Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

This site uses cookies. By using our services, you agree to our cookie use.
Learn more here.

The Architecture of Governance: Lecture transcript

James Stirling Memorial Lecture on the City. Harvard Law School Professor Gerald Frug on The Structure of Democracy

The Architecture of Governance

Part 2: The Structure of Democracy

I want to talk tonight about democracy.  In the UK and the United States, the democratic nature of our political system is often treated as unchallengeable.  I take a very different view.  I think that people are losing faith in elected democratic government, even at the local level.  Tonight, I want to describe and criticize some of the current ways in which elected city government is being undermined.  In this part of my talk, I will focus on the Localism Bill now under consideration by this country’s Parliament – a bill which is being trumpeted as increasing, not decreasing, local power.  In the second part of my talk, I will describe an alternative conception of the decentralization of power, one very different from the one embraced by the Localism Bill.  This alternative is enshrined in the South African Constitution.  The South African conception, I will argue, is a much better way to think about localism than the one offered by the Localism Bill. 

This is the second of two talks I am giving as the recipient of the Sterling Prize.   The first was in Montreal last October.  In Montreal, I focused on the ways in which local government decision making has been fragmented by, for example, allocating power to non-democratic forms such as quangos and public-private partnerships instead of elected government. Tonight’s talk will be on a different topic.  I want to focus here on elected democratic government itself.  As in Montreal, I start with the assertion that every form of urban governance has a structure.  By this I mean that there are rules that determine who has the power to decide how to deal with the problems cities face – whether the problems concern the environment, transportation, education, crime, housing or any other issue.  This governance structure is designed by people you can name – by Parliament, here in the UK, and by state legislatures in the United States.  I have entitled these lectures the architecture of governance in order to emphasize that this design is like an architectural design.  Every government structure embodies ideas that can be described, critiqued, and re-designed to make it work better.  These ideas are man-made; they are part of our built environment.  Like the physical construction of buildings and landscape, the design of the governance structure influences the way cities develop.  As a result, they shape the lives of all of us.  

The architecture of democratic governance in the UK, the United States, and in much of the rest of world, is built on a division of power among different levels of government.   National governments, state governments, regional governments, city governments, borough governments, and neighborhood governments can all be organized democratically.  Everywhere, more than one of these levels of government – in some places, all of them –make critical public policy decisions.  The result of this splintering of decision making authority has been bureaucratic delay, conflict, gaps in coverage, duplication, and –all too often — paralysis.   It’s therefore not surprising that there is a loss of faith in the ability of democratic government to make anything happen –that pessimism often turns out to be accurate. 

The key question is: what should we do about this problem?  Both the Localism Bill and the South African Constitution propose answers to this question.  Neither one embraces the centralization of power as the answer.  You’ve all heard people say that to make progress we need to be more like China or Singapore.  China can get things done, people say.  Here everything takes forever – or is impossible.  Both the Localism Bill and the South African Constitution rightly reject this appeal to centralized control: the history of unchecked centralized power in this world is not a happy one.  To reform democratic government, both the Localism Bill and the South African Constitution adopt a different course: both aim to strengthen local decision making.  To help you understand the architecture of governance, I will concentrate in this talk on the differences in the way they do so.  As in Montreal, I intend to discuss these alternative ideas at a particular level of generality –one that parallels Kevin Lynch’s book Good City Form.  I am not here to propose an ideal government structure for this city, let alone for every city in the world.  There is no one structure for democratic governance that works everywhere – just as there is no single good city form.  Still, as Lynch suggested about good city form, there are some general ideas about city governance that are worth exploring everywhere.  That’s what I plan to offer here tonight.

Let’s turn first to the Localism Bill.  I am going to criticize this bill even though, since I’m an outsider, you might think that I have no business doing so.  It’s true I might make mistakes in what I’m about to say.  But that’s OK; you’re here to correct me.  I am going to talk about the Localism Bill, despite the risks, because it so nicely captures what seems to be a general consensus about how to decentralize power –a consensus that I think builds the wrong structure for democratic governance.  I don’t see myself as taking a political position in the debates in this country.  When in power, the Labor Government embraced a lot of the conceptual framework of the current Localism Bill, albeit in a somewhat different form.  The key ingredients of this conception can be simply stated.  To decentralize power, no change whatsoever needs to be made to the organization of the national government or to its unlimited power to overrule local decision making.  All the national government needs to do is to enable more local government discretion, redesign the local government structure, and empower not just local governments but communities and neighborhoods to take more control of local services and decisions.  Once this is done, the sharp distinction between the central government and local governments –and the nature of the relationship between them –can continue without modification.  I’m going to describe the Localism Bill’s version of this architecture of governance in some detail.  You should know, however, that I’m not going to discuss the whole bill.  It’s more than 400 pages long.  I am going to focus only on a few of the key provisions that illustrate its way of allocating governmental power.

Chapter One of the Localism Bill gives local authorities –county councils, district councils, London boroughs, and other similar bodies, but, importantly, not the Greater London Authority –a general power of competence.   This provision is thought by both supporters and critics to be one of the most important features of the Bill.  A similar idea was supported by the Labor Government, although the different wording of its version turned out not to have the desired effect.  The proposed general power of competence is a reaction against the traditional notion, adopted in the United States as well as here, that localities have to rely on specific delegations of power from the central government to act on behalf of their constituents.  Giving localities a general power of competence is a way to avoid this requirement of an item-by-item delegation of power.  The Localism Bill describes this general grant of power in striking, and unusual, terms: localities, it provides, are given the power to do anything that individuals may do, and they may do so in any way whatever, anywhere in the UK or elsewhere.   At first blush, this seems like a striking grant of local power – the kind of empowerment that anyone who believes in the decentralization of power would applaud.

Giving localities a general power of competence is not a bad idea.  But there is a good deal less in this proposal than meets the eye.   The general power of competence is similar to what we call in the United States “home rule.”  Every major city in the United States has been granted the power of home rule.  But this grant has given them a lot less authority than people here seem to think.  One reason for this is that the power does not come with any financing to exercise it.  In the UK it’s worse than that, because it comes with a dramatic loss of funding for local governments.  A general grant of power means a lot less if you can’t pay for things you want to do.  And the financing problem derives not just from the diminished amount of support from the central government.  One item clearly excluded from the general power of competence – and also excluded from most grants of home rule in the United States –is the power to raise revenue.  If local authorities want new taxes, they will still have to get a specific delegation from the central government, and that is not likely to happen. 

This exclusion of the power over revenue from the general power of competence is not obvious from the first reading.  There are other limitations that are also not obvious.  In the United States, the home rule grant to cities is limited to local matters.  But most issues of concern to cities are not just local.  They affect the region, the state, and even the nation.  For that reason, a lot of what American cities want to do is not within their home rule power.  The general power of competence does not confine local authorities to local matters.  Instead, the Localism Bill gives localities the power of an individual that can be exercised anywhere in the country.  But local governments are not individuals.  They are not going to marry and have kids.  On the other hand, they may want to regulate the private sector–and individuals can’t do that.  The linking of the power of local authorities to the power of individuals is thus a limitation on local power as well as a grant of local power.  There is going to be a lot of controversy about what this limitation means, just as there is a lot of controversy in the United States over the question of what issues can be considered local. Every grant of power is simultaneously a limitation on that power – a limitation defined by the words of the grant.

Localism advocates think that the answer to these problems is to expand the definition of the general power of competence.  The limitation they particularly focus on is the ability of national government to overrule any local decision.  But no country empowers localities in a way that the central government cannot control.  Remember: the Localism Bill enables local authorities – by which we mean London boroughs as well as local councils elsewhere — to exercise authority anywhere in the country.  There has to be centralized control over this kind of local power or there would be chaos.  For similar reasons, state governments in the United States have the power to overrule most decisions made under home rule.  Moreover, central control means more than just overruling local decisions.  The Localism Bill makes clear, as it must, that any local decision has to be consistent with existing or future legislation enacted by the national government.  Given how much central government legislation there is, local authorities in the UK –just like localities in the United States — will have to be very careful to ensure that what they do is consistent with national power.  This kind of concern has dramatically limited the effect of home rule in the United States. 

The remarkable thing about the UK proposal is not the extent of this centralized control –that exists everywhere – but the way that it is to be exercised.  Under the Localism Bill, the Secretary of State may, by order, prevent a local authority from doing anything he specifies in the order.  This power is lodged in the Secretary of State without any legislative guidance or standards whatsoever.  This kind of centralized control does not exist in the United States; we do not give a Minister this kind of unlimited discretion.  This degree of control suggests that the increase in local power enabled by the general grant of competence is likely to be a lot less than its supporters suggest.

These limits on the general power of competence should be kept in mind when thinking about the Localism Bill’s proposal for directly elected mayors.  Like the Labor Government before it, the Localism Bill is not neutral about whether major cities in this country should have directly elected mayors.  Disappointed by the refusal of most localities so far to adopt the idea, they propose setting up a shadow mayor in a dozen cities before holding a referendum, apparently in the hope that people will get used to the idea.  The basic question is: are directly elected mayors a good idea?  I think it depends on what the idea is.  The Localism Bill limits the voters to two options: having a directly elected mayor or an indirectly elected leader with a cabinet.  What is left out of this list?  One way to begin to see what’s left out is to consider a question the Manchester Evening News asked David Cameron last month.  Why can’t the people of Manchester, he was asked, be able not just to choose one of the two centrally-prescribed options but to decide instead to elect a Mayor of Greater Manchester, creating a position like Boris Johnson now holds for Greater London?  Cameron did not respond to the question; he simply said you don’t have to have a Manchester Mayor if you don’t want one.  He then endorsed the idea of having an elected mayor, saying, and I quote: “I want Manchester to be up there with Beijing and Shanghai as a major world player, and I think having an elected mayor will really help.”  But Beijing and Shanghai don’t have elected mayors.  And their municipal governments control a territory much larger than greater Manchester.  Anyway, it seems odd for David Cameron to suggest that it’s important for cities to have a directly elected mayor in order to have a dynamic leader when he himself, as a leader, is not directly elected.

Still, having a directly elected mayor might be a good idea.  Whether it is depends on a decision about how best to organize the allocation of power within the local authority –how much power should be given to the mayor and how much to a local council or legislature.  There’s no one definition of what having an elected mayor means.  In the United States, some cities have strong mayors and others have weak ones.  It’s possible that the mayor will better embody the wishes of his constituents; it’s possible that representatives of the neighborhoods, in a local legislature, would better represent them.  The division of powers between the legislature and the executive defines what we mean by local democracy.  It might be thought to be the basic local issue.  But not under the Localism Bill.  Local citizens are asked to vote on whether to have an elected mayor without knowing his powers and without being able to decide what they should be.  Under the Bill, the extent of mayoral power is determined by the Secretary of State, again without any standards or guidelines whatsoever.  Indeed, the Secretary of State can enable the mayor to perform any public service function simply by giving an order.  True, once elected, a mayor could propose a list of the powers he should have, but, oddly, only in his first year of office; in any event, the Secretary of State need not adopt the proposal. This scheme is very unlike the way the authority of the Mayor of London was constituted.   His power derives from an Act of Parliament which is quite specific about what the London mayor can and cannot do.  To be sure, as for the proposed new referenda, the people of London voted for a mayoral system in a referendum before knowing what the later Act of Parliament would enable the mayor to do.

The upshot is that it’s hard to tell whether the shift to direct elections would enhance local power without knowing how the Secretary of State is going to decide these issues. Moreover, no matter how powerful the mayor is when compared to the council, he cannot do more than the local authority is allowed to do.  And that leads us back to the limits on the general power of competence. It has been suggested that the Secretary of State might give newly elected mayors powers now exercised by quangos; that would give them more power.  But if it’s a good idea, it might be a good idea everywhere.  And there is no indication that the same kind of transfer of power would be possible to strengthen the authority of an indirectly elected leader with a cabinet.  This is another indication of the Bill’s effort to stack the deck between the options for local organization. 

So far, in describing the Localism Bill’s conception of decentralization, I have concentrated on the power of the Secretary of State, emphasizing the continued centralized control of local governments.  But there is another innovation in the Localism Bill which threatens the power of elected local governments from another angle.  This is the Bill’s extensive reliance on local referenda.  The  use of this device as a way of choosing the nature of the local executive is just the beginning of the story.   Referenda are also prescribed as a vehicle for limiting the authority’s power to raise council taxes, and, more generally, as a mechanism for setting policy on local issues.  To limit taxes, the Localism Bill requires local billing authorities to determine whether a proposed council tax is excessive and, if it is, to submit the excessive tax to the voters for approval or rejection.  Who do you think defines what excessive means under the Bill?  Yes, the Secretary of State. The impact on this referendum procedure is not hard to guess: people vote taxes down.  The revenue problem for the locality, however, may be that it needs more money, not less.  Nothing in the bill allows a referendum to increase the amount of revenue that the central government provides – or stop cutbacks in their services.  And the central government is the primary source of local revenue.  The referendum procedure thus works only one way –it limits the authority of elected government to generate income but it doesn’t help it pay for the services it provides.

The referendum process for local issues is an even more important limit on the power of elected local authorities.  It requires a referendum whenever a valid petition asks for one.  The Localism Bill requires that the petition be signed by 5% of the voters, although the Secretary of State can raise or lower this number.  Once the petition is filed, the authority can refuse to hold the referendum on only four grounds: that it is illegal, that it is not a local issue (as defined by the Secretary of State), that the Secretary of State has determined it should not be held, or that it is “vexatious or abusive.”In other words, the local authority cannot reject the referendum on the grounds that it harms the social and economic well-being of the community or has a negative impact on good governance –grounds the Birmingham City Council has suggested adding. 

The Localism Bill’s expansion of the use of referenda has generally been embraced, by both supporters and critics of the Bill, as an attempt to promote local democracy.  I want to suggest that its primary effect is to limit the power of elected local governments.  Imposing on localities a referenda process is a way that central governments shift power from elected governments to popular votes.  In the United States, we have seen this all too clearly in California.   For thirty years, popular votes have controlled significant parts of the government agenda.  They have, for example, limited the government’s ability to raise revenue and, at the same time, mandated that it spend money for specific public purposes.  It is no surprise that this simultaneous limit on income and mandate of expenses has generated a fiscal crisis.  And many other local government policies, not just fiscal matters, are resolved by referendum in California.  The way I see it, the Localism Bill represents an effort to bring about the Californiazation of the United Kingdom.

Although the referendum has long been heralded as the true expression of democracy, it is a particular kind of democracy.  Referenda, usually written by interest groups, are often badly drafted and hard to implement.  Whether they pass is often affected by the amount of money available to the two sides.  Worse still, in the referendum process, unlike in the legislative process, there is no mechanism that engages every decision maker in a debate about the issue before the vote, no process that allows amendments based on the information generated by the debate, no process that allows negotiation between the two sides, no process that ensures that the issue being decided is put in the context of other demands on the government.  Ordinary people are asked for an up or down vote on a matter of great complexity, without relating the issue to other priorities, even other priorities voted on in the same election.  And they vote on the issue privately, in the isolation of the voting booth, with a secret ballot, without having to account to anyone for why they are voting the way they are.  No doubt, referenda have their virtues – but, as California demonstrates, they should also have their limits.

The Localism Bill also establishes another mechanism that transfers authority away from elected local governments.  This is the community right to challenge.  This provision enables a wide variety of people –a community group, a non-profit, a charity, two or more employees of the local authority, and others that the Secretary of State might specify –to express interest in providing a service that the local authority now provides.  The authority can reject the expression of interest only on grounds specified by the Secretary of State.  If it accepts it, the local authority must initiate a process, controlled by the Secretary of State, designed to transfer the service.  The exact nature of this process has not been specified, but it is bound to include other parties, including profit-making ventures, interested in offering the service.  Like the vote on the council tax, the community right to challenge works in only one direction: it can remove services from elected local authorities but it cannot add them.  Of course, even now, this kind of privatization of local services is well underway both in the United States and in the UK.  The innovation of the Localism Bill is that it takes the privatization decision out of the hands of local authorities and into the hands of a process initiated by petition and designed by the Secretary of State. 

Many supporters of the decentralization of power embrace this mechanism, as they embrace the referendum, as a step toward local empowerment.  But, as is the case of the referendum, I want to add a word of caution: this mechanism undermines the power of elected local governments.  Democratically elected governments certainly have their defects.  But we have an idea of how they can be held accountable.  That idea is called democracy.  We have much less of an idea when we use words like “community”and “local people”as the vehicles for exercising decentralized power. Here’s a quote from the Coalition Document, Building the Big Society:

We want to give citizens, communities, and local government the power and information they need to come together to solve the problems they face … We want society –the families, networks, neighborhoods, and communities that form the fabric of our everyday lives – to be bigger and stronger than ever before.  Only when people and communities are given more power and take more responsibility can we achieve fairness and opportunity for all.

Although local governments are included in this list, they are seen as something different from communities and people and networks.  What is a community?  Who legitimately represents the people?  What’s it mean to empower families or networks in a democratic society?  To whom are any of these groups accountable?  There are no answers to these questions in the Localism Bill.  There may be no answers to these questions at all.  But one thing is clear from the Localism Bill’s embrace of this community right to challenge: elected local government will be smaller.   Don’t get me wrong: I value local organizations, voluntary associations, and families.  But elected democratic government is a valuable human invention too, and we should be very careful before we discard it.

One of the longest parts of the Localism Bill –one I don’t have time to discuss –is the abolition of regional planning bodies and the emphasis instead on neighborhood planning.   Defenders of localism often embrace neighborhoods – like communities and local people — as a source of power.  One can see why they might: isn’t that the right group to decide what the surrounding area looks like?  But concentrating on the immediate neighborhood is also odd.  Neighborhoods can be very parochial.  What’s good for a neighborhood may not be good for the city or region as a whole. Indeed, the most significant limit on the power of neighborhoods and local authorities is the power of other neighborhoods and local authorities –their power to develop in a way that undermines the prosperity of their neighbors.  Local authorities have the power to prefer themselves over their neighbors only because the national or state government has allowed them to engage in this kind of competition.  Regional planning is one mechanism for dealing with the problems generated by this structure.  No doubt, it can be designed badly to be just one more bureaucratic centralized government undertaking.  But it can also be designed to enable neighborhoods and cities to work together rather than against each other. The Localism Bill seeks to recognize the importance of cooperation by imposing a duty on local authorities to cooperate with each other about sustainability –with guidance, of course, provided by the Secretary of State about what cooperation means.  But if the structure of local governance is set up to create a competitive, neighborhood-focused world, how will cooperation occur?  There is a difference between a duty to cooperate and an institution designed and empowered to make cooperation happen.

It’s time to summarize the architecture of governance embraced by the parts of the Localism Bill that I’ve discussed.  The Localism Bill provides for control over local authorities from above: the astounding power, exercised without any statutory criteria, by the Secretary of State.  I’ve given only a few examples of this power; the Local Government Association has found that the Bill contains 142 order and regulation-making provisions giving power to the Secretary of State.  The Localism Bill also provides for control over local authorities from below: the limitations on the elected government imposed by referenda and by communities and other groups seeking to take over local services.  Finally, there is the competition from neighboring jurisdictions, no longer subject to a regional process and limited only by the duty to cooperate.  Faced with these pressures from above, below, and the sides, and faced with a loss of funding and no revenue raising powers of their own, local authorities are likely to see privatization as the most reliable way forward.  Admittedly, local authorities have been armed with the general power of competence, and, perhaps, an elected mayor.  Even so, the Localism Bill does not seem to me the best way to strengthen the power of elected local governments.

I think that we need another way to think about the relationship between local and national governments.  This alternative requires rethinking the nature of the central government as well as local authorities and, at the same time, rethinking the relationship between them. To begin sketching this alternative architecture of governance, I want to quote from the South African Constitution.  Chapter 3, Section 40 of the Constitution provides that, in South Africa,

The “government is constituted as national, provincial and local spheres of government which are distinctive, interdependent, and interrelated.” 

We need to go over this sentence slowly – there’s a lot in it.  The sentence embraces the idea of national, provincial and local governments –but calls them spheres, not levels.  It’s not presented as a hierarchy.  Each has its own role to play, without one being the boss of the others.  This idea of spheres is very different from the traditional conception of government that I’ve discussed.  Under current law, Parliament — or a state legislature in the United States — can delegate power or refuse to delegate power to local governments in any way that it wants.  The South African Constitution opens up the possibility of another way to allocate powers to the different spheres of government.  Having labeled these aspects of government “spheres,”the South African Constitution then says that each of them is “distinctive.”  The word “distinctive”is designed to demonstrate the importance of the difference among the spheres –each of them has its own contribution to make.  This gesture toward distinctiveness does not grant local autonomy.  The very next word in the Constitution, after all, is “interdependent.”  If the spheres are interdependent, none of them can be autonomous.  The idea of interdependence recognizes that each of the spheres always affects what the others do.  Finally, the Constitution says that the spheres are “interrelated”–a word that suggests the need for cooperation among them.  Given their distinctiveness and interdependence, the different spheres of government need to be organized in a way that enables them to work together.  What this means, I think, is that the Constitution envisions managing inter-governmental relations through a political process rather than as a hierarchy or a fixed division of authority.  Now that I have read the provision in the South African Constitution slowly, it might be helpful for me to repeat the whole sentence again.  In South Africa, the Constitution provides, the “government is constituted as national, provincial and local spheres of government which are distinctive, interdependent, and interrelated.”

Ok, you might say, I get the words.  What do they mean?  To answer this question, we need to move from the very abstract language of the South African Constitution to the question of implementation, the puzzle of figuring out how to design an architecture of governance with this idea in mind.  As I have already said, I am going to discuss this puzzle as a general matter.  There are, and there should be, countless ways to design such a structure.  I am also not going to focus on the peculiarities of how the idea is working out in the South African context.  The issues of implementation in South Africa are not the same as those that will be faced elsewhere.  I want to explore the implementation question in a more fundamental way.  I’m going to do so through the words “distinctive,”“interdependent,”and “interrelated.” 

What’s distinctive about local government is its deep connections to people’s lives.  A large part of everyday life is affected by the nature of the area in which people live.  The quality of education, the safety of the streets, the affordability and quality of housing, the availability of clean water, the access to jobs, the vitality of public space –these are critical matters in defining who we are and who we become.  When these kinds of issues go unaddressed, the life chances of local residents are diminished.  If, on the other hand, local governments can demonstrate that they are making progress on these issues, trust in them is likely to increase.  These days, local governments’inability to connect with people on these kinds of matters generate the current demand for popular votes and the transfer of services to communities.  These alternatives are based on the idea that, since government can’t do anything, we have to take matters into our own hands.  I think that, instead, we need to make elected local governments themselves work better.

The distinctiveness of state, regional and national governments lies elsewhere.  It lies in their ability to take a broader view of the same problems that localities face.  Part of this broader view involves including the concerns of neighboring localities, concerns that are often overlooked by people worried about their own city.  Part also comes from the fact that we need expertise in public policy decision making.  Democracy is not simply a matter of public-opinion polls or snap judgments made by people when they cast a ballot.  Experiments in political science have demonstrated that people change their minds about issues when they are exposed to other people’s views and information provided by experts.  Other people know things you don’t.  It’s this quite obvious point that has helped generate the current reliance on quangos. The problem with quangos is that they tend to eliminate the role of ordinary people in public policy decisions.  The problem with referenda is that they tend to eliminate the role of experts.  It’s all too easy to value one of the distinctive ingredients of public policy decision making at the expense of the other. Democratic governance needs to be built on a different foundation.  Experts shouldn’t decide public policy, but neither should uninformed voters.  Democratic government has to bring the two together.  Each is distinctive, and each is necessary.

We reach quickly, then, the idea of interdependence.  We have to organize democratic government in a way that honors both local sentiment and the broader view at the same time.  You may think that I’ve have started to spout a meaningless democratic slogan.  But a recognition of both distinctiveness and interdependence is even now one of the basic theoretical building blocks of how state and national legislatures are constituted.  Central governments are now elected locally around the world.  Why is that? It’s critical that they are.  The reason is that central governments are thought of as ways to bring localities together.  Central governments are mechanisms that allow people who are locally responsible to hammer out agreements notwithstanding their differences.  Although the decision makers are chosen by localities, no locality is able to decide issues simply by itself. The other localities also are in the room –they too are part of the process.  So are experts.  Experts testify, executive officials push their agenda, the private sector lobbies.  This picture of democratic government is not accurately described as being either bottom-up or top-down.  It’s both.  It’s bottom up because the decision makers are local.  But it’s top-down because the decisions are made by all local representatives collectively and can be imposed on dissenting localities without their consent.   We should stop talking about governance as if the design choice were between bottom-up and top-down.  These two concepts are not the opposites of each other.  Both are part of the process. The critical issue is figuring how out to connect the top and bottom, how to deal with their interdependence.  That’s the way nations are formed.

Although the conception of democratic governance I have just outlined is familiar, the problem with the current architecture of democratic governance is that it is no longer working the way I have just described.  Parliaments and state legislatures may be locally elected, but they don’t actually represent localities in the decision making process.  Centralized government has taken on a life of its own, and the localities have become, at best, ineffective lobbyists. Central government decisions are not the product of inter-local negotiation and compromise.  Party discipline has replaced the idea of local representation. It is party discipline that enables the Secretary of State to wield the power conferred on him by the Localism Bill.  Given this structure, localities have sought to defend themselves against the central government, and their defense has been based on the language of local autonomy.  In other words, rather than attempting to reform the organization of the central government so that it better reflects their concerns, they seek to escape from its grasp.  I think that this is the wrong strategy.  As my discussion of the general power of competence should suggest, the goal of local autonomy cannot be achieved; in fact, it’s a bad idea.  But the current form of centralized decision making —  decision making made without genuine local input — is a bad idea too.

Let’s consider an example of how local and central governments can be organized in a better way, in a way that better recognizes their interdependence.  I want to use as my example the currently popular idea, if only in urban planning circles, of regional government.   These days regionalism is usually presented as a form of centralization.  Cities are no longer relevant, advocates insist – it’s the metropolitan area that is real now.  If so, we need to create a regional entity that can make effective decisions about transportation, housing, and jobs, decisions that will disregard the now-outdated city lines.  It’s not surprising that this conception of regionalism has gone nowhere in the United States –and has had difficulty gaining traction in many other countries as well.  Ordinary people in their neighborhoods and cities properly see this as just one more form of centralization –one more level of government, staffed by people who think of themselves as experts, who want to tell people how to lead their daily lives.

But there’s no reason that regionalism has to be defined this way.  I am a proponent of regionalism, but not as a form of centralized control.  I envision regionalism as a mechanism for exercising inter-local power –as a way of empowering the region’s cities, not eliminating them.  Localities currently lack power because they are often, quite properly, thought of as parochial.  Regionalism can allow them to become less parochial if the regional decision making body is organized to enable localities to meet with each other and forge a common policy.  One form this conception of regionalism can take in the United States, I have argued, is the creation of a regional legislature in which the cities themselves are represented.  Since the region’s cities vary widely in population, the larger cities would have more say than smaller ones.  But every city in the region would be in the room – and each would participate in the formulation of public policy.  This structure would not create a form of local autonomy.  Localities would not be able to do anything they want. The regional legislature would have authority over all of its member localities.  But the decision makers would be the localities themselves.  If the localities are in charge, they are likely to want to decentralize power as much as they can.  But they will be able to do so only if each locality is seen not just as distinctive but as interdependent.  To gain allies, localities will have to take into account the impact of decentralization not just on themselves but on others as well.

Why, you might ask, wouldn’t this structure create a form of paralysis of its own?  It might.  It depends on how it is designed.  And it depends on how we seek to retrofit it once it’s built and the leaks in the roof start to become noticeable.  I want localities to learn that the only way to increase local power is to work with their neighbors rather than to work against them.  To increase their power, they need collectively to become the centralized decision making body rather than seeking to protect themselves from it.  The regional legislature has to be designed to change what we mean by central government – and, through its collective decision making process, to change what we mean by local government too.  That’s what distinctive and interdependent mean to me –combining localized concerns with inter-local knowledge and perspective.  I want to be clear.   I am not proposing that this idea be adopted by every city in the world.  I am not even proposing it for every city in the United States.  Even in the United States, the proper form of metropolitan organization will differ from region to region and, in some places, my suggestion will have to be rethought completely.  What I suggesting to you is simply an idea –an architect’s drawing, one filled with problems and ambiguities, offered to generate thought rather than the start of construction. 

I don’t know enough to discuss examples of this kind of structure in this country.  But I do want you to see that the kind of idea that I’ve just proposed can be a useful starting point for the organization of spheres of government everywhere.  City councils and other local legislatures are usually elected by district.  But –as in London – the districts have no meaning whatsoever other than for the purpose of these elections.  Worse still, local legislatures are often given very little power – a strong Mayor is thought to be a better idea.  National governments are even less responsive to their constituent parts.  In recent years, this unresponsiveness has led to increasing calls for secession – the breaking up of countries around the world.  Proponents of secession have embraced autonomy as the answer to national conflict.  But in our inter-connected world, autonomy will not protect them.  They would get more protection, I think, if their effort instead was to change the way national governments operate, to get them to recognize both the distinctiveness and the interdependence of different people, including different people within the same country.

In case you think talking about the nation-state in this way is too romantic, I’d like to tell you a short story.  In mid-2010, the federal government in the United States promulgated nation-wide standards for English and Math from kindergarten through high school.  This was quite remarkable, since education has traditionally been thought of in the United States as the quintessential local function.  Nevertheless, these standards have been widely accepted.  Why is that?  I think that one important reason lies in the way that the standards were formulated.  Rather than being imposed on the states from above, they were generated by the states.  They were negotiated over the course of two years by state officials.  Then, rather than simply being issued as commands from the national government, the procedure for adopting them was to have the standards enacted state-by-state.  More than 40 states have done so.  Not all states agreed to them; some won’t adopt them.  The dangling of federal money was no doubt an encouragement.  Nevertheless, a key to the widespread support for them, I suggest, was organizing national policy by bringing the states together –by uniting the states – and having them work out policy differences themselves.  This story is very unusual in the United States.  It’s not how national policy is ordinarily made.  That’s why it’s important.

I want to end this talk with a few words about the third term in the South African Constitution: interrelated.  Current relations among different levels of government are rarely based on what one might call mutual respect.   Too often, local views are simply disregarded by central governments – or not even solicited.  In return, neighborhood people don’t respect central officials either.  Rage against others is all the rage these days.  At the same time, more and more people dismiss politics as way of solving social problems. To say that a proposal is just politics is to say it is not on the merits. 

Politics, however, has another meaning, one too many have forgotten.  Politics is our mechanism for collective problem solving.   A primary reason that politics is not performing this function, at least in the United States, is that politics has become just another vehicle for expressing rage and distrust.  This attitude has to change for democratic government to work better.  You may think I’m beginning to sound like President Obama.  In fact, I am.  His problem is that he doesn’t have a way to change how politics is practiced in the United States.  Political attitudes are shaped by the way we’ve organized the government – by the architecture of governance.  Changing this architecture will not be easy.  But doing so is a way to build a different kind of politics.  It would be quite a change if the national and state governments actually respected the ideas offered by the local governments.  And it would be quite a change as well if the respect was returned in kind.  I want to quote one last line from the South African Constitution, this time from its Bill of Rights: Everyone, the Constitution says, has inherent dignity and the right to have their dignity respected and protected.  It’s an amazing idea don’t you think?  It’s not the kind of right we usually think we have.  It would be particularly amazing if we applied it not just to each other but to the way we make public policy decisions.

Perhaps the biggest problem facing any reform of the structure of democratic governance is one not mentioned in the South African Constitution.  I’m referring not to the problem of designing a new governance system in any particular context – or even how to accomplish the reform politically.  Big as these problems are, there is one even more serious.  We need to design our governance system in a way that it can be changed when its defects become apparent. As architects know, when designing a building, it is not easy to ensure that it can be retrofitted –many buildings seem to defy being re-used for different purposes.  The only option, it seems, is to tear them down.  Retrofitting is harder still for governance mechanisms.  Many of our current government institutions were designed decades –even centuries — ago.  Everyone realizes that there are problems with how they work now.  And no one thinks that a radical change in the whole system –particularly along the lines I’ve been suggesting –could happen all at once.  The problem is that we do not have a way for re-designing the structure to be a regular, routine part of governance.  So, instead of trying to redesign it, we create a series of additions to an otherwise unquestioned structure.  We create a governance version of sprawl – badly organized, fragmented, dysfunctional.  We thereby feed the lack of confidence in governance and its reform, and, at the same time, we feed the dysfunction that makes reform harder to accomplish.  The only escape from this cycle is to begin –to begin to think about the architecture of governance and, then, step-by-step, government-by-government, work on redesigning it.

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.