U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590

Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Highway History


Thomas H. MacDonald on Toll Roads

Thomas H. MacDonald, who headed the FHWA's predecessor agencies from 1919 to 1953, and his top assistant, Herbert Fairbank, were recognized as highway authorities internationally. During the 1920s and 1930s, they helped the State highway agencies to create the Nation's first "interstate system" in the form of paved two-lane U.S. numbered highways. Fairbank wrote the 1939 report to Congress, Toll Roads and Free Roads in which he and MacDonald conceived what is now called the Dwight D. Eisenhower National System of Interstate and Defense Highways.

However, for all their visionary skills, they are sometimes criticized for their opposition to toll roads. For example, Phil Patton in Open Road: A Celebration of the American Highway (Simon and Schuster, 1986) described how Toll Roads and Free Roads rejected the toll option based on traffic figures showing low demand in most corridors, and predicted that the Philadelphia to Pittsburgh corridor would carry only 715 cars a day. "The BPR had no notion that the construction of new superhighways, like the introduction of such inventions as the telephone and the auto itself, might create its own demand." Similarly, Dan Cupper in The Pennsylvania Turnpike: A History (Applied Arts Publishers, Second Revised Edition, 2001) pointed out that ridership after the turnpike opened on October 1, 1940, "completely defied the pessimistic predictions - 715 vehicles a day - of the U.S. Bureau of Public Roads."

Financing the turnpike had not been easy. The State had created the Pennsylvania Turnpike Commission to issue bonds to finance construction, but had stipulated that the bonds would not have the full faith and credit of the State behind them. This stipulation made the bonds virtually unsaleable. In 1938, the U.S. Public Works Administration (PWA) gave a grant of $25.1 million to Pennsylvania for the turnpike, while the U.S. Reconstruction Finance Corporation purchased $35 million in bonds to complete the financing package.

The original 160-mile section of the Pennsylvania Turnpike, Irwin to Carlisle, was an immediate success, prompting a toll boom after World War II. While Pennsylvania extended its turnpike, Maine began construction of a 47-mile turnpike inland from U.S. 1 between Portland and the New Hampshire border. The Maine Turnpike opened in 1947. America's Highways 1776-1976, the FHWA's Bicentennial history, explained:

With the completion by Massachusetts of a free expressway to the New Hampshire line and the opening of the Maine Turnpike, traffic congestion became intolerable on New Hampshire's 15-mile portion of U.S. Route 1. Unable to obtain an increase in the gas tax from the Legislature to finance major improvements of the old winding road, the Highway Commission asked for authority to build a toll road financed by State bonds. The bonds, backed by the faith and credit of the State, sold for a net interest rate of only 1.58 percent. The toll road, built in only 1 year, opened to traffic in 1950 and was enormously profitable from the outset-so much so that the Legislature decided to build two more toll roads in other parts of the State. [p. 167]

While the debate in New Hampshire was underway, Ralph Blagden, Managing Editor of The New Hampshire Sunday News, wrote on April 8, 1947, to request MacDonald's views. Blagden's letter, consisting of only four paragraphs, prompted a six-page reply by MacDonald. The length of the reply suggests that MacDonald was concerned about the growing toll boom and aware that his predictions were not being borne out. Whether this speculation about the motive for such a lengthy reply is valid, the letter lays out MacDonald's thinking about toll roads during the crucial period between authorization in 1944 to designate a "National System of Interstate Highways" and creation of a program to build a "National System of Interstate and Defense Highways" in 1956. MacDonald's letter follows, with Blagden's letter presented at the end:

In The Roads That Built America: The Incredible Story of the U.S. Interstate System (Barnes and Noble, 2003), Dan McNichol put it this way:

By 1954, the turnpike-toll road boom was in full swing. Many of these toll roads brought in nearly ten times what they cost to operate, proving Chief MacDonald wrong in his belief that drivers would never pay to drive on a toll road, no matter how efficient, if a free road was nearby.


Washington 25

April 22, 1947

Mr. Ralph M. Blagden
Managing Editor, SUNDAY NEWS
48 Hanover Street
Manchester, New Hampshire

Dear Mr. Blagden:

We are in receipt of your letter of April 8 requesting information relative to toll roads, a request which you say is inspired by the bill pending in the New Hampshire legislature to provide for the construction of a toll highway and a recent article in the Saturday Evening Post.

The views of the Public Roads Administration concerning the advisability of a resort to toll financing for the construction of highways have been clearly presented on previous occasions, notably in the report entitled Toll Road and Freed Roads, submitted at Congressional request to the 76th Congress in April 1938. Copy of this report is supplied herewith.

Our views as presented in Toll Roads and Free Roads remain substantially unaltered. Even in the particular instances in which traffic can be attracted to toll highways sufficient in volume to repay the costs of financing, construction and operation, we regard the toll device as an inadvisable and expensive expedient. So far as we know, there is no existing toll road (except some short roads built to afford tourist access to mountain tops and coastal resorts) on which earnings have been sufficient to cover all costs of the undertaking. It is not generally understood that the Pennsylvania Turnpike was subsidized by a substantial PWA grant amounting to $29,250,000 of the original cost of $71,500,000.

For reasons inherent in a practicable application of the toll method of financing, no toll highway can be used by all of the traffic that would be served by a highway of identical design and location operated free of toll.

Some usage is inevitably discouraged by the heavy charge exacted in toll. How heavy that charge may be would be appreciated if it were understood that a rate of one cent a mile payable for passenger automobiles - the rate depended upon in most of the proposals - is equivalent to a gasoline tax of from 12 to 16 cents a gallon.

Another large fraction of the traffic that would use a free road, that cannot use a road of the same location and design standard operated as a toll facility, is made up of vehicles moving on trips of shorter lengths than the distance between toll gates. Entrance and exit are possible on a toll highway only at its gates. Every gate represents a substantial capital investment, and continuing large operating expense in salaries of attendants, etc. Since the best that can be hoped for in nearly all toll facility proposals is a bare possibility of financial soundness, the number of gates must be kept at a minimum, their spacing at a maximum. In some of the present proposals access points average as much as 25 miles apart. Analysis of the lengths of highway trips extending out of cities before the war in eleven representative States, including New Hampshire, showed that the number of trips less than 20 miles in length varied from a minimum of 79 to a maximum of nearly 88 percent of the total number of all such trips in the several States. Trips less than 10 miles in length varied from 56 to 70 percent of the total. In New Hampshire the corresponding percentages were 82.7 for trips less than 20 miles and 56.6 for trips less than 10 miles. These figures are averages representing the character of usage of rural roads generally. On particular highways the percentages may differ materially from these averages; but trips of short length are on all highways a large fraction of the total usage.

It follows, then, that apart from any consideration of the ability or willingness of travelers to pay the special charges imposed, no toll road can be expected to serve much of the traffic potential to a road in the same location operated free of toll. And this is true even if the so-called free road is designed on a controlled-access facility, because accesses to such a facility will invariably be spaced at much shorter intervals than is financially feasible on the toll road.

So, toll roads, if they are built, should be clearly understood to be partial facilities, capable of serving only a part of the traffic requiring service in the directions in which they extend and in the areas they traverse. A substantially paralleling free road must always be maintained to serve the remainder, probably the large remainder, of the traffic.

And here enters the most serious danger of the toll road. However great may be the remainder of traffic left to the parallel free road, however insistent may be the need for substantial improvement of the facility it affords to that traffic, the design standards of the free road must be held materially lower than those of the toll road, else the latter will be robbed of nearly all of its essential support.

Moreover, to this proposition there is a corollary equally damaging to the theory of the toll expedient as an economic solution. That is, that the toll road, in order to attract a barely sustaining traffic must be designed to standards adequate for the accommodation of a traffic far greater than that which it can actually attract. The Pennsylvania Turnpike, for example, could serve with complete convenience a traffic of 12,000 vehicles a day. Its actual average traffic is between three and four thousand vehicles a day, the bare minimum commonly considered essential to justify the construction of a 4-lane, divided highway.

But, it is asserted that resort to the toll expedient is the only way to get needed high-type facilities quickly; it is the only feasible way, it is said, to raise the necessarily large cost. Now, of course, the places where the most expensive new facilities are needed most quickly, namely, within and in the near vicinity of cities, are the very places where it becomes utterly impracticable to employ the toll method. The Post writer tells us that, and in this instance he is right. But, still, he accepts the argument that only by tolls are desired improvements to be obtained without intolerable delay. Like all of the other arguments advanced that, too, is fallacious. If the capacity and facility to be offered by any toll road are actually required, they can be provided in a so-called free highway more cheaply than in the toll road. Apart from the clear advantage of not having to maintain expensive toll collecting facilities, this is true. Toll roads are built with borrowed money, nearly always with money borrowed by way of what are called revenue bonds, that is, bonds predicated solely upon the problematical earnings of the roads themselves. Such bonds necessarily carry a relatively high rate of interest. If money must be borrowed to get the high-standard road now, it can be borrowed more cheaply (at a lower interest rate) in bonds predicated upon normal tax revenue and the faith and credit of the State at large than it can by the means commonly employed in toll road financing.

In sum, a toll road is bound to afford only a partial, and a communally unsatisfactory and expensive solution of the traffic problem existing in the area of its location.

It must afford only a partial solution because it can serve only a part of the traffic requiring service along its line.

It must afford a communally unsatisfactory solution because it requires for its protection a depression of the standard of improvement of alternate free roads.

It cannot fail to afford an expensive solution because:

  1. To attract any traffic at all it must be designed to standards beyond the need of the traffic it will get. This is signalized by the relatively high charges that must be exacted for its use.
  2. It must always remain an added facility incapable of dispensing with the need of alternative free facilities that must be continued in service at cost to the community.
  3. It must bear the large cost of toll collection.
  4. It is commonly, and almost necessarily financed with revenue bonds at interest rates inevitably higher than the rates the public would have to pay for money borrowed to create an identical free facility.

It has been stated above that the standard of alternate free roads must be kept depressed for protection of the investment in a toll road. But what is going to happen if, or probably I should say when, the traffic remaining to be served by a free alternate rises in volume to the point where only high-standard improvement can provide the facility and safety (we are more conscious of safety requirements nowadays) it required? That thought may well give us pause. Are we then going to improve the free road to its required standards and let the consequences to the toll road be what they may? Or are we just going to "free" the toll road, that is, take over its bonds as a public obligation and suspend the toll collection? The first course would mean inevitably large loss to the toll road investors. The second would mean either some reduction of the investors' expected incoming, or the public assumption of a debt bearing a higher rate of interest than it would have been necessary to pay had the road been built free in the first place. Moreover, if we are tempted to choose the second alternative we will probably find that the road built as a toll road is not located properly to serve as an acceptable all-duty free highway. So what can we do?

To revert to the argument that the toll expedient offers the only hope of immediately needed improvement, what does that imply? It implies that at a time when public highway authorities find it difficult to obtain at reasonable cost the materials, equipment, labor, engineers and contractual services sadly needed for the correction of serious deficiencies of the public road system, competing demand for these scarcities is going to be thrown upon the market. Scarcely a wise or a sound suggestion, I would say.

A matter of puzzlement to the author of the Saturday Evening Post article it seems, is the different attitudes assumed by the Public Roads Administration toward toll roads and toll bridges. The rather simple explanation lies very close to the surface where it can be discovered without great difficulty by anyone who is really looking for it. It is that a toll bridge can be, and for its financial success invariably must be a monopoly. It has no free bridge competition. Hence it exists as the complete and only facility serving a particular traffic stream. It should always be publicly owned and operated and freed when its tolls have paid for it. Under these conditions it is open mainly to the objection generally that it is built with money raised by revenue bonds, an objection which need not lie against it if, as is usually feasible wherever a toll bridge should be contemplated at all, tax-predicated bonds are issued.

Right here we come upon ground fertile for the raising of toll-road propositions. The general public, which wants and truly needs modernized highway and bridge facilities, is made to appear at times to be unwilling to pay for what it wants and needs. In California, recently, after a most careful study of the highway needs of the entire State by a joint legislative committee, aided by competent advisors, plans were projected for a program of general highway improvement extending over a future period of a number of years. The plans proposed, if permitted realization, would result in the meeting of all reasonably foreseeable needs of improvement of the entire State highway system, or all substantially important secondary rural roads, and of the arterial streets of cities, and, additionally, they would provide for the outright creation of a new system of high-standard expressways joining all of the State's larger cities. All of these benefits would result from an orderly program of effort, applied within a period as short as by any other means such a general improvement could be effected. For financing, a gasoline tax increase of 2 cents a gallon was proposed, and some increase in the taxes applicable to motor vehicles. The proposed tax increases met with most vigorous opposition from representatives of the petroleum industry and tax-affected road users. Yet, here there was the offer, not of a few stretches of highway, but of a systematic improvement of the whole network of roads and streets in accordance with need, at a cost equivalent to about 7 cents a gallon of gasoline consumed, compared with the 12- to 16-cent per gallon equivalent of practically all toll road proposals.

It generally appears that opposition to gas taxes and their increase arises from the petroleum industry rather than from the consumers of gasoline who actually pay the tax. Why this should be so is something of a mystery. We have looked long and hard to find evidence, if there be any, that a gasoline tax at any level thus far reached has had the least depressive effect upon the consumption of gasoline. We have found none. The tax is the ideal substitute for, and preventive of, toll financing. It is actually a form of general toll that has none of the defects of direct tolls. It has been the principal means by which the public has obtained the road service it now enjoys, if not the best and most desirable road service.

How a petroleum industry that insists as a cardinal principle upon an absolutely unrestricted competition of outlet will fare if many of our principal lines of highway are occupied by toll facilities is an interesting speculation. One thing is sure, toll roads, wherever they are built, are going to have to scratch for a living. The income derivable from exclusive concessions for various traffic-catering services, including the supply of gasoline, is going to be well-nigh irresistible. It has proved so in all undertakings to the present. It will continue to be so; and gasoline purveyors, those who choose to seek the privilege, are almost sure to find themselves saddled with a charge (or a tax) that they can't pass on; with death to the opportunity for competitive selling in the bargain.

Just one thing more. You ask if I think there is any foundation for your assumption that a factor in the current wave of toll road proposals is "some strong organized pressure possibly from powerful financial interested who underwrite the financing of the bonds." Isn't that a purely rhetorical question? I ask you.

You are at liberty to quote this letter if you wish to.


Very truly yours,

Thos. H. MacDonald
Commissioner of Public Roads


48 Hanover Street Manchester, New Hampshire

April 8, 1947

Federal Bureau of Roads
Washington, D. C.

Dear Sirs:

I am managing editor of The New Hampshire Sunday News in Manchester, New Hampshire. I am seeking information relative to toll roads. As you may or may not know, the State is considering a bill for a toll road across the seacoast area connecting with the Maine toll road running between Kittery and Portland.

You probably have seen the article in last week's Saturday Evening Post which asserts that the nation can get adequate roads only through a system of toll roads or through some assessment on road users. I wonder if you have any material which would help us to arrive at a correct judgment in this matter, and also if you could have somebody in your Bureau answer, on or off the record, some of the principal assertions in the Post article. Anything that you could furnish us that would clarify the record of these toll roads would be of great use to us in this matter.

I don't know whether you would be willing to comment on this or not, but you might give us some indication as to whether I am on the right track or not. This comparatively recent push behind toll roads has seemed to me to indicate some strong organized pressure possibly from powerful financial interests who underwrite the financing of the bonds.

Is there any foundation for my assumption that this is a factor in the picture?


Very truly yours,

Ralph M. Blagden
Managing Editor

Updated: 06/27/2017
Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000