The return of industrial policy
Robert Wade, LSE, Apr 2011
Neoclassical vision of price system
• Prices in well-functioning price system a
necessary & sufficient condition of efficient
coordination
• Prices accurately reflect relevant
knowledge of constraints & opp’ies.
• Prices direct self-interested actors to own
and societal advantage.
• Goal of policy: to protect system fr
distortions & establish system in DTCs
Larry Summers
• “The laws of economics, it’s often forgotten, are
like the laws of engineering. One set of laws
works everywhere”. Summers subsequently set
out the laws as the “three –ations: privatization,
stabilization, liberalization”.
• These ideas are so accepted as to be beyond
discussion, like “the idea that a huge spending
program is the way to stimulate the economy”.
Gregory Mankiw
• “Adam Smith was right when he said that
‘Little else is required to carry a state to
the highest degree of opulence from the
lowest barbarism but peace, easy taxes
and a tolerable administration of justice’” .
•
(The Wall Street Journal, 3 January 2006).
Gary Becker
• “The best industrial policy is none at all”,
whether for advanced or developing
countries.
•
(Business Week, August 25, 1985).
More
• World Bank: “For every Korea there are 100 failures.
Who would you put your money on?” (World Bank
economist speaking to Wade, 2010)
• “The government should be providing conditions that
help all businesses – namely, effective infrastructure, a
skilled workforce and better planning. We should make
no attempt to pick winners – whether individual
companies, specific sectors, or manufacturing as a
whole” (Tim Leunig, Reader in Economic History, London
School of Economics, 2010).
•
World Bank
• Country Policy & Institutional Assessment
formula crystallizes neoclassical /
Washington Consensus policy norms.
Stylized facts against neoclass
view
• C’ies (DTCs) tend to become what they X,
mainly b/c faster learning. Costs of not learning
in one period multiplied by costs of opp’ies
foregone in later periods.
• X success not come automatically fr good
endowments, instns, foreign techy
• Xs of a ctry “arbitrary” within broad categories:
eg Pakistan, BD
• Exploring new products full of uncertainty,
requires continuing learning.
Stylized facts (contd)
• Imported T needs (costly) adaptation
• Costs of discovering viable new products
& adapting foreign T generally not
recoverable by pioneers
• Therefore big problems of appropriation &
coordination in industrial diversification &
upgrading.
Policy response: IP
• IP = array of public supports: some sectorneutral, some sector-specific.
• Protection, credit subsidies, tax subsidies,
extension service, VC, public support for
R&D, fostering of networks of collaborating
firms
• Distinguish: hard & soft; micro & meso IP
East Asia IP
• A lot of hard IP (protection, subsidies,
SOEs, central planning implemented
through price system)
• Also a lot of soft-meso IP
• Eg Industrial Devt Bureau in Taiwan
• More recently, science parks for post-WTO
world
East Asia push into science
• Surge of “fraud in science” in Jpn, SK, China.
Why?
• (1) State has deployed IP tools in way wh
creates perverse incentives.
Grants, prizes given for publication in Western
journals (not exports). Encourages falsification
• (2) Authoritarian lab culture (German model)
provides little internal discipline on lab bosses
US IP: ineffective?
• “There is no serious American industrial policy;
this is left to the post-war powerhouses of the
US economy, the large corporations. Much of
this [industrial policy failure] is due to the radical
separation of powers enshrined in the US
constitution. A coordinated political economy
cannot easily be run by a President and his
cabinet, two Houses of Congress, a Supreme
Court and fifty ‘states’ (which are also
fragmented by the same separation of powers) –
especially when they belong to different political
parties”. (Michael Mann, 1997)
US IP: ineffective
• General argument: IP (targeting specific
industries) ineffective in the context of
• (1) a federal structure,
• (2) strong separation of powers between
legislature, executive and judiciary.
US IP: corrects network failures
• “The federal government has been pursuing
industrial policy within decentralized political
institutions for well over a generation…
American industrial policies go beyond
preservation of market competition,
maintenance of macro stability, and provision of
public goods to address firm-specific needs in a
host of different ways and through a variety of
different agencies”. (Schrank & Whitford, 2009)
•
US IP
• Operates at “meso” level.
• Comprised of programs run by national, state,
local govts, themselves uncoordinated.
• National level: NIH, ARPA, NIST, SBA, etc.
• Eg NIST organizes Manufacturing Extension
Partnerships, to link firms in specific locations
• SEMATECH as classic example.
US IP
• “Below the ideological surface, a powerful ‘jerrybuilt’ substrate has emerged of federal, state
and local government innovation support
programs each filling gaps in the other”.
• An official involved in these programs said, “We
definitely see the programs as a de facto
industrial policy, but we cannot use that term, so
we usually call it R&D policy”.
•
US IP: effective?
• Difft to evaluate effects of scattered
programs.
• Programs have developed valuable
products & processes
• Able to withdraw benefits from “losers”
• Publicly-supported networks have lower
rate of breakdown than others.
WB & return of IP
• WB sanctioned IP in 1950s-70s.
• Then devt economics merged with
mainstream neoclassical  “monoeconomics” & Washington Consensus
• From 2008 Justin Lin, chief economist,
Chinese, PhD Uni Chicago, nudging
thinking towards “development as
production” (not just “as consumption”) &
towards IP.
WB & IP
• Lin conceives economy as vast,
continuously improving Toyota-like
production system, with constraints &
opp’ies being discovered over time.
Learning & self-discovery central
processes.
• 6 steps for strategic government
intervention.
WB & IP
• Finance & Private Sector VP piloting IP in
several countries, including Kenya. Called
“competitive partnerships initiatives”.
• Many economists in WB strongly resist
Lin’s initiative.
Conclusion
• J. K. Galbraith:“The conventional wisdom”
gives way not so much to new ideas as to
“the massive onslaught of circumstances
with which [it] cannot contend”.
• GFC, uprisings in MENA (also about jobs),
“middle income trap”, and more, combine
to make massive onslaught.
• Opening way to rethinking of IP.