Paul Johnson, Steven Durlauf and Jonathan Temple 1153and has been explored in depth for the growth context by Lee, Pesaran and Smith
(1997, 1998). Since an absence of serial correlation in the disturbances is usually a
critical assumption for the GMM approach, parameter heterogeneity can be a seri-
ous concern. Some of the possible solutions, such as regressions applied to single
time series, or the pooled mean group estimator developed by Pesaran, Shin and
Smith (1999), have limitations in studying growth for reasons already discussed.
An alternative solution is to split the sample into groups that are more likely to
share similar parameter values. Groupings by regional location or the initial level
of development are a natural starting point.
Perhaps the state of the art in analyzing growth using panel data and allowing for
parameter heterogeneity is represented by Phillips and Sul (2003). They allow for
heterogeneity in parameters not only across countries, but also over time. Temporal
heterogeneity is rarely investigated in panel studies, but may be important.
One drawback of many current panel studies is that some of the necessary deci-
sions, and perhaps especially the construction of the time series observations, can
appear arbitrary. There is no inherent reason why five or ten years represent natu-
ral spans over which to average observations. Similarly, there is arbitrariness with
respect to which time periods are aggregated. A useful endeavor would be the devel-
opment of tools to ensure that panel findings are robust under alternative ways of
assembling the panel from the raw data; it is also possible that the field could draw
more heavily on the econometric literature on time aggregation than it does at
present.
More fundamentally, the empirical growth literature has not fully addressed the
question of the appropriate time horizons over which growth models should be
assessed. For example, it remains unclear when business cycle considerations, or
instances of output collapses, may be safely ignored. While cross-section studies
that examine growth over 30–40-year periods might be exempt from these consid-
erations, it is less clear that panel studies employing five-year averages are genuinely
informative about medium-run growth dynamics. As more data become available
with the passage of time, concerns over the scope for arbitrary choices can only
increase. It will also be important to develop robust methods for inference about
long-run effects.
24.5.3 The event study approach
Although we have focused on the limitations of panel data methods, it is clear
that the prospects for informative work of this kind should improve over time.
The addition of further time periods is valuable in itself, and the history of devel-
oping countries in the 1980s and 1990s offers various events that introduce richer
time series variation into the data. These events include waves of democratization,
macroeconomic stabilization, financial liberalization and trade reform, and panel
data methods can be used to investigate their consequences for growth. This can
proceed in a similar way to event studies in the empirical finance literature. In
event studies, researchers look for systematic changes in asset returns after a dis-
crete event, such as a profits warning. In other fields, before-and-after studies like
this have proved an informative way to gauge the effects of inflation stabilization,