I posted a while back Transportation Benefits Too Little.

Matt Logan sends me to another paper of interest: Understanding the Contribution of Highway Investment to National Economic Growth: Comments on Mamuneas’s Study by Randall Eberts
Introduction:

Since the early 1990s, various researchers have attempted to answer the question: “What is the rate of return of highway investments to the economy?” This follows from the simple proposition that highways provide the arteries for the flow of raw materials, and intermediate goods to producers and access to employment and markets for consumers. Without a highway system producers would not be able to obtain economies of scale to obtain the most efficient operation. Many economists agree that the rate of return to highways during the period 1950 – 1970 exceeded the rate of return available to alternative investments.
Recent studies have indicated that the rate of return to highways has decreased following the initial boost provided by national connectivity. There is strong debate about the actual level of returns in the late 90’s to the present day. The level becomes critical because some studies show the rate near or below the rate of return available from alternative investments. The Federal Highway Administration (FHWA) of the U.S. Department of Transportation seeks to continue to provide the best information available to policy makers to understand the benefits and trade-off involved with alternative uses of the scarce federal resources.
To that end, the Theofanis P. Mamuneas recently completed an update of his highway benefit research for the FHWA. During the 1990s Prof. Mamuneas conducted several studies with Prof. Nadiri based on a general equilibrium approach to modeling the returns to highway investments accounting for both consumer and producer benefits. This paper endeavors to place the most recent study authored by Prof. Mamuneas in context with previous results and results from alternative methodologies and data sets. Each study examining the returns to highways employs complex econometrical and data intensive modeling efforts. The studies also yield different and sometimes conflicting results, even though all employ generally accepted economic techniques. Policy makers need a technically skilled and unbiased review of the field in order to use the results of this literature for policy decisions.
The purpose of this paper is to review and summarize current literature relevant to contextualizing the recent update by Mamuneas to the previous Nadiri-Mamuneas estimates of the returns to highway investments. To provide this context, the paper will first provide an overview of the conceptual relationship between highways and output. The next section describes the highway capital stock estimated by Fraumeni and used by Mamuneas. Next, the paper describes the study conducted by Mamuneas and carefully analyzes the results for consistency within the modeling framework and in context with other studies. The paper then briefly summarizes the broad range of estimates from the literature to offer additional context. Finally, the paper offers an overview of the issues that still need to be addressed to provide more definitive estimates.

Conclusions say:

… While there may not be a consensus in the literature on the absolute value of estimates of the rate of return to highways, and perhaps that may not be possible given the different methodologies and levels of aggregation used in the literature, it appears that there is a convergence of thought that the U.S. highway system is maturing and that the system is no longer underbuilt. However, that does not mean that funding for the highway system should be reduced. As a mature system, highways require maintenance, upgrading and replacement to provide the same level of services as before. And with a large infrastructure, that requires considerable investment to maintain highways in the condition that they were originally built. In addition, as the economy continues to expand over the long-run, the proportion of highway stock to other factors of production will need to be maintained so that if the economy is at an equilibrium state in which the rates of return are roughly equal, that relationship can be maintained into the future.

I agree with the first part of the conclusions. I am less convinced the economy is ever in equilibrium nor that the share of highway stock in factor of production needs to remain constant. The whole point of dematerialization is to reduce transport and inventory costs, so I would hope transport costs might go down over time as more things can be done virtually.