CASA Seminar Series 2025-26

A Decade of OCOD: Analysing ten years offshore owned residential property in England and Wales 

London property is a magnet for the global elite, drawn by both the glamour and consistent rate of return of the housing market. Often, companies in tax or secrecy havens are used as the vehicle for these investments, making understanding this property class challenging. In 2015, the UK government began regularly releasing data on these offshore property investments.

This talk analyses the first decade of this valuable but difficult-to-use dataset, first parsing and classifying the data into a structured format, then identifying the major residential trends. Over the last 10 years, the absolute value of residential OCOD property has increased from £64 to £86 billion; however, the relative value has decreased from 2.4 to 1.9 times the national average. London is the hub of offshore residential property, accounting for 47% of all properties by volume and 81% by value.  

However, these overall changes obscure more complex dynamics as luxury properties, owned by companies incorporated in the British Virgin Islands, reduced in value and volume across the time period. In contrast, there was an increase in housing developments being built by companies incorporated in Jersey and Guernsey.  

As Named Entity Recognition using deep learning is a critical part of the address parsing pipeline, we test different approaches and find that although training a model using weakly supervised learning (F1=0.97) marginally outperforms a model trained on a small, high-quality dataset (F1=0.95), this does not outweigh the cost of creating the labelling rules. As a result, we believe that investing more time in high-quality manual labelling is a better choice. 

As a result of this research a pip-installable Python library and a pre-trained NER model that processes the datasets have been made available, increasing accessibility to this valuable resource for social and economic study. 

Speaker

Jonathan Bourne has an MEng from Manchester and spent 8 years in the Norwegian energy sector, In 2017 he started his PhD at CASA under the supervision of Elsa Arcaute graduating in 2021. He then spent time working in Marine decarbonisation, he currently is working on NLP and OCR problems with his company THE 3TC AI. 

From digital maps to digital space: the long history of placemaking technologies  

For much of their history, humans have used tools and technology to navigate, wayfind, and locate themselves in space. These can be a liberating force. For some, for example, the humble map was (and is) a tool of both exploration and reassurance, contributing to a “sense of place” to be shared and enjoyed. As well as finding their way, however, humans have also used tools and technology to modify, shape and control the landscape, and to appropriate ownership of it. This talk will offer a brief, and necessarily highly selective, survey of some of these “technologies of placemaking”, examining some of the tensions they create, and looking at what digital methods might contribute to our understanding of them.

Speaker

Stuart Dunn is Professor of Spatial Humanities in the Department of Digital Humanities (DDH) at King’s College London, where he is Head of the Humanities Cluster. He served as Head of DDH between 2019 and 2023. His interests revolve around the use and creation of space and place in the humanities, especially archaeology and history. His book, 'A History of Place in the Digital Age', was published by Routledge in 2019. He is currently working on the book, 'Spatial Narrative and the Technologies of Placemaking'. Follow Stuart on Bluesky.

From boom to bust: two decades of international student mobility to the UK (and a detour into tiny homes)

International students have become central to UK higher education but increasingly controversial. As universities face financial pressures and government policy becomes increasingly hostile to international students, understanding where students come from, why they choose the UK, and where mobility is heading has never been more urgent. 

Ruth’s PhD research built a comprehensive picture of student mobility to the UK, measuring undergraduate applications from over 150 countries between 2009 and 2024. The findings reveal how student mobility reflects non-linear development patterns, responds in unexpected ways to global shocks like Brexit and COVID-19, and points toward a paradoxical future: applications from key sending countries are set to decline, yet UK universities will become more dependent on them. Fewer students from fewer places, but greater reliance—a concentration paradox with implications for institutional sustainability and national policy.

Beyond forecasting flows, this research raises broader questions about what drives educational migration, how geopolitical disruptions reshape higher education systems, and what the next years hold for universities navigating between internationalisation ambitions and migration politics. To close, Ruth will discuss her current postdoctoral work at UCL, which shifts from student flows to mapping very small homes.

Speaker

Ruth Neville is a spatial data scientist working in population geography and demography, with a focus on migration and mobility. Her PhD research examined international student mobility, analysing how socio-economic, cultural, and institutional factors shape these flows and how shocks such as Brexit and the COVID-19 pandemic disrupt them. Her work combines advanced quantitative approaches, including machine learning forecasting and spatial interaction modelling, to analyse and predict population trends. She has also contributed to projects utilising innovative spatial and digital trace data to understand mobility during crises, with her work contributing to findings disseminated by international organisations such as ECLAC and IOM. Ruth is currently a Research Fellow at UCL on the No Space Like Home project, developing methods and data to understand the nature and spatial distribution of small homes in England. Follow Ruth on Bluesky.

GIS vs The City Council: Data Analytics, Collective Intelligence and Policy Dust-ups

This presentation explores the role of data visualisation, spatial analysis and modelling in developing an evidence base for a community collective intelligence, and how this can be used to directly challenge poor decision-making and disruptive policy proposals - generating a direct line to the policy makers themselves, the local and national press, and national institutions, while at the same time unearthing an intriguing story arc of ulterior motives and political chicanery. The case study is Brighton and Hove City Council’s (BHCC) recent attempt to re-organise secondary school admissions in the city. The case-study highlights the crucial role that open scientific practices and rapid online publication can play in helping communities better understand complex issues while at the same time underpinning and galvanising a local collective intelligence leading to the self-organisation of a community resistance movement, challenging those in power and the decisions they make. 

Speaker

Adam Dennett is Professor of Urban Analytics at the Bartlett Centre for Advanced Spatial Analysis, UCL. Adam’s research interests include migration, mobilities, housing, gentrification, urban modelling and beer geographies; these are tied together with a fondness for data, visualisation and computational methods, put to use in applied settings. 

The DfT Connectivity Tool 

In the UK we have a housing crisis – we need to provide more homes. However, pressing demand can drive development in areas where there is simply land available. This often means development is geographically remote and in poorly connected and less sustainable places. This serves to constrain land use and density ensuring development is not optimised. This places pressure on existing infrastructure and a focus on increasing road capacity to give people access to jobs, healthcare, education, shops, recreation and green space. 

This can have considerable knock-on effects, as cars are often the only viable means of access between homes and services. The absence of genuine choice in how people move means that we are having to travel further to access key services. This negatively impact upon health, transport outcomes, socio-economic status, place-making principles and the environment. 

So how can we tackle this challenge? Build the homes we need, maximise the development potential of sites, and provide the infrastructure and services needed to support sustainable growth? How do we improve connectivity to underserved communities and improve outcomes?

For DfT the answer lays in creating a commonly agreed definition of what a sustainable location is and measure the ability of the transport network to get you where you want to go. To create a mechanism to drive growth and empower policy makers and decision takers to do things differently and better. That is why we have developed an unambiguous metric that that combines transport and land-use data in an innovative way to understand the relative connectivity of places. By helping planners and place-makers consider how to shape their towns and cities, we can transform how we use the transport system to kickstart the economy and enable sustainable growth. 

Speaker

Robert Singleton (MRTPI, MCIHT) is a chartered spatial planner with over 19 years in the profession.  He has a wide range of experience across the public sector and more recently in central government.  Following a move to the DfT in 2019, Robert has established a centre of excellence for engaging with the transport and spatial planning systems capable of providing expert and specialist input to the full range of modal policy areas.  The key driver of the Planning Policy Team at DfT is to foster a step change to better align the spatial and transport planning systems to deliver on the Government's growth mission and drive truly sustainable patterns of development. 

Robert is the Head of the Spatial and Transport Planning Professions at DfT and its Arm’s Length Bodies and is the Product Owner of the recently launched Connectivity Tool – a landmark digital platform that is the most sophisticated measure of connectivity ever attempted and stands as an international exemplar. 

Rethinking Urban Governance for Social Justice 

Urban governance shapes how cities provide for their residents. Contemporary ways of governing cities are influenced heavily by neoliberal logics that ultimately work to extract and accumulate wealth. The business-as-usual paradigm exists in a realm of fragmented spatial planning and rise of platform urbanism. Whether it is innovative transitions like solar energy, or regeneration of neighbourhoods, or even transport connectivity improvements, history shows that transitions always leave communities behind. These spatial injustices are not accidental — they are systemically rooted in who owns the land and services, who gets to manage them, and who benefits.  

In this talk, Trivik will argue that rethinking urban governance requires confronting these underlying power relations through a justice-oriented, spatially explicit lens. Drawing on recent empirical work in The Hague, Rotterdam, and Cape Town, the presentation will demonstrate how spatial data science, reflexive modelling, and participatory foresight can surface inequalities deeply rooted in our cities. The talk will end with some experiences of how we can shift away from proprietary and predatory consulting practices within cities to planning in the public interest.   

Speaker

Trivik Verma is a Professor of Just Urban Futures at Loughborough University, UK. His research focusses on cities, inequalities, and justice. He studies various challenges of urbanisation such as segregation, inequities in access and wellbeing, transport and energy poverty, and climate-related vulnerabilities. Although he is trained in computational methods like quantitative human geography, he uses any combination of methods that are useful for evidence-based research, participatory planning, and public action to identify pathways for just urban futures. Follow Trivik on Bluesky.

The “Sinking City”: Integrating Flood Risk Modelling and Vision-Language Models for Urban Analysis in Jakarta

The Sinking City project examines how artificial intelligence can enhance the understanding of flood risks in rapidly subsiding coastal cities. Focusing on Jakarta, the seminar will discuss two strands of research within the Sinking City project, both utilising AI as a means of conducting urban analysis: (1) interpretable deep learning for dynamic flood risk prediction, and (2) Vision-Language Models (VLMs) for large-scale building typology analysis, functional for granular-scale flood analysis. 

Climate change and urbanisation have intensified global urban flood disaster risks; yet, existing models remain insufficient in terms of feature selection, interpretability, and the dynamic integration of socio-economic factors. This study developed a flood risk prediction framework that integrates deep learning with interpretability mechanisms to identify spatiotemporal patterns and incorporate social vulnerability, utilising Convolutional Neural Network (CNN) and Long Short-term Memory (LSTM) models. Using Jakarta as a case study, we combined multidimensional data from 2016 to 2024, selected topographic-hydrological, built environment, social exposure, and ecological buffering factors, and constructed a Flood Risk Index (FRI). We then used SHAP (Shapley Additive Explanations) to quantify variable contributions. This framework transforms models from “black boxes” to “transparent glass boxes,” providing data-driven strategies for coastal cities facing climate risks. 

By combining interpretable flood modelling with AI-driven urban analysis, the Sinking City project establishes a transparent, multi-modal framework for understanding and predicting future urban flood risk, offering transferable insights for resilience planning in vulnerable sinking megacities.

Speaker

Zara is an Associate Professor in Spatial Data Science at the Department of Geography, King’s College London. She co-directed the Spatial Data Science programme within the department and led the Geocomputation and Data Science Research Hub (2023-2025). She’s currently on teaching sabbatical until August 2026, allowing her to focus on her research. She obtained her PhD in Urban Analytics from the Centre for Advanced Spatial Analysis at University College London in 2020 and a master’s degree in urban planning from the University of Southern California, with a focus on the preservation and design of the built environment. Her work uses various big data and computational approaches to understand complex urban problems relevant to planning and policy development, with a focus on the United Kingdom and Indonesia. 

Avoiding Transparency through Offshore Real Estate: Evidence from the United Kingdom 

The 2014 Automatic Exchange of Information (AEoI) represents the most comprehensive global effort to fight tax evasion by enabling cross-border information exchange on financial assets. We examine how this policy shifted offshore investment behavior. While the AEoI mandates reporting of financial assets, it excludes real estate holdings. Using administrative data on UK real estate purchases by foreign companies, we show that offshore users substituted financial assets for real estate in response to the new transparency regime: real estate investment from tax havens more exposed to AEoI significantly increased after the policy's introduction. We estimate that around 9% of the offshore financial wealth that flowed out of tax havens due to AEoI was ultimately reinvested in UK real estate. Our findings suggest that real estate assets now account for a growing share of offshore portfolios, partly due to their exclusion from AEoI reporting requirements. 

Speaker

Jeanne Bomare is a Research Officer at the London School of Economics (International Inequalities Institute). Her research lies at the intersection of public economics and real estate economics, and relies on public policy evaluation to understand how international tax regulations shape local housing markets and social inequalities. She has a particular interest in offshore real estate and international tax evasion. She conducts projects studying the determinants and consequences of offshore real estate ownership and evaluating the impact of tax policies on individual and corporate tax noncompliance. Jeanne is also a research fellow at the EU Tax Observatory and at Skatteforsk - Centre for Tax Research. 

From Disobedient Objects to Disobedient Mapping 

This talk will look at intersections of critical art and design history with digital mapping.  The V&A exhibition Disobedient Objects addressed objects of art, design and material culture in grassroots social movements since the 1970s. The study of protest and political movements has often been poorly recorded, archived or recorded, or devalued by historians, leading to critical histories using conceptual frames such as ‘history from below’ from the 1970s onwards. I have framed my work with disobedient objects and after as ‘art history from below,’ and became interested in how critical digital mapping projects can intersect with this work of uncovering neglected cultural histories, and in the visual cultures of mapping itself. This led to two collaborative projects whose possibilities and difficulties I will discuss: 1) a short film recovering the disobedient objects of the 2011 Syrian Uprising using 3D reconstruction based on phone camera footage from its participants, and 2) the first full dataset and mapping of British public monuments related to slavery. 

Speaker

Gavin Grindon is a curator and a lecturer in cultural and creative industries at Queen Mary University of London. He co-curated Disobedient Objects (V&A, 2015), The Museum of Neoliberalism (London, 2022-ongoing), the museum in Banksy’s Walled Off Hotel (2017-present), Werbepause: The Art of Subverting (Kunstraum Kreuzberg, 2022) and the Museum of the Apocalypse (Glastonbury festival, 2025). He has published in Art History, Social Text, Transactions of the Institute of British Geographers and elsewhere. He is currently writing a history of activist art and preparing an exhibition on Surrealism and protest, both due in 2027. 

Does Financialisation Inflate Rents? Evidence from Shanghai’s Private Rental Sector

This paper investigates how financialization impacts rent levels and pricing mechanisms in Shanghai’s private rental sector (PRS), offering both methodological and conceptual advancements in understanding landlord behaviours. Drawing on a city-wide dataset of 49,698 rental listings from Lianjia, a leading renting listing platform in China, we develop an operational approach to distinguish between financialized and non-financialized landlords by tracing their capital sources and business models. Conceptually, we propose a framework that identifies the core motives and strategies of different landlord types, offering new insights into the differentiated market impacts of financialized corporate landlords (FCLs). Our findings reveal two key dynamics. First, financial capital has been mobilized into the rental sector through a combination of state-led incentives, institutional investment, and innovative financing extended to non-financial firms. Within this landscape, asset-light FCLs dominate, using economies of scale and market consolidation strategies to rapidly expand. These landlords often drive rent inflation by pricing premiums into structural housing features, such as design and amenities, rather than location. In contrast, asset-heavy developments—although slower to scale due to capital intensity and regulatory constraints—benefit from state support and rent caps (e.g., under the ARH program), resulting in generally lower rents. However, they still capture premiums through locational advantages, reflecting long-term asset appreciation goals. These findings carry critical policy implications. To foster a more equitable PRS, it is necessary to address the risks of monopoly through inclusive financing mechanisms and to introduce targeted demand-side subsidies. Such measures can better protect vulnerable tenant groups affected by increasingly strategic rent-setting practices in financialized rental markets. 

Speaker

Xiang Ao is a doctoral candidate in economic geography and transport geography at School of Geography and the Environment, University of Oxford. His research interests include urban spatial structures, regional development, and housing studies. Follow Xiang Ao on Bluesky.

Putting Geography on the Map at UKHSA 

A place based approach to public health has been in place for 20 years. This presentation will share the evolution of GIS for Public Health, UKHSA’s current work to cloud optimise GIS, and the benefits this will bring for enhanced location intelligence and analysis for public health surveillance and response. 

Speaker

James Lewis brings over 20 years of experience in the strategic development and delivery of GIS for Public Health. He led the Geospatial Team at PHE during the COVID-19 response and has since leveraged this expertise to design and implement integrated approaches to geospatial analysis and capability at UKHSA. James also brings experience in applying GIS within Housing and Local Government sectors, complemented by voluntary work with The Conservation Volunteers and Somerset Community Trust.