An interview with… Fidel Picos
Please tell us a few things about your background and research interests.
I am BSc and PhD in Economics. From the beginning of my academic career my research has been focused on the study of tax reforms and the analysis of the effects of taxation. My PhD thesis was a study of the Nordic model of Dual Income Taxation and its possible implementation in Spain, using microsimulation techniques. This work opened for me a broader research line that included not only income taxation, but also environmental and tourism taxation or consumption taxes, mostly using microdata and microsimulation techniques. I'm also interested and have research experience on analysing inequality and redistribution.
Since 2007 I am an Associate Professor at the Department of Applied Economics of the University of Vigo (Galicia, Spain). I have also been a fellow researcher at the Instituto de Estudios Fiscales (IEF, Spanish Insitute for Fiscal Studies) and at the OECD Centre for Tax Policy and Administration. In October 2015 I took a special leave at my University to work at the European Commission's Joint Research Centre in Seville, where currently I hold a position of Economic Analyst in the Fiscal Policy Analysis Unit.
Where are you currently based, how long have you been there and what exactly are you doing?
During 2017 and part of 2018 I have been based in Athens, in the Greek Ministry of Finance. I came here in the context of the Memorandum of Understanding between the European Commission and Greece. The objective of my mission, funded by the EC’s Structural Reform Support Service, is to give technical assistance to the Tax Policy Directorate and the Council of Economic Advisors on the use and development of microsimulation tools, with a specific focus on administrative microdata.
From the very beginning of my stay EUROMOD was used for tax-benefit simulations in the context of the third review of the Greek Programme. But my main contribution has been probably the development of an alternative version of EUROMOD using input data based on 2016 personal income tax returns. Both models have been intensively used for simulating changes on actual or planned reforms of taxes (personal income tax, solidarity contribution, property tax) and benefits (family, housing, social dividend, social energy tariffs). Actually, the alternative model does not replace the standard EUROMOD based on EU-SILC data, but it is an excellent complement: depending on the type of reforms, we have used only one of the models, or even both of them. For example, for the Social Dividend passed in December 2017, the use of tax data was much more adequate because the Social Dividend was a one-off means-tested benefit based on the last tax return of taxpayers. In this way we were able to predict the cost with a deviation of around 1%.
Most of the simulations have been run for the Ministry of Finance and the Ministry of Labour. In order to provide these Ministries with a detailed and complete output of the simulations, I developed a set of Stata do-files that generate more than forty tables with information on revenue, expenditure, distributional impact, inequality and poverty, for up to 12 reform scenarios at the same time. These tables have been widely distributed among policy-makers and representatives of the international institutions, and have been the common ground for the discussions about the reforms implemented during the Greek programme. The experience was very encouraging because the involved parties actually analysed the results, especially those concerning at-risk-of-poverty and the targeting and adequacy when discussing benefit reforms, and the decisions were based on these indicators.
In parallel to the development and use of EUROMOD, I've organised several trainings so that the staff of the Ministry of Finance is able to use the microsimulation tools autonomously when the technical assistance ends.
How is EUROMOD used at your day job?
As explained before, I use EUROMOD on a daily basis in Greece. This intensive use has lead me to become much more knowledgeable in EUROMOD, since I had to model all types of reforms, and even introduce completely new benefits from scratch. During my stay here I've been in permanent contact with Chrysa Leventi, co-ordinator of the Greek EUROMOD national team. This has been very useful for me to understand better the modelling of the Greek tax-benefit system, but also for the EUROMOD team, since it was possible for me to get information from the Government that has proved very valuable for improving some of the simulations.
Regarding my work in JRC Seville, I use EUROMOD mostly for the European Semester, which is the EU's annual cycle of economic and fiscal policy coordination and surveillance. Every year the European Commission analyses the reform plans of the member states and provides them with country-specific recommendations, which are related to the Europe 2020 strategy. My unit contributes to this process by conducting in-depth analysis of tax-benefit reforms, using EUROMOD and some of its extensions developed to simulate other taxes (consumption, wealth), sometimes jointly with models to assess behavioural responses (labour supply, macroeconomic effects). The reforms we assess can be either hypothetical (usually reflecting the recommendations of the Commission or ongoing discussions in the countries) or actual (already implemented or approved). Our analyses are used in the Country Reports, which are the technical basis for the Country Specific Recommendations issued in May each year.
Please share some highlights of your experience at the Greek MinFin
I think that the experience has been extremely positive. I would like to highlight three things. First, the fact that EUROMOD (in its two versions) has become the key tool for tax-benefit reform analysis in the Ministry of Finance, in a context where microdata and microsimulation tools were hardly used. Second, EUROMOD has been accepted by all the parties involved in the negotiations of the Third Greek Programme between the international institutions and the Greek authorities. This means that the results of EUROMOD were the starting point of the discussions and not a matter of discussion themselves. And third, the availability of a large sample of personal income tax payers made it possible to build an alternative input dataset for EUROMOD, and also to create an alternative EUROMOD spine adapted to it. This alternative version allowed us to simulate with high accuracy some policies that cannot be simulated with EU-SILC data due to lack of information.
Finally, I'd like to highlight the exceptional opportunity of being part of real policy-making processes. Maybe the best example is the implementation of the 2017 one-off Social Dividend. The Council of Economic Advisors (with my assistance) was involved in all the stages of the process: design of the benefit, simulation and analysis of different alternatives, implementation of a web-based application and drafting of the ministerial decision to implement it. This experience made me realise how much this kind of processes differ from the academic projects in which I used to be involved, because here any decision taken may have a huge impact on the Government budget and on the society as a whole. And since many decisions have to be taken with extremely short deadlines, being fast and accurate at the same time is really challenging.