Using Satellite Imagery to Revolutionize Creation of Tax Maps and Local Revenue Collection

Ali, Daniel Ayalew; Deininger, Klaus; Wild, Michael

doi:10.1596/1813-9450-8437

Cited by 4 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, our results are comparable to studies that estimate property value in low-income settings. Ali et al (2018) combine high-resolution satellite imagery with information on sales prices, targeted surveys and routine statistical data to predict property values in Rwanda, and find an R 2 of around 56 per cent when using a simple hedonic regression of property value on property characteristics using Ordinary Least Squares (OLS). 2020) and slightly underperform Knebelmann et al (2023), with an R 2 of 60 per cent and 22.1 per cent of the predicted values that are within 20 per cent of the target value (Table 5.1).…”

Section: Comparison With Other Studiesmentioning

confidence: 99%

“…The choice between different approaches typically depends on their cost, feasibility, speed, and how prone they are to corruption and leakages (Ali et al 2018).…”

mentioning

confidence: 99%

“…Inspired by these simplified hybrid approaches, recent studies have instead proposed to predict the value of properties using machine learning algorithms, which are typically trained using a small dataset of property values -obtained from transaction or market prices, or direct assessment by land assessors -and property characteristics available for all properties through administrative data or extracted from property photographs. predictions than area-based or points-based methods (Manwaring and Regan 2019), and might be associated with less collusion between tax assessors and taxpayers during inperson appraisal visits (Ali et al 2018). It is worth mentioning that in a context like the DRC, where increasing tax revenue and tax compliance is a high priority, policymakers need to properly weigh the benefits, in terms of accuracy and reduced leakage, against the complexity and opacity of these models -which might reduce taxpayer understanding and lower tax morale (Manwaring and Regan 2019).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Using Machine Learning to Create a Property Tax Roll: Evidence from the City of Kananga, D.R. Congo

Bergeron,

Fournier,

Kabeya

et al. 2023

View full text Add to dashboard Cite

Developing countries often lack the financial resources to provide public goods. Property taxation has been identified as a promising source of local revenue, because it is relatively efficient, captures growth in real estate value, and can be progressive. However, many low income countries do not collect property taxes effectively due to missing or incomplete property tax rolls. We use machine learning and computer vision models to construct a property tax roll in a large Congolese city. To train the algorithm and predict the value of all properties in the city, we rely on the value of 1,654 randomly chosen properties assessed by government land surveyors during in-person property appraisal visits, and property characteristics from administrative data or extracted from property photographs. The best machine learning algorithm, trained on property characteristics from administrative data, achieves a cross validated R2 of 60 per cent, and 22 per cent of the predicted values are within 20 per cent of the target value. The computer vision algorithms, trained on property picture features, perform less well, with only 9 per cent of the predicted values within 20 per cent of the target value for the best algorithm. We interpret the results as suggesting that simple machine learning methods can be used to construct a property tax roll, even in a context where information about properties is limited and the government can only collect a small number of property values using in-person property appraisal visits.

show abstract

Section: Comparison With Other Studiesmentioning

confidence: 99%

“…The choice between different approaches typically depends on their cost, feasibility, speed, and how prone they are to corruption and leakages (Ali et al 2018).…”

mentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Using Machine Learning to Create a Property Tax Roll: Evidence from the City of Kananga, D.R. Congo

Bergeron,

Fournier,

Kabeya

et al. 2023

View full text Add to dashboard Cite

show abstract

“…For instance, technology can help identify the tax base, help enforcement of tax collection, and decrease the cost of compliance. Technology can also be used for creating satellite imagery maps that can improve the coverage of tax collection, as Ali et al (2018) demonstrate for the city of Kigali in Rwanda. Within the context of our theoretical framework, such technology would reduce the scope for political influence and corruption (Kochanova et al 2020).…”

Section: Technology and Transparency Technology And Transparencymentioning

confidence: 99%

Political Distortions, State Capture, and Economic Development in Africa

Canen

Wantchékon

2022

Journal of Economic Perspectives

View full text Add to dashboard Cite

This article studies the role of political distortions in driving economic growth and development in Africa. We first discuss how existing theories based on long-run structural factors (e.g., pre-colonial and colonial institutions, or ethnic diversity) may not capture new data patterns in the region, including changes to political regimes, growth patterns, and their variation across regions with similar historical experiences. We then argue that a framework focused on political distortions (i.e., how political incentives impact resource allocation and economic outcomes) may have multiple benefits: it encapsulates many distortions observed in practice, including patronage, variations in contract enforcement and the role of political connections in firm outcomes; it unifies results in Africa and elsewhere; and it leaves a wide scope for policy analysis. We conclude by overviewing reforms that may curb such distortions, including changes to campaign financing rules, bureaucratic reform, free trade agreements, and technology.

show abstract

Design and Implementation of a System to Determine Property Tax Through the Processing and Analysis of Satellite Images

Silva

Solano

Jiménez

et al. 2020

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

One of the main objectives when implementing metaheuristics in engineering problems, is to solve complex situations and look for feasible solutions within a defined interval by the design dimensions. With the support of heuristic techniques such as neural networks, it was possible to find the sections that allow to obtain the characteristics of interest to carry out the study of the important regions of an image. The analysis and digital processing of images allows to smooth the file and to section the area of analysis in regions defined as rows and columns, results in a matrix of pixels, this way carrying out the measurement of the coordinates of the beginning and end of the region under analysis, taking it as a starting point for the creation of a frame of references to be examined. Once this requirement is completed, it is possible to return to the smoothed image with which the high edges of the image will be determined by means of the Gaussian function, thus finding the edges generated for the structures of interest.

show abstract

Using Satellite Imagery to Revolutionize Creation of Tax Maps and Local Revenue Collection

Cited by 4 publications

References 21 publications

Using Machine Learning to Create a Property Tax Roll: Evidence from the City of Kananga, D.R. Congo

Using Machine Learning to Create a Property Tax Roll: Evidence from the City of Kananga, D.R. Congo

Political Distortions, State Capture, and Economic Development in Africa

Design and Implementation of a System to Determine Property Tax Through the Processing and Analysis of Satellite Images

Contact Info

Product

Resources

About