Extreme Weather Events and Economic Activity: The Case of Low Water Levels on the Rhine River

Martin Ademmer; Nils Jannsen; Saskia Meuchelböck

doi:10.1515/ger-2022-0077

Article

Extreme Weather Events and Economic Activity: The Case of Low Water Levels on the Rhine River

Martin Ademmer , Nils Jannsen and Saskia Meuchelböck

Published/Copyright: March 14, 2023

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information Explore this Subject

From the journal German Economic Review Volume 24 Issue 2

Abstract

We make use of historical data on water levels on the Rhine river to analyze the impact of weather-related supply shocks on economic activity in Germany. Our analysis shows that low water levels lead to severe disruptions in inland water transportation and cause a significant and economically meaningful decrease of economic activity. In a month with 30 days of low water, industrial production in Germany declines by about 1 percent, ceteris paribus. Our analysis highlights the importance of extreme weather events for business cycle analysis and contributes to gauging the costs of extreme weather events in advanced economies.

Keywords: climate; extreme weather events; supply shocks; business cycle effects; transportation disruption

JEL Classification: E32; Q54

Corresponding author: Saskia Meuchelböck, Kiel Institute for the World Economy, Kiellinie 66, 24105 Kiel, Germany, E-mail: saskia.meuchelboeck@ifw-kiel.de

Acknowledgments

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. We thank the German Federal Institute of Hydrology (BfG) for providing the data on water levels on the Rhine river. We are also highly grateful for their explanations on the basics of hydrology. Moreover, we thank the Helmholtz Centre for Environmental Research for providing data on droughts in Germany. Finally, we thank two anonymous referees and the editor for very helpful comments and suggestions.

Declarations of interest: None.

Appendix A1: Variable Description and Summary Statistics

Table A1.1:

Variable description.

Variable	Description	Source
Low water (LW)	Number of days recorded in a month on which the water level measured at the Kaub gauging station was lower than 78 cm, an officially defined low water level threshold that serves as a benchmark for navigation	German Federal Institute for Hydrology
Industrial production	Monthly data on industrial production in Germany; excludes energy production by definition	Federal Statistical Office of Germany
Industrial production by sectors	Monthly data on industrial production by sectors	Federal Statistical Office of Germany
Global industrial production	GDP-weighted average of monthly industrial production in the 46 largest economies in the world	National statistical offices, own calculations
Water temperature	Water temperature (in degree Celsius) of the Rhine as measured near the city of Mainz. This location is relatively close to main power plants and industrial facilities in Germany. Since the data are only provided as 26 two-week averages per year and the measurement periods are somewhat irregular, we use for each month the data on the first two-week average available.	German Federal Institute of Hydrology
Air temperature	Average monthly air temperature in Germany (in degree Celsius)	German Meteorological Service (Deutscher Wetterdienst, DWD)
Rainfall	Average monthly amount of rainfall (in millimeters)	German Meteorological Service (Deutscher Wetterdienst, DWD)
Drought	Percentage of total area in Germany with a soil moisture index (SMI) below 0.2 in a given month	Drought Monitor Germany, Helmholtz Center for Environmental Research; Zink et al. (2016)
Drought intensity	Spatial mean of monthly drought intensity (measured as negative deviation from an SMI of 0.2)	Drought Monitor Germany, Helmholtz Center for Environmental Research; Zink et al. (2016)
Global trade	Monthly data on global trade in goods	IMF
Volume of inland water transportation	Monthly data, in tons	Federal Statistical Office of Germany
Volume of road transportation	Monthly data, in tons	Federal Statistical Office of Germany
Volume of rail transportation	Monthly data, in tons	Federal Statistical Office of Germany

Table A1.2:

Summary statistics.

Variable	N	Mean	Std. dev.	Min	Max
∆ days with low water	338	0.00	4.54	−27.00	30.00
Days with low water	339	1.04	4.27	0.00	30.00
ln(water level)	339	5.33	0.40	3.69	6.20
Growth in industrial production	338	0.09	1.42	−6.88	3.98
Growth in volume of inland water transportation	338	0.21	7.44	−30.25	44.49
Growth in volume of road transportation	263	0.44	8.80	−21.36	33.69
Growth in volume of rail transportation	170	0.22	5.65	−13.05	22.20
Growth in global industrial production	338	0.17	0.57	−3.52	1.45
Growth in global trade	338	0.75	6.95	−18.82	20.57
Positive water temperature anomalies	339	0.40	0.60	0.00	2.54
Temperature anomalies	339	−0.01	0.98	−2.74	2.24
Rainfall anomalies	339	0.01	0.99	−2.30	3.64
Drought	339	19.18	21.81	0.00	99.66
Drought intensity	339	0.02	0.03	0.00	0.18

Appendix A2: Specification in Differences versus Levels

In this Appendix, we compare a specification with the number of days ( L W t − j ) as explanatory variable to our baseline specification with the change in the number of days ( Δ L W t − j ) as explanatory variable. The former specification is given by

(A1) y t = α + ∑ j = 0 J θ j L W t − j + γ y t − 1 + ϵ t .

Table A2:

Low water levels and industrial production (differences vs. levels).

	Growth in industrial production
	(1)	(2)	(3)	(4)	(5)	(6)
∆ days with low water	−0.026^b	−0.029^b	−0.038^a
	(0.012)	(0.013)	(0.012)
∆ days with low water, t – 1			−0.026^c
			(0.014)
Days with low water at Kaub				−0.029^a	−0.040^a	−0.041^a
				(0.010)	(0.011)	(0.011)
Days with low water at Kaub, t – 1				0.023	0.019	0.010
				(0.018)	(0.017)	(0.020)
Days with low water at Kaub, t – 2						0.023
						(0.015)
Growth in industrial	−0.166	−0.346^a	−0.357^a	−0.167	−0.351^a	−0.357^a
production, t – 1	(0.117)	(0.043)	(0.042)	(0.119)	(0.043)	(0.042)
Growth in global		1.188^a	1.177^a		1.190^a	1.179^a
industrial production		(0.107)	(0.106)		(0.106)	(0.106)
Growth in global		0.529^a	0.553^a		0.541^a	0.555^a
industrial production, t – 1		(0.171)	(0.169)		(0.171)	(0.170)
Obs.	337	337	337	337	337	337
AIC	1183	1056.1	1055	1184.9	1056.7	1056.8
Adj. R ²	0.030	0.338	0.342	0.027	0.339	0.341

A constant is included in all regressions. Robust standard errors are given in parentheses. ^a/^b/^c indicates statistical significance at the 1%/5%/10% level.

The resulting estimates are shown in Table A2. The first three columns show for comparison the results for specifications with the first difference of the number of days with low water levels. Columns 1 and 2 illustrate that global industrial production as a control variable helps considerably to explain fluctuations in industrial production. The adj. R ² and the Akaike Information Criterion (AIC) favor our baseline specification with one lag of the differences in low water days (column 3) over the specification that considers this variable only contemporaneously. Columns 4 to 6 show the results for the corresponding specifications with levels of low water days for J = 1 and J = 2. In line with our baseline specification, the results indicate a strong contemporaneous negative effect of low water levels on industrial production that is statistically significant at the 1 percent level. It turns out that the estimated coefficients θ 1 and θ 2 for the lags of low water levels are positive, suggesting that industrial production recovers in the next months. However, these parameter estimates are not statistically significant at conventional levels.

When we compare the dynamic impact of low water levels on industrial production for the specification in differences (column 3) and in levels (column 6) of low water days – in a hypothetical scenario of two consecutive months with 30 days of low water levels in each month – in turns out that both specifications lead to very similar results for the growth rate of industrial production (Figure A2). However, while the specification based on the change in days implies that the level of industrial production turns back to its old level, the specification based on the number of days in levels implies that the level of industrial production is permanently reduced, albeit to a minor extent.

Figure A2:

Low water levels and industrial production (differences vs. levels).

The figure illustrates the predicted growth in industrial production (left-hand side panel) as well as the resulting level of industrial production (right-hand side panel) for a hypothetical scenario with two consecutive months with 30 days of low water for two alternative regression specifications. The differences specification corresponds to column (3) of Table A2 and the levels specification corresponds to column (6) of Table A2. The x-axis shows the number of months since the first month with low water.

The information criteria favor our baseline specification in differences of low water days over the specification in levels. In addition, we test the hypothesis ∑ j = 0 J θ j = 0 , which is implicit in our assumption of low water levels having only a temporary impact on the level of industrial production. We find that standard Wald tests cannot reject this hypothesis with a p-value of 0.13 for the specification with J = 1 (column 5) and a p-value of 0.54 for the specification with two lags (column 6). In general, it seems reasonable to assume that the relatively short-lived periods of low water levels do not have a permanent effect on the level of industrial production.

Appendix A3: Transportation Disruptions and Industrial Production

The empirical analyses in this paper show that low water levels lead to severe disruptions in inland water transportation and cause a significant and economically meaningful decrease of economic activity. In this section, we go one step further and examine the impact of shipping volumes on industrial production by using low water as an instrument. In particular, we first regress the changes in the transportation volume on the low water variable, along the lines of equation (3). We then estimate a regression of industrial production growth on changes in the transportation volume, replacing the latter by the predicted values from the first stage. Under the assumption that low water affects industrial production only through impaired inland water transportation, this extension allows for an application of the results to other transportation disruptions not stemming from low-water, such as an obstruction due to an accident as occurred in the Suez Canal in 2021.

Table A3:

Inland water transportation and industrial production.

	Growth in industrial production
	(1)	(2)	(3)	(4)	(5)
	2SLS	2SLS	2SLS	2SLS	LS
Growth in inland water	0.036^a	0.040^a	0.038^b	0.043^a	0.051^a
transport volume	(0.014)	(0.014)	(0.017)	(0.016)	(0.013)
Growth in inland water	0.027^a	0.030^a	0.028^a	0.031^a	0.032^a
transport volume, t − 1	(0.010)	(0.009)	(0.011)	(0.010)	(0.010)
Growth in industrial	−0.184	−0.361^a	−0.175	−0.351^a	−0.182
production, t − 1	(0.113)	(0.041)	(0.111)	(0.041)	(0.111)
Additional explanatory variables (I)	No	Yes	No	Yes	No
Additional explanatory variables (II)	No	No	Yes	Yes	No
Obs.	337	337	337	337	337
R ²	0.094	0.379	0.117	0.387	0.091
F-statistic (first stage)	44.12	47.33	39.53	41.77

Additional explanatory variables (I): contemporaneous and lagged global industrial production. Additional explanatory variables (II): contemporaneous and lagged anomalies of water temperature, air temperature, and rainfall. A constant is included in all regressions. All variables included in the second stage are also included in the first stage. Robust standard errors are given in parentheses. ^a/^b/^c indicates statistical significance at the 1%/5%/10% level.

The 2SLS estimates are presented in Table A3. The F-statistic of the first-stage regression indicates that low water levels are a strong instrument. Our results show that a decline in inland water transportation by 1 percent leads to a decline in industrial production growth by 0.036 percentage points in the same month (column 1). In the following month, industrial production growth is still dampened by a roughly similar magnitude. Again, the results are robust when we add contemporaneous and lagged growth in global industrial production (column 2), the set of weather-related variables (column 3), or both (column 4) as additional explanatory variables. Adding the full set of explanatory variables even results in a stronger effect of transportation disruptions on industrial production. For comparison, we additionally show the corresponding least squares estimates (column 5). They are somewhat larger in magnitude compared to the estimates in column 1, which is reasonable to the extent that there is a positive, simultaneous relationship between the transportation volume and industrial production. The 2SLS regression controls for the endogeneity of the transportation volume and therefore leads to smaller coefficient estimates.

In sum, this analysis documents that the shipping volume on the Rhine has a noticeable impact on economic activity in Germany.

References

Acemoglu, D., U. Akcigit, and W. Kerr. 2016. “Networks and the Macroeconomy: An Empirical Exploration.” In National Bureau of Economic Research Macroeconomics Annual, 30, edited by M. Eichenbaum, and J. Parker, 276–335. Chicago: University of Chicago Press.10.1086/685961Search in Google Scholar

Acemoglu, D., V. Carvalho, A. Ozdaglar, and A. Tahbaz-Salehi. 2012. “The Network Origins of Aggregate Fluctuations.” Econometrica 80: 1977–2016.10.3982/ECTA9623Search in Google Scholar

Barrot, J. N., and J. Sauvagnat. 2016. “Input Specificity and the Propagation of Idiosyncratic Shocks in Production Networks.” The Quarterly Journal of Economics 113 (3): 1543–92. https://doi.org/10.1093/qje/qjw018.Search in Google Scholar

BDB – Bundesverband der Deutschen Binnenschifffahrt e.V. 2019. Daten und Fakten 2018/2019. https://www.binnenschiff.de/service/daten-fakten (accessed January 07, 2020).Search in Google Scholar

Carvalho, V., M. Nirei, Y. Saito, and A. Tahbaz-Salehi. 2021. “Supply Chain Disruptions: Evidence from the Great East Japan Earthquake.” The Quarterly Journal of Economics 136 (2): 1255–321, https://doi.org/10.1093/qje/qjaa044.Search in Google Scholar

Cashin, P., K. Mohaddes, and M. Raissi. 2017. “Fair Weather or Foul? the Macroeconomic Effects of El Niño.” Journal of International Economics 106: 37–54. https://doi.org/10.1016/j.jinteco.2017.01.010.Search in Google Scholar

CCNR – Central Commission for the Navigation of the Rhine. 2019. Annual Report 2019: Inland Navigation in Europe – Market Observation. https://www.ccr-zkr.org/files/documents/om/om19_II_en.pdf#search=%22low%20water%20discharge%22 (accessed January 07, 2020).Search in Google Scholar

Contargo. 2017. Low Water. https://www.contargo.net/assets/pdf/Kleinwasser_Info-2017-EN.pdf (accessed January 07, 2020).Search in Google Scholar

Damania, R., S. Desbureaux, and E. Zaveri. 2020. “Does Rainfall Matter for Economic Growth? Evidence from Global Sub-national Data (1990-2014).” Journal of Environmental Economics and Management 102: 102335. https://doi.org/10.1016/j.jeem.2020.102335.Search in Google Scholar

Dell, M., B. Jones, and B. Olken. 2012. “Temperature Shocks and Economic Growth: Evidence from the Last Half Century.” American Economic Journal: Macroeconomics 4 (3): 66–95. https://doi.org/10.1257/mac.4.3.66.Search in Google Scholar

Dell, M., B. Jones, and B. Olken. 2014. “What Do We Learn from the Weather? the New Climate-Economy Literature.” Journal of Economic Literature 52 (3): 740–98. https://doi.org/10.1257/jel.52.3.740.Search in Google Scholar

ECB (European Central Bank). 2021. “Climate Change and Monetary Policy in the Euro Area.” In Occasional Paper Series No. 271. ECB Strategy Review. September.Search in Google Scholar

Felbermayr, G., J. Gröschl, and B. Heid. 2020. “Quantifying the Supply and Demand Effects of Natural Disasters Using Monthly Trade Data.” In Kiel Working Paper No. 2172. Kiel: Kiel Institute for the World Economy.10.2139/ssrn.3754689Search in Google Scholar

Felbermayr, G., J. Gröschl, M. Sanders, V. Schippers, and T. Steinwachs. 2022. “The Economic Impact of Weather Anomalies.” World Development 151: 105745. https://doi.org/10.1016/j.worlddev.2021.105745.Search in Google Scholar

Felbermayr, G., and J. Gröschl. 2014. “Naturally Negative: The Growth Effects of Natural Disasters.” Journal of Development Economics 111: 92–106. https://doi.org/10.1016/j.jdeveco.2014.07.004.Search in Google Scholar

Friedt, F. 2021. “Natural Disasters, Aggregate Trade Resilience, and Local Disruptions: Evidence from Hurricane Katrina.” Review of International Economics 29 (5): 1081–120. https://doi.org/10.1111/roie.12537.Search in Google Scholar

Heinen, A., H. Khadan, and E. Strobl. 2018. “The Price Impact of Extreme Weather in Developing Countries.” The Economic Journal 129 (619): 1327–42. https://doi.org/10.1111/ecoj.12581.Search in Google Scholar

ICPR – International Commission for the Protection of the Rhine. 2018. Inventory of the Low Water Conditions on the Rhine. https://www.iksr.org/fileadmin/user_upload/DKDM/Dokumente/Fachberichte/EN/rp_En_0248.pdf (accessed January 07, 2020).Search in Google Scholar

IPCC – Intergovernmental Panel on Climate Change. 2018. Global Warming of 1.5°C. https://www.ipcc.ch/site/assets/uploads/sites/2/2019/06/SR15_Full_Report_High_Res.pdf (accessed January 07, 2020).Search in Google Scholar

Islam, A., and M. Hyland. 2019. “The Drivers and Impacts of Water Infrastructure Reliability – a Global Analysis of Manufacturing Firms.” Ecological Economics 163: 143–57. https://doi.org/10.1016/j.ecolecon.2019.04.024.Search in Google Scholar

Jones, B., and B. Olken. 2010. “Climate Shocks and Exports.” The American Economic Review: Papers and Proceedings 100: 454–9. https://doi.org/10.1257/aer.100.2.454.Search in Google Scholar

Kim, H. S., C. Matthes, and T. Phan. 2021. “Extreme Weather and the Macroeconomy.” In Working Paper No. 21–24. Federal Reserve Bank of Richmond. August.10.21144/wp21-14Search in Google Scholar

Kurth, M., W. Kozlowski, A. Ganin, A. Mersky, B. Leung, J. Dykes, M. Kitsak, and I. Linkov. 2020. “Lack of Resilience in Transportation Networks: Economic Implications.” Transportation Research Part D: Transport and Environment 86: 102419. https://doi.org/10.1016/j.trd.2020.102419.Search in Google Scholar

Naguib, C., M. Pelli, D. Poirier, and J. Tschopp. 2022. “The Impact of Cyclones on Local Economic Growth: Evidence from Local Projections.” Economics Letters 220: 110871. https://doi.org/10.1016/j.econlet.2022.110871.Search in Google Scholar

Regmi, M., and S. Hanaoka. 2011. “A Survey on Impacts of Climate Change on Road Transport Infrastructure and Adaptation Strategies in Asia.” Environmental Economics and Policy Studies 13: 21–41. https://doi.org/10.1007/s10018-010-0002-y.Search in Google Scholar

Russ, J. 2020. “Water Runoff and Economic Activity: The Impact of Water Supply Shocks on Growth.” Journal of Environmental Economics and Management 101: 102322. https://doi.org/10.1016/j.jeem.2020.102322.Search in Google Scholar

Strobl, E. 2011. “The Economic Growth Impact of Hurricanes: Evidence from U.S. Coastal Counties.” The Review of Economics and Statistics 93 (2): 575–89. https://doi.org/10.1162/rest_a_00082.Search in Google Scholar

Todo, Y., K. Nakajima, and P. Matous. 2015. “How Do Supply Chain Networks Affect the Resilience of Firms to Natural Disasters? Evidence from the Great East Japan Earthquake.” Journal of Regional Science 55 (2): 209–29. https://doi.org/10.1111/jors.12119.Search in Google Scholar

Verschuur, J., E. E. Koks, and J. W. Hall. 2020. “Port Disruptions Due to Natural Disasters: Insights into Port and Logistics Resilience.” Transportation Research Part D: Transport and Environment 85: 102393. https://doi.org/10.1016/j.trd.2020.102393.Search in Google Scholar

Zink, M., L. Samaniego, R. Kumar, S. Thober, J. Mai, D. Schäfer, and A. Marx. 2016. “The German Drought Monitor.” Environmental Research Letters 11 (7): 074002. https://doi.org/10.1088/1748-9326/11/7/074002.Search in Google Scholar

Received: 2022-07-22

Accepted: 2023-02-14

Published Online: 2023-03-14

Published in Print: 2023-05-25

You are currently not able to access this content.

Articles in the same Issue

https://doi.org/10.1515/ger-2022-0077

Keywords for this article

climate; extreme weather events; supply shocks; business cycle effects; transportation disruption