This study examines the dynamic linkages between crude petroleum imports and GDP of Turkey. The vector autoregression analysis is carried on quarterly data for the period 1998Q1 to 2013Q2. This study utilized the generalized approach to forecast error variance decomposition and impulse response analysis which have many advantages against the traditional orthogonalized approach. The empirical results suggest that petroleum imports have positive impact on GDP until the second quarter. But, after the second quarter crude petroleum imports have negative impact on GDP. The results of the Granger causality test showed that crude petroleum imports granger caused GDP at 5% significance level, but not vice versa. Moreover, the generalized variance decomposition analysis exerted that the imports of crude petroleum shocks have only a small effect on GDP initially. However, after eighth quarters, the imports of crude petroleum shocks explain 31.7 pct. of the GDP, whereas 26.46 pct. of the variation in imports of crude petroleum shocks is explained by GDP shocks.