In sum, while we agree with Bohn’s (2007) critique of the use of purely time
series methods to test the IBC, we nonetheless believe time series properties of the
data can not be completely ignored and can provide a useful testing mechanism
especially when other factors which may have an important bearing on a test of fiscal
sustainability are brought to bear. In our study we believe that the interrelationship
between fiscal sustainability and the globalisation of international capital markets is a
crucial one. To account for common shocks to fiscal sustainability, which make
standard least squares inconsistent due to induced spurious regression problem from
unobserved stochastic trends, we apply, in particular, the Bai, Kao and Ng (2007)
panel cointegrating regression estimator which deals with cross sectional correlation
by utilising a factor model approach. We therefore see our approach as in the spirit of
Bohn (1998, 2007) who advocated estimating primary surplus response functions.
3. Econometric Methods
In this paper we utilise nonstationary time series methods to explore the
relationship between the primary surplus and debt and a factor model to elaborate the
importance of global capital market shocks in explaining the lack of fiscal
sustainability in emerging markets. Specifically, we test for a unit root in the variables
using the Bai and Ng (2004) PANIC test, and use panel time series estimators from
Bai and Kao (2006) and Bai, Kao and Ng (2007) to test the primary surplus
relationships. The former estimator adjusts for cross sectional correlation using a
stationary factor, whilst the latter contains a nonstationary factor. We now sketch our
empirical methods.
10