Central European Journal of Economic Modelling and Econometrics

The spectacular surge of Poland's VAT revenues after 2015 prompted a discussion about the role of the tax administration in collecting tax liabilities. Unfortunately, the scarcity of the available data prevents empirical studies from reaching reliable conclusions about the determinants of VAT revenues. Given that, this article presents a wider attempt at identifying the determinants of VAT revenues in the EU Member States. Using panel cointegration methods, several working hypotheses linking VAT gap to income factors, the business cycle, tax carousels, and an effectiveness of the government are evaluated. The results of the research provide evidence that the VAT gap in the EU countries is under a strong influence from variables approximating changes in per capita incomes, the business cycle, and the openness of an economy to intra-EU trade. The latter finding is a sufficient indication that the improvements made to Poland's tax system were both legitimate and effective.

In this work, we perform an analysis of the characteristics of the one-year and ultimate reserve risk distributions commonly used in actuarial science: duration, first development factor, coefficient of variation, skewness coefficient, skewness-to-CoV ratio, emergence factor, emergence pattern, and risk margin run-off patterns. Our study is based on empirical data for two European markets: the Polish and Slovak markets. We provide benchmarks and ranges for the considered characteristics, as well as analyse the relations between them. We study Solvency II lines of business and compare our coefficients of variation to the Standard Formula reserve risk standard deviations. We investigate more deeply the topic of emergence pattern and risk margin run-off patterns.

The purpose of this paper is to introduce and study a new generated family of distributions based on the type II transformation which is called the type II exponentiated half-logistic-Gompertz-Topp-Leone-G (TIIEHL-Gom-TL-G) family of distributions. We investigate its general mathematical properties, including, hazard rate function, quantile function, moments, moment generating function, Rényi entropy and order statistics. Parameter estimates of the new family of distributions are obtained based on the maximum likelihood estimation method and their performance is evaluated via a simulation study. For illustration of the applicability of the new family of distributions, four real data sets are analyzed.