This paper presents arguments that support the call for selective
industrial policy, drawing largely from the lessons of the East Asia.
Contrary to mainstream economic wisdom, policy interventions to promote
industries are necessary and cannot simply be generic. As market
failures are idiosyncratic, one needs a selective and targeted approach
in correcting these failures.

The author is Trade and Industrial Policy Analyst of Action for
Economic Reforms, and Instructor at the Department of Economics, School
of Social Sciences, Ateneo de Manila University.

After decades of continued economic liberalization and almost blind
faith in the free market, the Philippines finds itself today in a
quandary. With local industries exposed to harsh competition from the
global market, and emerging countries like China, India and Vietnam
eating away our competitive advantage in cheap labor, domestic
production is at a decline and jobs are constantly being lost.
Moreover, while overall export figures have been exhibiting signs of
robustness in its growth, a closer look will show that at best we have
but a small handful of export winners, overwhelmingly dominated by one
product: electronics, making electronics assembly and testing the
number one export activity in the country today. The cataclysm of
atrophied industries with shallow roots in the local economy has
finally come.

A large reason for such is the reluctance and arguably the inability of
our government to formulate and conduct an industrial promotion
strategy. Haunted by the ghosts of import substitution strategies of
earlier decades, many are uncomfortable with or even downright
disapproving of "picking winners." However, as we look to our more
successful neighbors in East Asia for pointers, we find that industrial
policy was indeed a crucial ingredient to their recipe of success.

It is now widely accepted that there was no unique "East Asian Model"
of industrialization success. Each of the tigers of East Asia had
different objectives of industrialization, and had different recipes
for each of these objectives. What was common among the success
stories, however, was clear export orientation, good human capital, and
very strong regional spillovers. A most crucial point to emphasize
though is that for none of the tigers, not even the least
interventionist, was simply "getting the prices right" a sufficient
explanation of industrial success.

Mainstream wisdom tells us of four instances when policy interventions
necessary to restore efficiency: 1) deficient capital markets, 2) lumpy
investments, 3) imperfect appropriable firm-level investments,
especially in knowledge and skills, and 4) inability of individual
investors to act rationally when there are technologically
interdependent investments. While the standard prescription of policy
interventions to correct generic market failures tries to address the
four aforementioned market failures, these generic remedies, such as
making capital markets more efficient via generic prudential
regulation, may be insufficient to banish the most important
distortions that impede industrial development. With a clearer view,
however, of the verity that the process by which firms in developing
countries become efficient is one that is slow, costly, and risky, we
find other instances when, in fact, interventions become necessary and
then spell the success or failure of industrial deepening.

TECHNOLOGY & LEARNING. The neo-liberal depiction of industrial
development assumes that technology is freely available via a menu from
which firms choose according to factor and product prices. Technology
is then absorbed with no cost or risk and used at best practice levels.
Any lag in efficiency can at most be only for a brief period in which
scale economies are fully realized or costs fall in a 'learning by
doing' process. The basic premise is that intervention cannot improve
on this outcome, and will only lead to inefficiency in the choice and
use of technology.

In reality, however, technology has many tacit elements and cannot be
transferred like a physical product. Its use requires the recipient to
invest in new skills, technical information, organizational methods and
external linkages. The process continues over time and is
technology-specific. As such, while firms are undergoing the process of
adapting to new technologies and learning its efficient use, good firms
will hardly be distinguishable from bad firms. The free market, if
tasked as a principal with purging the industry of bad firms, is then
confronted with the adverse selection problem. It is precisely because
of this that intervention becomes necessary.
Perhaps a more important point that needs to be made is that the
process of learning in developing countries may be distorted and
curtailed if firms do not know how to go about learning, how long it
will take, how much it will cost, and where to look for information and
skills: Joseph Stiglitz's 'learning to learn' process.

Dropping the assumptions of perfect information on and transferability
of technology therefore poses market failures in resource allocation
that go beyond the usual learning externalities. In free markets, firms
will tend to under-invest in technologies that have costly, risky, and
prolonged learning periods, even whilst they may be socially desirable.
Since it will mostly be NEW technologies that will have costly, risky,
and prolonged learning periods, under-investment will stunt the process
of technological deepening: the process of entering more complex
technologies, increasing local content, or undertaking more demanding
technological tasks (e.g. from final assembly and testing technology to
design and development activity). Supportive interventions,
custom-designed for technology-specific costs and risks, may be able to
persuade firms to undertake such risks and help them overcome learning

SCALE ECONOMY. Scale economy has an important implication for the cost
competitiveness of a country's industry. Economists may have
traditionally debated on whether the social cost from monopoly is 1% or
2% of total output, but in industries with significant scale economy,
choosing a suboptimal scale of capacity can often mean 30-50%
differences in unit costs. For this reason, the East Asian governments
used measures such as industrial licensing, government procurement,
export requirements, and subsidies in order to ensure that factories
would be built at scales which are not too much below (and hopefully
above) the minimum efficient scale. Their attitude was that
monopolistic firms producing at optimal scale would be much less of a
drag to the economy than "competitive" firms all producing at
suboptimal scales.

While it may sound less fancy than coordinating complementary
investments and giving industrial development a "big push", ensuring
the achievement of scale economy in key industries was in practice
probably a much more important aspect of East Asian industrial policy
than the former.

COORDINATING COMPETING INVESTMENTS. Going beyond the openly accepted
wisdom of coordinating of complementary investments particularly to
achieve economies of scale, we proceed into an under-explored rationale
for industrial policy – the genius of managed competition.
Oligopolistic competition that characterizes many modern industries
with significant scale economy often leads to excess capacity, unless
there is a coordination of investment activities across competing
firms. When excess capacity damages profits, firms may be forced to
scrap their assets. In a world without transactions costs and asset
specificity, asset scrapping can be a useful and costless way of
rearranging property rights. As the real world is not such, specific
assets involved in this process have to be scrapped or re-allocated to
alternative uses that can create much less value out of the assets
concerned, thus incurring a social cost. If the emergence of excess
capacity can be prevented through ex ante coordination, then the social
cost may be avoided.

Indeed, such coordination has been one of the most important components
in the industrial policy regimes of the East Asian economies, as
manifested in their continuous concern for "excessive" or "wasteful"
competition, and their attempts to minimize investments redundancy
through mechanisms like industrial licensing and investment cartels.

certainly a large element of truth in the view that industrial policy
in East Asia concentrated on "picking winners" rather than "protecting
losers", and one can even argue that East Asia did not "pick winners,"
but rather created them. However, protective industrial policies were
also widespread in East Asia, if less so than in other countries. And
in light of the long history of the import-substitution strategy
persecution, we need to go deeper if we are to understand why
protective industrial policy in East Asia did not end up blocking
structural change as in other countries.
One function of protective industrial policy was the more
short-term-oriented one of providing "social insurance" to firms which
were in temporary difficulty but could not borrow their way out of it
due to capital market imperfections. The practice can be justified
again in terms of asset specificity, in that it will be socially
inefficient to scrap specific assets in the face of a temporary
setback, if the net present values of their future income streams are
larger than the costs of supports needed to keep them employed in their
current uses.

Another function was the more long-term-oriented one of promoting
structural change. When an industry is in need of a large-scale
adjustment, those who had made specific investments in the industry
face the situation where their next best option is a total scrapping of
their assets and therefore a drastic reduction in their income. Unless
there is a mechanism that allows them an acceptable level of income
during the transition period when they run down their existing assets
and re-tool themselves, they will have an incentive to resist the
change by political means. In such a situation, measures to reduce the
impacts of adjustment on the owners of specific assets can accelerate,
rather than slow down, structural change by reducing the political
resistance to the change, if they also provide incentives for

What distinguishes these policies from similar policies in other
countries is that they were "forward-looking" in the sense that they
made it explicit that the aim of the protection was not to preserve the
industries concerned but to phase them out or to technologically
upgrade them. Perhaps more importantly, they also had well-specified
performance targets for the beneficiaries, thus preventing the policies
from turning into "nursing homes" for declining industries.

IDIOSYNCRACY AND FDI. A most important feature of market failures is
their idiosyncratic character. Market failures are particularly binding
for local enterprises, more so for new small and medium enterprises
entering modern industry. Foreign direct investments, most especially
those embodied in transnational corporations face fewer failures in
developing countries, largely due to the internalization of many
intermediate. This points to why TNCs can be a powerful engine to
launch industrialization in developing countries. Passive open door
policies, as we have here in the Philippines, may however not be the
best means to harness the benefits of FDI.

For one, passive liberal policy may only attract TNCs into areas of
static comparative advantage. Selective and functional interventions
can guide FDI into dynamic and more complex activities. Second, TNCs
tend to transfer operating know-how rather than complex technological
functions to developing host countries. However, as countries
industrialize it becomes increasingly important to develop R&D
capabilities, to keep abreast of and absorb technologies, deepen
industry and reduce the cost of importing technology. Again, there is a
need to intervene to induce an upgrading of TNC technological activity,
or to restrict foreign entry as local firms have to establish their own
innovative base.

A SUMMARY. Theory thus provides logical grounds for interventions to
selectively promote industrial development. In general, market failures
can occur in three spheres of firm activity: within firms, in
inter-firm relations and in factor markets. These failures are
inter-related, and their remedy calls for a range of selective and
functional interventions.