2014-03-25

Introduction

It’s a privilege to be a guest of the Employee Ownership Association and a particular honour to deliver the 2014 Robert Oakeshott Memorial Lecture.

The Association is the bulwark of this movement and its objectives show the time has come to seize this moment for the British economy. It’s also great to see representatives from John Lewis, Arup, Gripple and some of the other incredible private sector success stories here today.

Robert Oakeshott lived through one of the most polarised eras in our political history, yet his own place in that order was never fixed. His party political affiliations were with the centre-left but his writings found expression across the pages of the Spectator, the Economist and the Financial Times.

Independent-minded, he was a man of strong beliefs. Today the name Robert Oakeshott has become synonymous with one cause above all others – the cause of employee ownership.

Historically, employee ownership has never had a fixed ideological abode: often shunned by the left because of their dogmatic commitment to state ownership; down-played by the right in favour of what some saw as classic red-blooded capitalism. I remembered last week when reading Tony Benn’s obituaries how he tried to save 3 companies in the 1970s by turning them into workers’ cooperatives, notably the Triumph motorcycle works at Meriden. It was an experiment in syndicalist industrial reorganisation; and its failure set back the cause by years.

In the same vein we need to ensure that the problems faced by the Co-op Group, albeit failures of governance in a mutual of members rather than staff, are not allowed to contaminate the good name of mutual.

Public sector productivity

So how does all of this bear on the reform of public services? It’s because in 2010 we faced a public service crisis. We faced the biggest budget deficit in the developed world, a flat-lining economy and rising public expectations of the quality of services. Between 1997 and 2010, productivity in the public sector flat lined. Yet in the private services sector over the same period, it rose by nearly 30%. Even a back of the envelope analysis suggests that if productivity had risen by the same amount in the public sector, the annual deficit could have been at least a quarter smaller - a cut of around £50 billion; or the economy larger by a minimum of £2,000 for every household. However you calculate it our economic and fiscal position would have been radically different.

As a Treasury Minister in the early 1990s, I was charged with developing the Citizen’s Charter, an early programme concerned with the improvement of public services. I faced what felt like a tacit conspiracy of defeatism. The Treasury struggled with the idea that services could be improved without departments demanding more money. And departments, faced with demands for better quality, entirely predictably confirmed the Treasury’s gloomy prognosis by – yes – demanding more money.

Absent from both sides was any recognition that productivity could be improved; that more could be achieved for the same amount; that the same could be achieved for less money; least of all what we are now amply proving: that you can deliver more for less. This was the defeatist consensus that has held back public services for so long.

This approach reached its nadir in the last decade. The National Health Service budget more than doubled, but productivity if anything deteriorated. The apparent age of plenty relieved the public sector of the need to be creative. And then there was no money. As Ernest Rutherford famously is alleged to have said: “We’ve run out of money. Now we must think”.

Public service reform - 5 principles

Our thinking led us to propose public service reform which has followed 5 principles.

Openness, because transparency sharpens accountability, improves choice for the public and raises standards.

Digital by default, because as well as being an order of magnitude cheaper - 30 times less than by post and 50 times than face to face - services delivered online can be faster, simpler and more convenient to use.

A properly innovative culture, so public servants have permission to try sensible new ideas.

Tight control from the centre over common activities – like property, IT and procurement – because it reduces costs and encourages collaborative working. These tight central controls account for two-thirds of the £10 billion we saved for the taxpayer last year in central government spending.

Loose control over operations, which is where employee ownership steps in. The people who know how to deliver public services best aren’t the politicians in Westminster or the bureaucrats in Whitehall and in town halls, but the professionals working on the frontline. Tight control over the centre must be matched by much looser control over operations.

Opening up public services

For too long, delivery of public services has been shackled by a top-down, Whitehall-knows-best attitude. Public sector workers were left feeling alienated: dispossessed from effective control over their ability to shape the services they had responsibility for delivering.

But too often there seemed to be a binary choice: either the public sector as a bureaucratic in-house monopoly provider; or on the other hand full-blown red-blooded commercial privatisation or outsourcing.

Happily there are now alternatives. Social enterprises. Joint ventures. Voluntary and charitable organisations. And public service mutuals.

And it’s this last – the public service mutual – that is the fastest growing alternative, which is arousing most interest among governments abroad, and which I believe will increasingly be the way of the future.

Creating a new sector

Why? Because creating a mutual releases creative energy and entrepreneurialism. And that’s the problem many critics have - they either don’t believe entrepreneurs exist in the public sector, or they don’t think entrepreneurialism mixes with the public service ethos. Both are wrong. There are loads of latent entrepreneurs in the public sector. They may not think of themselves as entrepreneurs, but they have all that spirit of enterprise, the willingness to back their ideas, and invest their energy and creativity to make things happen.

It doesn’t mean they all want to be millionaires. In most spinouts the staff themselves have chosen that the entity should be a not-for-profit. They didn’t need to but that’s the choice they made.

Yes, you can get improved productivity through conventional outsourcing. This will often be the right option. But rarely does it deliver the almost overnight improvement that mutualisation can stimulate.

The last government started down the path of mutualisation. But their approach was too top-down, too prescriptive and bureaucratic; and resulted in no more than a handful of new mutuals.

I decided against this approach. So there was no White Paper; no all-encompassing strategy, or big bang media launch. It was the JFDI school of government. We didn’t start with the theory and move on to the practice. We did it the other way round. Find a hundred flowers and build a hothouse around them so they can bloom and grow.

So the first thing was to identify groups of workers who wanted to spin out from the public sector. As Pathfinders, we gave intensive support to these organisations, who in return shared their experiences with us and with others. Many of these Pathfinders are now among the country’s best performing mutuals.

Next, we made £10 million available through our Mutual Support Programme. Not a lot. But we’ve made it go a long way. The funding isn’t allocated directly. Instead it’s used to build the capability of these new businesses through professional expertise and advice. This way the government can negotiate the best deal and, over time, we’ve built up a valuable set of tools and templates which upcoming spinouts can draw upon for free.

And there’s no ‘1 size fits all’ format. Some are conventional companies; some companies limited by guarantee; some community interest companies; some choose to be charities. Some have 100% employee ownership; but to qualify there must be no less than 25% employee ownership so that staff can exercise at least negative control. There’s a whole spectrum of different models available, and each group must select the right course for their horse. Each spinout is a journey for its own staff. They’re the ones in the driving seat, leading the change.

Progress so far

And our approach is working.

4 years after the last General Election, the number of mutuals has increased tenfold. They employ more than 35,000 people, delivering around £1.5 billion worth of services. They’re in sectors ranging from libraries and elderly social care to mental health services and school support. They range in size from a handful of staff to upwards of 2,000.

Neither is this confined to any 1 region – it’s not a London niche – it’s a great national success story. The results are spectacular. Waste and costs down. Staff satisfaction up. Absenteeism falling. Business growing.

Staff engagement surveys bear out the simple truth that service improves and productivity rises when the staff have a stake; when they feel they belong; and that their individual voice and actions count.

Our latest data shows that after an organisation spins out absenteeism falls 20% and staff turnover 16%.

Take City Healthcare Partnership based in Hull as an example. 91% of staff said they now feel trusted to do their jobs – and this level of empowerment has had a knock-on effect in the care they give. Since they left the NHS in 2010, there has been a 14% increase in patients who have rated their care and support as excellent, and 92% say they would recommend the service to family and friends.

No wonder City Healthcare came 46th in the Times 2014 Top 100 Not for Profit Companies to work for.

At SEQOL in Swindon, a groundbreaking mutual formed out of the integrated PCT and social care arm of the council, staff proudly showed me the stockroom, where a nurse had painstakingly attached stickers with the unit price of each item. “Why did you do that?” I asked the nurse showing me round. “To make us more aware of saving money”, was the response. “But why?” I persisted. “You’re a not for profit organisation and none of you will benefit from the savings”. “No. But every pound we save makes us more competitive. And it’s a pound we can put straight back into better patient care”.

And that’s the point. There’s nothing wrong with better financial reward for public servants. But it’s not the biggest driver of better productivity. It’s the satisfaction people get from putting their ideas into action, and seeing swift results. It’s the sense of pride that it’s their organisation that is delivering the service. That they can make improvements quickly, taking responsibility for making things happen, without new ideas getting bogged down in bureaucratic treacle. Just looking at the Baxendale Awards for Employee Owned Businesses this year, you can see the spinouts dominating the innovation category. So public servants can give effect to their public service ethos with immediate and gratifying speed.

Whenever I visit a mutual, I always ask staff the same question. “Would you go back to work for the council/health authority/ministry?” The answer is always “No”. “Why not?” “Because in a mutual we can do things”. That’s why the public service mutual is the way of the future.

Growth

The public service ethos remains front and centre. But it doesn’t have to be at the expense of strong commercial instinct. Spinouts are winning new business – fast. A study from Boston Consulting Group projects an average increase in revenues of 10% this year.
MyCSP, which is responsible for administering the Civil Service Pension Scheme, was the first mutual joint venture to spin out from central government. Under its contract the cost of the service to the taxpayer will halve over 8 years. The private sector partner has taken on the transformation and IT costs that would otherwise have fallen to the Exchequer. Under its innovative equity structure, the government retains a 35% stake. The staff, with a 25% stake, all received a first year dividend of nearly £700. And in its first year of operation it gained 47 new clients.

Social AdVentures in Salford saw a 262% growth in revenues last year.

3BM in London increased their business by over 25% in their first year as a mutual.

Since spinning out in July 2011 - in the middle of the economic downturn - Allied Health Professionals in Suffolk has increased its number of staff from 63 to around 100 and its turnover from just over £2 million to £3 million – all on the back of winning new contracts.

And in Norfolk, East Coast Community Healthcare’s first year profit was 40% ahead of plans.

The point is that all of these new mutuals represent a new and dynamic enterprise in the market economy. They are all incentivised to improve and grow. They all strengthen the market for their services and make it deeper, wider and more competitive. And as we have seen all too clearly in the last year, public service commissioners have been too dependent in too many areas on too limited a range of suppliers. Every new mutual helps to remedy that deficiency.

Argument

So - cutting costs; improving quality and supporting economic growth and jobs: what’s not to like about all that?

I had a really interesting experience recently when successfully negotiating in the European Parliament much needed changes to EU public procurement rules. We wanted to shield future new public service mutuals from the full panoply of EU regulations while they established themselves as businesses. We had brilliant support from British Conservative and Liberal MEPs. But beyond them? The Socialists thought that it was promoting privatisation by the back door. And some on the centre-right thought it was a scandalous erosion of the pure milk of the competitive free market.

Let’s examine both contentions. Is mutualisation equivalent to privatisation? Technically yes. Mutuals are spinouts from the public sector into the private or social sectors so they get classified as non-public sector. It’s certainly not privatisation by the back door though. It’s as open a process as you could want.

Does on the other hand mutualisation by negotiation frustrate competition? Not a bit. It opens it up to a broader hybrid economy with a wider range of suppliers – there’s a place for mutual spinouts, joint ventures and charities and voluntary organisations, alongside private companies and the public sector.

So why is such a small proportion of public service delivered by suppliers outside the public sector? Because conventional outsourcing and privatisation is fraught with political and industrial relations risk. It can look ideological and dogmatic, and can arouse the hostility of the staff. It makes managers anxious because of the fear that the contract will be overpriced leading to excess profits and uncomfortable hearings at the Public Accounts Committee.

But mutualisation can square all these circles. If it is driven by the staff and is a not-for-profit then where’s the problem? And if it’s for profit then why not keep a stake for the state? Then the taxpayer benefits along with the staff and managers. And often the alternative to a mutual joint venture is a straight outsourcing. And I don’t come across many public servants who, if their operation is going to move out of the public sector, wouldn’t prefer to have a stake and some control over the new entity.

So for the staff it can feel like a lower risk alternative to straight privatisation and for managers a way of harvesting productivity gains without the downside of immediate competition. Because in nearly all situations, a negotiated spinout into a mutual or mutual joint venture must be followed after a few years by an unfettered competition, where the mutual will have the advantage of a track record and incumbent advantage.

Next steps

But while we have come a long way since 2010, we’ve only reached the end of the introductory chapter. For many years to come the state is going to be facing the same combination of tight budgets, rising expectations and challenging economic circumstances. We will continue to be expected to deliver more for less; so the transformation can never cease – spreading deeper and wider, further and faster. Public service mutuals should be front and centre of that transformation. So we now need to move on the next chapter in the story.

First, I want to remove the blocks that obstruct motivated staff from spinning out from public sector control.

I can today announce plans to establish a peer review scheme to support staff and local authorities interested and involved in developing mutuals. Working with the LGA, we are looking to establish a voluntary programme that will offer best practice advice and examine perceived barriers to spinning out.

And our new Commissioning Academy will continue to build knowledge and confidence among commissioners across the public sector in how to support and negotiate new spinouts.

Second, I want to encourage even more spinouts in those areas where significant numbers of mutuals have already been created, such as healthcare.

Next month, following his review of staff engagement in the NHS, Chris Ham will provide a set of recommendations to government. Chris’ review has looked at how to give staff a stronger role in their organisations, including through mutual models. This included looking at Circle’s strategy for empowering staff at Hinchingbrooke NHS Trust and supporting frontline staff in delivering service change.

And we’re also working with the Department of Health to explore options for increasing staff control across the NHS, including expanding the Right to Provide.

Third, we will focus our efforts on sectors with the greatest potential. I have spoken about the potential in youth services and adult social care. We are also working to expand the offer in children’s services and acute trusts. I have ensured that the current probation reforms have mutuals as a delivery option. We have even received interest from fire services.

Fourth, because what we’re seeing here is the creation of a whole new sector of organisations, I want to make sure that their progress and successes are recorded and underpinned by quality data which is freely available.

So far we’ve done much of the research in-house, in keeping with the start-up nature of this programme. But as with any successful policy, maturity is marked by the state taking a step back.

I am therefore pleased to announce that Cabinet Office will work with the Employee Ownership Association to set up a networked centre that brings together all the data on spinouts into one place, allowing everyone to see how they are performing and what they’re achieving.

Beyond the public sector

The potential for growth is echoed in the private sector experience too. The employee owned sector has been one of the quiet success stories of the British economy in the last 20 years. Companies which are employee owned, or which have large and significant employee ownership stakes, now account for over £25 billion in total annual turnover. And they’re helping to lead the economic recovery, by growing at a rate 50% higher than the economy at large.

So support for employee ownership across both sectors has been and will remain a priority for this government and a core part of our long-term economic plan, a point reinforced by the Chancellor in last week’s Budget, which confirmed 3 new tax reliefs to encourage and promote employee ownership; I’m sure you all look forward to seeing more detail on these in the Finance Bill later this week. This had a nostalgic resonance for me. I recall as Financial Secretary working on the tax measures to support employee ownership that appeared in the 1991 Budget! This will always be a work in progress, I suspect.

Conclusion

So, to conclude, 35 years on from Robert Oakeshott creating the Employee Ownership Association, now is the time to put employee ownership right at the heart of our public services.

Shifting power away from the centre and diversifying the range of public service providers is a historic opportunity to redesign how public services are delivered - not just to reduce costs, but to improve service, increase staff morale and stimulate growth.

And that’s a winning combination.

Show more