion of reluctant workers at the building's entrances. The upstream building-code violations — illegally added floors, heavy generators placed on a structure not designed for them, construction on filled pond land — are execution failures that long predated 23 April 2013 and that left the building with no residual safety margin to absorb the 24 April re-entry decision. Both the long-run execution failures in the building's construction and the acute execution of the forced reoccupation belong to Act. Act is classified Wrong at the easy end of the task-difficulty axis: a routine industrial-safety move — respect an engineer's written condemnation of a cracked structure — was not merely omitted but actively reversed. Act is a root-cause phase in this episode; Think and Act are jointly load-bearing, with Think carrying the interpretive decision and Act carrying both the pre-existing structural compromise and the coercive reoccupation.
4. Modality evidence
Direction.
The decisive directional act in this episode was a specific, attributable, dated choice: on the evening of 23 April 2013, building owner Sohel Rana told journalists that Rana Plaza was "safe" and that workers "should return" the following morning (Wikipedia, "Rana Plaza collapse"; HBS Case 514-034, Quelch/Rodriguez). This was not ambiguity or inaction — it was an explicit public commitment to the reoccupation posture made after an engineer had already delivered a written condemnation of the structure. The five factory managements operating on the upper floors translated that directional posture into operational orders on the morning of 24 April: produce was not to be lost, the morning shift was to proceed, and resistance from workers was to be overcome. The direction chosen was therefore clear, specific, and attributable to Rana as building owner and to the factory managements as operators, and it ran directly against the only expert input that had been sought (ILO 2020 Working Paper; HBS Case 514-034). The directional failure was not a failure to perceive the problem — Observe had already delivered an accurate signal — but a deliberate choice to treat production continuity as more valuable than an explicit structural condemnation.
Sohel Rana's political affiliation — as Senior Joint Convenor of the Jubo League, the youth wing of the ruling Awami League in the Savar municipality unit — is relevant directional context: it shaped his confidence that the consequences of ignoring safety enforcement would be manageable, a confidence that the subsequent decade-long delay in the murder trial retrospectively confirmed (OCCRP, "Bangladesh: Rana Plaza Collapse Result of Corruption"; Wikipedia, "Sohel Rana (businessman)"). Direction evidence passes the specificity, timing, and attribution bar: the choice is identifiable, datable to 23–24 April 2013, and attributed to named decision-makers in primary and secondary sources.
Structure.
Three structural features of the episode created the conditions in which the directional choice could be made and enforced without challenge. First, Rana Plaza's construction approval chain was corrupt and politically configured: the original permit covered a five-storey commercial building; three additional industrial floors were added without authorisation, approved by the Savar Mayor — an Awami League member who lacked the legal authority to issue such a permit — and enabled by bribes to local officials (Bangladesh Anti-Corruption Commission filing, 2014; Dhaka court indictment, 2016; OCCRP). The structural integrity of the building was therefore already compromised by the time 24 April arrived, because the governance mechanisms that should have blocked illegal construction had been captured by the building's politically connected owner.
Second, none of the five garment factories operating in Rana Plaza had a trade union, and so workers were powerless to resist their managers who ordered, threatened, and cajoled them to enter the doomed building a day after large cracks had appeared in it (Human Rights Watch, "Whoever Raises their Head Suffers the Most," 22 April 2015). The formal labour-relations architecture provided no organisational channel through which workers could collectively refuse the reoccupation order or escalate safety concerns to external authorities. Workers were structurally isolated as individuals facing coordinated management pressure.
Third, the building permit and inspection authority was fragmented across the Savar municipality, the Bangladesh National Building Code enforcement apparatus, and the industrial police, with no single body having clear, exercisable authority to compel closure when structural risk was identified (Bangladesh Anti-Corruption Commission filing; Dhaka court indictment; ILO 2013 news article). The engineer's recommendation had no formal enforcement mechanism attached to it; the industrial police had requested evacuation but could not compel it in the face of owner resistance.
Processes.
The pre-collapse industry audit and inspection regime failed to function as a process for surfacing structural risk. Social compliance audits conducted under the BSCI standard had been carried out at two Rana Plaza factories — including a TÜV Rheinland audit of Phantom Apparel Ltd. — before the collapse. Those audits did not cover structural building safety, a gap the auditing firms acknowledged after the fact; the BSCI audit questionnaire required auditors to check building permits, but discrepancies between the permit and the actual number of floors were not flagged (EHS Today, "Bangladesh: The Catastrophic Failure of the Apparel Industry's Factory-Inspection Regimes"; Accord on Fire and Building Safety founding agreement, 2013, which was constituted precisely because the pre-existing process did not work). The pre-collapse audit framework was therefore a process that certified factories for labour standards while remaining structurally blind to the physical buildings housing them.
At the building level, no routine structural inspection process for existing commercial-industrial buildings was in operation by the Savar municipality. The engineers called to Rana Plaza on 23 April 2013 were called ad hoc in response to the cracks appearing — there was no periodic inspection cycle that would have caught the progressive deterioration caused by illegal additional floors and rooftop generator loads (ILO 2020 Working Paper; HBS Case 514-034). The Accord established after the collapse specifically created the structural-inspection process that had been absent: binding, expert, public, and enforceable (Accord on Fire and Building Safety founding agreement).
Scoring note (zero-modality rationale): the Processes contribution described in this subsection is classified at the boundary with Culture in the scoring record — the §4 evidence locates the operative driver of the episode's failure causation in Culture rather than in a standalone Processes contribution. Processes is acknowledged in narrative as evidenced but does not carry independent weight in the scoring; weight is borne by Direction, Structure, Culture. Categorisation under METHODOLOGY-ota-scoring-v4.md §5: classification boundary with an adjacent modality.
Capability.
The capability most directly relevant to this episode was structural risk assessment — the ability to interpret visible cracking in load-bearing columns of a building with known illegal floors and heavy rooftop machinery as a signal requiring closure. That capability existed outside the building's ownership and management structure: the civil engineer called on 23 April 2013 possessed it and delivered an unambiguous verdict. What was absent was any structural risk assessment capability inside the factory owner and building owner group — no in-house engineer, no standing relationship with a structural consultancy, no internal competence that would have allowed factory managers to evaluate the engineer's verdict independently or to push back on Rana's "safe" characterisation (ILO 2020 Working Paper; HBS Case 514-034).
The capability gap is therefore not that the required expertise did not exist in Bangladesh's environment — it manifestly did — but that the factory operators had no capability to engage with it, no institutional knowledge about how to interpret structural warnings, and no prior experience of being asked to close production on safety grounds that would have built such judgment. This is a genuine capability absence rather than a structural mis-placement of existing capability: the relevant skill and knowledge did not exist within the operator group, and its absence left factory managers dependent on Rana's assertion rather than independent professional judgment (ILO 2013 news article). The capability gap is flagged as relatively thin in primary sourcing — no primary investigation specifically enumerates the technical competence profile of the factory operators.
Scoring note (zero-modality rationale): the Capability contribution described in this subsection is classified at the boundary with Culture in the scoring record — the §4 evidence locates the operative driver of the episode's failure causation in Culture rather than in a standalone Capability contribution. Capability is acknowledged in narrative as evidenced but does not carry independent weight in the scoring; weight is borne by Direction, Structure, Culture. Categorisation under METHODOLOGY-ota-scoring-v4.md §5: classification boundary with an adjacent modality.
Culture.
The cultural mechanisms operating on 24 April 2013 ran in two directions: downward coercion of workers and upward impunity of owners. At the worker level, multiple factory managements used direct economic threat to enforce reoccupation. Managers at Ether Tex threatened to withhold a full month's pay from workers who refused to enter the building; other factories made equivalent threats; contemporaneous accounts describe hired individuals at the building's entrances compelling reluctant workers to proceed through (Wikipedia, "Rana Plaza collapse"). [Citation corrected 2026-07-25: this specific claim was previously cited to HBS Case 514-034, Clean Clothes Campaign 2021, and the ILO 2020 Working Paper, none of which contain it; the claim is true and verbatim in Wikipedia's "Rana Plaza collapse" article.] Workers had no trade union protection and operated within an industry-wide norm in which publicly complaining about safety was associated with dismissal and informal blacklisting — a norm documented by Human Rights Watch in the Bangladesh garment sector (HRW, "Whoever Raises their Head Suffers the Most," 2015). The cultural environment for production-floor workers was one of structurally enforced powerlessness, in which even an individually rational and visible safety risk could not be declined without economic penalty.
At the ownership level, the cultural norm was one of impunity secured by political connection: Rana's Jubo League affiliation was not incidental but part of a broader Bangladesh ready-made-garment industry pattern in which building owners and factory operators who violated safety rules faced negligible enforcement, because the enforcement apparatus was politically penetrated and the garment industry's foreign-exchange earnings made it systemically protected from aggressive oversight (OCCRP; ILO 2020 Working Paper; ILO 2023 InfoStories). The combination — workers culturally prevented from refusing, owners culturally confident of non-enforcement — closed off every informal channel through which the expert warning of 23 April 2013 might have been honoured without the factory owners choosing to honour it.