Land Mines and Spatial Development< /strong> (with Giorgio Chiovelli and Stelios Michalopoulos)
Abstract:
We explore the economic consequences of
demining for both the affected districts and to the Mozambican economy as
a whole after collecting and digitizing data covering close to the universe
of demining activities in the country.
Our analysis proceeds in four
steps. First, we describe the self-collected and cross-validated data on t
he spatial distribution of land mines and unexploded ordnances (UXOs) at th
e end of the civil war (1992); we then provide fine geo-referenced informat
ion on land mine clearance operations. Thus we are able to provide to the M
ozambican government, the United Nations and the various NGOs involved in t
he demining process, a (almost) complete documentation of this gigantic pro
cess that lasted more than two decades.
Second, we exploit time varia
tion in the demining process -that appears to be largely non-coordinated an
d non-strategically planned- to assess its impact on economic activity, as
reflected on satellite images on light density at night. In this regard, we
estimate “difference-in-difference” specifications, that compare the evolu
tion of local development in areas (Mozambican municipalities), where demin
ing took place in a given period, to municipalities that either had no land
mines at the end of the civil war or municipalities that were mined but ha
d not been cleared. The analysis reveals small-to-moderate effects of the r
emoval of land mines on local economic development, as reflected in luminos
ity. We also examine heterogeneous effects in an effort to shed light on th
e mechanisms and guide policy on demining activities at other parts of the
world. Our tentative evidence points out that demining is associated with l
arge gains when it targets roads-railroads, areas with non-negligible popul
ation densities, and towns serving as local trade hubs.
Third, employ
ing recently developed techniques in quantitative trade literature, we exam
ine the economy-wide implications of land mine removals. This is crucial as
there are spillovers, general equilibrium effects, from the removal of lan
d mines in one district to others. Accounting for spillovers and estimating
the economy-wide effects of land mine clearance is also key for policy, as
the interest is typically on the aggregate impact. Our approach combines i
nformation on the evolution of land mine clearance with the transportation
network (roads, railroads and rivers) at independence. We apply a “market a
ccess” approach that quantifies the aggregate effects of land mines (and th
eir subsequent removals) on spatial development. This method, which is deri
ved from general equilibrium trade theory under flexible and widely-used as
sumptions, is well-suited to our research inquiry and application, as it al
lows us to shed light both on the direct and the indirect effects (stemming
from changes in the network structure) of land mine clearance and opening
of roads and railroads on economic activity. The “market access” estimates
reveal an economically large and statistically precise impact of land mines
removal on aggregate economic development.
To push on the causal int
erpretation of our estimates, we associate changes in development (luminosi
ty) to changes in “market access” coming from land mines cleared outside a
buffer zone of 50 or even 100 kilometers; this minimizes concerns that NGOs
and local agencies target districts those with high economic potential. We
then report results linking changes in regional development and changes in
market access in the set of Mozambican regions without any minefield and U
XO contamination at the end of the civil war. For these districts, changes
in market access stem solely from land-mine clearance in other regions. The
correlation between regional development and market access is quite strong
and the estimates are quantitatively similar to those derived in the full
sample.
Fourth, combining the spatial general equilibrium model of in
tra-country trade and the econometric estimates, we run counterfactual poli
cy simulations that allow us estimating the likely gains of demining if it
was centrally coordinated and planned targeting the key colonial developmen
t “corridors” or the roads and railroads connecting the three main ports. W
e also approximate the gains if demining was “optimally” targeting the main
hubs of the infrastructure system. Our analysis reveals large losses from
lack of central planning and strategic coordination among the demining oper
ators.