On the accuracy of cge forecasts in expansion and recession: spain 1990–1997

Clemente Polo, Raimundo Viejo

Research output: Contribution to journalArticleResearchpeer-review

4 Citations (Scopus)


© 2015 The International Input–Output Association. A recursive dynamic disaggregated computable general equilibrium model of the Spanish economy is used to compare the model predictions of endogenous variables with their observed values over the period 1991–1997. It includes 12 producers, 12 households, government and 2 external sectors. There are four types of labour and real wages that depend on unemployment rates. Private investment is determined by private savings and public and external surpluses. Domestic products and imports are imperfect substitutes. All exogenous variables and tax parameters are updated every year with the best available information. The model provides rather accurate predictions in 1991, a normal year, but it underestimates the intensity of the 1992–1993 recession. It also predicts dramatic reversals of trade balances in response to devaluations. These results suggest both that investment savings-driven models provide useful insights in the medium term but underestimate the consequences of downturns, and that Armington's elascitities typically assumed may be too large.
Original languageEnglish
Pages (from-to)525-542
JournalEconomic Systems Research
Issue number4
Publication statusPublished - 2 Oct 2015


  • Computable general equilibrium
  • Forecasting
  • Model validation


Dive into the research topics of 'On the accuracy of cge forecasts in expansion and recession: spain 1990–1997'. Together they form a unique fingerprint.

Cite this