Quarterly Estimates for Italy's Government Accrual Data

Abstract

In this paper we calculate quarterly estimates for ten yearly budgetary items of the Conto Economico Consolidato della Pubblica Amministrazione (CECAP) published by ISTAT, covering the period from 1980 to 2007. The unknown high frequency pattern of fiscal variables is derived by the pattern observed of quarterly related series available in the CON-ISTAT database. We apply a dynamic extension of the Chow-Lin (1971) temporal disaggregation model, which is the method actually used by ISTAT for the estimation of quarterly national economic account data. As dealing with flow variables, the time constraint underlying the model imposes that the sum of quarterly estimates is equal to the yearly observation. The estimation strategy is based on the state space representation of a first order Autoregressive Distributed Lag model (Adl), which transforms the distribution problem into an unknown observation one. The choice of data assures that our quarterly estimates are fully comparable with quarterly government accrual data of major industrialized countries based on ESA95 and SNA93, both for institutional coverage, accountancy rules and definition of budgetary items.