An Economics to Fit the Facts
CAMBRIDGE – The
economics profession was arguably the first casualty of the 2008-2009
global financial crisis. After all, its practitioners failed to
anticipate the calamity, and many appeared unable to say anything useful
when the time came to formulate a response. But, as with the global
economy, there is reason to hope that the discipline is on the mend.
Mainstream economic
models were discredited by the crisis because they simply did not admit
of its possibility. And training that prioritized technique over
intuition and theoretical elegance over real-world relevance did not
prepare economists to provide the kind of practical policy advice needed
in exceptional circumstances.
Some argue that the
solution is to return to the simpler economic models of the past, which
yielded policy prescriptions that evidently sufficed to prevent
comparable crises. Others insist that, on the contrary, effective
policies today require increasingly complex models that can more fully
capture the chaotic dynamics of the twenty-first-century economy.
This debate misses
the point. Simple models have their place. They are useful for making
the straightforward but counterintuitive points that distinguish
macroeconomics from other fields of economic analysis. We rely on such
models to explain, for example “the paradox of thrift,” whereby
individual decisions to increase saving can, by depressing spending and
output, result in the population as a whole saving less.
At the same time, complex models can be useful for illustrating special cases and reminding us that the world is a messy place.
Neither class of
models is useful, however, for providing the practical advice that
policymakers need in a crisis. Both are too stylized to be of use when
analyzed in the abstract. To make them useful, evidence is required.
In fact, largely
unbeknownst to the protagonists in this debate over models, an
evidentiary revolution is already underway. While older members of the
economics establishment continue to debate the merits of competing
analytical frameworks, younger economists are bringing to bear important
new evidence about how the economy operates.
For example, a
longstanding debate in macroeconomics has focused on how prices respond
to news about the economy, and whether companies pass through to
consumers changes in import prices that result from exchange-rate
movements. Today, “big data” promises to enhance our ability to
understand and even predict such responses. One application of this
approach, the Billion Prices Project at MIT, uses billions of observations from online retail websites to track inflation.
A second approach relies not on big data but on new
data. Economists are using automated information-retrieval routines, or
“bots,” to scrape bits of novel information about economic decisions
from the World Wide Web. Websites where commercial artists submit
designs for company logos and freelance editors offer services for
authors promise to shed new light on issues like the determinants of
innovation.
A third approach employs historical evidence. A number of commentators have observed
that the global financial crisis was good for economic history, because
it directed attention to previous crises and to the insights that could
be gleaned from studying them. In fact, economic history never stopped
playing its role in economic research. But the financial crisis served
as a useful reminder that history is replete with similar events and
with evidence concerning which policy responses work.
This realization then
dovetailed with the availability of more extensive historical data on
the operation of the economy. Economic historians have long gathered
information from parish registers, population censuses, and corporate
financial statements. But working in dusty archives has become easier
with the advent of digital photography, mechanical character
recognition, and remote data-entry services. Larger data sets are
enabling economic historians to address key questions – for example, how
aggregate economic conditions affect labor-force participation
decisions in different times and places – more effectively than ever
before.
This reference to
different times and places points to the fourth and final focus of the
new empirical research: institutions. Macroeconomic models have tended
to neglect the role of institutions, ranging from trade unions and
employer associations to property-rights regimes and mechanisms for
redistribution. Taking them seriously means considering long historical
intervals, because institutions change slowly and vary significantly
only over time. Renewed attention to history is thus allowing economists
to consider more systematically the role of institutions in
macroeconomic outcomes.
These developments
amount to a sea change in economics. As recently as a couple of decades
ago, empirical analysis was informed by relatively small and limited
data sets. To be sure, analytical frameworks are still needed to help
make sense of the data. But now there is reason to hope that, in the
future, economists’ conclusions and policy advice will be shaped not by
those frameworks’ elegance, but by their ability to fit the facts.
Read more at http://www.project-syndicate.org/commentary/big-data-economics-revolution-by-barry-eichengreen-2015-05#L49AF3QzRqGyLlEk.99
