Report on the Revised Forecast of States Income for Autumn 2020

Report on the revised forecast  of States income for autumn 2020

Income Forecasting Group

1 R.114/2020

Contents

IFG Autumn 2020 income forecast 4 Appendix A  Income tax forecast 18 Appendix B GST forecast 36 Appendix C Imp ts duty forecast 59 Appendix D Stamp duty forecast 64 Appendix E Other income forecast 69 Appendix F Terms of Reference 76

Income Forecasting Group

Report on the revised forecast of States income for autumn 2020

1. Executive Summary

1. Since the Income Forecasting Group (IFG) spring 2020 forecast , the Fiscal Policy Panel (FPP) have updated their economic assumptions to incorporate the latest data and economic analysis.
2. The Covid-19 pandemic has had an immediate and significant impact

on Jersey s economy and therefore its public finances. The Income Forecasting Group (IFG) has consequently reduced its autumn 2019

forecast for total States income in 2020 by £96 million (11.0%), when compared to the forecast made to inform the addendum to the

Government Plan 2020-23 (P.71-2019 Addendum). Around half of this reduction (£58 million) is due to downgrades to the forecast for income

tax, with £18 million from lower GST, £6 million from stamp duty, £1 million from reduced imp ts, £7 million from other income and £6 million from

an increase to the forecast for bad debts.

Figure 1 The variation in forecast States income in 2020

Year 2020 (£96m) Forecast variation to autumn 2019

-£7m, Other Income, 7%

-£6m, Bad Debts, 6% -£6m,Stamp Duty, 6%

-£1m,Impôts, 0%

-£18m,GST, 17%

-£58m, Income Tax, 64%

1. Report on the revised forecast for States income for spring 2020 - https://www.gov.je/Government/Pages/StatesReports. aspx?ReportID=5244
2. FPP Economic Assumptions August 2020 - https://www.gov.je/SiteCollectionDocuments/Government%20and%20administration/ ID%20Fiscal%20Policy%20Panel%20economic%20assumptions%20August%202020%2020200805%20CB.pdf

3. The Covid-19 pandemic has caused significant disruption to the economy, as necessary restrictions have been put in place to reduce its spread, which in turn affects the tax base. While restrictions have been partially eased, the speed of the recovery and the timing and extent of any further easing or re-imposition of restrictions remains uncertain. The IFG s forecast has been informed by the independent economic forecast produced by the Fiscal Policy Panel (FPP).
4. While Jersey has moved through the Safe Exit Framework, significant restrictions were in place for much of the year with Level 4 restrictions in place for six weeks, followed by four weeks at Level 3 and eight weeks at Level 2. This has a very significant impact on revenues in

2020 and, while revenues are expected to gradually recover, revenues are expected to be lower than previously forecast throughout the Government Plan period. The new IFG forecast for 2023 is around £85 million (8.6%) lower than the Government Plan forecast, primarily due to a reduction of £59 million in the income tax forecast with stamp duty £8 million lower, GST £9 million lower, other income £5 million lower and imp ts £4 million lower. This is consistent with the FPP forecast for the economy to be around 6% smaller than their pre-pandemic forecast, in the medium term.

Figure 2 The variation in forecast States income in 2023

Year 2023 (£85m) Forecast variation to autumn 2019

-£5m, Other Icome, 6%

£0m, Bad Debts, 0% -£8m, Stamp Duty, 9%

-£4m, Impôts, 4%

-£9m, GST, 11%

-£59m Income Tax 70%

5. The forecast remains highly uncertain, with a number of potential risks continuing to exist in relation to the Covid-19 outbreak:

The medical situation may require a second phase of restrictions

The potential for a larger impact on financial services, through a

contagion effect either from the global or local economies. This is a particular risk to corporate tax, as banks pay a significant proportion of corporate tax and may see reduced profits in the event of suffering significant loan impairments. Further, uncertainty remains over the future path of benchmark interest rates, which are a large driver of profits for the banking sector

Long-term structural impacts, for example due to the failure of some

key businesses

Further restrictions on travel or a permanent reduction in travel

following the pandemic. This would delay the recovery in the visitor economy and could hinder the development of new business for exporting sectors

A prolonged reduction in domestic demand due to a strain on

household finances and reduced consumer confidence

A prolonged reduction in business investment due to a strain on

corporate finances and reduced business confidence

Changes to consumer behaviour, for example a permanent increase

in the shift to online retail

In addition to downside economic risks, there is a further layer of

downside risks to tax:

- The potential that bad debts will increase more than in the base case forecast

- A potential trend for greater retention of profits in locally-owned firms, resulting in lower distributions.

6. In addition to risks from the Covid-19 outbreak, the longer-term risks remain:

The uncertainty regarding the UK s future trading relationship with

the European Union. While negotiations have been disrupted by the Covid-19 pandemic, the UK Government has ruled out any extension to the end of the transition period beyond the end of 2020. The IFG forecast assumes that the UK moves in an orderly fashion to a new trading relationship, at the end of the transition period.

Risks to financial services, including the challenge of compliance

with any changes to international treaty requirements. The pressure the current crisis has put on public finances around the world may result in an intensification of these requirements, which may have unintended negative consequences for Jersey s financial services sector. In addition, the risk to the Jersey financial services sector posed by the possible behavioural responses of multinational companies to changing international tax rules and, in particular, the OECD work on the Taxation of the Digitalised Economy.

Longer-term challenges including low productivity growth and the

impacts of an ageing demographic.

7. Under the downside scenario from the spring 2020 forecast, the forecast for total States income in 2020 was £163 million (18.6%) lower than the autumn 2019 IFG forecast. The majority of this reduction (£100 million) was due to downgrades to the forecast for income tax, with £24 million from lower GST, £16 million from stamp duty, £10 million from reduced imp ts, £6 million from other income and £6 million from a forecast increase in bad debts.
8. However, the autumn 2020 forecast is considered by the IFG as the central forecast, with the spring 2020 downside scenario included in Section 7 for information only. In future, several indicators will be closely watched to consider the extent to which the outcome may diverge from the IFG s central forecast.
9. Key indicators include:

The latest data and forecasts for the global economy

Sentiment in Jersey s key sectors informed by both discussions

with key business leaders and the results of the Business Tendency Survey

Local survey evidence on the impact of the Covid-19 outbreak Finance sector profit and employment projections

Revenue Jersey intelligence on corporate tax projections

Data on employment and on the number of Social Security

contributors

Number of people registered as Actively Seeking Work

Inflation

Average earnings

Residential property prices and transactions

Vacancies data from gov.je and other sources

Data from GST returns

Employer returns from the Income Tax Instalment System (ITIS) Aggregated data on credit card spending

Data from the support schemes that the Government of Jersey has

put in place:

- The Co-Funded Payroll Scheme

- The Business Loan Disruption Guarantee Scheme.

10. In the first instance, the Fiscal Policy Panel will be asked to update its forecast for the economy, using any new information that is available, plus the latest expectations for the extent and duration of any restrictions that remain in place. The income forecast will then be revised based on the updated FPP forecast and data from Revenue Jersey.

11. The autumn 2020 forecast should be read in conjunction with the spring 2020 forecast.

2. Purpose

2.1  The purpose of the Income Forecasting Group s report is to provide the revised forecast of total States income as at autumn 2020. The forecast reflects:

The Fiscal Policy Panel s (FPP) economic assumptions of August

2020 and other related economic data for Jersey

General revenues outturn for 2019

Initial information on general revenues for Quarters One and Two

2020

Forecasts from Treasury for other income

Latest available outturn data from Revenue Jersey

Intelligence from the IFG affecting future forecasts.

3. Background

1. The Terms of Reference (Appendix F) for the Income Forecasting Group requires that at least two forecasts are produced each year.
2. This is the second forecast produced in 2020 and covers the years 2020- 24 to cover the whole of the Government Plan period. Additional internal consideration is given to the forecasts throughout the year.
3. The IFG forecasts cover a large proportion of States income, including other States income from the Island-wide Rate, dividends and returns from States Investments, and other fees. The forecast for these income sources has been included in Appendix E.
4. The autumn forecast has been developed as a central forecast , to represent the IFG s view of the most likely outcome. The assumptions used for the downside scenario in the spring 2020 forecast have been reviewed and that forecast remains as the downside scenario for the autumn 2020 forecast (Figure 7).

4. Economic assumptions

1. The economic assumptions have been updated by the Fiscal Policy panel (FPP) based on the latest local and international developments to August 2020.
2. Further consideration has been given to the impact of the Covid-19 pandemic and data available from the restrictions in economic activity.
3. The central assumptions on which the autumn 2020 forecasts are based are shown in Figure 3.
4. The IFG has considered the economic assumptions from the FPP and have agreed that these assumptions should be used as the basis for the income forecast modelling, with any variations described in the relevant reports (appended).

Figure 3 FPP Revised economic assumptions as at August 2020

*Bank Rate Forecast for 2024 Only

Figure 4 Variation in FPP economic assumptions between March 2020 and August 2020

5. Risks and uncertainties

1. There are a number of specific risks and uncertainties applicable to each of the forecast components, and these are described in more detail within the appended reports.
2. The Covid-19 pandemic has had a significant impact on the forecast, with the International Monetary Fund forecasting the largest contraction in economic activity since the 1940s.
3. Prior to the Covid-19 outbreak, data on the local economy suggested there was a mild slowdown during 2019.
4. Uncertainty around the UK s future trading relationship with the EU continues. However, for the purposes of the forecast the IFG assumes that the UK will make an orderly transition to a new trading relationship
5. The increase in external regulation, particularly from the OECD, may have long-lasting consequences to Jersey s financial services sector.
6. Longer-term challenges, including low productivity growth and the impacts of an ageing demographic, will continue to cause uncertainty in the forecast.

6. Summary of autumn forecast

1. Due to the complexity of the adjustments made to each of the individual forecasts for spring 2020, the individual reports are appended, with the overall change to the autumn 2019 forecast summarised below.
2. Personal income tax (Appendix A) The personal income tax forecast comprises several components that have differing economic sensitivities. A number of individual adjustments suggested by the IFG result in a

£39 million decrease to the forecast in 2020 gradually decreasing to

£4 million over the following three years. However, these adjustments together with the updated FPP economic assumptions, and additional data, reduce the autumn 2019 forecast by £63 million in 2020, decreasing to a £42 million reduction in 2023.

3. Corporate income tax (Appendix A) The recent broadening of the corporate tax base, and the differential effects of the Covid-19 pandemic, has led to the IFG splitting the corporate tax forecast on a sectoral

basis. Additional adjustments have been made to reflect the impact

of restrictions on activity on sectors subject to a positive rate of tax. Updated information for the 2019 year of assessment has resulted in

an increase of £5 million to the 2020 forecast. However, from 2021 the forecast is reduced by £20 million with a £17 million reduction by 2023.

4. GST (Appendix B) The restrictions to activity arising from the Covid-19 pandemic have led to a reduction in sales revenues, and hence GST receipts, in 2020. Work has been undertaken to identify the sectoral impact of the restrictions. However, the IFG continue to believe that there will be a persistent impact on GST receipts continuing in 2021, with an implicit judgement that there would be a change in consumer behaviour. This leads to a reduction in the autumn 2019 forecast of £18.5 million in 2020, decreasing to £9.5 million in 2023.
5. Imp ts duty (Appendix C) Available in-year data suggests that the restrictions in place to mitigate the health impacts of Covid-19 have produced a varied effect on excise duties. For example, the closure of hospitality outlets resulted in a dramatic fall in on-trade sales of alcohol, and the closure of schools and businesses led to a significant reduction in fuel consumption. However, these decreases have been somewhat balanced by an increase in volumes of tobacco being put to duty. The revised FPP economic assumptions continue to give a reduction to the presumed duty rates in future years, with the autumn 2019 forecast being reduced by £0.5 million in 2020, £2.5 million in 2021, £3.0 million in 2022 and £3.7 million in 2023.
6. Stamp duty (Appendix D) Despite the Covid-19 pandemic causing significant disruption in the housing market, the easing of lockdown restrictions has permitted a resurgence in property transactions. The updated FPP economic assumptions for annual house prices and transactions have improved since the spring 2020 forecast, but now suggest a slower recovery. The in-year data and revised assumptions continue to cause a long-term reduction to the autumn 2019 forecast from £6.3 million in 2020 to £7.9 million in 2023.
7. Other Government income (Appendix E) The autumn 2019 forecast has been reduced for each year of the forecast, with a decrease of £7.0 million in 2020. The main driver of the reduction in the forecast

is reduced dividends from some States-owned entities and, more significantly, reduced investment returns on the Consolidated Fund and the Jersey Currency Fund.

8. Figure 5 provides a detailed analysis of the base case forecast which is summarised below.

Income Tax 585,000  554,000  558,000  603,000  642,000  674,000 GST 89,704  78,210  84,610  90,910  94,510  98,310 Imp ts Duty 62,879  67,716  67,986  69,979  71,037  71,485 Stamp Duty 34,898  29,083  30,953  30,249  31,118  32,023 Bad Debts (3,235) (9,000) (6,000) (6,000) (3,000) (3,000) Other Income 71,434  59,069  58,866  60,779  64,560  66,264

Total States Income 840,680  779,079  794,415  848,918  900,224  939,081

Autumn 2019 (forecast) 850,986  875,459  909,802  947,762  985,010

Variation (10,306) (96,380) (115,387) (98,844) (84,786)

 -1.2% -11.0%  -12.7% -10.4% -8.6%

Spring 2020 (forecast) 769,484 806,581  867,381  926,620  964,046 Variation 9,595  (12,166) (18,463) (26,396) (24,965) 1.2%  -1.5% -2.1% -2.8% -2.6%

Figure 5 IFG central forecast for autumn 2020

Central forecast

2019 Outturn (£ 000) 2020 Forecast 2021 Forecast 2022 Forecast 2023 Forecast 2024 Forecast

Income Tax

 467,000  Personal 434,000  461,000  500,000  534,000  561,000  475,000  Autumn 2019 497,000  524,000  550,000  576,000

 118,000  Corporate 120,000  97,000  103,000  108,000  113,000  113,000  Autumn 2019 115,000  117,000  121,000  125,000

 585,000  Total Income Tax 554,000  558,000  603,000  642,000  674,000

 588,000  Autumn 2019 612,000  641,000  671,000  701,000

 (3,000) Variation (58,000) (83,000) (68,000) (59,000)

-0.5% -9.5% -12.9% -10.1% -8.4%

Spring 2020 561,000  570,000  620,000  666,000  697,000 Variation -1.2% -2.1% -2.7% -3.6% -3.3%

GST

 80,793  GST 69,300  75,700  82,000  85,600  89,400  85,100  Autumn 2019 87,700  90,700  93,000  95,000

 8,911  ISE Fees 8,910  8,910  8,910  8,910  8,910  9,000  Autumn 2019 9,000  9,000  9,000  9,000

 89,704  Total GST 78,210  84,610  90,910  94,510  98,310

 94,100  Autumn 2019 96,700  99,700  102,000  104,000

 (4,396) Variation (18,490) (15,090) (11,090) (9,490)

-4.7% -19.1% -15.1% -10.9% -9.1%

Spring 2020 76,720  87,950  93,990  97,690  99,760 Variation 1.9% -3.8% -3.3% -3.3% -1.5%

Bad Debts

 (3,235) Bad Debts (9,000) (6,000) (6,000) (3,000) (3,000)  (3,000) Autumn 2019 (3,000) (3,000) (3,000) (3,000)

 (3,235) Total Bad Debts (9,000) (6,000) (6,000) (3,000) (3,000)

 (235) Variation (6,000) (3,000) (3,000) -

7.8% 200.0% 100.0% 100.0% 0.0%

Spring 2020 (9,000) (6,000) (6,000) (3,000) (3,000) Variation 0.0% 0.0% 0.0% 0.0% 0.0%

Imp ts Duties

 6,132  Spirits 7,544  7,185  7,293  7,476  7,701  6,375  Autumn 2019 7,268  7,441  7,635  7,841

 8,409  Wine 8,717  8,986  9,122  9,340  9,622  8,442  Autumn 2019 8,795  9,006  9,240  9,490

 832  Cider 851  860  855  858  868

 796  Autumn 2019 834  846  860  873

 6,204  Beer 6,031  6,569  6,633  6,691  6,791  6,339  Autumn 2019 6,628  6,719  6,827  6,941

 15,399  Tobacco 19,871  16,463  15,715  15,933  15,352  15,081  Autumn 2019 15,720  16,283  16,897  17,534

 22,685  Fuel 21,944  24,993  27,517  27,895  28,307  23,557  Autumn 2019 26,088  27,360  28,695  29,175

 235  Customs Duty 400  200  200  200  200

 200  Autumn 2019 200  200  200  200

 2,983  Vehicle Emissions Duty (VED) 2,358  2,730  2,644  2,644  2,644  2,948  Autumn 2019 2,730  2,644  2,644  2,644

 62,879  Total Imp ts 67,716  67,986  69,979  71,037  71,485

 63,738  Autumn 2019 68,263  70,499  72,998  74,698

 (859) Variation (547) (2,513) (3,019) (3,661)

-1.3% -0.8% -3.6% -4.1% -4.9%

Spring 2020 60,679  68,259  69,104  70,245  70,583 Variation 11.6% -0.4% 1.3% 1.1% 1.3%

2019 Outturn Central forecast  2020 Forecast 2021 Forecast 2022 Forecast 2023 Forecast 2024 Forecast

(£ 000)

Stamp Duty

 29,599  Stamp Duty 24,599  26,306  25,507  26,276  27,078  28,080  Autumn 2019 29,568  30,151  31,483  32,484

 2,548  Probate 2,400  2,400  2,400  2,400  2,400  2,400  Autumn 2019 2,400  2,400  2,400  2,400

 2,751  LTT 2,084  2,247  2,342  2,442  2,545  3,162  Autumn 2019 3,433  3,702  3,924  4,090

 34,898  Total Stamp Duty 29,083  30,953  30,249  31,118  32,023

 33,643  Autumn 2019 35,401  36,254  37,807  38,974

 1,255  Variation (6,318) (5,301) (7,558) (7,856)

3.7% -17.8% -14.6% -20.0% -20.2%

Spring 2020 20,318  26,851  29,040  31,009  31,910 Variation 43.1% 15.3% 4.2% 0.4% 0.4%

 769,246  General Tax Revenue 720,009  735,549  788,138  835,665  872,818

 776,481  Autumn 2019 incl GP measures 809,364  844,453  880,805  915,672

 (7,235) Variation (89,355) (108,904) (92,667) (80,007)

-0.9% -11.0% -12.9% -10.5% -8.7%

Spring 2020 709,717  747,060  806,134  861,944  896,253 Variation 1.5% -1.5% -2.2% -3.0% -2.6%

Other Income

 13,895   Island-wide Rates 13,286  13,486  13,809  14,155  14,523 13,870  Other Income - Dividends 9,330  8,133  8,568  8,918  9,347 13,722  Other Income - Non-Dividends 5,651  5,473  5,784  7,967  7,949

Other Income - Returns from

 29,947  Andium and Housing Trusts 30,802  31,774  32,618  33,520  34,445  71,434  Total Other Income 59,069  58,866  60,779  64,560  66,264

 74,505  Autumn 2019 66,095  65,349  66,957  69,338

 (3,071) Variation (7,026) (6,483) (6,178) (4,778)

-4.1% -10.6% -9.9% -9.2% -6.9%

Spring 2020 59,767  59,521  61,247  64,676  67,793 Variation -1.2% -1.1% -0.8% -0.2% -2.3%

 840,680  Total States Income 779,079  794,415  848,918  900,224  939,081

 850,986  Autumn 2019 875,459  909,802  947,762  985,010

 (10,306) Variation (96,380) (115,387) (98,844) (84,786)

-1.2% -11.0% -12.7% -10.4% -8.6%

Spring 2020 769,484  806,581  867,381  926,620  964,046 Variation 1.2% -1.5% -2.1% -2.8% -2.6%

* Some columns may not sum due to rounding

7. Downside scenario

1. The central forecast has been prepared based upon the latest FPP economic assumptions with additional consideration by IFG, as outlined in the separate reports.
2. The downside scenario determined for the spring 2020 forecast has been reviewed, and the IFG agree that the financial assumptions used remain appropriate for the autumn 2020 forecast.
3. The downside forecast assumes that the Covid-19 pandemic results

in significant structural impacts and, therefore, this should not be considered a default forecast if restrictions are extended.

4. The impact of the downside scenario results in a reduction of the overall base case forecast for autumn 2020 by £67 million in 2020 decreasing to £53 million in 2024 as shown in Figure 6.
5. The specific impact of the downside assumptions on the forecast are detailed in the relevant appended reports. In summary these are:
   1. Personal income tax - A c.£50 million decrease from the base forecast for spring 2020 arising from the extended period of reduced economic activity. The downside forecast for 2024 is £54 million less than the base case, reflecting the assumption of more significant structural impacts arising from the Covid-19 pandemic.
   2. Corporate income tax No further effect in 2020 due to this being the 2019 year of assessment. However, this is followed by a c.£16 million-£19 million reduction from the base forecast for each of the years 2021-24. This reduction is a result of a prolonged period of restrictions to economic activity, and an increased negative impact of the other risks to financial services firms.
   3. GST The downside forecast includes a weaker recovery and a permanently-lower level of receipts, beginning with a decrease of £4.3 million from the base forecast in spring 2020 reducing to less than £2 million below the base forecast in 2024.
   4. Imp ts duty The extended period of restrictions would only impact the 2020 year of the forecast, resulting in a decrease of £2.2 million from the base case in that year.
   5. Stamp duty A longer period of restrictions would likely result in fewer transactions of property, particularly those under £2 million or subject to Land Transaction Tax (LTT). The lower volume of transactions, and forecast decrease in property prices, further reduces the base case stamp duty forecast in 2020 by £1.2 million with an increased reduction of £2.6 million by 2024.

7. Other Government income The other income forecast is considered to be similar under both scenarios.

Figure 6 Downside scenario of IFG forecast

Downside Scenario  2020 Forecast 2021 Forecast 2022 Forecast 2023 Forecast 2024 Forecast (£ 000)

Income Tax

Personal 397,000  451,000  486,000  508,000  531,000 Corporate 115,000  74,000  83,000  86,000  93,000

Total Income Tax 512,000  525,000  569,000  594,000  624,000

GST

GST 63,530  72,240  80,740  86,120  89,040 ISE Fees 8,910  8,910  8,910  8,910  8,910

Total GST 72,440  81,150  89,650  95,030  97,950 Total Bad Debts (9,000) (6,000) (6,000) (3,000) (3,000) Imp ts Duties 58,432  67,986  69,979  71,037  71,485

Stamp Duty

Stamp Duty 15,608  20,755  22,219  23,736  24,431 Probate 2,400  2,400  2,400  2,400  2,400 LTT 1,111  1,777  2,126  2,343  2,442

Total Stamp Duty 19,119  24,932  26,745  28,479  29,273 General Tax Revenue 652,991  693,068  749,374  785,546  819,708

Variation to autumn 2019 (156,373) (151,385) (131,431) (130,126)

-19.3% -17.9% -14.9% -14.2%

Total Other Income 59,069  58,866  60,779  64,560  66,264 Total States Income 712,060  751,934  810,153  850,106  885,972

Variation to autumn 2019 (162,701)   (156,940)   (138,016)   (135,580)

-18.6% -17.2% -14.6% -13.8%

Variation to autumn 2020 (67,018) (42,481) (38,764) (50,119) (53,110) base case forecast -8.6% -5.3% -4.6% -5.6% -5.7%

Figure 7 Comparison of autumn 2020 and spring 2020 forecasts

Range of forecasts for States income (autumn 2020) £'000

,050,000

,000,000 950,000 900,000 850,000 800,000 750,000 700,000 650,000

600,000

208 209 2020 202 2022 2023 2024

Government plan 2020-2023 (autumn 20 9) Autumn 2020 forecast Downside scenario (spring 2020) Spring 2020 forecast Outturn

Appendix A Income tax forecast

Personal income tax forecast

The updated personal income tax forecast is summarised in Figure 1. The forecast has been revised down since May, primarily due to updated data from Revenue Jersey from both 2018 and 2019.

Figure 1: Changes to personal income tax forecast since May 2020

£m 2018 2019 2020 2021 2022 2023 2024

Personal tax

May 2020 forecast 453 480 446 480 519 561 585 New tax data -6 -12 -10 -12 -13 -15 -16 New FPP assumptions 0 -1 -7 -10 -8 -8 -9 Updated HVR forecast 0 0 +1 +1 +1 +1 +1 Change to IFG adjustments 0 0 +5 +2 +1 -4 0

New forecast 447 467 434 461 500 534 561 Change since May 2020 forecast -6 -13 -12 -19 -19 -27 -24

* Some columns may not sum due to rounding

The remainder of the note is set out as follows:

 Section 1 describes how the forecast is carried out

 Section 2 sets out the new economic assumptions and updates to tax

outturn data

 Section 3 explains the IFG s adjustments to the forecast to account for the

Covid-19 crisis

 Section 4 sets out the forecast

1 How the forecast is carried out

An overview of the personal income tax forecasting model is shown in Figure 2. There are two main elements forecasting taxable income and then forecasting the likely yield (i.e. tax collectable per £1 of taxable income) based on forecasts

of the value of deductions (including exemption thresholds for marginal rate taxpayers, and the various reliefs, credits and allowances claimed by taxpayers). The forecast of tax collectable is therefore the product of the forecasts for taxable income and yield.

Taxable personal income is estimated over the forecast period by taking outturn data provided by Revenue Jersey and projecting it forward on the basis of statistical relationships between taxable income and various economic variables. The economic variables include compensation of employees (CoE), company profits, employment, average earnings, inflation and interest rates. Forecasts of these variables are overseen by the independent Fiscal Policy Panel (FPP).

The yield is then forecast by taking the baseline data for the value of deductions and forecasting changes in these in line with assumptions about future taxpayer numbers, inflation, interest rates and policy changes announced in previous Budgets and Government Plans. So, for example, the aggregate value of the basic exemption thresholds might be assumed to rise in line with the lower

of RPI inflation and earnings (to represent the anticipated annual increase in the threshold), and employment growth (to represent the increase in taxpayer numbers claiming this threshold).

Figure 2: Model overview

BASELINE FORECASTS

Statistical Relationship

Income Forecast Income

Economic Assumptions

Minus Minus

Taxpayer Number Assumptions

Exemptions

Reliefs Forecast Reliefs

Known and Future Policy Assumptions

& Allowances

Economic Assumptions

Multiplied By Multiplied By Tax Rates Tax Rates

= =

Net Tax Forecast Net Collectable Tax Collectable

2 New economic assumptions and updated tax data

The Fiscal Policy Panel s (FPP) updated economic assumptions have been used in the model to update the income tax forecast. The economic assumptions were published in August 2020. The economic assumptions letter can be found at gov.je/fiscalpolicypanel

Updated information from Revenue Jersey Personal tax assessments

A full breakdown of personal income tax assessments for the 2018 year of assessment is not yet available. There are some differences in the way that data are reported between the previous system (ITAX) and the new system (RMS) with the new system providing greater granularity between taxable income types. Revenue Jersey are working through the data to produce a consistent series

of data from both systems on each of the income types reported on individual tax returns. At this stage, the priority has been producing consistent series for employment income and pension income, which represent around 80 per cent of total taxable income.

Based on this work, Revenue Jersey s analysis indicates that total income from employment grew by 4.6 per cent in 2018, compared to the 5.3 per cent growth assumed in the May 2020 IFG forecast. Pension income grew by 4.8 per cent, compared to the 7.5 per cent assumed in the May forecast.

This updated forecast continues to use 2017 as a base for the level of tax collected, but the actual growth rates in employment and pension income have been used to estimate the total amounts for each of these for 2018. The 2018 estimate therefore remains subject to some uncertainty.

Taxpayers on the high-value residency regime

Assessments have been completed for taxpayers on the high-value residency (HVR) regime. Revenue Jersey data show that £19.4 million was assessed for these taxpayers in 2018. As tax from these taxpayers is forecast separately, this has

been incorporated into the forecast.

ITIS data

Revenue Jersey has provided a revised figure for growth in employment income reported through the Income Tax Instalment System (ITIS) for 2019. This indicates that employment income grew by 3.2 per cent in 2019. This has been incorporated into the forecast, rather than the 3.1 per cent growth that would be estimated by the equation that underpins the tax forecasting model.

3 Adjustments to personal income tax forecast due to Covid-19 crisis

The personal income tax forecasting model is primarily based on estimates of the statistical relationships between economic variables and the tax base. This uses data over the last 20 years to forecast future tax receipts. While these statistical relationships are robust in the long run and in normal times, the IFG s previous forecast in May made a number of adjustments to these long-term relationships on the assumption that the collection of personal tax will diverge from the overall economic performance during the current period of economic distress and beyond.

Similar adjustments have been included in this forecast, but these have been revised to take account of the latest information. It is important to emphasise

that these adjustments are in addition to the downward impact of the revised FPP economic assumptions. So, the statistical relationships will already account for some lower growth in taxable income for example growth in taxable employment income will already be smaller because of the reduction in the FPP s forecast for growth in compensation of employees (CoE the national accounts measure of aggregate earnings) since the beginning of the pandemic.

In each case, the adjustments are assumed to gradually reduce and fall away

by the end of the forecast. Therefore, the forecast assumes that the tax base comes back into line with the economic variables, or into line with recent average growth rates for some of the smaller income types. There is a risk that the current economic disruption might result in a permanent adjustment to the relationship between economic variables and the tax base. Under the FPP forecast, the economy is smaller throughout the period, with a gradual recovery to a lower level, and this is considered sufficient to capture the impact on taxes in the medium term.

This is not intended to reopen the FPP forecast, but rather to seek to overlay additional impacts on the tax base. IFG judgement on these adjustments is set out below.

Employment income and other earned income (£2.3bn)

The IFG previously made a downward reduction to the forecast for employment income. This was due to the expectation that finance sector bonuses would be reduced, particularly in response to announcements from the Bank of England and European Central Bank urging financial institutions to show restraint with regard to bonuses. The expectation of lower bonuses has now been incorporated into the updated FPP economic assumptions and therefore this adjustment has been removed as it will already be captured by the model equations.

Figure 3: Impact of removing IFG adjustment to employment income

In May, the IFG made a further adjustment to the average effective tax rate (i.e. the amount of tax collected per £1 of taxable income). While the Co-Funded Payroll Scheme (CFPS) was expected to reduce the number of redundancies that otherwise would have occurred, the IFG expected that there would still be temporary lay-offs, reductions in hours and reductions in hourly wages, thereby reducing taxable income but not reducing the allowances that can be set off against that income. This would drive down the average effective tax rate.

The CFPS does appear to have been successful in reducing job losses, and registered unemployment has fallen back from its peak. However, it now seems likely that there will be a prolonged period of reduced hours and earnings, particularly in non-finance sectors, and this has been incorporated into the FPP forecast for a slower recovery in total earnings. This will put further downward pressure on the average effective tax rate, and the downward pressure will last longer than previously expected.

Therefore, the IFG s judgment is that the average effective tax rate will be lower, particularly next year, and the adjustment has therefore been increased in each year to 2023. The adjustment falls away from 2024 onward.

Figure 4: Impact of revised IFG adjustment to average effective tax rate

Pension income (£290 million)

The May forecast assumed that pension income would remain flat in 2020, equivalent to making a downward adjustment to the 2020 growth rate of 3.5

per cent. This was due to an expectation that, while pension income is driven

in the long run by both demographic change and growth in earnings, there are one-off factors that may result in slower growth in the short term. Specifically, the fall in financial asset valuations will have resulted in a reduction in the size of many pension funds, which may in turn result in new pensions paying out a lower amount than might have otherwise been expected and could in turn potentially lead to some deferral of retirement.

While financial markets have recovered partially, the IFG s view remains that there is likely to be limited growth in pension income this year. Therefore, an adjustment continues to be applied to reflect flat pension income in nominal terms this year. The updated FPP economic assumptions mean that the unadjusted growth rate

is now 4.5 per cent (rather than 3.5 per cent in May) so a larger adjustment is required to bring this to zero.

Figure 5: Impact of revised IFG adjustment to pension income

The adjustment falls away from 2023 onward as aggregate pension income is expected to return to following its long-run relationship with earnings and demographics.

Business profits (£160 million) and property income (£100 million)

In the absence of clear statistical relationships between the economic variables and taxable income in these categories, IFG has in the past assumed that this income grows in line with the respective recent average growth rates. Business profit (i.e. sole trader and partnership income) would therefore normally be forecast to grow at its five-year average rate (3.5 per cent per year) and property income at its 10-year average rate (6.9 per cent).

The IFG s May forecast made some adjustments to both of these. For business income, the adjustment was 30 per cent in line with the implied fall in non- finance profits from the FPP economic assumptions. The implied fall from the FPP assumptions is now 21 per cent in 2020, therefore an adjustment of -25 per cent has been made (i.e. to reduce the 3.5 per cent growth close to a 21 per cent fall). As before, this adjustment is assumed to gradually fall away over three years.

Figure 6: Impact of adjustments to personal business profits

For property rental income, the IFG previously made a downward adjustment of 15 per cent. The downward pressure remains, not least due to the impact of deferrals and waivers of rents. Therefore, the IFG adjustment is unchanged from May.

Investment and other unearned income (£100 million), shareholder income (£180 million) and income taxed at source (£20 million)

Investment and other unearned income includes a range of income, sources

such as bank interest, dividends and investment income plus a number of smaller sources of income. This income is generally grown in the IFG forecast using an established statistical relationship with interest rates (specifically Bank Rate). Shareholder income is a separate category, taxed on distributions of locally owned companies. The volatile nature of shareholder income, and the recent history of policy changes made to taxation of this area, means that in the past the IFG forecast assumed that this income remains flat in real terms. Income taxed at source, while not a substantial amount, has also generally been assumed to be flat in real terms.

The FPP assumption for Bank Rate has fallen further since May, which will reduce the forecast for investment income. However, a further adjustment is still required to reflect the expectation that dividends and distributions will fall sharply as

firms make significantly less profits and/or chose to retain more profits due to increased uncertainty over prospects. In addition, financial services firms make up a large proportion of global dividends and firms in this sector have been urged by regulators to reduce or suspend any dividend payments in the short term.

The May forecast made a downward adjustment of 20 per cent to investment income, distributions and taxed at source income. Developments since May suggest that a fall of approximately 20 per cent in each remains an appropriate judgement. This requires an adjustment of 16 per cent on investment income and an adjustment of 21 per cent on distributions and taxed at source, as set out below.

Equation based on

Investment Income Bank Rate -4% -16% -20% Distributions RPIY inflation +1.5% -21% -20% Taxed at source RPIY inflation +1.5% -21% -20%

Figure 7: Impact of adjustments to investment income, distributions and taxed at source

Overall IFG adjustments

Overall, the impact of the IFG adjustments is slightly smaller in most years than it was in the May forecast. This is for two reasons:

1. The reduction in employment income due to constraint on financial services bonuses has now been reflected in the FPP economic assumptions and therefore no further adjustment is required
2. The FPP forecast for the fall in non-finance gross operating surplus (the national accounts measure of profits) remains strongly negative in 2020 but less negative than May. The negative adjustment IFG made to sole trader/ partnership profits has therefore been adjusted

The upward impact of these is partially offset by an increase to the negative adjustment made on the average effective tax rate, and a larger downward adjustment to pension income. The aggregate impact of the IFG adjustments (i.e. the May adjustments plus the revisions made in this forecast) remains significant, as set out in Figure 8.

Figure 8: Aggregate impact of IFG adjustments (£m)

Adjustment 2020 2021 2022 2023+ Average effective tax rate -8 -16 -9 -4 Pension income -4 -2 -1 0 Profit -10 -5 -3 0 Property -4 -2 -1 0 Distributions and taxed-at-source income -10 -5 -3 0 Investment and other unearned income -3 -2 -1 0 Total -39 -32 -17 -4

Adjustments from May -45 -36 -18 0 Change in impact of adjustments +5 +3 +2 -4

Note: The impact of the May adjustments in this table is based on the underlying/unadjusted forecast as it was in May. Applying the May adjustments to the current forecast may result in a smaller/larger impact.

4 Updated personal income tax forecast

The personal tax forecast has fallen further in each year of the forecast. This

is primarily due to updated data from Revenue Jersey relating to 2018 tax assessments (based on individual tax returns) and 2019 taxable income reported through ITIS (based on employer submissions). The revised FPP economic assumptions reduce the forecast further. However, as some of the factors driving the change in the FPP assumptions were already factored into the IFG s May forecast through the IFG adjustments this means the impact of these adjustments reduces slightly in most years.

Figure 9: Changes to personal income tax forecast since May 2020

£m 2018 2019 2020 2021 2022 2023 2024

Personal tax

May 2020 forecast 453 480 446 480 519 561 585 New tax data -6 -12 -10 -12 -13 -15 -16 New FPP assumptions 0 -1 -7 -10 -8 -8 -9 Updated HVR forecast 0 0 +1 +1 +1 +1 +1 Change to IFG adjustments 0 0 +5 +2 +1 -4 0

New forecast 447 467 434 461 500 534 561 Change since May 2020 forecast -6 -13 -12 -19 -19 -27 -24

* Some columns may not sum due to rounding

New tax data

The partial data for 2018 assessments have been used to update the growth rates of employment income and pension income for 2019. The updated assumption for the growth rate of employment income results in a £5 million reduction in the estimate for 2018. This is assumed to recur, with £5 million-£6 million reduction

in each year of the forecast. Similarly, the updated assumption for the growth in pension income results in a reduction of £1 million-£2 million each year.

Further, the updated data have been used to re-estimate the equations used to forecast employment income and pensions. Growth in earnings, pension income and investment income are forecast using statistical relationships that are proven to be effective in explaining past growth rates. Estimates of these relationships are normally updated (i.e. re-estimated) when a new set of tax data are available. In this case only data for pension and employment income are available - re- estimating the equations used to forecast each of these results in small increase in the tax forecast for 2020 and 2021 but a reduction for 2023 and 2024.

Tax from taxpayers on the high-value residency scheme (HVRs) is forecast separately. Outturn data from Revenue Jersey indicate that the tax received from these taxpayers for 2018 was £1 million higher than the previous estimate, this higher tax is taken as a new baseline position and the increase therefore recurs in each year of the forecast.

The latest ITIS data for employment income in 2019 result in a reduction in the forecast of £5 million for 2019. This is assumed to be a recurring reduction and therefore the forecast is reduced by £5 million-£6 million each year.

New FPP economic assumptions

The new economic assumptions result in a reduction to the forecast in each year:

The reduced FPP assumption for finance sector profits result in a reduction

in the forecast for employment income. This is because statistical analysis demonstrates that growth in employment income can be explained by growth in both finance and non-finance earnings and also by profits in the finance sector (but not the non-finance sector). While the sensitivity of the forecast to changes in profits is relatively low (a coefficient of 0.07), the fall in the FPP assumption is very large and therefore results in a reduction to the forecast of £6 million-£7 million in 2020 and 2021. The FPP assumption is that finance profits will recover only partially and gradually.

The outturn for FTE employment in 2019 results in an assumption that

aggregate allowances/deductions will grow more quickly in 2019. However, sectoral analysis undertaken as part of the FPP forecast suggest that Compensation of Employees (CoE a measure of aggregate wages in the economy) will not have grown as a result as much of the FTE growth was in relatively low-paid sectors. This therefore reduces the forecast for 2019.

The FPP assumption for CoE in 2020 is slightly stronger, therefore increasing

the tax forecast by around £1 million. But going forward, the recovery in CoE is more gradual and this therefore has a downward impact on the tax forecast, reducing it by £4 million-£6 million in 2021-2024.

Inflation is now assumed to be lower in 2021, which reduces the forecast for

growth in allowances and therefore increases the forecast for tax by around £1 million from 2022 onward.

Lower interest rates result in lower claims for mortgage interest tax relief,

though this is partially offset by lower income from bank interest. The net impact is around £m-£1 million additional tax in each year 2021-2023.

The FPP assumptions now see house prices flat in 2020 and falling in 2021.

This has a small impact on the forecast through the expected impact on mortgage interest tax relief.

New HVR forecast

Data have been received from Locate Jersey on expectations for the number of new HVR in 2020. The net impact of these is a small increase to the forecast from 2020.

IFG adjustments

The impact of the individual IFG adjustments is set out in section 3. The aggregate impact of the adjustments remains significant, with a £39 million reduction to the forecast in 2020, a £32 million reduction in 2021, £17 million in 2022 and £4 million in 2023.

However, the impact of the reductions is smaller in most years than the impact in May. This adds to the forecast for 2020-22 but reduces the forecast for 2023.

As stated above, the positive impact of the revisions to the IFG adjustments is due to the fact that the factors that justified some of the adjustments from May have now been incorporated into the FPP s economic assumptions.

Policy changes

Two policy changes were already reflected in the previous forecast and remain in this forecast:

Revenue Jersey has estimated a reduction in personal tax of around £1

million from 2020 onward as a result of reduced distributions that may result from the extension of a positive rate of income tax to large corporate retailers ;

An increase of around £3million due to the decision to remove the higher

child allowance from 2019 as part of the new higher education funding scheme.

As these are unchanged from the previous forecast, they do not appear in Figure 8.

3. There may be some lag in the reduction in personal income tax due to this change, as distributions of profits that arose before 2018 but distributed in 2018 or later will not benefit from the tax credit. However, the most likely profile of this reduction is difficult to predict and therefore it has been assumed to be £1.2m in 2020, growing in line with whole economy profits thereafter.

Corporate income tax forecast

Figure 10 summarises the corporate tax forecast. The forecast has increased since May, primarily due to outturn data from Revenue Jersey.

Figure 10: Changes to corporate tax forecast since May 2020 forecast

£m 2019 2020 2021 2022 2023 2024

Corporate tax

May 2020 forecast 114 115 90 101 105 112 Revised outturn +5 +2 +2 +2 +2 +2 In-year data +3 +3 +3 +3 +3 Revised to financial services forecast 0 +3 -1 +1 +2 Revisions to other sectors 0 -1 -2 -3 -6

September 2020 forecast 118 120 97 103 108 113 Change since May 2020 forecast +5 +5 +7 +2 +3 +1

* Some columns may not sum due to rounding

Updated information from Revenue Jersey

Revenue Jersey has provided an update to the outturn for 2019. This suggests that tax from financial services firms was around £4 million higher than previously estimated, with the increase due to a small number of firms where the assessment was higher than the estimate. However, £3 million of this is expected to be a one-off increase and therefore only a small amount of the increase since May will be carried into future years. The estimated tax from corporate taxpayers outside financial services has increased by around £1 million.

Revenue Jersey has also considered in-year data using the payment-on-account made by corporate taxpayers for the 2019 year of assessment. The first instalment of 50 per cent was due in March/May and, based on analysis of payments by some large taxpayers in financial services, Revenue Jersey estimates that around an additional £3 million would be expected this year, on the basis of in-year data, and this is assumed to recur in future years.

Updated forecast Financial services

The forecast is now assumed to follow the FPP forecast for financial services profits. This differs from the last IFG forecast where an adjustment was made

to take account of recent cuts in benchmark interest rates and an expectation interest rates would remain lower for longer. Leading to markedly lower banking profits in particular.

As the March FPP forecast was finalised around the time of the cuts in UK benchmark rates, these cuts were not fully factored into the FPP forecast for

financial services profits. And the outlook for interest rates has deteriorated since then. On the basis of the latest outlook for interest rates and discussions with the financial services sector, the FPP has revised its financial services profit forecast sharply downward, bringing it closer to the forecast used by the IFG in May.

The FPP forecast for financial services profits covers the whole financial services sector, rather than just those sectors that are subject to corporate tax. Reweighting the 2020 forecast leads to a steeper fall in profits for the corporate tax-paying finance sector (as banking now has a greater weight) of 21.3%, rather than the 18.6% the FPP forecast for the whole sector.

Using the reweighted FPP assumption for 2020 and the FPP assumption thereafter results in a slightly less steep decline in profits this year (and tax next year), but a slower recovery such that tax in later years is broadly unchanged from the previous IFG forecast:

Figure 11: Forecast corporate tax from financial services (£m - budget year)

90 80 70 60 50 40 30 20 10

0

2020 2021 2022 2023 2024 Forecast using FPP March IFG May forecast

Forecast using FPP August Forecast with August FPP reweighted

Note: tax is collected one year in arrears, so tax in 2020 relates to profits in 2019

Large corporate retailers

The May IFG forecast included significant falls in profits in large corporate retailers in 2020 and therefore a large reduction in corporate tax from this sector in 2021. In the base case forecast, this loss of tax was gradually recovered over two years, by 2023, while in the downside forecast the recovery was slower with the loss of tax not recovered until 2024.

The updated forecast assumes that the recovery is slower than the previous base case, this reflects an assumption that the economy (in particular total wages) will

recover more gradually, in line with the updated FPP forecast. But it also assumes a larger permanent reduction, which reflects the potential for some of the recent shift to online shopping to be permanent. This is consistent with the GST forecast, which assumes a permanent reduction in GST due to increased household importation through the internet.

Figure 12: Forecast corporate tax from large corporate retailers (£m - budget year)

8 7 6 5 4

3 2 1 0

2020 2021 2022 2023 2024

May base case May downside September

Note: tax is collected one year in arrears, so tax in 2020 relates to profits in 2019

Property

The previous IFG base case forecast assumed that 2020 property income subject to corporate tax followed the same path as property income subject to personal tax. The downside saw a much sharper fall to a permanently lower level.

The IFG s view is that the factors driving the base case forecast from May remain valid and both office and retail rents and occupancy will be under pressure throughout the forecast period. Therefore, as in May, the revised forecast for 2021 follows the forecast for rental income in the personal tax forecast in 2020 i.e. a fall of 9 per cent.

Beyond 2021, any recovery is very uncertain. Therefore, the forecast assumes no increase in nominal terms in 2022 and thereafter following inflation. This is the same approach taken by IFG in May though the updated FPP assumption for inflation is a little lower so the forecast is reduced by a small amount.

Figure 13: Forecast corporate tax from property rental and development (£m - budget year)

25

20

15

10

5

0

2020 2021 2022 2023 2024

May base case May downside September

Note: tax is collected one year in arrears, so tax in 2020 relates to profits in 2019

Other corporate tax

Other corporate tax is from utilities, oil, mining and a few other small areas.

It represented around 7 per cent of corporate tax revenues in 2018 (£m) and therefore the overall corporate tax forecast is not particularly sensitive to changes in the growth rate applied to income for these firms.

This forecast previously assumed that tax from these firms followed the FPP assumption for growth in finance sector profits in the year previous, as tax is collected based on profits in the year before. This has been amended slightly for this forecast so that income rises in line with the FPP forecast for the overall level of gross operating surplus in the economy (a measure of total profits in finance and non-finance, including the rental income of private households). The forecast for large corporate retailers has also been amended to use this figure for tax in 2020.

Figure 14: Assumptions used in corporate tax forecast

Note: tax is collected one year in arrears, so tax in 2020 relates to profits in 2019

Figure 15: Forecast of corporate tax in key sectors (£m)

90 80 70 60 50 40 30 20 10 0

2019 2020 2021 2022 2023 2024

Financial Institution Property Large corporate retailers

Figure 16 sets out the new forecast in comparison to the previous IFG forecast. The revised outturn and in-year data result in small increases to the forecast. The revised FPP economic assumptions result in a small increase to the forecast for tax from financial services in most years, though a small reduction in 2022. The IFG s revised forecast judgements for other sectors result in less tax in each year.

Figure 16: Corporate tax forecast

£m 2019 2020 2021 2022 2023 2024

Corporate tax

May 2020 forecast 114 115 90 101 105 112 Revised outturn +5 +2 +2 +2 +2 +2 In-year data +3 +3 +3 +3 +3 Revised to financial services forecast 0 +3 -1 +1 +2 Revisions to other sectors 0 -1 -2 -3 -6

September 2020 forecast 118 120 97 103 108 113 Change since May 2020 forecast +5 +5 +7 +2 +3 +1

* Some columns may not sum due to rounding

Impact of Covid-19 on the forecast

In both the May forecast and this revised forecast, the IFG has split the corporate forecast into sectors so that the differential impact of the Covid-19 pandemic can be incorporated into the overall forecast. Figure 17 shows how the forecast for each sector has changed as a result of this new sectoral approach and the changes made to account for the impact of the pandemic. This demonstrates that the adjustments made since October 2019 to the forecast sectoral tax growth rates has reduced the corporate forecast by £27 million in 2021, £24 million in 2022 and £23 million in 2023.

The impact is greatest on tax from the financial services firms, though this is to be expected as financial services is responsible for the majority of corporate tax. This change is due to a sharp reduction in the FPP s assumption for financial services profits in 2020. Other sectors have smaller impacts in absolute terms but still large in proportionate terms. The reduction in the forecast for these sectors is due to the sector-by-sector approach taken by IFG to account for the impact of the pandemic on each.

Figure 17: Impact of revisions to sectoral forecast since October 2019 (£m budget year)

2021 2022 2023 Financial services -19 -16 -15 Property -3 -3 -4 Large corporate retailers -4 -3 -3 Others -1 -1 -1 Total  -27 -24 -23

Note: This is the impact of the changes to growth rates from 2021 onward (i.e. relating to profits from 2020 onward), relative to what the forecast would be using the growth rates for corporate tax as a whole in October 2019. This does not take into account the increase to the forecast as a result of outturn or in-year data (around +£6 million in each year).

Appendix B GST forecast

GST forecast note

This note sets out a provisional forecast of Goods & Services Tax (GST) for the Income Forecasting Group s (IFG) consideration. This updated forecast is based on new economic forecasts published by the Fiscal Policy Panel (FPP) in August 2020. The previous forecast in March 2020 included an extensive analysis of likely impacts of Covid-19, and the consequent restrictions on economic activity, on the GST forecast. Much of the analysis and that assessment remains valid so this forecast is an incremental revision that largely reflects the impact of changes to the economic forecast.

This note repeats much of the material contained in the March 2020 GST forecast note for IFG, but it is better to have all the material in a self-contained note as compared with a shorter note highlighting only changes to the forecast that would need to be read in conjunction with the March 2020 note.

The remainder of the note is set out as follows:

Section A1 describes how the forecast is carried out

Section A2 outlines the new economic assumptions

Section A3 sets out the new forecast

Section A4 covers the uncertainties and risks to the forecast Section A5 concludes.

A1 How the forecast is carried out

Goods & Services Tax ( GST ) was introduced in 2008 and is collected by Revenue Jersey. GST is collected from purchases of goods and services in the Island, and some imports from off-island. Initially introduced at three per cent, the GST rate was increased to five per cent in June 2011 with an effective rate of four per cent for 2011 overall.

There are three components to GST receipts:

- Normal GST (i.e. spend on Island)

- Import GST (i.e. imports from off the Island)

- International Services Entity ( ISE ) fees

Normal GST refers to purchases made in the Island and, as such, would typically refer to purchases by households. These receipts were around £77.5 million in 2018 and accounted for 84% of the total.

Import GST is paid on the importation of high-value goods directly, these receipts are volatile and in 2018 were £5.8 million and a little over 6% of the total.

ISE fees are paid by the financial services sector in Jersey and have been broadly stable at around £9 million since 2011, which accounts for 10% of the total.

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. Specifically, the IFG agreed forecast methodology was:

For years in which real economic growth is being predicted by the FPP:

increase normal GST revenues by 2%; and

For years in which real economic growth is not being predicted by the FPP:

increase normal GST revenues by 0.8%

Import GST used a five-year rolling average of growth rates ISE fees are held flat at the latest outturn.

This methodology is relatively crude, and simple comparison against an alternative model that forecast growth in GST with growth in Compensation of Employees (CoE) suggested the potential for a marked improvement.

To illustrate the relative performance of the two methods, simplifying assumptions are made here:

It is assumed the sum of normal and import GST grows at either 2% or 0.8%

rather than normal GST only. This is because it would require a great deal of additional work to obtain an historical import GST series. In any event since import GST is currently just 6% of total GST this simplification would not significantly change the result

The evaluation is against historical outturns and is conditional on the

economic determinant (CoE). Simply put when the economy is growing is it better to simply assume a 2% increase in GST or have a forecast consistent with the economy forecast conditional on the forecast for CoE being true

The agreed existing forecast methodology does not cover explicitly how to

model a change in the rate of GST. So, the method is extended to assume that the increase in revenue is proportional to the change in the rate e.g. an increase from 3% to 4% would mean, all else equal, an increase of 33% (4/3) in the revenue.

GSTX is defined as normal GST plus import GST or equivalently GST excluding ISE fees, and so the agreed forecast methodology when the economy is growing i.e. growth in real GVA is positive can be set out as:

GSTXt+1 = GSTXt * ( GST_RATEt+1 / GST_RATEt ) * ( 1 + 2%)

where GSTXt = the latest outturn for GSTX receipts at time t

GSTXt+1 = the forecast for GSTX receipts following the outturn

GST_RATE = the rate of GST in the outturn period and forecast period

The alternative method forecasts GSTX given CoE i.e. instead of assuming a growth rate of 2% or 0.8% it is the growth in CoE that drives the growth in GSTX, the details of this method are set out following a comparison of the two methods. A comparison of one-year ahead forecast errors is set out in the chart below.

It is notable that both methods have large forecast errors in 2015 and that Jersey hosted the Island games in that year which would have brought additional expenditure on-island by visitors.

Moreover, the latest outturn for GSTx in 2019 is markedly lower than would be expected. After growth in GSTx of around 4.5% in 2017 and 2018, GSTx fell by a little under 2% in 2019. Outturn data for CoE in 2019 is not available but conditional on the FPP forecast for CoE growth in GSTX for 2019 was forecast to be a little over 2%. A fall of 2% in 2019 compared with a forecast for a rise of 2% means the forecast error was +4%. The fall in GSTx revenues is at odds with the performance of the Jersey economy in 2019, and would be consistent with a fall in CoE that seems unlikely given wider indicators including business surveys.

The chart illustrates that the errors for the Agreed or old methodology are much larger than the Alternative i.e. new method. The error in 2011 is particularly large but receipts growth was exceptionally strong: GSTX receipts rose from £39 million in 2010 to £57 million in 2011 when the rate of GST rose from 3% to 4%. However, the growth in GSTX even after accounting for the rate change was

still exceptionally strong at more than 10%, the Alternative method explains this growth by changing the coefficients in the model used to explain the growth in GXTX whereas the agreed methodology is constrained to use growth of 0.8% as growth in real GVA was negative in 2011. As is often the case in considering the average forecast error, large negative forecast errors can be partly offset by large positive forecast errors, so it is useful to consider the average absolute forecast error that ignores the sign (positive or negative) of the error. The table below sets out the forecast errors of the two methods, it shows that the average absolute forecast errors for the Agreed method are five times larger than the Alternative method, and still four times larger after excluding 2011.

The IFG 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 could be improved. The next section of the paper sets out the new Alternative method for forecasting GST the results of which are shown in the table above.

Table 1: Comparison of forecast methods for GST receipts excluding ISE fees

Actual, £m Forecast, £m Forecast error, % Forecast error, £m

Agreed Alternative Agreed Alternative Agreed Alternative

2010 38.7 41.0 38.6 6.1% -0.1% 2.3 0.0 2011 57.1 52.0 57.2 -9.0% 0.2% -5.1 0.1 2012 70.7 71.9 70.5 1.7% -0.4% 1.2 -0.3 2013 68.6 71.3 69.2 3.9% 0.8% 2.7 0.5 2014 71.0 70.0 71.6 -1.3% 0.9% -0.9 0.6 2015 75.8 72.4 74.3 -4.5% -1.9% -3.4 -1.5 2016 76.0 77.3 75.6 1.8% -0.4% 1.3 -0.3 2017 79.5 77.5 80.8 -2.5% 1.7% -2.0 1.3 2018 84.6 81.1 83.5 -4.1% -1.3% -3.5 -1.1 2019 82.1 86.3 85.3 5.1% 3.9% 4.2 3.2

Average 2010-2018 -0.9% -0.1% -0.83 -0.07 Average absolutue value 3.9% 0.9% 2.51 0.64

Average excluding 2011 0.1% -0.1% -0.29 -0.10 Average absolutue value 3.2% 0.9% 2.18 0.71

GST receipts are a product of the GST rate and the expenditures on which GST is liable

T = t * C  [1] T  = GST receipts excluding ISE fees

t  = GST rate i.e. 5%

C  = GST expenditures, broadly equal to around 2/3 of household  

consumption, C = T/t

Consumption generally rises directly with income, and the largest part of income for the household sector is income from employment. It is therefore reasonable to use the compensation of employees as a proxy for total household income and hence a key determinant of household consumption.

C = c * CoE  [2] c  = the proportion of CoE that is spent on goods and services liable to GST

CoE  = Compensation of Employees

In this context it is assumed that ISE fees are driven by different factors than normal and import GST so GXTx is defined as GST excluding ISE fees. As a first step GST expenditures excluding ISE fees i.e. C above is compared against CoE for the years 2009-2018.

Chart 2: Compensation of employees and GST expenditures ex. ISE fees

£1,800

£1,700 £1,600

£1,500

R² = 0.9183

£1,400

£1,300

£1,200

£1,900 £2,000 £2,100 £2,200 £2,300 £2,400 £2,500 £2,600 Compensation of Employees (CoE), £M

As expected GST expenditures generally rise with income, so the higher

the Compensation of Employees (CoE) the higher the expenditure on which normal and import GST is paid and vice-versa so when CoE is lower so are GST expenditures. The coefficient of determination or R2 of 0.9128 estimates the fraction of the variance in GST expenditures that can be explained by a simple linear relationship using CoE.

In simple terms this means that GST expenditures and hence receipts can largely be explained by CoE, and given a forecast for CoE as an economic determinant then conditional on that forecast for CoE a good forecast for GST can be estimated. The next section sets out the estimation of such a model.

T = t * c * CoE   substituting [2] in to [1]   [3] Ln(T) = Ln(t) + Ln(c) + Ln(CoE)  taking logs  [4]

The model [4] can be estimated by standard Ordinary Least Squares (OLS) regression. The model is unrestricted in that lags of up to one year are allowed for in estimation and the best model among all the possible permutations is taken. NB the regressor INPT is a constant term in the regression.

Selected results are shown below: GST model estimated in Microfit

This statistical model is very robust and provides an excellent fit to the data. The lagged term on GSTx captures the dynamics between the income of households and their spending. The econometrics of this model are not discussed further but a fuller set of results is contained in the Annex. The performance of the forecasting model is shown in the chart below.

Chart 3: GST receipts ex. ISE fees, £m

£90

£80

£70

£60

£50

£40

GST receipts, data GST receipts, forecast model £30

£20

£10

£0

2008 2010 2012 2014 2016 2018 2020

The forecast model is derived from a well-structured theoretical model that is borne out empirically i.e. it explains the data. All variables are correctly signed and provide a clear intuition e.g. it implies that on average in due course that expenditure on goods and services that are liable to GST is equal to around three quarters the Compensation of Employees (CoE). The historical forecast performance given outturn CoE and the forecast conditional on the FPP forecast for CoE in 2019-24 is shown in the table below:

Table 2: GST receipts excluding ISE fees (GSTx), £m

Actual Forecast Forecast error Growth rates

% £m Actual Forecast

2009 40.7

2010 38.7 38.6 -0.1% -0.03 -5.0%

2011 57.1 57.2 0.2% 0.12 47.7% 48.1% 2012 70.7 70.5 -0.4% -0.29 23.9% 23.2% 2013 68.6 69.2 0.8% 0.53 -3.0% -1.8% 2014 71.0 71.6 0.9% 0.61 3.4% 3.5% 2015 75.8 74.3 -1.9% -1.46 6.8% 3.9% 2016 76.0 75.6 -0.4% -0.33 0.2% 1.7% 2017 79.5 80.8 1.7% 1.31 4.6% 6.8% 2018 84.6 83.5 -1.3% -1.12 6.4% 3.3% 2019 82.1 85.3 3.9% 3.17 -2.9% 2.2% 2020 85.2 3.8% 2021 85.8 0.7% 2022 88.1 2.6% 2023 90.1 2.3% 2024 92.2 2.3%

Average 2010-2018 -0.1% -0.07

Average absolutue value 0.9% 0.64

It should be noted that the new method aggregates normal and import GST whereas previously import GST was forecast as a five-year rolling average. Import GST is volatile but given the accuracy of the new method that forecasts normal and import GST as a total it is not proposed to forecast import GST separately. If IFG decided it would be worth forecasting Import GST separately further work would be required to evaluate the forecast accuracy of this approach.

A2 New economic assumptions

The Fiscal Policy Panel s (FPP) updated economic assumptions (Figure 18) have been used in the model to update the GST forecast.

Figure 18: FPP economic assumptions August 2020 (% change, unless otherwise stated)

*Bank Rate Forecast for 2024 Only

Note that the GST forecast uses only the forecast for Compensation of Employees (CoE). In 2020 CoE falls by less than the sum of the growth in average earnings and employment in 2020, whereas typically the growth in CoE is broadly equal to the sum of the growth in average earnings and employment. This is because the falls in employment are concentrated in non-finance that has much lower average earnings than the finance sector. However, the economic model for GST includes an implicit model for consumer spending that will not be applicable in 2020 due to the lockdown measures put in place to combat the global pandemic Covid-19. The hospitality industry i.e. hotels, restaurants and bars were closed along with many other industries, including non-food retailing from late March.

To estimate the impact of closures on revenues and hence GST receipts a judgement is made over what proportion of typical revenues will be received in each month through 2020. For example, it might be assumed that hotels will have no revenue from April to September. Two scenarios are considered for lockdown: the central scenario is for lockdown of three months followed by recovery over three months, while the more severe scenarios is for lockdown of five months followed by recovery over two months. These scenarios are identified as 3+3 and 5+2 respectively.

For each scenario four categories are identified with differing severity: Red, Amber, Green and Essential; Green and Essential industries are those sectors broadly unaffected directly by Covid-19 or providing critical services that will

be maintained through the crisis. This judgement can be applied to detailed industries within the broader sectors for the Standard Industrial Classification 2007 (SIC07). For example, the sale of motor vehicles can be identified within the sector G - Wholesale and Retail Trade .

The monthly profile and consequent annual impact under the two scenarios are set out below:

Impact of Covid-19 on revenues 2020

Severity 2020 Revenue proportions: 3+3 scenario

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Red 58% 100% 100% 50% 0% 0% 0% 25% 50% 75% 100% 100% 100% Amber 79% 100% 100% 75% 50% 50% 50% 50% 75% 100% 100% 100% 100% Green 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% Essential 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%

Severity 2020 Revenue proportions: 5+2 scenario

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Red 48% 100% 100% 50% 0% 0% 0% 0% 0% 50% 75% 100% 100% Amber 69% 100% 100% 75% 50% 25% 25% 25% 50% 75% 100% 100% 100% Green 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% Essential 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%

The assumptions over monthly revenue through 2020 imply that revenue in Red , Amber , Green and Essential will be 58%, 79% and 100% of normal through 2020 overall under 3+3.

GST receipts are only readily available by broad sector e.g. Wholesale and Retail Trade for 2018, and these are then apportioned using Full-Time Equivalent (FTE) employment for June 2019 to the detailed industries. Table 3 illustrates these judgments and estimates for the sector Wholesale and Retail trade . On this basis revenue and hence GST receipts in Wholesale and Retail is expected to be 23% lower than normal through 2020 under 3+3.

This method is extended to all sectors that expected to be significantly affected by the lockdown brought about by Covid-19, notably Construction along with accommodation and food services activities. The summary results for all sectors are shown in Table 4 and show that in aggregate across all sectors GST receipts would, all else equal, be expected to be 17% lower than would be expected in

a typical year without a lockdown under 3+3. Under the 5+2 scenario receipts would be 23% lower.

A3 Updated GST forecast

The GST forecasts are set out in the attached Table 5 comparing the Base and Revised March 2020 forecasts with the forecasts in the Government Plan 2019-23. The base forecast is the forecast for GST receipts that would be expected taken as given the FPP forecast for Compensation of Employees (CoE). However, as discussed, the lockdown means that sales, revenues and hence GST receipts will be markedly lower in 2020. The expected decline in GST receipts e.g. 17% due

to the 3+3 lockdown is then applied to the base forecast for GSTx to provide a

 Revised forecast; the 5+2 scenario provides a downside forecast.

Note that all forecasts include the proposed revenue measure to reduce from £240 to £135 of the de-minimis value of goods that can be imported before taxes i.e. GST becomes payable. This is forecast to raise £800,000 in a full year of implementation. Conceptually the value of this measure would, all else equal, rise with nominal expenditure but for simplicity it is held constant.

There are a number of unusually difficult judgements around forecasting the effects of the lockdown on GST, potential effects on forecast are noted in brackets:

1. In estimating the revenue loss for 2020 all months have an equal weighting. In practice the months of the potential lockdown e.g. Apr-Sep are peak seasonal months for tourism and hospitality and so should have a higher weight in deriving the annual figure. (Lower)
2. While hotels, bars and restaurants are closed, households may be increasing their consumption of alcohol and food at home. Alcohol duties that depend on the volume of alcohol would be less affected than GST that depends on the value of expenditure. Although a home-cooked meal would be likely to involve lower expenditures than a meal outside the home, there will be increased expenditures on food and alcohol in addition to any stockpiling that would probably be neutral over the year. (Higher)
3. Some consumption such, as durable goods, would likely simply be delayed while retail outlets are closed, and there would be higher sales with a catch- up when these stores reopen. Consumption of other goods and services, such as food and alcohol, is less easily postponed, although there would be expected to be increased consumption outside the home for a period when bars and restaurants reopen. (Higher)
4. The lockdown in Jersey will promote the purchase of goods and services via the internet, and delivery by post, which are not subject to GST if the value of the goods purchased are below or up to the de-minimus value. This could prompt a permanent change in consumer behaviour, with fewer physical purchases made in Jersey. (Lower). NB this is discussed further in Annex 2.
5. Fuel consumption may fall very sharply under lockdown. In the UK, figures taken from the Department for Business, Energy & Industrial Strategy (BEIS) survey conducted over the weekend 28-29 March showed that petrol consumption was down by 75% and diesel by 71% across supermarkets, oil company and independent sites. (Lower).

Following the first draft of this note, further work has been done to consider the seasonal variation in GST expenditures using new analysis by Revenue. Monthly

data on GST returns, payments and repayments by sector e.g. Wholesale and Retail are available for 2018. The data showed a stagger , with a high proportion of the value of returns at the end of calendar quarters i.e. March, June, September and December. This reflects the fact that most firms make a return for three months ending in March for example: not that most activity occurs in March.

The data showed seasonality in accommodation and food services with Q2

and Q3 (the quarters affected by lockdown) accounting for 62% of the total, by contrast the respective figure for wholesale and retail is 51% i.e. not seasonal. The lack of seasonality in retail may reflect offsetting effects between expenditure by tourists, business travellers and residents expenditure abroad. The seasonality

in GST receipts by sector is shown in Table 7. Accounting for the seasonality in Accommodation and food services makes little impact of the overall GST forecast, reducing revenues by a further £0.4 million. Given the wider uncertainties the forecast was left unchanged with no separate adjustment applied for seasonality.

A4 Uncertainties and sensitivities

Conditional on the forecast for compensation of employees the forecast for GST is extremely accurate with a standard error in the estimated equation of 1.25%.

Note that the standard error is different from the average forecast error. A general result in statistics for the normal distribution is that 95% of values lie within +/- 1.96 standard errors of the mean so, given the forecast for CoE, the model suggests

that forecast GST will be roughly within +/- 2.5% or £2 million 95% of the time or +/- 1.25% or £1 million 66% of the time.

Conditional on the FPP forecast for Compensation of Employees in 2019 the forecast error in GSTx of 4.4% i.e. forecast growth of +2.3% compared with a fall of -2.6% in the outturn data is statistically significant. On the one hand this error could be entirely a one-off outlier with no new information for the forecast i.e. noise , but on the other it could reflect new information i.e. news that should be reflected in the forecast. Note that the forecast model is estimated over the period 2010-2018 and is, by construction in the statistical method used, unbiased in that the average forecast error is zero. To take account of the likely news in the forecast error for 2019 the forecast model was adjusted by applying half the forecast error i.e. -2.2% as an adjustment in future years. This reflects a pragmatic judgement that half the forecast error in 2019 was noise but half the error was news.

To consider the uncertainty and sensitivities in the GSTx forecast the implicit expenditure liable to GST was compared over time as a proportion of Compensation of Employees that is the economic variable which is assumed to be the key driver of GSTx expenditures in Chart 4.

The 3+3 scenario (-17% in 2020) provides a central forecast and the 5+2 scenario with an additional 6% point fall in GSTx receipts in 2020 provides one part of a downside forecast. In addition, the downside forecast includes a weaker recovery and permanently lower level of receipts.

There is a notable downward drift in GSTx expenditures relative to CoE that

may reflect a rising proportion of expenditures imported below the de minimis threshold of £240, or alternatively a rising proportion of income (CoE) allocated to expenditures that are not liable for GST e.g. housing costs. This trend is reflected in the forecast model where the coefficient on the growth in CoE is around 85% which implies that 5% growth in CoE would be accompanied by growth of 4.25% in GSTx receipts. It should be noted that the forecast adjustment for 2019 data provides a more realistic trend.

Discussion at the IFG on 14/04/20 concluded that it was unlikely that following the fall in 2020 due to Covid-19 lockdown expenditure and hence GST receipts would recover completely to the forecast value in 2021. While the FPP economic forecast includes the effect of Covid-19 it was felt that there would be a persistent impact on GST receipts in 2021, with an implicit judgement that there would be a change in consumer behaviour. This judgement is consistent with subsequent reporting in the Financial Times on the challenges for retailing .

There was also agreement that the assumptions around lockdown should be amended to a central case of 3+3 and a downside case of 5+2. For the central case of 3+3 adjustments of -7% and -2% were applied to the GSTx forecast in 2021 and 2022 respectively that deliver a protracted recovery in GSTx receipts, and implicitly consumption, relative to income. This would be consistent with a rise in household saving following the forecast recession in 2020 as households recover from a significant rise in unemployment and a fall in incomes. It would generally be consistent with a more cautious consumer post-recession with a rise in precautionary saving. For the downside case of 5+2 larger adjustments were applied consistent with a slower and weaker recovery in GSTx receipts. These judgements are illustrated in Chart 4 above and set out in a revised Table 5 and

4. Retailers fear pain will last beyond the end of lockdown - https://www.ft.com/content/64dd785f-c08c-4c6c-a952-3bc088871cbd

Table 6 for the central and downside forecasts respectively. It should be noted that the first draft of the GST forecast had an estimated fall in receipts of 27% in 2020 that was revised to 22% with judgement, this included a more severe set of assumptions around lockdown in the monthly profile with six months of zero revenue for Red sectors .

Finally, on 17 April, the Government of Jersey has announced that it will be deferring the changes to the "de-minimis" level for paying GST on unaccompanied imported goods for personal use. This change is being deferred due to the Coronavirus pandemic. These changes, outlined in the Government Plan 2020-23, were due to take place from 01 July 2020 and would see the value of goods which could be imported - mainly through the postal system - without paying GST drop from £240 to £135. However, the Treasury and Resources Minister, Deputy Susie Pinel, has decided that the change should be deferred to January 2021 to help manage the current pressures on hauliers, postal workers and customs officers. So this is reflected with a scorecard loss of revenue of £400k in 2020 and an unchanged costing of £800k in 2021 and thereafter. Note that the IFG forecast does not include any estimates for the impact of policy measures that might be brought forward in the Government Plan for revised ISE fees.

The IFG meeting on 11 August considered the draft GST forecast. It was noted that the revised outturn data for GST in 2018 and 2019 did not change the parameters of the forecast model, and that the revised economic forecast with significant revisions to GVA and profits (gross operating surplus) but little revision to compensation of employees did not, by itself, change the GST forecast very much. Revised forecasts under 3+3 and 5+2 are set out in Chart 4 above and Tables 5 and 6.

The IFG meeting on 18 August agreed that in March at the start of lockdown it made sense to include a downside scenario for a longer lockdown but in August it made more sense to agree a single central scenario. The IFG discussed the prospects for retailing in the medium term given the likely impacts of Covid-19 e.g. the evidence in the charts presented in Annex 2, the IFG concluded that the central scenario should include the hit in the medium-term present in the 5+2.

A5 Conclusion

The IFG agreed that the base forecast driven by the economy forecast was reasonable and made a judgement to include an adjustment to the model to take account of the large forecast error in 2019 of just under +4%. Specifically, the IFG agreed apply half of that forecast error to the model for 2020 onwards, the result of which is to lower the forecast for GXTx, conditional on the economy forecast for compensation of employees, by around 2%. In considering the impact of Covid-19 and lockdown the IFG agreed on an adjustment to forecast GSTx receipts (accrued) in 2020 of -17%. After careful debate it was agreed that receipts were unlikely to recover quickly and that further downward adjustments to the base forecast were warranted from 2021 onwards. In the August meetings the IFG noted the downward trend in expenditure liable to GST relative to the compensation of employees. Reviewing the evidence around the switch to online purchases in the UK through lockdown e.g. Annex 2 it was judged that this made a return to the trend prior to Covid-19 unlikely. In addition, there were clear economic arguments e.g. the balance between forced saving in lockdown and

precautionary saving in the recovery with higher unemployment that suggested household consumption and hence GSTx receipts would be lower relative to income. This meant that the outcome in the 5+2 scenario where the ratio of GSTx expenditures to compensation of employees did not recover to its pre-crisis trend was the more likely outcome.

The IFG agreed a central forecast set out in Table 8 and illustrated in the chart below.

Table 3: Estimating the impact of Covid-19 on GST revenues in 2020, Wholesale and Retail trade; 3+3 scenario

RAG Rating for 2020 Revenue proportions in months of 2020 SIC Sub-Sector

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Red 58% 100% 100% 50% 0% 0% 0% 25% 50% 75% 100% 100% 100% Amber 79% 100% 100% 75% 50% 50% 50% 50% 75% 100% 100% 100% 100% Green 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% Essential 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%

FTE GST Revenue Wholesale and Retail Trade 6310 41.0 77%

Table 4: Estimating the impact of Covid-19 on GST revenues in 2020, All Sectors; 3+3 scenario

RAG Rating for

SIC Sub-Sector 2020 Revenue proportions in months of 2020

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Red 58% 100% 100% 50% 0% 0% 0% 25% 50% 75% 100% 100% 100% Amber 79% 100% 100% 75% 50% 50% 50% 50% 75% 100% 100% 100% 100% Green 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% Essential 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%

2018 £M Covid £M Impact of Covid-19, % and 2018 £M Total 78.0 64.5 -17% -£13.5

A - Agriculture, Forestry and Fishing  1.0 1.0 0% £0.0

B - Mining and Quarrying  1.0 1.0 0% £0.0

C - Manufacturing  3.0 3.0 0% £0.0

D - Electricity, Gas, Steam, and Air Conditioning  5.0 5.0 0% £0.0

E - Water Supply Sewerage Waste management  1.0 1.0 0% £0.0

F - Construction  6.0 5.3 -11% -£0.7

G - Wholesale and Retail Trade  41.0 31.7 -23% -£9.3

H - Transportation and Storage  2.0 2.0 0% £0.0

I - Accommodation and food services activities  9.0 5.5 -39% -£3.5

J - Information and Communication  6.0 6.0 0% £0.0

K - Financial and Insurance Activities  1.0 1.0 0% £0.0

L - Real Estate Activities  1.0 1.0 0% £0.0

M - Professional Scientific and Technical Activities  4.0 4.0 0% £0.0

N - Administrative and Support Service Activities  3.0 3.0 0% £0.0

O - Public Administration and Defence  -5.0 -5.0 0% £0.0

P - Education  0.0 0.0 0% £0.0

Q - Human Health and Social Work Activities  0.0 0.0 0% £0.0

R - Arts Entertainment and Recreation  1.0 1.0 0% £0.0

S - Other Service Activities  0.0 0.0 0% £0.0

T - Activities of Households as Employers  0.0 0.0 0% £0.0

V - Charity  -2.0 -2.0 0% £0.0

W - DIY Housebuilder  0.0 0.0 0% £0.0

X - Other  0.0 0.0 0% £0.0

Table 5: Income Forecast Group (IFG) meeting 24 August, interim GST forecast (accrual), £m

Total 2018 2019 2020 2021 2022 2023 2024 2019-23

GST 92.9 93.4 95.9 98.4 100.6 102.7 491 De minimis 0.4 0.8 0.8 0.8

ISE 9.0 9.0 9.0 9.0 9.0 9.0

GP 2019-23 GSTx 84.0 84.5 86.5 88.6 90.8 92.9

growth 0.6% 2.5% 2.4% 2.5% 2.4%

Base (Mar 20) GSTx 84.0 81.8 82.0 85.0 86.8 88.8 90.9 424 GST 93.6 91.0 92.5 93.9 96.1 98.1 100.1 472

De minimis 0.8 0.8 0.8 0.8

Base (Aug 20) ISE 8.91 8.91 8.91 8.91

FPP Economic  GSTx 9.0 8.9 8.91 84.2 86.4 88.4 90.4 determinants growth 84.6 -2.6%82.1 083.6.3% 3.6% 2.1% 2.3% 2.3% +0425.1%

Covid-19 effect: 3+3 scenario -17% -7% -2% 0% 0%

GST 93.6 91.0 78.0 88.0 94.4 98.1 100.1 449

De minimis 0.0 0.8 0.8 0.8 0.8

Revised (Aug 20) ISEGSTx 9.0 8.9

8.91 8.91 8.91 8.91 8.91

84.6 82.1 69.1 78.3 84.7 88.4 90.4

growth -2.9% -16% 13% 8.1% 4.4% 2.3%

Difference from GP in GST:

Base (Aug'20) - FFP forecast 0.6 -2.4 -3.4 -4.4 -4.4 -4.6 -19 Revised with Covid-19 effect 0.6 -2.4 -17.9 -10.3 -6.2 -4.6 -41

Note: Outurn data in black text and forecast values in grey

Source:  Proposed Government Plan 2020-23, Page 153

States of Jersey, Annual Report & Accounts 2019, page 161

Table 6: Income Forecast Group (IFG) meeting 24 August, interim GST forecast (accrual), £m

Total 2018 2019 2020 2021 2022 2023 2024 2019-23

GST 92.9 93.4 95.9 98.4 100.6 102.7 491 De minimis 0.4 0.8 0.8 0.8

ISE 9.0 9.0 9.0 9.0 9.0 9.0

GP 2019-23 GSTx 84.0 84.5 86.5 88.6 90.8 92.9

growth 0.6% 2.5% 2.4% 2.5% 2.4%

Base (Mar 20) GSTx 84.0 81.8 82.0 85.0 86.8 88.8 90.9 424 GST 93.6 91.0 92.5 93.9 96.1 98.1 100.1 472