Entry #14: Published Today - A Decade of OCOD
Analysing Ten Years of Offshore-Owned Residential Property in England and Wales
New Publication: Decade of OCOD
A Decade of OCOD: Analysing Ten Years of Offshore-Owned Residential Property in England and Wales
Jonathan Bourne (University College London), Andrea Ingianni (Kingston University), Rex McKenzie (Kingston University / Faultlines Research Collective)
You get an idea. It grows. You start talking about it - with colleagues, at seminars, over coffee that goes cold while the conversation doesn’t. You decide to work it up. After several months, sometimes years, it becomes a full-blown article or book. That is exactly how it went with this one, and the journey is worth telling because it explains what the paper is actually trying to do.
I came to this data in 2017-18 through a piece in Private Eye. The magazine had made a freedom of information request to the Land Registry and the result was the Overseas Companies Ownership Dataset - every property in England and Wales registered to an offshore company, released publicly for the first time. I read the article and something clicked. If you could make sense of the data - really make sense of it, not just skim the surface - it could reveal a great deal about how the offshore actually works in relation to property. The mechanics of it. The jurisdictional architecture. The money trails that are designed not to be followed.
The problem was that the data is genuinely difficult to use. The addresses arrive as unstructured free text. A single entry might cover hundreds of individual flats compressed into one line with no postcode and no standard format. Standard address parsers fail immediately. But I was convinced the effort was worth it.
Shortly after I started working with the dataset I met Rowland Atkinson, and we began trying to use the data in relation to West London. That collaboration produced my first paper on the subject - Anchoring Capital in Place, published in Urban Studies in 2020 - and gave the research programme a foundation to build on. That paper established the wealth chain framework in relation to offshore property: the idea that what looks like a property investment is better understood as a node in a chain of jurisdictional and financial arrangements designed to accumulate value at one end while obscuring it at the other.
Backed by that publication, I started talking up the ideas at work and found that Andrea Ingianni was receptive. Andrea is a highly skilled econometrician and statistician. Around the same time I met Jonathan Bourne - Jonno to anyone who works with him. Jonno is what I would describe as an engineer. His official designation is data scientist, but what that means in practice is that he thinks about problems the way an engineer does: how do you build something that actually works, that others can use, that does not collapse the moment someone tries to replicate it? When I explained what I was trying to do with OCOD, he understood immediately both the potential and the obstacle.
The three of us - Jonno, Andrea, and I - then produced What’s in the Laundromat? Mapping and Characterising Offshore-Owned Residential Property in London, published in Environment and Planning B in 2023. That paper was a significant step forward. We built a data-processing pipeline drawing on machine learning techniques to create a parsed and classified dataset from the raw OCOD addresses, and what came out the other end was striking: 138,000 properties - 44,000 more than the original dataset had ever suggested existed. The majority were residential, with a disproportionate number in London. The average offshore residential property in London was worth £1.33 million, and collectively the London portfolio amounted to approximately £56 billion. We identified two distinct types of offshore property - nested multi-unit developments and individual single holdings - each with different price and distribution characteristics. That distinction, which the paper named and defined, became foundational to everything that followed.
At the same time, Rowland, Andrea, and I were developing the Liverpool work. Applying the Global Wealth Chain Typology to Property Purchases in the Liverpool and Merseyside Area, published in Environment and Planning A also in 2023, took the wealth chain framework out of the prime London market and tested it in a historically deprived city region. What we found confirmed the framework’s reach. Offshore investors in Liverpool real estate were operating across identifiable wealth chain types - the dominant asset strategy, concentrated geographically in and around the city centre, pointed toward the kind of jurisdictional opacity that in the literature is associated with illicit wealth accumulation. The multinational corporate model, theoretically the main source of innovation in wealth chain operations, was also present, and its spatial footprint was distinct. The OCOD data, properly processed, could do this kind of forensic work outside London. That mattered.
In 2022, funded by Trust for London, I led a further report - Tax Haven Money in London Real Estate - that pushed the analysis into territory with direct policy traction. Sharda Rozena, a very talented researcher, worked on that project alongside Jonno, Andrea, and me, and her contribution was integral to what it produced. The report analysed offshore purchases of London residential property between 1999 and 2022. We identified British Overseas Territories and Crown Dependencies as the dominant offshore centres - raising questions about Britain’s imperial past and the role of former colonial jurisdictions in contemporary London’s housing market that the mainstream literature had largely avoided. And we found evidence that offshore investment had pushed hard-pressed Black and Brown working class communities toward the outer London boroughs, a displacement dynamic that the aggregate numbers on their own could never have revealed.
That report established the empirical scale of the problem in terms that a policy audience could engage with directly. The decade paper published today builds on all of it and extends it across time.
What we have managed to do together - a political economist, a statistician, a data scientist, in this paper - is untangle an opaque and obstinate dataset and lay bare the dynamics underneath it. The singular achievement of this collaboration is not just to reveal how the offshore works in relation to property in England and Wales. It is to make that methodology available to everyone.
What the data required
The Overseas Companies Ownership Dataset has been publicly released since 2015. Over a hundred editions now exist. But despite a decade of availability, rigorous longitudinal work using it has remained rare, for a simple reason: processing it requires custom infrastructure that most research teams, policy analysts, and local authority teams cannot build. The Laundromat paper built that infrastructure for a single snapshot. The decade paper transforms it into something durable. Our solution is enhance_ocod - a pip-installable Python library that handles the full pipeline, from data acquisition through address parsing using a pre-trained ModernBERT model, to geolocation and property classification. It is freely available. The pre-trained model sits on Hugging Face. Anyone with basic Python skills can now work with a dataset that previously required a small technical team to approach. That was always the point.
What the data shows
The offshore residential property portfolio in England and Wales currently stands at approximately 106,000 properties with a combined estimated value of £80 billion. That is between 1.2 and 1.4 percent of the total value of all residential property in the country - a fraction that has remained essentially constant across the decade despite dramatic shifts in what that portfolio actually consists of. Relative to the national average, offshore residential property has declined from 2.4 to 1.9 times the average value: still a substantial premium, but a narrowing one.
The concentration is extreme. London accounts for 45 percent of all offshore-owned residential properties by volume and 81 percent by value. Half of that total value sits in just two local authorities - Westminster at 34 percent and Kensington and Chelsea at 16 percent. In 2025, those two boroughs alone held an estimated £39 billion of offshore-owned residential property. Those familiar with the Trust for London report and the Laundromat paper will recognise the geography. What the decade of data adds is the trajectory.
Beneath the aggregate stability lie two diverging dynamics that the time series makes visible for the first time. Single properties - typically ultra-luxury holdings in central London incorporated through the British Virgin Islands - have been declining consistently in both volume and relative value since 2018, when legislation was announced requiring beneficial ownership registers in British Overseas Territories. The data are consistent with anti-money laundering policy having had measurable effect at the top end of the market, though the paper treats that interpretation with appropriate caution. Multi-properties - typically large-scale housing developments incorporated through Jersey and Guernsey - have been moving in the opposite direction, increasing steadily across the decade. The BVI vehicle was the instrument of individual ultra-luxury accumulation. The Jersey and Guernsey vehicle is the instrument of institutional portfolio acquisition at scale. These are different money doing different things, and the decade of data lets you see them separating out.
Two events in the series deserve particular attention. The Mauritius Cliff of 2017 to 2018 saw a 91 percent reduction in properties held through Mauritian companies - 8,620 properties with an estimated value of £2.6 billion - following an amendment to the UK-Mauritius double taxation agreement that closed a loophole allowing property income to be extracted tax-free. The speed and scale of the response is the finding. Offshore property architecture is not passive investment sitting quietly in a register. It is active jurisdictional management that reorganises rapidly the moment the fiscal terms of a particular routing change. The Guernsey Bump of early 2022 - a sudden 10 percent increase in Guernsey-held properties coinciding with the passage of the Economic Crime (Transparency and Enforcement) Act - points to the same logic operating in anticipation of transparency requirements, though the data do not allow us to confirm causation.
Outside London, the pattern concentrates in regional urban centres. Liverpool, Leeds, and Manchester show the largest increases over the decade. The West Midlands broadly shows reductions. In most regions outside London and the South East, offshore-owned property is valued at or below the local average. The offshore premium is a London phenomenon, driven by the ultra-luxury single property sector that is now contracting. The Liverpool finding reinforces what the 2023 Environment and Planning A paper established: outside the capital, it is the corporate-institutional wealth chain model that dominates, not the ultra-luxury individual holding.
What this means for the wealth chain programme
The headline finding is this: the total fraction of residential market value comprised of offshore-owned property has held constant across a decade in which its internal composition has shifted substantially. Anti-money laundering legislation appears to have compressed the ultra-luxury end. Institutional development finance has expanded the multi-property end. The aggregate looks stable. The dynamics underneath are anything but.
That tension between surface stability and underlying reorganisation is precisely what wealth chain analysis is designed to illuminate - and this paper gives the framework something it has never had before: longitudinal traction. The earlier papers were essentially still photographs. What the decade of data provides, and what enhance_ocod makes replicable going forward, is the ability to watch wealth chains move. To track jurisdictional substitution across time, not just describe its presence at a given moment.
Wealth chain analysis has always been vulnerable to the criticism that it captures structure but not dynamics - that it shows you the architecture of extraction without being able to demonstrate how that architecture responds to pressure. The Mauritius Cliff and the Guernsey Bump answer that criticism directly. Wealth chains are not static conduits. They are adaptive systems that reorganise rapidly in response to regulatory and fiscal intervention. That is a stronger theoretical claim than anything the earlier papers could support empirically, and the decade of data now puts it on solid ground.
What this opens up concretely is the Jurisdictional Re-dressing work currently in development with colleagues in Mauritius - a paper that examines how offshore property ownership reorganises across jurisdictions in response to transparency and taxation regimes. The quantitative backbone for that argument now sits in a peer-reviewed open-access paper with a reproducible methodology attached. The three-registry triangulation can draw on enhance_ocod directly, and the Mauritius-to-Guernsey substitution pattern is exactly the kind of jurisdictional mobility the paper needs to theorise.
More broadly, what this collaboration has produced is infrastructure that others will use and cite. Every researcher who picks up enhance_ocod to study offshore property encounters the analytical vocabulary of the wealth chain programme - jurisdictional management, the distinction between ultra-luxury single holdings and institutional multi-property development, the sensitivity of offshore architecture to regulatory pressure. That is field building in a concrete sense. The Digital Plantation and The Extractivist City are the theoretical superstructure. This is the empirical foundation that makes the broader programme legible to audiences who need the numbers before they will engage with the framework.
Almost a decade of work on this dataset. A collaboration between people who each brought something the others could not. And a methodology that is now open to anyone who wants to use it.
The full paper is open access at Environment and Planning B: Urban Analytics and City Science. The Laundromat paper is at doi.org/10.1177/23998083231155483. The Liverpool wealth chain paper is at doi.org/10.1177/0308518X231191934. The Trust for London report is at trustforlondon.org.uk. The library is at github.com/JonnoB/enhance_ocod. The model is on Hugging Face. Use them.
A note on collaborators: Rowland Atkinson (University of Sheffield) is not a co-author on today’s paper but has been a central collaborator in this research programme since 2018. His fingerprints are on the foundational thinking, and his contribution to Anchoring Capital in Place and the Liverpool wealth chain paper was essential to everything that followed. The work is better for his involvement.