Economics of Infrastructure

An NBER conference on Economics of Infrastructure took place March 1 in Cambridge. Research Associates Edward L. Glaeser of Harvard University and James M. Poterba of MIT organized the meeting, which was sponsored by the Smith Richardson Foundation. These researchers' papers were presented and discussed:

Abhishek Nagaraj, University of California, Berkeley

The Private Impact of Public Information: Landsat Satellite Maps and Gold Exploration

For centuries, the availability of maps of under-explored geographies has provided new opportunities for innovators, and yet mapping as a channel to enable discovery has been rarely examined. To shed light on this topic, Nagaraj focuses on the impact of the NASA Landsat satellite mapping program on shaping the level and distribution of new discoveries between firms in the gold exploration industry. Nagaraj finds that idiosyncratic gaps in mapping coverage (from technical failures and cloud-cover in satellite imagery) had important implications for gold exploration -- firms were almost twice as likely to report the discovery of new deposits once regions were successfully mapped and the mapping program disproportionately supported discoveries from smaller, entrepreneurial firms, especially in regions with high quality local institutions. These findings point to the important but underexamined role of mapping as an economic activity in shaping industry performance and entrepreneurship.

Shoshana Vasserman and Valentin Bolotnyy, Harvard University

Scaling Auctions as Insurance: A Case Study in Infrastructure Procurement

The U.S. government spends about $165B per year on highways and bridges, or about 1% of GDP. Much of it is spent through "scaling" procurement auctions, in which private construction firms submit unit price bids for each piece of material required to complete a project. The winner is determined by the lowest total cost -- given government estimates of the amount of each material needed -- but, critically, they are paid based on the realized quantities used. This creates an incentive for firms to skew their bids -- bidding high when they believe the government is underestimating an item's quantity and vice versa -- and raises concerns of rent-extraction among policymakers. For risk averse bidders, however, scaling auctions provide a distinctive way to generate surplus: they enable firms to limit their risk exposure by placing lower unit bids on items with greater uncertainty. To assess this effect empirically, Vasserman and Bolotnyy develop a structural model of scaling auctions with risk averse bidders. Using data on bridge maintenance projects undertaken by the Massachusetts Department of Transportation (MassDOT), they present evidence that bidding behavior is consistent with optimal skewing under risk aversion. They then estimate bidders' risk aversion, the risk in each auction, and the distribution of bidders' private costs. Finally, the researchers simulate equilibrium item-level bids under counterfactual settings to estimate the fraction of MassDOT spending that is due to risk and evaluate alternative mechanisms under consideration by MassDOT. They find that scaling auctions provide substantial savings to MassDOT relative to lump sum auctions and suggest several policies that might improve on the status quo.

Christoph Boehm, University of Texas at Austin

Government Consumption and Investment: Does the Composition of Purchases Affect the Multiplier?

Boehm shows that a large and conventional class of macroeconomic models predicts that short-lived government investment shocks have a smaller fiscal multiplier than government consumption shocks. Boehm tests this prediction in a panel of OECD countries using real-time forecasts of government consumption and investment to purify changes in purchases of their predicted components. Consistent with theory, Boehm estimates a government investment multiplier near zero and a government consumption multiplier of approximately 0.8. The findings suggest that fiscal stimulus packages which contain large government investment components may not be as effective at stimulating aggregate demand as commonly thought.

Pablo Fajgelbaum, University of California, Los Angeles and NBER, and Stephen J. Redding, Princeton University and NBER

Trade, Structural Transformation and Development: Evidence from Argentina 1869-1914

Fajgelbaum and Redding provide new theory and evidence on the relationship between economic development and international trade using Argentina's late-19th-century integration into the global economy. They show that structural transformation, from agriculture to non-agriculture, and across disaggregated goods within the agricultural sector, was central to Argentina's rapid export-led economic development. The researchers provide evidence that the reductions in internal transport costs from the construction of the railroad network were important in enabling interior regions to participate in this process of structural transformation and economic development. The researchers rationalize their empirical findings using a theoretical framework that emphasizes a spatial Balassa-Samuelson effect, in which regions with good access to world markets have higher population densities, urban population shares, relative prices of non-traded goods, and land prices relative to wages. In counterfactuals, the researchers find that the construction of the railroad network increases the total population of Argentina by 49 percent under free international migration and raises the common real wage across all Argentinian districts by 8 percent under restricted international migration.

Daniel Leff Yaffe, University of California, San Diego

The Interstate Multiplier

By considering the construction of the Interstate Highway System, Yaffe asks how big, if any, are returns to constructing new highways? In this context, Yaffe finds that the biggest threats to identification are endogeneity and anticipation. To overcome the first, Yaffe notes that a state's initial population and area shares played an important role in determining the assignment of interstate highway funds. To overcome the second, Yaffe proposes using news to identify the timing of shocks. The solutions are combined in an IV local projection framework, as in Ramey & Zubairy (2018), and estimate a relative multiplier of 1.7 at the 15 year horizon. Then the specification is extended to allow for spillover effects. Finally, using the neoclassical model in a multi-region setting Yaffe studies the channels through which highway spending impacts the economy. Both empirical and theoretical results suggest that in the case of highway spending the relative multiplier is a lower bound of the aggregate multiplier.

Aleksandar Andonov, University of Amsterdam; Roman Kräussl, University of Luxembourg; and Joshua Rauh, Stanford University and NBER

The Subsidy to Infrastructure as an Asset Class (NBER Working Paper No. 25045)

Andonov, Kräussl, and Rauh investigate the characteristics of infrastructure as an asset class from an investment perspective of a limited partner. While non U.S. institutional investors gain exposure to infrastructure assets through a mix of direct investments and private fund vehicles, U.S. investors predominantly invest in infrastructure through private funds. The researchers find that the stream of cash flows delivered by private infrastructure funds to institutional investors is very similar to that delivered by other types of private equity, as reflected by the frequency and amounts of net cash flows. U.S. public pension funds perform worse than other institutional investors in their infrastructure fund investments, although they are exposed to underlying deals with very similar project stage, concession terms, ownership structure, industry, and geographical location. By selecting funds that invest in projects with poor financial performance, U.S. public pension funds have created an implicit subsidy to infrastructure as an asset class, which the researchers estimate within the range of $730 million to $3.16 billion per year depending on the benchmark.

Christopher Severen, Federal Reserve Bank of Philadelphia

Commuting, Labor, and Housing Market Effects of Mass Transportation: Welfare and Identification

Using a panel of tract-level bilateral commuting flows, Severen estimates the causal effect of Los Angeles Metro Rail on commuting between connected locations. Unique data, in conjunction with a spatial general equilibrium model, isolate commuting benefits from other channels. A novel strategy interacts local innovations with intraurban geography to identify all model parameters (local housing and labor elasticities). Metro Rail connections increase commuting between locations containing (adjacent to) stations by 15% (10%), relative to control routes selected using proposed and historical rail networks. Other margins are not affected. Elasticity estimates suggest relatively inelastic mobility and housing supply. Metro Rail increases welfare $146 million annually by 2000, less than both operational subsidies and the annual cost of capital. More recent data show some additional commuting growth.

Nicolas Campos, Eduardo Engel, and Ronald Fischer, Universidad de Chile, and Alexander Galetovic, Stanford University

Renegotiations and Corruption: The Odebrecht Case

In 2016, Brazilian construction firm Odebrecht was fined $2.6 billion by the US Department of Justice (DOJ). According to the plea agreement, between 2001 and 2016 Odebrecht paid $786 million in bribes in 10 Latin American and two African countries in around 150 large projects. The DOJ estimated that bribe payments increased Odebrecht's profits by $3.2 billion. Judicial documents and press reports on the Odebrecht case reveal detailed information on the workings of corruption in the infrastructure sector. Based on these sources the researchers establish five facts. First, renegotiations amounted to 71.3 percent of investment estimated when contracts were awarded, compared with 6.5 percent for projects where Odebrecht paid no bribes. Second, Odebrecht's bribes were of the order of one percent of a project's final investment. Third, Odebrecht's profits were small, both in projects where it paid bribes (less than two percent of final investment) and in its overall operation. Fourth, following the creation of an internal unit to centralize bribe payments and substitute electronic payments into off shore accounts for cash, Odebrecht's sales increased close to ten-fold while its profits remained small. Last, net profits from bribing were Odebrecht's main source of profits during the period. Campos, Engel, Fischer, and Galetovic build a model where firms compete for a project, anticipating a bilateral renegotiation at which their bargaining power is larger if they pay a bribe. Conditional on paying a bribe and cost dispersion among firms being small, in equilibrium firms' profits are small, while lowballing and renegotiated amounts are large. Small bribes are necessary to produce large renegotiations. When one firm unilaterally innovates by reducing the cost of paying bribes, its market share increases substantially while profits, which are proportional to both the cost advantage and the magnitude of bribes, remain small. A parametrization with the DOJ's data suggests that Odebrecht enjoyed a substantial cost advantage in bribing, of the order of 70 percent.

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us