A Data-Driven Supply-Side Approach for Measuring Cross-Border Internet Purchases

TL;DR

This paper introduces a novel supply-side, data-driven method using tax records and machine learning to accurately measure cross-border internet purchases in the EU, revealing significantly higher values than previous estimates.

Contribution

It presents a new approach leveraging tax data, business registers, and internet data, overcoming survey biases to improve measurement of cross-border online shopping.

Findings

Cross-border internet purchases in the EU by Dutch consumers were over EUR 1.3 billion in 2016.

This value is more than six times higher than previous estimates.

The methodology can be adapted for other EU countries to improve overall measurement accuracy.

Abstract

The digital economy is a highly relevant item on the European Union's policy agenda. Cross-border internet purchases are part of the digital economy, but their total value can currently not be accurately measured or estimated. Traditional approaches based on consumer surveys or business surveys are shown to be inadequate for this purpose, due to language bias and sampling issues, respectively. We address both problems by proposing a novel approach based on supply-side data, namely tax returns. The proposed data-driven record-linkage techniques and machine learning algorithms utilize two additional open data sources: European business registers and internet data. Our main finding is that the value of total cross-border internet purchases within the European Union by Dutch consumers was over EUR 1.3 billion in 2016. This is more than 6 times as high as current estimates. Our finding motivates the implementation of the proposed methodology in other EU member states. Ultimately, it could lead to more accurate estimates of cross-border internet purchases within the entire European Union.