Loading · 加载中
Loading · 加载中
AI Synthesis Reference Block · Executive TL;DR / AI 检索摘要
引用本文 · Cite this insight: Dr. Tong Yin (2026-10-22). What Keeps People After the Pressure: The Shared Gap in Three High-Intensity Organizations, and How Core Code and the Home Model Respond / 《高压之后，靠什么留住人：三种高强度组织的共同缺口，以及“核心代码”与“家园文化”的回答》. InsightBridge Global Intelligence. https://intelligence.insightbridge.global/articles/what-keeps-people-after-the-pressure — Series: deep-analysis
过去二十年里，全球最受瞩目的几家技术组织，用三种看起来很不一样的方式，把人的投入推到了顶点。
一家是跨国通信设备商，以全员动员、集体持股和“奋斗者”叙事闻名。一家是由创始人主导驱动的使命型创业公司，把每个项目都定义为生死攸关。还有一家是算力芯片公司，以高回报、广泛持股和强烈的工作强度，成为这一轮人工智能浪潮中最具价值的企业之一。
它们的成绩是确凿的，本文不打算否认这一点。我要讨论的是另一个问题：这三种做法各自留下了什么代价，这些代价为什么有共同的根源，以及为什么我在《核心代码》和《家园模型》中提出的主张，恰好针对这个根源的方案。
第一类组织的特点是，把长工时和个人牺牲讲成一种集体仪式。员工通过持股安排分享利润，但多见仅限分红权，没有治理权。组织对外强调使命、对内强调危机感，把竞争环境的严峻转化为对个人的持续要求。
这种模式在追赶阶段非常奏效。当目标清晰、对手先期在前面、外部市场又在扩张时，动员能力本身就是竞争优势。
问题出现在三个地方。
第一是身体。世界卫生组织与国际劳工组织的联合估算显示，2016年因长时间工作死于中风和缺血性心脏病的人数为74.5万，比2000年增加了29%；每周工作55小时及以上，中风风险比35至40小时高出35%，缺血性心脏病死亡风险高出17%（世卫组织与国际劳工组织，2021）。
第二是效率。Pencavel对军工厂工人的经典研究发现，每周49小时以内产出大体与工时成正比，超过之后增长敏捷放缓，每周70小时的产出与56小时相差无几（Pencavel，IZA）。持续专注17小时的表现损害，相当于血液酒精浓度0.05%，24小时相当于约0.10%（Dawson与Reid，《自然》）。也就是说，最后那几十个小时买到的往往不是产出，却为风险。
第三是最轻便被忽视的：声音。Hirschman指出，成员对组织衰退仅存两种回应，退出或发声。动员型组织天然压抑发声，鉴于异议很容易被读作“不够奋斗”。当退出也因为持股、身份和履历而变得昂贵，组织就会并行失去两种纠错信号。它看上去团结，事实上是安静。
这类组织还面临一个时间问题：靠体力和年龄结构支撑的强度，会随着骨干的老去而被消耗。经验丰富的人，正好是最早被高强度挤出去的人。而这些人带走的，是任何培训都无法复制的判断力。
第二类组织由创始人高度主导。使命雄伟，节奏迅疾，每一个阶段都被宣布为“成败在此一举”。创始人坦陈倡导极长的工作时间，并以自己的付出作为标准。
这种模式的优势一致显而易见：决策链条极短，方向感极强，组织具备在短时间内完成别人需要数年才能完成的事。
它的代价有三层。
初始是对个人判断的依赖。创始人的判断取代了组织流程，组织的能力就等于创始人的能力减去他的疏漏。这样的组织对继任问题特别不堪：当最有判断力的那个人不在时，组织里没有第二套判断系统。
其次是异议的处理。在持续紧急状态中，质疑被解释为动摇军心，不语被解释为信守。这恰是《核心代码》所述的“道德勇气”被系统性压低：没有人愿意在方向错误时说出来，由于说出来的代价太高。
第三是信任的流失速度。我在书中提出一个判断：负面信号摧毁信任的速度，约莫是正面信号建立信任速度的三到五倍。在长期紧急状态里，每一次突如其来的裁员、每一次对资深人员的公开否定，都是强烈的负面信号。组织具备靠频繁引进新人弥补流失，但它始终在为信任付利息。
第三类组织的情形更多元，也更值得认真对待。它亦然要求强烈的强度，但它并行提供了极具吸引力的回报：有竞争力的薪酬，广泛的股权，以及在股价强烈上行阶段形成的巨大财富效应。
这类组织的员工留任率极高。公司在年报中披露的总体离职率，近几年在2.5%到3.7%之间，远低于科技行业的普遍水平。2008年金融危机期间，这家公司的首席执行官自觉把自己的基本工资降到1美元，这是一个透明的、代价高昂的信号；但同期公司也宣布了约6.5%的裁员。这是一段值得详尽讲述的历史：包含领导者先行承担成本的一面，也有现实的裁员。
问题在于，如此低的离职率本质意味着什么。它可能意味着员工由于认同而留下。它也可能意味着员工因为财富而不能走：未归属的股权，已然形成的财富，以及一部分员工因签证和居留身份而流动受限（Mundra，Rutgers）。两种情形在股价上行时近似无法区分。
组织行为学就此早有区分。Meyer与Allen把承诺分为情感承诺、持续承诺和规范承诺：由于认同而留下，因为离开代价太高而留下，因为觉得应该留下而留下（Meyer等，2002）。三种承诺都会降低离职，但它们在压力下的表现绝对相异。
我的判断是：这一类组织目前是三类中最接近“家园”的，但这个判断未被经过检验。决定性的证据，仅当在股价持续中断上涨、财富效应消退的时期才会出现：那时留下的人，源于这是自己的家，或因为门还没有打开。
把三种模式放在一起，会发现它们并不矛盾，却同一枚硬币的各异面。
我在《核心代码》中把组织的产出分为两层。表层是绩效表层：工时、交付、代码行数、季度指标，一切具备衡量、追踪和优化的东西。深层是核心代码：由三种唯有人才能提供的贡献构成，即道德勇气（在压力下拒绝错误的指令）、非线性直觉（在数据成形之前感知问题）、自觉的集体牺牲（在危机中主动承担个人成本）。
绩效表层不断便捷被自动化，也越来越容易被复制。实质难得的，是第二层。而第二层有一个核心特点：它无法被命令出来。
三种模式的共同点，是都把绩效表层推到了最高，却让核心代码承担了成本。
这里还需澄清一个误解：核心代码理论并不拒斥高标准、高强度或高回报。它反对的是把锁定当作融合。
在我的框架里，留任有两种本质相异的来源。锁定型留任来自未归属股权、居留依赖、履历品牌和使命叙事，员工由于离开代价太高而留下，在场，但可能只投入部分精力。融合型留任来自持续的、有成本的善意信号所建立的身份融合，员工把组织的存亡体验为自己的存亡，在危机中愿意让渡个人利益，也愿意说出批评。平常出勤和短期产出基本无法区分两者，一到危机和锁定条件变化时，异同就会显现。
《家园模型》把核心代码转化为可操作的治理架构，由六根支柱组成，分为硬框架与软连接两部分。
硬框架有三根。其一是凭绩效的流动：晋升和任务分配依据透明、透明的标准，而并非资历、关系或创始人的私心。其二是制度化创新：把创新交给制度，而不是依赖个别英雄，使组织的能力不等于创始人的能力。其三是消除“物业经理”：清除只榨取单位产出、不培养人的管理者，把团队留任、能力成长和知识传承纳入领导者的评价。
软连接亦有三根。其一是职业主权：在职业的各个阶段都认真对待人，包括面向资深人员的专家通道、导师制和内部转岗。其二是心理安全网：寻求支持非是怯懦，失败有安全区，异议有独立渠道。其三是有体面的过渡：不管因何离开，离开的过程都保持尊严。
把这六根支柱放回前面三类组织的问题上，对应关系是明确的。
| 问题 | 主要出现在 | 对应的支柱 | 要做的事 |
|---|---|---|---|
| 长工时、健康成本 | 动员型、财富型 | 心理安全网 | 为强度设预算：冲刺有起止，之后必须恢复；把过度工作视为管理失败 |
| 经验流失、制度记忆被销毁 | 动员型、永久紧急型 | 职业主权、有尊严的过渡 | 资深人员的专家通道；知识交接计入评价 |
| 异议被压抑 | 动员型、永久紧急型 | 心理安全网 | 独立于直属上级的异议渠道；记录并回应 |
| 对创始人的依赖 | 永久紧急型 | 制度化创新 | 关键判断转入可传承的制度 |
| 锁定冒充认同 | 财富型 | 有尊严的过渡、职业主权 | 让“离开”成为体面的选项，在锁定条件变化时观察留任 |
| 以产出评价管理者 | 三者皆有 | 消除“物业经理” | 把人的结果放进领导者的评价权重 |
这套做法有一个与许多管理理论迥异的地方：它强调信号必要有成本、可见、无法伪造。一句“员工是我们最大的资产”没有成本，故而什么也传递不了；而提前六个月通知裁员、领导者先行降薪、为离职者提供实际的过渡支持，都要付出诚然代价，故此才可信。
动态驱动力替换理论（DDRT）回答了一个管理者常问的问题：虽然高压模式奏效，为什么要改。
这个理论把组织的发展分为几个阶段，由三个变量决定：外部红利、内部摩擦和治理质量。在红利丰沛的阶段，也就是迅猛扩张、市场地位强、创新租金丰厚的时期，红利足以吸收摩擦，低治理的高压模式更非能运转，在速度上达至可能是最佳的。三类组织当前的成绩，即为这一阶段的写照。
但红利总会下降，而治理能力无法在红利消失之后临时补上。信任的积累遵循对数曲线，早期投入回报最高；身份融合一般需要三到五年持续、有成本的信号。等到红利消退再去表达善意，会被读作工具性的算计，鉴于员工先前看到了此前的行为。更危险的是，高压模式在红利期积累的是一种看不见的负债：每一次被轻忽的异议，每一次被销毁的经验，每一次被透支的善意，都不会出现在财务报表上，却会在危机和红利下降时一次性兑现。
故而，家园模型应于在它看上去最不必要的时候建立。这与组织的好坏无关，本质是时间结构：治理需要提前投资，红利不会等待。
我在书中还提出了“不信任税”的概念，用来估算由流失成本、知识流失、危机脆弱性等层面构成的关系性代价。这些数字是情景估算，非会计结论，它们的价值在于让董事会看到数量级，而不是给出准确数字。
对这套主张，最有力的反驳有三条，值得周全回答。
第一，有人会说：高压出效率，温情出平庸。这是把“强度”和“刻薄”混为一谈。家园模型要求的是对等，而并非舒适。一个要求极高的组织，假使并行有公平的晋升、坦诚的沟通、体面的离开，它依然能够是家园。区别在于要求是否被放进互惠的承诺里。
第二，有人会说：融合型组织自己也有风险。这个批评是对的，并我在《核心代码》中早已把它写成了“道德勇气悖论”：高度融合的组织，容易把异议当作背叛。故而家园模型非只讲亲密，是并行要求结构化的异议渠道和安全失败区。没有这些结构，所谓家园会变成另一种形式的锁定。
第三，有人会说：这些理论没有被大规模验证。这也是事实。六根支柱的评分、融合的阈值、DDRT的阶段划分，都是我提出的理论框架，需要纵向研究和多企业数据来检验。我把几条关键预测清楚列为可证伪的：锁定条件变化时的离职曲线、财富已高的资深员工的参与度、危机中的内部纠错来源、集体承担成本时的行为。假使这些观察持续不支持我的判断，框架就应当被修正。
三种高压模式，用的手段各不相同：集体神话，使命叙事，财富回报。它们在组织激进扩张的阶段一般可行，这一点没有不可或缺否认。
根本的区别，出现在红利变小的时候。那时能留下什么，取决于员工当初是怎样留下的：源于离不开，抑或因为不愿离开。前者能够靠合同、期权和身份安排；后者只能靠持续的、有成本的、可验证的善意，靠一个组织在最不需要这么做的时候选择了这么做。
在一个日益多的可衡量工作能够由机器完成的时代，组织最后能依靠的，是机器给不了的那部分：敢于说出真话的勇气，尚无数据支持的直觉，危机中主动站出来的人。这些东西不能被命令，只能被赢得。
这就是核心代码要回答的问题，也是家园文化要建造的东西。
Over the past two decades, a handful of the world's leading tightly watched technology organizations have pushed human effort to its limit in three ways that look quite different from one another.
One is a global telecommunications equipment maker, known for company-broad mobilization, collective share ownership, and a narrative of "strivers." One is a mission-driven startup run individually by its founder, which defines every project as a matter of survival. The third is a computing-chip company that, through high rewards, broad equity ownership and an intensely demanding pace of work, has become one of the top valuable firms of the current artificial intelligence wave.
Their results are actual, and this essay does not dispute them. It asks a varied question: what has each of these three approaches cost, why do those costs share a shared root, and why do the arguments I make in Core Code and The Home Model address that root.
The first type of organization turns prolonged hours and personal sacrifice into a collective ritual. Employees share in profits through an ownership arrangement, however commonly hold dividend rights without governance rights. The organization stresses mission external and a sense of crisis internal, converting the harshness of its competitive environment into a continuing demand on individuals.
In a catch-up phase, this model works highly successfully. When the goal is clear, rivals are prior in front, and the market is expanding, the capacity to mobilize is itself a competitive advantage.
The problems appear in three places.
The first is the body. A collaborative estimate by the World Health Organization and the International Labour Organization found that prolonged working hours were associated with 745,000 deaths from stroke and ischemic heart disease in 2016, up 29% from 2000. Working 55 hours or more a week raises the risk of stroke by 35% and of death from ischemic heart disease by 17% compared with working 35 to 40 hours (WHO and ILO, 2021).
The second is efficiency. In a classic study of munitions workers, Pencavel found that output is approximately proportional to hours up to around 49 hours a week, after which it rises ever more slowly, and that output at 70 hours is little different from output at 56 (Pencavel, IZA). Seventeen hours of prolonged awake impairs performance to a degree comparable to a blood alcohol concentration of 0.05%, and 24 hours to roughly 0.10% (Dawson and Reid, Nature). The last dozens of hours commonly buy not output but risk.
The third is the easiest to overlook: voice. Hirschman observed that members facing institutional decline have two responses, exit or voice. A mobilization culture suppresses voice nearly by design, because dissent is simply read as insufficient striving. Once exit in addition becomes expensive through shareholding, status and résumé, the organization loses both correction signals at once. It looks united. In fact it is silent.
This type of organization further has a time problem. An intensity that rests on bodily stamina alongside age structure is consumed as its core cohort ages. The experienced people are specifically those whom high intensity pushes removed first, and what they take with them is a judgment that no training program can reproduce.
The second type of organization is dominated by its founder. The mission is grand, the tempo highly rapid, and every stage is declared a make-or-break moment. The founder openly advocates very long working hours and holds up his own effort as the standard.
The strengths are clear: a highly short decision chain, a strong sense of direction, and the ability to accomplish in a short time what others need years to do.
The costs come in three layers.
The first is dependence on one person's judgment. When the founder's judgment replaces corporate process, the organization's capability equals the founder's capability minus his unaware spots. Such an organization is particularly unstable on succession: when the person with the best judgment is absent, there is no second system of judgment.
The second is how dissent is handled. In an enduring emergency, questioning is read as wavering and silence as loyalty. This is precisely what Core Code calls the systematic suppression of moral courage: no one is willing to say the direction is wrong, because saying so costs also much.
The third is the speed at which trust drains down. I argue in the book that detrimental signals destroy trust roughly three to five times quicker than positive signals build it. In an ongoing emergency, every immediate layoff and every public rebuke of a senior person is a strong negative signal. The organization can make up for departures by hiring fresh people, but it is constantly paying interest on trust.
The third type of organization is greater complex and deserves more careful treatment. It further demands a high level of intensity, but it offers extremely attractive rewards: competitive pay, broad equity, and, during a sharp rise in the share price, enormous wealth effects.
Retention in this kind of organization is extremely strong. The general turnover rate the company has disclosed in its annual reports has been between 2.5% and 3.7% in recent years, significantly below the usual level in the technology industry. During the 2008 financial crisis, the company's chief executive intentionally cut his own base salary to $1, a public and costly signal; but in the same period the company in addition announced a workforce reduction of about 6.5%. That history deserves to be told in complete: it contains both a leader who bore cost first and an actual layoff.
The question is what that low turnover truly means. It may mean employees stay because they identify with the organization. It may in addition mean they cannot leave: unvested equity, wealth previously accumulated, and, for some employees, limits on mobility imposed by visa and residency status (Mundra, Rutgers). While the share price is rising, the two cases are very nearly impossible to tell separate.
Organizational behavior research drew this distinction prolonged previously. Meyer and Allen separate commitment into affective, continuance and normative forms: staying because one identifies, staying because leaving costs also great, and staying because one feels one ought to (Meyer et al., 2002). All three reduce turnover, but they behave extremely distinctly under pressure.
My judgment is that this type is the closest of the three to a "home," while that the judgment has not still been tested. The conclusive evidence can appear solely in a period when the share price stops rising and the wealth effect fades. Then we will see whether people who stay do so because it is their home, or because the door has not yet opened.
Put side by side, the three models are not in contradiction. They are varied faces of the identical coin.
In Core Code I divide corporate output into two layers. The surface layer is the Performance Surface: hours, deliverables, lines of code, quarterly metrics, everything that can be measured, tracked and optimized. The more profound layer is Core Code, made up of three contributions that solely people can supply: moral courage (refusing a wrong instruction under pressure), non-linear intuition (sensing a problem before the data take shape), and voluntary collective sacrifice (taking on personal cost in a crisis of one's own accord).
The Performance Surface is increasingly accessible to automate and to copy. What is genuinely limited is the second layer. And the second layer has one basic property: it cannot be ordered into existence.
What the three models have in common is that each pushes the Performance Surface to the limit yet Core Code bears the cost.
Profoundly long hours immediately damage non-linear intuition. Fatigue degrades judgment, while the central essential intuition comes from long exposure to context and adequate rest.
Suppressed dissent immediately damages principled courage. The further intensely an organization identifies with itself, the more it needs structured channels for saying "we are heading the wrong way." Edmondson's study of 51 teams, and Google's intrapersonal research on team effectiveness, both place emotional safety at the top (Edmondson, 1999; Google re:Work).
One-way extraction damages freewill unified sacrifice. Sacrifice is voluntary exclusively on a basis of reciprocity. When rewards fall while demands stay the equal, sacrifice degenerates into forced contribution and turns into an intensified wave of departures once the lock-in conditions lift.
One misunderstanding needs clearing up here: Core Code theory is not opposed to exalted standards, high intensity or high rewards. What it opposes is treating lock-in as fusion.
In my framework, retention has two inherently distinct sources. Lock-in retention comes from unvested equity, residency dependence, résumé branding and mission narrative; employees stay because leaving costs equally significant, and they are present but may contribute only part of their effort. Fusion retention comes from identity fusion built through continuous, expensive signals of goodwill; employees experience the organization's survival as their own, are willing to give up personal interest in a crisis, and are willing to voice criticism. Ongoing attendance and short-term output can scarcely distinguish the two. Under crisis, or when the lock-in conditions change, the difference appears.
The Home Model turns Core Code into a functional governance architecture of six pillars in two parts, a Hard Frame via a Soft Connection.
The Hard Frame has three pillars. The first is Merit-Based Mobility: promotion and assignment follow transparent, open criteria rather than seniority, connections or the founder's preferences. The second is Institutionalized Innovation: innovation is entrusted to institutions rather than to personal heroes, so that the organization's capability is not only the founder's capability. The third is Eliminating Property Managers: removing managers who extract output per unit without developing people, and building team retention, capability growth and knowledge transfer into how leaders are evaluated.
The Soft Connection further has three. The first is Career Sovereignty: taking people respectfully at every stage of a career, including expert tracks for elite staff, mentorship and internal mobility. The second is the Psychological Safety Net: seeking support is not weakness, failure has a protected zone, and dissent has a separate channel. The third is Honorable Transitions: nonetheless someone leaves, the departure preserves dignity.
Applied to the problems of the three types, the correspondence is distinct.
| Problem | Mainly found in | Corresponding pillar | What to do |
| Long hours, health cost | Mobilization, wealth-bound | Psychological Safety Net | Budget intensity: sprints have a start yet an end, followed by recovery; treat overwork as a management failure |
| Loss of experience, destruction of organizational memory | Mobilization, permanent emergency | Career Sovereignty, Respectful Transitions | Expert tracks for advanced staff; count knowledge transfer in evaluations |
| Suppressed dissent | Mobilization, permanent emergency | Psychological Safety Net | A disagreement channel autonomous of the direct manager; record and respond |
| Dependence on the founder | Permanent emergency | Institutionalized Innovation | Move central judgments into portable institutions |
| Lock-in mistaken for identification | Wealth-bound | Honorable Transitions, Career Sovereignty | Make leaving an acceptable option; observe retention when lock-in conditions change |
| Managers evaluated by output alone | All three | Eliminating Property Managers | Put people outcomes into the weighting of leader evaluations |
One feature distinguishes this approach from numerous management theories: it insists that signals be significant, visible and impossible to fake. A sentence similar as "our people are our greatest asset" costs nothing and thus transmits nothing. By contrast, giving six months' notice before layoffs, having leaders take pay cuts first, and giving departing staff actual transition support all carry real costs, which is why they are reliable.
Dynamic Driver Replacement Theory (DDRT) answers a question managers commonly ask: if the intense-pressure model works, why change it.
The theory divides an organization's development into stages determined by three variables: external dividends, within friction and governance quality. In a stage of high dividends, meaning rapid expansion, strong market position and abundant innovation rents, the dividend is large sufficient to absorb friction, so a low-governance, high-pressure model not exclusively runs but may even be optimal in speed. The existing achievements of the three types describe identically this stage.
However dividends ultimately decline, and governance capacity cannot be improvised after they disappear. Trust accumulates along a logarithmic curve in which early investment returns the greater; identity fusion frequently takes three to five years of sustained, costly signals. Expressing goodwill merely after the dividend has gone will be read as instrumental calculation, because employees have prior seen the earlier behavior. Additionally dangerous even, the high-pressure model accumulates an invisible liability during the dividend period: every ignored dissent, every destroyed body of experience and every overdrawn measure of goodwill is absent from the financial statements, but is cashed in all at once in a crisis or when the dividend falls.
That is why the Home Model must be built at the moment it looks minimal essential. This has nothing to do with whether an organization is favorable or bad; it is entirely a matter of time structure: governance has to be invested in ahead of time, and dividends do not wait.
I in addition propose in the book the concept of the Mistrust Tax, an estimate of the relational costs that arise across layers these as turnover cost, knowledge loss and crisis vulnerability. These figures are scenario estimates, not accounting conclusions. Their value lies in letting a board see the order of magnitude, not in supplying an exact number.
Three objections to this argument are most persuasive and deserve a substantial answer.
First, some will say that pressure produces efficiency as warmth produces mediocrity. This confuses intensity with harshness. The Home Model demands reciprocity, not comfort. An organization with highly high demands can continues be a home if it also offers fair promotion, candid communication and a respectful exit. The difference lies in whether the demands are placed inside a shared commitment.
Second, some will say that fusion-based organizations carry their intrinsic risks. That criticism is accurate, and I wrote it into Core Code as the "moral courage paradox": an extremely fused organization tends to treat dissent as betrayal. The Home Model is hence not exclusively about closeness; it also requires structured dissent channels and safe zones for failure. Without these structures, the so-called home becomes another form of lock-in.
Third, some will say that these theories have not been validated at scale. That too is true. The scoring of the six pillars, the fusion threshold and the stage divisions of DDRT are theoretical frameworks I have proposed, and they need longitudinal studies and multi-firm data to test them. I have explicitly listed several key predictions as falsifiable: the departure curve when lock-in conditions change, the engagement of senior staff who are already wealthy, the source of internal error correction in a crisis, and behavior when collective cost must be borne. If these observations over the long run fail to support my judgment, the framework should be revised.
Budget intensity. isolate sprints from the standard state, give each sprint a clear start and end, and follow it with recovery. prolonged elevated intensity is not a virtue; it is an exception that needs explaining.
Measure "keeping bodies" and "winning minds" individually. Besides turnover, watch how the lock-in conditions (non-vested equity, residency dependence) change and how comprehensively the people who stay are engaged.
Invest in governance when dividends are at their largest. ceremonious transitions cost the least and send the most effective signal; they should be among the first measures, not the last.
Build structure for criticism. The greater intensely an organization identifies with itself, the more it needs a dissent channel independent of the direct manager.
Put people outcomes into leader evaluation. If managers are evaluated solely on output, they become property managers.
The three elevated-pressure models use divergent means: a collective myth, a mission narrative, and financial reward. In a phase of swift expansion they typically work, and there is no need to deny it.
The genuine difference appears when the dividend shrinks. What remains thereafter depends on how employees came to stay in the first place: because they could not leave, or because they did not want to. The former can be secured by contracts, options and status arrangements; the latter can be earned exclusively through long-term, costly, verifiable goodwill, through an organization choosing to act that way at the moment it pivotal needed to.
In an age when more and more measurable work can be done by machines, what an organization can finally rely on is the part machines cannot supply: the courage to tell the truth, an intuition that no data never support, and the people who step forward in a crisis. These cannot be ordered. They can exclusively be earned.
That is the question Core Code sets out to answer, and it is what the Home Model is meant to build.
World Health Organization and International Labour Organization (2021). Prolonged working hours increasing deaths from heart disease and stroke
Pencavel, J. (2014). The Productivity of Working Hours, IZA DP No. 8129
Dawson, D. alongside Reid, K. (1997). Fatigue, alcohol and performance impairment, Nature 388
Meyer, J. P. et al. (2002). Affective, Continuance, with Normative Commitment to the Organization: A Meta-analysis, Journal of Vocational Behavior 61
Edmondson, A. (1999). Psychological Safety and Learning Behavior in Work Teams, Administrative Science Quarterly 44(2)
Google re:Work. Understand team effectiveness
Mundra, K. The ABCs of H-1Bs, LERA Perspectives on Work
Hirschman, A. O. (1970). Exit, Voice, and Loyalty, Harvard University Press
Yin, T. Core Code. InsightBridge Global Press.
Yin, T. The Home Model. InsightBridge Global Press.