Fiscal impact analysis of 2017 election policies

Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance 1 Fiscal impact analysis of 2017 election policies for the Green Party of Aotearoa New Zealand September 2017 In Confidence

2 Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance All work and services rendered are at the request of, and for the purposes of the client only. Neither Infometrics nor any of its employees accepts any responsibility on any grounds whatsoever, including negligence, to any other person or organisation. While every effort is made by Infometrics to ensure that the information, opinions, and forecasts are accurate and reliable, Infometrics shall not be liable for any adverse consequences of the client s decisions made in reliance of any report provided by Infometrics, nor shall Infometrics be held to have given or implied any warranty as to whether any report provided by Infometrics will assist in the performance of the client s functions.

Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance 3 TABLE OF CONTENTS 1. Introduction... 4 Report Structure... 4 2. Fiscal Impacts... 5 Approach... 5 Fiscal Impact of Green Party policies... 6 3. Fiscal Graphs... 9 4. Behavioural Effects... 11 Fiscal assumptions are sensitive to behaviour changes... 11 Tax threshold changes could result in less (top) tax paid... 11 A capital gains tax could reduce sales, and in turn, revenue... 12 Removing benefit sanctions could increase time and costs... 12

4 Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance 1. INTRODUCTION The Green Party of Aotearoa New Zealand (Green Party) commissioned Infometrics to examine the fiscal impact of their policy proposals in the lead-up to the 2017 General Election. Fiscal costings for this review were provided to Infometrics by the Green Party. Although Infometrics has performed a high-level sense test for the magnitude of these costs, we have not analysed the costs in detail, and have relied on the data provided to us by the Green Party. Costing Green Party policies fall outside the scope of this review. The analysis undertaken by Infometrics analyses the fiscal outlook under different scenarios. It does not attempt to calculate any second-order impacts of Green Party policies. That is, no calculations have been made of the flow-on effects of projected differences to economic and fiscal tracks. Any informed judgement around these second-order impacts requires in-depth analysis, modelling, assumptions and sensitivity testing, all of which fall outside the scope of this project. The economic forecast tracks provided by Treasury s Pre- Election Economic and Fiscal Update (PREFU) Fiscal Strategy Model (FSM) are therefore assumed to remain the same over the forecast period in this report. Infometrics received progressive reports and updates of Green Party policy costings throughout mid-august and early-september 2017. Final costings used in this examination were received on 14 September 2017. Report Structure In reviewing the fiscal costings of the Green Party s 2017 election policies, Infometrics has accounted for the cost of individual policies in accordance with standard fiscal accounting practices. This helps ensure the correct impacts are attributed to revenue, expense, and capital allocations. The different scenarios, and their related fiscal tracks outlined throughout this report, are drawn from: 1. The 2017 Budget Economic and Fiscal Update (BEFU) published by Treasury on 25 May 2017 2. The 2017 Pre-Election Economic and Fiscal Update (PREFU) published by Treasury on 23 August 2017 3. Fiscal tracks from the 2017 PREFU adjusted for Green Party policies This report provides comparisons of various fiscal measurements established through analysis of the above three scenarios. High-level comments around the impact of behavioural effects on selected policies with implications for fiscal revenue or expense tracks are also provided. Behavioural effects are only discussed if there is potential for a large impact on the estimates of the policy costings. Effects are not costed and are provided to broadly inform the Green Party of potential implications.

Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance 5 2. FISCAL IMPACTS Approach Individual Green Party policies have been costed out by the Parliamentary Library or internally by the Green Party for the period Fiscal Year (FY) 2017/18 to FY 2020/21, and provided by the Green Party to Infometrics. All costs provided are assumed to be annual totals with any cumulative effects already accounted for in individual costings. Depreciation has been accounted for, where appropriate. Policies have been broadly categorised by the Green Party as either increased government funds (via new revenue or decreased planned expenses); decreased government funds (via new expenses or a reduction in revenue); or capital changes (an addition or removal of capital spending by the Crown). Infometrics reviewed the policy costing categorisation undertaken by the Green Party and correctly allocated them to different fiscal inputs according to established fiscal accounting processes. In some cases, we have reallocated policies as revenue (by Crown revenue stream), expenses, or capital. The sum of each of the three fiscal inputs have then been run through the PREFU 2017 FSM adjuster to produce comparable fiscal track figures. To ensure comparison on a like-for-like basis, it has been assumed for this project that the economic forecasts in the PREFU 2017 FSM remain constant. Capital injections to the New Zealand Super Fund (NZSF) have been treated differently as per the Fiscal Strategy Model, with both the usual fiscal effects and Super Fund effects accounted for in this analysis. Table 1: High-level fiscal inputs Fiscal Year 2017/18 2018/19 2019/20 2020/21 2018 2019 2020 2021 ($ billions), excluding NZS New Revenue 0.100 2.730 3.550 3.724 Green Party Election Priority Spending 0.119 4.991 7.807 9.585 Net impact -0.019-2.261-4.257-5.861 Existing Budget provisions for new spending 0.000 2.034 3.746 5.495 Impact on OBEGAL -0.227-0.511-0.366 Green Party costings decrease the 2017 PREFU operating balance before gains and losses (OBEGAL) forecasts as seen in Table 3. As a result, net debt is slightly higher than that forecast in PREFU 2017 by the end of 2021.

6 Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance Fiscal Impact of Green Party policies Over the four-year period, Green Party fiscal revenue sits higher than the current PREFU track, and will be at 31.1% of GDP in 2020/21. Expenses are higher than the PREFU track, sitting at 28.9% of GDP in 2020/21. Capital spending sees minor movements, alongside new capital injections to the NZSF in 2017/18 to 2019/20 before resuming the PREFU NZSF capital injections track in 2020/21. Under the Green Party costings scenario, the OBEGAL surplus sits at $6.05b in 2020/21, below the current OBEGAL surplus of $6.44b in 2020/21 forecast at PREFU 2017. Both gross Sovereign-issued debt (GSID) and net debt increase slightly from PREFU tracks, with GSID sitting at 26.5% of GDP in 2020/21. Net debt falls to 19.1% in 2020/21. Crown total net worth increases to 45.0% by 2020/21, slightly below the PREFU track of 45.2%. Table 2: Summary allocations of Green Party policy costings Fiscal Year 2017/ 18 2018/ 19 2019/ 20 2020/ 21 2018 2019 2020 2021 ($ billions) Additional expenses Associated with announced policies 0.123 6.050 8.037 9.372 Operating Allowance 0.600 1.200 1.800 2.400 Less expense savings 0.604 2.259 2.030 2.187 Total new expenses 0.119 4.991 7.807 9.585 Additional revenue New tax revenue 0.100 2.835 3.760 3.988 Non-tax revenue 0.000 0.000 0.000 0.000 Less tax reductions 0.000 0.105 0.210 0.264 Total new revenue 0.100 2.730 3.550 3.724 Net contribution to operating balance -0.019-2.261-4.257-5.861 Capital items Additional assets 0.200 0.479 1.064 1.677 Additional liabilities 0.000 0.000 0.000 0.000

Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance 7 The above fiscal tracks broadly adhere to the Budget Responsibility Rules (BRRs) agreed to by the Labour Party and Green Party in March 2017. An operating surplus is achieved in each forecast year, as well as expenditure kept below 30% of GDP. Net Core Crown Debt falls across the forecast period to 19.1% by 2020/21. The other two BRR targets dealing with long-term investment and taxation are outside fiscal scope. Included in Table 2 is a $600m per annum operating allowance. An operating allowance is the amount of new operating funding available each year, for either new policies or cost increases in existing policies. 1 This operating allowance replaces previously announced operating allowances in the 2017 PREFU. Other cost increases for health and education are already included in forward expenses. This is consistent with the Labour Party fiscal plan, where cost pressures are met within their existing line item funding of health and education, accounting for population and inflation rises. 2 1 Treasury. (2017). Operating Allowance. Treasury. https://www.budget.govt.nz/budget/guide/budgetingpractices/operating-allowance.htm (accessed 13 Sep 2017). 2 Robertson, G. (2017). Joyce must come clean on Health and Education funding. Labour Party. http://www.labour.org.nz/joyce_must_come_clean_on_health_and_education_funding (accessed on 13 Sep 2017).

8 Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance Table 3: Summary fiscal impact of Green party policy costings Fiscal Year 2017/ 18 2018/ 19 2019/ 20 2020/ 21 2018 2019 2020 2021 OBEGAL (Operating Balance before Gains and Losses), $ billion Green Party Costings 3.347 3.298 5.235 6.053 2017 PREFU 2.869 3.515 5.746 6.440 2017 BEFU 2.858 4.051 6.085 7.168 OBEGAL (Operating Balance before Gains and Losses), % of GDP Green Party Costings 1.2% 1.1% 1.7% 1.9% 2017 PREFU 1.0% 1.2% 1.9% 2.0% 2017 BEFU 1.0% 1.4% 2.0% 2.2% Gross Sovereign-issued Debt (GSID), $ billion Green Party Costings 89.411 90.016 90.265 85.572 2017 PREFU 89.689 90.297 89.950 84.757 2017 BEFU 90.657 91.073 90.494 85.465 Gross Sovereign-issued Debt (GSID), % of GDP Green Party Costings 31.6% 30.4% 29.1% 26.5% 2017 PREFU 31.7% 30.4% 29.0% 26.2% 2017 BEFU 32.2% 30.7% 29.0% 26.3% Net Core Crown debt (excluding NZS Fund), $ billion Green Party Costings 61.961 63.433 62.293 61.622 2017 PREFU 62.239 63.714 61.978 60.807 2017 BEFU 64.111 65.744 64.154 62.824 Net Core Crown debt (excluding NZS Fund), % of GDP Green Party Costings 21.9% 21.4% 20.1% 19.1% 2017 PREFU 22.0% 21.5% 20.0% 18.8% 2017 BEFU 22.8% 22.1% 20.6% 19.3% Crown Total Net Worth, $ billion Green Party Costings 120.828 127.178 135.818 145.554 2017 PREFU 120.350 126.918 136.069 146.192 2017 BEFU 111.442 118.390 127.829 138.671 Crown Total Net Worth, % of GDP Green Party Costings 42.8% 42.9% 43.7% 45.0% 2017 PREFU 42.6% 42.8% 43.8% 45.2% 2017 BEFU 39.5% 39.9% 41.0% 42.7%

Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance 9 3. FISCAL GRAPHS Graph 1: Marginally lower OBEGAL surplus in 2021 OBEGAL Operating Balance before Gains and Losses, $ billions 10 Green Party Costings 2017 PREFU 5 0-5 -10-15 -20 2007 2009 2011 2013 2015 2017 2019 2021 Graph 2: Net Core Crown debt decreases to $62b in 2021 19.1% of GDP Net Core Crown debt Excluding NZS Fund, as a % of GDP 30.0% Green Party Costings 2017 PREFU 25.0% 20.0% 15.0% 10.0% 5.0% 0.0% 2007 2009 2011 2013 2015 2017 2019 2021 Graph 3: Crown revenue increases to over 31.1% by 2020/21 Core Crown Revenue As a % of GDP 34.0% 33.0% 32.0% 31.0% 30.0% 29.0% 28.0% 27.0% 26.0% Green Party Costings 2017 PREFU 25.0% 2007 2009 2011 2015 2017 2019 2021

10 Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance Graph 4: Crown expenses remain below 30% of GDP Core Crown Expenses As a % of GDP 35.0% 34.0% 33.0% 32.0% 31.0% 30.0% 29.0% 28.0% 27.0% 26.0% Green Party Costings 2017 PREFU 25.0% 2007 2009 2011 2015 2017 2019 2021 Graph 5: 100 90 80 70 60 50 40 30 20 10 GSID declines to $86b by 2021, 26.5% of GDP Gross Sovereign-issued debt GSID, $ billions Green Party Costings 2017 PREFU 0 2007 2009 2011 2013 2015 2017 2019 2021 Graph 6: Net Crown total worth increases to $146b to be at 45.0% of GDP in 2021 Crown total net worth $ billions 160 140 120 100 80 60 40 20 Green Party Costings 2017 PREFU 0 2007 2009 2011 2013 2015 2017 2019 2021

Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance 11 4. BEHAVIOURAL EFFECTS Fiscal assumptions are sensitive to behaviour changes The costings for various policies included in this fiscal impact analysis provide a robust estimate of the initial or theoretical fiscal implications of a policy. They are however, likely to over or understate the effects for some proposals due to subsequent changes in behaviour. Given the importance of revenue and expense considerations for fiscal analysis, it is important to briefly consider high level additional behavioural effects on some costings, and the sensitivities of fiscal costs. This section briefly discusses policies where behavioural shifts by taxpayers could change Crown revenue or expense expectations. The effects outlined are not quantified, but are important to consider given their implications for fiscal policy setting and revenue/expense expectations. Tax threshold changes could result in less (top) tax paid Part of the Mending the Safety Net policy announced by the Green Party in July 2017 will reform the New Zealand tax system. Under the policy, the bottom tax rate on income up to $14,000 will be reduced from 10.5% to 9.0%. At the same time, a new top tax rate of 40% will be implemented on income over $150,000. Previous changes in the New Zealand tax system have seen shifts in taxpayers behaviour. Tax changes in the 2000 tax year saw a new 39% tax rate imposed on income above $60,000. Similarly, tax rate changes between 2008 and 2011 saw shifting tax rates and thresholds, culminating in the current system. 3 As Graph 7 outlines, these changes had a stark impact on the amount of declared income reported by taxpayers. Graph 7 Personal tax distribution $000 income brackets (horizontal), $m of taxable income (vertical) 3000 2500 2000 1999 2001 2012 1500 1000 500 0 5 15 25 35 45 55 65 75 85 95 105 115 125 135 145 3 The current personal tax system is progressive with the following rates: Threshold $0 $14k $14k $48k $48k $70k $70k + Tax rate 10.5% 17.5% 30.0% 33.0%

12 Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance The large spikes in the graph correlate with the changes in tax policy, resulting in a behavioural response by taxpayers to minimise their tax burden. The other spikes below $25,000 are due to transfer payments including welfare benefits and superannuation. As Infometrics outlined in its Fiscal impact of Green Party Policies prior to the 2014 election for a similar policy (increasing the top tax rate to 40% for marginal income over $140,000), we are concerned that that estimates made makes no allowance for behavioural responses to the tax change. The type of impact is demonstrated in Graph 7, which presents the way that declared income evolved following the introduction of a 39% tax for incomes over $60,000 in the early 2000s. Most of the more immediate post-1999 income growth (2001 line) occurs, at income levels below $60,000 while longer term behavioural changes are harder to pick up due to changes to the top tax bracket between 2008 and 2011. Of note is the development of an income spike precisely at $60,000 in the years immediately following the tax change in 1999. In subsequent years, this has shifted to $70,000 due to changes in the highest income tax bracket threshold. Within their policy, the Green Party has moved to limit the extent of possible tax avoidance, through harmonising the trust tax rate with the new top tax rate, alongside increased tax enforcement spending. If taxpayers consider the impact large enough, they will look to shift their taxable earnings into other areas including through companies to avoid personal income tax an economically rational and perfectly legal option. Given this, and the marked changes in tax distribution after 2000 and 2008, it is important to recognise a potential decline in the tax base as a result in any changes in tax policy. The revenue collected from this change should therefore be viewed as a high-end estimate. A capital gains tax could reduce sales, and in turn, revenue The implementation of a capital gains tax (CGT) excluding the family home would see a reduction in house sales over time as demand for housing dropped. This effect, due to an increase in the implicit price of a house (compared to a house sale without capital gains tax), would cause house sales to fall somewhat over time. We therefore advise caution around the revenue implications of a capital gains tax, as reduced house sales will cause a reduction in tax receipts from the CGT. 4 Depending on the elasticity of the market, a CGT may dampen the market below where revenue levels are estimated. This could result in a reduced revenue intake for the Crown, affecting fiscal balances. Again, the Green Party tax revenue estimates should therefore be viewed as high-end estimates. Removing benefit sanctions could increase time and costs Changes to social support through the Green Party s Mending the Safety Net policy include a provision to remove all obligations and sanctions that create an excessive burden on people. Although the direct impact of reducing sanctions and obligations appears as a net saving (without sanctions, each year the Government is expecting to pay out $8.8m more, but save $36.4m from enforcement costs), economic theory suggests that there may be a larger indirect effect. Reduced sanctions and obligations could 4 Adam Smith Institute. (2010). Estimated revenue losses from Capital Gains Tax increases Briefing Paper. Adam Smith Institute. https://www.adamsmith.org/research/estimated-revenue-losses-from-cgt-increases (accessed 6 Sep 2017).

Fiscal impact analysis of Green Party policies September 2017 $600m Operating Allowance 13 change behavioural responses from those either already accessing benefits or intending/needing to access benefits in the future. Government transfer payments could increase if the reduction in obligations and sanctions changes recipients (or potential recipients) responses to benefit uptake. This could be borne out by an increase in both the amount and length of time claimed for benefit transfers, pushing up the fiscal cost of this policy. Overall, a range of academic literature shows that imposing benefit sanctions can lead to an increase in transitions to work and a consequential reduction in benefits paid out. 5 Given this, it is assumed the reverse is also true where removing benefit sanctions could over time increase the length and amount paid out. We therefore believe the net savings estimates of this policy should be viewed as higher-range estimates. This is highly dependent on changes in preferences and responses by those receiving or potentially receiving benefit transfers. 5 Boockmann, B., Thomsen, S. L., Walter, T. (2009). Intensifying the use of benefit sanctions - An effective tool to shorten welfare receipt and speed up transitions to employment? IAW-Diskussionspapiere, No. 56