How much the Isle of Man, Jersey and Guernsey actually produce, who produces it, and how that mix has shifted since the 1990s — set against the UK as a reference economy. Three parts: Growth & Output (all three islands against the UK), Sector Mix (which industries produce it — Isle of Man focus, with cross-island comparisons in the last two tabs), and Structure (why the Isle of Man's mix keeps churning while its total barely moves).
View
Why doesn’t growth compound? → Structure
The chart shows how much output each island economy produces per person — in cash (nominal)
£ as published. Switch the Prices toggle to constant prices to strip out inflation and
compare across years on a fair footing.
All three Crown Dependencies produce substantially more per person than the UK.
In 2023, the Isle of Man, Jersey,
and Guernsey each generated roughly
twice the UK figure.
But high output per head doesn't mean residents are twice as well off.
These are small financial centres — much of the output is generated by banks, funds, and
online businesses whose profits ultimately flow to shareholders and clients who live elsewhere.
What the islands produce and what residents earn are two different things.
Put a number on that gap: in 2023 the Isle of Man's real GDP per head
sat around £71,000 against the UK's roughly
£38,000 — about 1.9× parity. Most of that isn't wages.
In the Island's own National Income accounts, company income (corporate profits
and retained earnings, heavily weighted toward finance and gaming) has run at roughly
three-fifths of GDP in recent years, against a little under a third for personal
(wage and salary) income — a split not seen in most economies of this size (2023/24 National
Income report, Table 4). A meaningful share of the headline figure is company profit passing
through Manx-registered entities, not money landing in residents' pay packets.
The Isle of Man's output per person
peaked around 2016/17 and has fallen since — two consecutive years of
contraction (2022/23 and 2023/24) have unwound most of the gains since the financial crisis.
⚠ The big step up in the Isle of Man line around 2007/08 is not a real economic
leap — it reflects a change in how GDP is measured (the adoption of ESA10
accounting standards), which mechanically inflated the headline figure by around 40%.
The two segments either side cannot be directly compared.
Nominal series. Isle of Man: GDP at market prices (income method), ESA10 basis from
2007/08 (IoM NI Accounts, Cabinet Office); pre-2007 factor-cost GDP from Digest of Economic and
Social Statistics 2009, Table 15.5. Per-capita denominator: published Table 4 (IoM NI Accounts
2012/13) for 2007–2012; census-interpolated population estimates for 2013–2023 (indicative, ±£1–2k).
Hard methodology break at 2007/08: the ESA10 adoption inflated nominal GDP substantially through
reclassification of FISIM, R&D capitalisation, and revised NPISH treatment; the two sub-series
are not chain-compatible. Jersey: GVA at basic prices (SIC2007, back-cast) ÷ end-year population
(Statistics Jersey, opendata.gov.je). Guernsey: GDP at market prices ÷ population (States of
Guernsey Statistics, GVA & GDP Supplementary Spreadsheet). UK: ONS ABML (GVA at current
basic prices, CP SA £m) ÷ EBAQ (resident population '000s, mid-year interpolated), release 31-03-2026.
Real series. Jurisdictions use different deflators and base years, all expressed
in approximate 2024-25 prices: IoM (pre-2008) — chain-linked backwards from the 2008 ESA10 anchor
using published real GDP growth rates (IoM NI reports, Table 3), extending to 1990/91. This uses
the IoM's own volume measures rather than deflating nominal levels by RPI, which would be less
appropriate given rapid sectoral change in the economy during this period. IoM (2008–present) —
IoM CPI (Statistics IoM, Jan 2008=100, rebased 2024-25=100). Jersey — Statistics Jersey implicit
deflator (constant 2024 prices). Guernsey — Statistics Guernsey implicit deflator (constant 2023
prices, one-year base offset vs other series). UK — implicit GVA deflator derived as ABML÷ABMI
(ONS chained volume measure, real-time database latest vintage, rebased 2024=100).
Cross-jurisdiction level comparisons in real terms are directionally indicative only; deflator
methodologies are not harmonised. IoM does not publish a GDP deflator.
IoM GDP uses the income approach only (no expenditure cross-check). Jersey and Guernsey
GVA excludes taxes on products net of subsidies; GDP at market prices adds these back.
The difference is small for these economies (<3%) but means the series are not
conceptually identical. UK comparator is GVA (basic prices), consistent with Jersey.
Personal vs. company income share of GDP
IoM National Income reports (Table 4, “Share of GDP by Source”) split GDP into
Personal Income, Company Income, and Other. Across the 2013/14–2023/24 constant-price series,
Company Income has run at roughly 58–65% of GDP; Personal Income at roughly 30–33%.
In 2023/24 (constant prices): Company Income £3,477m, Personal Income £1,921m, Other
£442m, GDP £5,840m — Company Income ≈59.5% of GDP. This is a National
Income concept (income approach to GDP), not a residency-adjusted GNI measure — it does not
by itself show how much of that company income is ultimately repatriated to non-resident owners
versus retained or distributed to resident shareholders/employees, only that it is not
compensation of employees.
⚠ Don't mistake output for prosperity.
High GDP or GVA per head is not necessarily a sign of a healthy, balanced, or sustainable economy.
In small, specialised jurisdictions like the Crown Dependencies it can equally reflect
high sectoral concentration (a single industry dominating output) and
volatility (large swings driven by a handful of firms or regulatory changes).
These figures also cannot capture regulatory and reputational factors
that affect the long-term sustainability of the business models behind the numbers.
⚠ Don't mistake output for prosperity.
High GDP or GVA per head is not necessarily a sign of a healthy, balanced, or sustainable economy.
In small, specialised jurisdictions like the Crown Dependencies it can equally reflect
high sectoral concentration (a single industry dominating output) and
volatility (large swings driven by a handful of firms or regulatory changes).
These figures also cannot capture regulatory and reputational factors — including ongoing international
tax reform (OECD BEPS), anti-money laundering frameworks (FATF), and automatic exchange of information
obligations (Common Reporting Standard) — which affect the jurisdictions' ability to sustain current
business models over the long term.
Economists in the Islands' governments, central statistics offices, and treasury departments
should treat these charts as a starting point for analysis, not a scorecard.
PricesSeries
Vertical line marks IoM ESA10 methodology adoption (2007/08); series not directly comparable across break
Annual percentage change in real GDP or GVA — the cleanest cross-jurisdictional comparison
because it abstracts from both price-level differences and deflator methodology.
The Isle of Man ran extraordinarily hot through the late 1990s and
2000s (eGaming, insurance, and finance expansion), with growth rates exceeding 5% most years
before the GFC. The pattern since 2015 is more mixed: a 7.4% spike in 2016/17 driven by eGaming
and corporate profits, followed by stagnation and two consecutive years of contraction
(−5.0% in 2022/23, −2.5% in 2023/24).
Jersey contracted almost every year from 2001–2013 in real terms —
a prolonged post-dotcom and post-GFC squeeze largely unreported in nominal figures — before
recovering strongly from 2014, including a remarkable +10.6% in 2021.
Guernsey and the UK track each other
broadly, though Guernsey is more volatile and recorded −2.4% in 2023.
Chart clamped to ±15%. The IoM recorded real growth of 13.5% (1998/99) and 13.7% (1999/00)
— these are in the data but clipped by the axis scale to preserve legibility of post-2000 variation.
IoM series. Source: IoM National Income reports, Table 3 (Growth in GDP at
Constant Prices). Annual percentage change in chain-linked real GDP at constant prices,
income approach. Financial year (April–March); plotted at the start year (e.g. 2023/24 → 2023).
Deflation via IoM CPI (Statistics IoM); IoM does not publish a GDP deflator.
Series runs 1990/91–2023/24 with consistent methodology from the 2010/11 report onward.
The UK real growth series shown here is derived from UK_GVA_PC_REAL_2024
(ONS ABML÷implicit GVA deflator) year-on-year, not from a published ONS annual GDP growth series —
differences vs ONS IHYP or ABMI YoY are typically <0.2pp but may diverge in data-revision years.
Jersey series. Year-on-year % change in GVA (basic prices) in constant 2024
prices (Statistics Jersey, opendata.gov.je). SIC2007 classification, back-cast to 1998 by
Statistics Jersey. Calendar year. The 2020 contraction (−9.5%) reflects a severe
COVID impact on financial services and hospitality; the 2021 rebound (+9.9%) is partly
a statistical recovery of deferred activity.
Guernsey series. States of Guernsey Statistics, GVA & GDP Supplementary
Data Spreadsheet, Real sheet. Annual change in GDP at constant 2023 prices. Calendar year.
Available from 2010 only. Guernsey's national accounts are produced less frequently than
Jersey's and subject to more substantial revision; treat year-on-year figures as indicative.
Comparability. Real growth rates are more directly comparable across
jurisdictions than nominal levels because they abstract from price-base differences.
However, different deflator methodologies (IoM CPI vs Jersey implicit deflator vs Guernsey
implicit deflator vs UK GVA deflator) introduce non-trivial differences, particularly in
years with housing cost divergence or large terms-of-trade movements. The series are
directionally comparable; point estimates should not be compared with precision <1pp.
IoM growth rates exceeding ±10% in a single year (1998/99: +13.5%; 1999/00: +13.7%;
2020/21: −8.0%) are clipped on this chart at ±15%. These are genuine published figures,
not data errors. The 1998–2000 episode reflects a step-change in online gambling
licensing; the 2020/21 contraction reflects COVID disruption to eGaming, tourism, and
aviation. Both are visible in the un-clamped data accessible via the governance panel.
Line up the Isle of Man, Jersey and Guernsey next to the UK on a map and they look like rounding errors — a few hundred square kilometres between them. Line them up on this chart instead, and each is out-producing the UK, per head, by margins that would make a G7 economy blush.
A reading of 100% means parity — the island produces the same per head as the UK average.
Above 100% means more; below means less.
Jersey has sat between 170–200% of UK for the
entire period shown — roughly double — reflecting its position as a major international
financial centre. Guernsey runs similarly high.
The Isle of Man crossed UK parity around 2005/06 and is now
estimated at roughly 190% of UK, though the 2016/17 peak has not been regained
in real terms. The jump in 2007 is partly a measurement change, not purely economic growth.
⚠ The dashed vertical line marks a change in how Isle of Man GDP is measured (ESA10 adoption,
2007/08). The two segments either side are calculated differently and the jump should not
be read as a sudden doubling of prosperity.
Construction. Ratio = jurisdiction nominal GDP or GVA per head ÷ UK GVA per head
(ONS ABML÷EBAQ), expressed as a percentage. IoM pre-2007: factor-cost GDP per head (Digest 2009,
Table 15.5) using the same ONS UK denominator series; the ratio was 86% in 1997/98, reaching 116%
in 2006/07. IoM 2007–2012: ESA10 GDP per head from NI Accounts 2012/13 Table 4, using the
UK comparator as published therein (ONS ABML÷EBAQ, 2013 Blue Book vintage); ratio 167% in 2007/08,
217% in 2012/13. IoM 2013–2023: derived nominal GDP pc ÷ ABML÷EBAQ (current vintage); ratio
approximate given population interpolation uncertainty. Jersey: GVA (basic) pc ÷ UK GVA pc.
Guernsey: GDP (market prices) pc ÷ UK GVA pc — Guernsey's numerator includes taxes on products,
UK's denominator excludes them, introducing a small upward bias (~2–3pp).
The 2007/08 IoM discontinuity reflects ESA10 adoption: FISIM reclassification,
R&D capitalisation (adding ~1.5–2% to GDP), revised NPISH treatment, and alignment to
European accounts standards. IoM Treasury estimated the modernised 2007/08 GDP at £2.84bn
vs the original £2.03bn — a 40% uplift from methodology alone. The pre- and post-2007
sub-series are shown in distinct line styles; they should not be interpolated.
Jersey's numerator is GVA (basic), which excludes taxes on products net of subsidies.
For a like-for-like comparison with Guernsey and with UK GDP (market prices), Jersey's
ratio would be marginally higher. The series are nonetheless presented on the same chart
because the difference is small and the directional comparison is valid.
Did the growth ever compound? The Isle of Man spent two decades posting
growth rates the UK could only envy; this tab asks whether they added up to a UK-beating
trajectory. The first chart answers directly: it divides the
Isle of Man's cumulative real growth by the
UK's, so the line rises only when the Island is growing faster
than the UK, runs flat when the two merely keep pace, and falls when the Island loses ground.
Changing the base year doesn't change the shape — it only moves which year reads 100.
What it shows (base 2009): a real climb through the 2010s — by the
2018 peak the Island's cumulative growth ran about 19% ahead of the UK's pace.
Ignore the 2020 spike (that's the UK's deeper COVID dip, not Manx strength). Then the unwind:
after two contraction years (2022/23, 2023/24), the Island sits roughly
4% ahead of where simply matching the UK since 2009 would have left it.
Nearly all the relative gains of the 2010s have been handed back — this is what
“growth that doesn't compound” looks like drawn as a single line.
The second chart shows the four jurisdictions separately, each rebased to 100 at the selected
base year. From 2009: Guernsey reaches ≈102 by 2023,
Jersey ≈107, the UK ≈114,
the Isle of Man ≈118 — everyone above the post-GFC
floor, by very different routes. Jersey's route is worth pausing on: itself a banking and
funds centre, it spent 2009–2020 mostly below its 2009 level (see the Real
growth tab) before a sharp recovery from 2021 that lines up with rising interest rates and
funds-industry growth — same underlying industry as the Isle of Man, different phase of
the cycle, not a ‘healthier’ economy. And note the absolute income gap never
closed either way: the same two years on the Output per head tab are £71k vs
£38k. Use the base-year selector to reframe — 2019 isolates COVID and after;
2014 shows the divergence since Jersey's trough.
Guernsey real series available from 2009 only. IoM built from chain-linked annual
growth rates (Table 3, IoM NI reports). Different deflator methodologies mean
cumulative levels are directionally comparable but not precisely equivalent. Series basis
mixes whole-economy (IoM, Jersey) and per-head (Guernsey, UK) measures; the IoM population
has been roughly flat, so its line reads similarly either way, but treat fine differences
as indicative.
Construction. Each series is indexed to 100 at the selected base year.
Divergence above or below 100 reflects cumulative real growth or contraction relative to
that anchor. The chart shows volume change, not absolute levels — a jurisdiction at 120
has grown 20% in real terms since the base year, regardless of whether it was richer
or poorer than another at the start.
Series construction by jurisdiction:
IoM — built from chain-linked annual growth rates (IoM NI reports, Table 3), anchored to
100 at the base year and compounded forward and backward. This propagates any measurement
error in individual growth-rate observations; accumulated error over 10+ years may be
material. Jersey — Statistics Jersey constant 2024 price GVA series, rescaled to base year
= 100; level-based, no chaining error. Guernsey — Statistics Guernsey constant 2023 price
GDP per head series, rescaled to base year = 100; available from 2009 only, so base years
before 2009 require backward extrapolation using growth rates, introducing additional
uncertainty. UK — ONS ABML÷ABMI implicit GVA deflator applied to nominal per-head series,
rebased to base year = 100; covers 1955–2025.
Ratio chart construction. ratiot = IoM indext ÷
UK indext × 100. Because both indices share the base-year anchor, changing
the base year rescales the ratio by a constant — the shape is base-year invariant,
which removes base-year choice as a degree of freedom in the headline comparison. Note the
numerator is whole-economy IoM real GDP while the denominator is UK real GVA per head; UK
population grew ~11% over 2009–2023 while IoM population was roughly flat, so this is
closer to a per-capita pace comparison than a total-output one. The 2020 observation is an
artifact of the UK's larger COVID contraction and should not be read as IoM outperformance.
Base year choice. The default base year of 2009 corresponds to the
post-GFC output trough — a natural common anchor from which diverging recovery trajectories
are analytically informative. It is also the first year for which all four real series
have data. Base years before 2009 exclude Guernsey; base years before 2001 exclude Jersey's
real series. Avoid anchoring at years with known outliers (e.g. 2020) unless specifically
examining COVID recovery, as this creates confounded index values in flanking years.
Cross-jurisdiction index comparisons accumulate the deflator methodology differences
noted in the Output per head tab. For trajectories longer than ~10 years, the
accumulated divergence between CPI-deflated IoM series and GDP-deflator-deflated
UK series may reach 5–8 index points. These indices should be read as approximate
cumulative volume change, not precise percentage differences.
Base year (= 100)Series
Isle of Man relative to the UK — cumulative real growth
IoM index ÷ UK index × 100. Rising = outgrowing the UK; flat = keeping pace; falling = losing ground. The 2020 spike is the UK's deeper COVID dip.
All four jurisdictions — real output rebased to the base year
Each series set to 100 at the selected base year; divergence shows cumulative growth trajectories, not levels.
Why does the mix churn? → Structure
Fromto
View
Industry
Jurisdictions
Loop sectors
Island
▼ Public Sector ▼ Finance sector (+Digital, IoM) ▼ Local Economy ··· UK Finance reference
2023/24
Groups
The Growth and Mix views show what happened: output roughly flat against the UK for two decades, while the industry mix churned dramatically underneath. This view is about why. The short version: the Island's economy runs on three pools that barely grow — people (population has risen 0.3% since 2011, with deaths exceeding births), housing, and capital. When one part of the economy expands, it draws workers, homes and investment away from the rest rather than adding new capacity. Three claimants compete for those pools: the public sector, the established private economy, and whichever new industry is being promoted at the time. The diagrams below map that competition; the charts show it in the data.
A qualitative system-dynamics reading of the Manx economy: a factor-constrained small open economy in which sectoral promotion produces reallocation, not growth. Three claimants — the public sector, the existing private sector, and the currently-promoted (“ephemeral”) sector — compete for quasi-fixed stocks of labour, housing and capital. Reinforcing loops (promotion R1; the VAT revenue engine R2) drive expansion phases; balancing loops (crowding-out B1, the housing brake B2–B3, wage bidding B4, tax-base erosion B5) enforce the ceiling. Population growth is migration-only (natural change negative since 2017) and migration is throttled by housing costs, so labour supply is effectively inelastic; census employment was lower in 2021 than 2001. The observable implication — sectoral GVA shares churn while UK-adjusted total growth ≈ 0 — is what the Growth and Mix views display.
Real GVA growth 1998–2022
+10%
UK over the same period: +54%
Population 2011 → 2024
+0.3%
84,497 → 84,756 · natural change negative
Employment 2001 → 2021
−574
32,293 → 31,719 (census sectors)
Median house price 2005 → 2025
+77%
£175k → £310k, land registry
Each new promoted industry — finance in the 1980s, then film, space, e-gaming, and now digital — follows the same arc. Promotion works: the sector grows and the headline numbers improve. But because workers, housing and premises are scarce, the same growth raises wages, rents and prices for everyone else, and the established economy shrinks to make room. When the new sector matures, attention moves to the next one. The result over 25 years: the pie barely grew, but the slices changed hands — see Sector Mix → Competing Out for the year-by-year picture.
The public sector plays this game differently: it doesn't sell anything, it taxes the other two — and its workforce has grown through every cycle, up around 2,600 jobs since 2001 in a labour market that shrank.
Figure S1 — two coupled loops sharing the promoted-sector node. R1 (policy attention → sector growth → headline GVA → political reward → attention) is reinforcing and fast; B1 closes through factor prices (staff/housing demand → wages & rents → existing-sector viability → GVA), converting promotion into reallocation. The rotation path (maturity ∥ → attention shifts) restarts R1 with a new protagonist, which is why the mechanism survives each individual sector's decline.
Figure S2 — the public sector's loop is asymmetric: it competes in factor markets (staff, housing) but not in product markets; its revenue derives from taxing the other two claimants, and its headcount behaves as a ratchet — expanding in revenue upswings, defended via reserves in downswings. B5's tax-base erosion operates with a long delay, which is what makes the ratchet politically stable and economically corrosive.
Figure S1 — The promotion engine
Promotion pays politically (R1) while crowding out the existing economy (B1); maturity rotates attention to the next sector.
Figure S2 — The public-sector asymmetry
Competes for people and housing, but is funded by taxing the other two sectors; headcount is a one-way ratchet.
Evidence → Sector Mix: Competing OutEvidence → Growth: Cumulative growth vs UK
How was a growing public sector paid for, in an economy that wasn't growing? Largely by the VAT-sharing arrangement with the UK. Under the historic agreement, the more the Island spent, the larger its measured share of pooled UK-and-Island VAT became — spending increased revenue, which financed more spending. VAT & duties receipts nearly tripled between 1998 and 2007, reaching 72% of all Treasury receipts. The UK then re-based the share in 2009 and 2011, cutting it by roughly a third. Since 2016 the arrangement (now FERSA) is re-set by expenditure survey roughly every five years, with back-dated corrections — the current agreement was signed in April 2025, with the next recalculation due 2028/29.
Figure S3 — R2: public spending → on-island final expenditure → measured VAT share → receipts → spending. Reinforcing and externally funded, so the domestic crowding-out cost of public expansion was invisible to local taxpayers while the loop ran (to 2009). The 2009/2011 rebasings are exogenous interventions, not endogenous corrections. Under FERSA (2016– ) the loop persists in throttled form: government consumption still enters measured final expenditure, but ~5-yearly surveys with straight-line retrospective true-ups bound the gain (B7); the 2023/24 survey produced back-dated adjustments and a rebalancing payment to HM Treasury in February 2025. Chart series: Customs & Excise receipts from Treasury accounts (Digest of Economic & Social Statistics 2009 t.16.5; Isle of Man in Numbers 2017 t.10; detailed government accounts 2014/15–2023/24). Nominal £m, fiscal years plotted on start year.
Figure S3 — The revenue engine: Common Purse → FERSA
Spend more → bigger measured share → more revenue (R2), until the UK re-based it; FERSA's survey-and-true-up now throttles the loop (B7).
Customs & Excise receipts, 1998/99–2023/24
The engine running (1998–2007), amputated (2009–2013), rebuilt under FERSA discipline. Cash terms, £m.
Receipts in cash terms (not inflation-adjusted). 2012/13–2013/14 as published (rounded); the 2019/20 spike and 2020/21 dip are COVID-era effects.
The binding constraints, directly. Population has been essentially flat since 2011; deaths have exceeded births every year since 2017, so all growth is inward migration — which is itself limited by housing costs. Meanwhile the median recorded property transaction has risen 77% since 2005. The employment chart shows where a fixed workforce went: out of retail, manufacturing, hospitality and farming; into the public sector and business services. The promoted sector of the last two decades, e-gaming, produces a large share of measured output with around 1,100 staff — it wins the statistics without needing many people.
Figure S4 — B2/B3/B4: migration is the only population inflow (natural change −351 in 2024) and is throttled by housing costs; housing supply responds with a multi-year delay; three sectors bid wages over a quasi-fixed labour pool. Sources: census-confirmed population plus experimental annual estimates (Statistics IoM Population Report 2025); IoM land registry completions (42,730 records, £20k–£5m filter, crude median — no mix adjustment); census 'Economic sector' employment (health/education/public administration combined by the source across reclassifications; e-gaming census employment 1,086 in 2021 vs 658 in 2016).
Figure S4 — The fixed factor pool
Migration-only population growth, throttled by housing (B2); slow supply response (B3); wage bidding over fixed labour (B4).
Resident population, 1951–2024
Census-confirmed points; experimental estimates from 2022.
Median transaction price, 2000–2025
Land registry completions; crude median, no mix adjustment.
Where the workforce went — census employment change, 2001 → 2021
Within a total that fell by 574. Health, education & public administration combined by the source.
Evidence → Sector Mix: GVA/Worker
The structural analysis in this view was first set out in a 2010 systems-dynamics study of the Manx economy and a discussion paper delivered to the Isle of Man Treasury in 2011. This page updates it with published data to 2026: IoM National Income accounts, census & population reports, land registry transactions, Treasury Pink/Blue Books, ONS national accounts, HMRC–IoM annual accounts and gov.im FERSA statements.