1120 The Methods of Growth Econometrics
24.1 Introduction
In this chapter we will review some of the methods that have been used in the
emerging field of growth econometrics. We define this field as the use of statistical
models to explain variation in growth rates and productivity levels across countries
or regions. This has long been an active field of research, especially since the mid
1980s, but the position of this field within economics remains somewhat uneasy.
There are thoughtful and well-informed observers who regard this form of evidence
as essentially inadmissible.
It is important to understand why this is so, and it is true that the field faces many
problems, not least the constraints imposed by the available data. We will review
these problems in detail, but argue that there are some grounds for optimism. At
least in adopting methods appropriate to these datasets and economic questions,
genuine progress has been made, and we will highlight areas where further progress
seems especially likely. Moreover, the prospects for useful findings seem likely to
improve over time, as more and better data become available, and more is learnt
about the appropriate methods for analyzing such data.
The abiding interest of the study of aggregate productivity, whether in levels or
growth rates, perhaps speaks for itself. Seeking to understand the wealth of nations
is one of the oldest and most important research agendas in the entire discipline. At
the same time, it is also one of the areas in which genuine progress seems hardest
to achieve. The contributions of individual papers can often appear slender. Even
when the study of growth is viewed in terms of a collective, incremental endeavor,
the various papers cannot easily be distilled into a consensus that would meet the
standards of evidence routinely applied in other fields of economics.
Faced with such criticisms, one traditional defence of empirical growth research
is phrased in terms of expected payoffs. Each time an empirical growth paper is writ-
ten, the probability of gaining genuine understanding may be low, but when and
where it does emerge, the payoff to that understanding could be vast. Moreover,
some contributions can be seen as stepping stones in the development of credible
evidence. That gradual process, working towards better methods and more reliable
findings, may take years, but could ultimately have a high payoff. From this per-
spective, even if much of this evidence lacks credibility, the literature is gradually
evolving towards methods and findings that should be taken more seriously.
These arguments are plausible, but rely on an important tacit assumption. They
all depend on the ability of researchers and policy makers to discriminate between
the status of different pieces of evidence – the good, the bad and the ugly – and
this process of discrimination carries many difficulties of its own. The accuracy,
or otherwise, of such judgments plays a key role in the overall development and
intellectual health of any empirical literature. This reinforces the case for building
an understanding of the relevant methods, their strengths and limitations, and the
ways in which the existing literature is often flawed or inconclusive.
Caution will be needed throughout. Rodriguez and Rodrik (2001) begin their
skeptical critique of the evidence on trade policy and growth with an apt quote
from Mark Twain: “It isn’t what we don’t know that kills us. It’s what we know