Federal Bridge Policy: Past and Future

 Response to 1967 Disaster Created A Program That Has Gradually Lost Its Focus

Originally published in Transportation Weekly, August 15, 2007

Imagine this: a highway bridge over a major river, in service for 39 years, suddenly collapses while full of afternoon rush hour traffic. A significant number of lives are lost. The White House and the Department of Transportation immediately order nationwide bridge inspections. The bridge collapse hits home for the chairman of a powerful Congressional committee that oversees the highway program, and he not only holds a series of hearings to draw attention to decaying bridges, he also introduces and eventually pushes through legislation establishing a special federal program to pay for bridge replacement.

Sound familiar? It happened almost 40 years ago, in 1967. The initial federal response to that bridge disaster was strikingly similar to what has happened thus far in response to the August 1 collapse of the Interstate 35W bridge in Minneapolis, and the lessons of the 1967 incident and the history of the resulting federal bridge programs may prove useful in the current debate.

The Silver Bridge. On Friday, December 15, 1967, at about five p.m., the so-called “Silver Bridge” spanning the Ohio River between Point Pleasant, West Virginia and Gallipolis, Ohio suddenly collapsed. 31 of the 37 vehicles then on the bridge fell, killing 46 people.

The bridge had been completed in 1928, making it 39 years old. It was called the “Silver Bridge” because of the color of the aluminum-based paint it used. It was a suspension bridge, but instead of the woven steel cables used on most other suspension bridges, it used chain-link cables held together with carbon steel “eyebars” (a rigid bar with a hole at each end, used to link the ends of two chains together). Unlike almost all other chain-link suspension bridges, instead of using many redundant eyebars to secure each joint, the Silver Bridge only used two eyebars made of new high-tensile-strength carbon steel for each joint.

Three full years later, the National Transportation Safety Board issued its final report on the incident, saying that one of the eyebars on the Ohio side of the bridge failed, forcing its companion eyebar to fail immediately and loosing a cascade effect that put the entire 1,368-foot bridge into the river in short order. The NTSB stated that the “fracture was caused by the development of a critical size flaw over the 40-year life of the structure as the result of the joint action of stress corrosion and corrosion fatigue.” Neither stress corrosion nor corrosion fatigue were understood when the Silver Bridge was built in 1928.

The Federal Response. Four days after the 1967 disaster, President Lyndon Johnson appointed a federal task force “to begin immediately an intensive study of the Ohio River bridge tragedy and to conduct a national survey of bridge safety.” The task force was chaired by Alan Boyd, the first Secretary of Transportation. The task force quickly split up into three committees: one to investigate the cause of the Silver Bridge collapse (with the NTSB taking the lead), one to plan a new bridge to replace the collapsed one, and the third (headed by FHWA) to address nationwide bridge safety issues.

The collapse of the Silver Bridge, with its high loss of life (it was the deadliest U.S. bridge collapse of the 20th century, including the Cypress Street Viaduct collapse during the Loma Prieta earthquake in 1989), did serve as a wake-up call to the nation regarding the poor state of bridge infrastructure. But in the 1967-1970 period, the United States was facing several other problems, including (in no particular order) inner-city riots, the Vietnam War, the War on Poverty, a series of high-profile political assassinations, and waves of domestic violence and unrest. Plus, there was a sexual revolution, and human beings walked on the moon.

The fact that Congress was able to stay focused on bridge needs during this period was primarily due to one man – Senator Jennings Randolph (D-WV), the chairman of the Senate Public Works Committee. Because most of the victims of the Silver Bridge collapse were Randolph’s constituents, and because the loss of the bridge placed significant handicaps on West Virginia’s interstate commerce with the other side of the Ohio River, Randolph held a series of hearings on bridge issues once Congress reconvened in 1968.

The March 1968 hearings revealed a lack of consistent standards and a gap between knowledge of Federal-aid and state/local bridges. Federal Highway Administrator Lowell Bridwell told the panel that:

…States are given the duty to maintain, or to cause to be maintained, any project constructed with Federal-aid financing under the provisions of the Federal-Aid Highway Acts so long as the project remains a part of a Federal-aid system…As a part of their regular duties, the Bureau of Public Roads engineers make periodic inspections to determine the condition of maintenance of completed Federal-aid improvements. In the event a project is found to be in an improper condition of maintenance, the State highway department is notified of such fact. If, within 90 days after receipt of this notice, the project has not been put in a proper condition of maintenance, approval of further projects of all types in that State is withheld until the project has been put into a proper condition of maintenance.

But these were cursory inspections. Bureau of Public Roads (BPR) Director Frank Turner added that during a typical BPR inspection,

…we will not be making an inspection of the kind that would reveal anything other than the general outward appearance and a visual kind of inspection rather than getting into the details of the kind that would be required to make an analysis which might have revealed deficiencies in the Point Pleasant Bridge.

Both Bridwell and Turner pointed out that since the Point Pleasant bridge was not on the Federal-aid system, there had been no federal inspection requirement at all. They also pointed out a lack of uniformity in inspection standards between DOT, states, railroads, and local bridge authorities.

A few months later, Congress enacted the 1968 highway bill, which contained a provision directing DOT to establish national bridge inspection standards for all bridges on the Federal-aid system and a national inventory of all Federal-aid bridges. DOT was also charged with training federal and state employees to carry out bridge inspections.

By January 1970, DOT was ready to report that:

There are some 563,000 bridges in the United States of which approximately 70 percent were built prior to 1935. The bridges on county and secondary rural roads account for about two-thirds of all the bridges, and of these, approximately nine out of ten were built prior to 1935. Many of these older bridges will require repair or replacement in the immediate future. Since the vast majority of these bridges are on non-Federal-aid systems the burden of any corrective action will fall almost entirely upon State and local governments. (1970 National Highway Needs Report).

The report concluded that there were about 24,000 “critically deficient” bridges on the Federal-aid system and that repair or replacement of the bridges would cost $6 billion. Another 64,900 critically deficient bridges were not on a Federal-aid system and would cost an additional $8.8 billion to repair or replace. The study estimated that another 900 new bridges crossing major waterways were also needed, costing $4 billion. Total bridge cost: $18.8 billion.

In other words, the “needs” were becoming clear. However, for the federal government to take a special role in repairing bridges for motorist safety would mean a fundamental shift in the federal attitude towards bridges, which had not changed substantially since the early days of the Republic.

Water vs. Land. The Constitution adopted in 1787 gave the federal government the power to regulate commerce between states. And at that time, that meant traveling by water. As far back as 1824, the U.S. Supreme Court ruled that this power included regulation of navigation, and a series of Court decisions in the 1850s and 1860s gave Congress sole authority over the “navigable waters” of the U.S. – including the ability to regulate, restrict or remove bridges over such waters – because low or collapsed bridges might block waterborne interstate commerce.

Even as railroads began to carry a larger share of commerce during the latter half of the nineteenth century, Congress’s attitude towards bridges remained hydro-centric, largely because railroads and their bridges were built by private interests (albeit sometimes with government subsidies). But no level of government directly participated in building railroad bridges. (Some mixed-use bridges were built by local toll authorities starting in the 1870s, like the Brooklyn Bridge, but they were usually authorized by special one-off Acts of Congress).

In 1899 and 1906, Congress passed laws standardizing bridge approvals, requiring the approval of Congress and the Corps of Engineers in advance of the construction of any bridge over a navigable waterway and allowing the Corps to force bridge owners to upgrade their bridges so as not to interfere with boats and barges.

As late as World War II, the Truman-Hobbs Act of 1940 and the General Bridge Act of 1946 stayed focused on waterborne commerce, allowing the federal government to pay for the share of improving a bridge “attributable to the necessities of navigation” and adding the Bureau of Public Roads to the list of permissions necessary to build a bridge over navigable water. The 1946 law also contained restrictions on interstate toll bridges and gave blanket authority for the Corps of Engineers to transfer all U.S.-owned bridges to States.

Between the WWII period and the reauthorization of the highway program in 1970, highways were about to overtake waterways as the second-biggest movers of domestic freight tonnage, and the progress in completing the Interstate system (combined with the postwar suburban housing boom) had pulled America fully into the age of the passenger automobile.

Domestic Freight Traffic

(By mode, billions of ton-miles)
1939 1969 Increase
Rail 370 774 +109%
Highway 53 404 +662%
Waterway 96 303 +216%
Pipeline 56 411 +634%
Air 0 3 + infinity
TOTAL 575 1,895 +230%

The First Program. In 1970, led by Randolph, Congress decided to give the federal government a role in keeping bridges safe for motorists and trucks (in addition to keeping them safe for boats and barges). Section 204 of the Federal-Aid Highway Act of 1970 created the Special Bridge Replacement Program (SBRP). DOT was to “inventory all bridges located on any of the Federal-aid systems over waterways and other topographical barriers,” classify the bridges, and assign each a priority for replacement. $250 million in contract authority was authorized over two years for the inventory and for DOT to make grants to states to replace Federal-aid bridges at the Secretary’s discretion, based on priority, at a federal cost share of 75 percent.

The law establishing the new program was taken almost word-for-word from a bill Randolph had introduced in December 1969 (S. 3242, 91st Congress). The House version of the highway bill had simply required that states spend between 5 percent and 10 percent of their regular apportionments on replacement or reconstruction of bridges (over waterways only).

Over the next eight years, Congress would extend the special bridge program several times and add additional money (another $585 million in contract authority over five years) while maintaining the original structure of the program – completely discretionary, for bridge replacement only, and only for Federal-aid bridges over waterways or other topographical barriers.

In the 1970 law and those immediate subsequent extensions, one other thing stayed constant – funding for the bridge program was always contained in title II of a highway bill, with NHTSA funding and other highway safety programs – not in title I of the bill, where the highway construction programs are funded. This along with other legislative history shows that the bridge program was intended to have safety as its focus, not commerce or convenience. Also, funding for the program was indefinite (“no-year contract authority”) so states did not risk losing the money for a project if it got delayed.

(For a table showing historic funding levels for the bridge program, see page 6 of the original PDF version of this article.)

The 1978 Program. By 1978, the first results of the National Bridge Inventory were in, and Congress decided to change the bridge program to reflect its results, greatly expanding the program’s scope and funding. The special discretionary bridge program became a formula-based Highway Bridge Replacement and Rehabilitation Program. Rehabilitation (repair) of bridges was made eligible for federal funding in addition to replacement. Many more bridges were made eligible: the existing inventory of Federal-aid bridges was amended to add

bridges over highways and railroads, and a second inventory was added for bridges that were not on the Federal-aid system at all (“off-system”). Between 15 and 35 percent of a state’s formula apportionment would have to be dedicated to these non-Federal-aid bridges.

Over $1 billion per year was granted at a federal share of 80 percent, and the money was not “no-year” so, like all other non-Interstate contract authority, it lapsed at the end of the third fiscal year after its issuance. $200 million per year was set aside at the discretion of the Secretary for bridge replacement or rehabilitation projects costing more than $10 million or which were double a state’s annual formula apportionment.

That formula was based on the results of the first National Bridge Inventory and gave each state a share of that year’s total formula money based on that state’s share of the cost to repair or replace all deficient Federal-aid bridges. However, Congress locked the formula in stone by writing the apportionment factors into a House Public Works and Transportation Committee print and also changed FHWA’s formula by giving each state a minimum 0.5 percent apportionment and a maximum 5 percent apportionment (which necessarily changed the apportionments of most states).

1978 To Now. The broad outlines of the program established in 1978 remain in place today. However, the periodic reauthorizations of the program have shifted it from its original focus (ensuring the safety of existing bridges) and made it more like the other major highway construction programs, which focus less on safety and more on capacity.

  • First, while the bridge authorizations in the 1970 through 1982 highway bills were under the safety title of the bill (title II), starting in 1987 they were moved to the construction title (title I). This parallels the movement of the other title 23 U.S.C. safety programs (rail crossing elimination and hazard mitigation) into title I. However, the grade crossing and hazard programs were, like the construction, based on demand-based or capacity-based formulas (like population, or lane-miles), not on specific dollar amounts needed for specific safety upgrades. The bridge program was special in that it was based purely on the cost to repair or replace each state’s deficient bridges, regardless of how many people used them or how much tax they paid. Grouping the safety-based bridge program in with all of the other, mostly capacity-oriented construction programs surely made Congress think of it as just another construction program.
  • Second, Congress in 1991 made bridge money fungible. The ISTEA law allowed a state to transfer up to 40 percent of each year’s bridge apportionment to the NHS or STP construction programs (and that amount was increased to 50 percent in 1998). This is a logical extension of Congress’s decision to put bridge money on the same “use it or lose it” four-year timetable as other title 23 apportionments. Some states take advantage of this flexibility more than others. Under the ISTEA and TEA21 laws, Pennsylvania chose to transfer almost $1.4 billion of its bridge money. California transferred $550 million, Massachusetts $407 million, and all other states and territories totaled $1.00 billion. (Ed. Note: Because Congress cares so much about its earmarks and Appalachian highways, it gives special, nontransferable, non-expiring funding to those projects, but bridge money expires and states are free to transfer half of it to the construction of new, non-bridge projects). (For a table showing which states transferred their bridge money to other programs, see page 8 of the original PDF version of this article.)
  • Third, Congress has gradually taken away all of DOT’s discretion in the bridge program. Under the 1978 law, the discretionary set-aside was about 20 percent of the total bridge program. In the 1980s, that dwindled to about 10 to 12 percent, and in the 1990s ISTEA cut it to under three percent. The 2005 SAFETEA-LU law eliminated the discretionary program entirely and instead earmarked $100 million of the program, much of it for bridges (like the one in Gravina, Alaska) that are entirely new construction and would not have been eligible for funding under a program dedicated to the safety of existing bridges.
  • Fourth, beginning with the 1982 law, the bridge program became part of the donor state versus done state fight over rate of return on fuel tax contributions. The bridge program was made subject to the 1982 and 1987 85 percent minimum allocations, the various ISTEA equity programs, the TEA21 90.5 percent minimum guarantee and the SAFETEA equity bonus. Before 2005, this made the bridge program somewhat unpopular with donor states since it disproportionately benefits donee states with lots of older bridges. And even today, under a dollar-based equity bonus formula that removes the perverse incentives of prior percentage-based equity programs, placing the bridge program in the formula fight contributes to the sense that the bridge program is just like all the other programs and that funding should be based on gas tax payments and political clout, not safety needs.
  • Finally, Congress has gradually expanded the focus away from the safety of Federal-aid bridges. In 1978, non-Federal-aid bridges were made eligible and given between 15 and 35 percent of annual apportionments. Over the years, Congress has added provisions for preservation of historic bridges, the replacement of destroyed ferryboats, a set-aside for timber bridges, and the earmarks mentioned earlier. Also, the SAFETEA-LU law lifted the 35 percent ceiling on funding for non-Federal-aid bridges, meaning that states are now free to spend all of their federal dollars on bridges that have nothing to do with the Federal-aid highway program. (In the 1970s, some prescient legislators worried that giving federal funds to states to fix non-Federal-aid bridges would create an incentive for states to shortchange bridges in their state funding plans).

We have obviously come a long way since 1967. Yet earlier this month a fatal bridge collapse, similar to the one of 40 years ago, reminded us that the bridge safety problem is still significant. The initial count taken in the aftermath of the Silver Bridge collapse showed almost sixteen percent of all U.S. bridges to be deficient, and current data shows the rate to be slightly above twelve percent.

What next? In the immediate aftermath of the I-35W bridge collapse, the Department of Transportation ordered immediate inspections of all 756 steel arch truss bridges in the United States (easy enough to do since there are relatively few of them). The DOT Inspector General was ordered to “conduct a rigorous assessment” of the bridge inspection program. And after a few days, states were warned to be careful when loading construction supplies on bridges, since over 100 tons of gravel had been loaded onto the I-35W bridge hours before the collapse as part of a repaving project. And Congress quickly enacted legislation (Public Law 110-56) authorizing a $250 million general fund appropriation to replace the bridge, waiving the statutory cap on FHWA emergency relief for a single incident, and giving $5 million in transit funding for displaced motorists to use. (Ed. Note: the $250 million authorization is redundant because a “such sums as necessary” authorization already exists for general fund ER appropriations.)

The author of that legislation was House Transportation and Infrastructure Chairman James Oberstar (D-MN) who (in the Jennings Randolph role as highway committee chairman from the state where the disaster took place) has also proposed new legislation dealing with the overall U.S. bridge safety spending backlog.

The centerpiece of his plan (which has not yet been fleshed out with legislative language) would be a new bridge repair and replacement program totally outside the existing highway program. The program would focus exclusively on bridges that are part of the National Highway System (a designated 162,000-mile system of which the entire Interstate system is a 47,000-mile subset). The NHS is only about 4 percent of total U.S. road-miles but it carries 45 percent of the vehicle traffic (and its bridges carry almost 70 percent of all traffic over bridges). Since NHS bridges are usually at least four lanes wide and tend to be longer than average, NHS bridges constitute half of the deck area of all U.S. bridges.

The latest DOT needs report indicated that there is presently about a $32 billion backlog of unmet needs on NHS bridges. Oberstar’s plan would create a new trust fund modeled on the Highway Trust Fund to pay for the bridge spending and would require a dedicated revenue source to raise the money (his plan did not make a specific revenue proposal but invited one to read between the lines and envision a gas tax increase in the neighborhood of five cents per gallon).

President Bush responded to Oberstar’s proposal by criticizing the level of earmarks in SAFETEA-LU and saying “I would strongly urge the Congress to examine how they set priorities. And if bridges are a priority, let’s make sure we set that priority first and foremost before we raise taxes.”

(Ed. Note: Bush’s categorical refusal to countenance a gas tax increase in 2003-2005 ensured that total highway spending, including bridge spending, would be much lower than his own DOT said was necessary. But once resources were constrained, the fact that SAFETEA-LU significantly cut back formula spending’s piece of the pie (including bridges) in order to pay for drastically increased earmarking — more than double TEA21’s earmarks, as a percentage of total highway spending — is indisputable.)

Oberstar responded by pointing out that his proposal would ban earmarking of the bridge money and would automatically shut down revenue transfers to the new trust fund if any earmarking took place.

(Ed. Notes: (a). This seems to be an implicit acknowledgement by Oberstar that the excessive earmarking in SAFETEA-LU did serious damage to the reputation of the highway program. (b). We can’t wait to see the legislative language on this.(c). Who watches the watchmen? Congress sets the rules, and its efforts to limit its own behavior are dependent on the goodwill of future Congresses. Case in point: in 1913 the House established a rule prohibiting any earmarking of any specific road projects. The rule stayed on the books, but the House ignored the rule more and more frequently until they finally repealed it in January 1999 after the passage of TEA21 made it look silly.)

Oberstar’s proposal contains worthwhile provisions that would surely pass on their own in short order (an updating of bridge inspection standards and the bridge inventory, better calculations and postings of weight limits, and better computerized access to bridge data). But some of the features point out shortcomings in the present bridge program that might also be remedied, including:

  1. Transferability. Presumably, states won’t be allowed to transfer funds under the new bridge program to non-bridge activities. Why should states be allowed to transfer their 2008 and 2009 SAFETEA-LU bridge apportionments to non-bridge needs?
  2. Federal focus. Oberstar’s plan focuses solely on NHS bridges, implicitly acknowledging that the existing bridge program no longer focuses enough attention on federal bridges. Congress might demand immediate data on how much federal bridge money each state now spends on non-federal bridges and, if necessary, bring back a cap on non-system bridge spending.
  3. Jobs or lives? New bridge spending may be necessary because existing dollars are buying fewer bridge repairs. The price of steel used in bridge projects has almost doubled in the last three years. Someone should commission a reputable study of how much of that increase is due to global increase in demand, and how much is due to Buy American requirements for bridge project steel.

Oberstar will hold a hearing on bridge needs when Congress returns on September 5 and says he will move his legislation as soon as possible (though the heart of the plan — some sort of tax increase — is out of his hands and requires Ways and Means Committee approval). Republican opposition is likely.

Perhaps all of the needs-based safety programs should be given their own safety trust fund outside the HTF. As long as present law highway program resources are constrained, highway and bridge safety will always have to fight congestion relief for its slice of the pie. And by the time the next highway bill is finalized sometime in 2010, the dozen or so motorists who tragically lost their lives on August 1 will have been gone for over two years, and in the interim, over 200 million American drivers will have spent billions of frustrating hours sitting in gridlocked traffic. They will be focused on their daily grind and demand that Congress make congestion relief its top priority. (And Congress usually gives a vocal, determined majority what it wants).

In the end, we may not be that far removed from this exchange between Sen. Hiram Fong (R-HI) and Frank Masters of the Consulting Engineers Council in March 1968:

Senator FONG. So, the collapse of bridges is few and far between; is that right?

Mr. MASTERS. Yes; few and far between.

I think it is important that we don’t panic at this time and frighten people around the United States into thinking that all the bridges that they are crossing may take their lives. That just is not so.

Senator FONG. Just because of this one major disaster that a bridge brought to our attention.

Mr. MASTERS. Yes.

By the same token, on a certain few structures I can honestly see where it is conceivable we could have a collapse this morning.


The above article was originally published on August 15, 2007. For a follow-up overview of how federal bridge policy has changed in the decade of 2007-2017, see this article. And also see this Timeline of Key Moments in Federal Highway Bridge Policy, 1789-2017.

Search Eno Transportation Weekly

Latest Issues

Happening on the Hill