Barcelona GSE Lecture – Using Internet Data to Understand Consumers and Markets

Lecture summary by masters students Hugo Kaminski ’15 and Yi-Ting Kuo ’15.


During the 31st Barcelona GSE Lecture, Professor Jonathan Levin, Holbrook Working Professor of Price Theory at Stanford University, discussed the advantages and challenges of using Internet commerce data for empirical research in economics.

Seller Experience – Let real online sellers run experiments for us

Economic research has long relied on public governmental institutions and organisations to collect empirical data, which are reliable but expensive. Data on the Internet could potentially be an alternative source as it can reduce the cost of varying parameters. However, it is also challenging to isolate specific effects as customization of products and services raises concerns about selection and endogeneity.

Prof. Levin and his collaborators found that the millions of listings on eBay could potentially serve as millions of small experiments with different seller choices. In collaboration with eBay, US data allowed them to run fixed effect regressions exploiting within-experiment variation in prices, fees, displays, and other parameters.

Auction and Demand – Never start with an intermediate price

The buyers market is very competitive. If an auction starts at a low price, the item will be highly sellable and the market will drive it to around 80% of its value. Starting at a high price will make the item harder to sell, but increases the probability of ending with a higher price. The resulting estimated demand curve was convex, which implies that for a seller to start an auction with either a low or high price is more profitable than an intermediate price.

As second example, Prof . Levin argued that the Internet market could be used to test behavioural hypotheses about consumers. By looking at the multiple auction listings with different (flat rate) shipping fees, their analysis suggests that people prefer free shipping so much that they are willing to pay a higher price for the goods with free shipping even if there is an equal good with lower total price including shipping costs.

By analysing Internet commerce data from 2003 to 2012, they observed the sellers’ learning curve and concluded that the demand for online auction has declined. One interpretation of this development could be that consumers grew accustomed to increasingly instant purchase options and tend to spend more time on other online activities thus losing the attention for online auction.

Sales tax – a problem across US states

In the United States there exists a sales tax of 8.875% on average among states for within state sales. It does not come as a surprise that the payment of this tax is viewed as a negative additional charge which online often appears just towards the end of the purchasing process. To analyse the impact of the ‘tax surprise’ Prof. Levin estimates the tax sensitivity by comparing purchase rates.

Prof. Levin’s research finds that both consumers and sellers are aware of this perceived “extra charge” and the fact that the tax does not apply on out-of-state sales. Their analysis finds two consumer trends: first there exists a preference for goods bought in geographic proximity; second, the rising item-level substitution which means the consumer chooses to buy the item from an out-of-state seller to avoid the tax. These two trends seem to form a paradox.

Professor Nezih Guner (right) with our speaker
Professor Nezih Guner (right) with our speaker

On the seller side, Prof. Levin illustrates the case of Amazon and the location of their distribution centres. Amazon takes advantage of California’s geographic shape by locating its distribution centres just outside state lines while keeping delivery times short. This results in being able to cater to consumers in California without paying the sales tax.

Since online shopping is a growing trend, taxation legislation changes have potentially great impact. Based on the data used, a 1% increase in current sales tax decreases online sales by 1.5-2.0% but increase online home-state sales by 3-4%; alternatively switching to national tax collection of internet purchases would decrease online sales by 12%.

Professor Levin’s presentation concluded that new large-scale data offers an opportunity to assess microeconomic theories of behaviour and market operation. The presentation was followed by a Questions & Answers session on data quality, auction information asymmetry and lessons learned from use of auctions. For Yi-Ting Kuo and Hugo Kaminski it has been an insightful experience to listen to Professor Levin’s talk presented with support from Banco Sabadell.

If you are interested to know more about the lecture, you may view it here:

Can big data be official? – Barcelona GSE Data Scientists

Originally posted by Stefano Costantini ’15 on the Barcelona GSE Data Scientists blog. Stefano is on Twitter @stefanoc.

Originally posted by Stefano Costantini ’15 on the Barcelona GSE Data Scientists blog. Stefano is on Twitter @stefanoc.

At the Renyi Hour on November 13th 2014, Frederic Udina gave a talk on big data and official statistics. Apart from being a professor at UPF and BGSE, Frederic is Director of IDESCAT, the statistical institute of Catalonia.

Frederic Udina presenting to BGSE Data Science students
Frederic Udina presenting to BGSE Data Science students

In his talk, Frederic compared the “traditional” official statistics – slow to produce, with well-defined privacy limits and access rights – to “big data”, which is fast to produce, volatile and with fuzzy privacy limits. Frederic highlighted the tension between these two worlds, focusing particularly on the need for official statistics to become easier to collect, organise and customise to the need of the final user. In particular, Frederic identified the opportunity for IDESCAT (and other statistical institutes) to integrate the officially collected information with alternative information sources, such as:

  • Administrative data
  • Data freely available from the society
  • Data from private companies

Frederic outlined IDESCAT’s plan to move away from the current data generation system (the ‘stove pipe model’) which is slow, expensive and inefficient as it does not re-use information already collected, towards a fully integrated model (‘Plataforma Cerdà’) where any new information needs to be integrated with existing data.

The Renyi hour crowd
The Renyi hour crowd

Frederic noted that data is becoming increasingly important in society, and this is beginning to be recognised by official statistical institution. In particular, Frederic discussed the Royal Statistical Society’s Data manifesto where the RSS notes that data is:

  • A key tool for better, informed policy-making
  • A way to strengthen democracy and trust
  • A driver of prosperity.
The Royal Statistical Society Data Manifesto
The Royal Statistical Society Data Manifesto

Frederic also stressed the importance of confidentiality and privacy issues with regards to data availability. While it is desirable for some data to be freely available to the public, confidentiality and privacy should always be protected. However, it is important to strike the right balance between access and privacy, ensuring that while personal sensitive data is protected, important information is not prevented from being used in ways that may ultimately help the wider society. Personal health records are a classic example of this.

Frederic concluded his talk by providing some example of national statistical authorities integrating official statistics with widely available information to carry out new interesting analysis. Examples include:

  • Production of origin/destination arrays between territorial units (usually municipalities) for working or studying reasons using trajectories of mobile phones (ISTAT, New Zealand Statistics)
  • Using Google Trends to estimate/predict labour market, monthly forecast, small-area estimation (ISTAT)
  • Measuring use of TCI in firms, by using web scraping and text mining techniques
Lunch with Frederic after his talk
Lunch with Frederic after his talk

Useful links: