Welcome!

This a website about my work and about me, Clemens Hoffmann. If you have any questions, please contact me.

European Electoral Law

As a European federalist, I advocate for the further democratization of the EU. In this context, I have focused extensively on the European electoral system and devised a way to implement reform without requiring unanimity in the Council.

The approach is based on the Tandem System proposed by Jo Leinen and Prof. Pukelsheim. In this system, the member parties of a European political party or a faction of the European Parliament form a transnational list coalition. The number of seats allocated to this list coalition is determined solely by its share of votes at the European level. These seats are then distributed among the member parties, leaving the seat quotas of individual member states unaffected.

The Tandem System can be modified so that opt-outs are allowed in distributing the list coalition’s seats to the member states. This means some member states distribute their seats as usual, while the participating member states ensure that the list coalitions receive the correct number of mandates. For this reason, I refer to it as a partial Tandem System.

This partial Tandem System does not rely on the participation of all member states. It can therefore be introduced via an intergovernmental supplementary treaty or the mechanism of enhanced cooperation without requiring the involvement of every member state.

Transnational lists can also be combined with the Tandem System. This is even possible with the partial system, where the member parties of a list coalition can start their lists with candidates from the transnational list. Upon successful election, candidates from the transnational list would thus receive “national” seats. An additional rule ensures that these are distributed evenly across the countries.

I give a simple explanation in the following video:

The partial Tandem System is also part of the resolutions of JEF Germany, JEF Europe and you can find an English article on the blog "Der (europäische) Föderalist".

The seat distribution under the Tandem System leads to different outcomes depending on whether, for example, electoral thresholds are introduced or diversity rules for the list coalitions are applied. With the following online calculator, you can test various scenarios.

Web App: Tandem System

Time Series Models

As part of my doctoral research, I investigate the effects of data aggregation on the analysis of price transmission processes. I use error correction models and examine how they change when data frequency decreases, e.g., from daily to weekly, or when multiple time series are merged, e.g., local prices aggregated into a national average price.

Temporal aggregation has been previously studied by Marcellino and Lütkepohl. Numerous issues arise when working with temporally aggregated time series. For instance, temporal aggregation can induce Granger causality: if two prices exist where one depends on the past values of both series while the other depends only on its own, this relationship is lost in aggregated data. Instead, both prices will appear to depend on all past values. The only component that remains unchanged is the long-term relationship between prices. In error correction models, these are the cointegration vectors. In general, it is not advisable to work with temporally aggregated data when the focus is on estimating short-term parameters.

Spatial aggregation or similar methods that reduce the number of variables in a time series model are less explored. Simply omitting variables has little effect on the remaining error correction model. Long-term relationships not involving the omitted variable remain unchanged. When aggregation is done via averages, the long-term relationship parameters correspond to average values. Overall, spatial aggregation appears less problematic, often resulting in offsetting effects.

Linear transformations like the aggregations considered here always produce autocorrelated error terms. Aggregated processes must therefore be estimated as ECVARMA models. Estimation functions and aggregation functions can be found in my R package. Visit my GitHub page to download the code.

Beyond aggregation, I have also worked with a smooth transitioning model. In a project for the World Bank, my advisor, a fellow PhD student, our World Bank contact, and I analyzed the behavior of various grain prices when world market prices skyrocketed. We found that markets become increasingly isolated, and international trade is restricted, which can negatively impact food security for poorer populations.

Published Articles

Hoffmann, C., & von Cramon-Taubadel, S. (2026). The Effects of Temporal Data Aggregation on Price Transmission Analysis. German Journal of Agricultural Economics, 75.

Hoffmann, C., Kastens, L., Portugal-Perez, A., and von Cramon-Taubadel, S. (2026). Food Price Crises and the Insulation of Domestic Grain Markets. Agricultural Economics 57, no. 4: e70125.

Conference Contributions

Hoffmann, C., & von Cramon-Taubadel, S. (2023). The Effects of Temporal Data Aggregation on Price Transmission Analysis. Presented at the NCCC-134: Conference on Applied Commodity Price Analysis, Forecasting, and Market Risk Management. St. Louis, 24.-25. April 2023.

Hoffmann, C., Kastens, L., Portugal-Perez, A., & von Cramon-Taubadel, S. (2024). Trade policies and the transmission of international to domestic prices. Presented at the International Conference of Agricultural Economists. New Delhi, 2.-8. August 2024

Web App: Partial Tandem

Note the web app needs 1-2 minutes to load and runs better in Chrome or Edge. The usage with mobile devices is possible.