Regulatory Impact Statement: Budget 2017 Family Incomes Package

Agency Disclosure Statement#

This Regulatory Impact Statement has been prepared by the Treasury.

It provides analysis of options for a package of tax and transfer changes to:

• improve work incentives;
• improve incomes for those in financial hardship; and
• simplify the tax and transfer system.

The analysis performed includes:

• Modelling of revenue costs;
• Modelling of distributional impacts at the family and individual level;
• Modelling of labour supply impacts;
• Consideration of compliance and administrative impacts; and
• Consideration of fairness impacts and horizontal equity.

The detailed analysis is in a number of papers on tax and transfers, outlined in appendix 1. These papers will be released by the Treasury as part of the Budget 2017 Information Release.

The options analysed were limited to those that could feasibly be included in Budget 2017 and implemented within a year. Options for tax and transfer changes that were not considered feasible, because they were not consistent with the Government's objectives, could not be implemented by Inland Revenue until it has completed its business transformation programme, were not within fiscal parameters, or required more detailed analysis, design or implementation have not been analysed, but still could be considered in the longer term.

A key constraint in the process was the need for Budget secrecy, which limited the extent of consultation on options. However, policy responses to the issues outlined in this statement have been widely debated in the public arena. In addition, the changes are, in general, to rates and thresholds rather than to underlying structure.

Further constraints included the timeframe for analysis. The timeframe for providing information was often extremely short, with the bulk of reporting taking place since late January 2017 and decisions having been taken by Ministers in mid April 2017.

The model used to estimate impacts makes a number of assumptions that may impact the accuracy of results. These assumptions include full uptake of targeted assistance by those eligible and that any individual wishing to supply more labour can do so. In addition, neither the Treasury's model nor the Ministry of Social Development's can capture all elements of the package. Changes to the Accommodation Supplement were modelled separately to the other elements, and the results cannot be combined.

Suzy Morrissey
Acting Manager, Tax Strategy
The Treasury

12 April 2017

Status quo#

Current tax and transfer settings#

Personal income tax

1. Ideally, a tax system should be designed to raise revenue as efficiently as possible while contributing to distributional objectives. New Zealand is generally regarded as having a relatively efficient and ‘low distortion' tax system as it is broad-based with relatively low rates.
2. Personal income tax is levied using a progressive rate structure from 10.5% up to 33%. Table 1 shows the rate structure.

3. Personal taxes are not indexed for inflation or wage growth. Rates and thresholds were last changed in October 2010.

Tax credits

4. The progressive rate structure is supported by social assistance through tax credits delivered via the tax system. These consist of payments made either weekly, fortnightly or in lump sum at the end of the year administered by Inland Revenue. Table 2 shows the suite of credits.

5. Working for Families tax credits provide targeted assistance to increase family income to support families to raise their children and provide incentives to move off benefits and into work. This is achieved through a combination of conditions on entitlement and income thresholds above which entitlement begins to abate. Payment rates are automatically revised to increase the degree of targeting when prices have increased by five percent since the last revision.
6. Some families are entitled to multiple credits (for example, the Family Tax Credit and In-Work Tax Credit). However, only the Family Tax Credit is available to beneficiaries.

Benefits

7. The benefit system has three tiers of assistance:
   • The first tier is the main benefits, providing a basic income for people who are not able to support themselves through paid work, which are income tested (except for the Emergency Benefit and Jobseeker Support Student Hardship which are both income and cash asset tested).
   • The second tier is additional assistance paid to people in particular situations or with on-going costs. For example, low-income people may be eligible for the assistance whether or not they receive a main benefit. This tier of assistance is mostly subject to income testing and may be subject to cash asset testing. This tier of benefits includes the Accommodation Supplement which is discussed further below.
   • The third tier is both income and cash asset tested and provided generally to people in hardship (whether on benefit or not) as one-off grants such as Special Needs Grands or may continue over a relatively short period. Such assistance includes the Temporary Additional Support which is discussed further below.

Housing assistance

8. Further assistance is provided in the form of targeted accommodation subsidies. The Accommodation Supplement is paid to eligible private renters, boarders and homeowners. Maximum levels of assistance (maxima) are set for different areas. As with Working for Families, the supplement is abated as income increases.
9. The maxima and areas were most recently updated in 2006, based on 2003 rents. Unlike Working for Families, the Accommodation Supplement is not indexed.
10. Eligible people in social housing pay an Income-Related Rent which limits the amount of rent that low-income tenants pay to no more than 25 percent of their net income. This rental payment is then topped up by the Income-Related Rent Subsidy paid to the registered housing provider (Housing New Zealand or a Community Housing Provider) which covers the balance between the tenant's rental payment and the market rent of the property. This subsidy abates much more steeply with income than the Accommodation Supplement.
11. Temporary Additional Support is also available for those who cannot meet essential living costs from their income and other resources. While available for a maximum of 13 weeks, clients with ongoing shortfalls may reapply. Temporary Additional Support abates dollar-for-dollar as income increases.

Fiscal strategy#

12. Tax changes must be considered as part of the Government's wider fiscal strategy. The Government has stated the following as shorter-term priorities:
    • maintaining rising OBEGAL surpluses over the forecast period so that cash surpluses are generated and net debt begins to reduce in dollar terms;
    • reducing net debt to around 20 percent of GDP in 2020;
    • if economic and fiscal conditions allow, beginning to reduce income taxes; and
    • using any further fiscal headroom to reduce net debt faster.
13. The fiscal strategy also includes long-term fiscal objectives for the next ten years, including to reduce net debt to within a range of 0 to 20 per cent of GDP. Future operating surpluses and operating revenues need to be consistent with this objective.
14. A family incomes package will impact on the fiscal outlook, including on the Crown's operating balance, revenues, expenses and net debt. As such, it needs to be consistent with the fiscal strategy. Packages considered had a fiscal cost in the range of $1-3 billion per annum, which was expected to be most likely to be consistent with the fiscal strategy.

Problems with current tax and transfer settings#

Work incentives#

15. Personal income tax rates were last changed in 2010. Since that time, median income has increased from $40,000 to $48,000.[1] Incomes across the distribution have also increased.[2]
16. As incomes increase, the marginal tax rates individuals face also increase. This has the effect of weakening work incentives by reducing the rewards of extra work received by the worker. High Marginal Tax Rates (MTRs) can also encourage people to structure their affairs to reduce their tax obligations.
17. In particular, where the median income earner in 2010 faced a MTR of 17.5 percent, they now face 30 percent. New Zealand's labour force participation rate is relatively high. However, if income tax thresholds are not adjusted periodically, work incentives could be diminished over time.
18. The interaction of tax and targeted transfers means some taxpayers face abating assistance. Abatement can push effective marginal tax rates (EMTRs) considerably higher than statutory MTRs. This reduces the rewards of extra work even further.
19. Around 11 percent of individuals face EMTRs higher than 33 percent.[3] Once entitlements are fully abated, EMTRs will return to statutory tax rates. This suggests high EMTRs, and so poor work incentives, will generally affect those on low to middle incomes.
20. Table 3 provides a decomposition of EMTRs for individuals with income between $14,000 and $60,000.

Note: Excludes benefits, student loans repayments and ACC

21. Over time, high MTRs may cause problems for New Zealand in maintaining its tax base, enhancing its productivity and maintaining or improving its living standards.

Reducing financial hardship#

22. The tax and transfer system is highly redistributive and is designed to provide assistance to those in financial hardship. Average incomes at each decile have grown since 2010, as shown in Table 4 below.

Source: Ministry of Social Development: Household Incomes in New Zealand: trends in indicators of inequality and hardship 1982 to 2015

23. However, increases in the cost of housing have in some cases outstripped income growth. As a result, some groups have seen declines in after-housing-costs, or 'residual,' income over the last ten years particularly in some geographic areas.[4]
24. Recipients of the Accommodation Supplement have seen their residual incomes fall on average by eight percent since 2006. Some groups, such as beneficiaries, have seen steeper declines. Around 40 percent (approximately 120,000) of recipients spend more than half of their income on housing costs. This indicates housing-related stress.

Objectives#

25. Changes to personal taxes and transfers involve trade-offs among key objectives of: efficiency/growth, equity, fiscal integrity, ease of compliance/administration, and fiscal cost. There is no perfect or optimal tax and transfer system, and any tax structure will incorporate value judgments. The Government's main objectives for the Budget 2017 package are:
    • improving work incentives; and
    • improving incomes for those in financial hardship.

The Government also has a secondary objective of simplifying the system where possible.

26. To achieve the stated objectives, a set of criteria has been established to assess the options. These are:
    • Impact on work incentives/employment outcomes: the extent to which the option will promote incentives (or remove disincentives) to take up and enter employment, or to increase the number of hours worked, with a focus on helping lower and middle income earners.
    • Impact on residual incomes: Low equivalised[5] residual income is used as the primary indicator of financial hardship. This measures income after housing costs and is adjusted for the size of the household supported by the income. Although not a perfect measure of hardship, the residual income measure recognises that housing costs are a significant proportion of most budgets, and similar quality housing can vary in cost among different areas. It therefore more closely resembles the resources available to a household to purchase other necessities than measures that focus solely on income.
    • Administrative complexity: A secondary consideration is improving ease of administration where there are no or minor impacts on the above criteria and cost.
27. Other key considerations within the context of this package were:
    • how quickly a package could take effect - Ministers expressed a strong preference towards options that could be fully implemented within a year of Budget 2017; and
    • the cost of a package - the fiscal range for the full package was $1-$3 billion per year for the forecast period.

Options and impact analysis#

Options considered#

28. As previously mentioned, options considered for inclusion in Budget 2017 were limited to those that could feasibly be implemented in the short term and were consistent with the fiscal strategy. Non-regulatory responses were not considered feasible and were not analysed further.
29. The broad options identified were:
    • changes to tax thresholds;
    • reprioritising Independent Earner Tax Credit; and
    • increasing targeted assistance.
30. The list of options was narrowed through discussion with Ministers with reference to the objectives to allow more thorough exploration of preferred options.

Tax thresholds

31. Options for adjusting tax thresholds were presented, with a focus on alleviating work disincentives caused by the 30 percent rate for middle-income earners, and improving incentives to enter work. Analysis indicates modest labour supply increases and that at least 1.4 million families would benefit from the adjustments. However, it is recognised that targeted payments would likely have a larger impact on income adequacy.
32. While changes to tax rates could be used to improve work incentives, they are less well targeted than threshold changes. That is, to alleviate pressure on the same target group, rate changes are more costly, with more spill-over benefits for higher income earners, than threshold changes. Rate changes were considered a less effective means of achieving the same objectives as threshold changes, and were not explored further.

Reprioritising Independent Earner Tax Credit

33. The Independent Earner Tax Credit provides up to $520 per annum to around 500,000 individuals earning between $24,000 and $48,000 that do not receive a benefit, Working for Families or superannuation. The credit can be claimed each week or in full at the end of the year. It costs around $220 million per annum.
34. The credit was introduced in 2009 as a means for increasing incentives for participation in the workforce by targeting those just below the full-time minimum wage at the time. However, wage growth since then means the credit is poorly targeted. It is likely improving work incentives could more effectively be achieved by removing the credit and using the savings to make other changes to tax and transfer settings.
35. Analysis indicates marginal impacts on labour supply if the credit were to be removed. While there is a small, positive labour supply impact for recipients, there is an offsetting negative impact as the credit abates, pushing EMTRs up to 30.5 - from 17.5 - percent for those with incomes between $44,000 and $48,000. In addition, nearly two-thirds of recipients receive the credit as a lump sum, indicating a low degree of recipient hardship.
36. Removing the Independent Earner Tax Credit would remove a complex layer of entitlement and abatement from the tax system, simplifying its administration.

Increasing targeted assistance

37. Targeted payments are likely to have greater impact on income adequacy than tax rate or threshold changes for a given cost. However, there are several delivery options for targeting these payments. These are discussed below.
38. Government transfers inherently reduce incentives to work for some recipients. The degree to which those incentives are reduced will vary depending on design; some transfers will provide greater disincentives than others.

Increasing main benefits

39. Increases to main benefits[6] are strongly targeted to those on the very lowest incomes, but not necessarily to those with low residual incomes. However, more-than-modest increases to benefits are likely to reduce work incentives.

Increasing Accommodation Supplement

40. Accommodation Supplement is strongly targeted to those with low residual incomes; the incidence of low residual incomes is nearly five times higher among recipients than non-recipients. The most support is provided to those with the highest housing costs.
41. The average Accommodation Supplement recipient spends more than half their family income on housing, and the proportion of recipients receiving the maxima has increased from 24 to 44 percent since 2006. Housing costs above the maxima, which were last adjusted in 2005, are unsubsidised and borne completely by the recipient. This indicates that increases to the maxima will tend to strongly target those in financial hardship.
42. Although work incentives are likely to be reduced in total, beneficiary recipients stand to gain from moving into employment as support for non-beneficiaries is withdrawn more slowly than for beneficiaries. In addition, increases in the supplement will tend to reduce the use of Temporary Additional Support - which abates more steeply with income - further improving work incentives. The Accommodation Supplement effectively lowers the relative price of employment, and could be expected to promote greater supply of labour than main benefits for a given level of assistance.
43. Although “landlord capture” is a risk, the evidence for this phenomenon is mixed. Rent subsidies are not unique in this respect; landlord capture is a possibility for any targeted transfer.

Increasing or better targeting childcare payments

44. Increasing or better targeting childcare payments may reduce barriers to labour force participation for those with children. While early childhood education (ECE affordability) has been declining since the introduction of 20 Hours ECE in 2007, it still remains more affordable relative to income than it was prior to the introduction of this policy. The cost of ECE does not seem to be a barrier to labour market participation, as ECE enrolment and participation levels have been increasing over recent years

Aligning Family Tax Credit rates

45. This option would raise payment rates for families with children under 16 by increasing rates for children under 16 to align with rates for children 16-18. Alignment would support working families as well as beneficiary families. Working for Families is less strongly targeted to those with low incomes than the Accommodation Supplement. Where 90 percent of Accommodation Supplement recipients have incomes less than $30,000, more than half of Working for Families recipients have incomes greater than $30,000.
46. Unlike the Accommodation Supplement, Family Tax Credit payment rates do not depend on housing costs. While it is not clear that the Family Tax Credit would target those with low residual incomes as well as the Accommodation Supplement, administrative data indicate that larger families, who are more likely to have multiple children under 16, are not correspondingly higher-income and accordingly are at greater risk of financial hardship. Benefits from raising rates for families with children under 16 will tend to accrue to those families and alleviate financial hardship.
47. It is difficult to assess the impact on work incentives in the presence of tax cuts without modelling specific combinations.

Simplifying In-Work Tax Credit formula

48. This option would replace the In-Work Tax Credit formula, which depends on family income and the number of dependent children, with a single rate for all recipients. The total impact depends on where the base payment is set relative to the status quo. To the extent that large families are compensated through raising the base payment there may be a positive labour supply effect for smaller families, although this would reduce targeting of low residual incomes relative to the Family Tax Credit and raise costs. If the payment is not raised, large families would be disadvantaged and their work incentives worsened.

Further analysis#

49. Four options were selected for further analysis:
    • changes to tax thresholds;
    • removing Independent Earner Tax Credit;
    • increasing Accommodation Supplement; and
    • aligning Family Tax Credit rates.
50. No single option considered can satisfactorily advance all of the policy objectives; each presents trade-offs. Removing the Independent Earner Tax Credit will worsen work incentives for those receiving full entitlement while improving them for those facing high EMTRs through abatement. Increasing the Accommodation Supplement and Family Tax Credit rates will alleviate financial hardship but worsen work incentives for recipients. Changes to tax thresholds will improve work incentives but do little to improve incomes for beneficiaries in financial hardship. Accordingly, the final package will comprise a combination of the above options reflecting the Government's preferred balance of the objectives.
51. A number of packages with various combinations of the above were presented to Ministers to allow modelling resource to be prioritised where estimated impacts were most consistent with objectives.
52. Impacts from the changes were estimated using the Treasury's microsimulation model, Taxwell. However, Taxwell is not able to accurately model the changes to the Accommodation Supplement. The labour supply impacts of the tax, Independent Earner Tax Credit and Family tax credit changes were modelled in aggregate using Taxwell, but we were not able to model the labour supply impacts of changes to the Accommodation Supplement. We expect the impact to be moderately negative for recipients.
53. The impacts on residual income of the tax, Independent Earner Tax Credit and Family Tax Credit changes were again modelled in aggregate using Taxwell. Presented alongside are estimates derived from MSD modelling of the changes to the Accommodation Supplement. However, the results from the two models cannot be combined.
54. Analysis was conducted mainly in terms of families rather than individuals. This recognises that family income more accurately reflects the resources available to its members than any member's own income. Note that the term family is used broadly for the purposes of this analysis. It comprises one adult, a partner (if any) and dependent children (if any). For example, five adults sharing a flat are considered five families, and a couple with one dependent child sharing their house with a boarder would be considered two families.
55. As with all models, Taxwell has some limitations that should be noted:
    • Only labour supply is modelled, not the demand side by firms.
    • The model estimates the long run response, not the adjustment through time. In reality, it takes time for workers to negotiate more/fewer hours of work or enter/exit the workforce.
    • There is no modelling of labour supply choices by self-employed people, retirees or their partners, full-time students or people with disabilities.
    • It assumes a fixed population (i.e., does not model migration or fertility rates) and fixed household composition.
56. Table 12 summarises the analysis.

Impacts of final package#

57. Ministers selected Package 6a as providing the preferred balance of objectives. Options for scaling and phasing components of the package were sought to ensure the package would remain within the fiscal envelope. The Accommodation Supplement increase was set to the 40^th percentile of 2016 rents. The package and its variations were subject to further analysis.
58. The overall impacts of the final package are very similar to those presented in Table 11 above and are discussed below.

Labour Supply

59. Hours worked are estimated to reduce in total. However, this result is dominated by the negative labour supply impact of increases to the Accommodation Supplement and Family Tax Credit for recipients. For a much larger number of non-recipients the labour supply impact is positive.

Residual Incomes

60. The tax, Independent Earner Tax Credit and Family Tax Credit changes benefit around 1.3 million families, who on average gain $26 per week. The Accommodation Supplement changes improve residual incomes for around 136,000 families, who on average gain $35 per week. This lifts around 14,000 families out of what the Ministry of Social Development considers is severe housing stress.[9]
61. A small number of families (around 4,500) are estimated to lose around $1.50 per week as a result of the changes. These losses arise from:
    • abatement changes to the Family Tax Credit;
    • increases to the Accommodation Supplement ‘entry threshold' as a result of increases to the Family Tax Credit; and
    • interactions between Temporary Additional Support and the Disability Allowance.

Aggregate Impacts

62. We have not modelled impacts at the macroeconomic level. The overall economic impact will depend primarily on:
    • the relative impacts on GDP of the increased labour supply induced by the tax changes and the decreased labour supply induced by the transfers; and
    • the extent to which GDP increases as a result of reduced influence of tax in resource allocation.
63. Identifying these impacts would require assumptions over the value of output produced through the extra hours worked, the value of output foregone through hours no longer worked, and the value of activities not being performed that will be after the changes are implemented. Those assumptions would be very difficult to produce credibly. Finally, these impacts are unlikely to be immediate and we are unable to allocate them to time periods with confidence.

Consultation#

64. The Ministry of Social Development contributed to the development of this package.
65. Due to the need for Budget secrecy, and the short timeframes involved in developing a family incomes package for Budget 2017, the ability to consult in the usual manner has been severely constrained. However, policy responses to the issues raised have been widely debated in the public sphere.

Conclusions and recommendations#

66. As previously noted, there is no perfect or optimal tax and transfer system. Changes to personal taxes and transfers involve trade-offs among key objectives of efficiency/growth, equity, fiscal integrity, ease of compliance/administration, and fiscal cost.
67. The packages discussed in the preceding pages comprising changes to tax thresholds, the Independent Earner Tax Credit, Family Tax Credit and the Accommodation Supplement, will, relative to the status quo:
    • improve work incentives;
    • improve incomes for those in financial hardship; and
    • simplify the system.
68. Accordingly, the packages are consistent with the stated objectives. However, the direction of the final package is shaped by the relative importance placed on these objectives and is a question of value judgements.

Implementation plan#

69. This package will be announced at Budget 2017. The components of the preferred option will be implemented from 1 April 2018.
70. Inland Revenue has advised that, with additional funding, it can implement the tax, Working for Families and Independent Earner Tax Credit changes. The costs are being worked through. These costs include systems changes and education and marketing to notify those taxpayers, tax credit recipients and third party withholders (for example, employers and banks) affected by the changes.
71. The Ministry of Social Development will require $3.9 million to implement the proposed Budget package. This includes implementing changes to the AS regions and the maxima as well as consequential changes from the proposed changes to the Family Tax Credit and tax thresholds (to ensure that the IT system that administers the welfare system calculates payments and debts correctly). This funding would cover the cost of the systems changes that would be required as well as a project team to manage the implementation, staff training, communications and stakeholder management, changes to operational documents, and post-implementation support.

Monitoring, evaluation and review#

72. There are no plans to specifically and separately monitor, evaluate or review the proposed family incomes package for Budget 2017. The post-implementation phase of the Generic Tax Policy Process will help in identifying and addressing any remedial issues post-Budget.

Appendix 1#