Can democracies adapt?

While reading Tim Harford’s book, ‘Adapt’, I was thinking about the implications for democracy of the theme running through the book. I have been thinking a lot about democracy recently.

The theme is what Harford refers to as Palchinsky’s principles:

1. Variation: seek out new ideas and experiment.
2. Survivability: ensure that the scale of experiments is not so big that failure leads to catastrophe.
3. Selection: seek feedback and learn from failures.

The theme is also captured fairly well by the books sub-title: ‘Why success always starts with failure’.

Peter Palchinsky was a talented engineer in the Soviet Union. He was arrested by the secret police in 1928 and never seen again. His main ‘crime’ seems to have been that he refused to recant after he had been denounced for objecting to looming technological disasters.

Harford applies Palchinsky’s principles to a wide range of things, including the management of the war in Iraq by the US miltary, the events leading to the design of the Spitfire in England prior to World War 2, provision of foreign aid, climate change mitigation, financial meltdowns, corporate strategy and self-management.

One of the points that came through strongly to me is that even hierarchical military organizations need to learn strategy from their experiences in the field. I knew that was true of big business organizations, but I had not previously thought of armies as having to experiment in order to find the best approach to adopt in different conflicts. It seems that senior people in the US military were not aware of that either during much of their occupation of Iraq.

Harford’s discussion of the global financial crisis seems sensible. He suggests that the failure of experiments with sub-prime mortgages would not have led to a major financial crisis in the absence of tight coupling in financial markets via credit default swaps. These appeared to provide the banks with a greater margin of safety, but actually linked their fortunes closely to insurance companies. As a consequence, the sub-prime mortgage problem was transmitted from insurance companies to banks (and then to governments). The solution is obvious with hindsight - decouple the system so that banks are less likely to fail and the failure of individual banks does not lead to system failure. I would add that governments should adopt a ‘no bailouts’ policy and back that up with a requirement for banks to guarantee immediate and automatic transfer of ownership of all assets to depositors and bond holders if they (the banks) are unable to meet their interest or repayment obligations.

The points about self-management were not new to me. But it didn’t hurt to be reminded of the importance of trying new things in a context where failure is survivable - and of being prepared to learn from failure. It helps to remember that the failure of our experiments does not necessarily reflect on our competence unless we fail to learn from them.

The closest Tim Harford comes to discussing democracy in this book is his discussion of charter cities. In his version, governments establish the rules of a new city as an experiment to see if citizens and business want to live there.  If the experiment succeeds it can be repeated elsewhere. If it fails, that failure is survivable.

 

In reading Tim Harford’s book I was reminded of a claim by Timothy Ferris:

‘A liberal democracy in action is an endlessly changing mosaic of experiments, most of which partially or entirely fail’ .

The claim is made in his book, ‘The Science of Liberty’ (2010). In my discussion I suggested that Ferris’ view of liberal democracy, as akin to scientific experimentation, was a desirable ideal rather than a description of how democratic systems actually work. I suggested that elections seemed to be a reasonably effective way for citizens to avoid being governed by despots, to get rid of politicians who are obviously corrupt and to experiment with new policies. However, democracy departs from the scientific ideal because the democratic process is not good at ending policy experiments that partially fail. The most radical reforms rarely involve more than a tweaking of failing policy experiments.

If we look at democracy in terms of Palchinsky’s principles there are problems all over the place. There are at least two problems relating to willingness to experiment. First, there is often a strong bias in favour of the status quo because of loss aversion. If there is potential for any politically influential group to be disadvantaged by a policy experiment then it is unlikely to be conducted.  Second, there is often a reluctance to conduct experiments unless they involve the whole of a jurisdiction - questions of fairness are often perceived to be involved.

There are problems relating to survivability. Social experiments are often entered into by governments without adequate consideration of longer term implications for system survival. For example, welfare assistance that is granted when it seems to be affordable is often difficult to withdraw when economic circumstances change for the worse. Some European democracies are currently faced with survivability problems as a result of excessively ambitious social experimentation in the past.

There are also problems in seeking feedback and learning from failures. As noted above, the democratic process is not good at ending policy experiments that partially fail. This reflects two problems. First, governments are often reluctant to accept policy failure because it may have electoral consequences. Second, those who benefit from failing policy experiments are reluctant to see them end and often have the political clout to prevent this from happening.

So, can democracies learn to adapt? We have natural experiments that help us to answer that question. The older democracies have so far shown a capacity to adapt. They have been able to learn from some of their mistakes by adopting institutional innovations to improve the effectiveness of government and, to a lesser extent, to contain the growth of government responsibilities.

There is also potential for the democracies to learn from the success and mistakes of other democracies. For example, there is probably a lot to be learned from the modern experience of democracy in Greece, as well as from ancient history. A recent article by Steve Horwitz makes the point very well.

The idea of learning from the mistakes of others has more general relevance. It might even be worth adding to Palchinsky’s principles. However, it is probably harder to learn from mistakes made by others than to learn from your own mistakes. When my father told me that I should learn from his mistakes, as he often did, my response was that it would be much more fun to make my own mistakes. Looking back now, though, I don’t think my father actually made many mistakes that I could learn from.

What implications does 'monitory democracy' have for the survival of democratic institutions?

I ended my last post promising to consider whether John Keane’s observation that we now have ‘monitory democracy’ has implications for the relationship between the responsibilities and effectiveness of government, and hence the survival of democratic institutions. Can monitory democracy be expected to move political systems towards or away from equilibrium between effectiveness of government and the responsibilities that government is expected to perform?

First, who is Keane and what does he mean by ‘monitory democracy’? John Keane is professor of politics at Sydney university and author of a major history of democracy, ‘The Life and Death of Democracy’ (2009), a book over 1,000 pages described by one reviewer as not ‘for the faint hearted’. (Unfortunately, I can’t claim to have read it.) In an article in the Griffith Review, Keane argues that from the middle of the 20^th Century representative democracy began to transform into monitory democracy – a new historical form described by ‘the rapid growth of many different kinds of extra-parliamentary, power-scrutinising mechanisms’.

Keane’s list of power-scrutinising mechanisms includes: advisory boards, focus groups, citizen juries, talk shows, think tanks, consensus conferences, teach‐ins, online petitions and chat rooms, public vigils, straw polls, summits, public planning exercises, public consultations, social forums and, of course, weblogs. About the only thing these inventions have in common is that they change the incentives faced by politicians and political parties. Keane suggests:

‘The central grip of elections, political parties and parliaments on the lives of citizens is weakening. Democracy is coming to mean more than elections, although nothing less. Within and outside states, independent monitors of power begin to have tangible effects. By putting politicians, parties and elected governments permanently on their toes, they complicate their lives, question their authority and force them to change their agendas …’

Monitory democracy is closely linked to the emergence of new communications media. In Keane’s terms, ‘monitory democracy and computerised media networks behave as if they are conjoined twins’. He observes:

‘The combination of monitory democracy and communicative abundance … produces permanent flux, an unending restlessness driven by complex combinations of different interacting players and institutions, permanently pushing and pulling, heaving and straining, sometimes working together, at other times in opposition to one another’.

In considering what implications monitory democracy might have for the survival of democratic institutions it seems to me to be worth recalling Joseph Schumpeter’s view, discussed here recently, that democracy is essentially a leadership contest in which the role of citizens ends after they have cast their votes.  The reality of democracy, as described by Keane, is light years away from Schumpeter’s view that democracy can only succeed if voters refrain from trying to tell politicians what to do after they have been elected.

How does monitory democracy actually impact on the balance between the responsibilities and effectiveness of governments?

Some of Keane’s comments suggest that monitory mechanisms might have a positive impact on the effectiveness of government. He points out that, ‘when they do their job well, monitory mechanisms have many positive effects, ranging from greater openness and justice within markets and blowing the whistle on foolish government decisions to the general enrichment of public deliberation and the empowerment of citizens and their chosen representatives through meaningful schemes of participation’. On the other hand, he suggests that nobody ‘should be kidded into thinking that the world of monitory democracy … is a level playing field – a paradise of equality of opportunity among all its citizens and their elected and unelected representatives’.

The combination of monitory democracy and communicative abundance is also likely to impact on voter rationality and the incentives for the major political parties to develop coherent policies to sell to the electorate. There is still a strong incentive for politicians to have regard to particular interests of voters, but it seems to me that their incentive to appeal to the reasoning powers of uncommitted voters may be diminishing. As Keane seems to imply, the new political environment may be discouraging uncommitted voters from taking an intelligent interest in policy issues:

‘Monitory democracy certainly feeds upon communicative abundance, but one of its more perverse effects is to encourage individuals to escape the great complexity of the world by sticking their heads, like ostriches, into the sands of wilful ignorance, or to float cynically upon the swirling tides and waves and eddies of fashion – to change their minds, to speak and act flippantly, to embrace or even celebrate opposites, to bid farewell to veracity, to slip into the arms of what some carefully call “bullshit”.’

This does not provide strong grounds for confidence that monitory democracy improves the effectiveness of government.

How does monitory democracy impact on the scope of responsibilities that governments are expected to perform? The discussion in a recent post about voter irrationality seems highly relevant. There is evidence that voters who say that politics is ‘not at all important’ to them are more likely than others to say that ‘the government should take more responsibility to ensure that everyone is provided for’.

On that basis, monitory democracy could be expected to thrust more responsibilities on governments than they can cope with effectively. This raises serious questions about the ability of democratic institutions to survive.

However, as more people perceive that existential threats are facing democratic institutions they might form new interest groups to foster norms of behaviour that might enable better outcomes to be achieved. Under favourable conditions monitory democracy might prove to be a system with self-correcting characteristics. If influential interest groups can form around the survival of some threatened species of animals, is it not reasonable to expect that stronger and more influential interest groups might form when social institutions that make a valued contribution to human well-being are seen to be threatened? 

Postscript:
I would like to draw attention to relevant comments by kvd and Jim Belshaw on Jim's blog. Jim now also has another relevant post: What would you nominate as the most asinine slogan?

In his comments below kvd has drawn attention to an article by Michael Pascoe entitled, 'Timid Governments Bow to Populism', SMH (9/7/2012). That article is well worth reading, as is the article by Thomas Friedman entitled 'The Rise of Popularism' NYT (23/6/2012) which is referred to by Pascoe. Please note the word used by Friedman is 'popularism'. It was only when my spell-checker objected that I understood why Friedman wrote: 'I heard a new word in London last week: “Popularism”.'

Neil Whitford is not faint hearted. He has read and reviewed 'The Life and Death of Democracy'. 

What determines the ability of democratic governments to cope with their responsibilities?

I ended my last post suggesting that the greatest threat to human progress over the next few decades is that democratic governments will not be able to cope with the increasing responsibilities that seem likely to be placed upon them, particularly in high-income countries.

The most straight forward aspect of this question is the economic effect of increases in government spending and taxation. (I have discussed this previously here and here). It is well known that the economic cost of taxation rises disproportionately with tax revenue because taxpayers substitute non-taxed goods (e.g. leisure) for taxed goods. (The economic cost is roughly proportional to the square of the tax rate.) There are also economic costs associated with government spending that inadvertently discourages participation in productive activity. These costs can be reduced to some extent by tax reforms (e.g. broadening the base in order to lower the rate) and various measures to reduce the disincentive effects associated with government spending. As the economic costs have risen, some countries with big governments (e.g. Sweden) have introduced various reforms that have had the effect of containing the growth of government spending and reducing marginal tax rates.

 

So, if the economic costs of big government become too high, governments can take action to reduce those costs. Where is the problem?

The problem is that countries differ greatly in their ability to cope with the problems of big government – in terms of both raising revenue and managing provision of services, and implementing tax reforms etc. if that becomes necessary. Sweden seems to be able to cope with government spending that is still around 50 per cent of GDP without too many problems, whereas Greece was on the way to becoming a basket case before its government spending reached that level.

The wide variation in the effectiveness of government in OECD countries with big governments is apparent from the chart below. Size of government is measured by OECD data on government spending (General Government Total Outlays) as a percentage of GDP averaged over the five years ending 2007. Government effectiveness is measured by the relevant World Bank governance indicator which reflects, among other things, perceptions of the quality of the civil service and its independence from political pressures, and the quality of policy formulation and the credibility of the government's policy commitments.

I chose the period prior to the global financial crisis (GFC) in making this comparison to see to what extent the indicator of government effectiveness predicts how countries have fared subsequently. It predicts that three of the PIIGS - Italy, Greece and Portugal - might have problems in coping, but that Ireland and Spain would be less likely to have problems. Failure to predict problems in Ireland and Spain could be attributed to deficiencies in both the indicator of government effectiveness and the indicator of size of government. In relation to effectiveness, it does not seem consistent with effective governance for bond holders of banks to be viewed as ‘consenting adults’ when the future looks rosy and then subsequently issued with guarantees when the future becomes uncertain. In relation to size of government, it could also be argued that government spending was a poor indicator prior to the GFC because it didn’t take account of the contingent liability associated with implicit guarantees for bond holders of banks. For example, in the case of Ireland, a better measure of the size of government might have anticipated in some way the subsequent increase in government spending from around 34 percent of GDP in 2006 to 67 percent in 2010.

Leaving measurement problems aside, it seems reasonable to argue that the ability of democratic governments to cope with the responsibilities being placed upon them will depend on the extent of those responsibilities and on the effectiveness of the governments concerned. The question that needs to be considered is whether democracies are changing in ways that are likely to contribute to or detract from the effectiveness of government, or add to or subtract from the responsibilities that are placed on government. If effectiveness is rising and responsibilities are contracting, the future of democracy looks rosy. If effectiveness is falling and responsibilities are expanding, democracies may be heading for collapse.

In my next post I will consider whether John Keane’s observation that we now have ‘monitory democracy’ has implications for the relationship between the responsibilities and effectiveness of government, and hence the survival of democratic institutions.

What is the greatest threat to human progress?

It seems pretentious to be posing this question for myself. But the question is difficult to avoid in the book I am writing. It might help me to take it out of the ‘too hard’ basket and get my ideas straight if I try to draft an answer here first.

The background is that the world has seen massive human progress over the last three centuries. Human societies have generally become much more peaceful; the opportunities for individuals to live lives that they value have expanded greatly in much of the world; and people in many countries now enjoy a great deal of economic security. From what we know about the drivers of this social progress it would be reasonable to expect this process to continue as long as conditions remain broadly favourable.

Whenever I read statements like ‘as long as conditions remain broadly favourable’ I begin thinking up reasons why conditions will not remain favourable. That response probably has something to do with being human. Loss aversion probably makes us sensitive to future threats to our quality of life, so that we can avoid them.

Matt Ridley suggests in ‘The Rational Optimist’ that loss aversion makes people ‘naturally gloomy’ about ‘the future of society and the human race’. He argues that for 200 years those preaching doom and gloom ‘have had all the headlines, even though optimists have far more often been right’. I think he overstates his case. I am old enough to remember what today seems like a naive faith in progress that seemed to pervade western society in the 1950s and ‘60s, despite a real risk of nuclear Armageddon during much of that period. Even today there is a fair amount of optimism in reporting of scientific progress. Slow progress is being made in treatment of cancer, for example, but press reports of breakthroughs often seem to paint an excessively optimistic picture.

I agree with Ridley that there has been a tendency to forecast the future on the assumption of no technological change and to find it dire. Ridley quotes Paul Romer: ‘We consistently fail to grasp how many ideas remain to be discovered’. Most gloomy prognostications about the world running out of resources and the potential for widespread famine should be heavily discounted on those grounds. Disease pandemics are a greater threat, but we have lived with this threat at least since the beginning of urbanization and scientific progress has been improving our ability to cope.

While optimists have generally been on a winning streak over the last few centuries that cannot be because perceived threats are always exaggerated by gloom and doom merchants. The threat of a major nuclear war is a case in point. As Ridley acknowledges: ‘There were very good reasons to be a nuclear pessimist during the Cold War’ (p 299). Everything that I have read about the Cuban missile crisis of 1962 suggests to me that this could easily have escalated to nuclear war, despite mutually assured destruction. Steven Pinker seems to me to make a strong case that a taboo against use of nuclear weapons emerged gradually as a result of pressure of public opinion rather than because military and political leaders decided spontaneously that use of such weapons of mass destruction would be futile (‘The Better Angels of Our Nature’, pp 268-271). In my view Gregg Easterbrook is right on target in his observation that ‘historians will view nuclear arms reduction as such an incredible accomplishment that it will seem bizarre in retrospect that so little attention was paid while it was happening’ (‘The Progress Paradox’, p 70).

It seems to me that the only known threat we face that could rival the threat of nuclear Armageddon in terms of severity of its impact is climate change. Even in that context, however, the threat of human-induced climate change seems likely to be much less severe than the aftermath of a major nuclear war (or infrequently occurring natural occurrences such as super-volcanoes and asteroids). I have written recently about human-induced climate change (here and here) so I will be very brief. The most likely outcome in my view is that the world will stumble on toward a reduction in greenhouse gas emissions, relative to what would otherwise occur. This will not prevent climate change, but the general story is likely to be one of successful adaptation. It is within the bounds of possibility, however, that climate change could accelerate and the costs of adaptation could begin to rise steeply within the next few decades. It therefore makes sense for governments to take precautions by contributing substantially to funding of research to mitigate climate change.

There are other potential threats to the quality of life over the next few decades that apply particularly to high-income countries. Globalization and technological progress will continue to have the potential to raise overall living standards in these countries, but there seems to be a fair chance that the distribution of benefits will become more unequal and jobs will become less secure. There are some good reasons why the aphorism, ‘a rising tide lifts all boats’ should be questioned. The forces of globalization are likely to subject an increasing proportion of occupations, including some occupations requiring substantial skills, to increased international competition. It is also possible that technological progress will impact unevenly in ways that benefit some groups (e.g. successful innovators, owners of robots) and disadvantage others (e.g. those whose skills are becoming redundant). If that happens, expressions of collective guilt/responsibility, like this one by Jim Belshaw on his blog yesterday, are likely to become more common:

‘We talk about the need for management and people flexibility in that most basic area, secure employment. We say that a young person will follow multiple career paths in their lives. Yet we do nothing to address the most basic questions: how might this actually work? How do we create a world that might provide the desired business and personal flexibility? How do we give people a degree of certainty about their own lives in an unstable world?’

It seems likely that governments in high-income countries will come under increasing pressure to provide people with a greater degree of economic security. If we are to avoid a return of protectionism – protecting  existing jobs at the expense of new job opportunities - more adjustment assistance may need to be provided. This could involve more government assistance for retraining of people whose skills are becoming redundant and broadening social welfare safety nets that are already coming under increasing stress as a result more predictable developments such as the rising age structure of populations.

Some people are concerned about a rather different potential threat to the quality of life in high income countries. This is a concern that rising affluence brings with it epidemics of obesity, diabetes, depression, and other ills of modern life. (See for example Jeff Sach’s introductory chapter in ‘World Happiness Report’, 2012.) Such concerns seem likely, increasingly, to convert the problems that individuals and families have in making good use of the opportunities available to them into ‘social’ problems that governments are called upon to address. For example, some governments (including the Australian government) are already taking increasing responsibility for preventative health care, weakening the responsibility of individual adults to manage their own lives.

It seems to me that the greatest threat to human progress over the next few decades arises because threats to human progress seem to provide a compelling case for collective action - the threats to human progress discussed above give rise to greater demands on government. As I see it, the greatest threat we are faced with over the next few decades is that democratic governments in high-income countries will not be able to cope with the increasing demands that seem likely to be placed upon them.

I will write about that in my next post.

Postscript:
For a highly relevant recent discussion of effects of globalization and technological change on income distribution see: Jonathan Haskel, Robert Z. Lawrence, Edward E. Leamer and Matthew J. Slaughter,
'Globalization and U.S. Wages: Modifying Classic Theory to Explain Recent Facts', Journal of Economic Perspectives (26) 2, Spring 2012.