Transportation Deployment Casebook/Streetcars in the Twin Cities

Introduction to Streetcars
Three different technologies were used by street railways in the Twin Cities between 1872 and 1954: horsecars, cable cars, and electric streetcars. The power was the biggest difference between these three; otherwise they were relatively similar technologies. The street railways in the Twin Cities ran on standard gauge (1435mm) rails. Some of the suburban lines ran on ballasted rails on private right of way, but for the most part the streetcars operated on embedded rails in mixed traffic. Electric power was generated at several steam and hydroelectric power plants in Minneapolis and Saint Paul. Several small substations converted the electricity to 600 V DC that was used to power the cars via overhead catenary.

For passengers, the advantages of streetcars were the speed of travel, low cost, and the number of destinations available. From the perspective of the operating company, the advantages were the efficiency of vehicles on fixed rails and the ability to operate on schedule despite muddy streets and winter weather.

The main market was the working classes. The rich have always had transportation choices, from horses and carriages in the early years to automobiles later on. The working classes could not afford these options and did not purchase automobiles en masse until after World War I. Streetcars provided frequent service throughout the day and were used for both work and shopping trips. Twin City Rapid Transit (TCRT) actively developed a market for leisure trips to resorts out in the country and profited from special event services to downtown nightlife, sports games, and the Minnesota State Fair.

Before Streetcars
At least in the Twin Cities, not much existed before the first horsecar line developed in 1872. At the 1870 Census, Minneapolis (including the City of Saint Anthony, with which it merged in 1872 ) had a population of 18,079 and Saint Paul had a population of 20,030. Minneapolis and Saint Anthony had only been incorporated since 1858, and Saint Paul had been incorporated since 1849. The Mississippi River was the primary means of transportation at this time, and it was very effective at bringing settlers to Minnesota as well as carrying furs, lumber, and flour away. However, navigation was difficult above the confluence with the Minnesota River, short of Minneapolis, and the river was not a practical solution for intracity travel. A few people might have used horses and carriages to get around town, but most people just walked everywhere. This limited the size of both cities, since workers needed to locate within a reasonable walking distance of their jobs.

The first rail line, the Chicago and Rock Island, was built into the area in 1854. The first line connecting Saint Paul to Minneapolis, the Minnesota and Pacific, was completed in 1862. At this point, it would have been possible to travel from Minneapolis to Saint Paul by train, but the line was built with freight opportunities in mind. This connection encouraged the cities to grow toward each other and sparked the development of Minneapolis as a milling center. These developments, as well as the rapid growth of the Minneapolis-St. Paul area from a couple thousand settlers in 1850 to about 40,000 people in 1870, encouraged local politicians and business leaders to plan for future growth. Saint Paul granted its first horsecar franchise to a group of businessmen in 1866, and Minneapolis did the same in 1867 although neither started operation right away. On July 15, 1872, the opening day for Saint Paul's first horsecar line, the Saint Paul Pioneer wrote that the city "has seen the necessity of streetcars to reduce its length and breadth of 'magnificent distances.'" This quote reflects the realization that both cities needed a faster mode of intraurban transportation in order to expand beyond what have become their central business districts today.

Invention of Streetcar Technology
Streetcar service began in 1872 with horsecars. By this point, horsecars were a mature technology, having started in New York in 1831. They were a hybrid of other technologies, using the horses that were already owned to pull stagecoaches and omnibuses but running on rails instead of a carriage base for a smoother ride. Horsecars in the Twin Cities ran on tracks fashioned out of iron straps attached to wooden stringers, usually at a narrow gauge of 1067mm. This led to economies of scale later on since these rails were being produced in massive quantities. The cars themselves were 3 meters long and the driver sat outside on a platform. The passenger compartment had straw on the floor and a small stove, but was notoriously cold in winter. Cars frequently derailed due to the track conditions. Additionally, the use of horses caused problems. They were expensive to care for, had to be switched out after 4-5 hours due to fatigue, left manure on the streets, and had to be replaced when they died from disease or old age.

Although steam power was used on mainline railroads at this time, it was considered undesirable for use on street railways due to the noise, soot, and potential to spook horses. Nevertheless, the Minneapolis, Lyndale, and Lake Calhoun Railway built a steam powered line from downtown Minneapolis to Lake Calhoun in 1879, which was later extended to Lake Minnetonka. The steam engine was enclosed in a box in an attempt to address concerns, but it was still disliked and the steam trains were retired in 1890. (The line itself was taken over by the Minneapolis Street Railway Company in 1887.)

The next technology to enter the market was cable cars. Cable car technology was invented in California for use in gold mines and was first applied to street railways in San Francisco in 1873. Downtown Saint Paul is in a low-lying area surrounded by hills. Many of these were too steep for horsecars, so the first cable route was built in 1887 up Selby hill. Another line was built on E 7th Street in 1889. Most of these lines were removed a few years later, although the section up Selby hill stayed until a tunnel was complete in 1906. Cable cars were powered by a central power plant and could climb steep hills, but had a number of disadvantages as well. The cable ran in a slot below the tracks and cars could grip or release the constantly running cable to control speed. Although having one central power plant was good for efficiency, one cable break or a problem at the power plant would shut the whole line down. There were problems with snow and ice getting into the cable trench, and the cable cars were not very fast.

Electric power was the next advancement, and it went through several revisions before it was implemented in the Twin Cities. Ernst Werner von Siemens presented a third-rail powered locomotive in 1879 in Berlin. This was a successful technology, and third-rail power is used today in subway systems around the world. However, it was unsuitable for streetcars due to the danger to horses and pedestrians. Charles Van Depoele in Minneapolis, and others around the country, tested electric cars with overhead power in the 1880s. None of these were very successful until Frank Sprague solved both the power collection and electric motor questions in 1885. His design collected power from a single overhead wire and transmitted the power from the motor to the axle through gears and bearings (as opposed to belts and chains, which were prone to breaking). A demonstration line was built on 4th Avenue S in Minneapolis in 1889. It was so successful that both cities converted their horsecar lines to electric power between 1889 and 1892, switching to standard gauge (1435mm) in the process. The Minneapolis and Saint Paul Street Railways were also consolidated into Twin City Rapid Transit (TCRT) in 1891 to help obtain financing for electric conversion. Electric power was 600 V DC, supplied by coal-fired steam or hydroelectric plants. As the system expanded, the power plants switched to AC transmission since DC can only be transmitted over short distances. Substations were built around the city to transform the AC from the power plants to the 600 V DC for the streetcars. After this point (1892), the technology remained more or less the same. The rolling stock consisted of 13.1m long cars with controls at one end and air brakes (after 1902). The vast majority of cars owned by TCRT throughout its history were homebuilt in its Snelling Shops.

Early Market Development
During the market development phase, the streetcar lines served the entire urbanized area of Minneapolis and Saint Paul and extended beyond it in places. It is hard to identify a "market niche" during this period because the system served the entire city and could conceivably have been used for any trip purpose where the origin and destination were both in the same city. In 1890, the systems in Minneapolis and Saint Paul were connected with a line along University Avenue, capturing the market for travel between the cities.

Functional enhancement was important throughout the lifecycle of the streetcar, but especially in the early period. As can be seen in Figure 2, there were four years between 1880 and 1900 that had a net loss of track. While the loss in 1891 was part of the shift from horsecars to electric streetcars, the losses in the other years were due to reconfigurations of service. The street railways were the first urban mass transit in the area, and the Twin Cities essentially developed around streetcar lines. Most of the streetcar lines were radials out from downtown along a particular street or set of streets. TCRT (and its predecessors) generally paired two radial lines together to form a route. This was so that cars operated through instead of terminating in downtown. Turnarounds and layovers could then be done at the ends of lines, which were not as busy as downtown. However, these line pairings changed over time. Geography, service frequency, and logical destination pairs were considered for service pairings. As the service pairings changed over time, some downtown links were no longer needed and were subsequently abandoned, while others were added.

Functional development was key to the development of the region and the success of the streetcar system. Thomas Lowry, the businessman who merged the Minneapolis and Saint Paul Street Railways into TCRT, also had real estate interests. As a result, streetcar extensions were done with land development in mind and often occurred before the surrounding land developed. In other words, streetcar development led land development, and streetcar lines have also been shown to be a significant predictor of residential density. In addition to 'streetcar suburb' development, TCRT attempted to capture the leisure market by developing lines to Excelsior, White Bear Lake, and Stillwater. The company built amusement parks on Big Island in Lake Minnetonka and at Wildwood on White Bear Lake. It also operated ferries and sightseeing tours on Lake Minnetonka. TCRT lines served the Minnesota State Fairgrounds as well as the baseball stadiums for the Minneapolis Millers and Saint Paul Saints. The company ran special event trips to serve these events and carried a significant portion of the people attending.

The Role of Policy
Since the horsecar companies in Saint Paul and Minneapolis were founded about 40 years after the first horsecar company, they had the opportunity to learn from established systems. Due to the status of the horsecar as a hybrid mode, it is likely that the first systems drew from the operational policies of stagecoach lines as well as from railroads. By the time of the conversion to electric power, TCRT owned many of the same assets as a railroad (rolling stock, tracks, shops) and its management were also involved with railroads. Lowry was the president of the Soo Line Railroad at the same time TCRT was converting to electric streetcars. Because of the similarities of these organizations and the familiarity Lowry and other management had with railroads, TCRT developed the same kind of hierarchical structure.

The Minneapolis and Saint Paul companies began as separate entities, but followed a similar process. The Minneapolis Street Railway Company organized in 1873 and started building track. It was granted a franchise by the City of Minneapolis in 1875, and this was confirmed by the Minnesota Legislature in 1879. The franchise agreement with the city specified the locations of the lines it was to build, the types of service, and set deadlines. The agreement gave the company a monopoly, but set fares and required the company to maintain the city streets around the tracks. These agreements could be, and were, renegotiated as conditions changed. However, TCRT never managed to get out of the fixed fare and street maintenance clauses, although it would have liked to.

Growth of Streetcars
The growth of streetcars was mostly fueled by private investment. In the horsecar period, the Minneapolis Street Railway Company chose not to pay dividends, opting to reinvest all of its profits in system expansions. The system grew very quickly during this period and regularly expanded into undeveloped areas in an attempt to spur development. Therefore, the franchise agreement with the city served mostly as a minimum level of service. The companies were interested in expanding well beyond what the contract required, but had to maintain service on their original lines. In general, the cities were supportive of system expansions because they allowed the population to spread out. As a result, the Twin Cities became characterized by houses and small apartment buildings instead of the large tenements typical of eastern cities like New York.

However, when the conversion to electric power began, both cities required their street railway companies to electrify all the lines in a short period of time. Both companies took on debt to do this, and merged to form TCRT in 1891 as a way to get financing from eastern capitalists. The conversion to electric power would have happened anyway, but it might have taken longer without the contractual requirement. It was certainly a good investment, as companies throughout the country that switched from horse to electric power observed an increase in gross receipts of 30-300%. An analogy could be drawn to the conversion of railroads from steam to diesel or electric power. The most important lines switched over early on, but less important lines kept the old source of power for years after the first conversions. TCRT would have been unlikely to keep horses around for very long since they were expensive to operate, but the company might have tried to avoid taking on so much debt without the public intervention.

The Mature Phase
As the system matured, it began to feel pressure from automobiles and two clauses of its municipal contracts: set fares and street maintenance. Henry Ford developed his first car in 1896 and others soon followed. The car was initially a toy for the rich, but as it became mass produced it turned into a commodity that more and more people could afford. The automobile competed directly with the streetcar for passengers, but the increased prevalence of automobiles also made streetcar operation more difficult because of the increased traffic on the roads.

The fixed fare policy took away the flexibility of TCRT to set its own price. These types of price controls were common at the time and applied to railroads as well. For a period of several years, the fare was set at 5 cents. This meant that, even if ridership stayed the same on a particular line, it became less and less profitable as inflation increased the operational expenses. Fare increases had to be negotiated with the Railroad and Warehouse Commission and the Commission often refused. When a fare increase was refused, TCRT had to respond by trying to increase ridership, or cutting service and deferring maintenance.

Finally, street maintenance became a burden in the later years. In the 1870s or 1880s when the contracts were first developed, most streets were unpaved so the street maintenance clause meant that TCRT was responsible for maintaining its own tracks and plowing the streets after winter storms (extensions were used to plow the right lane since streetcars typically ran in the left lanes). However, the way these clauses were written meant that TCRT was required to remove its tracks and replace them at its own expense if the city decided to pave a street. This was a significant expense when Minneapolis and Saint Paul began to pave all their streets. It became even more expensive as automobile traffic increased, since streets needed to be repaved more often. On at least one occasion, the conversion of 6th Avenue N to Olson Memorial Highway, TCRT refused to replace its tracks and converted the line to buses instead.

A report to the Mayor of Minneapolis in 1952 summarized the problems facing the streetcar system. Cars had taken riders away and increased the level of traffic congestion. The streetcar equipment was outdated and needed reinvestment. The franchise agreements had been developed very early on and now severely limited the actions TCRT could take to improve profitability. Finally, traffic patterns had changed faster than the company could reconfigure its network. The rapid development of suburbs after World War II was primarily responsible for this. This report and a subsequent one recommended changes to the regulatory framework and suggested other changes like distance-based fares. They recommended regulation on a metropolitan, as opposed to a municipal level, which was later implemented in the form of the Metropolitan Council.

The streetcar network is gone now, so the system cannot be reconfigured to meet the needs of transportation in 2011 and beyond. However, some of the lessons from streetcar operations are applicable to its successor, buses, and others can be applied to efforts to reestablish a rail transit network in the Twin Cities. First, government policies should be reevaluated periodically to see if the are still applicable to the current market. Policies developed at a time when streetcars had a monopoly on urban transportation hurt their ability to compete when the monopoly eroded away. Transit systems are publicly owned today, making negotiations about government policy a little easier. Next, infrastructure costs should be assessed in an equitable manner. In its last few years, TCRT was essentially subsidizing car travel by paying to relay its tracks every time the road was repaved, and by continuing to plow snow in winter. When it was the primary user of the roads, it made sense for TCRT to pay for maintenance and plowing. As cars became the dominant user, the costs should have been shifted to car owners, but they weren't. Today, transit mode share is very small in the Twin Cities. If an effective transit system is a priority, then the subsidy on car travel should be reduced by increasing car-related taxes or implementing road pricing. It makes very little sense for government to be subsidizing car travel and simultaneously trying to convince people to take other modes.

Data and Curve Fitting
The data set used in this analysis is a shapefile of historical transit routes in the Twin Cities produced by the Metropolitan Council. This file has separate records for each link in the regional transit network from 1872 to 1954 showing the year the link opened and the year it shut down. Horsecars, cable cars, streetcars, ferries, and 1948 bus routes are included. The bus routes were eliminated for this analysis since they were a replacement technology, but the ferries were left in since there were relatively few of them and they developed and were operated in conjunction with the streetcars. Using the shapefile, the length of each segment was computed and the file was dissolved by starting year, then by ending year to obtain the length of track added or subtracted from the system each year. The track kilometrage in the system is displayed in Figure 1 and the change each year is shown in Figure 2. These values represent centerline mileage, so two tracks on the same street are not double counted. This measure might be more accurately described as system extent as opposed to kilometers of track.

The curve parameters are shown in Table 1. Because streetcars have gone through birth, growth, maturity, and decline already, K, the carrying capacity, was already known to be 602.162km. A three parameter logistic function of the form $$S(t)=\frac{K}{1+e^{-b(t-t_{0})}}$$ was fitted to the data. This was done in two phases: one representing system growth to maximum extent (before 1931), and the second representing system decline (after 1931). A least squares regression was used to determine the optimal value of b and to find $$t_{0}$$, the inflection point using the relation $$t_{0}=\frac{intercept}{-b}$$. For the growth phase, b was found to be 0.161 and the inflection point was in 1899. For the decline, b was determined to be -0.0586 and the inflection point was in 1974.

The S-curve for the growth period seems to be a good fit. The $$r^{2}$$ value was 0.928, the t-statistic for the intercept was -27.0, and the t-statistic for b was 27.1. Knowing that the maximum $$r^{2}$$ is 1.0 and the level of significance for the t-statistic is greater than 2, the curve is a reasonably good fit and the estimated variables are statistically significant. Qualitatively, the curve follows the data very well in the birth and maturing stages, but not as well in the growing stage. The rapid growth of the system in the 1880s seems to have happened around 5-10 years before the curve predicted it would. The halfway point, 300km, was reached in 1890 as opposed to the inflection point of 1899 suggested by the regression. By the 1900s and 1910s, network growth lags slightly behind what the curve predicted. However, we can also see that TCRT made major capital investments about once every ten years in this period: large sections of new track opened in 1905, 1906, 1913, and 1914. Given that new lines tended to open in fairly large segments, we would not expect the data to follow the curve exactly. With a sufficiently large sample size, like all street railroads in the United States, the data should be smoother and might fit the curve better. On the other hand, some of the anomalies are caused by external factors, such as the financial markets. The slowdown in the early 1890s, odd because it occurs right where the logistic curve predicts the highest rate of network expansion, can be explained by the Panic of 1893 which made financing for large capital expenses nearly impossible to find.

On paper, the curve for the decline period is also a very good fit. The $$r^{2}$$ value was 0.928, the t-statistic for the intercept was 16.28, and the t-statistic for b was -16.02. The $$r^{2}$$ value is exactly the same, while the t-statistics are lower but well above the statistically significant threshold. Qualitatively, the curve is not such a good fit because it is nowhere close to the starting and ending points. However, because these are only two points, the least squares method favors a curve that best matches the rest of the data. Looking at what actually happened, a negative exponential function might better represent the decline. In any case, the regression produced a curve that represents the slow decline of the system in the 1930s and 1940s. The facts that this curve's inflection point is in 1974 and that it would never go down to zero  suggest that TCRT's streetcar network was abandoned prematurely. This view has been held by several analysts, and is supported by the fact that several of the managers who oversaw the conversion to buses (including Fred Ossanna and Barney Larrick) were later convicted of fraud and sent to prison. Canals and railroads, which are other examples of mature transportation systems, declined to about half of their maximum extent and have continued to operate. It might have been possible for TCRT to cut some services and continue to operate the most profitable lines, but public opinion in the 1950s was not very supportive of transit and especially not of streetcars.