Transportation Deployment Casebook/2020/Rhode Island Streetcars

Introduction
This case study reviews the policy and technological developments behind the deployment of streetcar technology in Rhode Island, USA. Whilst the quantitative analysis primarily focuses on the period 1885 - 1920, the qualitative review considers events preceding and succeeding this period to identify the causes and catalysts for the growth, maturation and decline of this mode of transport.

Historical Context
Prior to the introduction of the horse-drawn streetcar in circa 1840 the predominant mode of urban transport was by foot. Omnibuses and hansom cabs were available, but fares were too expensive for the average city dweller. Whilst horse-drawn streetcars sped up journey times fares were still out of reach of the working class.

A key step in the technological development of the streetcar was the initiation of cable car services 1870’s whereby the passenger car was pulled along the track by a cable located under the roadway. This for of technology had significant advantages over horse-drawn carriages in cities with steep topography. Cable car systems suffered from mechanical reliability issues which were compounded during inclement weather.

Streetcar technology resulted in 1884 from union of innovations in electrical traction technologies, with the horse-drawn streetcar passenger wagon, as a mode of mass transport. In many forms the basic technological platform remains similar to this day. The emergence of the electric streetcar coincided in a time of rising wages which when coupled with significantly lower fares made this new form of transport accessible to the masses, helping to establish it and promote the rapid expansion of urban systems. Early systems

In high level terms it can be considered that the horse drawn carriage established the drawn passenger car as a unit of mass public transport. The passenger car motive effort was mechanised through the introduction of cable technology, and then improved through harnessing electrical power. Along this technological journey the quality of passenger transport increased through improved reliability, comfort and speed. This established the electric streetcar as the primary technological platform, and further technology advances were evolutionary rather than revolutionary.

Technological Evolution
Early trials of electric streetcars used existing horse-drawn track and modified passenger cars as the test bed for this new technology. As such the track gauge, a key technical parameter of the new technology was 'locked-in', or pre-determined, by the previous mode of transport. Indeed following early adoption of the electric streetcar in 1887 the Woonsocket Street Railway Company reverted to horse-drawn power before later reintroducing electric traction, displaying the interoperability of these two technologies. Cable cars were also developed to run on the same gauge track as horse-drawn and electric streetcars.

In Rhode Island many early electrical streetcar systems were borne out of the electrification of existing horse-drawn system. Alongside common technological features the operators therefore transferred across many policy and policy characteristics including organisational structure, pricing mechanisms, ticketing and timetabling.

The Union Railroad Co. in Providence, R.I., experimented with two forms of electrification: battery and electric wires. Whilst batteries removed the need to install unsightly cables they required frequent charging and could not provide sufficient power to overcome moderate inclines. Overhead wire electrification therefore became the preferred approach for all systems in Rhode Island, although some battery powere systems were operation in other part of the USA.

The early electric streetcar technology deployed in Woonsocket dragged a small truck that run above the power cable, and provided power through a slack cable back to the streetcar. These truck units could derail from the power cable and were difficult to reinstate. They were replaced by trolley-pole collectors whereby a sprung pole engaged a rolling wheel to the underside of the power cable, later iterations replaced the wheel with a grooved shoe. Photo examples of both modes of power transference are shown in the images below.



The American Context
Wide ranging technological advances at the turn of the 20th century established the growth of cities. Innovations including structural steel and the elevator eliminated the historic eight storey building height ceiling. The adoption of cheaper electric lighting removed constraints on the working day and working hours increased.

The introduction of electric streetcar technology in the 1880s both coincided with, and facilitated a period of significant urban growth in the US, and by the 1920 census the majority of the population resided in urban areas for the first time. This rapid growth was experienced by Rhode Island, being a key manufacturing state and financial centre. The speed, capacity and fare cost of streetcar technology facilitated a horizontal spreading of cities and was the catalyst for residential decentralisation. Streetcar technology enjoyed a symbiotic relationship with this newfound economic propensity. The increase in employment opportunities led to an increase in personal wealth, facilitating an increased uptake of transport services. There was a positive feedback loop between transport demand and the horizontal expansion of residential areas. This was exploited by promoters and developers of streetcar systems which sought to capitalise on land value capture from expansion of systems into undeveloped areas.

Local politicians were instrumental in promoting the early local streetcar development, and provided private enterprises with low cost loans, cash subsidies, parcels of public land and long term franchises at healthy operator profit margins. The politicians enjoyed significant public support in bringing this new transport technology to their cities and streetcar systems were a source of civil pride and city leaders boasted of the extents of their respective networks.

However in many cities, including in Rhode Island, a number of competing streetcar operators installed multiple lines that often converged in city centres, causing disruption during construction and congestion when operational. There was little incentive for the competing operators to coordinate their systems and the result was a incoherent system operated to maximise private sector profits.

Over time previously supportive politicians campaigned to reduce fares and nationalise the streetcar operators. Whilst popular with the general public this criticism often led to backlash from an influential upper class. This reversal in public and political approval led created a more adversarial relationship between the private streetcar operators and public officials who were more liable to reject fare increase and franchise requests. As a result private firms identified more profitable unregulated enterprises to invest in.

Issues with implementation
The following section describes the specific issues and policies that were encountered, overcome and developed to implement electric streetcar service in the key Rhode Island cities. It can seen that there are many shared experiences in terms of policy challenges and organised opposition to the introduction of services across the state.

Woonsocket
Prior to electric streetcar implementation in Rhode Island there were a number of suburban horsecar operators typically restricted to relatively small extents within each urban area; the largest operator was the Union Railroad Co. of Providence, that state capital. Slow speeds limited public appeal and the realistic extent of the networks. The first attempt at an electric service was by the Woonsocket Street Railway Co. in Woonsocket, a small regional R.I. town, which experimented with the new form of technology shortly after commencing operations over a 5 mi line in 1887, one year after the first electric streetcar revenue service commenced in Alabama. Given the short line distance the electric service garnered novelty appeal rather than providing an improved transport service over horsecars. The trial attracted opposition from local business owners, the fire brigade and citizens. Criticism included objection to street rails, claims the electric vehicles scared horses and the electrical risks of live wires. These fears were likely compounded by technical issues and safety incidents experienced during the trial of the immature technology. After approximately 3 months the Woonsocket trial was abandoned and the railroad returned to horsecar operations, before reintroducing electric services in 1893.

Newport
As a major 18th century seaport Newport hosted an affluent upper class, principally as a summer destination for resident of New York. Despite attempts to introduce horsecar services in 1864 and 1884, the upper classes and private carriage operators opposed the development, objecting to the street rails damaging private carriages, journey disruption, and improved public access to private beaches. Despite the minority exerting sufficient opposition to influence city transport policy during this period; by 1885 sufficient local support existing to pressure the Newport city council to hold a vote on the introduction of a streetcar system which was passed. The wealthy continued to oppose the newly formed Newport Street Railway Co., establishing a lobby group that included William Astor who was undertaking similar action to prevent subway construction in New York - demonstrating the similarities facing transport developments targeted at the public across wealthy US cities. Despite extensive opposition the inaugural service began in August 1889.

Providence
In 1882 the Union Railroad Co. was the sole provider of streetcar services in Providence, however it did not service College Hill, a wealthy enclave, due to the adverse incline of the streets. The Providence Cable Tramway organisation was established in 1882 and petitioned the city to introduce a cable car system in this suburb based on the success of similar technological deployment in San Francisco. There was extensive opposition from the wealthy residents of the suburb to the introduction of mass transport, and the Union Railroad protecting its monopoly position. Despite this opposition the concept enjoyed significant public support and was approved by the city. At this time Union Railroad was experiencing growing criticism around enriching its shareholders whilst paying negligible taxes and failing to provide promised expansions to it's network. Capitalising on this discord the tramway proponent announced an extension of the cable tram network, over Union Railroad track, which was granted for free by the city despite provisions for reimbursing the track owner within streetcar franchise agreements. The cable system opened in 1989 and was bought out by the Union Railroad Co. in 1890 to re-establish their monopoly. The Union Railroad Co. commenced electrification of their network in 1890 and the cable system was converted in 1895.

Final public acceptance
The Union Railroad Co. was the largest streetcar operator in Rhode Island in the early 1890's undertook experiments of electric operation to garner support for network wide upgrades. They had provided reduced fares on two lines, ostensibly to buy local goodwill support for the experiment. This is evidence for the monopoly position that the company enjoyed since in a truly competitive environment there would be limited scope for price reduction in this manner. The experiment was a success, with public feedback noting the enhanced quality of the electric street car in terms of speed and rise comfort of the ride over horsecars. However it was likely the increase in land values, and real estate development opportunities, that solidified support from a previously opposing wealthy elite.

Market expansion & consolidation
As shown in the quantitative analysis below streetcar systems enjoyed significant growth in Rhode Island and by 1902 there were over 15 streetcar companies operating within the state, although often under a complex lease or ownership arrangements. By 1902 the majority of the streetcar systems in the state were consolidated into The Rhode Island Co. which leased the United Traction holdings and their leases through a network of companies. The complexity can best be described as follows: "J.P. Morgan owned stock in NY, New Haven and Railroad Company which owned all the stock of the Providence Securities Company. The Providence Securities company held securities and owned stock in the Rhode Island Company which in turn owned the Columbian Street Railway Company, Providence and Burrillville Street Railway Company and Woonsocket Street Railway Company.  The Rhode Island Company also leased and operated from the United Traction and Electric Company who owned shares and bonds of the Rhode Island Suburban Railway Company, Pawtucket Street Railway Company and the Union Railroad Company.  The Union Railroad Company owned stock and leased the Providence Cable Tramway Company.

To meet federal antitrust laws, in 1914 the New Haven Railroad was forced to divest its interests in all New England streetcar systems, including the Rhode Island Co. Following this the Rhode Island Co. faced financial difficulties in maintaining lease payments and investments required to adapt to the introduction of early automotive transport. The company's leases were terminated and the majority of its streetcar systems were consolidated into a new entity established in 1919, the United Electrical Railways Co. (UER). To mitigate against further risks or disruption to the Rhode Island public transport system from the intensive mergers and acquisitions of the previous two decades, and lack of investment UER was placed under the regulatory control of the Public Utilities Commission. This was a significant political act to provide public control over the private sector which had historically enjoyed high levels of autonomy and political influence.

Quantitative Analysis
An analysis of the lifecyle of the streetcar system in Rhode Island was undertaken to identify the following system parameters:


 * Growth rate;
 * Mid-point/inflection year;
 * Developmental phases (birthing, growth and maturity);
 * Predicted system size (was likely to be the maximum size of the installed system in the study period).

Methodology
A common lifecycle feature of transportation modes throughout history has been the S-curve (or Sigmoid) profile that can be used to describe the rate of adoption of different transport technologies. The Logistic function provides a common S-curve profile that can fitted to the system growth data. For this assessment the installed extent of track is used as the measure of system growth. The lifecyle model can be characterised using the logistic equation:

$$S_t= S_{max}/1+e^{(-b(t-t_i))}$$

Where:

St = system extent at time t in terms of installed track

Smax = maximum system extent (the peak length of installed track in Rhode Island)

t = time/year

ti = time/year of inflection (the year of peak growth of system deployment)

b = logistic growth rate (i.e the steepness of the S-curve)

This can be re-written in the form:

$$ln \frac{S_t}{S_{max}-S_t}=bt+c$$

Where:

c,b = model coefficients

The left hand side of the above equation can now be expressed in the form of the straight line equation ($y=mx+c$ ) ,where m is the gradient and c is the intercept with the y-axis. in terms of Y using a single linear regression of the known values to allow values to be estimated for coefficients c and b.

$$y=bt+c$$  &   $$y=-ln \frac{S_t}{S_{max}-S_t}$$

The value for Smax should be the maximum extent of installed track in Rhode Island. Whilst this peak may have occurred during the study period, 1885–1920, this was verified using sensitivity analysis for the value of Smax by considering varying values of Smax and adopting the value with the highest R2 value to fit the closest S-curve to the measured data.

The original form of the logistic equation can then be used to plot the model S-curve for review against measured data.

Analysis Data
The McGraw Electric Railway Manual and state archives provide a comprehensive record of track installed by each streetcar operating company on an annual basis over the study period - reproduced in the table below. The installed track extent is expressed in terms of single track equivalent installed in Rhode Island state. The McGraw manual and state records provide consistent measures of installed track at each year. A copy of the full dataset and analysis spreadsheet can be found here.

As expected the track extent increases year on year throughout the study period, with the exception of decreases in 1919 and 1920. During the latter part of the study period a number of the streetcar lines were under the operational control of the Rhode Island Co., which was in financial difficulties and ceded control of a number of lines in 1919 and 1920. It is likely that the lines were still operational under an alternate company but this could not be tracked through the records and therefore a drop in total track extent is recorded.

Results
The results of the S-curve fitting analysis are shown in the table below. The following graph shows the actual extent of streetcar track installed based on historic records and the model derived installed extent based on the logistic equation S-curve profile.



The graph below shows the year on year change in actual installed track extent based on historic records.

Discussion
The various stages of the Rhode Island streetcar development lifecycle have been interpreted on the S-curve profile above. The transition from birthing period appears to be approximately 1890 and the system reached a mature stage of development from approximately 1909.

It can be seen that the The optimal value of Smax is 20 mi larger than the maximum installed system extent by 1920.This may reflect some latent system expansion that had not been realised by 1920. Even when the model was analysed excluding the years 1919 and 1920, which displayed an unexpected reduction in installed track extent, the Smax value exceeded the total installed track extent and supports a latent expansion theory. The R2 value of the modelled S-curve indicates a confidence >95% which indicates a the model is a close representation of the measured data. However it can be seen from the graph that whilst the model S-curve is reasonably representative of the the birthing and maturity stages of the lifecycle, it tends to under-represent the rate of track installation during the growth period, and in particular the period 1897-1902. Whilst this was a period of rapid construction in a highly competitive environment the methodology of reporting the historic installed track in this period appears to be skewed by the listing of the Rhode Island Co. streetcar system only once it became operational, rather than spread the track installation over a pre-operational period as for the majority of developments. As such the actual curve may over-represent the peak rate of track installation.

Alternate scenarios
In addition to reviewing the lifecycle of the statewide streetcar system an analysis was also undertake on individual streetcar systems, and the systems within each urban area.

Due to the extensive consolidation of streetcar operators in R.I. over the study  period only the Union Railroad Co. and Rhode Island Co. have sufficient data  to undertake a meaningful lifecycle assessment at an individual system level. This appears to be due to the face that these companies operated the majority of services on all the  systems over their respective periods of operation as discussed above. Any system expansion during this time was undertaken through these two companies whilst the remainder of the lines under their control remained close to their as-built configuration at the point of sale. Of these two systems a well fitted S-curve could be fitted to the Union Railroad Co. data whilst the curve fitted to the Rhode Island Co. was only weakly fitted. The weak fit is likely because this system only contains data from the late growth and mature stages of the lifecycle.