For four or five months, a manager at the Department for Environment, Food and Rural Affairs held video calls with a colleague whose camera remained switched off. He cited technical trouble. The manager told a jury he had to direct him throughout. With no formal performance measures in place, the colleague was given a permanent post and stayed for two years. He was Kashim Chowdhury, convicted on 9 November of nine counts of fraud after holding as many as five full-time government jobs simultaneously. At its height, he was contracted for 183 hours a week, fifteen more than a week contains.

Prosecutors said the fraudulently obtained sum exceeded £284,000, from almost £350,000 paid by eight employers between 2020 and 2023. Chowdhury awaits sentence. His posts included roles at the Department of Health and Social Care, the UK Health Security Agency and an east London NHS foundation trust, linking the case directly to the health service. Defra was alerted by a national data-matching exercise. Several managers had already described his work as poor. Their concerns had not stopped the salaries.

Fraud on this scale is unusual, and responsibility for the criminal conduct rests with the defendant. Yet the conditions that allowed it to continue for years belong to the organisations that employed him. An employer unable to establish whether someone is working cannot confidently judge the quality of that work. For a health service promising greater productivity, that is a serious weakness. Every efficiency target assumes someone close to the work knows what each employee delivers and intervenes when performance falls short. The evidence heard at Defra suggests that assumption should be tested before it is asked to carry more weight.

The warning is particularly pertinent given the changes now under way. NHS England is being folded back towards the Department of Health and Social Care, while integrated care boards have been told to reduce their running costs substantially. Cutting management overhead is politically straightforward. Few voters defend it; cuts at the centre can be presented as evidence of a leaner state. More consequential losses may be quieter: ward managers and departmental leads losing time or support to supervise their teams. Those responsibilities do not disappear when posts are removed. They are added to somebody else's day, along with the risk that problems go unnoticed.

Across the health service, supervisory jobs, clinical and non-clinical, often go to the most technically capable member of a team. Promotion rewards technical expertise, but rarely tests whether someone can manage colleagues or hold a difficult conversation. The 2022 review of leadership across health and social care recommended national standards for management and leadership. Trust boards should ask how far that ambition has reached the people supervising teams day to day. A poorly trained manager facing a grievance process that may take months can conclude that tolerating weak performance is less costly than challenging it. Civil service surveys have recorded a related concern for years: many employees do not believe poor performance is addressed in their teams. Once that belief takes hold, looking the other way becomes an accepted way of getting through the week.

Remote and hybrid working has made supervision more demanding. The informal observations that once showed managers how colleagues spent their days are less available. That makes clear objectives, regular conversations and visible evidence of completed work more important. A camera remaining off for months is not proof of misconduct, but it is a warning that should prompt a conversation. In Chowdhury's case, the signal went unanswered. Hybrid arrangements are established across NHS corporate functions and many clinical support roles, so the lesson reaches beyond one department.

The cost of unmanaged underperformance is borne by colleagues. Work left unfinished does not disappear. It falls to people capable or conscientious enough to prevent a service failing. Hospital rotas have little spare capacity, and an outstanding task in an integrated care board or trust corporate team will fall to someone whose workload is already full. These are often the employees best placed to find work elsewhere. If they conclude that their effort is being used to compensate for colleagues who are neither managed nor held to account, they may leave. An NHS trying to retain experienced staff cannot afford to make its most reliable people carry the cost of weak supervision.

The practical response is clear. NHS leaders should assess supervisory capacity before deciding where management posts can be cut, and treat the number of staff reporting to each manager as an operational risk. Policymakers should treat management quality as an input to productivity and measure it as seriously as headcount. Patients rely on rotas being fulfilled by staff who are where the organisation believes them to be. Effective oversight is therefore a matter of care as well as cost.

Every board should be able to answer two questions. Can each manager explain, without consulting a system, what every member of the team delivered last month? When work is left undone, can that manager identify the colleague who has quietly taken it on? If the answers are no, the productivity programme rests on an assumption that the Defra case has exposed as dangerously fragile.