The second round of debate on free trade and protectionism completes itself with Prof. Lui's publishing his very critical Reply to Prof. Hui in Mingpao a few days ago:
http://www.inmediahk.net/public/article?item_id=110990&group_id=77
History of the Debate. The debate began, as the reader should recall, as an exchange between Prof. Lui and Leung Kai Chi, in Mingpao, respecting the merits of opening the agricultural market in South Korea to foreign producers. The exchange touched, naturally, upon the various arguments for and against free trade in general, especially regarding its effect on alleviating poverty. After that first round of engagement, Prof. Hui made a substantial intervention by supplying many pieces of historical evidence purporting to demonstrate that free trade did not necessarily conduce to higher rates of economic growth; and in various periods, it was in fact associated with lower rates of observed growth, compared, of course, with countries following protectionist policies. Prof. Lui's present rebuttal attacked this observed association, arguing that no causal relationship could ever be demonstrated by amassing historical evidence of this kind; for concommitant factors were not well controlled for.
Structure of the Debate. It is imperative (I think) to disentangle a number of themes in the debate, whose logical relationships with each other are often lost in the heat of engagement. I propose to read the divers contributions as revolving around these four questions:
(1) Is it the case, that free trade causes a higher rate of GDP growth, with all other conditions being equal, and so universally (in the sense that this effect of free trade can be established regardless of the absolute condition of a country, and of its relative position in the world trading system)?
* Imbedded in this question are in fact two sub-questions. Free trade may be shown, for instance, to cause a higher rate of GDP growth, provided that a country be above a certain per capita GDP level. Restricted to the set of rich countries, then, free trade can be said to cause higher economic growth ceteris paribus. But the statement would not be true across all parameter values; and so not universal.
(2) If not, under what circumstances (including both absolute condition and relative position) does free trade cause a higher rate of GDP growth?
* Imbedded in this question is a complex thought, concerning the domestic institutional setting (e.g. industrial policies, distributive policies, etc.) and technological level in which trade is to take place; and the sectoral impact of trade opening. Moreover, GDP, being an aggregate measure of the volume of market transactions, does not reflect how a particular sector fares under a free trade regime.
(3) In light of a comprehensive answer to (2), can it be concluded, that further trade opening is beneficial to South Korea as a whole?
* An answer to this question shall take into account of constraints imposed by economic as well as political bargaining between trading partners. If only economic well-being in the aggregate sense be considered, then perhaps opening the agricultural sector, whereby concessions in non-agricultural sectors might be gained from trading partners, should be adopted by the South Korean government.
(4) To the extent, however, that trade opening might cause some people to win, and some to lose, on what ground can further opening be recommended, seeing that it will also have significant distributive consequences?
* The distributive question can, and should, be separated from the efficiency question, the latter being that respecting the relationship between free trade and economic growth in the aggregate.
A Political Case. If my reading of the structure of the debate is right, then one obvious conclusion is that, two major reactions to free trade (both in the South Korean case specifically, and in the world trading system in general) are very often conflated, by supporters as well as opponents of free trade.
Many commentators who have been sympathetic to the South Korean farmers oppose free trade because of the harm it shall cause to these farmers. Absent suitable governmental intervention - gradual opening, interim subsidy, technological support, etc. - the farmers appear, to these commentators, to suffer a great deal of injustice.
Some commentators, starting with Prof. Hui perhaps, try to draw our attention to the claim that free trade may be injurious, not only on distributive grounds, but also on efficiency grounds: that it can cause in fact a lower rate of economic growth compared with a protectionist regime. In other words, free trade may cause not only a smaller share of the pie for the farmers, but in fact a smaller pie for the whole country.
And yet, it is clearly the case, that even if free trade be recommendable on efficiency grounds (after, say, the claim that it is conducive to higher GDP growth be established), there can still be a political case against certain kinds of trade opening, on distributive grounds.
I call it a political case deliberately, in order to highlight the fact that whether this case be accepted, is something which econmoic science, at least as conceived by Prof. Lui, can never decide. It goes beyond economics, so to speak. The idea is that, the version of economics espoused by Prof. Hui and followers is one which would uphold the strict separation between positive analysis and normative prescription. Methodology, on this view, should be free of ideology. Hence, the said economics can only recommend policies which are strictly a Pareto improvement: no one is worse off, and at least some one is better off.
Economists who are sensible to the extreme hurdle this criterion poses would replace the word "strictly" in the last sentence by "potentially." And that is the Kaldor-Hicks criterion. An obvious attack at adopting such a criterion is, of course, that in reality the kind of compensation that must be effected in order to make a Kaldor-Hicks improvement into a true Pareto improvement - which, we note, is always the goal - is rarely effected in full, if at all. Absent such compensation, a policy recommended on Kaldor-Hicks grounds must have tacitly made a value judgement - who is to be worse off - which Prof. Lui's version of economics must needs forbid.
In brief, taking Prof. Lui's version of economics seriously, namely a value-free scientific enterprise, one can make a strong case, that the Professor would have no ground on which he could recommend the further opening of the agricultural market in South Korea. But of course, he could acknowledge that he had but insinuated either one of two innocuous minor premises into this analysis, viz, either that the South Korean government is supposed to effect a Kaldor-Hicks compensation in some form, or that aggregate economic well-being be accepted as the legitimate concern of the kind of policy analysis at hand. In either case, the analysis must go beyond pure economics.
The Economic Case. Under this head I mean to examine the various arguments respecting the effect of free trade on GDP growth. Prof. Lui's Article has several themes. But the major criticism he made against Xu was, that the evidence put forth by the latter has not isolated the causal effect of free trade on GDP growth by suitably controlling for other factors. To the extent that the criticism is directed to Bairosch, to whom Prof. Hui appealed in his earlier article, I think it has some force. Bairosch argued, for instance, that in the late 19th-century Western Europe, periods of high growth coincided with periods of protectionism. But it is not clear, whether a causal relationship can be established.
It is, unfortunate to say, really difficult to establish conclusively the effect of free trade on economic growth. The problem, chiefly, is that trade opening is often accompanied by other major policy changes, like internal liberalization or institutional reforms; that it is not easy to identify econometrically what role trade opening really played in affecting the growth rate. Worse still, the decision to open up a country for trade is frequently affected by patterned political or economic developments in the country; in other words, the decision is itself endogenous, which makes it even harder to ascertain what causes what. There are just not enough countries in the datasets, and not enough variation thereby, for empirical economists to say confidently whether free trade is good or bad for growth. The safest conclusion, in light of historical evidence as well as econometric analyses, is that the effect is ambiguous. Rodrik's paper must be read as a sincere report of ignorance. Bhagwati, whom Prof. Lui cited enthusiatically, is well-known, in economic circles, for his reluctance to engage with cross-country analysis, believing that it is too unreliable given the state of the art.
Theoretically, one can certainly make a case for trade opening being beneficial for economic growth, under certain assumptions. But these assumptions go, needless to say, to the heart of the
question, What does one mean by free trade?
Amsdem, Wade, and others, have investigated closely the industrial policies followed by regions or countries like Taiwan and Japan. In one sense, these regions or countries engage actively in trade - their economies are very much export-led. But internally, their governments deliberately directed investment to certain key industries, and intervened in various ways (from tax subsidies to closing up the market by non-economic means). Free trade, in these cases, is coupled with intense domestic interventions. A case may be made, that such interventions have successfully harnessed the beneficial effects of free trade to promoting economic growth. The experience of China in the past three decades should also be read in this light. Absent appropriate industrial policies, free trade might be detrimental.
Future Directions of the Debate. In sum, there are two major bones of contention in the present debate, as far as trade opening is concerned:
1. (The economic case) Whether free trade do help GDP growth;
2. (The political case) Whether free trade could be recommended on strictly scientific grounds.
I think one can answer conclusively respecting 2.; but 1. must remain a very murky subject, given the reasons I outlined in the preceding section. Opponents to free trade are advised to distinguish clearly the economic from the political case; but supporters thereof should also acknowledge the assumptions and limits of their arguments.
There can probably be no single law declaring that free trade is universally good; but nor is it always unacceptable to have freer trade. One should be hopeful, that the debate be taken in such a direction, that both the internal and external limits of the economic case - the uncertainty of the conclusion, and the self-imposed boundary of positive analysis - may be illuminated. Further conflation of these two threads, as well as further boasting of intellectual lineage, must be deemed unconstructive.
Y.T.