Sunday, January 22, 2017

EU Economy - Revenue elasticities measure the reaction of public revenues to changes in their base- Relying on static revenue elasticities for cyclical adjustment is likely to lead to systematic biases especially for those countries for which we found large differences between short- and long-run elasticities (i.e. Spain, Greece, Luxembourg, Malta and Cyprus). -

Publication - Revenue elasticities in euro area countries An analysis of long-run and short-run dynamics



Non-technical summary

Revenue elasticities measure the reaction of public revenues to changes in their base. They play a key role in forecasting, monitoring and analysing public finances. They are at the core of cyclical adjustment methods of public finances and are therefore especially important under the European fiscal framework, which largely builds on cyclically adjusted indicators. 


This paper investigates whether there is evidence for dynamic – instead of the frequently used static – elasticities in euro area countries. While static elasticities assume that revenues only react to changes in their base in the contemporaneous period, dynamic elasticities allow for time lags in the reaction of revenues to changes in their base and hence also for a variation between the short-run and long-run revenue elasticities. 

Based on country-specific error correction models we find evidence for dynamic revenue elasticities in a majority of euro area Member States (12 out of the 18 countries analysed). Relationships between macroeconomic and public revenue developments are found to be strongly dynamic in Spain, Greece, Luxembourg, Malta and Cyprus. A second group of countries with smaller, but nevertheless significant dynamics in the relationship between GDP and public revenues includes Estonia, Belgium, Austria and Italy. Only small differences between short- and long-term elasticities and hence very limited dynamics are found for Slovenia, Finland and Latvia. In quantitative terms, the difference between the short-run and the long-run revenue elasticity varies between 0.7 (in the case of Spain) and 0.1 (as in the case of Latvia). For six countries (France, Ireland, Germany, Portugal, the Netherlands and Slovakia) however, we find no evidence for dynamic revenue elasticities. Relying on static revenue elasticities for cyclical adjustment is likely to lead to systematic biases especially for those countries for which we found large differences between short- and long-run elasticities (i.e. Spain, Greece, Luxembourg, Malta and Cyprus). 


Our analyses reveal that adjustment patterns between short- and long-run elasticities also differ strongly across countries and need to be taken into account for forecasting and cyclical adjustment of public finances. In a majority of those countries with evidence for dynamic revenue elasticities (Greece, Luxembourg, Malta, Estonia, Belgium, Austria, Italy, Slovenia and Finland), the short-run elasticity is found to be below the long-run elasticity leading to an “undershooting” of revenues after a shock in aggregate income. For three countries (Cyprus, Spain and Latvia) short-run elasticities are significantly higher than long-run elasticities – leading to an overshooting of revenues in the short run. 



A comprehensive out-of-sample forecast evaluation based on an ex-post dataset shows that applying the identified dynamic elasticities (instead the frequently applied static ones) reduces revenue forecast errors for all euro area Member States except for Italy. For three of the five countries with strongly dynamic revenue elasticities (Cyprus, Greece and Luxembourg) the forecast evaluation shows that a dynamic model performs significantly better than a model with static elasticities. Moreover, the dynamic models consistently outperform benchmark random walk forecasts in all euro area countries. Based on real-time data, random walk benchmark forecasts turn out to be generally harder to beat and there is less evidence for the superiority of dynamic over static elasticities for current year forecasting. Taken together our findings indicate that the dynamic revenue elasticities derived by us could in several countries help to improve cyclical adjustment methods and could also help to substantially reduce forecast errors – with the evidence being substantially stronger based on ex-post than on real-time data.




Copyright: European Central Bank
Directorate General Communications 
Sonnemannstrasse 20, 60314 Frankfurt am Main, Germany 
Tel.: +49 69 1344 7455, E-mail: media@ecb.europa.eu


page source http://www.ecb.europa.eu/pub/