Labour Productivity - harnessing the power of the workforce
Key research questions
In most public services, the workforce is the largest cost factor and the greatest determinant of productivity, and like most service industries it is the greatest determinant of quality. The NHS directly employs over 1.7 million people, or a quarter of the 6 million public sector workforce, while local government employs almost half of this. This workforce represents a significant resource in enabling change but also a significant proportion of costs. For instance, salaries in the NHS accounted for 60% of the NHS budget in 2009 and an even higher proportion of variable costs and staff costs are almost 70% of the education budget (DCSF, 2009). According to the Baumol cost disease model, public service worker productivity will lag behind that of the non-service sector due to the labour-intensive nature of the work yet rising labour productivity in other sectors will force average wages to rise. The monopsonist nature of public service employers combined with relatively strong unions exacerbates this problem. One way to overcome the cost disease model is to optimize the skill mix of the workforce.
Skill-mix change and role substitution have been central features of the public services’ workforces recent years. For example, healthcare assistants for nurses,
police community support officers for police officers, or teaching assistants for teachers. For example, the education headcount has increased by a quarter between 1999-2009 but largely with less skilled workers including teaching assistants who have tripled in number and support staff who have doubled in number (DCSF, 2009). Over the past decade as many new police community support officers as police officers were appointed (Home Office, 2010). In recent years, there have been significant alterations to the skill mix throughout the NHS with a broad shift from medical practitioners to registered nurses resulting from changes to junior doctors working practices, for example, as a result of the EU Working Time Directive. The NHS operating framework 2010/2011 identified the need to help local managers to identify optimum skill mix to ensure high quality care and to improve productivity (Department of Health, 2010). While there may be some potential to generate savings through simple reductions in headcount, it is more likely that a reconfiguration of the workforce by altering the skill mix both within roles (e.g. from senior to junior registered nurses) and between roles (e.g. from a doctor to a nurse, or a registered nurse to an unregistered assistant practitioner) will be necessary to increase productivity while maintaining, or even enhancing, quality. A similar story can be seen in other public services. For example, faced with a pressure for more frontline officers while under significant cost pressure police services may employ more police community support officers.
These questions directly affect productivity (e.g. the number of patients seen) and effectiveness (the quality of that healthcare). From a theoretical perspective, the impact of the skill mix on an organisation’s productivity depends upon the degree of complementarity and substitutability between skills i.e. whether the different staff roles are “aiding” or “competing” (Stern, 2004 p. 5) in production. In essence, whether the production of the public service requires workers of similar or diverse skill levels. These issues have received much attention in the private sector, especially manufacturing (e.g. Milgrom and Roberts, 1990). A more recent literature also suggests that skill mix can explain the large differences in efficiency between organisations within sectors (Fox and Smeets, Forthcoming).
However, relatively little is known about the complementarity or substitutability of these different staff groups and their optimal combination, and what is known comes from the US setting. For instance, Thurston and Libby (2002), estimate the staffing relationships for primary care services in the US, and Jensen and Morrisey(1986) investigate medical staff speciality mix in US hospitals. Thurston and Libby (2002), find that in the production of primary care visits, qualified nurses are complements (aiding inputs) for doctors, while technicians and unqualified nursing assistants are substitutes (competing inputs) for qualified nurses. Investigating only medical specialities, Jensen and Morrisey (1986) found that most hospital doctors are substitutes for one another, while primary care doctors are complements for hospital-based specialities. There is some limited evidence, therefore, that staffing inputs are substitutes within their broad roles (e.g. surgeons and obstetricians) but complements between roles (e.g. nurses and doctors). There are also clinical and workforce literatures that investigate skill-mix change and task delegation (see Buchan and Calman, 2004 for a survey) but these studies tend to focus on narrow interventions (e.g. using nurses rather than junior doctors in surgical pre-assessment clinics) and on individual outcomes (e.g. missed care).
Dr. Ioannis Laliotis