OLR Research Report

August 22, 2008




By: James J. Fazzalaro, Principal Analyst

You asked a number of questions relating to the proposal to operate commuter rail service between New Haven, Hartford, and Springfield. You wanted to know who owns the rail line and the associated transportation assets and if this could be transferred to the state. Additionally, you asked:

1. What the recommendations for the project are,

2. What rail equipment will be required to start up the service and its possible cost,

3. What other improvements or facilities may be required,

4. What the projected ridership is for the proposed start up service,

5. What are the subsidy needs,

6. What are the projected capital construction and operating costs,

7. What is the implementation timeline,

8. What environmental and other approvals may be required, and

9. What some possible sources of funding may be used to pay for the project.


The New Haven-Hartford-Springfield commuter rail service project involves a proposal to provide train service during the morning and afternoon peak commuting period along the existing Amtrak-owned rail line connecting these three cities. Service would also be provided to the communities of North Haven, Wallingford, Meriden, Berlin, Newington, Windsor, Windsor Locks, and Enfield. Stations would be constructed in North Haven, Newington, and Enfield where none currently exist. All of the other towns currently have rail stations, but they all would undergo improvements and enhancements under the proposal.

The current proposal for recommended start up service involves bi-directional weekday service during morning and evening commuting periods on a 30-minute schedule. A possible future “Full Build” scenario could bring service up to as much as every 15 minutes.

Current estimates for construction for the recommended start-up service are approximately $291 million. Current estimates of operating costs are approximately $10 million annually. At the fare schedule currently being estimated for the service, an annual subsidy of approximately $8.96 million would be required to cover operating deficits. The implementation study estimates daily ridership of approximately 2,200 for the recommended start up service.

Because of the broad range of questions you asked about the project, we have divided this response into several sections that discuss each topic in significant detail. We have also added an explanation of a provision contained in the pending Amtrak reauthorization legislation currently in conference committee in the U.S. Congress. This provision states a sense of Congress that the proposed New Haven-Hartford-Springfield rail service is considered an “important transportation priority.” Should this provision remain in the final legislation, it could enhance the possibility of some federal funding for the project.


Amtrak owns the 62-mile rail line from New Haven to Springfield Amtrak also owns the rail stations in Berlin, Meriden, and Windsor Locks. The stations in Wallingford and Windsor are owned by the respective municipalities. The New Haven and State Street stations are owned by the Connecticut Department of Transportation. The Hartford station is owned and operated by the Greater Hartford Transit District.


Amtrak also owns the branch platforms at the Berlin, Hartford, Meriden, Wallingford, Windsor, and Windsor Locks stations. The line is single track in some locations and double track in others. (The line was originally entirely double track, but Amtrak removed certain sections of the second track a number of years ago in a cost reduction effort. All of the single tracking is located in the Connecticut portion of the line. The entire track located in Massachusetts is double track.)

The New Haven-Springfield line is currently classified as Class 4 track. Under Federal Railroad Administration standards, Class 4 track can provide a maximum allowable operating speed for freight trains of 60 miles per hour and a maximum allowable speed for passenger trains of 80 miles per hour.

Since Amtrak is a federally-funded quasi public corporation, acquisition of its assets might be accomplished differently than if the state were trying to acquire private assets. Congress, in effect, controls Amtrak through the legislation that sets it up, establishes goals for its operation, and funds its capital and operating costs. It is possible that a transfer of ownership to the respective states, if it were to be done, could be done by Congress through an amendment to the federal laws. While there has been some mention made of the Amtrak possibly transferring the line and stations for a token amount, I am not aware of any serious proposal in this regard, nor any analysis that attempts to assign an acquisition value for the track, signal system, right-of-way and station assets. Thus, it is difficult to speculate on what this might cost Connecticut.

The NHHS rail service implementation study did not base any of its recommendations on a change in ownership status of the rail line and other assets.


Before addressing some of your other questions, it may be helpful to summarize the service plan recommendations that were made in the NHHS implementation study done by Wilbur Smith Associates for the DOT. The final report was issued in June 2005. The study examined a range of possible implementation strategies ranging from a “Minimum Build” scenario that involved no additional track construction, no off-peak service, and peak hour train frequency of 30-35 minutes to a “Maximum Build” scenario that involved complete double tracking of the line, hourly weekday plus weekend service, and peak hour train frequency of 15 minutes. The steering committee and study team reviewing the options developed a recommendation action plan for start-up service that involved the following:

● Bi-directional service, Monday through Friday, on a 30-minute peak hour schedule (at least 14 one-way trips) between New Haven and Springfield

● Adding at least 18 miles of extended double track (i.e., restoring the second track at selected locations to improve reliability and provide the necessary 30 minute headways at critical times in the three anchor cities for the service)

● New service to supplement Amtrak's existing intercity service in the corridor with, if possible, adjustments to Amtrak's schedule

● Augment the nine existing stations on the line (counting Springfield) with construction of three new stations in North Haven, Newington, and Enfield

● Improve the current station in Windsor Locks by providing facilities to accommodate a waiting area and transfers between the train and a shuttle bus connection to Bradley International Airport

● Modify local bus service to provide service to the stations

● Modify all 12 stations to provide high-level platforms, grade-separated pedestrian facilities and amenities, bicycle racks, and additional parking

The new station in North Haven is planned for the area near the Route 40 connector. The Newington station is planned to be adjacent to the proposed New Britain-Hartford Busway. The Enfield station is planned for the area in Thompsonville at Bigelow Commons.

The service level recommendation assumes stops at all stations. Based on this assumption, the estimated travel time from New Haven to Springfield, including all stops, is 90 minutes. Should consideration be given to operating an express train that would stop only at the three anchor cities—New Haven, Hartford, and Springfield—the average travel time would be reduced by 15 minutes; however, overall ridership might be reduced.

Future Improvements to Reach Full Build Scenario

The implementation study envisioned the following additional improvements for future service beyond the proposed start-up service.

Double-track the remaining 20.6 miles of single track sections so that service could become as frequent as every 15 minutes

Construct second high-level platforms and grade-separated pedestrian facilities at Wallingford, Berlin, and Windsor Locks to accommodate the additional double-track segments

Construct an additional station in the Wharton Brook area on the former Pratt and Whitney property as development takes place

Provide new commuter rail parking in the new Meriden parking structure to be built with downtown development plans in Meriden


The NHHS implementation study looked at two types of train equipment that could be used on the new service—conventional commuter rail equipment in the form of a locomotive and three passenger cars and a set of three self-propelled rail cars known as Diesel Multiple Units or DMUs. The study's cost estimates for rail equipment were based on an assumption of using the conventional rail equipment.

The study concluded that the amount of new equipment that would be required would depend on which of two possible scenarios was selected. In one scenario, the DOT and Amtrak would maintain separate equipment pools. In the other, there would be a single combined equipment pool (either DOT or Amtrak). In the separate pool scenario, the study concluded that DOT would need six train sets plus two spare sets and Amtrak would need three train sets plus one spare set for a total of 12 sets of equipment. In the combined pool scenario, the estimate was that a total of 10 train sets would be required (eight for the service plus two spare sets).

A traditional train set on the service would use two types of cars—cab cars and coaches. The cab car has an engineer's compartment while the coach does not. Using a cab car on the opposite end of the train from the locomotive creates a “push-pull” configuration that allows the train to operate in either direction without moving the locomotive to the other end of the train when changing directions.

The study assumes an average cost of a coach of $1.37 million and an average cost of a cab car of $1.87 million. The assumed cost range for a diesel electric locomotive was from $2.8 million for a basic “no frills” locomotive to $4.5 million for a high-end AC power locomotive. The rolling stock cost estimate uses the $4.5 million cost for the locomotives.

The cost estimate for new equipment required for the start-up service is a total of $70,140,000 broken down as follows—$36 million for eight locomotives at $4.5 million per unit; $19.18 million for 14 coaches at $1.37 million per unit; and $14.96 million for eight cab cars at $1.87 million per unit.

The cost for a single three-car DMU set was estimated to be $6.8 million.

Each type of equipment has distinct advantages and disadvantages for different types of service. DMUs or other types of self-propelled equipment was estimated to be less expensive to purchase and maintain than locomotive-hauled equipment and has some operating efficiency advantages for the type of service being proposed for the NHHS line, but equipment availability is less certain and operating experience is more limited. The implementation plan states that “To be conservative, this study has developed its operating plan using conventional commuter rail equipment.” (p. 8-21)


One major element of the proposed project is construction of a maintenance facility in the New Haven area for the rail equipment. The implementation study discusses several possible locations for the facility and estimates costs of land acquisition, construction, and possible environmental clean-up at approximately $20.7 million.  I believe that the analysis for this aspect of the project was probably conducted separately from any consideration being given for the new maintenance facility proposed for servicing the new M-8 cars being purchased for the Metro North service because the decision to make the large M-8 purchase had not been made by the time the report was finalized. This aspect of the project may require further consideration given the plans for the new New Haven facility. It may be possible to accommodate the NHHS equipment differently that contemplated in the implementation study.


The implementation study identified the following four types of passengers who might be attracted to the NHHS service.

Commuters accessing employment hubs in the three anchor cities

Intercity rail ridership to points off the corridor, specifically connections to Amtrak service in New Haven and Springfield

Riders going to Bradley International Airport

Off-peak non-commuter and weekend users

The primary tool for estimating ridership was the Connecticut DOT Statewide Travel Model that is primarily a commuter model which uses population and employment to calculate expected trips. While the calculations for the Full Build scenario included additional modeling to estimate off-peak and weekend potential ridership, the analysis for the recommended action plan includes primarily peak hour service, plus Amtrak mid-day and weekend service. Thus it is primarily commuter ridership.

Based on the DOT model, the year 2025 projected daily commuter ridership for the recommended plan was 2,208. This compared to 1,606 for the Minimum Build scenario and 3,440 for the Maximum Build scenario. In addition to the projected commuter ridership, 10% was added to account for non-commuter ridership. Thus the projected 2025 average daily ridership was 2,428 (2,208 commuters/220 off-peak riders). Using the standard of 254 days of weekday service projected in the plan, this equates to approximately 616,700 annual trips. These would augment an estimated 616 daily trips being taken on Amtrak's Vermonter service.

To address concerns that the ridership projections might be too conservative, additional analysis was performed using a DOT review of the 2000 Census Journey to Work data for towns being served by the Shore Line East (SLE) commuter service. DOT's review indicated that the SLE service capture rate of the potential commuter market is approximately 5%, which is more than twice the projection from the Statewide Travel Model for the NHHS service plan. While the study states that the two systems are not directly comparable, it concluded that the higher captured rate for SLE could be considered the higher end of a range for the NHHS plan. While the study concluded that this should be considered an optimistic figure, it resulted in a high end estimate of 5,000 daily trips.

As noted above, the Statewide Travel Demand model uses population and employment data to calculate expected trips. To my knowledge, neither the expected fare structure nor the influence the price of gasoline may have on travel choices are part of this model. However, as indicated by the range of projected ridership based on the Minimum Build, Maximum Build, and Recommended scenarios indicates, service frequency is accounted for in the projections. Please note that the ridership analysis was conducted primarily in 2004 (long before the current run up in fuel prices).


The one-way fare schedule proposed for making revenue projections was a range of $2.25 to $9.50. The monthly fare range was projected as $51 to $210. This is based on a formula of $2.419 plus 14.6 cents per mile over 10 miles rounded to the nearest 25 cents. The DOT developed this fare structure, which is similar to the 2005 SLE fare structure. Monthly fares were calculated at 50% of the one-way fares based on 42 trips per month. The assumption was made that commuters would all use a monthly pass and non-commuters would pay a one-way fare.

Based on this fare structure and the 2,428 projected weekday trips, the plan estimates daily revenue of $4,400. Annual revenue based on 254 days of weekday service was projected as $1,117,600 for the recommended service plan. Annual operating costs based on using conventional rolling stock were projected at $10.079 million, for a farebox recovery ratio of 11%. The annual operating deficit was calculated at $8.96 million; making the subsidy per passenger $14.53.

Annual operating costs for the connecting bus services proposed as part of the implementation plan was estimated to be $3.8 million. The 2004 Connecticut Transit farebox recovery rate was applied to calculate the annual operating deficit for these services. This was estimated as $2.75 million annually.


The implementation plan projects the recommended action plan capital costs to be a total of $291,259,000. However, you should be aware that these costs resulted from estimates made primarily in 2004 and early 2005 and recent experience with cost escalations experienced in the planning for major capital projects such as the New Haven harbor crossing and the new rail car maintenance facility in New Haven suggest that actual costs are likely to be higher than those projected in the implementation plan.

The capital costs were broken down as follows.

● $70,140,000 for train equipment

● $20,696, 000 for the maintenance facility

● $80,966, 000 for station construction and improvements

● $33,235,000 for double track construction

● $505,000 for bridge improvements

● $2,500,00 for Amtrak flagmen

● $83,217,000 (40% additive) for contingencies

TOTAL COST $291,259,000

Capital costs for the bus connecting services recommended in the implementation plan were estimated as $3.6 million.

The additional capital costs estimated in the implementation plan for the future Full Build option are approximately $96.2 million. As noted above, this includes full double tracking and a new station at Wharton Brook.


The operating costs for the recommended start-up NHHS service were calculated by multiplying the projected annual train miles for the start-up service times (eight round trips per day) by a representative cost per train mile. Annual train miles were estimated to be 251,968 not counting movements necessary for maintenance purposes.

The cost per train mile was calculated using the SLE service as a model. The analysts used SLE as a model because they felt it most closely resembled how the NHHS service would operate. The estimate was based on an average of the actual SLE costs per train mile for 1999, 2000, and 2001. This was calculated at $40 per train mile. Multiplying this by 251,968 annual train miles yielded the annual costs of $10,078,720.


The implementation plan contains a projected timeline for the various stages of project implementation. The assumption made in the projected timeline appears to be that actions would commence immediately following publication of the implementation plan final report in June 2005. The timeline as originally envisioned called for completion of:

● Environmental assessment by June 2006

● Final design and property acquisition by April 2008

● Equipment procurement by the end of 2009

● All required construction by April of 2011

● System testing by October 2011.

Operations would presumably commence thereafter. Thus the projected timeline shows a period of approximately five years and nine months from the completion of the environmental assessment to the end of system testing and beginning of operations.

These timeframes are clearly behind schedule as the DOT is currently still in the environmental assessment phase of the project in mid-2008. The aforementioned potential delay in this phase should EPA insist on preparation of a full EIS makes it appear unlikely the project can be fully implemented before late 2013 or 2014. One possible area where the implementation timeline might potentially be compressed is in the area of equipment procurement. The proposed timeline estimates a period of three years for procurement of equipment—from January 2007 through December 2009. Depending on the decisions that are made in this area, there may be reason to believe that it would not take that long to acquire the necessary equipment.


As you know, DOT has initiated the environmental assessment for the project. The preliminary review of the potential environmental issues contained in the implementation plan concluded that addition of the double track sections was not expected to result in any significant adverse environmental impacts since they would be within the existing railroad right-of-way.

The primary environmental impacts are likely to be associated with the station construction aspects of the project. However, the feasibility study concluded that the adverse impacts were minor in nature, due largely to the fact that it was located within an existing rail line and avoided significant environmental recourses in placing the stations. The study also notes that there are broad transportation benefits of the project, it has substantial public support, and it is not expected to be highly controversial.

The implementation study concludes that the magnitude of the impacts identified through the preliminary analysis “indicate that extensive and/or detailed impact studies at the level of an Environmental Impact Statement (EIS) would not be necessary for this project and that an Environmental Assessment (EA) would adequately address the project's potential minor impacts.” (Final Report p. 8-5)

This is perhaps why the implementation study's proposed timeline for the project allocates only 12 months for an environmental assessment rather than 18 months to two years, which is typically the timeframe required for a full EIS. However, as you know, there have recently been published reports that the EPA intends to require a full EIS for the project regardless. If this decision cannot be reversed, the environmental analysis phase of the project would clearly become longer than envisioned. But, as noted above, the implementation timeline is relatively long and it might be possible to save time in other aspects of the implementation process. For example, DOT may want to assess whether it could shorten the period for rail equipment acquisition by looking at acquiring existing rolling stock available from other sources, such as it did when acquiring the VRE rail cars, rather than having the equipment built new.

The following is a list of some of the possible, though not definitive, environmental impacts associated with some of the station construction the implementation study identified:

Potential 100-year floodplain impacts should be coordinated with the Connecticut DEP at the State Street, Meriden, Newington, Windsor, and Windsor Locks stations.

The State Street and the proposed North Haven stations are located within the coastal boundary and thus subject to the Connecticut Coastal Management Act.

The proposed improvements at the Berlin station on land with farmland soils require review by the National Resources Conservation Service

Possible wetlands impacts at the Windsor and Windsor Locks stations must be further evaluated and, if they cannot be avoided, construction will require an Inland Wetlands and Watercourses permit from DEP and, possibly, a U.S. Army Corps of Engineers permit.

DEP's Stream Channel Encroachment Lines list indicates that encroachment lines exist all along the Connecticut River within Windsor Locks. If construction impacts cannot be avoided, a DEP Stream Channel Encroachment permit will be required.

Activities at the Wallingford, Berlin, Hartford, and Windsor Locks stations, and at the proposed station areas in Newington and Enfield, will have to be coordinated with the State Historic Preservation Office either because the stations are historic buildings (Wallingford and Hartford) or there are other historic properties near or on the station sites. Additional environmental documentation and mitigation activities may be required.

Due to their proximity to the Connecticut River, the report recommends that the State Archeologist investigate the Windsor and Windsor Locks sites as an archeologically sensitive area before the ground is disturbed as required by the Section 106 regulations of the Advisory Council on Historic Preservation.


The NHHS commuter service proposal is designated by state law as one of more than 25 strategic transportation projects and initiatives (CGS 13b-79p). These are generally referred to as “Tier 1” transportation strategy projects. Another of the Tier 1 initiatives authorizes use rehabilitation of rail passenger coaches for use on SLE, NHHS, and the Metro North branch lines. Except in a few instances, the law does not authorize specific amounts for these initiatives. Instead, they are eligible for use of $1 billion in bond funding that was authorized in 2006.

The implementation plan assumes a 90/10 split of capital costs between Connecticut and Massachusetts. This is based largely on the fact that 10% of the line is located in Massachusetts and 90% in Connecticut. The plan assigns none of the costs for double tracking to Massachusetts.

The division of operating costs between the two states would have to be negotiated as part of a service agreement. The implementation plan makes an initial recommendation of a 90/10 split, but this could also be based on some criterion other than the relative track mileage.

Operating costs would likely be a state responsibility. There are generally no sources of federal transportation funding for operating expenses, although under certain conditions, the implementation plan indicates that funding received under the Congestion Mitigation and Air Quality program (CMAQ) may be available for some operating costs. The plan projects an annual operating deficit of almost $9 million for the start-up service, none of which is currently budgeted for in the DOT operating budget. For the Full Build service alternative, the annual operating deficit was estimated to be $44.7 million.

As already mentioned, the NHHS project is already designated as eligible for state bond funding. How much of the capital costs will have to be funded through this source will have to be determined. As indicated above, the capital cost estimate for the start-up service is $291.26 million and, for the Full Build scenario, $387.4 million. There are several sources of federal funds that could provide for some of the capital costs, but they vary in terms of the state's likely ability to access them.

Some of the most significant possible sources of federal funding identified in the implementation plan are summarized below.

FTA Section 5309 New Starts program—This is a large funding program, but extremely competitive and it utilizes a travel demand forecasting model to determine user benefits attributable to projects that differs significantly from the model DOT uses. The New Britain-Hartford Busway project is already a possible New Starts funding candidate.

FTA Section 5307 Urbanized Area Formula Grants Program—Available federal funding is allocated by formula to all areas with populations over 50,000. Funding is used for locally determined capital projects and transportation-related planning.

FTA Section 5311 Non-urbanized Area Formula Grants Program—Similar program for assistance for public transportation in areas under 50,000.

Flexible Federal Highway Funding—Some of the federal funds administered by the Federal Highway Administration and apportioned to the state under certain programs (for example, the Surface Transportation Program) can be transferred between highway and transit projects based on state and regional priorities. Flex funding designed for use in transit capital projects must derive from priorities established in the metropolitan and statewide planning process and must be included in the Statewide Transportation Improvement Program before they can be transferred.

Railroad Rehabilitation and Improvement Financing (RRIF) Program—The RRIF program is administered by the federal Railroad Administration and provides loans and loan guarantees for rail capital projects for things like acquiring, improving, or rehabilitating intermodal or rail equipment or facilities, including track, buildings, and shops; refinancing outstanding debt for these purposes; and developing new intermodal or rail facilities. Eligible borrowers can be railroads, state and local governments, government-sponsored authorities or corporations, and joint ventures that include at least one railroad.

TIFIA Financing—The Transportation Infrastructure Finance and Innovation Act of 1998 established a federal credit program administered by the U.S. DOT that provides direct loans, loan guarantees, and standby lines of credit for surface transportation projects of national or regional significance. This includes transit projects. A project's eligible costs, as defined by federal regulation, must be expected to total at least $100 million or, in the alternate, be at least 50% of the state's federal-aid highway apportionments for the most recent fiscal year, whichever is less.



Congress is currently considering S 294 and HR 6003, known as the Passenger Rail Investment and Improvement Act. This is a multi-year reauthorization and appropriations act for Amtrak and for other rail purposes. The legislation is currently in conference committee. Among its provisions is a section making certain findings regarding public transportation generally and specifically with respect to public transportation in Connecticut. It also makes a Sense of Congress statement relating to the proposed NHHS project. The provision (Section 226) states:

“It is the Sense of the Congress that expanded commuter rail service on the rail line between New Haven, Connecticut, and Springfield, Massachusetts, is an important transportation priority, and Amtrak should work cooperatively with the States of Connecticut and Massachusetts to enable expanded commuter rail service on such line.”


The bill also requires Amtrak to submit a report to Congress and to the two states' transportation departments on the total cost of uncompleted infrastructure maintenance on the New Haven-Springfield line.


While it is not clear what the full implications of this Sense of Congress provision may be, designation of the expanded commuter service proposal as an important transportation priority is likely to have some significance with respect to the possibilities of some federal funding for the project.