Several limitations of the data used in this paper should be mentioned. We found
extreme temperatures do impact agricultural output positively, but this output could
be volatile, which implies that if other years’ data were to be used, the output may
be negatively impacted (in this paper, agricultural and nonagricultural output for
2003–04 was used, and exogenous variables were used for 2001, for a lagged
effect). It is to be noted that extreme temperatures in this paper are measured by the
difference between average maximum and mean minimum temperatures at the level
of the district, not at the level of the city, due to the selection problem which was
discussed earlier.
It should also be noted that climate change might be a threat if it affects mean
maximum and minimum temperatures eventually, many studies indeed confirm that
this is the case (see Deschênes and Greenstone 2007). If there are other forms of
climate change, such as the risk of flooding, unseasonal rainfall, and related events,
which do not significantly impact temperatures, then the findings in this paper
would not throw light on those aspects of climate change and their impacts.
The other limitation we are faced with is that several states did not report data on
DDP for their districts for various reasons. Only when these data are complete, we
would be able to say something definitive about the economic impacts of climate
change.
Further, the data on rainfall and temperature differences are from the Census of
India’s town directory only for 2001. A better indicator to use may have been
percentage changes in temperature over a long period of time to understand whether
global warming has been really taking place. We expect that at least in the relatively
more industrialized cities, global warming would be much more pronounced.
Unfortunately the data on minimum temperatures required to make an assessment
are not available free of cost from the Indian Meteorological Department.
Additional funding would be required to obtain these data.
Finally, the estimations were conducted at the level of districts because there are
econometric problems with the choice of cities as the unit of analysis (since only
growing areas are designated as cities).Hence for purposes of the district-level data,
the average rainfall and temperature differences are averaged for all cities within a
district. Obviously future research in this area should build upon the limitations of
this work.
References
Beeson, P., DeJong, D. N., & Troesken, W. (2001). Population growth in U.S. counties, 1840–
1990. Regional Science and Urban Economics, 31, 669–699.
Cohen, B. (2004). Urban growth in developing countries: A review of current trends and a caution
regarding existing forecasts. World Development, 32(1), 23–51.
Deschênes, O., & Greenstone, M. (2007). The economic impacts of climate change: Evidence
from agricultural output and random fluctuations in weather. The American Economic Review,
97(1), 354–385.
294
K.S. Sridhar
extreme temperatures do impact agricultural output positively, but this output could
be volatile, which implies that if other years’ data were to be used, the output may
be negatively impacted (in this paper, agricultural and nonagricultural output for
2003–04 was used, and exogenous variables were used for 2001, for a lagged
effect). It is to be noted that extreme temperatures in this paper are measured by the
difference between average maximum and mean minimum temperatures at the level
of the district, not at the level of the city, due to the selection problem which was
discussed earlier.
It should also be noted that climate change might be a threat if it affects mean
maximum and minimum temperatures eventually, many studies indeed confirm that
this is the case (see Deschênes and Greenstone 2007). If there are other forms of
climate change, such as the risk of flooding, unseasonal rainfall, and related events,
which do not significantly impact temperatures, then the findings in this paper
would not throw light on those aspects of climate change and their impacts.
The other limitation we are faced with is that several states did not report data on
DDP for their districts for various reasons. Only when these data are complete, we
would be able to say something definitive about the economic impacts of climate
change.
Further, the data on rainfall and temperature differences are from the Census of
India’s town directory only for 2001. A better indicator to use may have been
percentage changes in temperature over a long period of time to understand whether
global warming has been really taking place. We expect that at least in the relatively
more industrialized cities, global warming would be much more pronounced.
Unfortunately the data on minimum temperatures required to make an assessment
are not available free of cost from the Indian Meteorological Department.
Additional funding would be required to obtain these data.
Finally, the estimations were conducted at the level of districts because there are
econometric problems with the choice of cities as the unit of analysis (since only
growing areas are designated as cities).Hence for purposes of the district-level data,
the average rainfall and temperature differences are averaged for all cities within a
district. Obviously future research in this area should build upon the limitations of
this work.
References
Beeson, P., DeJong, D. N., & Troesken, W. (2001). Population growth in U.S. counties, 1840–
1990. Regional Science and Urban Economics, 31, 669–699.
Cohen, B. (2004). Urban growth in developing countries: A review of current trends and a caution
regarding existing forecasts. World Development, 32(1), 23–51.
Deschênes, O., & Greenstone, M. (2007). The economic impacts of climate change: Evidence
from agricultural output and random fluctuations in weather. The American Economic Review,
97(1), 354–385.
294
K.S. Sridhar