In most states, if you want to add lanes to a new freeway, or build a new bypass or highway exit, and the project is eligible for federal-aid matching funding, you (the state) pay 20 cents of every dollar of total project cost.

But not if you are one of the 14 states subject to the “sliding scale” calculation in 23 U.S.C. § 120 . If you are one of those states, in which the federal government owns a significant percentage of the land, you pay as little as 5.1 cents on the dollar for your federal-aid highway projects (lucky Nevada).

The decision to give these 14 states preferential treatment in the funding of their federal-aid highways  was made 102 years ago, and the rationale is outdated. Yet the policy remains in place.

Why the sliding scale was created

While the federal-aid highway program dates back to 1916, it was the Highway Act of 1921 that created the program more or less as we know it today.

The way in which states financed their road programs prior to 1921 was completely different than today. As Brian D. Taylor, Eric A. Morris, and Jeffrey R. Brown point out in their excellent new book The Drive for Dollars: How Fiscal Politics Shaped Urban Freeways and Transformed American Cities, during the Good Roads Movement era, “the use of property taxes steadily increased, comprising 34 percent of state highway revenue by 1904 and ultimately becoming the dominant funding mechanism in the early decades of the twentieth century.” (p. 80)

Road financing based on property taxes disadvantaged states where a large amount of the real estate was not subject to state or local property tax. Accordingly, the Highway Act of 1921 contained this proviso in section 11: “…in the case of any State containing unappropriated public lands exceeding 5 per centum of the total area of all lands in the State, the share of the United States payable under this Act on account of such projects shall not exceed 50 per centum of the total estimated cost thereof plus a percentage of such estimated cost equal to one-half of the percentage which the area of the unappropriated public lands in such State bears to the total area of such State.”

The House Roads Committee’s report on the 1921 highway bill justified the sliding-scale proviso this way: “More than 90 percent of all the lands in Nevada are unappropriated public lands. The taxes raised from the other 9 or 10 percent of the lands of Nevada will not produce sufficient funds to make any material progress toward the construction of roads in that State. This situation, although not to such a marked degree, is found in 10 other Western States that have large areas of unappropriated public lands.” (H. Rept. 162, 67^th Congress, p. 6.) The sliding scale provision clearly referred to states paying for their road programs by collecting property taxes.

Why that reasoning is no longer valid

But a funny thing happened in the years following passage of the Highway Act of 1921 –  states pivoted to replacing their reliance on property taxes for raising highway revenues with the gasoline tax (first pioneered in Oregon in 1919, and adopted by all 48 states by 1929). According to Table SF-201 in Highway Statistics Summary to 1945, while in 1921 states raised the bulk of their revenues for highways from property taxes (at a ratio of $21 from property taxes – either dedicated, or through general fund appropriation – to every $1 raised by gasoline taxes), by 1931, the ratio had flipped – with states raising nearly $11 in gas taxes for every $1 raised for roads by property taxes.

And, as The Drive for Dollars points out, property values were part of the bubble that burst in 1929 – the Great Depression hammered property values, and thus property tax receipts, for decades. (The book points out that it was the loss of city and county property tax funding for local roads, then not eligible for federal funding, that precipitated the expanded federal and state roles in highway building and led to reduced local input on the location and design of urban freeways).

Adoption of the gasoline tax, on top of existing auto registration fees, cemented the “user-pay, user-benefit” model for state (and later, federal) highway funding, rendering the issue of untaxed lands moot. A state raises the same gas tax revenue whether driving occurs across public lands or whether it occurs on federally-owned right-of-ways. In addition, driving across state lines was a relative rarity in 1921, and long-haul trucking did not exist. Today, states receive the financial benefit of user taxes levied on people who are not even state residents – just driving through.

(There is a separate issue regarding taxation on Indian lands, which is complicated and deserves further study. Some kind of sliding-scale treatment might be consistent with the user-pay paradigm for states with a high percentage of non-taxable Indian lands and/or population. But according to the map, the states with the largest percentages of their land owned by BIA for Indian tribes are Arizona (#1) and South Dakota (#2), yet on the sliding scale rates, Arizona is #2 and South Dakota is #12, while Nevada, with a far smaller share of its land held by Indian tribes than either AZ or SD, gets the most benefit from the sliding scale.)

Given that property taxes have not been a significant state source of highway funding for decades, is the preferential treatment in federal-aid highway funding afforded to the 14 based on the sliding scale actually unfair?

Ranking states by actual ability to raise matching funds

First, let’s look at the five states that currently have the most difficulty raising money to support their road system, in terms of total state highway user tax receipts and tolls (FWHA Table HDF) divided by total state road lane-miles (FHWA Table HM-60).

Under the sliding scale, South Dakota and Montana are privileged while North Dakota,Kansas, and Nebraska, are not.

Then look at the five states that raise the most money from highway users, per lane-mile (excluding Hawaii, because Alaska, Hawaii, and the District of Columbia are outliers in many highway funding calculations).

California is somehow on the sliding scale and only pays 16.4 cents per dollar of a federal-aid project, but still manages to raise 27 times as much money for its road program, per lane-mile, as Kansas, which has to pay 20 cents on the dollar because it is not on the sliding scale.

But wait, you say. Since the sliding scale only applies to federal-aid projects, we should only look at lane-miles on the federal-aid highway system, not the much larger system of all roads. All right. Applying federal-aid lane-miles from FHWA Table HM-48:

Nevada occupies the highest spot on the sliding scale, paying just 5.1 cents on the dollar for its federal-aid highway projects, but still manages to raise over ten times as much highway user tax and toll revenue per federal-aid lane-mile than North Dakota, which has to pay 20 cents on the dollar for its federal-aid projects.

For those looking for a statistical correlation, the precise federal lands formula used under 23 U.S.C. § 120 includes Indian lands but, in its most common variant, excludes national parks, and that calculation is not available for all 50 states, only for the 14 sliding scale states. However, a recent report from the Congressional Research Service ranks all 50 states on their percentages of federally-owned land.

I ran statistical correlations both ways –­ first, by ordering all 48 contiguous states based on its CRS report ranking, and then by ordering all 48 contiguous states using the the percentages given in FHWA Notice 4540.12 for the 14 sliding-scale states and the ordering based on the CRS report for the other 34 states(Again excluding AK-DC-HI as outliers.)

If the logic behind the sliding scale is correct, and states with lots of federally-owned or Indian lands have a harder time raising revenue for highway projects, then there should be a strong negative correlation between higher percentages of federally-owned and high dollar amounts of tax revenue raised per road lane-mile.

There was no strong negative correlation, only a very weak negative correlation.

Conclusion

The sliding scale applied to the federal share of highway projects in states with vast federal land holdings is an outdated relic of the age when states used property taxes, instead of gas taxes, to pay for roads. And its continued use creates fundamental unfairness, where some states get a financial discount for their road projects irrespective of their ability to raise state revenues to match funding for federal-aid highways.

One thing the Federal Highway Administration could do on its own is update its numbers. The rates paid by sliding scale states are still based on the amount of federally-owned and Indian lands in that state as of FHWA Notice 4540.12, dated March 17, 1992. There have been some significant land transfers in the last 31 years. That same CRS report notes that “since 1990, the total federal land ownership has declined by 31.5 million acres… BLM ownership decreased by 27.6 million acres (10.2%), in large part due to the disposal of BLM land, under law, to the State of Alaska, Alaska Natives, and Alaska Native Corporations. DOD land ownership also declined, by 11.7 million acres (56.8%)… In contrast, the NPS, FWS, and FS expanded their acreage during the period, with the NPS having the largest increase in both acreage and percentage growth─3.8 million acres (5.0%).”

But the sliding scale still continues as it was in 1992, as if none of those changes had taken place.

As far as reforming the law itself, the Indian land question should be considered separately (as noted above), but there is little justification for continuing to give privilege to states with high shares of other federal lands. Significant changes to the law, however, are unlikely. In practice, legislators from the 14 sliding-scale states would oppose the removal of the preferential treatment afforded to them in 23 U.S.C. § 120, and they may be willing to “spend” political capital to garner enough support from additional states to block reform (especially in the Senate, where it only takes 41 votes to kill a bill, and the sliding scale starts out with 28 Senators presumably in support).