Briefing Paper: Elected Mayors

Prof. Catherine Staite and Jason Lowther

 

In this long-read, INLOGOV’s Professor Catherine Staite and Jason Lowther provide an in-depth brief on the role of the new elected mayors, how they relate to the devolution agenda and the things we should watch out for ahead of the upcoming mayoral elections on May 4th. 

 

1. Introduction

The role of elected mayor for regions, such as the West Midlands and Greater Manchester, has been created as part of a move to greater devolution of power over resources and policy, from central government to consortia of local authorities known as Combined Authorities, through which individual authorities have agreed to collaborate in applying these new powers and resources. The Combined Authorities have negotiated individual ‘devo deals’ with central government and, as a result, the extent of their devolved powers varies enormously (see Table 1 below). For example, the Greater Manchester Combined Authority, formed in 2011, has been granted the most extensive powers of any Combined Authority, including powers over the NHS in the GM region. One of the prerequisites of the devolution of significant powers and resources to Combined Authorities has been the creation of a new elected office – that of a directly elected regional mayor.

Screen Shot 2017-04-04 at 08.44.59.png

Table 1: Powers to Devolved in Devolution Deals. 

Of course, mayors are nothing new. Joseph Chamberlain, who led the foundation of this University, was elected Mayor of Birmingham in 1874 and acted as catalyst for hugely significant improvements to the lives of the people of Birmingham in the 1870’s and 80’s; clean water, better pavements and roads, as well as the iconic municipal buildings which still give the city its distinctive character today.   District, Borough and City Councils across the country already have civic mayors, who are appointed from among the council members, not directly elected by the public. They are easily identified, when carrying out their largely ceremonial roles, by their robes and chains of office.  More recently, directly elected “executive” mayors have been created in some local councils.

The question about whether we should have more elected mayors has been hotly contested. Conservative governments have demonstrated a surprisingly enduring enthusiasm for elected mayors for many years, in the face of opposition from many of their own MPs, local politicians of all political hues and the demonstrable apathy and mistrust of the public.

 2. What can we learn from international comparisons?

 There is an international trend towards directly elected mayors, especially in cities. The number of countries and cities that have decided to introduce directly-elected mayors has increased markedly since the 1980s (Hambleton and Sweeting, 2014).   In Europe, directly elected mayors were introduced into systems of local government in Slovakia (1990), Italy (1993), Germany, Hungary (1998), the UK (2000) and Poland (2002).

Outside Europe directly-elected mayors are now in place in many countries including New Zealand.

Hambleton and Sweeting (2014) suggest the mayoral trend is linked to four key themes in urban leadership:

  • The institutional design of local government: the attempt to enable effective civic leadership.
  • The drive for outward-facing leadership: responding to the trends of global competition and the need for “networked” governance where local authorities work with other statutory and non-statutory providers in multi-agency partnerships to deliver social policy.
  • The changing relationships between politicians and officers: including issues around the potential politicisation of the bureaucracy.
  • The relationships between city leaders and followers: with direct election challenging traditional party political models.

Hambleton and Sweeting (2015) summarise the arguments for and against directly elected mayors. Arguments in favour of directly elected mayors include:

  • Visibility – citizens and others know who the leader of the city is, generating
  • Interest in public issues
  • Legitimacy and accountability – arising from the direct election process
  • Strategic focus and authority to decide – a mayor can make tough decisions for a city and then be held to account
  • Stable leadership – a mayor typically holds office for four years and this can underpin a consistent approach to government
  • Potential to attract new people into politics – creative individuals may be able to stimulate innovation in citizen activism and business support
  • Partnership working – a mayor is seen as the leader of the place, rather than the leader of the council. This can assist in building coalitions

The arguments they present against directly elected mayors are:

  • A concentration of power – the model could place too much power in the hands of one person, who is overloaded
  • Weak power of recall – elect an incompetent mayor and the city is stuck with this person for four years
  • Celebrity posturing – the model could attract candidates more interested in self-promotion than sound policy-making
  • Wrong area – the Localism Act 2011 provided for mayors to be elected for unitary authorities when many consider that metropolitan mayors, covering a number of unitary areas, on the London model are needed
  • Cost – having a mayor will cost more money if the rest of the governance architecture of an area is unchanged
  • Our over-centralised state remains – without a massive increase in local power to decide things, the mayor will be a puppet dancing on strings controlled in Whitehall

Gains (2015) suggests that the current weak engagement between electors and representatives argues for a more visible and accountable leadership. She argues that calls for more participation require an activist leadership reaching out to citizens and bypassing entrenched interests such as political parties.

The Warwick Commission on “Elected Mayors and City Leadership” argued that “directly elected mayors offer the possibility of greater visibility, accountability and co-ordinating leadership as well as re-enchanting the body politic, and much of this derives from their relative independence from party discipline through their direct mandate and through their four year term. But they also hold the dangers of electing mayors whose popularity obscures their inadequacy in leading their communities” (Warwick Commission, 2012:7).

They pointed to five reasons often cited for the rise of the elected mayor as follows:

  • A response to the rise of the network society that otherwise disperses responsibility and a demand for greater accountability from political leaders
  • An attempt to reinvigorate democratic politics and civic engagement in the face of apparently widespread political apathy
  • A localist and decentralising reaction against the rise of the centralising power of the state or super state (European Union)
  • The realisation by some local politicians in certain areas that they can make the most impact through elected mayors, not traditional party politics
  • The return of ‘personality’ to the political agenda in place of depersonalised party systems.

 

 International case studies

  • Italy
  • Directly elected mayors since 1993.
  • Mayor appoints executive including non-councillors (often during the election campaign).
  • Limited to 2 consecutive terms.
  • Wide executive powers including roads, education, social services, housing, social security, planning, police, transport.
  • Mayor “acts as a powerful focus point of political decision making and is able to speak to all tiers of Italian government as a legitimate political leader and ambassador for the area. Indeed, mayors are often important players in the distribution of national resources to the localities” (Copus, Leading the Localities, 2006:145)
  • Council can either approve Mayor’s programme or table “no confidence motion” which results in resignation of both the Mayor and the council.

 

  • USA
  • “Strong” mayors predominate in larger cities, directly elected with mayor-council form of government (“weak” mayoral model in smaller towns with mayor indirectly elected by council).
  • Mayor acts as chief executive officer, directs administrative structure, sets policy agenda for the city, determines the details of the budget, and has a veto over council decisions (though may be over-ridden by two-thirds council vote).
  • New York City Mayor elected for maximum of three 4-year terms. The Council is a “deliberative and investigative body” monitoring performance, making land use decisions and passing local legislation.

 

3. How widespread are elected mayors in the UK?

 

The first directly elected mayor in the UK was introduced in Greater London in 2000 as part of the statutory provisions of the Greater London Authority Act 1999.

In England, elected mayors were established by the Local Government Act 2000. Eleven councils adopted a mayoral system (3% of councils), with over 80% adopting the leader-cabinet system.

As of May 2016, there had been 52 referendums on the question of changing executive arrangements to a model with an elected mayor. Of these, 16 have resulted in the establishment of a new mayoralty and 36 have been rejected by voters. The average “yes” vote was 45%. Typical turnout was around 30%, varying from 10% to 64%. There have been six referendums on the question of removing the post of elected mayor, of which three have been disestablished.

The Localism Act 2011 permitted central government to trigger referendums for elected mayors in 10 large English cities. On 3 May 2012, referendums were held in these cities to decide whether or not to switch to a system that includes a directly elected mayor. Only one, Bristol, voted for a mayoral system.

In 2014 it was announced that a Mayor of Greater Manchester will be created as leader of the Greater Manchester Combined Authority. From 2017 onwards there are expected to be directly elected mayors for Greater Manchester, the Liverpool City Region, the West Midlands, and Tees Valley as part of the devolution deals introduced by the Cities and Local Government Devolution Act 2016.

3.1     UK case studies

Greater Manchester

Greater Manchester (GM) has a long history of cross authority working and infrastructure. In 2011 they became the first group of authorities to establish a combined authority. Recently GM has been granted devolved decision-making which is (in UK terms) remarkably extensive. The “price” of this has been to agree to the imposition of a “metro mayor”.

The GM mayor will have devolved powers around housing, transport and (subject to unanimous approval by the constituent councils) spatial planning. They will also become the Police and Crime Commissioner for GM. They will chair the GM Combined Authority (GMCA).

GMCA will have responsibilities around devolved business support, further education, skills and employment, and housing investment.   It will jointly commission (with DWP) the next stage of the Work Programme, and has recently taken on responsibilities around health and social care integration.

In GM, the mayor’s decisions can be rejected by two-thirds of the cabinet consisting of the leaders of the ten constituent councils. The Statutory Spatial Framework is subject to unanimous agreement by this cabinet.

The new elected mayor will be subject to scrutiny by the existing scrutiny committee of the GMCA: the ‘GMCA Scrutiny Pool’, made up of 30 non-executive councillors drawn from the ten Greater Manchester boroughs.

The Government passed an amending Order to create an eleventh member of the GMCA (alongside the ten borough leaders) to be the ‘interim mayor’ until the first mayoral election. Tony Lloyd, Greater Manchester Police and Crime Commissioner, was appointed to the post (by the existing members of the GMCA) on 29 May 2015.

The March 2016 Budget announced the following additional powers for the GMCA:

  • bringing together work on Troubled Families, Working Well, and the Life Chances Fund into a single Life Chances Investment Fund;
  • working with the Government and PCC on joint commissioning of offender management services, youth justice and services for youth offenders, the courts and prisons estates, ‘sobriety tagging’, and custody budgets;
  • taking on adult skills funding
  • further discussion over approaches to social housing.

 The 2016 Autumn Statement further announced devolution of the budget for the forthcoming national Work and Health Programme and the beginning of talks on future transport funding in Greater Manchester.

West Midlands

The West Midlands mayor will represent a population of over 2.8 million people, compared to the average MP parliamentary constituency of under 96,000 people – almost 30 times as significant. The powers of the elected mayor are not yet proportionately significant.(see https://westmidlandscombinedauthority.org.uk/media/1572/adocpackpublicversion0001.pdf)

The West Midlands mayor will have limited independent powers, mostly relying on building consensus with local council leaders.

The constitution of the WMCA was approved on 10th June 2016 and published here:

https://westmidlandscombinedauthority.org.uk/media/1171/ca-draft-constitution-24-5-16.pdf

The constitution suggests that “any matters that are to be decided by the Combined Authority are to be decided by consensus of the Members where possible”. Where consensus is not achieved, each Member is to have one vote and no Member including the Chair is to have a casting vote.

Usually votes will require a two-third majority of constituent members, however several areas required a unanimous vote of all members, including:

  • approval of land use plans;
  • financial matters which may have significant implications on Constituent Authorities’ budgets;
  • agreement of functions conferred to the Combined Authority;
  • use of general power of competence within the Local Democracy Economic Development and Construction Act 2009, including in relation to spatial strategy, housing numbers and the exercise of any compulsory purchase powers;
  • approval to seek such other powers
  • changes to transport matters undertaken by the Combined Authority.

Non-constituent members will be able to vote on defined issues (where a simple majority is required) including around:

  • adoption of growth plan and investment strategy and allocation of funding by the Combined Authority
  • the super Strategic Economic Plan strategy along with its implementation plans and associated investment activity
  • the grant of further powers from central government and/or local public bodies that impacts on the area of a Non Constituent Authority
  • land and/or spatial activity undertaken by the Combined Authority within the area of a Non-Constituent Authority
  • Public Service reform which affects the areas of Non-Constituent Authorities
  • all Combined Authority matters concerned with education, employment and skills, enterprise and business support, access to finance, inward investment, business regulation, innovation, transport, environmental sustainability, housing, economic intelligence, digital connectivity and regeneration
  • future use of business rate retention funding generated beyond that retained within new and existing Enterprise Zones

The WMCA “cabinet” (council leaders) will examine the Mayor’s draft annual budget and the plans, policies and strategies, as determined by the Mayoral WMCA, and will be able to reject them if two-thirds of the Mayoral WMCA Cabinet agree to do so. In the event that the Mayoral WMCA reject the proposed budget then the Mayoral WMCA shall propose an alternative budget for acceptance by the Cabinet, subject to a two-thirds majority of those present and voting. The Mayor shall not be entitled to vote on the alternative Mayoral WMCA proposed budget. In terms of specific functions:

  • “Mayoral functions” will be devolved to the Mayoral WMCA by central government, exercised by the Mayor and subject to the provisions in the Scheme.
  • “Mayoral WMCA/Mayoral joint functions” are subject to the Mayor’s vote being included in the majority in favour with the two-thirds of the Constituent Members voting.
  • Mayoral “WMCA functions” are not subject to the Mayor’s vote being included in the majority in favour with the two-thirds of the Constituent Member voting. The items reserved for unanimous voting of the Constituent Members are also not subject to the Mayor’s vote in favour.

The functions which are proposed to be “Mayoral functions” are:

  • HCA CPO powers (with the consent of the appropriate authority(ies)
  • Grants to Bus Service Operators
  • Devolved, consolidated transport budget
  • Reporting on the Key Route Network (in consultation with the authorities)
  • Mayoral precept
  • Raising of a business rate supplement (in agreement with the relevant LEP Board(s) and the Mayoral WMCA)
  • Functional power of competence (but no general power of competence).

 

4       How do mayors fit with the wider devolution agenda?

The Government’s approach to devolution has been to negotiate the transfer of powers through a series of “devolution deals” or agreements. The House of Commons Communities and Local Government Committee concluded that “the Government’s approach to devolution in practice has lacked rigour as to process: there are no clear, measurable objectives for devolution, the timetable is rushed and efforts are not being made to inject openness or transparency into the deal negotiations” (CLG Committee, 2016).

The 2015 devolution agreements are a development of a series of “city deals” between 2011 and 2015; first with the eight core cities and later with 20 smaller cities and city regions.

The devolution deals agreed so far have many similarities in terms of powers to be devolved (Sandford, 2016). The core powers devolved include the following:

  • Restructuring the further education system.   Some areas will also take on the Apprenticeship Grant for Employers.
  • Business support. In most areas, local and central business support services will be united in a ‘growth hub’.
  • The Work Programme. This was the Government’s main welfare-to-work programme, subsequently replaced by a much smaller Work and Health programme. Many areas are to jointly develop a programme for ‘harder-to-help’ benefit claimants.
  • EU structural funds. A number of areas are to become ‘intermediate bodies’, which means that they, instead of the Government, will be able to take decisions about which public and private bodies to give EU structural funds to. The future of these funds is of course in doubt following the EU referendum.
  • Fiscal powers. Many deals include an investment fund, often of £30 million per year. Elected mayors will have the power to add a supplement of up to 2p on business rates, with the agreement of the relevant Local Enterprise Partnership.
  • Integrated transport systems. Many deals include the power to introduce bus franchising, which would allow local areas to determine their bus route networks and to let franchises to private bus companies for operating services on those networks. Each deal also includes a unified multi-year transport investment budget.
  • Planning and land use. Many deals include the power to create a spatial plan for the area.

Further details are provided in the Annex to this paper.

 

5       How well have other elected mayors performed?

Mayors in England have had a mixed picture of performance. In Stoke and Doncaster they did not deliver improvement, but in some areas they are linked to significant progress. The Warwick Commission concluded “our evidence suggests that elected mayors offer a real opportunity for change in a place where change is needed and also a way of invigorating a body politic”.

Gains (2015) concludes that “the evidence base for improved performance under mayoral governance is weak”. However, reviewing evidence on the introduction of the first city mayors she notes that “compared to areas operating a leader/cabinet model where the leader was indirectly elected, respondents to surveys of councillors, officer and local stakeholders in mayoral authorities agreed more strongly that there was quicker decision-making, that the mayor had a higher public profile, that decision-making was more transparent, that the council was better at dealing with cross cutting issues that relationships with partners improved and disagreed more strongly with the statement that political parties dominated decision-making”.

The Bristol Civic Leadership Project has explored the question “What difference does a directly-elected mayor make?” since September 2012. An early analysis published in 2014 identified that the Mayor had enjoyed access to central government ministers, that he had emphasised leading the city rather than the council, and that he was a more prominent public figure in Bristol city life than any previous leader.

The project’s final report in Sept 2015 (Hambleton and Sweeting, 2015) concluded that there has been a changed perception of governance in Bristol, in particular:

  • Many perceive an improvement in the leadership of the city, in areas such as the visibility of leadership, there being a vision for the city, the representation of Bristol, and leadership being more influential than previously was the case.
  • However, there are areas where the model is seen as performing inadequately. There are concerns about the levels of representation of views within the city, trust in the system of decision-making, and the timeliness of decision-making.
  • Frequently there are considerable differences of view about the mayoral model of governance from those situated in the different realms of civic leadership in Bristol. Councillors tend to display considerably more negative views about the impacts and performance of the new model compared to those in public managerial, professional, community and business realms.
  • Members of the public in different parts of Bristol tend to think somewhat differently about the impacts of the reform. Often, but not universally, those people living in better off parts of Bristol are inclined to see the move to, and the impacts of, the mayoral model more positively than those living in less well off parts of Bristol.

Assessments of the impact of the London Mayor are complicated by the evolving powers linked to this role. The initial model was largely restricted to transport, and led to the successful introduction of the congestion charge and cycling initiatives. The subsequent successful bid for the London Olympics 2012 perhaps demonstrates the wider “power” of the role.

Analysis suggests that leadership turnover in places with mayors is 50% lower than those with council leaders (Warwick Commission, 2012:29).

6       What issues remain to be resolved?

 

6.1     Scrutiny, checks and balances

The Warwick Commission argued that the relationship between mayor and full council needs to be constructed so the mayor is visibly held to account, yet their mandate should not be undermined by a body which has been separately elected. There needs to be an appropriate recall process which enables the removal of an elected mayor in office in extremis.

Gains (2015) argues that democratic considerations initially received insufficient attention in Greater Manchester. These relied on the cabinet of local council leaders provided strong veto powers and the four-yearly direct election of the mayor. However, she points out that the potential for wider and innovative public engagement and effective formal scrutiny were not fully explored initially. The latter could not rely on the cabinet because “their executive role precludes the kind of independent scrutiny expected elsewhere in local government”. She points out the more active public engagement and transparency arrangements are now being developed in GM.

The CLG Select Committee review of Devolution Agreements found “a significant lack of public consultation and engagement at all stages in the devolution process” (CLG Select Committee, 2016)

The Cities and Local Government Devolution Bill 2016 sets out key requirements for overview and scrutiny arrangements. Each combined authority will be required to establish at least one overview and scrutiny committee, consisting of backbench councillors from the constituent councils, to review and scrutinise its decisions and actions and those of the elected mayor.

Alternative models have been suggested through local decisions on a clear governance framework (Centre for Public Scrutiny) or introducing “second chambers” of people from the business, voluntary and community sectors and citizens’ panels (Institute for Public Policy Research North).

6.2     Public engagement and consultation

 A number of criticisms have been made of the lack of public consultation in most devolution negotiations. The House of Commons Local Government Select Committee found “a significant lack of public consultation and engagement at all stages of the devolution process” (CLG Committee, 2016).
There have been some examples of innovative engagement, for example the University of Sheffield and the Electoral Reform Society, with other partners, held two “citizens’ assemblies” in autumn 2015, in Sheffield and Southampton. Over two weekends, invited members of the public discussed devolution options in their local areas. Details of the assemblies and the outcomes of the public discussions can be found at http://citizensassembly.co.uk/. Similarly, Coventry held a one-day citizens’ panel on 9 September 2015, discussing whether the city should participate in the West Midlands combined authority. (Sandford, 2016).

6.3     Mayoral Powers

The Warwick Commission stressed that “the difference between ‘powers’ and ‘power’ is critical in discussing elected mayors. Whilst the debate about clarity over which powers (and budgets) Whitehall will hand to cities with directly elected mayors will continue, it is also important to recognise the soft and invisible power that has often been accumulated by elected mayors that sits outside their statutory remits has been considerable. In many cases, it has led to the granting of more powers” (Warwick Commission, 2012:8).

That said, they argue that “Mayors should examine the totality of the public spend in a place and hold bodies over which they do not have budgetary control to public account in a wider sense, e.g. the combined impact of social care, recidivism amongst low level offenders, impact of welfare and work and training”.

In terms of the national legislative framework, many powers are now available to elected Mayors. The list in Table 1 (below) is taken from NLGN’s publication “New Model Mayors: Democracy, Devolution and Direction” (2010) updated for powers subsequently provided to elected mayors.

 

 

Table 1: Comparison of current mayoral powers with NLGN proposals

 

NLGN Proposal Current position
The financial flexibility to balance budget over the 3 final years of a term, instead of being limited by in-year balancing No
The creation of a single capital investment pot for the area, so that all relevant monies are pooled and control over spend maintained by the mayor Yes?
The power to introduce a supplementary business rate of up to + or – 4p, with any extra funds raised to be spent on economic development within the locality as deemed best by the mayor Partly – currently limited to 2p and subject to agreement with the local business-led LEP.
Permission to use TIF mechanism through the establishment of an ADZ Yes (through New Development Deals)
Ability for mayor to appoint or dismiss Chief Executive, giving the council an advisory role but the final decision to rest with the Mayor No
Similar transport powers to those that the Mayor of London currently enjoys, in particular to have a say in local transport provision within the authority’s boundaries through chairing (or the nomination of chair) of the local transport body Yes?
The introduction of a new post of Police Commissioner, with the Mayor taking up this position or appointing a councillor to this position Yes
The power of appointment for the position of PCT Chief Executive and in addition power to nominate one person to sit as a non-executive member on the board of the PCT No
Alignment of PCT priorities with local Mayoral health priorities GM only
Responsibility, powers and funding for 14-19 and adult skills Yes
The formation of a statutory Employment and Skills Board, chaired by the Mayor or a representative of the Mayor, to devise strategy Yes?
Fast-tracked to a devolved commissioning model for welfare-to-work provision No – DWP resist devolved commissioning but promise to engage with local areas.
A seat in the second chamber of the Houses of Parliament No

 

6.4     Gender balance

Recent research by the Fawcett Society (Trenow and Olchawski, 2016) concludes that the current approach to devolution “risks handing power to male-dominated structures and shutting women out of the decision making process”.

Their analysis shows that for the Northern Powerhouse area (NP in the chart), 40% of councillors are women, rising to 50% in Manchester City Council. In this respect they outperform Westminster, where only 29% of MPs are women, and Police and Crime Commissioners (16% women).

Diagram 1: Representation of women

Screen Shot 2017-04-04 at 08.55.53

However, the proportion of women falls significantly when considering senior positions in the Northern Powerhouse. For these roles the figures are: to 28% of senior leadership roles and 14% of chairs of established and proposed combined authorities. More generally, so far only four out of 16 existing directly elected mayors in England are women.

 

 

 

References

 

CLG Select Committee, Devolution: the next five years and beyond, First Report of Session 2015–16, January 2016.

Gains, Francesca. “Metro mayors: devolution, democracy and the importance of getting the ‘Devo Manc’ design right.” Representation 51.4 (2015): 425-437.

Hambleton, Robin, and David Sweeting. “Innovation in urban political leadership. Reflections on the introduction of a directly-elected mayor in Bristol, UK.” Public Money & Management 34.5 (2014): 315-322.

Hambleton, Robin, and David Sweeting. “The impacts of mayoral governance in Bristol.” (2015).

Osborne, “Chancellor on building a Northern Powerhouse”, HM Treasury and The Rt Hon George Osborne MP, 14 May 2015

Sandford, Devolution to local government in England, House of Commons Library briefing paper number 07029, 5 April 2016

Svara, James H. Official leadership in the city: Patterns of conflict and cooperation. Oxford University Press on Demand, 1990.

Trenow, Polly and Jemima Olchawski, The Northern Powerhouse: an analysis of women’s representation, Fawcett Society, 2016

Warwick Commission. “Elected mayors and city leadership summary report of the Third

Warwick Commission.” Warwick, Warwick University (2012).

 

 

 

As DirectoCatherine Staite 02r of Public Service Reform, Professor Catherine Staite leads the University’s work supporting the transformation and reform of public services, with a particular focus on the West Midlands.  Her role is to help support creative thinking, innovation and improvement in local government and the wider public sector. As a member of INLOGOV, Catherine leads our on-line and blended programmes, Catherine teaches leadership, people management, collaborative strategy and strategic commissioning to Masters’ level.  Her research interests include Combined Authorities, collaboration between local authorities and the skills and capacities which elected members will need to meet the challenges of the future

lowther-jason

Jason Lowther is a senior fellow at INLOGOV. His research focuses on public service reform and the use of “evidence” by public agencies.  Previously he led Birmingham City Council’s corporate strategy function, worked for the Audit Commission as national value for money lead, for HSBC in credit and risk management, and for the Metropolitan Police as an internal management consultant. He tweets as @jasonlowther

 

 

De-reformation of the Local Government System in Turkey?

Saban Akca

Local government in Turkey has a two-headed structure: a centrally appointed governor with a plethora of responsibilities on the one hand and popularly elected municipal authorities on the other. This dualism dates back to the days of the Ottoman Empire, but in this blog I am not going to barge into historical details; rather I will discuss the current situation of the Kurdish municipalities that are now being seized one by one by Mr Erdogan, the Turkish President, and his political party, the AKP (Justice and Development Party), particularly after an infamous coup attempt last July. An upcoming referendum on 16 April this year – which is considered a pivotal move for transforming the country’s democratic parliamentary system to a presidency and creating a powerful sultan-president – will be also taken into consideration.

In the immediate aftermath of the coup, Mr Erdogan and his government started a brutal witch-hunt against all dissenters in Turkey, including liberals, leftists, seculars and followers of the Gulen Movement. Tens of thousands of public servants have been dismissed for allegedly being behind the coup and most of them arrested on not-so-clear charges. Their names have been publicised on the Turkish Official Gazette and almost everybody now treats them as terrorists or traitors. Companies that are deemed to be associated with the Gulen Movement have had their assets seized, been filled with appointed trustees and finally sold. Journalists have been jailed as their newspapers and TV stations have been closed down. The country is now in a constant state of emergency rule, in which everything is done with decrees streaming down from the presidential palace and where all kinds of illegalities committed by authorities are protected.

Municipal authorities in predominantly Kurdish towns and cities have their shares in these atrocities. After sending Kurdish deputies to jails, the government began to seize power in Kurdish local governments, particularly in the east and south-east of the country. Elected mayors have been replaced by appointed trustees, chosen from local governors in districts or provinces. However, removing mayors did not suffice, so they were jailed too.

Under the stretched state of emergency rule, with no visible and effective opposition remaining, Turkey is holding a referendum to turn its current political system into a presidency. This, unsurprisingly, will work best for the benefit of Mr Erdogan, who is trying to establish a de-facto sultanate. However, one of his referendum promises is to increase the number of greater municipal authorities by over 20, from the present 31. Considering that the country has 81 provinces, increasing the number of popularly elected local municipalities and enlarging their responsibilities across the provincial border is a bold move and may transform the local administration system and add more fuel to the discussion of devolution that has emerged since the 1980s.

The President’s referendum promises and his seizing of popularly elected municipal authorities in the country’s Kurdish regions represents a very problematic sphere in Turkish politics in terms of the empowerment of local governments. Increasing the numbers of democratically and locally elected mayors and devolving power to municipal authorities, as well as enlarging their territories and overlapping provincial borders is a powerful move in the case of strengthening local government reforms. However, only time will tell whether Mr Erdogan and his government are only merely tolerating those who he considers to be his political allies in local governments across Turkey. What we can say though is that this move will certainly bring more uncertainty to the country’s already troublesome local government system.

 

Saban Akca is a doctoral researcher in INLOGOV at the University of Birmingham. His research focuses on leadership in public administration, and his doctoral thesis examines the leadership exercises of local-level district governors in Turkey.

Troubled Families: So what can we learn?

Jason Lowther

Over the last five blogs I have looked in some detail at the Troubled Families Programme and in particular its independent evaluation. I’ve argued that the evaluation shows some important impacts from the programme, but has so far missed valuable learning by failing to capture the local angle, covering too short a time horizon, and not designing in a theory-informed experimental approach. This week I want to reflect on four lessons from the experience.

The TFP has delivered real impacts. We know that the TFP has changed how services for these families are delivered. The independent evaluation finds it has mainstreamed “whole-family” approaches, stimulated local multi-agency working, opened up previously impossible data sharing and made employment support more responsive. Families on the programme feel (and told the researchers) that it’s made a big difference to their lives. And the figures local authorities submitted about the changes in families who were classified as “troubled” (out of school, out of work, committing crime, etc) are audited and truthful – they do represent actual changes in people’s circumstances.

The TFP evaluation questions whether these impacts would have occurred in any case, without the TFP itself. But the evaluation was hamstrung by being undertaken too early and for insufficient time, by limited data (for example because academy schools are not required to co-operate on sharing vital information), and by the lack of an experimental and theory-based approach.

So what can we learn from the TFP experience?

First, the TFP isn’t the panacea ministers claimed – trumpeting an incredible 99% success rate whilst delaying publication of the independent evaluation set up the department to face a storm of media criticism. But it has made a big difference: the TFP changed how these services are delivered, the families noticed a significant improvement, and councils have rightly claimed for progress made.

Secondly, the department and evaluators have done a good job at trying to rigorously assess whether the TFP worked better than “business as usual”. Next time, it would be best to build a rigorous experimental approach into the programme design up front – and to develop some testable theories of how the programme is supposed to effect change.

Thirdly, national summaries can only take us so far. The real diamonds of learning are at local level. Departments should fund and support local areas to learn quickly from the natural experiments that happen when different councils adopt and adapt national policy which is based on limited prior knowledge and evidence.

Fourthly, although challenging for politicians with an eye on their ministerial career, pilots need to be given chance to bed-in before being pulled up for evaluation, and evaluation needs to run long enough to know whether we are getting results. Evaluators can learn from past experience and “new” approaches such as theory-based evaluation.

As TFP and other government programmes roll out in future, these four lessons can make sure that we learn and improve outcomes as quickly as possible.

 

lowther-jason

 

Jason Lowther is a senior fellow at INLOGOV. His research focuses on public service reform and the use of “evidence” by public agencies.  Previously he led Birmingham City Council’s corporate strategy function, worked for the Audit Commission as national value for money lead, for HSBC in credit and risk management, and for the Metropolitan Police as an internal management consultant. He tweets as @jasonlowther

Why co-produce? Accounting for diversity in citizens’ motivations to engage in neighbourhood watch schemes.

Carola van Eijk, Trui Steen & Bram Verschuere

In local communities, citizens are more and more involved in the production of public services. To list just a few examples: citizens take care of relatives or friends through informal care, parents help organizing activities at their children’s school, and neighbours help promoting safety and liveability in their community. In all these instances, citizens complement the activities performed by public professionals like nurses, teachers, neighbourhood workers and police officers; this makes it a ‘co-productive’ effort. But why do people want to co-produce? In our recently published article in Local Government Studies we try to answer that question by focusing on one specific case: local community safety. One of the main conclusions is that citizens have different incentives to co-produce public services, and local governments need to be aware of that.

Simultaneous to the international trend to emphasize citizens’ responsibilities in the delivery of public services, there are also concerns about the potential of co-production to increase the quality and democratization of public service delivery. One important question pertains to who is included and excluded in co-production processes. Not all stakeholders might be willing or feel capable to participate. So, acknowledging the added value of citizens’ efforts and the societal need to increase the potential benefits of co-production, it is important to better understand the motivations and incentives of citizens to co-produce public services. A better insight not only can help local governments to keep those citizens who are already involved motivated, but also to find the right incentives to inspire others to get involved. Yet, despite this relevance, the current co-production literature has no clear-cut answer as the issue of citizens’ motivations to co-produce only recently came to the fore.

In our study, we focus on citizens’ engagement in co-production activities in the domain of safety, more specifically though neighbourhood watch schemes in the Netherlands and Belgium. Members of neighbourhood watch teams keep an eye on their neighbourhood. Often they gather information via citizen patrols on the streets, and report their findings to the police and municipal organization. Their signalling includes issues such as streetlamps not functioning, paving stones being broken, or antisocial behaviour. Furthermore, neighbourhood watch teams often draw attention to windows being open or back doors not being closed. Through the neighbourhood watch scheme, the local government and police thus collaborate to increase social control, stimulate prevention, and increase safety.

The opinions of citizens in co-producing these activities and their motivations for getting engaged in neighbourhood watch schemes are investigated using a ‘Q-methodology’ approach. This research method is especially suitable to study how people think about a certain topic. We asked a total of 64 respondents (30 in Belgium and 34 in the Netherlands) to rank a set of statements from totally disagreement to fully agreement.

Based on the rankings, we were able to identify different groups of co-producers. Each of the groups shares a specific viewpoint on their engagement, emphasizing for example more community-focused motivations or a professional attitude in the collaboration with both police and local government. To illustrate, in Belgium one of the groups identified are ‘protective rationalists’, who join the neighbourhood watch team to increase their own personal safety or the safety of their neighbourhood, but also weigh the rewards (in terms of safety) and costs (in terms of time and efforts). In Netherlands, to give another example, among the groups identified we found ‘normative partners’. These co-producers are convinced their investments help protect the common interest and that simply walking around the neighbourhood brings many results. Furthermore, they highly value partnerships with the police: they do not want to take over police’s tasks but argue they cannot function without the police also being involved.

The study shows that citizens being involved in the co-production of safety through neighbourhood watch schemes cannot be perceived as being similar to each other. Rather, different groups of co-producers can be identified, each of these reflecting a different combination of motivations and ideas. As such, the question addressed above concerning why people co-produce cannot be simply answered: the engagement of citizens to co-produce seems to be triggered by a combination of factors. Local governments that expect citizens to do part of the job previously done by professional organisations need to be aware of the incentives people have to co-produce public services. Their policies and communication strategies need to allow for diversity. For example, people who co-produce from a normative perspective might feel misunderstood when compulsory elements are integrated, while people who perceive their engagement as a professional task might be motivated by the provision of extensive feedback.

 

Foto Carola %28bijgesneden%29.jpgCarola van Eijk holds a position as a PhD-candidate at the Institute of Public Administration at Leiden University. In her research, she focusses on the interaction of both professionals and citizens in processes of co-production. In addition, her research interests include citizen participation at the local level, and crises (particularly blame games).

 

trui02a

Trui Steen is Professor ‘Public Governance and Coproduction of Public Services’  at KU Leuven Public Governance Institute. She  is interested in the governance of public tasks and the role of public service professionals therein. Her research includes diverse topics, such as professionalism, public service motivation, professional-citizen co-production of public services, central-local government relations, and public sector innovation

 

Bram Verschuere 2.jpg

Bram Verschuere is Associate Professor at Ghent University. His research interests include public policy, public administration, coproduction, civil society and welfare policy. 

Troubled Families: How experimenting could teach us “what works?”

Jason Lowther

 

In this blog on 3rd Feb, I explored the formal Troubled Families Programme (TFP) evaluation and looked at the lessons we can learn in terms of the timing and data quality issues involved. This week I want to consider how designing a more experimental approach into this and future programmes could yield lots more insight into what works where.

The idea of an “experimental” approach to policy and practice echoes enlightenment period thinkers such as Francis Bacon (1561—1626), who promoted an empirical system built on careful experimentation. Donald Campbell’s ideas[1] on ‘reforms as experiments’ argued that social reforms should be routinely linked to rigorous experimental evaluation. ‘Social engineering’ built on ‘social experiments’ became a popular concept in the USA and social science.

Social experiments in America included work in response to a concern that providing even modest income subsidies to the poor would reduce motivation to find and keep jobs. Rossi and Lyall (1976) showed that work disincentives were in fact less than anticipated. In the field of prison rehabilitation, Langley et al. (1972) tested whether group therapy reduced re-offending rates. The results suggested that this approach to group therapy did not affect re-offending rates.

Unfortunately, meaningful experiments proved more difficult than anticipated to deliver in the field, and even robust experiments were often ignored by policy makers. As a result, until recently this experimental approach fell out of favour in social policy, except in the field of medicine.

The term ‘evidence-based medicine’ appears to have been first used by investigators from a US university in the 1990s where it was defined as ‘a systemic approach to analyze published research as the basis of clinical decision making.’ The evidence-based medicine movement considered experiments – specifically, collections of Randomised Controlled Trials (RCTs) subject to systematic reviews – as the “gold standard” of proof of whether interventions “work” or not.

Randomised controlled trials are sometimes not easy to undertake in social policy environments, but they can be done and they can provide surprising results. Starting in 2007, Birmingham City Council evaluated three evidence-based programmes in regular children’s services systems using RCTs[2]. We found that one programme (Incredible Years) yielded reductions in negative parenting behaviours among parents, reductions in child behaviour problems, and improvements in children’s relationships; whereas another (Triple-P) had no significant effects.

What was interesting for practitioners was that the children in all the trials had experienced improvements in their conduct. Only by use of a formal “control” group were we able to see that these “untreated” children were also improving, and so we were able to separate out the additional impacts of the intervention programmes.

There are a number of lessons from this and other past experience that can help practitioners wanting to deliver robust trials to test whether innovations are working (or not). The most important point is: build the evaluation testing into the design of the programme. The Troubled Families Programme could have built an RCT into the rollout of the programme – for example, selecting first year cases randomly from the list of families who were identified as eligible for the scheme. Or introducing the scheme in some council areas a year earlier than others. Or councils could have done this themselves by gradually rolling out the approach in different area teams.

Sandra Nutley and Peter Homel’s review[3] of the New Labour government’s Crime Reduction Programme stressed the importance of balancing the tensions between fidelity to “evidence based” policy (to maximise the chance of impact) and innovation (to ensure relevance to the local context), short-term wins and long-term learning, and evaluator independence (to ensure rigour) versus engagement (to help delivery).

In my final blog on the TFP next time, I explore the potential for “theory-based” approaches to evaluation helping us to understand “what works and why?” in this and other policy areas.

Campbell, D. T. and Russo, M. J. (1999) Social experimentation. Sage Publications, Inc.

Langley, M., Kassebaum, G., Ward, D. A. and Wilner, D. M. 1972. Prison Treatment and Parole Survival. JSTOR.

Nutley, S. and Homel, P. (2006) ‘Delivering evidence-based policy and practice: Lessons from the implementation of the UK Crime Reduction Programme’, Evidence & Policy: A Journal of Research, Debate and Practice, 2(1), pp. 5-26.

Rossi, P. H. and Lyall, K. (1976) ‘Reforming public welfare’, New York: Russell Sage.

Sanderson, I. (2002) ‘Evaluation, policy learning and evidence‐based policy making’, Public administration, 80(1), pp. 1-22.

White, M. (1999) ‘Evaluating the effectiveness of welfare-to-work: learning from cross-national evidence’, Evaluating Welfare to Work. Report, 67.

[1] Campbell, D. T. and Russo, M. J. (1999) Social experimentation. Sage Publications, Inc.

[2] Little, Michael, et al. “The impact of three evidence-based programmes delivered in public systems in Birmingham, UK.” International Journal of Conflict and Violence (IJCV) 6.2 (2012): 260-272.

[3] Nutley, S. and Homel, P. (2006) ‘Delivering evidence-based policy and practice: Lessons from the implementation of the UK Crime Reduction Programme’, Evidence & Policy: A Journal of Research, Debate and Practice, 2(1), pp. 5-26.

 

 

lowther-jason

 

Jason Lowther is a senior fellow at INLOGOV. His research focuses on public service reform and the use of “evidence” by public agencies.  Previously he led Birmingham City Council’s corporate strategy function, worked for the Audit Commission as national value for money lead, for HSBC in credit and risk management, and for the Metropolitan Police as an internal management consultant. He tweets as @jasonlowther

Councils going bankrupt? Don’t be Scilly!

Chris Game

 

Remember the old London bus joke: you wait for ages, then three come along at once? Well, some local government finance anoraks have been waiting ages for a 114, and now two 24s arrive, almost in mini-convoy.

Not buses, of course – but key sections of Acts of Parliament. Section 114 of the Local Government Finance Act 1988 requires a council’s Chief Finance Officer (CFO) to issue a s114 Notice reporting to all elected members an actual or impending seriously unbalanced budget.

As indicated, they’re infrequent, issued only in what CIPFA terms (p.3) “the gravest of circumstances” – current spending way beyond budget, reserves virtually exhausted, no imminent solution. Their impact too is grave – effectively freezing spending until councillors agree measures to achieve a balanced budget. But … they do keep the crisis in-house.

The alternative – being required to wash your dirty linen in public – is the dreaded Section 24 of the Local Audit and Accountability Act 2014. Here, a council’s external auditors append a written s24 ‘Recommendation’ to their Annual Audit letter, “copied to the Secretary of State”. The recommending and copying may sound chummy, but it’s the bullet-shaped chumminess of a Mafia ‘message job’. S24 Notices are very nasty, and happily very rare – or were.

Recently the Isles of Scilly Council received one, which is interesting in itself, since not everyone’s clear what it actually is: not part of the already enormous Cornwall Council, but a sui generis unitary authority – in a class of its own.

It’s considerably smaller than my own authority, Birmingham City Council (bear with me, it’s not that daft a comparison), with one-480th of its population – though, interestingly, one councillor for every 109 residents, compared to Birmingham’s post-2018 ration of one for every 10,900. And, as a genuine unitary, it has Birmingham’s major functions, including an airport, plus fire services, water supply, sea fisheries and coastal defence.

Unfortunately, Scilly’s unique status doesn’t exempt it from the austerity pressures and grant cuts faced by all English councils. It’s suffering badly, and its external auditors concluded that – with £3 million needed to pay staff and suppliers, no council tax income for the last two months of the financial year, and reserves down to £0.5 million – the law required Section 24, ‘recommending’ in terms (p.37) that the Council get its whole financial act together, extremely pronto.

Its impact, warning notwithstanding, can be surmised from the reaction of the eminent former Labour leader of the previous council to have received a s24 missive, just last November from coincidentally the same auditors, Grant Thornton. “The most concerning audit letter I have seen in all my [36] years on the council” was Cllr Sir Albert Bore’s verdict – the council being Birmingham, and the equivalent budget black hole not £3 million but pushing £38 million.

Just as Aristotle’s single swallow did not a summer make, two s24s don’t themselves make a systemic winter crisis. They’re surely, though, a sign – given that not one such report was issued to any council during the whole four-year 2010 spending review period, and we’ve now had two in two months.

But a sign of what? That’s rhetorical. I’m categorically not a local finance expert, and in this blog’s limited remaining space there are no answers – just three observations.

  1. Why no s114s?

 In the past we’d see them at least occasionally. Now there are rumoured sightings – e.g. in Northamptonshire – yet what have materialised are the two s24s, the proverbial nuclear option. It’s been suggested that the (post-1988 Local Finance Act) statutory duties of Social Services Directors mean a s114-prompted total spending freeze could prevent, say, a vulnerable child being placed in care. But CIPFA Chief Executive Rob Whiteman has rejected this interpretation.

  1. Please, not the B-word

There’s unfortunately no way of avoiding the media headline, but Scilly, Birmingham and probably any financially struggling English councils aren’t about to ‘go bankrupt’ in the sense in which the word is commonly understood. UK councils, unlike US local government or our own national government, are statutorily required to set each year a balanced budget. Running a deficit of the smallest fraction of Detroit’s nearly $400 million in 2013 is simply not possible – indeed, a very s(c)illy idea.

3. But watch for those ‘statutory duties’

So not bankruptcy in the normal legal sense, but almost daily signs and public warnings – Conservative Surrey’s seriously contemplated 15% council tax referendum, tax hikes all round of approaching 5%, hitherto protected adult care budgets now being cut and 13% of responding councils in the MJ/LGIU 2017 State of Local Government Finance Survey –reporting “a danger they would no longer have enough funding to fulfil their statutory duties in the coming year”.

Which would mean facing legal challenge for failing to meet those statutory duties and/or declaring ‘technical insolvency’. Not ‘bankruptcy’, note; but, as the famous duck test puts it: if it looks like, walks like, and quacks like a duck ….

 

Chris Game - pic

Chris Game is a Visiting Lecturer at INLOGOV interested in the politics of local government; local elections, electoral reform and other electoral behaviour; party politics; political leadership and management; member-officer relations; central-local relations; use of consumer and opinion research in local government; the modernisation agenda and the implementation of executive local government.