White Paper on Universal Credit
Written evidence submitted by Reform
Thank you for your correspondence dated 18 November 2010 in which you invited Reform to submit written evidence to the Work and Pensions Committee’s inquiry on the White Paper on the Universal Credit. This letter gives Reform’s response.
The key points that this letter raises are, in summary:
§
The belief in the likely benefits of smart automation in the White Paper is seriously overstated. Smart automation is not a silver bullet.
§
The proposed Universal Credit is poor value for money with no real evidence it will improve labour market outcomes.
§
Efforts overseas to introduce a single core benefit have amounted to little.
§
Simplification is the right goal but the White Paper takes the wrong approach.
This letter contains material from the United Kingdom, United States and New Zealand that highlights concerns with the depth of the economic analysis of the Universal Credit in the White Paper. The White Paper appears to overlook important institutional features of labour markets, such as the presence of segmented labour markets and discontinuities in labour supply, and key empirical results. The White Paper also gives inadequate consideration to the economic costs of potentially discouraging second earners in households from working.
Given the outlook for the public finances it would be difficult to find any other area of government expenditure where proposals with a total fiscal cost of over £2 billion per annum are being advanced without first undertaking a robust review of the evidence base for reform.
As well as this potential financial cost the time spent on pursuing the introduction of a Universal Credit will come at an opportunity cost and mean that more fruitful areas for welfare reform could go unaddressed.
Background material outlining Reform’s work on this issue is included as an attachment to this letter. This background material includes:
§
A copy of Reform’s recent welfare report, The Money-Go-Round.
§
A copy of the transcripts from Reform’s four post-election conferences on welfare, education, public services and the deficit and health. These transcripts, the Reform commentary on the Coalition’s first hundred days and a summary of the transcripts are contained in the document titled The First Hundred Days.
Smart automation is not a silver bullet
1.
There are practical barriers to the introduction of the Universal Credit. This includes the need to develop a new information technology system.
2.
The proposals in the White Paper are based on a view that "smart automation" would allow changes in circumstances, such as earned income, to be monitored close to real time and the levels of entitlement and abatement to be automatically updated.
3.
This belief in the likely benefit of smart automation is seriously overstated. This is aside from the poor record of attempts at implementing national information technology programmes in the United Kingdom.
4.
Experience shows how difficult it can be to accurately monitor fluctuations in income and adjust levels of assistance. This can be seen in the difficulty in adjusting tax credit entitlement with fluctuations in incomes and household characteristics.
5.
Further, not all important criteria for determining assistance can easily be automated without incurring significant administration and compliance costs. While, for instance, earned income and ages of children could be monitored automatically, criteria such as marital status, the length of time a child resides with a caregiver in a separated household, and hours of work are more difficult to keep track of.
6.
If these criteria are to continue to be used in the assessment of entitlement then the benefits of smart automation will be limited. Failing to use these criteria in the assessment of entitlement would be seen as unfair. This would mean that, for example, in a shared custody arrangement the levels of support paid for children would not reflect the length of time for which the parent was the primary caregiver. Taking such criteria into account with an automated system would require high levels of voluntary compliance.
7.
These limitations of smart automation will become more apparent as families become more diverse and labour markets more flexible.
The Universal Credit is poor value for money with no real evidence it will improve labour market outcomes
8.
The trade-off between objectives for poverty alleviation, financial incentives and fiscal cost is often referred to as an iron triangle. Yet although the iron triangle is a relatively old idea, analysis based on it is limited in important ways.
9.
In many cases the key trade-offs in the design of a welfare policy are not between objectives but within them. For example, improving financial incentives through a more generous earnings’ disregard may improve incentives to work for very few hours of work a week but would discourage full-time and higher wage employment.
10.
This is why international evidence shows that increasing earnings’ disregards is relatively ineffective at increasing labour supply.
11.
Blank and Matseidura (2008) found that, based on the Temporary Assistance to Needy Families reforms in the United States, increasing earnings’ disregards had little effect on the labour supply among single mothers. They noted that an increase in earnings’ disregards could motivate non-workers or low hour workers to increase their work effort, but other people with slightly greater hours of work or higher wage rates might be inclined to reduce their incomes.
12.
When combined with lower rates of abatement, a major effect of more generous earnings’ disregards would be to encourage low wage, low status work. Increasing wages would be discouraged in many cases and the costs of in-work benefits would increase.
13.
It may be argued that encouraging low wage, low status work would be beneficial as this work ("mini jobs") could be a stepping stone towards full-time work. Yet research commissioned by the Joseph Rowntree Foundation has shown that the evidence base for this claim is not strong. Many workers fail to progress from mini jobs to full labour market participation. This could be explained by the segmented nature of the labour market.
14.
The New Zealand coalition government has established an independent Welfare Working Group to develop a menu of practical options to reduce welfare dependency. The Working Group’s November 2010 options paper argued that emphasis should be on improving financial incentives to increase hours of work. This stands in contrast to the emphasis on more generous earnings’ disregards in the White Paper.
15.
Indeed, the approach recommended in the New Zealand Working Group’s report is to improve incentives for full-time work through potentially reducing earnings’ disregards. This makes sense as most people who consider entering work will do so at a point where they earn more than the earnings’ disregard. For these people the earnings’ disregard would be "infra-marginal" and reducing its value would mean that they would be more likely to leap over the tail of benefit abatement. They would thus face stronger incentives for meaningful and full-time work.
16.
This evidence from the United Kingdom, United States and New Zealand highlights concerns with the depth of the economic analysis of the Universal Credit in the White Paper. The White Paper appears to overlook important institutional features of labour markets, such as the presence of segmented labour markets and discontinuities in labour supply, and key empirical results.
17.
The White Paper was, for example, silent on the trade-off between improving incentives to work for different population groups. To illustrate, Table 1 of the White Paper showed that around 300,000 people in work earning below the tax threshold would face an increase in the rate at which they lose income in benefit abatement and taxes. Yet the implications of this, particularly for second earners in families, were not discussed.
18.
While it is important to ensure that more families have at least one adult with a strong connection to the labour market, at least some recognition should be given to the costs of discouraging second earners from working in the labour market (including the potential loss of employment-related skills).
19.
Given the outlook for the public finances it would be difficult to find any other area of government expenditure where proposals with a total fiscal cost of over £2 billion per annum are being advanced without first undertaking a robust review of the evidence base for reform.
Efforts overseas to introduce a single benefit have amounted to little
20.
In New Zealand in the two decades to 2008 the idea of introducing a single core benefit was given serious consideration. These 20 years of experience with proposals for a core benefit have highlighted how slow and difficult designing and implementing a core benefit is. The amount of work involved in fleshing out the details of such a system should not be underestimated.
21.
The failure to adopt a single core benefit in New Zealand has not just reflected political constraints but fundamental problems with the concept. A key problem, particularly for the 2005 proposals, is the constraint that these reforms must have "no losers" (defined in a narrow sense of no family receiving less in transfers (before behavioural changes are taken into account) from government after reform).
22.
This constraint, which has also been adopted by the UK coalition government, means that it is not possible to simplify the welfare system without significantly increasing costs to the government. The fiscal costs of ensuring no losers in the move to a universal system should not be underestimated as this implies that the minimum transfer under the new system needs to be raised to be at least equal to the largest transfer in the old system.
23.
The New Zealand experience also shows how the difficulties in introducing a core benefit cannot be simply assumed to fall as information technology develops. Although technology has advanced, families and labour market outcomes have also become more diverse. Increasing population heterogeneity is a major explanation for the growing complexity of welfare systems.
24.
Unless the role of the welfare system is retrenched to an unreasonable degree then some complexity is necessary in the design of welfare programmes. Different programmes exist for different purposes and employ different units of assessment (e.g., individual or family), periods of assessment (e.g., weekly, monthly or annually) and definitions of income and treatment of assets.
Simplification is the right goal but the White Paper takes the wrong approach
25.
The poor take up of some benefits can partly be attributed to the complexity of the welfare system. Yet while the White Paper is right to highlight the need to reduce complexity the approach taken is the wrong one. Simplification needs to be an ongoing process not a one-off change (big bang).
26.
Reform, following Martin (2009), believes simplifying the system in stages would be a more fruitful approach. This could involve:
·
Firstly, rules for benefit eligibility should be made to conform to each other. For example tax rules should be aligned and benefits should be set on the same periodic basis.
·
Secondly, overlaps between benefits should be gradually eliminated and benefits combined if appropriate.
·
Lastly, the distinction between contributory and non-contributory benefits should be reviewed to allow for further simplification.
27.
Some measures along these lines have already been introduced: new arrangements have been made to simplify benefit payments, improvements have been made to the way customers who move in and out of work receive their benefits and credits, including streamlining the need for officials to provide information to officials. Nevertheless, these initiatives need to be combined with simplification of the rules of programmes if they are intended to be more than mere band-aids.
28.
There is a real risk that time spent on working on the introduction of a Universal Credit will mean that more fruitful areas of reform of the welfare system go unaddressed.
December 2010
|
