Statement of
Scott Bernstein
President, Center for Neighborhood Technology
On behalf of the Surface Transportation Policy Project

"Planning As If People And Places Matter:
Surface Transportation Research Needs and Performance for The Next Century"

Before the Senate Committee on Environment and Public Works Friday, March 15, 2002

Mr. Chairman and Members of the Committee, thank you for the invitation to join you today to discuss the surface transportation research programs of the Transportation Equity Act for the 21st Century. I am Scott Bernstein, President of the Center for Neighborhood Technology and a co-founder of the Surface Transportation Policy Project or STPP.

STPP is an eleven year old coalition of over 800 groups whose mission it is to ensure that transportation policy and investments help support a strong economy, promote energy conservation, protect environmental quality and help build more equitable and livable communities. We are vitally interested in transportation research and development, because we believe that the research investments of today lay the foundation for the transportation system that our grandchildren will enjoy throughout this century.

Our comments today are based on our experience since the passage of ISTEA and throughout most of TEA-21. Our members have had the privilege of participating in many committees of the Transportation Research Board, concerning topics ranging from global warming to pavement condition and technology to the role of transportation investment in economic productivity; in the advisory committees to the Federal Highway Administration and the Federal Transit Administration; and as members of the White House Advisory Commissions on Sustainable Development and Greenhouse Gas Emissions. One of the benefits of a well-rounded membership is that we have also enjoyed appointments to analogous bodies in fields ranging from housing policy to welfare reform to telecommunications to community economic development. That range of experiences gives us a basis for comparison for judging the performance of the federal government in meeting the goals of Title V of TEA-21 on Research, and for suggesting areas for improvement as the committee moves forward on reauthorization.

The stated purposes of ISTEA and TEA-21 should guide the federal research, information and development agenda. Those purposes - intermodalism, economic efficiency, environmental quality and equity - should guide the expenditure of the billions of dollars authorized by Title V on research as part of the Department's strategic plan.

But as we've thought about the challenge you laid out in your invitation today, we are troubled by the extent to which the federal information system on which this kind of analysis depends is lacking. There are five areas we want to call to your attention today. They include: transparency; data quality and new measurements needed; the need to distinguish between regional and long distance travel, particularly in light of the events of September 11; the value of systems preservation and maintenance; and lack of attention to end user requirements such as understanding travel expenditures.

Transparency

Americans spend 1.5 trillion dollars annually for all forms of transportation, capital and operating, governmental and private expenditures. This is very large, but much smaller, for example, than the federally regulated depository financial services industry. That industry, under the direction of the Home Mortgage Disclosure Act of 1975, is required to report, on demand, the origins and destination of all deposits on a small geographic basis - a census tract. Every day in America, hundreds of thousands of inquiries are logged on the Internet, or taken in person at bank branches, from persons who want assurance that a fair share of their deposits are returning to their communities. A companion piece of legislation, the Community Reinvestment Act, requires affirmative efforts by these institutions to reinvest in their primary service territories. These two acts have been responsible for several trillion dollars of reinvestment in our nation's communities. The Federal Reserve Bank found that lenders who did the best job of complying with these two statutes were significantly more profitable than those who met the minimum compliance requirements.

The experience is somewhat different for the public's transportation expenditures. The Intermodal Surface Transportation Efficiency Act of 1991 painted a vision of research and information provision that was not just modally based, but rather would be presented in a comprehensive and easy to understand format. In reality, USDOT provides extremely limited access to data on expenditures of federal funds. Even though the Financial Management Information System has been in place for more than a decade, information by recipient and by program is limited and often several years behind. In some cases, the way the data is collected fails to respond to legislative mandates. For example, states report expenditures back to FHWA using hundreds of codes for activities. These codes do not necessarily correspond to programs and required set-asides, such as the 10 percent set aside from the Surface Transportation Program for safety and for enhancements. A request for enhancements expenditure data in 2000 resulted in a coding chart with more than 70 codes for things ranging from roadside landscaping to bicycle trails. This makes tracking performance virtually impossible. The data is also effectively inaccessible to citizens, local officials, and even Congressional staff. The FMIS is not available on the web, or even in paper form for those who would like to simply look up how their communities are spending federal transportation dollars. Every two years, the public interest non-profit group, the Surface Transportation Policy Project, spends months decoding the FMIS data for the general public: this is a job the public sector should be doing. We are nowhere near the ability to track revenues versus expenditures for zip code or census tract level geographies, and so the ability to understand these investments at larger aggregations, such as in corridors, metropolitan statistical areas or states, is suspect.

Bad Data Supports Bad Science and Bad Decisions

Despite years of characterizing national transportation trends and use patterns, we know very little about multi-modal trips, the full range of users, the role of intercity travel in overall transportation demand, or the travel behavior of persons in institutions (to name a few of the gaps). This information is crucial to adequate policy making.

Our knowledge gap is largely due to the structure of the two national surveys conducted by U.S. DOT (and previously by Census in some years): the National Personal Transportation Survey (NPTS), conducted under the auspices of FHWA, and the American Travel Survey (for trips 100 or miles more that are not a commute trip). These surveys are infrequent and do not include all groups who travel, even though they are widely quoted and used as representative of all personal travel.

The National Personal Transportation Survey (NPTS) was conducted in 1969, 1977, 1983, 1990, 1995 and 2000 (2000 results not yet available). It provides substantial data on the travel behavior of some households, but there are significant segments of the population that are undercounted or missing. Here are some of the gaps. As a household survey, NPTS is not intended to include on-base military personnel or residents of group quarters - nursing homes, college dormitories, long-term medical institutions or prisons. (See Summary of Travel Trends, 1995 Nationwide Personal Transportation Survey, December 1999, USDOT/FHWA, page 2). According to the 1990 census, these groups account for over six million people. Due to the survey methodology, which was changed in 1990 from a face-to-face survey to a computer-assisted telephone interview, immediately approximately 6 percent of all households (i.e. those without telephone service) are left out, disproportionately affecting minority households. Critically, the survey is administered only in English - thus missing many in the growing immigrant population. In 1993, over 13 percent of the population reported living in a non-English speaking household, and in our larger cities that percentage is as much as 25 percent. The 2000 survey finally had Spanish language surveyors and instruments. The net result of these shortcomings is that low-income populations, Native Americans, many immigrants and persons in institutions are not represented or underrepresented, even though I would argue that they do travel.

It is unclear why these gaps exist, although the lack of funds is often cited. It is important to fill the gaps, and we would argue that additional funds could be extremely well spent. We would also support any such surveys being conducted by USDOT's statistical arm, the Bureau of Transportation Statistics (BTS). The current practice of housing the NPTS in the FHWA, which is focused primarily on automobile travel, leads to a narrow focus on a single mode. This is not reflective of true travel patterns.

In the meantime, however, it is important to be clear that the survey does not capture a representative sample of all people in the country, but is in fact a survey focused on the middle class household. And yet the sponsors of the survey promote the use of the survey data for: "travel demand forecasting, identifying modal choices and use factors, and examining facility accessibility and use issues for elderly, low-income, minorities, and recent immigrants . . ." (this statement is from a recent FHWA presentation to TRB) and as comparisons to validate local travel data. In addition to money to conduct the survey properly, we would also argue for proper use of the data and better labeling so that researchers do not use it to characterize "trip making of low-income population" or "auto-less households" or other recent research topics.

The fact that the NTPS is conducted only every five to seven years makes it far less useful for policy than surveys by the Department of Health and Human Services that are far more frequent, such as the Annual Drug Use survey that is a household-based survey covering more than 70,000 households. To make matters worse, methodological changes in each NTPS survey make it almost impossible to rely on this data for trend analysis.

The data gaps left by the NPTS and the American Travel Survey are many. One very important finding of the NTPS that we'd like to point out is that less than one trip out of five taken by metropolitan households is for the journey to work. The decennial census is the largest coverage survey on travel patterns, but it only asks questions concerning the work commute. It does not survey respondents on non-work travel purposes such as shopping, getting kids to school, attending worship, recreation or visiting friends. This is a missed opportunity to gather a wealth of travel data.

In one case, another government agency is providing useful travel data, but even this has limitations. In its attempt to implement the Clean Air Act, the USEPA pays for the recording of odometer readings for cars undergoing enhanced inspection and maintenance under Clean Air Act Amendments for non-attainment areas. This data has been very useful for establishing mileage-based programs in mortgage lending, such as the Location Efficient Mortgagesm, and recently introduced pay-as-you-go insurance initiatives. Vehicle Miles Traveled (VMT) is also estimated (not really measured) by the Office of Highway Research and used, for example, in the annual congestion index reports of the Texas Transportation Institute. But this data is still too focused on a single mode, and fails to provide answers to the more fundamental question: what are the opportunities for trip substitution by mode in particular areas? Vehicle miles are the wrong measure for making this assessment; trip making is the correct one. We support the use of expanded resources to address this need, and to identify creative approaches to data collection at the local level that can provide a truer picture of the variation in regional travel demand by purpose, by traveler and by place.

There are other omissions from national data that bear attention. Mass transit transfer data is almost non-existent; we have no idea of the extent to which travelers use a combination of automobiles, transit, walking, or even bicycling to complete a single trip. Data for "other providers" is non-existent (for example, from airport, hotel, auto rental and shuttles). New York pedestrian trips during the day probably dwarf both transit and auto trips, but who would know? And from what data? Church provided para-transit and employer supported para-transit are probably both large numbers, but who would know. How about transportation in the new "active life style senior communities"? For those of us who haven't learned how to stop growing older, we might want to know.

Intercity Travel Information and September 11

We lack good data on intercity travel patterns, data that is now essential if we are to adjust our transportation system in response to September 11th. The American Travel Survey (ATS) covers trips of 100 miles in length or more. The results of the 1995 survey (the only one available so far) shows that 81 percent of such distance trips are taken by passenger vehicle; 16 percent by airplane; two percent by motor coach, and 0.5 percent by rail. Yet this survey provides precious little more information about long distance travel. It does not indicate what portion of trips are of shorter distances that can plausibly be made via a variety of modes, including rail, air, or automobile. We are also clueless as to how much measured urban travel is through-travel from other locations, and therefore there is a serious flaw in the models used to estimate travel demand and the results applied to State and MPO budget allocations.

A look at some private sources show us just how valuable additional data could be. For example, it turns out that most rail trips and 56 percent of airline trips are less than 500 miles in length. I know this because we paid to purchase the commercial data-base of the Official Airline Guide (apparently use of this data is a tradition, all major studies of trip length from the 1960's on, such as those performed by Brookings and by the National Bureau of Economic Research, needed to rely on this source). This information is simply not available in this format or quality from the Federal Aviation Administration or from the Bureau of Transportation Statistics. Combine this with research by the Stanford Research Institute, another private institution, and by the property casualty insurance industry, and we discover that risks from both accidents and terrorism are lower in surface travel modes than in aviation, and in public transportation modes than private ones. We would include true intermodal linkages, such as air/rail combinations now available at 22 European airports, in the lower risk categories. These findings indicate the potential for making our travel system far more secure by shifting from air to surface modes. Yet the Congress, it seems, has been asked daily to support constantly increasing multi-billion dollar expenditures to protect public safety and security in inherently risky airline travel, without being asked to develop strategies for risk reduction through travel demand strategies. September 11 offered a painful set of reasons to catch up with the requirements of real time risk identification, and to use such information to help set new policy going forward. Is it too much to ask that in the face of a declared national emergency, new forms of knowledge be assembled on which to base transportation security policy?

The Value of Fixing It First

At their core, both ISTEA and TEA-21 stress systems maintenance and preservation. The Northeastern Illinois Planning Commission, with support from FTA and using carefully constructed data sets, showed that in metropolitan Chicago, from 1990 to 1998, the percentage of public transportation funds programmed for systems maintenance and preservation increased from 55 to 80 percent of total capital funds, an increase of 45 percent in total capital devoted to the Acts' "fix it first" philosophy.

Similarly, a story can be told of the need to spend more time accounting for the quality of our infrastructure investments. An economist from Northeastern University in Boston noticed that based on FHWA data, the capital stock accounts of the national income and product accounts kept by the Bureau of Economic Analysis, assumed a useful life of 60 years for all roads. She didn't believe this to be plausible, and through determined inquiry succeeded in getting FHWA to survey state DOT's. They conducted an analysis of the useful lives of roadbeds, pavements, surfaces and structures. Their weighted average finding was 45 years per road measured (still a one-size-fits-all but closer to what is intuitively correct). But a road that lasts 15 years should be worth less, so on the last day of 1999, an adjusting entry was made in the National Income and Product Accounts - and the net worth of the United States dropped by $140 billion. The study also did us the favor of estimating the value of extending the life of the nation's highways. A reasonable estimate is that $140 billion is what it would cost to extend the life of the highways another 15 years, or that it should be worth at least $9.3 billion more in enhanced maintenance of the nation's highways, per annum.

The Government Accounting Standards Board (GASB) in 1999 issued Statement 34, on accounting for capital assets of state and local governments. It takes a systems maintenance and preservation point of view, requiring the nation's 88,000 units of state and local government to disclose their budgets for maintaining current investments. This was developed over a period of 15 years because it was noticed that with no standards for such reporting no two financial statements are comparable; similarly, the federal government has no such standard either. Starting this year, state and local governments will spend an estimated one-time $2.9 billion over the next three years to do their capital accounting consistently. Recent events in the energy futures industry demonstrate the value of this exercise.

The federal government has an inherent interest in helping state and local government get this one right, and can do so in several ways. First, by sponsoring quality technical assistance; second, by issuing consistent guidance for life cycle costing of infrastructure investments; and third, by getting its own investments into compliance with an accounting standard based on the very same kind of preservation and maintenance philosophy upon which ISTEA and TEA-21 are based. It can also take steps to ensure that major research and development initiatives, such as the so called F-SHRP (pronounced F-sharp) pavement research program, helps public agencies meet these kinds of commonsense requirements. If such programs intended to result in high performance initiatives are not well aligned with national policy goals, particularly ones adopted by the private investment marketplace, such as the GASB 34 standards, then why should they continue to enjoy funding?

Similar questions could be asked about pedestrian facilities such as sidewalks, intersections, trails, and recreational facilities including parks and bicycle paths. The Surgeon General of the United States and the Centers for Disease Control and Prevention find that Americans are suffering an obesity epidemic partially as a function of limited physical activity, which is related to a lack of safe places to walk or bicycle. The Robert Wood Johnson Foundation responded by reprogramming over $50 million dollars to promote physical activity in American communities. I=m unaware of any similar response to date by any agency of the United States Department of Transportation. It's bad enough that there are different standards for the Conditions and Performance Reports for highways, transit and aviation (see for example, OMB Analytical Perspectives, section on capital budgets) but there is no standard or requirements for the infrastructure needed for the most basic form of transportation of all: walking.

Transportation and Household Economics

The Bureau of Labor Statistics of the Department of Labor is the only agency that conducts consumer expenditure surveys. Using this data, STPP and the Center for Neighborhood Technology produced Driven to Spend, which developed a way of comparing household expenditures between metropolitan areas. Transportation expense as a portion of household budgets ranged from 12 percent in New York in 1998 to 23 percent in Atlanta. These variations are accounted for by the degree to which households are dependent on excess automobile ownership. The study shows that households who choose to own too many cars too soon in life are making a poor choice, compared to the opportunity to invest in an appreciating asset such as homeownership. Cars depreciate from day one at approximately 10 percent of their value per annum, while homes appreciate at a net rate of 3.3 percent per annum (better than my savings account).

The study used a method of mapping expenses within metropolitan areas, the same one that the Federal National Mortgage Association chose to use to underwrite Location Efficient Mortgages^sm against. These mortgages count the savings from living in transit and amenity rich communities as income, allowing first time homebuyers to qualify for homeownership sooner. Much of the data used in developing this product was federally collected. But it cost almost $1 million to collect and format it in a form that could be used by the marketplace for these purposes. Data collection in information technology (IT) systems costs 100 times the value of hardware and software and people to run these tools, partly because of the lack of attention to what users really need. This particular effort should have cost less than $100,000; the difference was an unreasonable, if funded, information mandate. Studies suggest that excess car ownership is one of the principal barriers to homeownership, and a principal contributor to the low U.S. household savings rate. By contrast, counting these savings as income to help families qualify for homeownership is an excellent example of how federal information and research could produce the tangible benefit of helping moderate income Americans - those in the second and third income quintiles - achieve the American dream.

No less a transportation expert than Will Rogers noted during the Depression, "we will be the first nation in the history of the world who drove to the poorhouse in an automobile." Credit for car purchase is equal in volume to credit for credit card purchases, and may be as responsible for delinquencies, foreclosures and bankruptcies. It would seem that the Department of Transportation has an interest in tracking and disclosing this kind of information to better serve consumers and other end users.

Recommendations

The question before the Committee today boils down to this. Does our transportation data help answer commonsense questions necessary to achieve the goals committed to in the Act? For example, how do people get around in their communities? How many people have access to transit service that provides enough convenience to compete with auto trips? Can better land uses in places without good mass transit serve similar purposes through convenience and accessibility? What do we know about how much people bike or walk for transportation purposes? How do these demand factors vary by income? We don=t know the answers to these questions, because our data collection, our data formulation, our level of cooperation between agencies and between the federal government, other levels of government, and the private sector, is so poor. We can, however, tell you exactly how much was spent on building "highways over waterway of timber and slab construction" (see STPP, Changing Direction, page 31).

Our information system and research apparatus fails to provide the level of service to the public at large provided by HUD and banking regulators under the Home Mortgage Disclosure Act and the Community Reinvestment Act, or by USEPA under the Toxics Release Inventory, or by financial services industries and credit counselors to millions of Americans every day. We call for citizen participation in the planning process to be "early and continuous;" can we afford to settle for a lower standard in the provision of basic user oriented information?

We recommend the following 3 actions:

1. Information and research should be less modally and technology-oriented. This was one of the promises of the Bureau of Transportation Statistics. But whereas modal and technology-based research received $450 million per year, BTS got by on just $31 million.

2. Information and research should be approached as a comprehensive effort that involves both the federal government and leading research institutions. This would follow the example set by best practices in other fields. Private foundations, in anticipation of welfare reform reauthorization, spent tens of millions of dollars jointly with the federal government and leading research institutions, such as Brookings, the Center on Budget and Policy Priorities, and the Urban Institute over the past two years to be able to evaluate results and provide the public and Congress with information needed for informed decisions. A similar process could be crafted in anticipation of surface transportation reauthorization, which will occur simultaneously with the reauthorization of the aviation bill, AIR-21, and the debate over the future of passenger rail service. Astronomers plan years in advance for a lining up of the planets; this may be the equivalent of that kind of epochal event.

3. Data collection should be structured to answer critical questions faced by policy makers and citizens to ensure that the transportation system meets community needs. Currently, much federal transportation data collection is designed to meet either bureaucratic requirements, or narrow highway engineering specifications. This data is ill-suited for use by planners, citizens, or lawmakers, yet it is all that is available. A fundamental change in mindset is needed, so that transportation data collection and research are designed from the beginning to answer some of the fundamental questions outlined in this testimony.

Walter Lippman once noted, "democracy means paying attention." Every week in America, due to ISTEA and TEA-21, there are thousands of meetings where people, in spite of the lack of information, are muddling their way through local transportation plans. We can do better, and owe it to our selves and our future generations to do so.

On behalf of the many affiliated organizations and members of the Surface Transportation Policy Project, many thanks for the opportunity to present our views on the successes and failures of transportation research to the Committee.