David Miliband: The decade of disorder

As he prepares to leave for his new life in America, the former foreign secretary explains why the financial crisis has not created an upsurge of support for the centre left – and why defensive opposition is not enough.

You are invited to read this free preview of the upcoming New Statesman, out on 8 August. To purchase the full magazine - with our signature mix of opinion, longreads and arts coverage, plus columns by Laurie Penny, Will Self, Caroline Criado-Perez, Rafael Behr and John Pilger, as well as our cover features on John F Kennedy and all the usual books and arts coverage - please visit our subscription page

 

 

People sometimes ask why – or lament that – the financial crisis of 2008 has not led to an upsurge of support for the centre left. One answer is that in many countries parties of the centre left were in government, and so got the blame, yet this is insufficient. In truth there was a major market failure and in particular a failure of under-regulated markets. But there was state responsibility, too. Moreover, as Stan Greenberg has written, “a crisis of government legitimacy is a crisis of liberalism”. That is why reforming the state as well as reforming the market is a necessity, not a luxury. Where centre-left parties do get elected, they quickly find that, while the nostrums of what Tony Judt called “defensive social democracy” can strike a chord when a party is in opposition, they are insufficient for government. The welfare state needs defence in the face of unfair attacks. But it also needs reform. Inequality is undoubtedly a modern menace exacerbated by austerity, but macroeconomic policy and even full employment will not tackle its roots. Those at the bottom need a voice, but the middle class is also feeling squeezed. The public sector and its services have been a civilising force in western societies, but they need to operate very differently to meet the twin challenges of changing public expectations and needs and limited public funds.

Francis Fukuyama has laid down the challenge as follows: “It has been several decades since anyone on the left has been able to articulate, first, a coherent analysis of what happens to the structure of advanced societies as they undergo economic change and, second, a realistic agenda that has any hope of protecting a middle-class society.”

We who are on the centre left should take this challenge seriously. There is not an “off-the-shelf” model of social democracy, though there are powerful examples of what can make a positive difference (as well as what doesn’t). The starting point for the modern left has to be to understand the new world.

There are many ways of describing the changing context for what the US columnist David Brooks has called the debate about how to rejuvenate mature industrialised societies. To my eye, there are four key global changes under way that explain why this feels like a decade of disorder. 
 
There is the shift in economic power from west to east (and to some extent from north to south), as outsourcing and what Al Gore calls “robo-sourcing” (that is, technological change) alter the economic prospects of millions of people. As many as 50 million people a year are joining the global middle class in Asia, while norms of employment are being overturned. This has huge domestic consequences in industrialised societies, but is also creating a new power structure in the international system.
 
There is the resource crunch, which predated the credit crunch and is exemplified by a mismatch in the supply and demand for resources that is driving up their prices. Shale gas discoveries in the US are important but they do not invalidate this issue. The oil price remains high and non-oil commodities have risen fast enough in the past decade to eradicate the price cuts of the previous hundred years. As the urban middle-class global population is continuing to grow, it is hard to see this trend going into reverse. 
 
There is the radical extension of the open society to the nooks and crannies of economic and social life, public and personal. This is fuelling democratic revolutions in the undemocratic world and frustration with democracy as conventionally practised in the established democracies. Across the public-private divide, the bar for the legitimate exercise of authority is being raised. This is a good thing.
 
And then there is the pressing nature of interdependence, local and international, that demands new forms of co-operation at precisely the time when people want more power for themselves. The Lehman moment showed that the pre-existing definition of which institutions posed systemic risk was not properly thought through. 
 
It is the interaction of these global changes that has reframed the political equation in western democracies. The welfare state is costing more, but investment in supporting wealth creation constitutes an increasingly urgent claim on public funds. Wages are being squeezed and immigration is causing resentment. There is a crisis of employment and opportunity among young people, but they don’t vote in heavy numbers, so politics is skewed towards the old. The shift to low carbon is a necessity, but there is a collectiveaction problem when global negotiations are stuck. Societies have become more diverse in cultural as well as racial terms, and this has opened up a challenging axis of division in politics, which makes the construction of coherent electoral and governing coalitions more difficult. The putative conflict between cosmopolitans and communitarians has been a real one.
 
In the 1990s it seemed that economics and politics were pushing in the same direction. Globalisation was driving down inflation by producing goods in developing countries at low prices to satisfy western demand; education and training were offering opportunities for greater social mobility; social justice seemed like the counterpart of economic efficiency. The heart of the conundrum facing centre-left parties today is that economics and politics are systematically pulling in opposite directions. The economics calls for more sharing of sovereignty, but the politics is more nationalist or localist. The economics calls for more migration, the politics for less. The economics says switch welfare spending from the old to the young, but the politics of that are horrible. And so on.
 
Meanwhile, politics itself is increasingly seen as broken – and not merely seen as broken but actually broken. This is not just about corruption scandals. It is also about political systems that, for too many people, have lost their capacity to engage and include. Party systems are controlled by small groups (President François Hollande’s exemplary election by open primary excepted). Big money creates a sense of popular disempowerment. Parliaments are gridlocked and enfeebled. For parties of the left, which take pride in the power of politics to check the abuse of power in markets, these are not peripheral concerns. They are the basis for rethinking.
 
It is not that the centre left is wrong to continue to see the importance of Keynesian insights, or wrong to see an important role for government in economy and society, or wrong to defend the poorest, who rely on the welfare state. It is that these articles of faith are insufficient for navigating the rapids of government. 
 
I was once told by a Swedish friend of mine – a successful politician in his own right – that good politics starts with empathy, then proceeds to analysis, then sets out values and establishes the vision, before getting to the nitty-gritty of policy solutions. A lot of politics focuses on the last point, but maybe that is why it fails to connect to a lot of the public. It seems to me that if this is right, three sets of issues, interlocking in nature, will be defining for the future of the centre left.
 
The first is how we respond to the people’s desire for more protection from risks beyond their control – without falling into the trap of big-state solutions that are neither affordable nor deliverable. These can be global risks of an economic, environmental or securityrelated nature. They can be personal risks, such as unemployment, dementia or crime. In each case the old mechanisms for collective insurance against risk through the state are under pressure. 
 
The second is how we meet people’s expectations for more power and control over their own lives – at a time when interdependence and inequality are making this seem like a pipe dream. The sense of disempowerment can be economic, in workplaces that lack the collective power of trade unions and where employee satisfaction and opportunity are limited. It can be social, in communities where crime or neglect sap individual and collective strength. Or it can be in public services, where budgets or decisions are out of the reach of the people who use the services.
 
In truth, the “Third Way” discussions of the 1990s were too weak in addressing issues of economic power, in the workplace and elsewhere. There seemed to be sufficient common ground to facilitate win-win solutions. But the stagnation of wages, among other matters, has shown that this perception was optimistic at best. Economic empowerment needs a harder edge. 
 
The third is how to foster a sense of belonging when people are more mobile and identities more complex. This is partly, but not only, about immigration. It is also about the loneliness of too many elderly people and the lack of security of people of all ages. 
 
Each of these areas of study captures socialdemocratic concerns with equity and mutuality. But they look at them in different ways, through the lens of the way people lead their lives, rather than the overall structure of society. What is more, they challenge the stilldominant paradigm of Anglo-Saxon politics: for or against state or market, for or against public or private sector. They speak much more clearly to the continental European notion of a social-market economy, in which public values and mores define market rules and norms, and in which the state is respectful of private enterprise and personal liberty. Where Labour had ended up in Britain by 2010 – too hands-on with the state and too hands-off with the market – shows the perils of not getting this clear. 
 
So what is the way forward? Away from the tactical battles that dominate day-to-day politics, the lessons from the successes – like that of President Obama – and the struggles, seem to me as follows.
 
The first is that, while fiscal stimulus may be a short-term remedy for low demand, fiscal prudence is a medium-term necessity. Keynesianism is after all a hard taskmaster. 
 
The second is that the politics of production requires social democrats to address the structural challenges to western economies as engines of wealth creation. Sometimes this challenge is reduced to a debate about the importance of “manufacturing”, but that is not enough, as the border between manufacturing and services becomes blurred, and as services take up a much larger share of the economy and require ever higher degrees of skill. 
 
The third is that the tendency of markets towards inequality and instability needs to be countered by action upstream in order to curb abuse of power, and to empower citizens and employees with rights, information and control. “Predistribution” does not fit on an electoral pledge card, but the idea is right and so is the challenge of developing new ways, either through individual rights or collective organisation, to tilt the balance of power towards ordinary people.
 
The fourth is that the state needs to do more with less, and that requires big reform. Reform is one of those words that can lose meaning through excessive and loose use. But for social democrats the specifics of decentralisation of power, transfer of budgets from managers to people, integration of back offices, and challenge from private- and voluntary-sector providers should all be geared towards the goal of a public sphere that can serve its founding purpose: representing the public and its interest against threats of injustice and insecurity. That the Tories will try to use their bungled NHS changes as an excuse for an assault on the very universality of the health service shows how high the stakes are. 
 
The fifth lesson is that fighting globalisation is futile and damaging, but the processes of global integration nevertheless need to be managed by regional co-operation in the absence of stronger global governance. It is striking that, despite the travails of the European Union, around the world there is a tendency towards regional association – in the Pacific Alliance in South America, for example (its GDP is 20 per cent greater than Brazil’s). This is the way for medium-sized states to have influence.
 
The sixth message is that the challenge for parties and politics is as big as the challenge to policy. Assembling a winning coalition is one thing; holding it is harder. That makes the achievement of the Obama campaign quite remarkable. He was helped by his opponent, but he made his own luck through a drive street by street and voter by voter to maintain his winning coalition. Presidential elections are different from parliamentary systems, but there is read-across nonetheless. 
 
Politics is more open than for many years – and there is a lot to fight for. I will be fighting a different set of battles, but I leave Britain confident that Labour can win.
 
David Miliband becomes president and chief executive of International Rescue Committee next month. This essay is an edited version of the foreword to Policy Network’s “Progressive Politics After the Crash: Governing from the Left”(edited by Olaf Cramme, Patrick Diamond and Michael McTernan), which will be published by I B Tauris on 4 September

 

Illustration: Francesco Bongiorni / Eastwing Illustration

David Miliband is the  President and CEO of the International Rescue Committee
He was foreign secretary from 2007 until 2010 and MP for South Shields from 2001 until this year. 

This article first appeared in the 12 August 2013 issue of the New Statesman, What if JFK had lived?

Jamie Squire/Getty Images
Show Hide image

Clinton and Trump: do presidential debates really matter?

The ability of the candiates to perform in front of the cameras is unlikely to impact the final result.

The upcoming televised presidential debates between Hillary Clinton and Donald Trump are undoubtedly the most eagerly anticipated for many years. No doubt there are various surprises in store – this has been, after all, the most surprising of campaigns.

People will be particularly fascinated to see if Trump dials down his bombastic rhetoric and perhaps even adds some substance to the vague policy pronouncements he has made so far. To a lesser extent, many will also be interested in whether Clinton can add the necessary zest to what some consider her lacklustre style, and whether she can prove she’s made a sterling recovery from her recent bout with pneumonia.

It’s possible that some voters may in fact change their minds based on what they see in the two’s only on-camera encounters. And yet, barring a true disaster or devastating triumph, it’s unlikely that anything the candidates say or do will make much difference to the overall result.

This might not seem all that surprising for these two candidates in particular. Leaving aside how long they’ve both been in public life, social media and the 24-hour news cycle have put Clinton and Trump under incredible scrutiny ever since they announced their respective candidacies – and their every sentence and gesture has already been analysed in the greatest detail.

Trump in particular has received more free publicity from the networks and Twitter than even he could afford, and it’s highly unlikely that he will say anything that the US public hasn’t heard before. Similarly, voters’ impressions of Clinton are apparently so deeply entrenched that she probably won’t change many people’s minds.

Yet there are also broader reasons why presidential TV debates are less important than we might imagine.

Looking the part

Even before the media environment became as saturated as it is today, debates were rarely, if ever, decisive in presidential elections. The exception was possibly the very first TV debate in 1960, which pitted the then vice-president, Richard Nixon, against John F. Kennedy.

At the time, the election was so close that the young, relatively inexperienced but highly telegenic Kennedy was able to reap the benefits of putting his case directly to viewers. He was the underdog; a relative unknown in comparison to Nixon and so had more to gain from such national exposure. Nixon, as the establishment figure, had a lot to lose.

In the end, Kennedy’s narrow victory may well have been because of his debate performances. But his success also demonstrated another important feature of television debates: that viewers take more notice of what they see than what they hear.

Notoriously, television viewers responded very favourably to Kennedy’s film-star good looks, but were turned off by Nixon, who refused to wear make-up and looked sweaty and uncomfortable under the studio lights. In contrast, those who listened on the radio believed that Nixon had come out on top. It seems that viewers saw Kennedy as more “presidential” than Nixon, especially given his calmness under pressure. Kennedy did work hard to exploit some of Nixon’s weaknesses on policy, but in the end, that turned out not to be the point.

Kennedy’s success was one of the reasons that neither of his two successors, Lyndon B. Johnson and then a resurgent Nixon, participated in any such events when they were running for the presidency. Although some debates were held in the primaries, there were no face-to-face contests between presidential candidates in 1964, 1968 or 1972.

The next debates were held in 1976, another tight campaign. These yielded a notorious moment in the second encounter between Gerald R Ford and Jimmy Carter, when the incumbent Ford appeared to throw the election away with a poorly judged remark declaring that there was no Soviet domination of Eastern Europe. As myth has it, this gaffe stalled Ford’s polling surge; he ultimately lost the election.

Yet even this was not decisive. Although the comment did the president no favours, it’s highly debatable whether it in fact had an impact on the overall result; Ford actually closed the polling gap with Carter between the debates and the general election. People’s reactions to the debate had less to do with the substance of his remark and much more with the media’s constant replay and analysis of that moment, which continues to mar Ford’s reputation to this day.

Selective memory

This pattern has continued in the election cycles that have followed, as slips and awkward moments rather than substance provide the media with dominant themes. Many people recall vice-presidential candidate Dan Quayle’s cack-handed attempt to compare himself to Kennedy in 1988, or George Bush senior’s ill-judged glance at his watch when listening to a question in 1992; few probably remember much about what policies they discussed, or whether, if they won, they carried them out.

If anything, the shortcomings of the TV debate format have become more pronounced in the current cycle. Although neither of the main candidates in this year’s election wants for national exposure, the primary debates have tended to favour the underdog and those who claim to be outsiders.

On the Republican side, Trump’s various moderate competitors were one by one hobbled and engulfed; Clinton, for her part, spent months slugging it out with her remarkably successful left-wing rival Bernie Sanders, never quite landing a televised knockout punch and ultimately only defeating him properly after six months of primaries.

While credible policy proposals seem to matter less than ever, things that would have once been considered catastrophic gaffes have become par for the course. Indeed, one could argue that Trump’s success so far is because he has built his campaign on half-truths and outright lies without care for the consequences.

So despite all the anticipation, this year’s debates probably won’t tell us very much about what will happen after the president takes office next January; the analysis will almost certainly focus less on what the candidates have to say and more on how they say it. Voters will no doubt tune in in great, possibly record-breaking numbers, but they’ll come away with precious little sense of what’s in store for their country.

Equally, the spectacles we’re about to witness might be pyrotechnic enough, but they’re unlikely to decide the result in November. And in the unlikely event that they do, it won’t be for the right reasons.

Andrew Priest is a lecturer in Modern US History at the University of Essex

This article was originally published on The Conversation. Read the original article.