The Harvard and Michigan Models of HRM: Competing Perspectives on People Management

Article 16 traced the idea that HR should be matched to strategy, and Article 11 introduced John Storey's distinction between hard and soft HRM. This article examines the two founding frameworks themselves, in the books' own wording wherever it could be checked.

Managing Human Assets, by Michael Beer, Bert Spector, Paul Lawrence, D. Quinn Mills and Richard Walton, grew out of teaching. A view had developed among Harvard Business School alumni and faculty that a new required course in human resource management was needed; "That course was launched in 1981," the authors write, the first new required course since Managerial Economics "nearly 20 years ago" (Beer et al., 1984). The book addressed general managers, and its central claim concerned them: "General managers, not personnel or labor relations specialists, make conscious or unconscious choices" that shape the employment relationship. Its definition of the subject was correspondingly wide. HRM "involves all management decisions and actions that affect the nature of the relationship between the organization and employees."

Strategic Human Resource Management, edited by Fombrun, Tichy and Devanna, collected research and practitioner chapters. Its framework chapter, "A Framework for Strategic Human Resource Management" (Devanna, Fombrun and Tichy, 1984), built on articles the three had published since 1981. The "Michigan" label is a convenience. When the book appeared, Tichy was a professor of organizational behavior at the University of Michigan's Graduate School of Business, Fombrun was on the faculty of the Wharton School, and Devanna was research coordinator at a centre in Columbia University's Graduate School of Business.

The books ask different questions. The Michigan framework asks how HR systems can be designed to deliver a given strategy. The Harvard framework asks what management's choices about people do, to whom, and with what long-term consequences. Most of the contrasts that follow come from that difference.

The Harvard framework is a diagram, the "Map of the HRM Territory" (Beer et al., 1984, Figure 2-1): five groups of elements joined by arrows, with a feedback loop from long-term consequences back to the conditions that shape later choices.

Three features of the map are easy to miss when it is reduced to a slogan.

First, the policy areas are not the familiar list of HR department functions. Employee influence comes first, and the book frames it as "a key question that all managers must ask: How much responsibility, authority, and power" the organization should delegate to employees. Human resource flow concerns the movement of people "into, through, and out of" the organization. Work systems concern how work is organized and jobs are designed, and the book gives particular attention to what it called "high-commitment work systems." Selection and appraisal, which head the Michigan cycle, do not appear as headings of their own.

Second, the four Cs are intermediate outcomes. Commitment and competence describe what employees bring; congruence asks "What levels of congruence do HRM policies and practices generate or sustain between management and employees"; cost effectiveness counts turnover, absence and strikes as well as wages. The authors said the four Cs "are not exhaustive of the criteria" managers might use. They ask whether a policy is building or wearing down the employment relationship before its long-term consequences show.

Third, the long-term consequences include "societal well-being" alongside individual well-being and organizational effectiveness. A redundancy programme that improves this year's costs while damaging the local labour market has, on the map, costs the income statement does not record. This is what makes the model stakeholder-pluralist: employees are one of several parties whose interests count independently of the owners'.

The book did not assume those interests coincide. "Organizations constantly make trade-offs, either implicitly or explicitly," the authors wrote, between business goals and the needs of employees and society, and they warned that "Some managers may be lulled into believing that all stakeholders have the same interests and points of view." Their answer was top management "balancing and, where possible, integrating the interests of the many stakeholders," through general managers who "must develop a role and perspective that allows them to manage the trade-off process in such a way as to minimize differences between stakeholders." Section 5 asks whether that answer is enough.

Devanna, Fombrun and Tichy started from strategic management, not from the employment relationship. Their framework treats mission and strategy, formal structure, and human resource management as interrelated elements of an organization "embedded in a turbulent environment," which the book describes in terms of the "turbulent economic, political, and cultural forces of the 1980s." The managerial problem follows: "The critical managerial task is to align the formal structure and the human resource systems so that they drive the strategic objectives of the organization" (Devanna, Fombrun and Tichy, 1984, p. 37).

The cycle's internal logic is tighter than the Harvard map's. Performance is the product of the other four systems, and each feeds the next: appraisal supplies the information for reward and development decisions, and development changes who can be selected later. Following Robert Anthony, the chapter examined each system at the strategic, managerial and operational level, so that selection, for instance, involves both long-range questions about the people the business will need and the filling of today's vacancies.

Two features explain why the framework has been read as "hard." The first is its dependent variable. The Harvard map ends in three kinds of long-term consequence; the Michigan cycle ends in performance, and the strategy defines what performance means. The second is what the cycle leaves out. Employees are selected, appraised, rewarded and developed, but the framework has no component through which they influence any of those decisions. Neither feature makes the model indifferent to employees. Its reward system aims at "the equitable distribution of rewards," and development looks ahead to employees' future positions. But both concerns are justified by what they contribute to performance.

The table is this article's summary of the two frameworks, not a reproduction of either book.

The usual mapping, Harvard soft and Michigan hard, has scholarly support. Karen Legge (1995) identified two normative models in the HRM literature: the "soft," or "developmental humanism," model of the Harvard School and the "hard," or "utilitarian-instrumentalism," model of the Michigan School, taking the labels, as she states, from John Storey's 1987 work and from Chris Hendry and Andrew Pettigrew (1990). Article 11 discussed Storey's distinction, so two cautions will do here.

The first is that the mapping concerns emphasis, not exclusive content. The Harvard model counts cost effectiveness among its outcomes; the Michigan model includes equitable rewards and development. Either could justify cost cutting or heavy investment in people.

The second is that the real dividing line is the table's fourth row. For the Michigan model, employee interests matter because, and to the extent that, they serve the strategy. For the Harvard model they matter in their own right, and an HR policy that serves the strategy while damaging employee or societal well-being has failed on part of the model's own criteria. That is a disagreement about the purpose of management. Research on HR and firm performance can show whether particular practices pay, but it cannot settle whose interests a manager ought to count.

The Harvard model and the problem of conflicting interests. The strongest criticism of the Harvard model is that it identifies competing interests without saying how to decide between them. Its authors, as Section 2 showed, acknowledged trade-offs and warned against assuming harmony. Their remedy, though, was a perspective, not a rule: general managers were to balance interests, integrate them "where possible," and "minimize differences between stakeholders." The map gives no guidance where differences cannot be minimized, as with a closure that serves shareholders and removes a town's main employer. And the person doing the balancing is not neutral: management is itself one of the stakeholder boxes.

Legge (1995) pressed the point further. In her reading, the normative HRM models, like the normative models of personnel management before them, "assume a unitary frame of reference": that "there is a long-run coincidence of interests between all the various stakeholders of the organisation," as Beer and Spector (1985, p. 283) put it. "Even where potential union problems are recognised," she added, "cooptation is identified as the way forward." Richard Walton's (1985) statement of the commitment model shows the underlying theory: policies of "mutuality" were expected to "elicit commitment which in turn will yield both better economic performance and greater human development" (as quoted in Legge, 1995). The language is pluralist; the expected end state is harmony.

Legge also asked what becomes of the organization that "chooses to compete in a labour intensive, high volume, low cost industry, generating profits through increasing market share by cost leadership." For such a firm the commitment route may be unaffordable. Section 7 meets that firm in practice.

The model has defenders. John Bratton and Jeff Gold (2017) accept that it "is still vulnerable to the charge of 'unitarism'" but judge it "a much more pluralist frame of reference than is found in later models." Peter Boxall (1992), as summarized by Michael Armstrong and Phil Long (1994), credited it with incorporating "recognition of a range of stakeholder interests" and with recognizing trade-offs. A map that makes conflict visible is still more useful than one that leaves it out.

The Michigan model and people as a resource to be matched. The analytical criticism of the Michigan model, examined in Article 16, concerns the matching assumption itself. Legge (1995) put her version sharply: consciously matching HR policy to business strategy "is only relevant if one adopts what, empirically speaking, is the least realistic model of the strategy-making process."

The normative form concerns what the model makes of employees. In the cycle, people are inputs selected, measured, paid and developed for purposes set elsewhere: Legge's "utilitarian instrumentalism." Bratton and Gold (2017) list the model's weaknesses as "its prescriptive nature and its focus on just four HR practices," and add: "It also ignores different stakeholder interests, situational factors and the notion of management's strategic choice." They also credit its strength, which is that it expresses "the coherence of internal HR policies and the importance of 'matching'" them with business strategy.

The objection is more precise than a charge of coldness. A framework whose only criterion is fit cannot distinguish a well-aligned system that develops people from a well-aligned system that wears them out. It can say which is better aligned, but not which is acceptable.

The answer depends on where influence is measured, and part of what follows is interpretation. In research, the Michigan question has framed most of the field. Patrick Wright and Michael Ulrich (2017) describe the strategic HRM that emerged after the 1984 books in terms of three shifts: the level of analysis moved from individuals to organizations, attention moved from single practices to systems of practices, and HR came to be discussed in the context of the firm's strategy. The empirical programme of the 1990s onward, discussed in Article 15, asked whether HR systems aligned with strategy improve firm performance: the Michigan question, asked with better data. Thirty years on, Beer, Paul Boselie and Chris Brewster (2015) criticized "the assumption that the business of HRM is solely to improve returns to owners and shareholders" and called for "a wider, more contextual, more multilayered approach founded on the long-term needs of all relevant stakeholders." Beer (2015) went further: "we do not need more proof that HRM practices are associated with financial performance."

That verdict is contested. The same 2015 issue of Human Resource Management carried Bruce Kaufman's thirtieth-anniversary review, which used two recent strategic HRM books as a benchmark for tracing how the field had developed since the founding books, "largely in an American context" (Kaufman, 2015). Wright, David Guest and Jaap Paauwe (2015), authors of one of those recent books, replied that Kaufman had chosen "the wrong target" and that his analysis "seemed to miss the mark," because the field had already been dealing with the issues he raised for years. Wayne Cascio (2015) asked for "models that reflect a broader view of performance." Whether the field narrowed or specialized remains a live argument.

In teaching, the Harvard model has fared better. The British textbooks consulted for this article (Armstrong and Long, 1994; Legge, 1995; Bratton and Gold, 2017) all present the two models side by side, and Bratton and Gold describe the Harvard model as "one of the first comprehensive statements" of what HRM is. Whether either model has shaped what managers actually do is harder to establish. No systematic evidence on practitioners' use of either framework was found for this article, and claims that a company "follows" one model usually rest on labels applied after the fact.

The events below are real and dated, drawn from press reports and an ILO study listed in the references. Reading them through the two models is this article's interpretation.

Tea workers' daily wages in Bangladesh were set by agreement, normally every two years, between the Bangladesh Cha Sramik Union and the owners' Bangladesh Tea Association (BTA). By August 2022 the previous agreement period had expired more than 19 months earlier (Gain, 2022a), and the cash wage in top-category gardens was Tk 120 a day. Two-hour daily stoppages began on 9 August, and an indefinite stoppage in 166 gardens on 13 August, demanding Tk 300 (The Daily Star, 13 August 2022). The BTA replied that wages and benefits, counting subsidized rations, housing and medical care, were worth "at least Tk 400 per day" (The Daily Star, 17 August 2022). An interim rise to Tk 145, announced on 20 August on the Prime Minister's directive, did not end the stoppage (The Daily Star, 20 August 2022).

On 27 August Prime Minister Sheikh Hasina met garden owners and fixed the daily wage at Tk 170, with increases in other benefits such as the plucking bonus (Prothom Alo, 27 August 2022). Workers returned the next day after 15 days; most of those The Daily Star spoke to were dissatisfied and said they had no alternative (The Daily Star, 28 August 2022). Two days later the BTA claimed that the new package was worth nearly Tk 540 a day. Philip Gain, who has researched tea workers' conditions for over three decades, called such calculations "absurd," arguing that the Labour Act's definition of wages excludes most of the items the owners had priced (Gain, 2022b).

The wage then moved into the state's machinery. A gazette notification in August 2023 fixed Tk 170 as the minimum daily wage in category-A gardens, with a 5 percent annual increase and a review every two years involving owners, workers and government (The Financial Express, 13 August 2023). By 2025 the wage was Tk 187 (The Daily Star, 4 April 2026). In August 2026 the BTA declared Tk 196 under the annual increment; workers' leaders called it unilateral, and protests spread behind a demand for Tk 500 (The Daily Star, 17 August 2026). No settlement had been reported in the sources reviewed.

One structural fact runs through the dispute. An ILO study found that anyone not working, or not registered as a permanent worker, in a garden "is not permitted to live in the labour line," and concluded that "Tea garden workers are like ancestral contract labour as their employment also secures their accommodation" (Ahmmed and Hossain, 2016).

Through the Harvard lens. Every stakeholder box on the map is occupied: owners, garden managers, permanent and casual workers, a union, government from the Department of Labour to the Prime Minister, and the labour-line communities. The situational factors are unusually binding: few alternative jobs, housing tied to employment, and a legal definition of wages. The map also shows why the dispute could not be confined to pay. Because employment secures accommodation, a reward-system decision is at the same time a human resource flow decision and a decision about community welfare. On the four Cs, the owners' valuation argument was a claim about cost effectiveness; the strike was evidence that congruence had broken down, and the book itself counts strikes among the costs of a policy.

What the map cannot do is say what the wage should be. Tk 120, Tk 145, Tk 170 and Tk 300 can each be presented as a balance of stakeholder interests. Managers were in no position to "minimize differences between stakeholders": the parties disagreed about the facts (what the benefits were worth), the law (what counts as a wage) and the outcome. The decision came from the party with authority over the others, first the Prime Minister and then a statutory wage process. The Harvard model treats government as a stakeholder whose interests management balances; here government set the outcome. Recognizing interests is not the same as having a way to decide between them.

Through the Michigan lens. The cycle describes the gardens' system as coherent. The BTA presents tea as an import-substituting industry, with Bangladesh producing about 3 percent of world output (The Daily Star, 17 August 2022). A low-cost strategy matched by a low cash wage, in-kind benefits that bind workers to the estates, a workforce drawn from families who have lived in the labour lines for generations, and little development beyond the job is, in Michigan terms, aligned. The model explains the owners' position well: the wage is a design variable set by what the strategy can bear. This is Legge's low-cost firm, where a framework whose only criterion is fit cannot say whether the system is acceptable, and can treat the state's intervention only as a change in the environment.

Three cautions apply. Reported counts of gardens in 2022 ranged from 166 to 241 between sources; the owners' valuations of in-kind benefits are disputed; and the union's representativeness has itself been questioned, since its last election was held in 2018 (The Daily Star, 17 August 2026). The comparison's point is narrower than a verdict on the industry. The Harvard map shows who is affected and why the issue is more than pay; the Michigan cycle explains why the system persists. Neither says how to decide when interests cannot be reconciled.

For each practice, decide which model's reasoning it follows most closely, then read the commentary. Some practices can be defended under either model, and the commentary explains what would tell them apart. The practices are constructed for teaching.

This case is fictional and was constructed for teaching. The company, people and figures are invented.

Kopotakkho Seafoods, a shrimp processor near Khulna, has exported block-frozen shrimp for fifteen years. Its new managing director wants to move into peeled, cooked and ready-to-cook products for supermarket chains, which pay more but require consistent quality, traceability and a stable workforce. About 900 people work in the plant, most of them women; around 400 are hired each season through labour contractors and paid piece rates for peeling. Turnover among contract workers is high, and the rejections in the first supermarket trial were traced partly to newly hired peelers. The HR manager, Shirin Akhter, has been asked to put an HR plan to the board.

What the exercise is designed to show. The Michigan cycle asks what HR must deliver; the Harvard map asks what each way of delivering it does, and to whom. A plan built on one model alone tends to fail in a predictable way: the first ignores the causes of the problem, and the second never states what success requires.

The Harvard and Michigan models are usually presented as rival answers to the same question. Read closely, they answer different ones. Devanna, Fombrun and Tichy asked how HR systems can be designed to drive a given strategy, and their human resource cycle remains a clear statement of that task. Beer and his colleagues asked what management's choices about people do to everyone with a stake in the organization, over the long term as well as the short. The first question set the research agenda; the second survived in teaching and has been revived by critics of the field's narrowing.

Neither model supplies a way to decide when interests cannot be reconciled. The Michigan model avoids the problem by making strategy the only criterion. The Harvard model sees the problem and hands it to the general manager, who is also an interested party. In Bangladesh's tea gardens, where the conflict was open and the parties disagreed even about what the wage was, the state made the decision. For managers elsewhere the implication is practical: a stakeholder map shows whose interests are at stake but is not a method for choosing between them, so anyone using it should be clear about who has the authority to decide, and on what grounds. Article 18 turns to the AMO model of how HR practices shape individual performance.

Select a significant HR decision taken by a real organization in the last five years: a restructuring, a pay settlement, an automation programme, a change in employment contracts, or a plant closure. You may use your employer (disguised if necessary) or a decision documented in published sources.

Prepare a report of 2,500 to 3,000 words that:

1. describes the decision, its business context and the evidence available about it, distinguishing documented facts from your inferences;
2. analyses the decision using the Harvard map: identify the stakeholder interests and situational factors, the policy areas affected, and the likely effects on the four Cs and on long-term consequences;
3. analyses the same decision using the Michigan model: state the strategy it was meant to serve and assess how far the selection, appraisal, reward and development systems were aligned with it;
4. identifies at least one point where the two analyses lead to different judgements, and explains which assumption about employee interests produces the difference;
5. evaluates how conflicts between stakeholders were in fact resolved (by management, negotiation, regulation or another route) and whether either model helps explain or improve that process; and
6. recommends how the organization should approach a similar decision in future, stating the trade-offs openly and who should decide them.

Use at least five academic sources, including at least three from this article's reference list. Quote the models' authors only where you have checked the wording, and say where your evidence about the decision is incomplete.

Assessment rubric (100 marks)

1. Beer, M., Spector, B., Lawrence, P. R., Mills, D. Q., & Walton, R. E. (1984). Managing human assets. Free Press. Internet Archive record: https://archive.org/details/managinghumanass00beer
2. Devanna, M. A., Fombrun, C. J., & Tichy, N. M. (1984). A framework for strategic human resource management. In C. J. Fombrun, N. M. Tichy, & M. A. Devanna (Eds.), Strategic human resource management (pp. 33–51). John Wiley & Sons. Internet Archive record: https://archive.org/details/strategichumanre00fomb
3. Legge, K. (1995). Human resource management: Rhetorics and realities. Macmillan Business.
4. Beer, M., Boselie, P., & Brewster, C. (2015). Back to the future: Implications for the field of HRM of the multistakeholder perspective proposed 30 years ago. Human Resource Management, 54(3), 427–438. https://doi.org/10.1002/hrm.21726
5. Kaufman, B. E. (2015). Evolution of strategic HRM as seen through two founding books: A 30th anniversary perspective on development of the field. Human Resource Management, 54(3), 389–407. https://doi.org/10.1002/hrm.21720
6. Wright, P. M., Guest, D., & Paauwe, J. (2015). Off the mark: Response to Kaufman's evolution of strategic HRM. Human Resource Management, 54(3), 409–415. https://doi.org/10.1002/hrm.21723
7. Wright, P. M., & Ulrich, M. D. (2017). A road well traveled: The past, present, and future journey of strategic human resource management. Annual Review of Organizational Psychology and Organizational Behavior, 4, 45–65. https://doi.org/10.1146/annurev-orgpsych-032516-113052
8. Bratton, J., & Gold, J. (2017). Human resource management: Theory and practice (6th ed.). Palgrave.
9. Ahmmed, F., & Hossain, M. I. (2016). A study report on working conditions of tea plantation workers in Bangladesh. ILO Country Office for Bangladesh. https://www.ilo.org/media/423846/download

Other sources cited

• Armstrong, M., & Long, P. (1994). The reality of strategic HRM. Institute of Personnel and Development.
• Armstrong, M., & Brown, D. (2019). Strategic human resource management: Back to the future? A literature review (IES Report 517). Institute for Employment Studies and CIPD. https://www.employment-studies.co.uk/system/files/resources/files/517_Strategic-Human-Resource-Management-Back-to-the-future-IES-CIPD-2019.pdf (source of the page number for the Devanna, Fombrun and Tichy quotation)
• Beer, M. (2015). HRM at a crossroads: Comments on "Evolution of strategic HRM through two founding books: A 30th anniversary perspective on development of the field." Human Resource Management, 54(3), 417–421. https://doi.org/10.1002/hrm.21734
• Beer, M., & Spector, B. (1985). Corporate wide transformations in human resource management. In R. E. Walton & P. R. Lawrence (Eds.), Human resource management: Trends and challenges (pp. 219–253). Harvard Business School Press.
• Boxall, P. F. (1992). Strategic human resource management: Beginnings of a new theoretical sophistication? Human Resource Management Journal, 2(3), 60–79. https://doi.org/10.1111/j.1748-8583.1992.tb00260.x
• Cascio, W. F. (2015). Strategic HRM: Too important for an insular approach. Human Resource Management, 54(3), 423–426. https://doi.org/10.1002/hrm.21735
• Hendry, C., & Pettigrew, A. (1990). Human resource management: An agenda for the 1990s. International Journal of Human Resource Management, 1(1), 17–43. https://doi.org/10.1080/09585199000000038
• Walton, R. E. (1985). Toward a strategy of eliciting employee commitment based on policies of mutuality. In R. E. Walton & P. R. Lawrence (Eds.), Human resource management: Trends and challenges (pp. 35–65). Harvard Business School Press.

Bangladesh news sources

• Gain, P. (2022a, 14 August). Why are the tea workers on strike? The Daily Star. https://www.thedailystar.net/opinion/views/news/why-are-the-tea-workers-strike-3094846
• Gain, P. (2022b, 3 September). Tea workers' strike ends. What's next? The Daily Star. https://www.thedailystar.net/opinion/views/news/tea-workers-strike-ends-whats-next-3110056
• "Tea workers go on indefinite work abstention demanding pay hike." The Daily Star, 13 August 2022. https://www.thedailystar.net/news/bangladesh/news/tea-workers-go-indefinite-work-abstention-demanding-pay-hike-3093956
• "Tea garden workers get Tk 400 in daily wages, benefits." The Daily Star, 17 August 2022. https://www.thedailystar.net/business/global-economy/news/tea-garden-workers-get-tk-400-daily-wages-benefits-3097606
• "PM directs to raise tea workers' wage to Tk 145; strike called off." The Daily Star, 20 August 2022. https://www.thedailystar.net/news/bangladesh/rights/news/pm-directs-raise-tea-workers-wage-tk-145-strike-called-3099201
• "Tea workers' daily wage increased to Tk 170." Prothom Alo (English), 27 August 2022. https://en.prothomalo.com/bangladesh/c511ril245
• "Tea workers return to work after 15 days." The Daily Star, 28 August 2022. https://www.thedailystar.net/news/bangladesh/news/tea-workers-return-work-after-15-days-3105546
• "Tea workers' daily minimum wage Tk 170." The Financial Express, 13 August 2023. https://thefinancialexpress.com.bd/national/tea-workers-daily-minimum-wage-tk-170
• "Out of the tea gardens." The Daily Star, 4 April 2026. https://www.thedailystar.net/business/news/out-the-tea-gardens-4142726
• "Tea workers intensify protests for Tk 500 daily wage." The Daily Star, 17 August 2026. https://www.thedailystar.net/news/bangladesh/news/tea-workers-intensify-protests-tk-500-daily-wage-4249716

Verification notes: quotations from Beer et al. (1984), Devanna, Fombrun and Tichy (1984), Legge (1995), Bratton and Gold (2017) and Armstrong and Long (1994) were checked against page-level full-text search of scanned copies held by the Internet Archive, not by reading the whole books; the page number for "The critical managerial task" is taken from Armstrong and Brown (2019). Beer and Spector (1985) and Walton (1985) are quoted as they appear in Legge (1995). Boxall (1992) is cited as summarized in Armstrong and Long (1994). Kaufman (2015), Beer (2015), Beer, Boselie and Brewster (2015), Wright, Guest and Paauwe (2015) and Cascio (2015) are quoted from their published abstracts. Hendry and Pettigrew (1990) and Storey's 1987 paper are cited as the sources of Legge's labels, as she states, and were not read; no full reference is given for Storey (1987) because its details could not be checked. The comparison table, the classification exercise, the Kopotakkho case and the reading of the tea dispute through the two models are original teaching material for this series.

Next in The Human Capital MBA series: Article 18 examines the AMO framework of ability, motivation and opportunity.