Let’s roll back auto-enrolment
30 June 2020
SUGGESTED
Economic Theory
Society and Culture
Healthcare
4 July 2024
Pension auto-enrolment was one of the early significant policies arising from the government’s “nudge unit”. The argument was that employees suffer from inertia and therefore do not save as much as they would rationally like to. Enrolling employees in a pension scheme and then allowing them to opt out, it was suggested, would lead to individuals acting in ways that better matched their real, underlying preferences. This is a classic policy adaptation of Thaler and Sunstein’s “Libertarian Paternalism” agenda.
Policy insights based on behavioural economics receive a mixed response from free-market economists, but it is worth noting that Richard Thaler proposes it as an alternative to regulation and compulsion. He believes that, if people are nudged, they will take rational decisions and there will be no need for compulsory pension provision (to take this particular example).
It should also be noted that this rationale for intervention is different from the one usually given in relation to compulsory pension provision. It is often argued that we should force people to save because, if they don’t, they will be a burden on the welfare state. Now that the basic state pension is higher than welfare levels (rightly or wrongly), this particular argument is no longer relevant.
Despite its merits, there are a number of arguments against auto-enrolment, and especially the way it has been implemented.
To begin with, governments, over the decades, have been removing various structures that used to exist within employment contracts to encourage the kind of behaviour that it is now regarded as appropriate to nudge. Until the Social Security Act 1986, it was possible for an employer to require employees to join their pension scheme. This was a voluntary paternalistic device which overcame the supposed myopia of potential members. The government not only made that illegal, they retrospectively over-wrote freely negotiated employment contracts that allowed employers to require their employees to join their pension scheme. This ruling was responsible for about 90 per cent of the pension mis-selling crisis in the early 1990s. Eventually, 30 years on, auto-enrolment was introduced, thus introducing as a legal requirement a practice that the government had previously prohibited.
Secondly, an intrinsic problem with all nudge policies is that, as well as ensuring that people save who may otherwise fail to do so because of inertia, they also encourage saving amongst those who should not do so. For example, auto-enrolment nudges people into saving who should be paying off debts. There will be other people who might be saving for a house and for whom pension saving is not appropriate. These people are more likely to be young than old.
Indeed, the government is not just nudging people in relation to pension auto-enrolment, it is giving them a big shove. As well as auto-enrolment leading to a 4 per cent contribution from the employee, it requires a 3 per cent contribution from the employer. As it is illegal for the employer to offer additional salary to those who opt out, if people cannot afford additional pension contributions under auto-enrolment, they are penalised by losing the employer contribution. The requirement for an employer contribution goes way beyond the nudge philosophy. It also raises employers’ costs and creates pressures that are likely to lead to lower wages at a time when real earnings growth is under pressure.
Remarkably, nearly all assessments of the success of the auto-enrolment scheme focus almost entirely on the proportion of people who do not opt out (it is assumed the higher that is the better). This includes the assessment by the Cabinet Office’s Behavioural Insights Team.
The correct measure of success would compare the number of people who would like to save but whom inertia stops saving and who therefore benefit from auto-enrolment (the higher the better) with the number of people who would rationally prefer not to save but who have not opted out because of inertia (the lower the better). Of course, this is harder to measure, but it would provide information that would have the advantage of being relevant.
It is also worth noting that auto-enrolment imposes burdens on employers and employees in relation to ludicrously low pension contributions. Auto-enrolment applies to anybody who has earnings of more than £10,000 per annum. And this threshold is not adjusted for inflation. Contributions are only payable above the national insurance threshold. This means that those on low earnings might be making pension contributions each year that would earn them an annual pension of about £10 to £15 from the age of 65.
Without dismantling auto-enrolment, reform should attempt to maximise the number of people who have a preference to save but choose not to because of inertia whilst minimising the number of people who remain contributing due to inertia but for whom it is not rational to save. The following three reforms should move the scheme in that direction whilst reducing burdens on employers:
Number 2 would be difficult to achieve politically. However, 1 and 3 would be a welcome (though small) deregulation of the labour market which, during a time of declining productivity growth, has suffered from more and more regulation in the last 12 years.
Policy insights based on behavioural economics receive a mixed response from free-market economists, but it is worth noting that Richard Thaler proposes it as an alternative to regulation and compulsion. He believes that, if people are nudged, they will take rational decisions and there will be no need for compulsory pension provision (to take this particular example).
It should also be noted that this rationale for intervention is different from the one usually given in relation to compulsory pension provision. It is often argued that we should force people to save because, if they don’t, they will be a burden on the welfare state. Now that the basic state pension is higher than welfare levels (rightly or wrongly), this particular argument is no longer relevant.
Despite its merits, there are a number of arguments against auto-enrolment, and especially the way it has been implemented.
To begin with, governments, over the decades, have been removing various structures that used to exist within employment contracts to encourage the kind of behaviour that it is now regarded as appropriate to nudge. Until the Social Security Act 1986, it was possible for an employer to require employees to join their pension scheme. This was a voluntary paternalistic device which overcame the supposed myopia of potential members. The government not only made that illegal, they retrospectively over-wrote freely negotiated employment contracts that allowed employers to require their employees to join their pension scheme. This ruling was responsible for about 90 per cent of the pension mis-selling crisis in the early 1990s. Eventually, 30 years on, auto-enrolment was introduced, thus introducing as a legal requirement a practice that the government had previously prohibited.
Secondly, an intrinsic problem with all nudge policies is that, as well as ensuring that people save who may otherwise fail to do so because of inertia, they also encourage saving amongst those who should not do so. For example, auto-enrolment nudges people into saving who should be paying off debts. There will be other people who might be saving for a house and for whom pension saving is not appropriate. These people are more likely to be young than old.
Indeed, the government is not just nudging people in relation to pension auto-enrolment, it is giving them a big shove. As well as auto-enrolment leading to a 4 per cent contribution from the employee, it requires a 3 per cent contribution from the employer. As it is illegal for the employer to offer additional salary to those who opt out, if people cannot afford additional pension contributions under auto-enrolment, they are penalised by losing the employer contribution. The requirement for an employer contribution goes way beyond the nudge philosophy. It also raises employers’ costs and creates pressures that are likely to lead to lower wages at a time when real earnings growth is under pressure.
Remarkably, nearly all assessments of the success of the auto-enrolment scheme focus almost entirely on the proportion of people who do not opt out (it is assumed the higher that is the better). This includes the assessment by the Cabinet Office’s Behavioural Insights Team.
The correct measure of success would compare the number of people who would like to save but whom inertia stops saving and who therefore benefit from auto-enrolment (the higher the better) with the number of people who would rationally prefer not to save but who have not opted out because of inertia (the lower the better). Of course, this is harder to measure, but it would provide information that would have the advantage of being relevant.
It is also worth noting that auto-enrolment imposes burdens on employers and employees in relation to ludicrously low pension contributions. Auto-enrolment applies to anybody who has earnings of more than £10,000 per annum. And this threshold is not adjusted for inflation. Contributions are only payable above the national insurance threshold. This means that those on low earnings might be making pension contributions each year that would earn them an annual pension of about £10 to £15 from the age of 65.
Without dismantling auto-enrolment, reform should attempt to maximise the number of people who have a preference to save but choose not to because of inertia whilst minimising the number of people who remain contributing due to inertia but for whom it is not rational to save. The following three reforms should move the scheme in that direction whilst reducing burdens on employers:
- Increase the minimum earnings threshold that requires auto-enrolment to £15,000 a year and index it to inflation. This would roughly triple the minimum contribution and make it meaningful.
- Reduce the employer’s contribution to zero so that the scheme is a genuine “nudge” scheme.
- Increase the age at which individuals are auto-enrolled to 25.
Number 2 would be difficult to achieve politically. However, 1 and 3 would be a welcome (though small) deregulation of the labour market which, during a time of declining productivity growth, has suffered from more and more regulation in the last 12 years.
SHARE
