Articles



Economic Productivity, Transportation, & Jobs

U.S. Department of Transportation (USDOT) guidelines for economic evaluation of transportation projects, both written and presented by USDOT in a series of webinars, have increasingly focused on the dos and don’ts of incorporating wider economic benefits from transportation investments in benefit cost analysis (BCA). This is particularly, but not exclusively, reflected in the Transportation Investment Generating Economic Recovery (TIGER) guidelines, which emphasize the importance of transportation’s contribution to “national economic competitiveness.” Implicit in this is the idea that transportation projects are most valuable when they increase national economic output and gross domestic product (GDP). As economists hold that economic gains derive mainly from productivity growth, it follows that transportation projects that increase economic productivity are the ones that are most likely to increase GDP. Ultimately, transportation project sponsors, while they are interested in demonstrating productivity and output gains, are much more interested in how, and to what extent, such gains translate into jobs.

Definitions of Economic Productivity

What is productivity? Simply put, productivity is a measure input. Output relates only to market valued goods and services; for better or worse, charitable work and unpaid work at home do not count. While many types of inputs go into production, the basic categories used by economists are labor and capital. Labor is easy to grasp hours worked on the job. To standardize the measurement of labor as a factor input, the U.S. economic agencies—Bureau of Labor Statistics and Bureau of Economic Analysis—adjust labor hours by workforce factors including age, education, and gender.

Capital or capital services are defined as the value of physical assets (for example, see U.S. Bureau of Labor Statistics, News Release, Preliminary Multifactor Productivity Trends—2010) It includes computers, software, communications and other information processing equipment, other fixed business equipment, structures, inventories, business related rental residences, and land. Within these categories, investment, depreciation, and capital income are estimated to arrive at the stock value of capital services. Years ago, most of this would be industrial machinery and structures; now, with a service economy, computers, software, and IT equipment predominate.

How is productivity measured? It is measured basically in two ways: individual factor productivity and multifactor productivity. Individual factor productivity—for example, labor productivity—is total output divided by labor hours. Capital productivity is output divided by the value of capital stock. Multifactor productivity (MFP), sometimes referred to as Total Factor Productivity, is a more synthetic number. It is designed to measure the change in output for labor and capital combined. Capital and labor are combined by statistical means, reflecting the share of each to output for a given industry. As defined by the U.S. Bureau of Labor Statistics (BLS), MFP measures the joint influences of technology change, efficiency improvements, returns to scale, reallocation of resources, and other factors of economic growth, after allowing for the effects of capital and labor itself. MLF is a synthetic measure because the combined input variable is derived, rather than observed. For that reason, MFP is best understood as an index, and the trend is most important (see Box 1)

Labor productivity measures are critical, as they relate to employment and wages, and have to be thought about carefully. Labor productivity can increase because labor itself is inherently more productive—for example, the labor force may be better educated or healthier. Economists have long held that wage growth can occur only when labor productivity grows. At the same time, it does not necessarily follow that higher labor productivity means commensurate employment growth. Indeed, it has long been observed that employment tends not to keep pace with output, particularly beginning after World War II. So there is a long trend for productivity and output to run ahead of employment.

U.S. Productivity Trends

The productivity of the U.S. economy, as shown in Box 2 on the following page, has fluctuated over the last decade. Productivity grew modestly after the 2001 recession through mid-decade. Then, after some years of stagnation, a striking jump in productivity occurred between 2009 and 2010—in the teeth of the worst U.S. economic recession since the Great Depression. Notably, throughout the decade, labor productivity consistently trended above both MFP and capital productivity. Indeed, labor productivity has been conspicuously strong throughout the past decade.


Box 3 illustrates this trend dramatically for manufacturing. Since 1970, U.S. manufacturing has rapidly downsized its workforce, while output has actually grown in real terms, contrary to what most people expect to see from U.S. manufacturing. As seen, this trend is not transitory; it has been going on for decades.A good portion of the explanation for the rise in labor productivity is, paradoxically, bad news for American workers—substitution of equipment, computers, and other information technologies for employees has continued and, more important, employment has continued to be shed in key sectors of the economy, as corporations merge, and down size work force. The “jobless” recovery of the decade has been replaced by a “job shedding” recession.

While employment to output ratios have declined by far the most for manufacturing, it is true of the economy overall. Between 2000 and 2010, U.S. economic output per hour increased at a rate of close to three percent per year (i.e., work to output declined by about three percent). At the same time, MFP has been closer to one percent. Overall, the economy has become more capital and information intensive, production has become more efficient, while employment has languished, increasing by only 1.6 percent over the decade. 

What has occurred in manufacturing is likely to be less extreme for a service economy; unlike manufacturing, there are more limits to automating a service economy. However, globalization allows for a stratification of activities even within the service and information sectors, where low pay, labor intensive work—for example, call services—can be increasingly sent off-shore. The net effect is inevitably an increase in labor productivity, as it is measured, and a lower employment/output ratio, even for services. Workers are as good or even better, as jobs are filled by higher skilled workers, but there are fewer of them, as machines and computers are vastly more efficient, and make unnecessary jobs that people once filled.1

With worker productivity on the rise—at least statistically—some increases in real average wages should be expected. While wages have grown for some sectors, labor productivity for the economy as a whole has not always translated into commensurate gains in wages. Research by the St. Louis Federal Reserve found that between 1995 and 2006, U.S. labor productivity increased about 35 percent, compared to about 15 percent growth in real average hourly earnings.2 However, when total compensation was considered, productivity and compensation tracked more closely. Whether this effect occurred across the board at all income levels is another question.

The implications of all this for transportation are important. To what extent may transportation projects increase economic productivity and output? How? And if transportation improvements increase the productivity and size of the economy, will jobs and wages grow proportionally?

Relationship between Transportation

Investments and Economic Productivity

 

How may transportation improvements generate higher economic productivity? New transportation investments represent a form of public capital investment that can stimulate productivity of labor and private capital by expanding markets, creating economies of scale, and producing benefits from proximity. A series of econometric studies over the past two decades have consistently found a positive relationship between transportation investment (typically measured as highway capital stock) and economic productivity and output. Studies undertaken on behalf of the FHWA in the 1990s focused on aggregate national measures; econometric analyses by various academic  economists, such as Ishaq Nadiri,3 Alicia Munnel,4 and David Aschauer5 all found positive relationships between highway “capital stock” and national economic productivity and output. These and similar studies, while interesting and important, have limited applicability in understanding the impacts of specific transportation projects or programs of projects (clearly, only projects of significant scope can be expected to have productivity benefits; a new highway interchange is probably not worth looking at in this way.) Thus, they are less useful in expanding project evaluation procedures.

A range of dynamics that can lead to productivity and output increases from transportation investments have been noted in the literature. They include:

Market Expansion

Transportation and accessibility improvements allow firms to expand their market for sales, while allowing these same firms to buy intermediate inputs from a wider range of suppliers. Expanded markets allow firms to buy intermediate goods more cheaply and to sell to a wider market, expanding output and creating potential for significant economies of scale in production. Regions can become more specialized, and the overall mix of production activities among regions becomes more efficient.

Supply Chain Effects

Transportation’s impact on logistics costs are probably more studied than other impacts. In addition to lowering freight transport costs, reduced transit times and improved reliability in delivery allow shippers to lower inventory and increase just in time delivery. Improved logistics, including a more efficient mix of transportation and warehousing, allows some firms to reorganize their production processes, source intermediate goods from almost anywhere, all leading to increased productivity and lower production costs.

Labor Market Effects

By expanding access to jobs, employees can better match their skills to the types of jobs available. From the employer’s perspective, the effective labor supply expands, skilled positions can be filled more easily, and employers may not have to bid wages up as much to fill these types of high value added positions.

Work Related Travel

For transportation projects such as high speed rail,6 more comfortable travel with business amenities such as free Internet provides opportunities for business travelers to be productive, as compared with driving, or even flying, where transfers, inconvenience, lack of amenities, and delays can be onerous and cut down productivity while traveling.

Agglomeration Effects

Benefits from the spatial concentration of economic activities, especially jobs, have long been a central feature of urban economic models, and recent studies have revived interest in the role of density in aggregate economic growth. Agglomeration benefits refer to a range of effects, including denser interactions among firms of similar types—leading to knowledge sharing, labor pool sharing—and increased access to key inputs such as content contributors, research, financing, legal services, etc. An illustration of such increased interactions, in this case due to the California High Speed Rail project, between Silicon Valley/Northern California “tech” firms and Southern California content and high value added resource contributors is shown in Box 4.

Research in this area is perhaps more advanced than the other areas.7 Studies by the British economist Daniel Graham have been pioneering. Graham’s econometric experiments utilize the concept of “effective density”—by improving accessibility, markets are “effectively” denser, even though the actual spatial location of activities does not change. The econometrics relates variations in accessibility for small zones to variations in productivity of industries, and to wages of workers. The magnitude of agglomeration effects on productivity is typified by TCRP Project J-11 (2009),8 which estimates that a five percent increase in effective density yields a 0.1 percent change in productivity, a productivity elasticity of around 0.02 with respect to density. Other studies show somewhat more robust (output) elasticities, in a range of 0.08 to 0.3, with averages for an economy around 0.12 (Graham, 2007). In the United Kingdom, the Department for Transport has begun to incorporate agglomeration effects in their benefit cost guidelines.

Employment Benefits

To what extent can we expect transportation induced economic productivity gains to generate increased employment? It seems intuitive that if you can increase economic productivity and expand output, employment gains should follow directly. This may not be the case, however, as employment effects may be minimal, certainly less than proportional to the output gains.

As indicated earlier, employment/output ratios have declined in the U.S. Old economic models, which do not reflect this reality, can overstate the employment benefits of expanded output. Indeed, if transportation investments create economies of scale, improve production or logistics processes, or create scale economies, the productivity benefits may entail virtually no increases in employment—MFP just improves, without requiring more labor; productivity improvements can allow firms to expand output for the same level of spending on labor and equipment. In other cases, if transportation investments lower the costs of transportation (e.g., by reducing truck travel times), more transportation may be purchased, but less warehousing will be needed. The net effects on employment and wages are not clear. Wage impacts may be significant for some types of employees, as transportation projects can increase the range of choices available to workers (i.e., some workers can trade up to more productive and higher paying jobs). At the same time, an expanded labor market may increase supply and exert downward pressure on wages in some locales.

All of this is theoretical and speculative. Good hard research is needed to trace the relationship between transportation induced productivity gains, output increases, and employment.

Current Interest in Productivity Benefits for Project Evaluation

Demonstrating net economic gains, measured by productivity and ultimately output and GDP, is important to demonstrating that the societal benefits of a particular transportation investment exceed costs. And it seems clear by now that standard and very conservative approaches to benefit cost analysis, which focus on direct user benefits and externalities such as emissions, noise, and safety, are not capturing everything. Even expansion of benefit cost procedures to capture long term sustainability benefits may not be enough.

More currently, USDOT has sought to grapple with this issue through a series of studies and program guidance. Focusing on impacts of highway projects on freight, the Office of Freight Management and Operations has sponsored a series of research studies, the Freight Benefit Cost Study.9

This includes a number of monographs, such as Studies on Economic Growth/Productivity and Social Impacts. Further, BCA guidelines issued by USDOT as part of the TIGER program seek to specify conditions under which net economic productivity gains might be realized, and some criteria by which these gains might be included in a benefit cost analysis, required under the TIGER program guidance. TIGER guidance has focused on broadening the BCA benefits set only where “national productivity” gains are demonstrated. What this means is not always easy to discern. For example, the TIGER program guidance indicates that applicants may include workforce productivity benefits where it can be demonstrated that transportation investments enable workers to move into higher productivity jobs or transportation projects increase service to high unemployment areas.

A recently released TRB RFP, NCHRP 02-24 Economic Productivity and Transportation Priorities, is pending. It specifically asks for research leading to a better and more economically sound basis for incorporating transportation induced productivity benefits in benefit cost analysis procedures.

 

Notes:

  1. For a discussion of causes of increased labor productivity see the St. Louis Federal Reserve, National Economic Trends @ stlouisfed.org
  2. St. Louis Federal Reserve, National Economic Trends
  3. Nadiri, Ishaq M. (1996), Contribution of Highway Capital to Industry and National Productivity, for FHWA.
  4. Munnell, Alicia H. (1990), How Does Public Infrastructure Affect Regional Economic Performance?, New England Economic Review, (p. 11-32).
  5. Aschauer, David (1989), Is Public Expenditure Productive?, Journal of Monetary Economics 23, (p. 177-200).
  6. Parsons Brinckerhoff, 2011 California High Speed Rail Business Plan @ cahighspeedrail.ca.gov
  7.  Graham, Daniel (2007), Agglomeration Economies and Transport Investment, Discussion Paper 2007-11 (International Transport Forum).
  8. TCRP Project J-11 (2009), Task 7 Economic Impact of Public Transportation Investment.
  9. http://ops.fhwa.dot.gov/freight/freight_analysis/cba/index.htm 

 

Image Header Source: Ben Lerchin (Creative Commons)