November 18, 2015
Talking Freight: International Urban Freight Conference Presentations - Warehouse Locations and Community Impacts
Audio files require the Windows Media Player.
Nicole Coene
Good afternoon or good morning to those of you in the West. Welcome to the Talking Freight Seminar Series. My name is Nicole Coene and I will moderate today's seminar. Today's topic is: International Urban Freight Conference Presentations - Warehouse Locations and Community Impacts.
Before I go any further, I do want to let those of you who are calling into the teleconference for the audio know that you need to mute your computer speakers or else you will be hearing your audio over the computer as well.
Today we'll have three presentations, given by:
Sönke Behrends is an assistant professor in urban logistics at Chalmers University of Technology, and the manager of the Urban Freight Platform (UFP) based in Gothenburg, Sweden. His primary research interest is to advance the understanding of the interaction between logistics, transport and land-use and to develop solutions to improve the sustainability of urban logistics operations. His expertise is in impact assessment of logistics measures, urban freight planning and intermodal transport.
Jean-Paul Rodrigue received a Ph.D. in Transport Geography from the Université de Montréal (1994) and has been a professor at Hofstra University since 1999. Dr. Rodrigue's research interests mainly cover the fields of transportation and economics as they relate to logistics and global freight distribution. Specific topics over which he has published extensively cover maritime transport systems and logistics, global supply chains, gateways and transport corridors.
Sanggyun Kang is an advanced Ph.D. student in urban planning at the University of Southern California. His field specialization is urban freight systems. He is interested in documenting sub-metropolitan spatial characteristics of warehousing and distribution centers and testing how changing spatial patterns may affect urban freight activities over time.
Quan "Jack" Yuan (Yawn) is a PhD candidate in Urban Planning and Development at Price School of Public Policy at University of Southern California. His research interests include urban transportation and land use, urban freight planning, and parking policy. He got his Bachelor's degree in Urban Planning from Peking University in 2011 and Master's degree in Urban and Regional Planning from UCLA in 2013. He is now working with Professor Genevieve Giuliano on the MetroFreight Project in USC METRANS Center.
Today's seminar will last 90 minutes, with 60 minutes allocated for the speakers, and the final 30 minutes for audience Question and Answer. If during the presentations you think of a question, you can type it into the chat area. Please make sure you send your question to "Everyone" and indicate which presenter your question is for. Presenters will be unable to answer your questions during their presentations, but I will start off the question and answer session with the questions typed into the chat box. If time allows, we will open up the phone lines for questions as well. If we run out of time and are unable to address all questions we will attempt to get written responses from the presenters to the unanswered questions.
The PowerPoint presentations used during the seminar are available for download from the file download box in the lower right corner of your screen. The presentations will also be available online within the next few weeks, along with a recording and a transcript. I will notify all attendees once these materials are posted online.
Talking Freight seminars are eligible for 1.5 certification maintenance credits for AICP members. In order to obtain credit for today's seminar, you must have logged in with your first and last name or if you are attending with a group of people you must type your first and last name into the chat box. I have included more detailed instructions in the file share box on how to obtain your credits after the seminar.
For those of you, who are not AICP members but would like to receive PDH credits for this webinar, please note that FHWA does not formally offer PDHs, however, it may be possible to receive PDHs for your participation in Talking Freight if you are able to self-certify. To possibly receive PDHs, please download the agenda from the file download box and submit this agenda to your respective licensing agency.
Finally, I encourage everyone to please also download the evaluation form from the file share box and submit this form to me after you have filled it out.
I'm now going to turn it over to Tom O'Brien from METRANS for opening remarks.
Tom O'Brien
Thank you. My name is Tom O'Brien and I am the Executive Director of the Center for International Trade and Transportation at California State University, Long Beach and the Associate Director of the METRANS Transportation Center. I wanted to take two minutes to provide some context for the presentations you are about to hear. These are presentations that were featured at METRANS' Six International Urban Freight Conference which was held last month in Long Beach, California. METRANS is the home to four research centers. The METRANS University Transportation Center, which is a US DOT UTT and we are part of the National Center for Sustainable Transportation housed at UC Davis. The focus of our work is on sustainable freight movement. Metro Freight which is the Center of Excellence in Urban Freight is funded by the Research Education Foundation and the FHWA Southwest Transportation Workforce Center.
The conference is the home to research for these research centers and for an international audience of researchers, practitioners, and public sector agency representatives to explore various aspects of urban freight. The presentation from the conference of which these three are included are available on the METRANS website are available at METRANS.org. All of the background on the conference can be found there. You will find presentations that relate to areas of interest including local pickup and delivery, trade routes and hubs, different freight modes, urban modeling and planning, new technology, changing consumption, production and spatial organization, and best practices. If there is any need for additional information you can contact us through that website. At this point I would like to turn it over to Dr. Sonke Behrends.
Sonke Behrends
Thanks Tom for that introduction. My name is Sonke Behrends. I am happy to be a part of this seminar. I will take a closer look on the differences between freight distribution in metropolitan areas as well as suburban areas with lower density. We will discuss if we are actually observing an emerging dualism in urban freight distribution between these areas. First during my presentation I will provide background to the research and then define the fundamental differences between city logistics and urban logistics. We will focus on suburban logistics and identifying some trends increasing the relevance of this type of logistic operations in urban areas. We will further discuss whether we see an emergent dualism between these two areas and finally we will conclude whether we see this emerging area of urban freight research.
To provide some background freight is very important to economic development of urban areas. On the other hand it is obvious that freight contributing to congestion and air pollution is seen as a barrier to the sustainable development of cities. What we have seen in recent years is that there has been growing city logistics research and polices in cities around the world. We also see that the focus when it comes to logistics research is mainly on last miles deliveries. That is the focus of research. We tend to focus on the last mile delivery from the distribution center into the center area. Freight operations in the suburban areas on the other hand are widely underrepresented. What we are trying to do with this presentation is to put logistics into the whole metropolitan area, so going beyond a focus of last-minute deliveries and including other urban operations in the areas for example, large terminal facilities such as ports, rail terminals or airports.
From this we have formed a number of questions. The main question is that: are there underlying or different forms of distributions in urban areas? If there is a dualism, what is the focus of policies on the area concerning logistics and will it have an impact? And finally from that, will carriers develop different distribution channels, which use different transport modes compared to the current approach?
Jean-Paul Rodrigue
Hi. This is Jean-Paul Rodrigue speaking from another time zone. Before moving on with this topic it might be a good idea to start with the perspective about what this city logistics and urban freight flows entails and that chart here resumes the situation. It is a variety of urban freight flows that are taking place. If you want more information I will provide at the bottom of the slide a hyperlink to further explanation. The point I would like to make is that more often city logistics per say tends to focus on what we call urban deliveries which are that little green line that we see over there which is mostly from distribution centers to final consumers and consumer related shopping trips. But of course we have noticed that it is self-evident there is an all array of product related flows which are linked with transfer terminals, manufacturing clusters of freight distribution clusters which are taking place in suburban areas and as this is taking place we tend to observe and emerging dichotomy between the different forms of city logistics. As a Sonke underlined before, we tend to focus on this last mile and sometimes the last feet of logistics while we tend to ignore some kind of the big picture taking place in suburban areas. That is what we want to focus on in this presentation.
To move on further with this rationale or this perspective about suburban logistics, this is from speculation, but there is evidence behind it but it needs to be further underlined. The point is that when we deal with city logistics we observe very interesting differences in the cost structure and challenges based upon the respective densities in which the urban freight distribution is taking place and paradoxically logistics or central area logistics are the most expensive and tend to be the most challenging. Because of all of those series of factors which are identified on the slide: higher density and you have limited space available in storage so more frequent deliveries. Parking difficulties: those are well-known. As opposed to this, the system of freight distribution which is taking place in suburban areas I would say is more effective because of the difference in densities which is offered in suburban areas. That is something which is worth underlining. The paradox behind it is the suburban freight system which is an environment that is suitable for a much more suitable for urban freight distribution than what takes place in central areas. That is what we are questioning.
Are we observing and emerging dualism? We observe different separate freight distributions one being called city logistics which seems to be the focus of the attention because this is where the low hanging fruit is. This is where benefits need to be realized in terms of improving the efficiency. However you have a whole emerging suburban logistics system where the growing population tends to be living in suburban areas which seem to be involving its own paradigm or operations and that seems to be a question we are asking ourselves will this create separate urban freight distribution systems because of that.
Sonke Behrends
We will continue with discussing current trends. Suburban areas have a great impact on logistics which is an ongoing consumption pattern. When we look at the perspectives on the structure we usually see the core center and suburban spaces as a dichotomy. We have the core center area and the consumption patterns that are simple, that the retail activities are in the central area and consumption takes place in the suburban residential areas. We have clear interaction between the center and its related suburbs. This dichotomy of city and suburbs has been replaced by a highly dispersed and differentiated polycentric region. These are no longer dominated by a core city but increasing the characterized by decentralization. As we have learned we see much more dispersed and differentiated interactions between the core city and its suburbs.
Furthermore, resulting from this we see suburban consumption patterns everywhere in North America and Western Europe so we have out of town shopping malls and large format stores which have emerged in suburban areas. Logistics was one of the driving forces because there were established at suburban areas which were cheap to buy and have a high accessibility.
Another feature of suburban consumption patterns is the increase of e-commerce. We see especially in areas where we have characteristics like higher average incomes and larger households and composition of the area. We see in these areas high levels of e-commerce. The point I want to make is the suburban consumption patterns may be different from the central area.
Another trend of logistics for all that is the suburbanization of logistics. Freight volumes have been growing and increasing consumption across the world affected which means more space and bigger places for handling the logistics at urban locations. There have been moved out of the central areas and located to more accessible places where land-use was cheap to buy and you had access the interregional transport network. We can observe two effects: The first is increased distribution distances leading to more congestion and emissions and more noise and so on. This also leads to higher amenities in urban areas. On the other hand, moving logistics facilities to suburban locations we had the reduction of traffic by large trucks serving the warehouses and facilities in central locations.
[ Indiscernible ] logistics could be more space in suburban areas and so on. We also see many new forms of interactions between suburban and city centers that instead of having this trend of increasing usage of small vans for urban distribution, logistics operators they made the switch again using larger vehicles for the transfer from suburban to central areas, but then reloading at urban distribution centers where it leads to more suitable vehicles for last mile delivery, for example cargo bikes or small electric vans. Suburbanization of logistics may in this way lead to new central distribution centers and new forms of distribution operations.
Jean-Paul Rodrigue
Another issue which is lingered in the suburban logistics paradigm is the fact with the out sourcing and offshoring of globalization. These things are well-known. You see across the world, especially large in urban areas such as Los Angeles and New York and Chicago, the emergence of a large gateway function which involves a complex interaction between intermodal terminals and clusters of freight distribution centers. This brings into the equation I would say an additional dimension to suburban logistics which is what we can call terminal-centric which starts out at a significant impact on the distribution aspects and congestion and is often a problem. Most terminals tend to be located in proximity to central areas such as New York and Los Angeles for instance. All of this creates additional traffic and complexities and conflicts and the potential for additional regulation. That could impact the dimension of suburban logistics. That needs to be kept in mind which brings us to the question, are we observing a convergence or divergence in city logistics? The point we are putting forward here is in our opinion for a large metropolitan area it seems to be a divergence taking place.
The question is to what extent what is divergence and what it will create, and it might create two separate urban freight distribution channels. One adapted to the central areas where you will have specific vehicles and smarter vehicles in the sustainability strategies with electric vehicles. Parking, let's say deliveries outside of regular hours. City logistics and strategies are well-known and the regulations that are linked to try to mitigate these impacts. But while this happens it forces urban freight distributors to implement a dual strategy. A Central City strategy and then I Swill keep in suburban areas my regular freight distribution strategy. That is in a sense but we are wondering. This remains to be further explored but that is something which I would say interesting. We observed two forms of city logistics. Another thing we are observing and it may be upsetting to some of you is the smart growth strategies and complete street strategies and they appear to be in contradiction or undermine city logistics strategies. In the sense that in suburban areas or in areas where the strategies are establishing smart growth strategies are going to undermine the efficiency of urban freight distribution because of the design and that freight appears not to be part of the design.
That is something that needs to be further discussed and addressed. This has been taking place in suburban areas. It may be a contradiction to a freight distribution strategy. Another one, last but not least is of course is e-commerce issue. This can be very problematic in the sense that it creates a benefit of consolidation in central areas because of the density factor so the consolidation factors are positive while the delivery factors are cheaper in suburban areas.
Sonke Behrends
What could we in the conclusion? So we thought about the question, should city logistics emerge a new field of freight policy and research? I think we can conclude, yes. We see this divergent distribution. Suburban logistics opens up a paradox in city logistics meaning that it is becoming more of the norm rather than the exception. On the other hand we see that suburban areas represent an ideal environment for freight flows mainly due to their lower densities and all of the related problems that we see such as higher densities for parking and congestion have less relevance here.
We see a developing dualism here and what are the possible implications? What we also see is compartmentalization of distribution in central and suburban areas. This brings about the question, will this divergence simply be functional or will it involve costs? Another question which we would look further into is what is the driver of the developing divergence? We found two factors which made the driver: one is the density and the other is regulatory. If we have high density we see more developing strategies. The question remains is it only relevant for the high density areas like Manhattan and other cities or do we see this also in smaller towns and cities. In terms of regulation we see as a consequence of the problems in urban logistics as cities want more advantaged solutions like access restrictions, regulating access, and banning them during different times of day and so on. This creates a burden for operators and carriers inside this industry. The question remains, what will lead to this divergence? These are questions we need to look further to into the future.
We are happy to answer questions and you can direct the questions to Jean-Paul or to me directly. Thank you.
Nicole Coene
Thank you, Gentlemen. We will now move on to Sanggyun Kang from the University of Southern California.
Sanggyun Kang
Thank you. My name is Sanggyun Kang, PhD student at the University of Southern California. This presentation is about changes in warehouse location and their impacts on metropolitan areas. There is growing concern that warehousing activities are moving to the edges of the metropolitan areas, further from their markets. These locations are generating more truck traffic and related impacts. However, the evidence on warehouse location changes is limited. In this presentation I will discuss the reason for decentralization and how we should measure location changes. I will present some preliminary results for four metropolitan areas in California.
This presentation is organized around three questions. First we need to consider the impacts of warehousing location changes. Also in order to determine where warehousing activities are decentralizing, we need consistent measures. Then, we need to apply these measures across different metro areas to see if the trends are consistent.
Question number one. There is anecdotal evidence that warehouses are moving to the urban periphery. It implies more truck activities. Again, more truck activities suggest more potential negative externalities on communities such as congestion, fuel consumption, air pollution, greenhouse gases, noise, and environmental justice issues. Then why would that move?
The supply chain is restructuring. The restructuring goals are to move things in large volume and high speed at the right time. They need large volume-operation because they have to fulfill national and global goods movement. It is required to respond immediately to diversify customer tastes; and the product shelf life is getting shorter and shorter; also just-in-time production requires a reliable delivery of resource and products. These factors require different types of warehouses: such that it is big and technology intensive. It became more about movement of goods rather than storing them. So the result is system-wide integration, but geographical separation. All this search for low cost began in the 1970's and this has been well understood.
What is not well-understood is metropolitan-level supply chain restructuring. The current understanding so far is that in order for the logistics industry to achieve these goals: it needs larger facilities, hence larger land parcels, and lower per unit land costs. They constantly seek lower overall costs, and maybe automation lowers the cost, but only with large volumes. As transport costs decline, distance to market becomes less valuable relative to other costs. Travel distance became relatively flexible thanks to information, communication and transportation advancement. Having access to local markets became less important than having access to the global supply chain. All of these location factor trade-offs suggest warehousing decentralization.
Let's think about how to measure this. What are the issues in measurement? Let's begin from a simple example. Let's look at a CBD within a city which is the major marketplace. We have a high density urban core and a lower density urban area. A simple system. If there were imports and exports, they came from some facility and were transported to the market to be sold over there. What's been produced in here typically went in the opposite direction. The world changes for all reasons. We may have a very different kind of system. The metro area has changed, and the supply chain has changed as well. The metro areas are far more dependent on import and export flows because metro areas don't produce much of what they consume these days. Or metro areas are often producers of products, for which there is an international market. We can imagine all kinds of flows now that are going in and out in addition to move around within them.
Now let's consider a new activity. New warehouses are built in different places where land is cheaper and more available, land use regulation is less stringent, or they have greater access to transportation infrastructure. If we consider their market as the CBD, their distance from the market has increased, which, in turn, implies more truck activity. However, we do not really know whether in fact distance from market has increased. Location changes could take many forms. The distance from the market might increase, but at the same time, due to the land availability, regulation and all other factors, warehouses end up locating in similar places. They ended up concentrating in a specific place. This suggests that location changes can take many patterns, and these patterns have implications for impacts.
Warehouses are not standalone facilities, but connect supply chains. It is located between the market and freight infrastructure. If population locations have changed, we should consider its location change. Even though it looks like there has been this dramatic change in warehousing location, it is possible that the population and employment distribution has changed so much, that warehouses are not further away from markets on average than this was previously. These are all the things we want to think about.
In summary, I generated these four measures as to how the location change might be measured. As in Measure 1, we can think about a measure of straight decentralization: which is relative to a specific location: such as from the CBD or from the geographical center of warehouses. We can think about relative decentralization; as in Measure 2. We can consider the distribution of employment and population. Similarly in Measure 3, we think about a straight measure of concentration of warehouses. I used the Gini coefficient. Or as in Measure four, I can consider the degree to which all employment is concentrated relative to the warehouses. There have been a few empirical studies. Their limitation is that they have used many different measures; but they have not consistently used one; they have not used multiple measures in one place.
I used the zip code business patterns from 2003 to 2013. It is based on ZIP codes, and it provides the number of warehouses by ZIP code. The definition is from NAICS code 493-Warehousing and storage. I used four metro areas in California as case study areas: which are Los Angeles, San Francisco, Sacramento, and San Diego. I will explain this in detail.
The four metropolitan areas are quite diverse, and thus provide good cases for testing spatial measures. Los Angeles is the second largest CSA in the United States. It has roughly more than twice the population and employment as in San Francisco. Sacramento and San Diego are much smaller in population and in employment. The metropolitan areas differ in other ways. Los Angeles is the nation's largest international trade node in the U.S. San Francisco is also a major international trade center, but in higher value goods. Sacramento is a trade node for the Central Valley. San Diego serves as a hub for cross-border trade and industry. These differences should lead to different development density patterns and freight flows.
The warehousing industry differs across metro areas. Los Angeles has the largest warehousing industry as it is the number one international trade node. San Francisco follows, however the size difference is greater than that of population. Sacramento is the third, followed by San Diego. All but San Diego show a significant industry expansion. The dramatic change in Sacramento is from Apple's new distribution center that I will explain with a map later.
This map shows the distribution of Los Angeles warehouses employment in 2003. The circles are the zip codes that have at least one warehouse, and its size corresponds to the number of warehousing employment. In 2013, we can see the apparent industry expansion in several specific places. This is before in 2003 and this is after in 2013. The expansion is concentrated in Inland Empire and also Santa Clarita. The addition of new warehouses and warehousing employment in these distance places should increase the average distance to the market, and thus it implies more truck activity. This map shows the Sacramento warehousing employment with the same scale and same size classification. It shows the situation in 2003. In 2013, we have captured a large increase in size Elk Grove. Apple constructed their new distribution center in 2012 and 2013, and it added a large number of warehousing jobs.
I showed very distinctive patterns of warehousing location change with the maps. Could the spatial measures capture this disparity? These are the results for Measure 1: Decentralization. It shows the changes between the two periods 2003 and 2013. I put it with highlights if the changes were statistically consistent. As you can see, Los Angeles is the only place where it shows a consistent pattern of decentralization from the CBD and from the geographical center of warehouses. However, if we considered the warehousing employment, all four metro areas show a similar change between the two periods.
Measure 2 quantifies the degree of decentralization considering distribution of employment and population. The result is similar to measure 1. Los Angeles is the only place with a consistent trend of relative decentralization. Warehouses in Los Angeles decentralized more than employment and population have done. Other areas show varying results.
Measure 3 quantifies the degree to which warehouses are concentrated. In Los Angeles and San Diego we see a consistent pattern of concentration. In Measure 4 I could not conduct a statistical test, but they all changed in the same direction, namely they all became more clustered than employment as a whole.
I have done case studies of four metropolitan areas. When I used these multiple measures, which is a comprehensive way of measuring different aspects of spatial organization, there is no consistency across four cities. Los Angeles is the only place with strong decentralization and concentration patterns. San Francisco shows some evidence of decentralization, but it is only true for the employment. Sacramento and San Diego show varying results. The measures definition can change your understanding of the nature. Accordingly, about it is important to use multiple measures. Warehousing employment allocation is much more flexible than the actual movement of warehouses. I surmise that this might be one of the supply chain operators' strategy of allocating labor intensive- or larger scale- warehousing operations in distant places.
Decentralization happened in some places and not in others. It's happened when I measured in one way, but not when I measured in another way. The suggestion is that different trends are happening in different places. Why? It is because these metro areas differ from each other. Their land rent is very different from others, not to mention land availability. As I explained earlier, their role in international trade differs widely. The sizes of local markets also differ. Their industry composition is different. The transportation environment is different. For example, people in Los Angeles suffer from congestion and delay. Of course land use regulation, tax policies, and other local labor pools are all different. These factors are worth further research. I look forward to more empirical evidence.
That is it for my presentation. If you have questions, I am happy to answer. Thank you.
Nicole Coene
Thank you, Sanggyun. We will now move on to our final presentation given by Quan Yuan from the University of Southern California.
Quan Yuan
Thank you. Hello everyone. My name is Quan Yuan. I am from the University of Southern California. The topic of my presentation is Location of Warehouses and Environmental Justice. I would like to present empirical evidence on spatial relationships between warehouse locations and the communities near the warehouses. Here is an overview of my presentation. The topic originates from the discussion on environmental justice and transportation activity. As we know, different types of transport facilities such as airports, rail yards, and high-volume freeways can bring about substantial environmental impacts and they are regarded as locally undesirable. We have seen evidence showing the co-location of these facilities and disadvantaged population. In the world of freight, warehouses are major truck terminals and attractors. Truck movement around warehouses generates environmental externalities including air pollution, noises, and et cetera. The rapid growth of freight demand and warehousing industry provides us incentives to examine the environmental justice issue from the perspective of warehousing location. From the studies on truck-related traffic congestion and accidents, they have caused many concerns in our neighborhoods. Such problems are majorly severe in large metropolitan areas where there is much freight demand.
Therefore the first question I would like to ask is, are warehouses locally undesirable? According to existing studies and data, trucks are major emitter of toxic diesel PM and NOx in California. They generate high levels of noises during operation and are the primary contributor to pavement damage. Based on our case study in the city of Compton, residents living near a warehouse cluster complain regularly about highway truck traffic and associated noises. Shown on the map, the location of warehouses and emissions, we can find spatial relationships between them. The blue dots identify warehouse locations. The red hot spots are where the diesel PM levels are high. They appear to be the areas where the warehouses are clustered. Then the second question is: are warehouses disproportionately located in disadvantaged neighborhoods? Traditionally studies focused on minority and poor population from the literature where three factors are identified including economic, sociopolitical, and racial. To be specific, low land rent and ineffective political power of local population attract the siting of locally undesirable land uses. The following decrease in land values due to environmental externalities further affects the housing dynamics and attracts the inflow of the disadvantaged population. Therefore we finally see the co-location of locally undesirable land uses and the disadvantaged population groups. The case study in the city of Rialto and the county of San Bernardino, people living on one side of the jurisdiction of boundary are not able to exert political power over the siting of warehouses on the other side. Lack of zoning coordination can lead to environmental injustice as well. On top of question 2, I would like to ask, among the neighborhoods with warehouses, does disadvantaged neighborhoods have higher warehousing activity intensity? Warehousing activity intensity is associated with truck footprints and disadvantaged neighborhoods may be more subject to a high concentration of warehousing activity.
I developed a set of models to examine the relationship between warehouse locations and population characteristics. Here are my dependent variables. I used binary, discrete, and continuous dependent variables to measure warehousing locations. This is a list of my independent variables. Population characteristics include the minority ratio and the median household income. Transport variables include access, zoning and economic attributes. All of these variables are associated with warehousing locations. My study area is Los Angeles region, which is the second-largest metropolitan area and the biggest trade gateway in the United States. It has a long history of warehousing industry development and a very comprehensive database for the analysis. The number of observations I included was 3,709 Traffic Analysis Zones (TAZs).
I found a strong linear relationship between household income and minority ratio in the region. I divided the neighborhoods into categories regarding minority dominance and household income levels. I achieved a medium or low income white group as the reference group. I tried to figure out whether other four groups would have higher probability of containing warehouses than the reference group.
Here are some statistics that can help us understand the data. Out of the 3,709 TAZs, about 3,005 were without any warehouses. 134 TAZs with at least 10 warehouses. We measured the distance from each TAZ to the nearest freight related facilities. We see those TAZs with warehouses have shorter distances to all the four type of facilities on average. Those TAZs with warehouses also have high population shares of Latino and black people, lower medium household income and low median housing values. These statistics seem to show that race and economic status are both associated with warehouse location. However, as I mentioned earlier, these factors are intertwined spatially with each other and we need stronger statistical tools to separate them.
Here is a map showing the location of warehouses and different types of neighborhoods. Blue dots are warehouses. Green areas are minority dominant and orange areas are white dominant. We can find the spatial relationship between minority dominance and warehousing location. Meanwhile in the areas surrounded by red circles, where low-income minority people are concentrated, there are very few warehouses. It is not consistent with traditional environmental injustice predictions.
Let's take a look at some regression results and see whether they can provide us some evidence on the minority-income puzzle. The first model is on the probability of obtaining at least one warehouse in a certain TAZ. We can see out of the four types of neighborhoods only medium-income minority neighborhood has higher probability of containing warehouses, compared to medium income white neighborhoods that are referenced. We can also find the coefficients for most of the control variables have expected signs and significances.
When we use the number of warehouses as the dependent variable in the second model the regression results are very similar and again the medium income minority variable stands out as a significant factor in estimating the number of warehouses in a neighborhood. When estimating the warehousing activity intensity both high income minority and medium income minority have significant coefficients. This means compared to the reference group these two groups have higher warehousing activity intensity.
Putting all these results together, here is some discussion. Given the spatial overlap of minority ratio and the household income, I disentangle the two factors using dummy variables and regression models. In most cases only median income minority neighborhoods have higher probability of containing warehouses while low income minority neighborhoods never show such tendency. This is different from traditional environmental injustice literature. One explanation is that in the Los Angeles region poor neighborhoods are concentrated in certain locations either dense South Los Angeles or the periphery of the region where warehouses are not present. Meanwhile zoning requirements and the availability of land development can be part of the story here. Secondly, I use different regression models to estimate the number of warehouses and no matter how I specified the models the results are similar. In the environmental injustice studies using discrete variables is not common. This reminds me that compared to traditional local undesirable land uses such as landfills and toxic facilities, warehouses are much more widely distributed and the scale of the environmental impacts would be even greater. Finally, among the factors that are associated with warehousing locations, population characteristics are not trivial at all. We can see this by comparing the beta coefficients of different factors. Therefore we should not ignore population characteristics when examining the environmental impacts of warehouse location.
To sum up, the bottom line of my research is warehouses are disproportionately located in minority-dominant neighborhoods but the relationship between locations and household income is not clear. In other words, environmental injustice exists in warehousing locations but not in a traditional way. In the near future I would like to further expand the scope of my study. Los Angeles is unique in terms of the demographics and I would need to include more metropolitan areas to test my findings. I will need to track the historical choices of warehousing location and provide more evidence on the mechanisms behind the spatial patterns. This is the end of my presentation. Thank you.
Nicole Coene
Thank you, Quan. I'd now like to start off the Q&A session with the questions posted online. The first question, Jean-Paul has answered but I wanted to give you an opportunity to discuss it further over the phone. Can you give some specific examples of "smart growth" principles that you feel are more significantly on a "collision course" with freight/warehousing systems and physical development of these structures?
Jean-Paul
I was expecting this question. That is why I had an answer already. The point was that what we have noticed again it varies. We didn't do a review of all of the smart growth strategies but it looks like smart growth strategies are focused on passengers and proving livability to quality of neighborhood which is all fine and perfect. But the only issue is those strategies and policies which are improving the quality of life of residents are many ways impairing the efficiency of urban freight distribution percent that is the concern we have and that is the point I was making. All the designs are impairing freight delivery and it looks like the strategies assume that people do not consume. That is paradoxical. You don't see the word freight ever mentioned. You say what is happening here and some people are involved in the data consumption activities and that is the point I want to make. This might be better part of a discussion on our point of view.
Sonke Behrends
What we observe when looking at the architecture visualization of new developments including high density development, they are often dominated by walking and driving cycles but also one or two electric cars. In all of these methods of transportation which I have seen I have never seen a commercial vehicle in them. This represents the awareness of the architects and urban planners for freight.
Nicole Coene
The next question is for Sanggyun. Do you think the warehouse location trends in different-sized metro areas in California also hold in different-sized metro areas around the U.S.?
Sanggyun Kang
Yes, I think these observations that I made in the study is quite consistent with other places. I have conducted preliminary studies using metropolitan areas in the United States. What I found is only five metros which are Atlanta, Chicago, Detroit, Los Angeles, and New York: they only show that the distances from the CBD can increase when counting warehouses. But when I count the warehousing employment, 19 of 20 largest metro areas show an increase in distance of warehousing employment from the urban center. So my answer is yes, I am expecting similar patterns and I think the use of four metro areas provide a general understanding of the situation right now.
Nicole Coene
Thank you. I have another question for Sanggyun Kang. How was the Alameda Corridor considered in the growth of warehousing in the "Inland Empire"?
Sanggyun Kang
The Alameda Corridor is built to accommodate very large freight volume from the port to the warehouses and the production (manufacturing) site near Los Angeles downtown. It is quite different from the Inland Empire which is mainly in Riverside and San Bernardino. So warehouses located specifically in that place really look for large parcels and they look for large operations.
Nicole Coene
Thanks Sanggyun. Please discuss the use of the CBD in the research. Is CBD data being kept up to date? Malls and other centers of business are often not included in many CBDs.
Sanggyun Kang
CBDs have been widely used as the highest density places. It is representing the central market of a metro area. I define this as the highest employment density zip code in the metro area. I found a consistent pattern of downtown (CBD) places in metro areas: Downtown Los Angeles and Downtown San Francisco, Downtown Sacramento, and Downtown San Diego have been all consistently picked up for being the highest density locations, which is interesting. For Measure 1 - the distance to a specific location from the CBD that are defined as the market' - it calculates the average distance to every warehouse. I think the point of question is that - Maybe we have to consider that a metro area might have multiple numbers of employment concentration centers. We see this in Los Angeles. We have multiple employment concentration clusters. That information is very valid in that we have to consider using Measure 2 "relative decentralization": it covers the aspect of having multiple numbers of employment centers. I hope that answers your question.
Nicole Coene
We have another question. How likely do you think it is that the higher income of minority residents is because of the location of the warehouses, rather than the other way around? The absence of warehousing, as well as absence of access to other industrial or warehousing/distribution employment drives down minority income?
Quan Yuan
Thank you. Thank you for this question. It is an interesting question. The interdependence issue has been discussed in the environmental injustice literature. I think in this case although there are several other factors to consider such as the growth of the warehousing industry in the last few years, the influence of warehousing the industry on employment attractiveness is not that important compared to other industries. I don't think the absence of warehousing industry would have a large influence on the minority incomes. I would need to take a look at a structural model when I have more data and time. Thank you.
Nicole Coene
Thank you, Quan. The next question is: are the conclusions valid for non-coastal ports (e.g. Great Lake ports, inland port sites, etc.?
Sanggyun Kang
As I mentioned in another question, I conducted a preliminary study using Chicago. I found a similar pattern of decentralization which means: 1) Chicago is one of the major international freight nodes and 2) it shows a significant increasing distance from CBD. Whereas in Cleveland, it did not show. This could be because of the size of the local markets and its role in international trade, but it requires more testing I think.
Nicole Coene
I would like to open up the phone lines. Please press * 1 to be put into the queue. This is open to all of the speakers. As you conducted your earlier research, did you find that cities were wowed by the potential for economic boosts like Amazon, IKEA, Apple, etc. to the extent that zoning traffic and other issues were ignored, or did you find that even the larger warehousing entities had to follow existing regulatory rules? In other words, did you see abnormalities that may indicate a pattern due to the new quick to knock issues with e-commerce?
Jean Paul
I already provided a brief explanation. It is a tough one to answer because it is very often the outcome of not that much public negotiation. But what I know based upon the Intermodal terminal industry especially with terminal operators and ports that there is an asymmetric relationship between the investor and the recipient. Under circumstances the investor has some kind of a leeway to negotiate favorable rule bending issues. Again that is more difficult to move through when you have a number of logistic zones and they want to attract potential investors and there is room there for the issue you have identified.
Quan Yuan
From my suggestion I think for the big companies when they want to do something in the neighborhood, communities may not have adopted political power on the siting of warehouses especially where the city of county that they are living in the warehouses is favorable. People need more information over the location division warehousing.
Sonke Behrends
I can also add some other dimension. Maybe it is not only municipalities wowed by the fact that logistic companies want to register. There are also some cases where there are already many logistics activities at very attractive areas because they are highly accessible to different mode choices. I know from some cases in Germany we saw that there are municipalities that are opposing additional development of logistic companies because first they are afraid of the traffic that will be generated. The jobs generated are not the highest paid jobs insinuating in modest income. They want to keep the attractive areas at a high level manufacturing zone.
Jean Paul
I can give you another suggestion. I will type it in the chat field. Dr. Laetitia Dablanc conducted a few interviews with municipalities in cities and counties regarding their attitude towards the warehousing industry. Maybe her study and her paper can explain and tell you the answer to your question.
Nicole Coene
Thank you. We don't have any questions at the moment but there are folks typing into the chat pod. I will pause to give them a minute. If you like to ask a question over the phone please press * 1.
The next question is for Sonke or Jean-Paul: is it fair to say that when evaluating industrial land use-related transportation/freight logistics strategies the focus should be on how communities are laid out or developed rather than their pure size? As an example, there are significant differences between the "urban" areas, including some older suburban communities, between an "older" city like Philadelphia and a "newer" city like Houston?
Jean-Paul Rodrigue
I will start. What we have noticed when we look at city logistics around the world and it is always context specific. The best way I can answer that question is each time you have a freight distribution strategy it is strongly contingent upon the urban spatial structure and you could say in some way the history of the land-use and the land-use regulations. That is in a sense what I could have contributed to this answer, that indeed the context specific part is quite important.
Sonke Behrends
I will add to that. Not only are a city as a whole, but there large differences. You can have a very small city but the tracks maybe very narrow or where you have problems observed in Manhattan or other metropolitan areas. You have in metropolitan areas maybe logistic areas. I guess to answer the question; there are major differences in small cities vs urban areas.
Jean-Paul Rodrigue
We try to assess to start the issue, to what extent the density as well as regulation are factors and we can use the term segregation or divergence and we can start to narrow this down and we will have some good pointers in terms of what level of density you could expect these issues to arise. Eventually you can move further on to identify this specific with land-use patterns. This could be prone are less prone to these strategies. We know very little at this point because this is a recent field.
Nicole Coene
Thank you. The next question is: in your investigations, how were Foreign Trade Zones brought into the equation? Was warehousing value looked at differently within a FTZ versus outside of an FTZ?
Sanggyun Kang
I'm not sure if this question is directed to me but I haven't looked at Foreign Trade Zones yet.
Jean-Paul Rodrigue
Some FTZs are more related to port activity and some are related to airport activity so that is a factor which will be part of the equation. Some are more used as a way to store and release and transform to change taxation regime of the commodities. I cannot answer that query in a satisfactory manner.
Nicole Coene
Thank you. It looks like we have gotten to the end of questions and I don't see anything else coming in. I think we will go ahead and close out. The recorded version of this event will be available within the next few weeks on the Talking Freight website. The next seminar will be held on December 16, 2015 and the topic is Highlighted FHWA Freight Transportation Initiatives - Intermodal Connectors and Freight Analysis Framework. I encourage you to join the Freight Planning Listserv if you have not already done so. Thank you to the presenters and to everyone for attending. Enjoy the rest of your day.
