Comment / Market forces: search for the X factor

The MFF, effectively a multiplier for tariff income, has long produced discussion about its accuracy in reflecting unavoidable cost differences. David Parkin explains the latest revisions

The new weighted capitation formula for resource allocation made changes to one of its most controversial parts – the market forces factor (MFF). Like the formula in general, everyone has always found its aims desirable, but it is despised by those who think it disadvantages them. Since it became part of providers’ payment by results tariffs, it has become more visible to more people, leading to more pressure on the Department of Health for it to be reformed.

Few would argue with its rationale or aims. The idea is that provider costs vary over different geographical areas, so that in high cost areas each NHS pound buys less healthcare for a given level of need than elsewhere. The MFF exists to deal with these ‘unavoidable’ costs, so that every PCT can deal with local levels of need to the same extent. The question is: do actual MFFs do that?

MFFs do impact on PCT allocations for both hospital and primary care services and cover staff, buildings and land. However, it is the impact of the staff MFF for hospital services that is seen as most contentious. In particular it is the impact this has on the money paid by PCTs and received by NHS providers under the national tariff. Its two most contentious features are the theory underlying the way that it is calculated and the cliff-edge effect that it produces, whereby neighbouring PCTs and providers have radically different MFFs.

The theory is plausible. The NHS competes with other industries to employ staff. Local labour markets have a ‘going rate’ for different kinds of workers. If NHS wages are below this, providers have problems in recruitment and retention as well as productivity.

To deal with these problems, good managers find ways around the rigidity of nationally-determined NHS pay scales, so that they can do their job of delivering services. All of this raises costs. The MFF adjustment in effect aims to allow providers to pay the locally-determined ‘going rate’.

Critics of the MFF raise some arguments against this. First, is the theory true in practice? Second, if it is true do the differences in local wage rates translate into the same differences in actual costs? And third, does it apply to all types of staff?

The latest research commissioned by the Department, from Crystal Blue Consulting in London and the Health Economics Research Unit

in Aberdeen, shed some light on these issues and has led to changes. An important finding is that local NHS costs do vary in line with variations in local rates of earnings. This gives some confidence in the actual method used to calculate MFFs, the so-called general labour market (GLM) approach. This estimates the ‘going rate’ based on private sector wages, which are available for each PCT in the form of standardised spatial wage differentials (SSWDs).

A second finding was that it is not possible using existing data and methods to estimate directly the precise impact on costs that market forces have – the so-called specific cost approach. This means that we do not know how accurately the MFFs based on the GLM approach reflect changes in costs.

However, that is of course a feature of all aspects of the resource allocation formula – for example, we do not know how accurately the various needs factors in the formula reflect changes in real needs.

A third finding was that the GLM approach is, as has always appeared obvious, not really applicable to medical staff. As a result, they have now been removed from the MFF, except for London weighting. But it does work for nursing staff, and as these are the largest staff group, the staff MFF remains an important factor in allocations.

What of the cliff-edge problem? Rather like premium vodka, the raw ‘going rate’ data are now triple smoothed to produce more palatable MFFs. First, a PCT’s SSWD is smoothed by taking account of all other PCTs’ SSWDs, with the weight given to others reducing the further away they are. Next, a provider’s MFF is calculated from the smoothed SSWDs from both the PCT in which it is located and neighbouring PCTs, again weighted by distance. Finally, a PCT’s MFF is calculated from the MFFs of its providers, weighted by inpatient activity but taking account of other services provided in the PCT’s area.

The Department will hope that the evidence it has provided and the changes that it has made will lower concerns about the MFF. Questions remain about whether it accurately measures what it is intended to measure, but in that the MFF is no different to any other part of the weighted capitation formula.