Is Saudi Real GDP difference or trend stationary? evidence and implications

Abstract EN

This study is an attempt to analyze the time-series behavior of the kingdom's real GDP covering the period spanning 1980-2011.

behavior in this context refers to the underlying stochastic process that characterizes the GDP series over this period.

more specifically, the focus is primarily on the extent to which GDP behavior over this period may be characterized by the presence of a deterministic time trend process (known as trend stationary process, tsp) or alternatively by the presence of stochastic trend process (known as difference stationary process, DSP).

these two processes have different time series properties with important economic implications and consequences.

building on the work of nelson and plosser (1982) and employing ADF univariate unit root test, our analysis shows that movement in the natural log of KSA real GDP is characterized by a difference-stationary process (DSP) rather than evolving as trend-stationary process (tsp).

the KPSS null stationarity test is also performed on the same set of data.

results from KPSS test are consistent with those obtained by ADF test providing further evidence for our findings.

our finding implies that temporal shocks to real GDP tend to have permanent effects.

the economic implication is that supply or real shocks being random in nature, such as technology shocks, are contributing much more to output path relative to nominal demand shocks.

the conclusion reached is consistent with findings reported by nelson and plosser (1982) that us GDP is difference-stationary i.e.

has a unit root in its AR representation.