Income Forecast Group (“IFG”) Report on the Revised Forecast of States Income from Taxation and Duty for spring 2019

STATES OF JERSEY

r

INCOME FORECAST GROUP ("IFG") REPORT ON THE REVISED FORECAST

OF STATES INCOME FROM TAXATION AND DUTY FOR SPRING 2019

Presented to the States on 15th August 2019 by the Minister for Treasury and Resources

STATES GREFFE

2019  R.107

Ite

Income Forecast Group ("IFG") Report on the Revised Forecast of States Income from Taxation and Duty for spring 2019

1. Purpose

1. To provide the revised forecast of States income from taxation and duty as at spring 2019. The forecasts reflect:
   1. the Fiscal Policy Panel's ("FPP") economic assumptions of March 2019[1] and other related economic data for Jersey;
   2. general revenues outturn for 2018;
   3. initial information on general revenues for Quarter One 2019;
   4. information from Revenue Jersey for Year of Assessment ("YOA") 2018;
   5. intelligence from the IFG affecting future forecasts; and
   6. the latest accounting treatment regarding the recognition of personal income tax
   3. Background

1. The agreed Terms of Reference for the IFG (see Appendix A) requires that at least two forecasts are produced each year. This forecast will provide important background to the development of the Government Plan 2020-23.
2. The forecast period covers the years 2019-2023 in order to cover the whole of the Government Plan period.
3. The next scheduled revision of income forecasts by the IFG will be an update carried out in August/September 2019.
4. The IFG forecasts cover a large proportion of States income, but exclude other States income from Island Wide Rate, dividends and returns from States Investments, and other fees which are produced and reported separately.

3. Summary of Revised Forecasts of States Income from Taxation and Duty

1. The revised forecast has been prepared by officers and reviewed by the IFG and is considered as the central scenario within a range for 2019-2023.

Uncertainties and range of forecasts

2. The FPP's Advice for the 2020-2023 Government Plan set out a number of risks around Jersey's economy and fiscal position over the short, medium and long term. At a high level, these risks included Brexit, risks to financial services, challenges around productivity and demographic pressures.
3. The IFG consider that each of these risks result in considerable uncertainty in income forecasts

– with the balance to the downside, but also with some upside potential. Productivity and demographic pressures are likely to be more long-term risks, but Brexit and risks to financial services have more potential to cause uncertainty over the forecast period to 2023. However, the IFG notes a modest improvement in the FPP's longer-term forecasts for productivity growth in financial services.

4. A number of specific factors could affect the income tax forecast in particular:
   1. Variations in employment numbers/earnings both in level and distribution;
   2. Impact of unforeseen changes in interest rates on investment incomes and on banking profits;
   3. Uncertainty in yield forecasts due to a large number of component factors;
   4. Uncertainty regarding the amount of shareholder income arising in a particular year;
   5. Impact of unforeseen events and potential regulatory impacts on the financial services sector;
   6. The possibility of the UK leaving the European Union without a deal, potentially resulting in a further depreciation of sterling (and potential spike in inflation) and knock-on effects from any resulting downturn in the UK economy; and
   7. Broader risks to the global economy, including potential trade disruption.
5. The IFG continues to emphasise the need to include flexibility within future financial planning given the risks above, and this is particularly reflected by the range around the income forecast.

Income forecasts for spring 2019

6. This forecast from the IFG follows the forecast update prepared in September 2018 to inform the draft Budget 2019. Since that time the IFG has received further information based on:
   1. Updated FPP economic assumptions, as contained in their report "Advice for the 2020- 23 Government Plan" published in March 2019. Importantly the FPP has now made specific forecasts for 2020-2022, replacing the trend assumptions that were previously used for these years.
   2. The financial outturn for 2018 for all States income and detailed tax data for YOA 2017. Financial outturn for 2018 in several income areas was in excess of the September 2018 forecasts, and these variances are explained later in this report. Consideration has then been given to the implications for the forward forecasts.
   3. Initial data on personal income tax and, notably, corporate income tax for YOA 2018 from Revenue Jersey. Initial information for some other income areas for Quarter One 2019.
   4. Social Security Contributions data for 2018.
   5. The income tax forecasting model has been updated to reflect the latest FPP economic assumptions. The personal income tax model was adapted in 2017 to incorporate the recommendations from the external review by Oxera, which were accepted by the IFG.
7. A summary of the revised forecasts for taxation and duty for spring 2019 are shown in Figure 1, together with a comparison with the Budget 2019 (September 2018) forecast.

Figure 1: Revised forecast (spring 2019) for States income from taxation and duty

Summary of variations in 2018 outturn compared to Budget 2019 (September 2018) forecast

8. Actual revenue from taxation and duty for 2018 was £18 million more than the forecast prepared in September 2018. Fuller explanations are provided later in this report, but the key aspects to note are:

1. Personal income tax broadly turned-out on forecast, with actual personal income tax revenue exceeding the forecast by less than 1 per cent.
2. Corporate income tax performed more strongly than forecast with an additional £4.6 million of income being received in respect of YOA 2017 close to the surcharge deadline (i.e. November/December 2018) and £2.5 million being assessed in respect of earlier YOAs.
3. "Normal" GST (i.e. on Island spend) performed more strongly than forecast, particularly when adjusted for a single very large transaction, with an additional £1.9 million of income being received during 2018.
4. Work undertaken in relation to taxation debts has resulted in additional income being booked to the States income statement in 2018. This is primarily a "one-off" benefit and unlikely to occur in future years and hence the forecast reflects a provision for bad debts based on historical trends.

Movements in forecasts since the Budget 2019 (September 2018) forecast

9. The revised forecast shows a number of variations compared to the Budget 2019 (September 2018) forecast and which in overall terms produce an improvement in 2019 and all forecast years to 2023.
10. The main forecast variations, which are described and explained in more detail later in this report, can be summarised as follows:

Personal Income Tax – accounting adjustment

1. Consistent with international accounting standards, the Treasury has determined that all personal income tax should be recognised in the year in which the income arises (i.e. the personal income tax arising on income earned in the 2019 YOA should be recognised in the States accounts for 2019).

2. The high level impact of this accounting adjustment is to accelerate the recognition of personal income tax by one financial year (i.e. personal income tax relating to the 2019 YOA was previously recognised in the States accounts for 2020 but will now be recognised in the States accounts for 2019).
3. Since 2015 the personal income tax arising on those taxpayers who pay their tax by way of ITIS on a current year basis has already been accelerated by one financial year – this acceleration has been achieved by way of the "current year adjustment".
4. This income forecast, and future income forecasts, will used the personal income tax model to forecast the personal income tax due in each YOA/financial year without the need to include a "current year adjustment".
5. This change in accounting treatment increases the personal income tax forecast as follows: 2019: £12 million; 2020: £10 million; 2021: £12 million; 2022 £12 million; 2023: £13 million.

Personal Income Tax – underlying forecast from the personal income tax model

6. The revised underlying forecast from the personal income tax model is broadly consistent with the September 2018 forecast, however there are a number of changes within the forecast which broadly net to zero.
7. The outturn data for the 2017 YOA results in approximately £1 million less tax in the 2018 YOA, with the impact of this rising to £4 million by the 2022 YOA.
8. However the latest employment income data from ITIS for 2018 results in a £3 million increase to the forecast in each year.
9. The combination of new economic assumptions from the FPP and updated estimations of their relationship with taxable income results in a net decrease of -£2 million in the forecast from the 2019 YOA onward.

Corporate Income Tax

10. The improved outturn for 2018 and intelligence from the YOA 2018 appeals result in an improved base of £98 million; this improvement in the base broadly reoccurs across the forecast period.
11. The small downward revision in financial services profits growth in 2019 has resulted in slower growth in corporate income tax across the forecast period.

GST, ISE Fees and Import GST

12. On Island GST revenues are forecast to increase in excess of the September 2018 forecast due to the improved base from 2018 and the updated FPP assumption that there will be real GVA growth in the years 2020-23, resulting in additional growth in GST revenues from 2020 onwards.

Impôts Duties

13. The duty increases in Budget 2019 and the updated FPP assumptions increase the September 2018 forecast by circa. £1.5 million in 2019 onwards.

Stamp Duty

14. Stamp duty is forecast to increase in excess of the September 2018 forecast as a result of the 2018 outturn in House Price s and Housing Turnover and the updated FPP assumptions relating to the housing market.

Range of forecasts

11. The range around the central forecast has been extended, along with the central forecast, to 2023 and has been updated and re-modelled to reflect the revised range of economic assumptions (see Figures 2 and 3).

Figure 2: Range of income forecasts from taxation and duty (spring 2019)

Figure 3: Range of income forecasts from taxation and duty (spring 2019)

4. Economic Assumptions

1. The economic assumptions have been updated by the FPP based on the latest local and international developments to March 2019.
2. The main variations to the economic assumptions used in September 2018 are summarised in Section 5 of this report.
3. The central assumptions on which the spring 2019 forecasts are based are shown in Figure 4.
4. The IFG have considered the economic assumptions from the FPP and have agreed that these assumptions be used as the basis for the income forecast modelling.

Figure 4 – FPP Revised Economic Assumptions as at March 2019 (% change except where stated)

Trend 2017 2018 2019 2020 2021 2022 2023+

Real GVA 0.4 1.6 1.0 1.4 1.3 0.8 0.6 RPI 3.1 3.9 3.1 2.6 2.5 2.6 2.6 RPIY 3.2 3.5 2.9 2.5 2.4 2.5 2.5 Nominal GVA 3.6 5.1 3.9 3.9 3.7 3.3 3.1 GOS (including rental) -0.7 5.8 3.2 3.6 3.5 3.2 3.2

Financial services profits -6.6 4.0 2.0 2.9 3.1 3.3 3.4 Compensation of employees 7.6 4.5 4.5 4.2 3.9 3.4 3.1 Employment 2.3 1.0 0.5 0.7 0.8 0.5 0.4 Average earnings 2.6 3.5 4.0 3.5 3.2 2.9 2.7 Interest rates (%) 0.3 0.6 0.8 0.9 1.0 1.1 1.1* House prices 2.9 7.1 6.3 5.4 4.5 3.6 2.7 Housing transactions 6.7 7.2 3.0 4.0 3.2 2.3 1.5 *Interest rate assumption for 2023 only

Figure 5 – Variations in FPP Economic Assumptions between August 2018 and March 2019 Trend

2017 2018 2019 2020 2021 2022 2023+

Real GVA 0.3 0.0 -0.5 1.4 1.3 0.8 0.6

RPI 0.0 -0.3 -0.3 -0.7 -0.8 -0.7 -0.7 RPIY 0.0 -0.3 -0.1 -0.5 -0.6 -0.5 -0.5 Nominal GVA 0.3 -0.3 -0.6 0.9 0.7 0.3 0.1 GOS (including rental) 0.1 0.8 -0.8 0.6 0.5 0.2 0.2

Financial services profits -0.4 -0.3 -1.5 -0.1 0.1 0.2 0.4 Compensation of employees 0.7 -1.3 -0.4 1.2 0.9 0.4 0.1 Employment 0.0 -0.5 -0.5 0.7 0.8 0.5 0.4 Average earnings 0.0 -0.7 0.1 0.5 0.2 -0.1 -0.3 Interest rates (%) 0.0 0.0 0.0 -0.1 0.0 0.1 0.1* House prices -0.1 2.1 2.3 2.4 1.5 0.6 -0.3 Housing transactions N/A N/A N/A N/A N/A N/A N/A

5. Income Tax Forecast 2019-2023 (spring 2019) Introduction

1. This section summarises the income tax forecast (both personal and corporate) at spring 2019. The revised forecast is based on:
   1. Updated FPP economic assumptions for 2020-2023;
   2. Outturn income tax data for YOA 2017; and
   3. Available information from Revenue Jersey on both personal and corporate income tax for YOA 2018.
2. The rest of this section is set out as follows:

• Overview of the FPP's revised economic assumptions that have been used to update the income tax forecast, and the reasons for any changes.
• Information from Revenue Jersey on the most recent outturn data.
• Revised output from the income tax forecast model on the basis of new data and new economic assumptions.

FPP revised economic assumptions

3. The FPP's updated economic assumptions (Figure 4) have been used in the tax model to update the income tax forecast. The economic assumptions were published in March 2019.
4. When compared to the previous (August 2018) assumptions (see Figure 5), the main changes are:
   1. Lower forecasts for inflation, reflecting new analysis that indicates Jersey's RPIX broadly tracks CPI inflation in the UK, though a little faster on average.
   2. Slower growth in employment in 2018 and 2019, due to the relatively low June 2018 outturn and due to Brexit uncertainty. A small bounce back is expected in the following years.
   3. Higher forecasts for earnings due to a tightening labour market; recent high inflation; and the impact of significant increases in the minimum wage.
   4. Slower growth in profits in the short term, reflecting Brexit uncertainties.
5. The changes in these assumptions have had knock-on effects on the nominal and real economic growth (GVA) assumptions, with real growth expected to be around ½ per cent slower in 2019 but higher for the remainder of the forecast.
6. The FPP has now made specific forecasts for 2020-2022, replacing the trend assumptions that were previously used for these years. There has also been an increase to the FPP's estimate of the future trend rate of growth in real GVA, from 0 per cent to 0.6 per cent. However, the trend rate of growth in nominal GVA has increased only slightly from 3.0 per cent to 3.1 per cent.

Personal income tax

Personal Income Tax - Outturn data for YOA 2017

7. Personal income tax for YOA 2017 was £1 million (0.25 per cent) lower than forecast. Taxable income was around £8m higher than forecast but the yield was 0.1 percentage points lower than forecast.
8. Within taxable income, earned income was generally lower than forecast, but unearned income was generally higher than forecast. The main individual income lines are considered in more detail below.

9. Employment and other earned income grew by 3.6 per cent, compared to a forecast growth of
   4. per cent based on the latest data at the time from ITIS. The long-term (15-year) growth rate is around 3.8 per cent, though this was significantly lower during the 2009-2013 period.
10. Pension income was £17 million lower than forecast. Total pension income fell slightly compared to YOA 2016, but this was due to some one-off pension income for taxpayers on the High Value Resident scheme (HVRs) in YOA 2016, which fell out in YOA 2017. Excluding the £18 million fall in HVR pension income, the remainder of pension income grew by 4.8 per cent.
11. It is noted that during some recent years (in particular YOA 2014 and YOA 2016) growth was slower due to an exercise to reduce the number of tax returns sent out to taxpayer entities who were in receipt of relatively small amounts of (pension) income that did not exceed their allowances.
12. Personal business profits were £3 million above forecast. Annual changes in personal business income can be volatile year-to-year, though there is a strong upward trend in the last four years. Given the recent strong growth, IFG forecasts growth to continue at the five-year average (3½ per cent).
13. Income from distributions was £7 million lower than forecast. This income is volatile from year- to-year due to the number of large distributions in any one year, many of which appear to be one-off distributions. Due to this volatility, the IFG continues to forecast this income stream to remain flat in real terms.
14. Property income (including Jersey property, furnished letting and income from lodgers) returned to strong growth in YOA 2017 after a slowdown in YOA 2016, and was £3 million above forecast. The previous forecast used a more prudent rate of growth (a 15yr average rather than a 10yr average) to reflect the potential that YOA16 reflected an easing of the recent strong rate of growth. However, as this easing appears to be temporary, IFG has reverted to using the 10yr average.
15. Other unearned income (including bank interest and dividends) was £26 million above forecast. However, this reflects significant dividend income for HVR taxpayers. Excluding HVRs, this income has fallen over the last two years. In particular, there have been significant recent falls in income from bank interest, from £43 million in YOA 2013 to £16 million in YOA 2017.

Personal Income Tax – Provisional data for YOA 2018

16. ITIS data for 2018 show tax paid increasing by 7.1 per cent and taxable employment income increasing by 5.3 per cent. IFG has previously found that initial estimates from ITIS are often subject to significant revision – though this generally increases as further ITIS returns are received from employers. The model forecasts employment and other earned income to grow by 4.7 per cent – which is lower than the 5.3 per cent suggested by the latest ITIS data.
17. Therefore, this forecast uses a 5.3 per cent growth rate for income from employment and other earned income, on the basis that Revenue Jersey believe this is the lowest growth rate likely to be seen from ITIS once all returns are received. If the 4.7 per cent from the model had been used, this would have reduced the forecast by around £3 million per year.
18. Looking at ITIS data covering the earlier part of 2018, which are likely to represent a more complete sample, there is potential for a higher rate of growth. Both the first half of the year and the first three quarters have seen 5.7 per cent growth compared to the same period in 2017. Using a 5.7 per cent growth in the forecast would result in an additional £2 million of income tax per year.

5.19. Some additional analysis has been carried out to determine the source of the strong growth in ITIS income in 2018, and particularly whether this has been driven by deferred payments to public sector employees. This analysis shows that while the public sector did see strong growth, this was not the main driver as financial services saw strong growth, as did private non-finance

– in particular construction. It is understood that the deferred and one-off payments in the public sector amount to around £5 million and therefore this is not a significant element of the total employment income (£2.2 billion).

5.20. On this basis the forecast for 2018 employment income is taken as the new base, rather that assuming that significant elements of this represent one-off payments.

Updated Personal Income Tax Forecast

21. The personal income tax forecast has not changed significantly since the forecast prepared for Budget 2019, with a £2m downward revision by the end of the forecast period.

Figure 6: Changes to personal income tax forecast since Budget 2019 (September 2018)  

New data

22. The outturn data for YOA 2017 result in approximately £1 million less tax in YOA 2018, with the impact of this rising to £4 million by YOA 2022.
23. The latest data from ITIS for YOA 2018 result in an increase in a £3 million increase to the forecast in each year.

New assumptions

24. The combination of new economic assumptions and updated estimations of their relationship with taxable income results in a net decrease in the forecast from YOA 2019 onward:

• The changes to employment and earnings assumptions reduces the forecast by around £1 million in the initial two years but is positive thereafter, peaking at £6 million in YOA 2022. These assumptions primarily affect forecasts for employment income and pension income, but also have an impact on the forecast for tax allowances.
• The lower forecasts for inflation result in slower increases in allowances and therefore increase the amount of tax by around £1 million in YOA 2021 and £2 million in YOA 2022.
• The lower forecast for financial services profits and interest rates each result in a reduction of around £½ million by the end of the forecast.

25. Because of the new outturn data, the equations used to estimate investment income, employment income and pension income have been re-estimated, as have the average growth rates used to forecast personal business profits and rental income. This results in an increase to the forecast of £1 million in YOA 2018, but a decrease of £2 million in YOA 2019, £4 million in YOA 2020, £6 million in YOA 2021 and £8 million in YOA 2022. The reduction is primarily because the coefficients from the Oxera equations have fallen, i.e. the responsiveness of taxable income to change in the economic variables has weakened. There are two main reasons for this:
    1. The income of HVR taxpayers has been excluded from the equations and this has been forecast separately (see below). This removes the effect of large increases in HVR income (driven by both value and volume) and therefore reduces the coefficients.
    2. The variables used to forecast employment income (primarily Compensation of Employees ("CoE")) and unearned income (Bank Rate) both saw significant proportionate increases in 2017 but this did not feed into quite such strong growth in taxable income. To partially account for this in the employment equation, an additional variable has been added to also include the previous year's CoE for financial services[2].
26. As income and tax from HVRs is no longer included in the main model, a separate forecast needs to be made of this. IFG forecasts tax from HVRs to grow at its five-year average growth rate (16 per cent). This reflects the potential for further ongoing increases in the average tax paid (due to the new regime being in place from YOA 2018) and steady increases in numbers of HVRs.

Policy changes

27. There are no significant changes to the personal income tax forecast as a result of the 2019 Budget (the impact of the changes to non-residents relief was estimated at £100k per year). Two policy changes were already reflected in the Budget 2019 forecast and remain in this forecast:

• A reduction in personal tax of around £1 million from YOA 2019 onward as a result of reduced distributions which may result from the extension of a positive rate of income tax to large corporate retailers; and
• An increase of around £3½ million due to the decision to remove the higher child allowance from YOA 2019 as part of the new higher education funding proposal.

Corporate income tax Outturn data for YOA 2017

28. Corporate income tax was around £5 million higher than forecast for YOA 2017. The previous forecast of £92 million was based on in-year information from Revenue Jersey, which at the time was considered a baseline position with the potential for a slight improvement. The increase from the in-year position was split relatively evenly between financial services and property companies.
29. Year-on-year, corporate income tax increased by £5 million, with £4 million of this from financial institutions.

New data

30. The latest data from Revenue Jersey show that, as a baseline figure, a minimum of £106 million is expected for YOA 2018. Analysis of past trends, however, suggest that this figure is likely to continue to grow as more information becomes available during 2019. At this time, there is uncertainty over the amount of growth that will occur before (a) the surcharge deadline in December 2019, and (b) the completion of the assessing cycle for YOA 2018. However, past trends  indicate  that  the  growth  of  corporate  income  tax  in  line  with  the  FPP  economic assumptions  (i.e.  forecast  corporate  income  tax  of  around  £110  million  in  2019)  is  not unreasonable.

Changes to Corporate Income Tax Forecast

31. The forecast for corporate income tax has increased over the forecast period, when compared to the previous (Budget 2019) forecast.
32. Corporate income tax for YOA 2017 was £98 million - approximately £5 million higher than the Budget 2019 forecast. This has been built into the base, in the absence of any significant evidence that this was due to one-off factors. The IFG previously included additional income from improvements and increases in compliance activities, which is now considered to be achieved and therefore included in the base.
33. From 2019 onward, the forecast assumes that corporate income tax grows in line with the FPP economic assumption for financial services profit growth. The initial information from Revenue Jersey for 2019 (based on YOA 2018) appear to be in line with this forecast. The downward revision to the forecast growth in financial services profits for 2019 results in a lower forecast from 2020 onward.
34. The 2018 Budget extended the reach of corporate income tax to include additional financial services firms and large corporate retailers. Based on the estimate in the 2018 Budget, this adds around £9m to the forecast from 2019 onward. Revenue Jersey has confirmed that this remains their best estimate based on the information currently available to them.

New forecast

35. The net impact of the above is to increase the forecast by £5m in 2019 and by around £3 million

- £4 million for the remainder of the forecast period.

Figure 9 – Changes to corporate income tax forecast since Budget 2019 (September 2018)

Revised Income Tax Forecast 2019-2023 (spring 2019)

36. The net impact of all of the above elements is to increase the forecast. Figure 10 – Revised Income Tax Forecast 2019-2023

Income Tax Forecast Range 2019-2023

37. The IFG previously agreed a forecast range starting at +/-2 per cent in the first year of the forecast, rising to +/-9 per cent by the fifth and final year of the forecast, with this continuing to be used in respect of corporate tax. However, the change in accounting treatment to accelerate the recognition of personal tax into the year in which the income arises results in the need to model an additional YOA, which correspondingly requires an increase to the range around the personal tax forecast to 11% in that fifth and final YOA modelled.

Figure 11: Revised Forecast range 2019-2023

6. GST Forecast 2019-2023 (spring 2019) Introduction

1. There are three components to the GST forecast:

• "Normal" GST (i.e. spend on Island)
• Import GST; and
• International Services Entity ("ISE") fees

"Normal" GST on purchases on Island

2. Goods & Services Tax ("GST") was introduced in 2008 and is collected by Revenue Jersey. GST is collected from purchases of goods and services on the Island. Initially introduced at 3 per cent, the GST rate was increased to 5 per cent in 2011.
3. The IFG considered as part of its draft MTFP 2016-2019 (June 2015) forecasts, changes to the forecast modelling of GST. The previous assumptions to increase GST forecasts by RPI were replaced by assumptions reflecting information on general trends in GST relative to the overall economic situation.
4. Consideration has also been given to trends by individual market sector but there were no obvious correlations identified that would improve the forward forecasts.
5. The IFG has agreed to review the modelling assumptions applied in the GST forecast in advance of the completion of the next full forecast (i.e. spring 2020), to determine whether the assumptions can be improved.
6. The outturn for GST on Island for 2018 was higher than forecast in September 2018. Other than simple price inflation there are no known significant factors influencing the year-on-year increase and at this time nothing is expected in 2019 or beyond that might affect the IFG's current forecast.
7. However, the latest FPP economic assumptions indicate that they now predict real economic growth in all years of the forecast period (i.e. for all years up to and including 2023), whereas previously the assumption had been that no real economic growth would occur from 2020 onwards. This revision, which increases normal GST revenues from 2020 onwards, has been factored into the latest forecast which is shown in Figure 12 below.

GST on imports

8. Import GST outturn for 2018 was £5.82 million. This was £0.46 million (8.6 per cent) more than the in-year forecast (updated September 2018) and £0.87 million (17.5 per cent) more than outturn for 2017.
9. Import GST is susceptible to year-on-year volatility due to the impact of the importation of very high value goods. A single, very high value transaction has resulted in £425k of import GST being paid in 2018. If the import GST on this transaction was removed, the outturn for 2018 would have been £5.39 million (which is almost exactly equal to the 2018 forecast for import GST produced in September 2018).
10. For income forecasting purposes it is therefore considered appropriate to use a figure of £5.4 million as forecasting outturn for import GST for 2018.
11. The current forecast methodology utilises a five year rolling average to forecast the increase in import GST. As noted above, excluding the single, very high value transaction, this methodology produced an accurate forecast for 2018.  

ISE Fees

12. ISE fees outturn for 2018 was £8.99 million. This was £0.13 million (1.4 per cent) less than the in-year forecast (updated September 2018) and £0.54 million (6.4 per cent) more than outturn for 2017.
13. Recent history indicates a steady decline of ISE fees over the period from 2013 to 2017 followed by the modest increase in 2018, reflecting the 2018 Budget changes to increase some ISE fees and widen the tax base.
14. The existing forecast methodology assumes £9.0 million of ISE fees per annum across the forecast period. It is acknowledged that history has shown that ISE fees have steadily declined over time, most likely as a result of consolidation within the financial services sector; however, in light of the small amounts involved, the IFG agreed to leave the forecast at £9.0 million per annum at this time.

Summary of GST forecast

Figure 12 – Summary of GST for 2018 – 2023

Forecast Range

15. The basis on which the forecast range is produced is unchanged and remains based on:
    1. A lower range 1 per cent below the central assumption and a higher range 1 per cent above the central assumptions is used for forecasting "normal" GST.
    2. A wider 2 per cent range above and below the central forecast is used for import GST, reflecting the higher trend growth assumption for this component.
    3. ISE fees have been relatively stable between years, and a 0.5 per cent range above and below the central forecast is used.
16. The overall effect of the range of forecasts is shown in Figure 13. Figure 13 – Summary of GST forecast range for 2019 – 2023

7. Impôts Duty Forecast 2019-2023 (spring 2019) Introduction

1. Impôts duties are levied on a range of commodities imported to the Island. The duties on the various commodities, principally alcohol, tobacco and road fuel, are reviewed annually. The duty increases for alcohol and tobacco are influenced by the strategies for particular health improvement and reduction in consumption policies.
2. The policies in that regard can be considered fairly successful based on the importation trends. These show that for most alcohol and tobacco commodities, the long-term trend is for reduced importation. There is some evidence from monitoring and feedback from retailers to suggest an increase in duty free for tobacco goods consumption but this is actively policed by Customs.
3. The basis of the Impôts duty future forecasts is to take the latest full year outturn and to apply past importation trends to forecast the future volumes and past Budget experience to forecast future duty rates.
4. Customs maintain records going back a number of years and on statistical advice, use a 10 year average of importation trends to forecast future volumes.

Increases in Impôts duty rates

5. The IFG has recommended that it remains appropriate to assume that recent policies in Budgets would continue in the absence of any updates to the existing tobacco and alcohol and licensing strategies. Analysis of recent Budgets shows that broadly RPI increases for tobacco and alcohol were common and that increases to road fuel and other commodities were less likely. Consequently, the forecasts only assume RPI increases for alcohol and tobacco goods.

Variation from Budget 2019 (September 2018) to spring 2019 forecast

6. The 2018 outturn was £62,462,842. The forecast (as at September 2018) was £63,305,000.
7. It is difficult to be accurate as to some of the precise reasons why the outturn differs from the forecast. Consumption of excise goods remains a matter of personal choice, particularly for alcohol and tobacco products. The quantities consumed throughout any given year and the subsequent Impôts yield can depend greatly on a number of divergent factors. These include changes in local legislation, changes in the duty collection regime, changes in attitude to health and the environment, as well as the general economic environment.

Future forecast

8. The forecasts have been revised and reflect:
   1. adjustments to the 10 year average trend of all imported goods to include the 2018 outturn; and
   2. the updated FPP economic assumptions for RPI (as at March 2019).
9. The duty rates for 2019 were agreed in 2019 Budget and the future duty rates for 2020-2023 are based on the RPI figures provided by the FPP in March 2019. The impact on the 2019-2023 Impôts Duties forecast can be seen in Figure 14.

Implications of Brexit

10. The implications of Brexit should be considered against future Impôts forecasts. The Customs and Immigration Service together with other States departments are currently engaged in operational workshops with the UK government, namely HMRC, HM Treasury and the Home Office regarding plans for Brexit.
11. The approach during such discussions is that the Island will continue to set Impôts duty rates and in this respect it is thought that future Impôts duties forecasts will be mostly unaffected by Brexit.
12. This position will be reviewed during subsequent IFG forecasts. Figure 14: Summary of Impôts duties for 2018 – 2023

13. The overall variations from the September 2018 forecast are primarily due to the higher base for tobacco from 2018.

Forecast range

14. The IFG is proposing to maintain the provision of a range around the Impôts duty forecast which uses the variation around the RPI assumptions compounded by a +/-1 per cent variation on future importation assumptions. A summary of the forecast range is shown in Figure 15.

Figure 15: Summary of Impôts duties forecast range for 2019 – 2023

8. Stamp Duty Forecast 2019-2023 (spring 2019) Introduction

1. Stamp duty is charged on property, equity and share transfer transactions according to the value of the transaction. It is also collected on Wills, Probate and Obligations. The stamp duty forecasts are separated into three components: general stamp duty, stamp duty on probate and stamp duty on share transfer property transactions.

2018 Outturn for all components

2. The 2018 outturn was £0.2 million below the Budget 2019 forecast (September 2018) however this was made up of variations across each of the components of the forecast:

• The main variance is for LTT, showing a 13.7 per cent (£0.3 million) increase over the Budget 2019 forecast.
• The increase over forecast in Stamp Duty received from Wills is 7.5 per cent, however this is only a £57k variance.
• The outturn for Stamp Duty transactions under £2 million is £0.4 million (<2 per cent) below forecast.
• Outturn for Stamp Duty transactions over £2 million is £0.1 million (1.6 per cent) below forecast.
• Probate duty shows a decrease below forecast of £81k (3 per cent).

General Stamp Duty

3. The main component of general stamp duty is duty on property transactions; in addition the forecasts allow for a relatively fixed forecast of stamp duty on Obligations and Wills. The duty on property transactions has been particularly volatile in the recent past, falling from over £14 million in 2009 to £10.7 million in 2013, a fall of 25 per cent, and then increasing to over £17 million in 2014, and then to £23 million in 2017.
4. The forecast for the MTFP 2016-2019 and Budget 2016 was based on a considerable analysis of the past years' data. This identified some key trends which informed the assumptions by the IFG for the forward forecast, in particular to identify an approach which separates the forecasts for properties under £2 million and those for higher value properties (over £2 million). The forward forecasts are then produced in two parts for these two sets of data with the economic assumptions for the housing market only applied to the under £2 million property transactions.
5. Considerable analysis has been carried out with Locate Jersey to establish the potential impact of its HVR policies. The current level of activity in the over £2 million property market is expected to continue for next 18-24 months and the view of Locate Jersey is there are sufficient properties to accommodate that demand.
6. The specific stamp duty economic assumptions for the spring 2019 forecast for house prices and housing turnover have changed since the Budget 2019 forecasts (September 2018) as shown in Figures 4 and 5.
7. Budget 2019 increased the rates of Stamp Duty on properties over £500k however the revenue effect of this was balanced by an extension of relief available to first-time buyers.
8. Due to the increase in housing turnover and house price assumptions, the spring 2019 forecast shows higher forecasts than the Budget 2019 forecast for under £2 million property transactions in 2019-2023.

Stamp Duty on Share Transfer – Land Transaction Tax (LTT)

9. The majority of share transfer property transactions are for flats and apartments, and likely to be lower value properties (on average) than non-share transfer property transactions. Therefore they are less likely to be subject to the anomalies and volatility seen on general property transactions.
10. The 2018 outturn for Land Transaction Tax was £0.3 million above the Budget 2019 forecast (September 2018). The same approach as for general stamp duty has been applied to the 2018 outturn for LTT using the revised March 2019 economic assumptions for the local housing market.

Probate duty

11. Probate duty is historically difficult to forecast. It is a duty payable from individuals who die and are domiciled in Jersey, or where the individual is not so domiciled but have Jersey moveable property. Between 2009 and 2015 however the number of transactions have remained steady, at broadly 2,000 per annum. Anomalies in income were seen in 2009 and 2012 due to one-off large transactions, but changes in the 2013 Budget have capped probate duty to £100,000 per estate as a competition measure to attract greater investment in to the Island, so these anomalies will not be seen in future.
12. The 2018 outturn was £2.6 million which continues to provide a range of quite variable outturns since the probate cap was put in place. The spring 2019 forecast has been proposed for 2019

– 2023 based on the five-year average of £2.4 million per annum. There is no mechanism proposed to attribute an annual increase or to vary a range of forecasts to probate duty.

Summary of Stamp Duty forecast

13. The resulting spring 2019 forecast, update in Figure 16, shows material increases in 2020. The primary influences are the forecasts for property transactions under £2 million, driven by the improved FPP assumptions, and the general increase in Land Transaction Tax forecasts in each year.

Figure 16: Summary of Stamp Duty forecast for 2019 – 2023

Forecast range

14. The IFG has maintained the current approach to providing a range around the Stamp Duty forecast. This uses the variation around the economic assumptions on house prices. The impact on the central forecasts is shown in Figure 17.

Figure 17: Summary of Stamp Duty for 2019 – 2023

9. Social Security Contributions 2018

1. The IFG has been informed by a summary of Social Security Contributions for 2018. These figures provide a useful check and balance for aspects of personal tax and employment income trends, although ITIS information should generally be more up to date.
2. The increase in number of contributors at 1.3 per cent and underlying earnings of 4.5 per cent do not significantly differ from the employment and earnings assumptions used by the IFG for the purposes of this forecast.

10. Recommendation

1. The IFG presents its revised forecasts for States income derived from taxation and duty for the period 2019-2023 as at spring 2019, as a range around a central scenario (see Figures 2 and

3).

2. In light of the continuing uncertainties identified in the outlook, the IFG would recommend that it is imperative that the Council of Ministers continues to maintain sufficient flexibility in the development of the Government Plan 2020-2023.

Appendix A – Terms of Reference

Terms of Reference for the Income Forecasting Group

Purpose

The group is established as an advisory function on the forecasts of all States income from taxation and social security contributions which will be informed by economic assumptions produced by the Fiscal Policy Panel.

Objectives

To produce an absolute minimum of two forecasts each year

• A full review of states tax, social security contributions and duty revenue forecasts will take place following the provisional outturn and no later than March of each year.
• A further forecast to inform the Budget debate, no later than September, including any revised economic assumptions and experience from the current year actual revenues.
• In an MTFP year, a further update will be considered between the full review in March and the actual lodging of the MTFP.

To produce reports on the forecasts of states income from taxation and social security contributions, including:

• Forecasts for income tax revenues
• Forecasts for goods and services tax and ISE Fees
• Forecasts for impots duties
• Forecasts for stamp duties
• Forecasts for social security contributions
• Economic assumptions used; and
• Factors and risks that should be considered

The forecasts will cover a period of at least four years and include a range within which a central forecast can be applied

Reporting

The reports will be presented to the Treasury and Resources Minister in advance of the Council of Ministers consideration.

Once a report is approved by the Treasury and Resources Minister it will be published alongside the Medium Term Financial Plan and the Budget.

Other reports can be prepared on the request of the Treasury and Resources Minister.

Administration

All meetings will be minuted with agreed actions.

Quorum – at least six of the ten members be present for the meetings to be considered quorate Quarterly internal review meetings will also be held.

Any variations to the group membership once established to be agreed by the Treasury and Resources Minister or Chief Minister.

It will be the responsibility of the Chief Executive and Treasurer of the States to ensure that the group has sufficient resources to fulfil its responsibilities.

Group Membership

The members of the group are:

• Treasurer of the States of Jersey (Chair)
• Comptroller of Taxes
• Director of Financial Services
• Chief Officer, Economic Development
• Chief Officer, Social Security
• Adviser, International Affairs
• Deputy Comptroller (Tax Policy)
• States Economic Adviser
• Up to two external members appointed by the Treasury and Resources Minister

The meetings of the group will be attended by the following officers in a supporting role:

• Head of financial planning (secretary)
• Finance director, Income Tax
• Economist

The group will invite other officers and external advisers to attend as appropriate which will be documented.

The group will operate independent of any political influence.