Policy Watch Infrastructure Investment and Economic Growth Author

Document Sample
Policy Watch Infrastructure Investment and Economic Growth Author Powered By Docstoc
					Policy Watch: Infrastructure Investment and Economic Growth
Author(s): Alicia H. Munnell
Source: The Journal of Economic Perspectives, Vol. 6, No. 4, (Autumn, 1992), pp. 189-198
Published by: American Economic Association
Stable URL:
Accessed: 16/06/2008 12:52

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We enable the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact

                         Perspectives- Volume Number4-Fall 1992-Pages 189-198
       Journal of Economic                  6,

Policy Watch
Infrastructure Investment and
Economic Growth
Alicia H. Munnell

   P       ublic policies are often made without much reliance on economic
        reasoning. Economists are often unaware of what is happening in the
        world of public affairs. As a result, both the quality of public decision-
making and the role that economists play in it are less than optimal. This
feature contains short articles on topics that are currently on the agendas of
policymakers, thus illustrating the role of economic analysis in current debates.
Suggestions for future columns and comments on past ones should be sent to
Isabel V. Sawhill, c/o Journal of Economic Perspectives, The Urban Institute,
2100 M Street N.W., Washington, D.C. 20037.


    The stock of public capital in the United States is very large, $2.7 trillion in
1991 or about one-half the size of the private capital stock, according to data
from the U.S. Bureau of Economic Analysis.' About $2.2 trillion of that

IThe capital stock figures are calculated from historical investment figures revalued to current cost
that are then cumulated and depreciated using a perpetual inventory method.

*Alicia H. Munnell is Senior VicePresidentand Directorof Research,Federal Reserve
Bank of Boston, Boston, Massachusetts.
190   Journal of Economic Perspectives

Table I
Nonmilitary Nonresidential Net Public Capital Stock by Type of Asset, 1991

                Public                                    Billions of     Percent of
              Capital Stock                                Dollars          Total

Highwaysand Streets                                         $722.0            32
Waterand Sewer Systems                                       306.3            14
Buildingsand Other Structures
  Schools, Hospitals,and Other Buildings                      665.4           30
  Conservationand DevelopmentStructures                       177.6            8
  Electricand Gas Facilities,
    Systems,Airfields,etc.                                   163.6             7
Equipment                                                    209.7             9
Total                                                     $2,250.5           100

Source:U.S. Bureau of EconomicAnalysis,unpublisheddata.

public capital stock is nonmilitary, with $1.9 billion held by state and local
governments. Table 1 presents a breakdown. Despite its magnitude, public
capital had been virtually ignored by the economics profession until the late
1980s. At that time, David Aschauer (1989) triggered a long overdue dialogue
among economists and political leaders when he published a study arguing that
much of the decline in U.S. productivity that occurred in the 1970s was
precipitated by declining rates of public capital investment. My own work
confirmed these results (Munnell, 1990a).
     Spending advocates seized on these findings as support for increased
public investment. Transportation Secretary Samuel Skinner and New Jersey
Governor James Florio joined traditional interest groups to argue that more
public investment in infrastructure would help the economy. Prominent
economists signed a national petition for increased infrastructure spending.
Several congressional committees held hearings on this topic. The U.S. Council
of Mayors (1992) called for stimulative infrastructure spending earlier this year,
and Governor Clinton has made infrastructure spending a major part of his
economic plan. Felix Rohatyn (1992) argued for very large increases in infra-
structure investment in a recent article in New YorkReview of Books.
     The enthusiasm among policymakers for the early Aschauer results was
matched, if not surpassed, by skepticism on the part of many economists.
Critics of these studies charged that the methodology was flawed, that the
direction of causation between public investment and output growth is unclear
and that, even if the historical empirical relationships were estimated correctly,
they provide no clear indications for current policy.
     Who's right? What do we know and not know about the link between
public infrastructure and productivity? And what are the implications of these
results for policy?
                                                                          Alicia H. Munnell        191

Public Capital and the Production Process

      Everyone agrees that public capital investment can expand the productive
capacity of an area, both by increasing resources and by enhancing the produc-
tivity of existing resources. A well-constructed highway allows a truck driver to
avoid circuitous back roads and to transport goods to market in less time. The
reduction in required time means that the producer pays the driver lower
wages and the truck experiences less wear and tear. Hence, public investment
in a highway enables private companies to produce their products at lower total
cost. The condition of the highway, of course, is just as important as its
existence. Similar stories can be told for mass transit, water and sewer systems,
and other components of public capital.
      Beginning with Aschauer's work, a number of studies have estimated
regressions where the dependent variable is output within some area, and the
independent variables are private capital, labor, public capital, and a constant
for the level of technology.2 In such regressions, the levels of public capital are
generally significant, and the consensus is that Aschauer made a significant
contribution by drawing attention to the importance of public infrastructure
and by adding public capital to the conventional production function. The
controversy arises about the method of estimating this expanded function and
about the interpretation of the results.
     Aschauer's original aggregate time series estimates (1989), my reestimates
(1990a), and earlier work by Holz-Eakin (1988) suggest that the impact of
aggregate public capital on private sector output and productivity is very large.
My own equations indicate that a 1 percent increase in the stock of public
capital would increase output by .34 percent. Given the size of the public capital
stock and output, these figures imply a marginal productivity of public capital
of roughly 60 percent; that is, a $1 increase in the public capital stock would
raise output by $.60. The marginal productivity of private capital estimated
from these equations is about 30 percent. Looking at similar numbers,
Aschauer (1990, p. 16) concludes that "increases in GNP resulting from in-
creased public infrastructure spending are estimated to exceed those from
private investment by a factor of between two and five."
      In my view, the implied impact of public infrastructure investment on
private sector output emerging from the aggregate time series studies is too
large to be credible. It does not make sense for public capital investment to

2Treating  public capital as an input whose services enhance the productivity of both capital and
labor yields the equation Q = (MFP)*f(K, L, G), where Q is output, MFP is the level of technol-
ogy, K is the private capital stock, L is labor, and G is the stock of public capital. Assuming a
generalized Cobb-Douglas form of technology yields a more specific relationship between inputs
and outputs: Q = MFP*KaLbGC. Translating this equation into logarithms produces a linear
function that can be estimated: ln Q = ln MFP + a ln K + b ln L + c ln G. The coefficients a, b, and
c are the output elasticities of factor inputs. In other words, the coefficients indicate the percentage
change in output for a given percentage change in factor input.
192    Journal of EconomicPerspectives

have a substantially greater impact on private sector output than private capital
investment, particularly considering that so much public investment goes for
improving the environment and other goals that are not captured in national
output measures.
     To obtain more evidence, I looked at the relationship between public
capital and measures of economic activity at the state level (Munnell, 1990b).
Since no data on state-level public or private capital stocks were available, the
first step was to construct stock estimates; these estimates were then used in
three separate exercises. The first, parallel to the national work, estimated
production functions for states and found that public capital had a significant,
positive impact on output, although the output elasticity was roughly one-half
the size of the national estimate.
     The second analysis examined the relationship between public and private
investment, which is characterized by two opposing forces. On one hand, public
capital enhances the productivity of private capital, raising its rate of return
and encouraging more investment. On the other hand, from the investor's
perspective, public capital acts as a substitute for private capital and "crowds
out" private investment. The estimated equations confirmed both forces
but suggested that, on balance, public capital investment stimulates private
     The third exercise used a business location model to explore the relation-
ship between public capital and employment growth. Here the average annual
change in employment was estimated as a function of variables reflecting input
costs (labor, energy, land), market size, tax burden, and public capital stock.
The results showed that, after accounting for all the other factors that affect
employment, public capital had a positive, statistically significant effect on
employment growth.3
     Taken together, these three analyses indicate that public capital has a
positive impact on several measures of state-level economic activity: output,
investment, and employment growth. The magnitudes of these effects are
considerably smaller than those found at the national level; for instance, the
elasticity of public capital with respect to output was .15, roughly half the
estimate at the national level. These estimates are consistent with the estimates
of other researchers working at the state level (Mera, 1973; Costa, Ellson, and
Martin, 1987; Garcia-Mila and McGuire, 1992).

Enter the Critics

     Critics have leveled three major changes at the results emerging from
estimated production functions. First, they contend that common trends in the
output and public infrastructure data have led to a spurious correlation.
Second, they argue that the wide range of estimates emerging from the various

3This estimated effect on state employment may overestimate the national impact if employment is
merely diverted from one area to another.
                                                                              Policy Watch 193

studies renders the coefficients suspect. Finally, they suggest that causation
runs not from public capital to output, but rather in the other direction.
     The most vociferous critics, concerned about the seemingly clarion call for
dramatically increased public investment, focus mainly on the aggregate time
series; they argue essentially that the equations should be estimated in the form
of first differences (Aaron, 1990; Hulten and Schwab, 1991; Jorgenson, 1991;
Tatom, 1991). Specifically, they contend that the data are not stationary but
tend to drift over time, and that it is necessary to remove this trend to eliminate
spurious correlations and determine the true relationship between the two
variables. This means specifying the relationship in terms of first differences,
which often yields results showing that public capital's effect is quite small,
sometimes negative, and generally not statistically significant.
     The first-differencing specification has its problems, however. After all, no
one would expect the growth in the capital stock, whether private or public, in
one year to be correlated with the growth in output in that same year. In fact,
equations estimated in this form often yield implausible coefficients for labor
and private capital as well as for public capital (Evans and Karras, 1991; Hulten
and Schwab, 1991; Tatom, 1991). However, researchers do not conclude from
these misspecified equations that private capital and labor do not contribute to
private sector output.
     In addition, first-differencing destroys any long-term relationship in the
data, which is exactly what one is trying to estimate. Instead of just first-dif-
ferencing, the variables should be tested for co-integration, adjusted, and
estimated accordingly. That is, researchers should examine not just whether
the variables grow over time, that is, the extent to which they are non-
stationary, but also whether they grow together over time and converge to their
long-run relationship, that is, the extent to which they are co-integrated.4
     The second broad criticism is that the wide range of estimates of public
capital's impact on output makes the empirical linkages fragile at best. In my
view, the critics are seriously misreading the evidence. In almost all cases the
impact of public capital on private sector output and productivity has been
positive and statistically significant. This finding is amazing, given that much
public capital spending is designed to alleviate environmental problems or
enhance the quality of life, and therefore contributes little to national output as
conventionally measured.
     Furthermore, the coefficients at each level of government tend to be very
similar across studies, as shown in Table 2. The variation between estimates
occurs as the unit of observation moves from the nation to states to cities. As the
geographic focus narrows, the estimated impact of public capital becomes
smaller. The most obvious explanation is that, because of leakages, one cannot

4 ratom (1991) adjusts for co-integration by including in the equation all significant leads and lags
of the independent variables and finds public capital's effect to be negative and insignificant.
However, Tatom's effort has been criticized for including the price of energy in an equation with
the quantities of other factor inputs. Clearly, more work is needed, and it should focus on adjusting
for co-integration rather than mechanically first-differencing all variables.
194    Journal of Economic Perspectives

Table 2
Production Function Estimates of the Output Elasticity
of Public Capital by Level of Geographic Aggregation

                               Level of                                                Elasticityof
      Author                  Aggregation                   Specification             Public Capital

Aschauer (1989)            National                  Cobb-Douglas; Log levels              .39
Holz-Eakin (1988)          National                  Cobb-Douglas; Log levels              .39
Munnell (1990a)            National                  Cobb-Douglas; Log levels              .34
Costa, Ellson,
  Martin (1987)           States                     Translog; Levels                      .20
Eisner (1991)             States                     Cobb-Douglas; Log levels              .17
Mera (1973)               Japanese regions           Cobb-Douglas; Log levels              .20
Munnell (1990b)           States                     Cobb-Douglas; Log levels              .15
Duffy-Deno and
  Eberts (1989)a           Metropolitan areas        Log levels                            .08
Eberts (1986, 1990)        Metropolitan areas        Translog; Levels                      .03

Note: Results from the first-difference model used by Aaron (1990), Hulten and Schwab (1991), and
other critics are not included because they yielded implausible coefficients on the private inputs.
aThe authors do not estimate a production function, but instead use personal income as the
dependent variable.

capture all of the payoff to an infrastructure investment by looking at a small
geographic area.
     The third major criticism is that the direction of causation may run from
high levels of output to greater public capital investment, rather than the other
way around. The criticism is legitimate. Capital investment, private as well as
public, goes hand in hand with economic activity. However, this mutual
influence can exist without necessarily tainting the coefficient on public capital,
or, for that matter, private capital in estimated production functions.
      Eberts and Fogarty (1987) examined the question of causality by looking at
public and private investment data from 1904 to 1978 for 40 metropolitan
areas. They found causation running in both directions. Their analysis indi-
cated that public investment led private investment in cities that experienced
most of their growth before the 1950s, while the reverse was true for southern
cities and cities that grew faster since 1950.
     To examine the simultaneity issue, I reestimated some equations using the
state data, but included only the value of public capital at the beginning of the
period, which foreclosed the possibility of any feedback effect of output growth
on public capital investment. Nonetheless, public capital continued to exhibit a
large, positive, statistically significant effect on output. This small exercise does
not put the question to rest, but does suggest that the coefficient of public
capital is not seriously tainted by the simultaneity problem.
      Other critics have suggested that the production function framework is
inadequate, both because it omits input prices (which affect factor utilization
                                                                        Alicia H. Munnell       195

and bias the estimated coefficients) and also because it places too many restric-
tions on firms' technology and behavior (Friedlaender, 1990; Morrison and
Schwartz, 1992). They believe that researchers should instead estimate cost
functions, which allow one to disentangle the effects of infrastructure, scale
economies, and fixed effects on costs and the cost-output relationship.
Dalenberg and Eberts (1992), Morrison and Schwartz (1992), and Nadiri and
Mamuneas (1992) all adopt the cost-function approach and find that public
capital significantly reduces the costs of private production.
     In summary, the critics are correct that the numbers emerging from the
aggregate time series studies are not credible, and that more evidence is
needed on the causation issue. But the tendency to throw the baby out with the
bathwater should be resisted. At this point, an even-handed reading of the
evidence-including     the growing body of cross-sectional results-suggests that
public infrastructure is a productive input which may have large payoffs.

Does the Nation Need More Public Investment?

     If infrastructure has substantial payoffs, does this imply that public capital
is undersupplied and higher levels of investment are warranted? At this time,
production function estimates provide little guidance on this point, but some
other evidence seems to support the notion that profitable public investment
opportunities exist.
     For example, cost-benefit studies reported by the Congressional Budget
Office (1988) indicate that the return to projects designed to maintain the
average condition on the federal highway system could be as high as 30 to 40
percent. The CBO (1991) also suggests the likelihood of substantial bene-
fits from increased outlays for both air traffic control and airport capacity
     Another piece of evidence comes from the work of George Peterson (1990,
1991). He explored voters' preferences for public capital investment as ex-
pressed in bond elections and other referenda. If public officials were trying to
satisfy the median voter, as theory suggests, they would submit frequent bond
proposals for consideration to assess voter demand. As a result, bond elections
should be closely contested with bond approval rates and margins close to 50
percent. Instead, Peterson found that 80 percent of infrastructure bond pro-
posals were approved between 1984 and 1989, and that the margin exceeded
66 percent on average.5 Although only 25 percent of capital spending passes
through the referendum process, this experience suggests that voters are
willing to pay for more infrastructure spending.

5Aaron (1991) offers a note of caution concerning Peterson's work. Specifically, he points out that
elected officials control the supply of proposals and it is very difficult to make any determinations
about voters' demands for infrastructure without analyzing changes in supply. Aaron concludes,
however, that he would not be surprised if Peterson's claims are entirely correct, but does not feel
that the case has been proven yet.
196   Journal of EconomicPerspectives

     Despite such findings, several voices urge caution when considering in-
creased spending for public infrastructure. For example, Clifford Winston and
his colleagues contend that the condition of the nation's highways could be
improved and congestion reduced with the same or less investment by making
three changes (Winston, 1990, 1991; Small, Winston, and Evans, 1989). First,
building roads thicker than prevailing engineering standards would produce
great savings. Second, shifting from a tax on the number of truck axles to one
on weight per axle would greatly encourage efficient use of highways and
minimize damage. Third, increasing the use of congestion taxes would reduce
peak-period congestion and increase the services provided by existing capacity.
Others point to inefficiencies in our federal grant programs, where matching
rates are probably much higher than can be justified on the basis of interjuris-
dictional spillovers (Gramlich, 1990, 1991). As a result, many states pay as little
as 10 cents on the dollar for new highways, giving little reason to believe that
they will make economically sensible decisions about infrastructure investment.
     Those who worry about the incentives to spend, the efficiency of design,
and the appropriateness of the prices charged tend -to believe that infrastruc-
ture policy should focus on eliminating current distortions and inefficiencies.
They seem to believe that once the perversities in the existing system are
removed, the present stock of infrastructure can meet most of the nation's
needs. Additional investment at this time will divert attention and alleviate the
pressure to make needed pricing reforms. Although these concerns about
pricing and grant programs should be addressed, my hunch is that these
reforms will be slow in coming and the country will still need more spending to
repair roads and bridges, treat wastewater, dispose of trash, and improve the
quality of the nation's lakes and rivers.


     Some proponents of expanded infrastructure investment are truly inter-
ested in the effect that additional spending will have on economic activity;
others are at least as interested in the potential or lucrative contracts coming
their way. The immediacy of policy implications places great demands on
economists, both not to oversell preliminary results and not to dismiss a
growing body of evidence. It has proved a difficult environment in which to
make a calm assessment of the evidence.
     That said, my views are as follows. This is an area ripe for research that
could have important policy implications. Researchers should focus on explain-
ing the variations in the coefficients by level of government, disentangling the
causation question, and examining the cointegration issue. Aggregate results,
however, cannot be used to guide actual investment spending. Only cost-benefit
studies can determine which projects should be implemented. Finally, while
reforms to grant programs and pricing should occur, it is probably neither
realistic nor desirable to hold off on infrastructure investment until such
                                                                            Policy Watch 197

reforms are adopted. For the evidence suggests that, in addition to providing
immediate economic stimulus, public infrastructure investment has a signifi-
cant, positive effect on output and growth.

   Aaron, Henry J., "Discussion of 'Why Is           Eberts, Randall W., "Estimating the Contri-
Infrastructure Important?'" In Munnell,           bution of Urban Public Infrastructure to Re-
Alicia H., ed., Is There a Shortfall in Public    gional Economic Growth," Working Paper No.
Capital Investment? Conference Series No. 34,     8610, Federal Reserve Bank of Cleveland, De-
Federal Reserve Bank of Boston, June 1990,        cember 1986.
51-63.                                               Eberts, Randall W., 'Public Infrastructure
   Aaron, Henry J., "Discussion of 'Historical    and Regional Economic Development," Eco-
Perspectives on Infrastructure Investment:        nomic Review, Federal Reserve Bank of Cleve-
How Did We Get Where We Are?'" Paper              land, Quarter 1, 1990, 26, 15-27.
presented by George E. Peterson at the Amer-         Eberts, Randall W. and Michael S. Fogarty,
ican Enterprise Institute Conference on "In-      "Estimating the Relationship Between Local
frastructure Needs and Policy Options for the     Public and Private Investment," Working Pa-
 1990s," Washington, D.C, February 4, 1991.       per No. 8703, Fedral Reserve Bank of Cleve-
   Aschauer, David Alan, "Is Public Expendi-      land, May 1987.
ture Productive?" Journal of Monetary Eco-           Eisner, Robert, "Infrastructure and Re-
nomics, March 1989, 23, 177-200.                  gional Economic Performance," New England
   Aschauer, David Alan, "Does Public Capital     Economic Review, Federal Reserve Bank of
Crowd Out Private Capital?" Journal of Mone-      Boston, September/October 1991, 47-58.
tary Economics,September 1989, 24, 171-88.           Evans, Paul and Georgios Karras, "Are
   Aschauer, David Alan, Public Investmentand     Government Activities Productive? Evidence
Private Sector Growth. Washington, D.C.: Eco-     from a Panel of U.S. States," unpublished,
nomic Policy Institute, 1990.                     Ohio State University, October 1991.
   Congressional Budget Office, New Direc-           Friedlaender, Ann F., "Discussion of'How
tionsfor the Nation's Public Works.Washington,    Does Public Infrastructure Affect Regional
D.C.: Government Printing Office, September       Economic Performance?"' In Munnell, Alicia
1988.                                             H., ed., Is There a Shortfall in Public Capital
   Congressional Budget Office, How Federal       Investment?Conference Series No. 34, Federal
Spending for Infrastructureand Other Public In-   Reserve Bank of Boston, 1990, 108-12.
vestments Affectsthe Economy.Washington, D.C.:       Garcia-Mila, Theresa and Therese J.
Government Printing Office, July 1991.            McGuire, "The Contribution of Publicly Pro-
   Costa, Jose da Silva, Richard W. Ellson,       vided Inputs to States' Economies," Regional
and Randolph C. Martin, "Public Capital, Re-      Scienceand UrbanEconomics,forthcoming 1992.
gional Output, and Development: Some Em-             Gramlich, Edward M., "How Should Public
pirical Evidence," Journal of Regional Science,   Infrastructure Be Financed?" In Munnell, Ali-
August 1987, 27, 419-37.                          cia H., ed., Is Therea Shortfall in Public Capital
   Dalenberg, Douglas R. and Randall W.           Investment?Conference Series No. 34, Federal
Eberts, "Estimates of the Manufacturing Sec-      Reserve Bank of Boston, June 1990, 223-37.
tor's Desired Level of Public Capital: A Cost        Gramlich, Edward M., "U.S. Infrastructure
Function Approach," Paper presented at the        Needs: Let's Get the Prices Right." Paper pre-
Annual Meeting of the Western Economic As-        sented at the American Enterprise Institute
sociation, San Francisco, California, July        Conference on "Infrastructure Needs and Pol-
10-13, 1992.                                      icy Options for the 1990s," Washington, D.C,
   Duffy-Deno, Kevin T., and Randall W.           February 4, 1991.
Eberts, "Public Infrastructure and Regional          Holz-Eakin, Douglas, "Private Output,
Economic Development: A Simultaneous Ap-          Government Capital, and the Infrastructure
proach," Working Paper No. 8909, Federal          'Crisis'" Discussion Paper Series No. 394, New
Reserve Bank of Cleveland, August 1989.           York: Columbia University, May 1988.
198     Journal of EconomicPerspectives

   Hulten, Charles, R. and Robert M. Schwab,           25, 1992.
"Is There Too Little Public Capital? Infra-               Small, Kenneth A., Clifford Winston, and
structure and Economic Growth." Paper pre-             Carol A. Evans, Road Work: A New Highway
sented at the American Enterprise Institute            Pricing and InvestmentPolicy. Washington, D.C.:
Conference on "Infrastructure Needs and Pol-           The Brookings Institution, 1989.
icy Options for the 1990s," Washington, D.C.,             Tatom, John A., "Public Capital and Private
February 4, 1991.                                      Sector Performance," Review, Federal Reserve
   Jorgenson, Dale W., "Fragile Statistical            Bank of St. Louis, May/June 1991, 73, 3-15.
Foundations: The Macroeconomics of Public                 U.S. Conference of Mayors, Ready to Go: A
Infrastructure Investment." Paper presented            Survey of USA Public WorksProjects to Fight the
at the American Enterprise Institute Confer-           RecessionNow. Washington, D.C.: U.S. Confer-
ence on "Infrastructure Needs and Policy Op-           ence of Mayors, 1992.
tions for the 1990s," Washington, D.C.,                   U.S. Congress, Office of Technology As-
February 4, 1991.                                      sessment, Rebuilding the Foundations: State and
    Mera, Koichi, "Regional Production Func-           Local Public WorksFinancing and Management.
tions and Social Overhead Capital: An Analy-           Washington, D.C.: Government Printing Of-
sis of the Japanese Case," Regional and Urban          fice, March 1990.
Economics,May 1973, 3, 157-85.                            U.S. Department of Commerce, Bureau of
    Morrison, Catherine J. and Amy E.                  the Census, Government Finances: 1988-89.
Schwartz, "State Infrastructure and Produc-            Washington, D.C.: Government Printing Of-
tive Performance," Working Paper No. 3981,             fice, February 1991.
 Cambridge: National Bureau of Economic Re-               U.S. Department of Transportation, Mov-
search, January 1992.                                  ing America: New Directions, New Opportunities.
    Munnell, Alicia H., "Why Has Productivity          Washington, D.C.: Government Printing Of-
 Declined? Productivity and Public Invest-             fice, February 1990.
 ment," New England Economic Review, Federal              U.S. House of Representatives, Committee
 Reserve Bank of Boston, January/ February             on the Budget, Task Force on Community
 1990a, 3-22.                                          Development and Natural Resources, "Hear-
    Munnell, Alicia H., with the assistance of         ings on Investments in America's Hometowns:
 Leah M. Cook, "How Does Public Infrastruc-            Community Development, Issues in Urban
 ture Affect Regional Economic Performance?,"          Areas and Other Places." Washington, D.C.:
New England EconomicReview, Federal Reserve            Government Printing Office, September 26,
 Bank of Boston, September/October 1990b,               1991.
 11-32.                                                   U.S. House of Representatives, Committee
    Nadiri, M. Ishaq and Theofanis P. Ma-              on Public Works and Transportation, "Hear-
 muneas, "The Effects of Public Infrastructure         ings on Our Nation's Transportation and Core
 and R&D Capital on the Cost Structure and             Infrastructure." Washington, D.C.: Govern-
 Performance of U.S. Manufacturing Indus-              ment Printing Office, February 20, 1991.
 tries," mimeo, New York University, February             U.S. House of Representatives, "The Status
 1992.                                                 of the Nation's Highways and Bridges: Condi-
    National Council on Public Works Im-               tions and Performance." Washington, D.C.:
 provement, Fragile Foundations: A Report on           Government Printing Office, September 1991.
America'sPublic Works.Washington, D.C.: Gov-              U.S. Senate, Committee on Environment
 ernment Printing Office, February 1988.               and Public Works, Subcommittee on Water
    Peterson, George E., "Is Public Infrastruc-        Resources, Transportation, and Infrastruc-
 ture Undersupplied?" In Munnell, Alicia H.,           ture, "Hearings on Infrastructure, Productiv-
 ed., Is There a Shortfall in Public Capital Invest-   ity, and Economic Growth." Washington, D.C.:
 ment? Conference Series No. 34, Federal Re-           Government Printing Office, February 5,
 serve Bank of Boston, June 1990, 113-35.               1991.
    Peterson, George E., "Historical Perspec-             Winston, Clifford M., "How Efficient Is
tives on Infrastructure Investment: How Did            Current Infrastructure Spending and Pric-
We Get Where We Are?" Paper presented at               ing?" In Munnell, Alicia H., ed., Is There a
the American Enterprise Institute Conference           Shortfall in Public Capital Investment? Confer-
 on "Infrastructure Needs and Policy Option            ence Series No. 34, Federal Reserve Bank of
 for the 1990s," Washington, D.C., February 4,          Boston, June 1990, 183-205.
 1991.                                                    Winston, Clifford M., "Efficient Trans-
    Rohatyn, Felix, "What the Government               portation Infrastructure Policy," Journal of
 Should Do," New YorkReview of Books, June             EconomicPerspectives,Winter 1991, 5, 113-27.