Economic Consequences of Climate Change Impacts on Transportation: The Case of Atlantic Canada Yuri V. Yevdokimov University of New Brunswick/Department of Civil Engineering, Fredericton, Canada Email:
[email protected]
Mikhail Zhukov and Oleg Zaytsev University of New Brunswick/Department of Economics, Fredericton, Canada Email:
[email protected],
[email protected]@unb.ca
Abstract — climate change impacts such as an increase in mean temperature, change in precipitation patterns and sea level rise are affecting regional road transportation network in Atlantic Canada. These impacts cause direct and indirect economic consequences for the network and regional economy. Currently a dynamic general equilibrium model (GEM) is being constructed. In this paper, basic principles of the GEM are discussed and the model’s architecture is presented. Climate change impacts are regarded as productivity shocks, and their dynamics is analyzed on the basis of advanced time series techniques. Further these impacts will be imposed on a dynamic regional economic system expressed via our Dynamic GEM to trace their economic consequences. Index Terms — climate change impacts, regional transportation network, economic consequences, dynamic general equilibrium model, statistical analysis, climate variables
I. INTRODUCTION According to the latest 5th Intergovernmental Panel on Climate Change (IPCC) Assessment Report, the main climatic drivers in North America are temperature warming and drying trends, extreme temperature events, extreme precipitation events, damaging cyclones and rising sea level [1]. In this regard, Atlantic Canada as a part of North America represents a high risk area with respect to the climate change. Atlantic Canada is comprised of four provinces - New Brunswick, Nova Scotia, Prince Edward Island, and Newfoundland and Labrador. Over 2.3 million people live in urban areas and small communities along the coast with a span of more than 40,000 kilometers of predominantly rocky coastline [2]. Climate change impacts here include rising sea levels and temperatures, change in precipitation patterns and more frequent extreme weather events such as thunderstorms, snowfalls, hurricanes, tornados and others. There are over 1.3 million registered passenger vehicles and over 40,000 registered trucks in Atlantic
Canada. They operate on 55,000 km of provincially maintained roads and highways, including nearly 8,300 bridges. This also includes 4,700 km of National Highway System highways and intermodal linkages. These highways, connected by key ferry services, border crossings, and the Confederation Bridge, connect all communities within each province, and via the National Highway System, to points throughout the region, Canada, and the United States. Urban transit also plays an important role for travelers in municipalities in Atlantic Canada, moving over 24 million passengers a year [3]. As a matter of fact, regional transportation network (RTN) is a part of the so-called Atlantic Gateway and Trade Corridor, which is made up of Ref.[4]“a system of major ports, international airports, key border crossings, and road and rail connections between Atlantic Canada and North America’s major markets”. It is a Ref.[4]“strategic, integrated, and globally competitive transportation system for international commerce to and from North America”. More than $45 billion worth of goods involved in international trade moves through the Atlantic Canada Gateway and Trade Corridor annually. Based on the above facts, it is obvious that climate change impacts imposed on the RTN can bring significant economic damage if this problem is not addressed in a timely manner. This paper presents our approach to the evaluation of economic consequences of the climate change impacts on the road RTN in Atlantic Canada. II. CLIMATE CHANGE IMPACTS VERSUS ECONOMIC CONSEQUENCES OF THESE IMPACTS
It is necessary to distinguish between climate change impacts and their economic consequences. According to the literature review, there are four major physical or biological climate change impacts relevant to the RTN under study: 1. 2.
Sea level rise. Change in precipitation patterns.
3. 4.
Increase in temperature. Increase in frequency of extreme weather events.
Sea level rise can lead to flooding of transportation infrastructure and eventually to soil erosion in the area. Change in precipitation patterns can affect infrastructure and vehicles as well as their maintenance. It can also change driving conditions in the area. Increase in temperature can affect transportation infrastructure and vehicles operation. Increase in frequency of extreme weather events can affect transportation infrastructure, its maintenance as well as driving conditions. As a matter of fact, all these impacts combined affect RTN directly in terms of transportation infrastructure, associated services and operations and indirectly through the associated with RTN economic activities. Accordingly economic consequence or what is further called economic effects of these climate change impacts on the RTN can be divided into direct and indirect. Direct economic effects are the ones associated with production of transportation services. In an economic sense, those are impacts on inputs of production and services produced by these inputs of production. In this regard, by inputs of production first of all we mean capital assets and labor. Climate change impacts directly affect transportation infrastructure and vehicles (capital assets) and labor involved in production of transportation services. As well climate change impacts affect transportation services produced by transportation capital (infrastructure and vehicles) and transportation labor. From an economic standpoint direct effects are associated with supply side of transportation. There are two types of direct economic effects on transportation infrastructure and vehicles: (i) direct damage, and (ii) increased maintenance costs. In turn, transportation labor is affected via negative health effects. It is also possible to identify two types of direct effects on transportation services: (i) temporary or permanent disruption of transportation services, and (ii) deterioration of the quality of transportation services. Indirect economic effects of climate change impacts are associated with consumers of transportation services such as households and economic sectors other than transportation. Households mostly consume passenger transportation services while economic sectors mostly consume freight transportation. Decrease in production of transportation services due to climate change impacts may lead to a loss of labor income and profits in sectors other than transportation. In addition, interruption of transportation services due to climate change impacts can lead to increased travel time of all users of a transportation network. Finally, since these consumers of transportation are affected by climate change impacts directly as well, they may change their consumption and production patterns which eventually may lead to a change in their demand for transportation services. From an economic standpoint, indirect effects are associated with demand side of transportation. Special consideration should be given to the link between climate change impacts and transport accidents.
Literature on transport accidents suggests that an increase in precipitations, temperature and frequency of extreme weather events lead to an increase in the number of transport accidents. These accidents are associated with all users of a transportation network – producers of transportation and consumers of transportation – and as such should be also incorporated into our framework. III. MODELING CLIMATE CHANGE IMPACTS ON TRANSPORTATION Our analysis of the existing literature on economic evaluation of climate change impacts in various segments of an economic system, led us to the conclusion that dynamic general equilibrium model (GEM) is the best framework for our analysis. Ref. [5] is a seminal work in this regard followed by more recent studies [6], [7] and [8]. These studies helped us formulate basic principles for our model as follows: 1. Both micro- and macroeconomic aspects of the RTN should be incorporated in our model 2. Modified Computable General Equilibrium Model (CGEM) is our fundamental tool since it allows us to incorporate micro- and macroeconomic dynamics as well as dynamics of climate change impacts 3. Modeling has to address the following two goals: (i) design of the appropriate architecture to include microeconomic and macroeconomic dynamics affecting our RTN; (ii) modeling of the dynamics of climate change variables. Once both goals presented above are achieved, dynamics of the climate change variables will be imposed on the dynamics of the basic CGEM to trace economic consequences of the climate change impacts. In this regard, the following architecture of our CGEM was constructed: Transportation module is the centerpiece of the CGEM. It consists of a system of dynamic equations for price and volume of regional transportation. These equations were obtained on the basis of Vector Autoregression estimation in statistical software package STATA using historical data. Within the module, this system produces equilibrium price of regional transportation for the RTN as a whole based on major economic determinants such as regional traffic, value added at regional hubs, overall price level, regional GDP, oil price, population and some other. At each hub, the above price of transportation is taken as an input to define the value added produced by the hub based on this price as well as geographical and industrial characteristics. Each hub is further disaggregated with respect to major industrial consumers of transportation in that sub-region. It means that initially the value added and traffic at each hub were estimated via panel Vector Autoregression based on historical data using major economic determinants and major industrial consumers of transportation. For example, the largest consumers of transportation at Fredericton hub were identified as follows:
- Manufacturing - Forestry and logging - Retail trade - Wholesale trade Further, all sectors’ values added are then summed up at each hub and sent back to the transportation module along with the hub’s traffic volume to determine next period equilibrium price of transportation for the RTN. This process repeats itself to produce time paths of our major economic and transportation variables associated with the RTN over time given current and expected micro- and macroeconomic conditions. In line with the CGEM framework, eventually we receive a time path which is a set of short-run equilibria represented by the transportation module and five regional hubs. Microeconomic forces are internal in this model since they are based on the supply/demand dynamics of the regional industries/sectors while macroeconomic forces are based on external dynamics given exogenously. In other words, given initial conditions, our model replicates evolution of the RTN, driven by internal microeconomic and external macroeconomic forces, with all elements estimated on the basis of statistical analysis and historical data outside the model. Statistical analysis of the model’s elements was done via comprehensive time series analysis on the basis of Vector Autoregression (VAR), panel VAR and cointegration using methodology described in detail in [9]. The designed model was tested with the help of exogenously imposed shocks. For example, our previous analysis of climate change impacts on the RTN showed that approximately 1.5% of the traffic is lost due to these impacts. Therefore, a 1.5% negative quantity shock was imposed on the Halifax hub – one of the five regional hubs in our RTN. As a result, this negative shock led to the loss of $503.1 million in value added over first 15 years at this hub with spillover effects to the entire RTN. For example, this spillover impact on Fredericton hub was estimated as $22.29 million over first 15 years. These values give us some benchmarks for the investments into climate change mitigation measures. In parallel, dynamics of climate variables in the region has been analyzed. Fundamental climate variables in our study are: temperature, precipitation and sea level. Some statistical work on dynamics of these variables in Canada is presented in [10] and [11]. Our analysis of climate change dynamics is based on stationary linear models widely used in time series econometrics: autoregressive moving-average (ARMA) and generalized autoregressive conditional heteroskedastic (GARCH) models with ARMA errors. By the time of this paper, we have completed time series analysis of the regional temperature using monthly time series data for the 1871-2012 period. According to our analysis, obtained linear trend is upward sloping, which suggests that the mean monthly temperature in Atlantic Canada will be increasing during 1872-2101 period. For example, we predict that cumulative increase in mean monthly temperature at Saint John hub will
amount to +3.25°C by January 2101 compared to the initial temperature level observed in January 1872, or to +1.43°C compared to the initial level of January 2000. We also found statistical evidence that the underlying data generating process of mean monthly temperatures had started to change in the mid-1950s. Eventually, we found the best-fit models to describe dynamics of temperature at various regional hubs of our RTN such as Northern New Brunswick or Edmundson-Bathurst area, Moncton, Saint John, Fredericton and Halifax. Currently we are analyzing dynamics of precipitation and sea level rise in Atlantic Canada using the same methodology. With respect to precipitation, common knowledge is that climate in Atlantic Canada is getting wetter. However, there is a substantial variety in the tendencies over the areas. We use the following time series: rainfall, snowfall, and total precipitation, which is summation of the two. The data is taken from the second generation Adjusted Precipitation for Canada (APC2) dataset. This is the data extracted from National Climate Data Archive, and corrected to take into account equipment measurement errors, and aggregated over closely-related stations to receive longer time series. Data is monthly, constructed from daily values of rainfall, snowfall, and total precipitation in millimeters. The data was taken from meteorological stations close to five hubs of our RTN. In this regard, we have individual station data for Fredericton, Saint John, Halifax, and Moncton. The data for the Northern New Brunswick and Miramichi is given by three stations located in Edmundson, Bathurst, and Miramichi. Number of observations varies from 1,080 for Edmundston to 1,692 for Halifax. Since there are some missing observations, time series approximation procedure was applied to fill the gaps. The following table summarizes our data: TABLE I. METEOROLOGICAL STATIONS
Station ID
Provi nce
8101500
NB
8104900
NB
8202250
NS
8103200
NB
810AL00
NB
8100503
NB
8101000
NB
From / To 1874 / 2009 1871 / 2011 1872 / 2011 1898 / 2011 1916 / 2005 1884 / 2012 1873 / 2004
Number of observations
missing,
missing,
snow
rain
1632
13
16
1692
1
1
1680
27
22
1368
16
16
1080
29
30
1548
252
251
1584
2
2
Since global sea level rise is projected to accelerate over the next century because of continued thermal expansion and melting of land ice, Atlantic Canada will be also a subject to the rising sea level. In this regard, several sources of the sea level data were analyzed. The data set provided by Canadian Hydrographic Service was chosen as the main source since it provided the most consistent data set. The monthly mean sea level data for
three tide-gauge stations were obtained from the Atlantic Zone Monitoring Program (AZMP) websites. The stations were chosen in accordance with transportation hubs of our RTN. Sea level is measured by tide-gauges relatively to land. Stations at Halifax and Saint John hubs contain a long data set with 1,128 and 962 observations respectfully. Shediac station at Moncton hub is characterized by a lot of missing observations. Rimouski station was implemented only in 1984; it contains 349 observations. Since the missing observations are present in the data, some time series procedures to fill the gaps are required. The information about stations is summarized in the following table: TABLE II. SEA LEVEL STATIONS
Station ID
Station name
Province
From/to
Obs
Missed, obs
00490
HALIFAX
NB
1920/2014
1128
26
00065
SAINT JOHN
NB
1920/2014
962
79
02985
RIMOUSKI
NB
1984/2014
349
16
1326
SHEDIAC BAY
NB
1972/2011
480
198
Once consistent datasets are constructed, comprehensive time series analysis will be applied to capture dynamics of these climate variables similar to the temperature to impose that dynamics on previously developed CGEM. IV. CONCLUSION It is an established fact that regional transportation networks all over the world are vulnerable to the climate change impacts such as increasing temperatures, changing precipitation patterns and rising sea level. However, according to our literature review, there is no comprehensive model to trace economic consequences of these climate change impacts over time. This study is the first attempt to produce such a model. The model is based on dynamic general equilibrium framework associated with Atlantic Canada regional road transportation network that includes microeconomic and macroeconomic aspects. Climate change impacts in this framework are regarded as productivity shocks imposed on the regional transportation network. This approach allows us to trace economic consequences of the climate change impacts on regional transportation network as consequences of the productivity shocks imposed on a dynamic economic system. In this regard, basic principles of the model’s architecture are presented; first version of the model is designed and tested. In parallel, dynamics of climate variables such as temperature, precipitations and sea level is being analyzed and some first results of this analysis are presented. In the future, dynamics of the climate variables will be imposed on the designed dynamic model of the regional transportation network.
REFERENCES [1] Intergovernmental Panel on Climate Change (2013), 5th IPCC. Available: Assessment Report (AR5), http://www.ipcc.ch/report/ar5/wg1/ [2] Environment Canada (2012), Celebrating our coastlines from rivers to oceans. Available: http://www.ec.gc.ca/envirozine/default.asp?lang=en&n=82B538B3-1 [3] New Brunswick Department of Transportation (2008), The Atlantic Canada Transportation Strategy 2008-2018, Available: http://www2.gnb.ca/content/dam/gnb/Departments/trans/pdf/en/Publicat ions/2008-2018AtlanticCanadaTransportationStrategy.pdf [4] Canada’s Atlantic Gateway (2010), Atlantic Gateway and Trade Corridor Strategy: Connecting Canada with the World, Available: http://www.atlanticgateway.gc.ca/media/documents/en/brochure.pdf [5] W.D. Nordhaus, and Z. Yang, “A regional dynamic generalequilibrium model of alternative climate-change strategies”, The American Economic Review 4(86), pp. 741-765, 1996 [6] A. Rezai, L. Taylor, and R. Mechler, ”Ecological macroeconomics: An application to climate change”, Ecological Economics 85, pp. 69–76 2013 [7] W.S. Jaglom, “Assessment of projected temperature impacts from climate change on the U.S. electric power sector using the Integrated Planning Models”, Energy Policy 73, pp. 524-539, 2014 [8] V.J. Nannen, van den Bergh, and A.E. Eiben, “Impact of environmental dynamics on economic evolution: A stylized agent-based policy analysis”, Technological Forecasting and Social Change 80, pp. 329-350, 2013 [9] W. Enders, Applied econometric time series, 3rd edition, John Wiley & Sons, Inc., 2010 [10] E. Mekis, and L.A., Vincent, “An overview of the second generation adjusted daily precipitation dataset for trend analysis in Canada”, Atmosphere-Ocean 49(2), pp. 163-177, 2011 [11] L.A. Vincent, X.L.Wang, E.J.Milewska, H.Wan, F.Yang, and V. Swail, “A second generation of homogenized Canadian monthly surface air temperature for climate trend analysis. Journal of Geophysical Research 117, pp.1804-1807, 2012
Yuri Yevdokimov Yuri is a Professor at the University of New Brunswick (Fredericton, Canada). Due to his degrees in economics and engineering, he holds joint appointment with two Departments – Economics and Civil Engineering. Dr. Yevdokimov teaches courses in undergraduate and graduate programs in both Departments, and transportation economics is among them. His primary research interest is in economic evaluation of climate change impacts on regional transportation. Dr. Yevdokimov has participated in three projects dedicated to this topic over last four years and, as a result, published several articles, papers and chapters in monographs both nationally and internationally. In general, Dr. Yevdokimov has more than 20 publications. One monograph, two textbooks, twelve refereed journal articles, and nine chapters in books are among these publications. Oleg Zaytsev Oleg is a Ukrainian born economist who in 2010 received a joint MA degree in economics with minor in financial economics from Kyiv School of Economics (Ukraine) and University of Houston (USA). In 2014, Oleg received another MA degree in economics from the University of New Brunswick (Fredericton, Canada). During his studies at that University, as a research assistant he participated in the Project “Modeling economic consequences of the climate change impacts on ground transportation in Atlantic Canada” financed by provincial government. Specifically, Oleg was estimating evolutionary dynamics of monthly temperature in Atlantic Canada on the basis of comprehensive time series analysis. The main focus of his research was on ARMA and (G)ARCH class of models. In general, Oleg’s professional interests lie in macroeconomics, economics of climate change, time series econometrics and statistics. Mikhail Zhukov Mikhail’s first Master’s degree was in applied mathematics received from Moscow Institute of Physics and Technology (Russia) in 2009. Later in 2014, he graduated with a Master’s degree in economics from the University of New Brunswick (Fredericton, Canada). During his studies at that University, Mikhail has participated in the Project “Modeling economic consequences of the climate change impacts on ground transportation in Atlantic Canada” financed by provincial government. Specifically, he was designing a dynamic general equilibrium model of the regional economy in Atlantic Canada. In 2013, he presented his research in this area at the 5th Annual Conference of Acadia Economic Society in Atlantic Canada. Currently Mikhail is a Ph.D. student at McGill University (Montreal, Canada).
