In recent years, leadership instability has become more visible across many countries. In the United States, for example, government control has shifted between parties, regulatory priorities have reversed, and university and agency leadership has changed. In the UK and elsewhere, political turnover and organisational restructuring have become common. Chief executives come and go. Ministers are replaced. New managers rewrite rules.
When leadership changes, long-term promises can change too. Employment contracts agreed under one regime may look different under the next. In research that I co-authored with Liang Dai and Yenan Wang, we ask a simple but important question: how should contracts be designed when employers cannot fully commit to keeping long-term promises? (Dai et al, 2026)
Broken promises: job type matters
Our study builds a framework to understand how limited commitment shapes employment contracts. We focus on a workplace where worker can either work diligently or shirk. Good effort benefits the employer, but it may not always be observed. Slacking may bring a small private benefit to the worker and, again, it may or may not be detected. In our framework, the employer can reward good performance, punish misconduct or dismiss the worker.
The key feature is that at random times the employer can revise the contract. These ‘revision opportunities’ capture events such as leadership changes or political turnover. The more frequently these opportunities occur, the weaker the employer’s commitment.
Our main result is striking. There are two fundamentally different kinds of jobs, and they respond very differently to weaker commitment by employers.
Guardian jobs
The first group includes what we call guardian jobs. In these roles, misconduct is easier to observe than steady good work. Think of airline pilots, compliance officers or safety inspectors. A serious mistake is likely to be noticed. Reliable day-to-day effort may not be.
In these jobs, the best contract is permanent employment with modest performance rewards and immediate dismissal if misconduct is detected. The threat of being fired provides discipline.
Weaker commitment has little effect on the optimal contract in these roles. Even if leadership changes allow contracts to be revised, the basic structure – permanent employment combined with clear dismissal rules – should remain stable because it is closely aligned with how guardian jobs function.
Recent high-profile layoffs or contract revisions in roles like air traffic control in the United States highlight the practical relevance of this insight. Our theory serves as a warning: altering well-designed guardian contracts for political or short-term budgetary reasons risks weakening performance in precisely those roles where reliability and public safety matter most.
Entrepreneurial jobs
The second group of roles that we define is entrepreneurial jobs. In these roles, exceptional performance is more visible, but poor effort may go undetected. Think of salespeople landing big deals, researchers publishing breakthroughs or founders building new ventures.
In these jobs, motivating effort requires strong rewards. Permanent employment quickly becomes expensive. To keep incentives strong, the employer would need to pay large bonuses repeatedly.
Instead, the optimal design begins with a probation period. During probation, the worker receives little or no pay. If strong performance is observed, the worker is promoted and receives a bonus. If not, employment ends. After promotion, employment becomes effectively permanent – unless a revision opportunity appears.
When commitment becomes weaker, probation becomes even more central. Each time that a revision opportunity arrives, the employer may restart probation. Our research suggests that weaker commitment leads to larger promotion bonuses, shorter probation periods and faster declines in bonuses over time.
One surprising result stands out. It is often assumed that more commitment is always better. But our study shows that partial commitment can create tension between the employer’s present and future selves. At a later revision point, the employer may regret earlier generosity. This internal conflict can make some long-term contracts unsustainable. In certain cases, reducing commitment further – even to zero – can improve outcomes for both employer and worker.
Optimal employment contracts
In guardian-type roles, dismissal threats are credible because misconduct is easy to observe. The optimal contract remains stable even in politically unstable or rapidly changing organisations. But stability depends on adherence to that design. When contracts are revised for unrelated reasons – such as ideological shifts – performance may suffer.
In entrepreneurial roles, job security is naturally more fragile. Leadership changes increase uncertainty, and they may lead to renewed probation and stronger performance-based pay. This aligns with what we see in sectors such as start-ups, creative industries and academia, where visible success drives careers and tenure systems often include lengthy evaluation periods.
Our research also contributes more broadly to our understanding of commitment. Much previous work assumes either perfect long-term commitment or constant renegotiation. Our study examines the middle ground by introducing a continuous way to measure how much commitment exists.
Our analysis shows that weakening commitment does not have uniform effects across jobs, nor are the effects always monotonic: in some cases, reducing commitment further may improve the outcomes. This non-monotonic pattern would be invisible in models that compare only the two extremes of full commitment and no commitment, highlighting the value of analysing commitment as a matter of degree rather than as an all-or-nothing assumption.
Most importantly, our results challenge a simple narrative. Political instability or leadership turnover does not automatically destroy job security. Its effects depend on whether strong performance or misconduct is relatively easier to observe in a given job type, and therefore on how incentives must be structured to sustain effort.
Implications for employers and policy-makers
Policy-makers and organisational leaders should recognise that employment protections and contract design cannot be uniform across all roles.
For guardian-type jobs, maintaining clear dismissal standards and stable long-term contracts supports performance – even in uncertain environments. For entrepreneurial roles, transparent probation systems and clear promotion criteria are essential. Since weaker commitment increases reliance on probation and large bonuses, organisations should consider how frequent leadership turnover shapes career incentives.
Future research could explore how these dynamics operate in public sector agencies, universities and/or rapidly changing political systems. As leadership instability becomes more common, understanding which promises can remain stable – and which are inherently fragile – becomes more important.
Job security is not just about goodwill. It depends on whether strong performance or misconduct is relatively easier to observe, and on how incentives are structured to sustain effort. When leadership changes, contracts may change. The real question is how they should change, and in which jobs altering the established structure risks making performance and welfare worse.
