M anaging competition: Allocative efficiency vs

productivity

If scale economy is taken seriously by the policy-makers of a late-developing country with small dom estic m arkets and thus the number of producers in each industry is restricted, an obvious prob­

lem that its governm ent faces is the existence o f oligopolistic, or even monopolistic, markets. While there is no one-to-one relationship be­

tween the num ber of firms in an industry and the intensity o f com pe­

tition between them (in fact East Asia provides many examples of fierce competition in oligopolistic markets), oligopolistic markets pose a greater challenge to the designers of competition policy. And in countries with serious infant-industry programmes, the resulting ab­

sence o f competition from imports makes competition policy even more challenging.

In the mainstream competition policy discourse, the allocative inefficiencies that the market power possessed by oligopolistic firms creates are regarded as the most serious problem facing the com peti­

tion policy-makers, and thus a vigorous Anglo-Saxon (or rather Ameri­

can) style anti-trust policy is recommended to deal with the problem.

Thus, even many neoliberal supporters of far-reaching deregulation would concede that anti-trust policy is one area where government activism is legitimate and necessary (Chang, 1997c). Restricting the abuse o f power by dominant firms through anti-trust action has cer­

tainly been an issue in East Asia, but it was only a minor part o f its

The East Asian Mo delof Economic Policy 4 1

competition policy, and its focuses were elsewhere.

First of all, as we pointed out earlier, the East Asian govern­

ments have deliberately created oligopolistic, or even monopolistic, market structures, in order to exploit scale economy as much as pos­

sible, if that was regarded as important in the particular industry con­

cerned. M ainstream economists frequently ignore this point, but many estimates o f the allocative inefficiencies arising from ‘non-com peti­

tive’ markets suggest only modest figures (1-2% usually), whereas the cost increase that follows from sub-optimal scale o f production is known to be very significant in many industries. To put it differently, scale-economy considerations were given clear priority over market- power considerations in the East Asian competition policy regime.

Secondly, the East Asian governments have been deeply con­

cerned with ‘excessive’, ‘wasteful’, or ‘destructive’ competition. The notion of ‘excessive’ competition has often been dism issed as a no­

tion based on ‘irrational’ fears of competition by ignorant bureau­

crats, but it makes perfect sense once we acknowledge the impor­

tance o f dedicated physical and human assets in m odem industries (or what W illiamson would call ‘specific assets’; see W illiamson, 1985). Given the existence o f such ‘specific’ assets, any failed project that follows an ‘excessive’ entry (compared to what is warranted by the dem and condition) leads to a ‘w aste’ o f resources in the sense that the ‘specific’ assets em ployed in the project may not be transferred to other activities without significant losses in their economic values (for further details, see Chang, 1994a: Chapter 3). As a result, the East Asian governments have tried to coordinate investments ex ante in order to prevent excess entry, but when excess entry materialised for whatever reason (e.g., erroneous projection, sudden changes in world m arket conditions, some firms defying the government plan, etc.), they organised (explicit and implicit) recession cartels, negoti­

ated capacity-scrapping arrangem ents, or even forced m erger and market-sharing programmes, to reduce the ‘wastes’ from excessive competition.

Thirdly, the East Asian governments have willingly suspended anti-trust actions and allowed collusive behaviours by firms, when it was thought that a suspension of competition was necessary for rais­

ing productivity (for a more detailed theoretical account, see Chang,

4 2 The East Asian Development Experience: The Miracle, the Crisisandthe Future

1994a: Chapter 3 ).18 For example, the Japanese state has frequently allowed, and often took initiatives in organising, various types of car­

tels - to weather recession, to coordinate capacity expansion, to en­

courage joint R&D, to allow collusion in export markets, to promote technological upgrading by small firms, to phase out declining indus­

tries, just to name a few (M agaziner and Hout, 1980: Dore, 1986). In Korea, there existed no anti-trust legislation until 1981. and even af­

ter that, collusive behaviours were explicitly allowed in ‘promising industries’ which needed to ‘increase R&D, improve quality, and at­

tain efficient production scale’, and in ‘declining industries’ which needed to ‘scale down their capacities’ (the quotes are from the 6th Five Year Plan (1987-91) document: for further details, see Chang, 1993). In Taiwan, where many large firms have been public enter­

prises, anti-trust policy has had a different dynamic, but the Taiwan­

ese state did not hesitate to promote mergers if they were deemed necessary for exploiting scale economy (see Wade, 1990: pp. 186-7).

The above-mentioned policies frequently ossified the cartels and re­

sulted in industrial stagnation in other countries, but the East Asian countries avoided such danger, because the suspension o f competi­

tion was regarded as a temporary measure to achieve relatively well- specified goals that were deemed necessary for productivity enhance­

ment (although it could sometimes last quite long), and thus did not result in a general suspension o f competition.

Many commentators have criticised the ‘lax’ attitude by the East Asian governments towards anti-trust policy as evidence of the preva­

lence of corrupt collusion between the state and big business - or

‘crony capitalism ’, to use the currently popular terminology. How­

ever, the above discussion suggests that it resulted more from the (non-neoclassical) view o f competition held by the East Asian policy­

18 The German anti-trust legislation, which has served as a model for the Japanese legislation since the latter's 1953 amendment away from the Anglo-Saxon elements imposed earlier by the American Occupation Authority, also provides many similar

‘escape' clauses to cartel and other collusive behaviours, especially by small firms, when they are related to aims like ‘rationalisation’, ‘specialisation’ (i.e., negotiated market segmentation), joint export activities, and structural adjustments (Shin,

1994: pp. 343-355).

The East Asian Mo delof Economic Policy 4 3

makers that is very different from the one held by mainstream econo­

m ists.11' The emphasis of their (broadly defined) competition policy was on raising productivity in sunrise industries and managing smooth resource transfers out of sunset industries without the ‘w astes’ that could result from unrestrained competition. And if it was necessary for those purposes, these governments were quite willing to restrict entry and capacity expansion (thus creating oligopolistic m arkets) and allow collusive behaviours among the firms, although these actions were taken with clearly defined productivity-related goals and with a relatively clear time limit.20

5.4 Technology policy: Controlling the inflows and raising