Why are public sector efficiency savings so hard? (Part 1 – The processes)

With the government demanding that the public sector do more with less, much has been written about public sector productivity and efficiency recently. I sometimes wonder if there are more people writing about it than actually doing it.

That might be because doing it is damned difficult. Even seasoned corporate trouble-shooters have walked off the job when they have seen the size of the task and the obstacles in the way. Doing more with less is a tall order. Even doing the same with less is likely to prove beyond many public sector organisations

Why should this be? Well that’s such a big question that I have decided to divide this post into two. This post will look at the ‘micro’ reasons – why public sector processes are so difficult to improve and why methods used for making manufacturing firms more efficient don’t always translate to the public sector. In the next post I will look at the ‘macro’ reasons – the wider public sector environment and the political and cultural issues. It is, of course, an artificial distinction because all these things affect each other but I had to find a way of chopping this subject up, otherwise the post would have been very long.

Almost all public sector organisations are service providers and it is far more difficult to make productivity improvements in the service sector than in manufacturing. According to the CBI, manufacturing made average productivity gains of 4 percent per year during the last decade. Over a similar timescale, KPMG calculates that private sector service organisations only increased productivity by twenty-percent.

Most of the theories, methods and case studies about productivity improvement come from manufacturing. Over the past couple of decades, they have proved extremely effective in helping manufacturing companies to drive down their costs. But many doubt that such techniques can be applied to service organisations – at least not without significant adaptation.

Service processes differ from those in manufacturing in that the customer is actually an actor in the process, rather than someone consuming the product from it at the end. People being what they are, their needs and requirements are always slightly different so, while a process might look the same on paper, it is never quite the same for each customer. It will vary each time depending on how the customer interacts with the service provider. Service processes, therefore, are rarely as controllable and predictable as manufacturing ones.

This is exacerbated in the public sector because public-facing organisations have to deal with whoever comes through the door. While price acts as a gatekeeper for private service organisations, public sector organisations must deal with the poorest, least educated and least articulate people, some of whom may not speak English as their first language. Consequently, the customer’s impact on the process is even more significant. There might be a neat box on the process map saying ‘assess customer requirements’ but, in practice, this might take five minutes or five hours.

To standardise and streamline processes, service organisations often try to design out the complexity and unpredictability – because simple equals cheap right? However, this often means designing out the main source of the complexity and unpredictability – the customer. By channelling people through pre-determined options in call centres and forcing them to use standard forms or websites, the providers can make their processes more uniform and therefore cheaper.

Or so they think. According to John Seddon, standardised processes lead to ‘failure demand’ – the demand caused by failing to meet the customer’s need the first time around. Many organisations, he says, fail to account for failure demand when calculating their unit costs.

For example, let’s say that I run a claims operation. I have led an efficiency programme which has reduced my unit costs to £100 per transaction. My boss says, “Bloody impressive, Rick. Well done! Have this huge bonus.” What I haven’t told him (because it hasn’t even occurred to me) is that 80 percent of transactions have to be reworked because they haven’t met the customer’s need and so the customers go back through the process again. For every ten customers I am therefore spending not £1000 but £1800. My unit cost per transaction might be £100 but the cost per customer is actually £180.

An extreme example? Well this head of revenues at a local authority reckons that 80 percent of their contact requires re-work so perhaps my scenario isn’t that far-fetched.

While the privileged few can turn failure demand into additional fees or call-out charges, most service organisations have to take the cost on the chin. For public sector organisations, where services are free at the point of consumption, increases in demand translate directly into increased costs. Failure demand can therefore have a huge impact on public sector finances and, here’s the rub, investing in process improvement programmes might actually make failure demand even worse.

As if that isn’t bad enough, public sector processes are often complicated even further by the involvement of a number of different organisations. NHS trusts and local authorities often have to work closely together and with charities, the police and increasingly autonomous schools. Wherever you hear the term ‘multi-agency’ there will be an extra layer of complexity. Of course, they are all supposed to be working towards the same goals but, inevitably, the players will all have different interpretations of those goals. Under such circumstances, just understanding a process can be difficult enough, without trying to standardise and control it. A process map for manufacturing a PC or a car would be relatively straightforward compared to one for, say, taking a child into care. The latter would look like spaghetti on the page and, even then, would tell you little about how the process actually worked in reality because every time it would be different.

Service sector processes, then, are complicated by the fact that the end users are part of the process. Attempts to standardise these processes can lead to expensive failure demand. Because demand equals cost for public sector bodies, failure demand is especially damaging. Add in the complexity of cross-organisational working and big P and small p politics, and it is amazing that public sector productivity programmes achieve anything at all.

Applying the approaches and methods used in manufacturing firms will only go so far in the public sector. Things are a lot more predictable than people and public sector processes are crammed full of people at every turn, all with different interests and agendas.

Because improving productivity and process efficiency in the public sector is difficult, and difficult to understand, politicians, journalists and even some public sector leaders tend to look for magic bullets. Management costs, bureaucracy and back-office costs are favourites. But though these figures may look impressive, they are often relatively small compared to the overall cost of managing the service. NHS management costs, for example, are only 3 percent of the service’s overall running costs.

The awkward truth is that most of the public sector’s costs are buried deep in its frontline services and that significant efficiency savings can only be made by improving the way these services operate. That requires a lot of hard detailed work.

But that’s not enough on its own. The workings of public sector organisations are further complicated by the environment in which they operate. As well as grappling with process improvement, public sector leaders face a number of cultural and political factors, both national and organisational, which make improving the productivity of their organisations that much more difficult. That will be the focus of the next post.