Private versus public nudging
14 January 2011
SUGGESTED
Uncategorized
Uncategorized
Government and Institutions
Last month I attended a speech by Richard Thaler, of ‘Nudge’ fame. Professor Thaler is an engaging speaker. His assertion that ‘libertarian paternalism’ is merely an extension of methods widely adopted in daily life is a beguilingly simple, yet ultimately dangerous one. According to Thaler since good mothers constantly attempt to ‘nudge’ their offspring in the ‘right direction’ then we should not worry about governmental nudging. Whether it is ‘fat taxes’ to discourage unhealthy eating induced by ‘weakness of will’ or compulsory enrolment in savings schemes to induce less ‘short-sighted’ and more thrifty conduct, people should learn to love a government that coaxes them towards what they themselves would recognise is in their own best interests.
The classical liberal tradition is not, and never has been suspicious of the idea that individual behaviour can or should be ‘steered’ and that conscious attempts might be made to ‘shape’ our character. Outside of the nudges provided by family life, people may join voluntary associations which exert different forms of social and peer-group pressure to steer them in one direction or another. Religious associations have long discouraged alcohol consumption and have favoured abstinence in various aspects of life. Many of the mutual aid associations that thrived during the heyday of liberalism in the nineteenth century modelled themselves explicitly as ‘character-building’ associations that encouraged thrift and an independent spirit. And, in the contemporary world the various reward schemes offered by health insurers to encourage personal fitness, devices such as minimum term gym memberships, and penalty charges for early withdrawal of investment funds are all examples of privately generated strategies to cope with ‘weakness of will’ and ‘short-sightedness’.
Listening to Professor Thaler I was reminded of the claim made by many socialists in the past – Lenin being perhaps the most prominent – that since private firms routinely engage in ‘planning’ there should not be any concern about the state ‘planning’ on a society wide scale. Yet, as Hayek noted on numerous occasions, to recognise that ‘planning ‘ is an essential element of a progressive society tells us nothing about ‘who should plan’ and ‘for whom’. Likewise, to acknowledge that ‘nudging’ strategies may be an aid to effective decision-making in the context of limited rationality, tells us nothing about ‘who should nudge’ and ‘for whom’. It does not follow that since some nudging may be desirable that we should automatically favour governmental nudging. On the contrary, there are several reasons to suggest that ‘private’ nudging should in fact be preferred to the statist variety.
First, in a context of limited knowledge and limited rationality, we do not know which nudges are most appropriate and for which particular types of behaviour. It makes sense, therefore, to rely on a decentralised process which reduces the possibility of erroneous nudges being imposed on a society wide basis – and this requires that no particular nudge is imposed by law. In the same way that planning by private firms is preferable to planning by government’s precisely because it is competitive, decentralised and voluntary planning, so competitive nudging in markets and civil society is to be preferred to ‘central nudging’ by the state. That the consequences of misplaced nudging by government agencies tend to be far more pronounced than equivalent failures in the private sphere is all too evident in what has happened to savings ratios across much of the developed world. It is odd, to put it mildly, that governments which have ‘nudged’ people towards immediate consumption through a combination of inflationary monetary policies and taxes on capital should now be trusted to encourage more frugal habits.
Second, while behavioural scientists in general (such as Thaler) may have better knowledge about the biases that limit people’s choices, the specific knowledge needed to cope with particular instances of irrational conduct and detailed knowledge of the reasons why people make the choices that they do is more likely to reside with those who actually make these choices. Classical liberalism does not claim that people are fully rational but that people are more likely to be aware of their own character traits than distant experts, however technically skilled in behavioural economics. Someone who is, for example, prone to ‘weakness of will’ may find it advantageous to join an organisation such as the army, or an amateur boxing club – a life choice that may be entirely inappropriate for an individual who already exhibits the necessary strength of character. Centrally imposed mandates and taxes are not, however, best placed to identify those choices which reflect preferences free from behavioural bias from those that are not, and are likely to place obstacles in the way of those who have genuinely unbiased preferences to act in ways of which officially sanctioned nudgers disapprove.It might of course be suggested that people may lack the knowledge or determination to avail themselves of private nudging strategies in shaping their own lives. Yet, if people are as lacking in motivation to address their character problems as this analysis implies then there would seem no more reason to believe that they would vote for a government proposing to subject them to ‘libertarian paternalism’.
Finally, people who are to be charged with nudging in the public sector may themselves be subject to the same or perhaps different behavioural biases to those of the population at large. That Professor Thaler seems unwilling to consider the dangers of behavioural bias in the institutions of government and those of democratic politics might be considered to indicate a cognitive blind spot on his behalf…
Read the full article on the Pileus blog.
Mark Pennington’s new book, Robust Political Economy: Classical Liberalism and the Future of Public Policy, will be launched at the IEA on 20 January 2011. Click here for details of the event.
The classical liberal tradition is not, and never has been suspicious of the idea that individual behaviour can or should be ‘steered’ and that conscious attempts might be made to ‘shape’ our character. Outside of the nudges provided by family life, people may join voluntary associations which exert different forms of social and peer-group pressure to steer them in one direction or another. Religious associations have long discouraged alcohol consumption and have favoured abstinence in various aspects of life. Many of the mutual aid associations that thrived during the heyday of liberalism in the nineteenth century modelled themselves explicitly as ‘character-building’ associations that encouraged thrift and an independent spirit. And, in the contemporary world the various reward schemes offered by health insurers to encourage personal fitness, devices such as minimum term gym memberships, and penalty charges for early withdrawal of investment funds are all examples of privately generated strategies to cope with ‘weakness of will’ and ‘short-sightedness’.
Listening to Professor Thaler I was reminded of the claim made by many socialists in the past – Lenin being perhaps the most prominent – that since private firms routinely engage in ‘planning’ there should not be any concern about the state ‘planning’ on a society wide scale. Yet, as Hayek noted on numerous occasions, to recognise that ‘planning ‘ is an essential element of a progressive society tells us nothing about ‘who should plan’ and ‘for whom’. Likewise, to acknowledge that ‘nudging’ strategies may be an aid to effective decision-making in the context of limited rationality, tells us nothing about ‘who should nudge’ and ‘for whom’. It does not follow that since some nudging may be desirable that we should automatically favour governmental nudging. On the contrary, there are several reasons to suggest that ‘private’ nudging should in fact be preferred to the statist variety.
First, in a context of limited knowledge and limited rationality, we do not know which nudges are most appropriate and for which particular types of behaviour. It makes sense, therefore, to rely on a decentralised process which reduces the possibility of erroneous nudges being imposed on a society wide basis – and this requires that no particular nudge is imposed by law. In the same way that planning by private firms is preferable to planning by government’s precisely because it is competitive, decentralised and voluntary planning, so competitive nudging in markets and civil society is to be preferred to ‘central nudging’ by the state. That the consequences of misplaced nudging by government agencies tend to be far more pronounced than equivalent failures in the private sphere is all too evident in what has happened to savings ratios across much of the developed world. It is odd, to put it mildly, that governments which have ‘nudged’ people towards immediate consumption through a combination of inflationary monetary policies and taxes on capital should now be trusted to encourage more frugal habits.
Second, while behavioural scientists in general (such as Thaler) may have better knowledge about the biases that limit people’s choices, the specific knowledge needed to cope with particular instances of irrational conduct and detailed knowledge of the reasons why people make the choices that they do is more likely to reside with those who actually make these choices. Classical liberalism does not claim that people are fully rational but that people are more likely to be aware of their own character traits than distant experts, however technically skilled in behavioural economics. Someone who is, for example, prone to ‘weakness of will’ may find it advantageous to join an organisation such as the army, or an amateur boxing club – a life choice that may be entirely inappropriate for an individual who already exhibits the necessary strength of character. Centrally imposed mandates and taxes are not, however, best placed to identify those choices which reflect preferences free from behavioural bias from those that are not, and are likely to place obstacles in the way of those who have genuinely unbiased preferences to act in ways of which officially sanctioned nudgers disapprove.It might of course be suggested that people may lack the knowledge or determination to avail themselves of private nudging strategies in shaping their own lives. Yet, if people are as lacking in motivation to address their character problems as this analysis implies then there would seem no more reason to believe that they would vote for a government proposing to subject them to ‘libertarian paternalism’.
Finally, people who are to be charged with nudging in the public sector may themselves be subject to the same or perhaps different behavioural biases to those of the population at large. That Professor Thaler seems unwilling to consider the dangers of behavioural bias in the institutions of government and those of democratic politics might be considered to indicate a cognitive blind spot on his behalf…
Read the full article on the Pileus blog.
Mark Pennington’s new book, Robust Political Economy: Classical Liberalism and the Future of Public Policy, will be launched at the IEA on 20 January 2011. Click here for details of the event.
9 thoughts on “Private versus public nudging”
Comments are closed.
SHARE
Firsty, a company can plan as it is planning its own assets and property and it suffers the mistakes. The State does not “own” us, so has no right to “plan” us, either. We are the ones who tend to suffer the mistakes it makes, though.
While, as a Minarchist, I have nothing against the State dismantling perverse incentives or distortions that encourage harmful behaviour, I am not supportive of the State being given permission (and that is how we and it should view this) to positively encourage certain behaviours. Who decides?, how will it be encouraged? These questions expose the flaw, the systemic dysfunction of State nudging. The State should keep off. Must keep off.
Private/voluntary/mutual planning is fine, as you say, as it will not be enacted in law. As you also say, it will be pluralistic or open to pluralism at any time, unlike State planning which tends, because of monopoly, to be a monoculture. One head is worse than two.
Nudging, as it is emerging, is also nothing to do with Libertarianism. The term “Libertarian Paternalism” is as near an oxymoron as to make no difference. It appears to be a land-grab of the term “Libertarian” just as Fabians grabbed “Liberal”. It is as much at risk of being bastardised.
Just to pick up on a finer point (though not unimportant), Thaler believes in auto-enrolment and not compulsory enrolment (to take the pension scheme example). I am not sure that he believes in fat taxes either (that is not a behavioural solution, as such). Thaler would argue that people are still free to not enrol under auto-enrolment and either enrolment or auto-not-enrolling has to be the default position (which the individual can change). I argued against this saying that many (especially young) people should not be in a pension scheme and therefore the nudge was in the wrong direction. He replied saying that the costs of the young person opting out would be small so it does not matter. Well, if it does not matter, there is no point nudging! However, is your argument subtly changed in these cases – in general, Thaler wants to allow people to do whatever they wish but change the way options are presented to influence the choice. He also wants to do this sort of thing INSTEAD of regulation – but that is never the result in practice!
Philip, thanks for the comment – v. useful. On compulsory enrolment – this is not so clear. If Thaler is simply advising employers to autoenrol people then no problem – but Sunstein and Thaler are also on record as saying that the law ‘might require employers to provide automatic enrollment and allow employees to opt out’ (Univ of Chicago Law Review, 70: 1176). They are not clear on this – but if employers are mandated to enrol people in pension schemes then this is compulsion – exercised against the employers -though obviously employees do retain the option of leaving the relevant savings scheme. On fat taxes, I’m not sure of Thaler’s position on this – but certainly other ‘behavioural economists’ have advocated sin taxes – which include ‘fat taxes’ – eg Camerer et al (2003) Regulation for Conservatives, Univ of Pennylvania Law Review, 1211-1254.
Even in the case of auto-enrolment, the case for intervention is weak. Recall that ‘libertarian paternalism’ purports to enable people to achieve the preferences that they themselves have – as opposed to ‘traditional paternalism’ – which claims to know what preferences people should have. For this to be operationalised then policy-makers must be able to distinguish between those agents who have genuinely unbiased preferences in favour of say not saving much (prefering present to future consumption), from those who fail to save owing to weakness of will etc. Yet, there is little reason to believe that policy-makers can do this.
To require auto-enrolment may make it easier for those who would otherwise not save to do so – but it will also make it more difficult for those who have a genuine preference for short-term consumption to achieve their preferences – they will face the costs of having to go through the bureaucracy of exiting from the scheme that they otherwise wouldn’t face. And, if they have a problem of ‘weakness of will’ in coping with such bureaucracy, that will make their position even worse. As a consequence, a proposal which purports to be neutral with respect to people’s preferences is in effect an attempt to bias people towards the preferences of ‘libertarian paternalists’ – whether this is for a particular rate of saving or anything else – and of course that is just plain old fashioned paternalism.
Taken in isolation the costs from introducing any one nudging measure in terms of unsatisfied preferences are likely to be marginal – as is the case with any one tariff or protectionist regulation. As we know though, in practice, governments don’t content themselves with intervening in just the odd one or two areas – and if you look at the agenda of Thaler’s colleague Cass Sunstein, he doesn’t want them to confine their ambitions in that way either.
thanks for the clarification. Very interesting – and it was in trying to deal with this point (that people may be nudged in the wrong direction that he seemed to go all at sea by saying that it would not affect people very much and so on.
Another strongly argued post by Dr Pennington. I think we can look at nudging from two directions. This is ‘practical’ or empirical examination of why the state ought not nudge us.
We also ought to look at the question of nudging from a more ‘theoretical’ perspective. In short, it comes down to the question of knowledge. As information is distributed throughout society and as human actions are a process of discovery, for the state to assume that it can know better than the distributed knowledge of individuals and free institutions is clearly erroneous. Moreover, for the state to assume that it can know the best outcome and that is what everyone in a society ought to aim for is clearly the fatal conceit. This, I think, would be von Mises’ position.
Secondly, we should think about ‘nudging’ from a constitutional position (as in something that is concerned with the proper role and powers of government). Put simply, it is not the proper duty of government to control and direct behaviour. That may be a role for independent and free social institutions and norms, but it is not a duty of government. This stems from both the practical and knowledge-limitation issues that government faces. However, even if government did not face these problems and could know that it was nudging in the right direction, there is also the question of where it ought to stop. After all, if government can intervene in one way, why not another and then in every way? Also, we know that bureaucratic agents seek to empire-build and expand beyond their powers – or at least behavioural scientists should know! What is required is a constitutional limit beyond which government cannot step – my belief is that limit should be to issues where property rights are disputed or being violated that government may intervene, and provide genuine public goods (defence, justice). In these areas we might think about nudging, although clearly the scope for nudging will be much more limited and should also be limited by the practical constraints Dr Pennington highlights.
“they will face the costs of having to go through the bureaucracy of exiting from the scheme”
That’s not true. Thaler says in the introduction that a “nudge” is only a nudge if it’s easy to opt out. So in this case, it would merely be a matter of unticking a box. If somebody has a preference for not saving or not having health insurance now, even if it is a weak preference, then they could act accordingly at a minimal effort.
The target group are those whose reason for not saving/insuring is just plain apathy. What’s wrong with nudging people who are apathic? Yes, this is a small group. Nudging (if I understood it correctly) is not about imposing a grand scheme that results in measurable differences in variables like the savings ratio.
I don’t buy the knowledge argument either. Of course, we cannot know what the best savings or health insurance plan for everyone is. So? There are a few things which we do know. One of them is that never saving anything for old age, or having no healthcare coverage whatsoever, can barely be in anybody’s interest. Attempts at rationalising such behaviour look hugely constructed to me. Hence, a poor value savings/insurance plan is better than none at all.
Yes, we could construct some hypothetical situation of somebody who has a perfectly good reason for not saving/insuring,but who would nevertheless not opt out under a nudging plan. That hypothetical person would then be worse off with government nudging than without. Or we could just admit that this scenario is highly improbable, and that the far more likely reason is apathy. If people have a reason for opting out, they will opt out, and the apathic few will have some minimal coverage.
I go with some of what you say, Kris, but it is surprising how much governments do not know (but which government thinks it knows). Of course, there will be apathetic non-savers who probably should be saving. But there will be apathetic non-opting-outers who should opt out. Particularly, this will include people who have a mortgage (though some people with a mortgage like to save in any case just to discipline themselves – showing again how complex it all is). It seems to me that savings behaviour is quite rational and the groups that the government highlight as not saving sufficiently have huge marginal withdrawal rates from the income that is generated from saving. You can justify compulsion for this group under some assumptions (not as a libertarian but as an economist thinking about welfare implications in a second-best environment) but nudging is the wrong tactic. Nudging, as you point out, is the tactic for the apathetic. If you assume apathy and nudge when the correct motivation is self interest in a means-tested social security system then you simply get a different group suffering (the apathetic non-opting-outers as I put it). One of the problems is that the government is assuming that people are apathetic rather than astute. People are perhaps not as daft as the government likes to think they are.
Ok Kris – I take the point. But if you are right about Thaler’s view here on what constitutes nudging – and I think he emphasises different aspects to different audiences – then this surely runs up against the point Philip was making. If it is so easy to exit from the scheme – then what’s the point of having it in the first place – those who suffer weakness of will but who do have a genuine preference to save for the longer term, will simply exit alongside those who prefer current consumption. If you’re right about what is a nudge – then the schemes that private investment funds offer – penalty charges etc. for early withdrawl would not count as ‘nudges’ – but they might be far more effective. And, they have the virtue that people enter them on a voluntary basis.
Mark P.
@Kris – you also need to acknowledge that many of the bad behaviours are the consequences – often unintended but foreseeable – of government interventions. Very often government behaviours are contradictory. For instance, government is encouraging consumption by running low and negative interest rates, say, or providing welfare payments which discourage saving. It is then rather perverse if it then decides to nudge people to save more. In the same way that government simultaneously subsidises poor health and smoking by providing ‘free’ healthcare then attempts to punish smokers and the overweight and pay them for adopting better diet. Better, surely, to remove the initial cause of such behaviours – government induced – and allow free institutions to develop mechansisms, which we may not be able to foresee, for solving them.
It’s unlikely that government is producing the better results, and even if we could know it I find it highly unlikely given the frequent and evident costs of intervention. But even if it were, we should also be asking whether it is the duty of government to do so, or else why doesn’t government simply take over every posible choice?