NBER WORKING PAPER SERIES
WHAT YOU EXPORT MATTERS
Ricardo Hausmann
Jason Hwang
Dani Rodrik
Working Paper 11905
http://www.nber.org/papers/w11905
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
December 2005
The views expressed herein are those of the author(s) and do not necessarily reflect the views of the National
Bureau of Economic Research.
©2005 by Ricardo Hausmann, Jason Hwang, and Dani Rodrik. All rights reserved. Short sections of text,
not to exceed two paragraphs, may be quoted without explicit permission provided that full credit, including
© notice, is given to the source.
What You Export Matters
Ricardo Hausmann, Jason Hwang, and Dani Rodrik
NBER Working Paper No. 11905
December 2005
, Revised March 2006
JEL No. F1, O4
ABSTRACT
When local cost discovery generates knowledge spillovers, specialization patterns become partly
i
ndeterminate and the mix of goods that a country produces may have important implications for
economic growth. We demonstrate this proposition formally and adduce some empirical support
for it. We construct an index of the "income level of a country's exports," document its
properties, and show that it predicts subsequent economic growth.
Ricardo Hausmann
John F. Kennedy School of Government
Harvard University
ricardo_hausmann@harvard.edu
Jason Hwang
Department of Economics
Harvard University
[email protected]rvard.edu
Dani Rodrik
John F. Kennedy School of Government
Harvard University
79 JFK Street
Cambridge, MA 02138
and NBER
dani_rodrik@harvard.edu
What You Export Matters
Ricardo Hausmann
John F. Kennedy School of Government
Harvard University
Jason Hwang
Department of Economics
Harvard University
Dani Rodrik
John F. Kennedy School of Government
Harvard University
December 2005
Abstract
W he n local cost discovery generates kn owledge sp illovers, specializa-
tion patterns become partly indeterminate and th e mix of goods that a
country produces may have im porta nt implications for economic growth.
We demonstrate this proposition formally and adduce some empirical sup-
port for it . We construct an index of the "income level of a c ou ntry’s
exports," document its properties, and show that it predicts subseq ue nt
economic growth.
1 Introduction
Why do countries produce what they do, and doe s it matter? The conventional
approach to these questions is driven by what we might call the "fundamentals"
view of the world. In this view, a country’s fundamentals–namely its endow-
ments of physical and human capital, labor, and natural resources along with
the overall quality of its institutions–determine relative costs and the patterns of
specialization that go with them. Attempts to reshape the production structure
beyond the boundaries set by these fundamentals are likely to fail and hamper
economic performance.
We argue in this paper that specialization patterns are partly indeterminate
and may be shaped by idiosyncratic elements. While fundamentals play an
Hausmann and Rodrik thank the Cente r for Inte rnational Development for …nancial sup-
port. Oeindrila Dube and Bailey Klinger provided excellent research assistance. We also
thank Ralph Ossa and Li u Chunyong for cat ching ty pos in an ear lier version.
1
important role, they do not uniquely pin down what a country will produce and
export. Furthermore, not all good are alike in terms of their consequences for
economic performance. Specializing in some products will bring higher growth
than specializing in others. In this setting, government policy has a potentially
important positive role to play in shaping the production structure.
Our analysis is driven by what Hausmann and Rodrik (2003) have earlier
called "cost discovery." An entrepreneur who attempts to produce a good for
the …rst time in a developing ec onomy necessarily faces considerable cost un-
certainty. Even if the good c omes with a standard technology ("blueprint"),
domestic factor endowments and institutional realities will require tinkering
and local adaptation (see Evenson and Westphal 1995, Lall 2000). What the
entrepreneur e¤ectively does is to explore the underlying cost structure of the
economy. This process is one with considerable p ositive externalities for other
entrepreneurs. If the project is successful, other entrepreneurs learn that the
product in question can be pro…tably produced and emulate the incumbent. In
this way, the returns to the pioneer investor’s cost discovery become socialized.
If the incumbent ends up with failure, on the other han d, the losses remain
private. This knowledge externality implies that investment levels in cost dis-
covery are sub-optimal unless the ind us try or the government …nd some way in
which the externality can be internalized.
In such a setting, the range of goods that an economy ends up producing
and exporting is determined not just by the usual fundamentals, but also by the
number of entrep reneu rs that can be stimulated to engage in cost discovery in
the modern sectors of the economy. The larger this numb er, the closer that the
economy can get to its productivity frontier. When there is more cost discovery,
the productivity of the resulting set of activities is higher in expectational terms
and the jackpot in world markets bigger.
In this pape r we provide a simple formal model of this process. We also
supply some evidence that we think is suggestive of the importance of the forces
that our formal framework identi…es. We are interested in showing that some
traded goods are associated with higher productivity levels than others and
that countries that latch on to higher productivity goods (through the cost
discovery process just described) will perform better. Therefore, the key nov-
elty is a quantitative index that ranks traded goods in terms of their implied
productivity. We construct this measure by taking a weighted average of the
per-capita GDPs of the countries exporting a produc t, where the weights re‡ect
the revealed comparative advantage of each country in that product.
1
So for
each good, we generate an associated income/productivity level (which we call
P RODY ). We then construct the income/productivity level that corresponds
to a country’s export basket (which we call EXP Y ), by calculating the export-
weighted average of the P RODY for that country. EXP Y is our measure of
the productivity level associated with a country’s specialization pattern.
While EXP Y is highly correlated with per-capita GDPs, we show that there
1
A very similar index was previously developed by Michaely (1984), whose work we are
happy to acknowledge. We encountered Michaely’s index after the working paper v ersion of
this paper was co mpleted and distributed.
2
are interesting discrepancies. Some high-growth countries such as China and
India have EXP Y levels that are much higher than what would be predicted
based on their income levels. China’s EXP Y , for example, exceeds those of
countries in Latin America with per-capita GDP levels that are a multiple of
that of China. More generally, we …nd that EXP Y is a strong and robust
predictor of s ubs equent e conomic growth, controlling for standard covariates.
We show this result for a recent cross-section as well as for panels that go back
to the early 1960s. The results hold both in instrumental variables s peci…cations
(to control for endogeneity of EXP Y ) and with country …xed e¤ects (to control
for unobserved heterogeneity).
Our approach relates to a number of di¤erent strands in the literature. Re-
cent work in trade theory has emphasized cost un certainty and heterogeneity
at the level of …rms so as to provide a better account of global trade (Bernard
et al. 2003, Melitz and Ottaviano 2005). In contrast to this literature, we focus
on the spillovers in cost information and are interested in the e conomic growth
implications. There is also an empirical literature on the so-called natural re-
source curse, which examines the relationship between specialization in primary
products and economic growth (Sachs and Warner 1995). The rationale for the
natural resource curse is based either on the Dutch disease or on an institu-
tional explanation (Subramanian and Sala-i-Martin 2003). Our approach has
di¤erent micro-foundations than either of these, and yields an empirical exam-
ination that is much more …ne-grained. We work with two datasets consisting
of more than 5,000 and 700 individual commodities each and eschew a simple
primary-manufactured distinction.
The outline of the paper is as follows. We begin in section 2 with a simple
model that develops the key ideas. We then present the empirical analysis in
section 3. We conclude in section 4.
2 A simple model
We are concerned with the determination of the production structure of an
economy in which the standard forces of comparative advantage play some role,
but not the exclusive role. The process of discovering the u nd erlying cost struc-
ture of the economy, which is intrinsically uncertain, contributes a stochastic
dimension to what a county will produce and therefore how rich it will be.
We normalize units of goods such that all goods have an exogenously given
world price p. Each good is identi…ed by a certain prod uc tivity level , which
represents the units of output generated by an investment of given size. We
align these goods on a continuum such that higher-ranked goods entail higher
productivity. The range of goods that an economy is capable of producing
is given by a continu ous interval between 0 and h, i.e., 2 [0; h] (see Figure
1) We capture the role of comparative advantage by assuming that the upper
boundary of this range, h, is an index of the skill- or human-capital level of
the economy. Hence a country with higher h can p roduce goods of higher
productivity ("sophistication").
3