Notes from the IIPF Annual Congress
14 September 2016, by Peter Varela

I recently had the pleasure of attending the 2016 International Institute of Public Finance Annual Congress in Lake Tahoe, USA – this blog post is a short summary of the experience.

The Congress is the annual meeting of the IIPF, and was attended by 358 people from a variety of different backgrounds, including academics, public servants and members of international organisations such as The World Bank, the IMF and the OECD. 39 countries were represented, with the US (84 attendees) and Germany (75 attendees) and Japan (33 attendees) representing the largest contingents.

The theme of the congress was Entrepreneurship, Innovation and Public Policy with Keynote addresses from Roger Gordon, Robert W Fairlie, Dietmar Harhoff and Hal Varian.

In all there were 282 papers presented at the conference, and all of them were in some way related to the field of tax and transfer policy. A full list of papers and abstracts from the Congress can be found here.

My first major takeaway from the Congress was the quality of data utilised in the presentations. It was common to use hear of researchers using the ‘universe’ of personal or corporate tax returns (I also attended two different presentations in which researchers commented that they had access to a large sample of administrative records, but had to create a smaller sample because of computer processing issues).

There were also a number of papers that collected their own data, either through ‘scraping’ websites (see this paper by Eleanor Reed Wilking on the incidence of Airbnb taxes), or by historical research (such as this paper by Olga Malkova that used declassified Soviet Union planning data to examine the effect of a large change in pension policy in the 1960s).

My second takeaway is that a significant number of papers used this new access to data to examine the effect of certain administrative features of the tax system. These features are often overlooked in analysis because in standard optimizing models they have no impact.

For instance, Anne Brockemeyer from the World Bank used administrative sale tax records from Costa Rica to examine whether the rate of withholding applied to businesses affected the final rate of tax paid (it does, and it lead to a large increase in tax revenue).

Another example came from Youssef Benzarti of UC Berkley who used bunching estimator techniques to examine the number of people who file tax returns just above the standard deduction in the US. This paper found that there were very few people who claim in this range, and shows that the implied cost of filing in an itemized tax return in the US is substantial.

Read the full article at Austaxpolicy blog.