What does the concept of Next Practice offer to people using social services? - Geoff Mulgan

Composing the Future, Innovation by Design
Published on 25 Mar 2010

Part of the Iriss Masterclass series, Geoff Mulgan.

Geoff Mulgan is director of the Young Foundation. Between 1997 and 2004 he worked in the UK Prime Minister's office and Cabinet Office and before that was the founding director of the thinktank demos. He is a Visiting Professor at LSE, UCL, Melbourne University and the China Executive Leadership Academy. He also works as a part time adviser to Prime Minister Kevin Rudd in Australia. His latest book is The Art of Public Strategy: mobilising power and knowledge for the common good.

Gathering information and data about good and poor services

Geoff Mulgan talks in his lecture about the concept of Failure Demand: demand caused by a failure to do something or do something right for the service user or client and often involves miscommunication or lack of action. This is different to Value Demand - demand responding to what the service exists to provide. Overcoming failure demand is important to the experience of the service and is about improving quality. Where an organisation is failing can be identified in a number of ways - from quantitative data including escalating costs and data collection, to qualitative methods including the number of patients return.ing to a hospital or investigating services using such methods as ethnography. 

Positive deviance is the name for a service or organisation that despite the having the same resources to other comparable operations achieves much better outcomes through a combination attitudes, behaviours and practices. The Positive Deviance Initiative aims to help organisations understand why this is happening and provides a variety of tools and advice.

Both these approaches provide public services with greater understanding of how to improve the experience of services and their success.

Engaging with citizens on what they want from services

Geoff Mulgan recommends the Kafka Brigade, a network of individuals who champion person-centred approach to innovation and use methods similar to action research to involve citizens. They are most interested in working in areas where bureaucracy and regulation has stifled innovation, and services are failing to meet the needs of individuals. Also mentioned is the Complaints choir's unique approach - using song - to enable individual citizens to feedback their opinions about public services 

Gathering ideas from citizens

South Korea and the World Bank have both set up Idea Banks as way of providing citizens to contribute their thoughts and ideas online. 

Innovating with citizens

Some organisations have started to use tools and methods used usually by designers to engage and collaboratively develop public services with citizens. Mind Lab in Denmark and the 27th Region project in France are both using this approach.

The Living Labs approach to innovating brings in people with different experience that reflect the diversity of regions to be involved in real life testing and experimentation of new ideas.

Another model for innovating that Geoff Mulgan mentions is Social Innovation Camp which has produced several new organisations with very specific social aims to date including: My Police and the Good Gym

Social enterprise

Social enterprise is suggested as model for delivering different kinds of services - with examples such as Alcoholics Anonymous which has successfully used this approach. If you are interested in finding out more about social enterprise you can contact the coalition or if you would like to find a local social entrepreneur get in touch with Social Enterprise Ambassadors.

Organisation models that challenge the status quo

Models that change the system are also discussed in the lecture. The free area of Freiberg in Germany and Hammarby Sjostad - a new community which has been designed holistically to consider every resident as part of an integrated eco system. In a different field the School of Everything challenges the notion of education establishments and teacher-pupil roles. Everybody is able to sell their services as an educator.

Recorded at Composing the Future, Innovation by Design.

Date of recording
Audio transcript

What I'm going to talk about two things which I suspect are, if they're not already, will be quite live in your sort of day jobs. One is about productivity and how do you cope with pressures on money and outputs and outcomes and so on, and the other is tools of innovation to help you get to better productivity.

I'm going to take you through some methods and frameworks and then apply it to one or two specific topics that at least some of you will have far more knowledge than me. So I'm partly trying to get out of you some interesting answers and then we'll go on to a next stage and reflect on some things. 

I'm going to start off by talking about innovation and my interest is really on how do public services or professions deal with what are called still by some as a 'wicked' problems, deep ones, intractable ones; like climate change, ageing and isolation, waste, drugs, obesity, poverty and you know you can add on to your list as many of those as you want. But in particular how do you deal with these in times of very heavy pressures on money and if you're under pressure to deliver rapid improvements in effectiveness in for example services in relatively chaotic families or care for isolated older people perhaps with dementia or something, what are your tools for doing it. If the current models, service models, simply can't achieve enough against the backdrop of probably both rising demand and reducing resources - the scissors seem to be a dominant issue across the public sector, not just in the UK but elsewhere - what do you do, apart from panic? It may well be worth panicking, but apart from that, I'm going to try and talk through some methods.

My starting point is if you think of the scale of impact that these changes can have very crudely from small to large and the speed with which they can have an effect from one year, two years, three years, four years. There are methods around improvement and you know many people in this room will be very familiar with them, around performance management tools, uses of price pressures and tariffs and contracting to drive improvement. There's a set of tools around speeding up adoption, adoption of better tools, better practice, either from elsewhere in the country or the world or from other sectors. But most of the larger impacts on productivity will tend to come from innovation. They always have in the past, they probably will in the future. These take a bit longer to have effect, but are key if you are faced with a 20, 30 percent change in the underlying economics and that's what I'm going focus on. I see this as entirely complimentary to classic improvement tools, but in a way improvements tools are used during the good times, when there's lots of money and it's okay to go for one, two, three percent increments. In harder times, all the research shows it's in times of financial stringency that the attention shifts to innovation just because there's no option. And that's been a repeated finding of studies. In boom times people talk about innovation a lot but they don't really do it because it's a bit too threatening or challenging or difficult, but if you have got 10, 20, 30 percent cuts you've got no choice.

I'm going to talk you through a loose framework for thinking about methods for innovation and I'm going to go through each of these and come back to it at the end. As an example to explain what I'm trying to describe would be something like a rising number of people in a city like this over the age of 70 experiencing severe loneliness and isolation. That's what I would call a prompt, a kind of new fact, something that which in a sense throws out the challenge to the professionals, the decision makers, the managers, to do things in a different way. From an issue like that and they're will be dozens I'm sure in your minds, to how do you come up with ideas, proposals, things which might be a way of addressing the issue in the case of elder isolation it might be, befriending schemes or time banks or it could be technological solutions, they're will be a whole host of things which at least in theory could help make people feel less isolated, less cut off in their homes. 

You can turn these ideas into things you could specifically trial ie. prototypes where it's clear who would do what, what the cost would be, a time scale for it, a pilot and so on. Some of those will work, some of those won't. The ones which do work, you would try and hopefully sustain, embed in the service, or in policy or perhaps in a market and then if it's really working well you'll try and grow it perhaps across a city, across a nation or perhaps across many nations. And then if it really works well you may move on to the next stage where you actually change the systems around this model. So it could be in the case of elder isolation you start off with something like a phone based sort of volunteer, befriending scheme, you try that, it works quite well, it's quite cheap, it gets in a sense a budget line attached to it. It turns out there are quite big economies of scale for the project but then around it you discover there are all sorts of related needs and you find that the system you've got with the telephone befriending service can do lots of other things as well, perhaps handle meals on wheels or handle some kind of diagnostic. Anyway that's the basic framework I want to take us through and I'm expecting also to challenge it and pull it apart.

So starting off with prompts - what are the things which might act as the trigger to do things differently? This is a whole series of things, which might work in this way, it could be feedback from your users, from citizens. It could be a new government is elected with a mandate to do something in a radically different way. 

In health and in manufacturing there's a concept of failure demand - are any of you familiar with failure demand as an idea? Failure demand is basically the idea that the demands coming in are basically the result of a previous intervention not working. So it's like if you're a car manufacturer it's the number of cars that within three months that have come back because they're not working. Now ironically this is a concept which comes from Toyota which is really good at dealing with failure demand and is now the sort of paradigm case of the problems of failure demand. In hospitals it's a very dominant issue where you know a high proportion of particularly older patients will be back within 3 months because of something not having been done right in their discharge. 

Every public sector call centre will have between 20 and 80 percent calls coming in basically because of something done wrong in a previous part of the service interaction. This is also another trigger from data. 

A lot of the public sector is now using ethnography as a prompt, looking at what the day-to-day experience of a service is, how it's felt as well as how it looks from a sort of managerial point of view.

Cost escalation is also an increasingly important one so where you know that over the next three to four years - other things being equal - your costs will go up. That's a pretty strong prompt to want to innovate in a different way. 

Are any of you familiar with positive deviance as an idea - the origins of this actually came from Vietnam and it's the idea that if you got a system, are there places that are doing much better than they should be? And looking at why are they succeeding against the odds? So it may be 16 year olds that have got a very serious criminal record but nevertheless are a year later in a job and doing quite well. What do we learn from those or some.thing that is succeeding against the odds. 

The other triggers are visits, so going and seeing something completely different acts as the most useful trigger for innovation in the public sector. Perhaps you go and visit, I don't know, a bank or another country or a community project that makes you see things in a different way.

What's different about these prompts is whether they help you get to the right diagnosis, what's the real underlying cause of a problem rather than just tackling the symptoms. But it isn't always the case and there's some fields where there's good reasons for only deal.ing with symptoms because you can never know the causes because of slight subtleties.

There are a few other examples that I quite like. There's a thing called the Kafka brigade, an idea which came from the Netherlands and is now working in Wales. They get groups of citizens to work with a public service, investigating a particular service and breaking it down into its component parts and getting rid of unnecessary bureaucracy, that's why it's called Kafka. Domestic violence is the main area they have been working on but they've got a method for prompting different perspectives. 

There are also complaints choir, an idea invented by a student in Finland about four years ago which has now spread to about 20 countries at the last count where people come together to sing about the things that really annoy them in a city. It's a very therapeutic but actually quite helpful kind of feedback mechanism to the city. I was in Seoul in Korea last year and went to a festival of complaints choirs and they had 12 competing choirs and a prize for the best complaints.

Alternatively, the prompts might be something quite different like insights from neuroscience. A lot of social services and care is being at least challenged and sometimes being re-thought from insights from neuroscience. The slight problem is that the insights seem to change every few years but certainly new knowledge can often be a prompt.

A very different example is forum theatre, which was developed by Augusto Boelle in Brazil and is similar to complaints choirs because it's a way of uncovering the real experiences of a community and their real needs but through the use of theatrical methods. It's another method which has been used in other countries as a way of getting under the surface of problems and not just using the formal data - which public services use - to get richer insights.

That's a set of the prompts and I'm going to come back and ask you to think through what might be some other kinds of prompts. Off the back of those how then do you turn a prompt into an idea, a possible answer. You'll all be doing this all the time in some ways and these are just a few of the methods, which we found out there. You can do a competition; you can ask the public or indeed your own staff if they've got a really good idea and perhaps give them a bit of time off or a few thousand pounds to work it up.

The city of Seoul in Korea has an ideas bank where any citizen can propose an idea. Last year they had about a hundred thousand ideas were submitted and 2,000 they put straight into practice, usually very simple little ideas but doable. The World Bank also runs an ideas marketplace. There are some governments that use A Teams where they get a group of more junior officials to work on problem solving a topic. One case I worked on in Australia they also linked in the local art school and film school and so on, so when they made a presentation to the Premier they also had a sculpture, exhibition and a film and so on alongside their recommendations which made them much more powerful.

In Information Technology there are living labs. In NASA and other technology orientated companies, skunkworks where you deliberately create a team slightly outside the line management to work up ideas. New York's Waste Department had an artist in residence to look at things in a different way and the first thing she did was actually go - well it took some time - and meet all 10,000 people working in the waste system in New York and she just said 'thank you' to all of them. That was such a shocking experience for them that the management were able to open up a quite different discussion about what could be done differently.

There's a lot of interest at the moment in design methods. For example using visual techniques to make, for example the experience of going through a hospital or experience of being a child in care, very visual and visible. The design world has also been good at the idea of rapid prototyping where you quickly put things into practice even in quite a half-formed way. And examples include the Mayo Clinic (health care) in the States, in Denmark there's a thing called Mind Lab which works in four government departments as the kind of in-house creativity consultancy. France has a thing called the 27th Region, they have 26 regions and all 26 regions share this kind of virtual space for doing the things they find hard to do for real in the other 26. 

And I don't know whether any of you heard of the last one of these in Glasgow, called Social Innovation Camps (SI Camp) , which get groups of web designers and programmers together with volunteers from the public sector and elsewhere over a weekend and they're set challenges to design a website which meets a social need. And by the Sunday evening they have to have a working website and then the top three get a prize. The one which won in Glasgow was a feedback site for the Police which is called MyPolice. It's worth looking at and the winner of the very first SI Camp was a very simple site for disabled people to design products together online, called Enabled by Design, and they've now got a whole series of really good products have come out of that. Another example is called The Good Gym, which is about elder isolation and said why don't we link together joggers and lonely old people so you jog to go and visit someone and I thought, who wants to be visited by someone who is really sweaty, you know. But it seems to be work.ing rather to my surprise and it's now a separate organisation with staff and so on.

But there's then the question, if you've got a good idea how do you trial it? And I'll be interested in what are your current methods of testing things. You know you can do the formal pilot, you can set up a 2 year pilot and try and freeze the model and evaluate it. There's randomised control trials at one extreme that are used in Welfare to Work and medicine, though they actually originated in agriculture. There are experimental zones like employment zones which Glasgow's had. In say the design and service industry world there's rapid prototyping where you may quickly put something into practice in a very small scale but learn by doing. And quite a few public agencies are using simulations where you actually, for example, before you spend millions on a new building you get a sort of simulation of the building and get some of the people who might work and live in it to really play in the virtual building and that then reveals some of the things that might go horribly wrong once it's a real building. Simulations were also used for the Health Market Reforms to get the participants to game how they would behave in a quasi market you couldn't have worked out on paper about how they act.

And I'll give one example which we're doing for speeding up this sort of prototyping process, we're created a role called the Social Entrepreneur in Residence in Birmingham in the Health Service and in Kingston. It's a joint appointment of the Local Authority and the Health Service and their job is to find the promising ideas in both the public sector staff, but in social entrepreneurs as well, and help them on a journey of better defining the idea, trying it out on a very small scale, turning it into a venture and so on. It's a way of bridging the inside of the public sector and some of the potential creativity outside and seems to work really well.

The fourth sort of tools again you will all have ones in your own organisations, is about if something does work how do you then make it mainstream? You may need some kind of formal validation process which really says it does work, you know, it's cost effective. You may simply make it a policy commitment again a national government or a city can just sort of announce a policy with a budget attached or programme funding. There's growing interest in investment models for sustaining things, so you'd turn the model into a social enterprise or a kick or something like that with public loans or equity as well as commissioning. There's whole sort of issues around the skills and training you may need to sustain a new model. Usually anything radical will require a new set of skills as well as a new model. And lots of different examples from the way commissioning works in the NHS and I now south of the border much better than here there are programmes like Sure Start that was attempted to be embedded initially it was embedded largely outside the public sector through voluntary organisations and then migrated back into the public sector through children's centres. My own organisation in the past promoted the idea of extended schools, which again was an idea which grew some way outside the system and then got legislation in England and Wales which made it mandatory for every school to provide a longer day. And there are things like social enterprise investment funds, the Department of Health in London now has £100 million social enterprise investment fund to sustain promising new models in a, more like a venture.

The more sort of difficult and interesting territory is when you want to grow something large. And again I'd ask you each to reflect when something is clearly working in one place what are the methods in your organisation or the bit of the systems you work in which then take things to a larger scale. Now some of them will be again just a sort of policy directive or decision making at the top and budgets attached. In some fields there's a key role played by arms length bodies like Inspectorates or Auditors or NICE in health, National Institute of Clinical Excellence, which rules on what is a cost effective approach and in theory at least Commissions, Purchasers are meant to act on that information, although even in the case of NICE it takes on average 7 to 8 years before there's large scale implementation.

There are lots of, in a sense, business ways of scaling things. You can do it through a franchise or licensing. So the Wise Group in Glasgow tried franchising its model to other cities, with some difficulty, but that was its approach. Licensing is used for things like parenting classes, Webster Stratton Parenting Classes, I don't know if any of you have purchased those but that's a classic sort of social innovation which then was turned into a quite formalised model which then public agencies buy, you know, in a very prescriptive form.

Other ways things scale is through take-over, someone just takes an idea or just takes over the organisation which promoted it. Professional networks obviously play a very important role with professional advocates promoting a new model and other people just copying it. In some fields intermediaries play a critical role, in technology that's certainly the case that there's usually an intermediary body that just gets good at linking the sup.pliers and the demanders. And there's a lot of interest now in new kind of financing tools for scaling. The Ministry for Justice recently signed the first ever Social Impact Bond, which is an idea that we promoted about two years ago for essentially raising money for investment in a group, for example 14 year olds at risk of prison or people being released from hospital at the risk of readmission. The Government re-pay according to milestones reached in reducing that group coming back and as cash is short there's a lot of interest in these kinds of new financing tools to better incentivise outcomes. These in principal help to scale and grow effective models of action.

Another very different example is NHS Direct, obviously a large scale example, in some ways its origins lay with Michael Young who set up a thing called Health Line, that's a phone based diagnostic service with a handful of people and then government sort of did it on a much larger scale as single entity, but organised outside the mainstream NHS Management channels. 

A very different example is M-PESA, have any of you come across M-PESA? Does anyone use M-PESA in Glasgow? It's a very interesting story and I think the inventor was from Strathclyde. A Manager in Vodafone had the idea in Kenya of using the phone as a bank. And this was for lots of people in Kenya didn't have a bank account but lots of them did have a phone so he said, why don't we make the mobile phone basically a banking service for transmitting money. Vodafone put a little bit of money into it, it now has 7 million customers including half a million in London and probably 20,000 in Glasgow. It's often used for remittances back to a home country. That's a rapid scaling model and it's an interesting one because it's one of many examples of where there was a technology which was widely used, you probably all have a mobile phone in your pockets or your bag, but much of its potential was not being mined at all. It hadn't occurred to Vodafone that it could be a bank, just as it hasn't occurred to them lots of other things which you could use it for.

I like Alcoholics Anonymous as a model that was developed in the thirties but which has scaled dramatically to the present existing in a hundred countries. But the precise opposite of those other two in that it has no head office really, it has a very simple set of principals, the 12 step principals and a cellular structure. It was actually based on models of religious organisations, non-conformist churches which wanted to avoid having centralised hierarchy and structure and I think quite a lot of things are now growing and scaling more like Alcoholics Anonymous than like NHS Direct and it's just worth keeping that in mind particularly for social projects.

And the final group I wanted to talk about it what happens when whole systems change? When really the very ways in which something is done transforms itself. An example - I think there are lots of examples around - is waste and re-cycling. Everyone encounters this when you go into your kitchen and you have to separate out your plastics and your paper and so on into different bags. In a sense we've moved from a system where waste was collected by the Council and buried in a big hole, somewhere out in the ground, to a system where a lot of waste at least is re-used. That is a true systemic change that required all sorts of things to change in tandem. So the law had to change including our penalties for not doing it. Regulations had to change so regulations on the newspaper industry but is just I think they're now required to recycle 70, 80 percent of newsprint. EU Directives which require car manufactures to recycle most of their cars. There's usually a changed script and this is a concept I would like us to come back to, so a script is really describing the ways in which people behave day to day. So the waste script used to be, you get a bit of rubbish and you chuck it in a bin, and that's pretty much the end of it from your point of view. And instead for recycling the script had to be, you've got a bit of rubbish and you have to think about what it goes with, what it doesn't, you have to put things out and there are different scripts for the collectors as well.

Shopping would be another example. So the script 40 years ago was you go into a shop, you ask the assistant to fill a bag for various things from their shelves and then give it to you and then you pay them. And then with supermarkets the script became, you went around the shelves, filled up your trolley and then paid for it and in fact the latest script is of course you pay for it without a human being being there, through a semi-automated payment system. 

The really interesting thing in public services, what are the scripts that could be changed? We've come to assume a certain set of behaviours for example Probation Officers have a particular script of how they interact with someone on probation, weekly meetings, certain kind of forming filled out, not at all obvious that's a sensible way of doing it and we tend to become prisoners of scripts. And anyway in these kinds of systemic change you end up with different kind of markets, different metrics, different things are measured, different kind of finance. And I've talked about waste, I think a very similar, potentially similar example is happening around personal budgets for care which have been now introduced in some states in the US, Sweden and in the UK because once you get a personal budget and you can, to a degree, control your mix of services potentially all sorts of things have to change around that including the markets for services, including information systems, including advice systems and so on. 

Eco towns would be another example of systemic change. If you're building an entirely new development you have to think of it completely systemically. The best examples in Europe are, in Germany in Freiberg, and Sweden, Hammarby Sjöstad, which are really the sort of purest examples of trying to build cities as we will be living in them in 30 or 40 years time. In those cases they have very dictatorial town planners who make it very hard for people to use their cars or even to have a car have very strict rules on things like recycling and energy and so on. People who live in these places absolutely love them and they are now being copied and so it's a very different sort of systemic view about what a city should look like.

Another example of systemic innovation, and one which I've been involved in, is one of the most challenging for public sector Managers. This is called the School of Everything and it was prompted by the realisation that in the UK there are about a million teachers who weren't in the public system. They were teaching music or yoga or sports or car mechanics, a hundred and one things and so the simple idea was could we create an organisation that linked all those teachers to people who wanted to learn but without necessarily going ever through a school or an FE college or university, an institution. So that's the School of Everything so you could log on to it and see what it is and all it does is link person to person to learn and you then actually have a face to face sort of lesson but it's a - and it has a few tens of thousands on it, it's still really in a sort of trial form, but what's interesting is having put that out there as a platform on the web it now has users all over the world. We had a visit from a Chinese delegation last week and almost as an experiment I showed them the School of Everything website and then tapped in their cities in China and we found teachers in all the cities they were from on the School of Everything even though we'd never promoted it in China. But as I say that's a very radical innovation to deal with an education issue but not coming at it from the point of view of, let's provide a new set of courses, a new school, a new college and so on. Having then created that it then does raise the question; well could we use variants of that within a social services department as a platform for linking experiences and knowledge between different members of staff? It opens up a different way of thinking.

What I thought we'd do is just perhaps have a few minutes of any sort of comments or questions and then what I'd like to do is take a specific example and ask you in groups to go through what might be the prompts, proposals, prototypes and so on. And the issue which I thought we might do is services for families at risk of a child being taken into care. It's a live issue everywhere. If you're asked, in a sense, to come up with options for more radical innovation in that field, what would be the prompts, the things which would as it were define the right way of shaping the question, how would you come up with proposals, how would you test them out, sustain them and so on? 

So the question is what money do you need - actually in each of these stages I would ask you to focus on what resources are needed both for the prompts and proposals but certainly for systemic change as in the case of direct payments, often you do need quite a bit of sort of transitional resource because you're keeping an old system alive while you're also introducing a new system and it's very hard to do that without extra money, although sometimes it does happen. And the related question is what I call the decommissioning question, so you say lots of money tied up in buildings and salaries and staff in a particular model and you can only create the space for the new model, like direct payments, if you're willing to actually shut down and decommission the acute hospitals etc and that is politically always very hard, and hard in terms of professional power as well. But I would ask you to keep that in your minds as you go through that, what are the implications in terms of money and power essentially at each step of the way. 

Q    A lot of people are working in this field at the moment and it's one of the biggest points for research...I think the Big Lottery Fund and there are problems when...really good projects but in some cases they don't bring the wider local authority along with them on the journey and they're presented with findings at the end which they're not prepared for and if they'd been involved in the journey at an earlier stage would make adoption much faster.

Yeah, if you reach as it were the key budget and power holders only when you get to point 4 and they've had no involvement in points one, 2 and 3, they'll be baffled and probably hostile. 

Q    You said earlier that innovative change driven by hard times may take longer but gives a bigger impact but no hard times are unprecedented in local government, you kind of feel that innovation taking time maybe brings bigger rewards, is that something that under the current circumstances is realistic in that what we're talk.ing about is...today, if not today tomorrow and certainly not the day after.

We've been thinking about our work with local authorities facing their 20, 30 percent cuts and how do they handle that immediate impact. You've got to act now and what we've done is we've tried to break down what we call 12 different kinds of economy or saving which a local authority needs to be using and planning and thinking through. So the first ones we call the traditional economies, and the fact they're traditional doesn't mean you can avoid doing them. So the local council is thinking what things are we doing which are nice but we simply can't afford so we stop them. And one extreme it might be library provision or it could be weekly bin collections as opposed to biweekly. There are a whole series of economies of trimming where there's the odd slicing bits off, could be pay freezes, efficiency savings, it's kind of applying Easy Jet, Ryan Air thinking to your service. There are also economies of delay where you delay capital programmes, pay rises, recruitment and so on. There are lots and lots of other things in this kind of basket which pretty much everywhere is doing most of these, it is the standard response to money being tight. 

This is what we call five sets of organisational economy because these don't usually deliver you enough. They start delivering you some savings but they don't deliver you the big sustainable gains or they may so damage service quality that you get a public back.lash. So these ones are looking at how can you reorganise things in ways which deliver you savings but actually may deliver you as good a quality of service to the public or as good outcomes.

If you ask management consultancies they'll always give you economies of scale as the answer. So you aggregate back offices or your call centres. Sometimes there are savings there, in my experience these are usually exaggerated but there are undoubtedly many bits of the British public sector where there are untapped economies of scale.

Another approach is economies of scope and that means when you bring different things together in ways which save money. So a one-stop-shop should give you economies of scope. If the pension service visitor to a household also deals with a number of other things in the same visit like for example housing benefit etc then that saves money without in any way reducing the quality of the service for the individual.

A fifth approach is called Total Capital which tries to look at where different public organisations are commissioning new buildings or refurbishment in the same place and there are lots of examples where you might have four or five different agencies all duplicating the specifications the architects and all of that - pure waste and quite substantial economies of scope to be realised.

The sixth one is economies of flow which is a slightly different approach which really comes from experience in some parts of manufacturing and parts of health where if you actually look at maximising through-put of a particular service you can achieve quite big savings and a better quality. In health the purest things are like eye operations, or hip operations, if you get the same surgeons specialising in one type of activity not only can you greatly increase the through-put going through them, they're success rates are much higher than if they're trying to be general purpose surgeons. It may be slightly more boring for them but there are quite dramatic economies of flow in surgery and there almost certainly are in other kinds of service. We're working on a thing for neurological conditions where we're creating a service which brings together specialist doctors and nurses, mainly working from a call centre, to people with multiple sclerosis and interacting with them directly and particularly managing crisis directly rather than them each going to their GP or going to hospital appointments and it appears that that saves about 40 to 50 percent of cost and sharply improves the quality of service experience for the patient and that's essentially by concentrating flow in a particular point rather than aggregating it.

There's another set of economies, economies of penetration, which is where you actually use locality to concentrate services. So a street concierge is an example where someone is paid maybe half a day a week to keep an eye on other people in the street, like isolated older people, perhaps to take in deliveries, to do low level fixing, handyman kind of services. It's much more economic if you have one person per street doing that rather than organising a whole service across a large area. In telecoms and energy they're very big on economies of penetration, that's why combined heat and power makes sense, but it requires you to require everyone in the street or in the estate to take up the CHP offer.

A different approach is circuit economies where you try and reduce failure demand. You try and stem the flow of costly events onto the public service and the purest example would be dealing with desisting from crime. You're investing money in diversion, prevention and so on in order to reduce the need to pay £40,000 a year for someone to be in a prison.

And then the next set are called relational economies, where the saving comes from slightly changing the relationship with the citizen. So there's some pure ones that are like responsibility where you just pass out a responsibility to citizens, so it could be more self testing, self management for people with long term conditions. It could be requiring citizens to separate their waste rather than the local council. Some councils are saying maybe we should say to people, you've got to sweep your own streets and we'll only come and sweep them every 2 weeks if you as a community sweep them the other week.

There are economies of visibility, MP's expenses is the best recent example. You make something visible, you save money. The Conservative Party in London is proposing that all public contracts over £25,000 will be put on the web. This causes horror to many people and much of the consultancy business, just simply the fact of visibility and shaming will change behaviour.

There's economies of regulation where you actually adjust your, in a sense, your appetite for risk downwards because there's less money around, rather than assuming you should have the same sort of level of regulation at all times.

And finally economies of commitment where you move functions perhaps from less highly committed organisations to more committed ones. An example would be a rural bus ser.vice where you can run that as a classic private company or public service with fully full paid staff and you then cut it when the money runs out or you can try and recreate it with a mix of paid staff and volunteer staff from the communities being served. That is achieving an economy of commitment.

These are all elements of an answer to how do you think about savings under pressure and some of these take quite a long time to realise. Some of the scale ones and scope ones are quite hard to do in less than a year although there are examples. But some of the approach can be done quite quickly or at least the first steps can be made quite quickly even though all the emphasis initially is bound to be up here.

Geoff Mulgan