HRM304 · Session 2 · After-class reference

What the scores explain — and what they miss

Six cards, six situations. For each one, this page lays out what the cultural frameworks predict, what they do not predict, and a single take-away.

Use it as a reference during the rest of the module and when preparing for the final exam. The four limits below are the names for what frameworks miss — they are exam-relevant.

Frameworks used on this page

  • Hofstede — six country-level dimensions, 0–100 scale. The cards show these directly.
  • GLOBE — leadership preferences at the level of cultural clusters (Nordic, Confucian, Anglo, etc.), 1–7 scale. Also on the cards.
  • Meyer's Culture Map — an applied framework used widely in practice. Eight scales that overlap with Hofstede but add useful ones, especially Trusting (task-based ↔ relationship-based) and Evaluating (direct ↔ indirect negative feedback). Not on the cards, but worth knowing for the cases.
  • Gelfand's tight ↔ loose — how strongly social norms are enforced in a culture. Tight cultures punish deviation; loose cultures tolerate it. Introduced in chunk 1 of this session.
Ecological fallacy
生态谬误
Treating a country average as a description of every individual. The individuals in the room may sit anywhere on the distribution.
Time-frozen data
数据滞后
The scores are a snapshot from prior decades. Cultures, generations, and policies move — sometimes fast.
Western framing
西方视角
The dimensions, questions, and assumed "good" HR practice were built in Anglo / Northern-European frames. What was never asked?
Culture ≠ institutions
文化≠制度
Many "cultural" differences are actually legal, economic, or political. Hofstede has no row for labour law, religion, or codetermination.
Card 1 Automotive · R&D · Chengdu

Sweden → China

Erik rotates junior leaders in the meeting. Juniors freeze, then start cc'ing seniors on every email. The project director objects to HR. Deadlines slip.

What the scores do predict
  • Power Distance Big gap (31 vs 80). A Swedish manager's "flat" instinct collides with a Chinese team that reads visible hierarchy as a sign the system is working. The cc'ing behaviour is what high power distance looks like when someone tries to flatten it — re-establishing the chain.
  • Individualism 71 vs 43. Swedish individualism says: speak as yourself, take credit yourself. Chinese collectivism filters speech through the senior who carries face for the unit.
  • Achievement / Motivation 5 vs 66 is the most dramatic gap on the card. Sweden's strong "care / quality of life" orientation makes Erik comfortable with consensus and ambiguity; the Chinese team is reading a competitive achievement environment and wants clear status signals.
  • GLOBE Participative Nordic 5.8 vs Confucian 5.0 — Erik's instinct (rotate juniors) is exactly the leadership style Nordic culture rewards.
Other frameworks add
  • Meyer · Communicating (low- vs high-context). Sweden is among the most low-context cultures; China is high-context. Erik's literal question — "what do you think?" — is itself too direct in a high-context room. The team reads silence, body language, and seniors' cues; Erik is asking them to break the channel.
  • Meyer · Evaluating. Sweden direct, China indirect — Erik's withdrawal of his own authority reads as a confusing signal, not as openness.
What the scores miss
  • Culture ≠ institutions The director is not just "a Chinese person who likes hierarchy". He has formal authority in the project structure. Erik's rotation undermines a role, not just a preference.
  • Ecological fallacy These are R&D engineers in Chengdu, many with overseas experience. The 80 score is a national average. Erik could probably find three engineers who personally test as low on power distance — the constraint is the group dynamic, not each individual.
  • Time-frozen Post-90s and post-00s engineers in tech sectors push back more than the score implies. The score doesn't know about generational change.
  • The cc-everything pattern is also a face-management tactic — there is no Hofstede dimension for face (面子).
Take-awayThe Hofstede scores would have warned Erik that "more participation" might not land the way it does in Gothenburg. They do not tell him the better move — which is to share authority with the senior, not in spite of him. Frameworks generate hypotheses; the design is on you.
Card 2 Engineering · Appraisal · São Paulo

Germany → Brazil

Munich rolls out a standardised appraisal with forced ranking. São Paulo completes it. Three top performers — including one in the succession plan — resign within four months.

What the scores do predict
  • Power Distance 35 vs 69. A German form treats each report as an independent unit to be assessed; the Brazilian frame reads "who is allowed to rank whom" as a question about position, not procedure.
  • Individualism 67 vs 36. Forced ranking is individualism in form: it requires you to separate your colleague from yourself in order to mark them lower. Brazil's collectivism makes this feel like betrayal — and exit interviews say exactly that.
  • Achievement Germany 66 vs Brazil 49. The "best to worst" logic aligns with achievement orientation; Brazil's lower score signals more pull toward relational maintenance.
  • GLOBE Team-Oriented L. America 6.0 (the highest in the cluster table). A ranking-based appraisal is structurally anti-team in a team-prioritising culture.
Other frameworks add
  • Meyer · Trusting (task-based ↔ relationship-based). This is the key one. Germany sits at the task-based end — trust is built by delivering on promises. Brazil sits at the relationship-based end — trust is built through personal connection that is then carried across tasks. Forced ranking rewards individual task delivery and ignores the trust network. Ranking colleagues against each other doesn't just feel cold; it damages the substrate on which the work runs.
  • Meyer · Evaluating. Germany direct, Brazil indirect — "two specific incidents to justify each score" is direct critique committed to paper. That format is exactly what an indirect-evaluating culture works to avoid.
  • Gelfand tight ↔ loose. Germany leans tight; Brazil leans loose. A uniform, standardised global form is tightness imposed on looseness — even before you get to the content.
What the scores miss
  • Culture ≠ institutions Brazil's CLT (Consolidação das Leis do Trabalho) strongly protects workers against dismissal. If bottom-ranked people can't easily be moved, what is the ranking for? Hofstede has no answer.
  • Culture ≠ institutions (parent-sub power) "Standardisation across 12 subsidiaries" is a control claim, not a culture claim. The conflict is partly Munich-vs-São-Paulo over local authority — not Germany-vs-Brazil over values.
  • Ecological fallacy São Paulo engineering is not Brazil-on-average. White-collar São Paulo professionals look much more individualist than the country score.
Take-awayMunich's response — "using the same form everywhere is the whole point" — is the textbook universalist error. The deeper question is what an appraisal is for in a relationship-trust culture with strong labour-law protection at the bottom of the ranking.
Card 3 Tech · Peer feedback · Tokyo

USA → Japan

A San Francisco firm rolls out anonymous quarterly peer feedback. Tokyo gives almost everyone 4s and 5s, with empty comment boxes. The Tokyo lead says: "the system is being used as designed."

What the scores do predict
  • Individualism The biggest individualism gap on any card (91 vs 46). US peer-feedback assumes individuals reporting on individuals. Japanese collectivism makes critique of in-group members near-impossible to write down — even anonymously.
  • Uncertainty Avoidance 46 vs 92. "Anonymous, but goes somewhere, and might affect career" is exactly the kind of ambiguous-consequence system Japan's very-high uncertainty avoidance rejects.
  • Long-Term Orientation 26 vs 88. A quarterly cycle is short-term; a working relationship is long-term. The cost of damaging a relationship that lasts 30 years is large; the cost of an empty comment box is small.
  • GLOBE Self-Protective Confucian 3.7 vs Anglo 3.1. Face-saving behaviour is leadership-rewarded in the cluster; this maps directly onto "all 4s and 5s, no comments".
Other frameworks add
  • Meyer · Communicating. Japan is the highest-context culture Meyer measures. "Empty comment boxes" is the communication — the absence is the signal, not the failure.
  • Meyer · Evaluating. Japan also the most indirect on negative feedback. Written, anonymous, traceable critique violates a deep workplace norm. Concepts like honne (本音, true feeling) and tatemae (建前, public stance) describe the same gap from inside the culture.
  • Gelfand tight ↔ loose. Japan is one of the tightest cultures Gelfand measured; the US is loose. In a tight culture, the right thing to do is follow the visible norm — and the visible norm is "don't write critical things about colleagues". The Tokyo lead's reply ("the system is being used as designed") is itself a tight-culture statement.
What the scores miss
  • Western framing The whole system design — "feedback = anonymous critique in writing" — is an Anglo management-school assumption. Japanese firms have plenty of feedback; it just isn't this form of it. Hofstede tells you the format will fail without telling you the format is provincial.
  • Culture ≠ institutions Japanese internal labour markets (long tenure, internal promotion) make written records of peer-critique far more career-consequential than they are at a US SaaS firm with high turnover. Same form, very different stakes.
  • "Anonymous" in a four-person random pool in an 80-engineer office is not anonymous if writing style is identifiable. A design flaw, not a cultural one.
Take-awayThis is the clearest "Western framing" card. The CPO's question — "why is Tokyo not using the system as designed?" — assumes the design is universal. The Tokyo lead's reply is correct: the team is giving honest feedback. Just not on this form.
Card 4 Retail · Graduate recruitment · Riyadh

UK → Saudi Arabia

600 applicants, 380 shortlisted, only 47 attend the assessment day, 12 accept. Top applicants from elite Saudi universities withdraw before the assessment day.

What the scores do predict
  • Power Distance Largest power-distance gap on any card (35 vs 95). A UK assessment centre with role-plays and group exercises lacks visible authority — to a candidate who expects clear hierarchy that signals "the people running this don't quite know what they want".
  • Individualism 89 vs 25. UK assessment centres surface individual potential through group exercises. Saudi collectivism makes the format conceptually contradictory.
  • Uncertainty Avoidance 35 vs 80. Saudi candidates expect clear criteria and structured process. "Show us how you'd handle this" reads as: "we have no system, and you may be judged on something arbitrary".
  • Indulgence 69 vs 52 — the informal/playful UK style reads as unserious.
Other frameworks add
  • Meyer · Trusting (relationship-based). Recruitment in a relationship-trust culture runs heavily through family, alumni, and wasta networks. Family approval and university reputation are not extra factors around the process — they are the process.
  • Meyer · Communicating. Saudi is high-context; the UK is lower. The signalling around an unfamiliar, low-status-looking format is itself information — candidates read "this firm doesn't know what it's doing" from cues the recruiter didn't even know they were sending.
  • Gelfand tight ↔ loose. Saudi Arabia is widely treated as a tight culture (norms strongly enforced, especially around gender and propriety). The UK leans loose. A tight culture's elite candidates have most to lose from engaging with a process that looks improper — even if it isn't.
What the scores miss
  • Culture ≠ institutions Mixed-gender role-plays are not a dimension issue. They run into gender norms grounded in religion, family structure, and (until recently) law. Hofstede has no row for any of these.
  • Time-frozen Saudi 2026 is not Saudi 2010. Vision 2030 has reshaped the labour market — women in the workforce, mixed workplaces opened, expanded private-sector hiring. The scores trail the change.
  • Ecological fallacy + signalling "Top applicants from elite Saudi universities withdrew" is a reputation story. Their outside options are real; engaging with a process that looks unserious is itself a reputational cost. Country averages don't capture market signalling.
Take-awayHofstede tells the London team that the format will feel wrong. It does not tell them which feature is the killer — and the killers are institutional (law, family, religion, reputation), not dimensional. "Adapt the form" is a useful brief; "build a Saudi recruitment process from the criteria up" is the harder, better one.
Card 5 E-commerce · 996 · Jakarta

China → Indonesia

Chinese product directors export the 996 rhythm to a Jakarta hub. A sprint review is scheduled for 7 pm on the second day of Ramadan; the room is half empty. The director tells Shanghai: "this team is not committed."

What the scores do predict
  • Power Distance Indonesia 78 vs China 80 — nearly identical. Hofstede would predict no clash on hierarchy. And there isn't one.
  • Individualism 14 vs 43. Indonesia is even more collectivist than China; the score would predict more deference to a foreign director, not less.
  • Uncertainty Avoidance 48 vs 30. Mild difference, but if anything the Indonesian team should prefer more structure.
  • On Hofstede, this card looks like the easiest of the six. The scores predict alignment, not friction.
Other frameworks add
  • Meyer · Scheduling (linear-time ↔ flexible-time). Indonesian (Southeast Asian) work is more flexible-time; Chinese tech is increasingly linear-time. The 7 pm meeting during Ramadan assumes that "the calendar is the schedule" — flexible-time cultures read time as social first, clock second.
  • Meyer's other scales (Trusting, Communicating, Disagreeing) actually also place China and Indonesia close together — relationship-based, high-context, conflict-avoiding. So Meyer doesn't differentiate them much either. Which is exactly the lesson: the major source of friction here sits outside the framework set entirely.
What the scores miss
  • Western framing — and a framework limit Indonesia is the world's most populous Muslim-majority country. Ramadan, iftar timings, and prayer schedules are non-negotiable, not preferences. Hofstede has no dimension for religion, and Meyer doesn't either. This is the strongest single illustration that the framework set is incomplete.
  • Culture ≠ institutions 996 is not "Chinese culture" — it is a contested corporate practice. In August 2021, the Supreme People's Court and the Ministry of Human Resources and Social Security jointly published typical labour-dispute cases ruling specific 996 schedules illegal under existing labour law. Enforcement is uneven and 996 persists in parts of the tech sector — but a workplace norm that is formally unlawful at home cannot be exported as "how China works".
  • The power-distance similarity actually worsens the misreading: the director assumes "they're like us, so the style works" — and the absence of pushback up the hierarchy looks (to him) like agreement, when it is the silence of a high-power-distance team in front of a foreign boss.
  • Ecological fallacy Jakarta tech engineers are a globally-mobile, mostly-urban workforce with options. The 14 individualism national score doesn't predict their willingness to walk.
Take-awayThis is the card where the scores don't tell you what's going on — and that is the lesson. Cultural frameworks (any of them) are not the full taxonomy of difference. Religion is the obvious miss here; there are others.
Card 6 Automotive · Restructuring · Stuttgart

China → Germany

Mr. Liu wants two roles merged "this week". Frau Bauer says it's a works council matter. Three weeks pass; the works council objects. Liu asks why Bauer "took the works council's side".

What the scores do predict
  • Power Distance 35 vs 80. Liu's "parent decides, HR implements" model is exactly what high power distance looks like. Bauer's consultative pacing is what low power distance plus codetermination produces.
  • Uncertainty Avoidance 65 vs 30. Germany's high uncertainty avoidance explains the demand for full documentation and statutory timelines; China's lower uncertainty avoidance tolerates "we'll work it out on the way".
  • GLOBE Participative Germanic 5.9 vs Confucian 5.0 — the German leadership ideal includes formal voice. Liu's directive style maps onto the Confucian cluster's lower participative score.
Other frameworks add
  • Meyer · Deciding (consensual ↔ top-down). The most useful Meyer scale for this card. Germany sits at the consensual end (decisions take longer but bind firmly once made); China at the top-down end (the boss decides; speed matters). Liu's "implement this week" doesn't just clash with Bauer's preferences — it skips an entire decision step.
  • Meyer · Leading. Germany egalitarian relative to most of the world; China hierarchical. From Liu's vantage, Bauer's deference to a works council looks like a manager refusing to manage.
  • Gelfand tight ↔ loose. Germany leans tight, especially around procedure. The "seven days with full documentation" is the visible form of that tightness, but the deeper tightness is the assumption that procedure is binding.
What the scores miss
  • Culture ≠ institutions This is the textbook example. Frau Bauer is not taking the works council's side. She is complying with the Betriebsverfassungsgesetz — the Works Constitution Act — which requires her to consult before personnel measures. The "seven days with full documentation" is a statute, not a preference.
  • Codetermination (Mitbestimmung) is the post-WWII institutional architecture of the German labour market. You can like it or hate it; you cannot opt out by being a foreign owner.
  • Ecological fallacy Mr. Liu is from an SOE (state-owned enterprise) — a specific organisational sub-culture, not a stand-in for "Chinese business". SOE governance formally includes a Party committee role in major personnel and strategic decisions (a feature of Chinese company law for SOEs, not informal influence), alongside longer planning cycles than private firms. "China score 80" overstates the cultural similarity of Liu to a private-firm Chinese manager, who might be just as frustrated with Bauer's pacing.
  • Time-frozen Codetermination has a specific history — post-war reconstruction, IG Metall, the 1970s Mitbestimmungsgesetz. The score doesn't tell you why; without the history, it looks like a "preference".
Take-awayFrau Bauer's "consensus-seeking" style is not personality and not really culture. It's law. The right question for Mr. Liu isn't "how do I work around her?" — it's "what does the law require, and what does it forbid?"

Patterns across the six cards

The four limits below are the names for what cultural frameworks miss. Each is anchored to the cards where it bites hardest.

Ecological fallacy · 生态谬误

Strongest in: Card 1 Card 2 Card 5

Every card features urban professionals with options, often globally mobile, often with outside experience. None of them are "the national average". The score is a hypothesis about distributions; the room contains individuals.

Time-frozen data · 数据滞后

Strongest in: Card 4 Card 6

Saudi Arabia 2026 is not Saudi Arabia 2010. German codetermination has a history that the score doesn't carry. Even China's Power Distance score of 80 is the product of long historical and institutional forces that are themselves in motion. Frameworks lag.

Western framing · 西方视角

Strongest in: Card 3 Card 5

The Tokyo card surfaces the framing problem inside the HR instrument (peer feedback as Western management technology). The Jakarta card surfaces it inside the framework set itself — religion is simply not in the dimensions, in Hofstede or in Meyer. Ask: what was never asked?

Culture ≠ institutions · 文化≠制度

Strongest in: Card 2 Card 5 Card 6

The biggest "cultural" differences on these cards turn out to be law: Brazil's CLT, Saudi religious-and-legal norms, Indonesia's Ramadan as a workplace fact, Germany's Betriebsverfassungsgesetz, China's own ruling on 996. If you only used Hofstede, you would mis-name every one of these as "culture".

Where this goes next

Frameworks are hypotheses, not answers

What the six cards show: the scores give you a shape of likely friction. They rarely give you the cause, and they almost never give you the fix. The cause is usually a mix of individual variation, institutional rules, and recent change — none of which the frameworks measure directly.

This is why scholars now use cultural frameworks alongside an institutional lens: law, labour markets, political economy. When something looks "cultural", it is worth asking which of those is also pushing on it. Most of the time, more than one.

The institutional lens returns in Session 9 (varieties of capitalism). For now: cultural frameworks are one tool, not the toolbox.