In a comment on my recent post on the contribution of trade liberalisation to Asia’s success, Jim takes issue with my contention that
China and India are very powerful examples of the benefits of
liberalisation: before they started their reforms, growth was slow, but
as they have opened up their economies, growth has accelerated.
He argues that growth speeded up before they opened up, and that their growth since they opened up is not due to freer trade.
Jim makes three specific counter-claims:
1. ‘Both India and China grew strongly through the 1980s, well before any significant trade liberalisation.’
2. ‘While they did eventually liberalise significantly and kept
growing, the evidence that the trade opening caused the growth is
pretty thin.’
3. ‘China is at best middling in terms of trade openness in comparison to
other developing countries.’
I’ll address each in turn.
1. It is not true that China and, to a lesser extent, India did not undertake any significant trade liberalisation in the 1980s.
In the mid-1970s, all trade was centrally controlled in both countries. But starting from December 1978, China began to open up. In the 1980s:
- The central government’s monopoly on foreign trade was broken.
- A large number of foreign trade corporations were established (there were over 5,000 by 1988), with trading rights also granted to big companies.
- Special economic zones and open cities were established.
- Joint ventures with 25% foreign investment were permitted, and these were granted the right to trade directly.
- Companies were allowed to keep some of the foreign currency earned through trade.
- The remninbi’s artificially high official rate was devalued from 1.5 to the dollar in 1979 to 8.7 in 1991.
In short, in place of the central government’s absolute control over foreign trade, a large number of companies were permitted to trade, with trade regulated through import licences and tariffs instead. Free trade it wasn’t, but it was certainly a much more liberal regime than before.
India’s trade liberalisation in the 1980s was more modest – by 1990 some 30% of imports, mainly capital and intermediate
goods, could be imported without specific licences and import quotas
were replaced with less harmful, albeit very high, tariffs – but then so too was its economic growth.
Whereas China’s economy grew by an average of over 9% a year in both the 1980s and the 1990s (subject to the caveat that applies to all official Chinese statistics), India’s managed only 4.8% in 1981-88.
That was an improvement on the wretched 3.2% recorded in 1965-81, but still pretty poor, especially given the rapid rise in India’s population, which meant that living standards rose much more slowly than GDP.
True, India’s economy boomed between 1988 and 1991 – but this growth spurt proved unsustainable and ended in financial crisis. Since the economic reforms in 1991, which ended import licensing and slashed tariffs, India’s growth has averaged 5.8%.
As Arvind Panagariya quite rightly points out:
Bradford DeLong and Dani Rodrik have argued that
reforms in India cannot be credited with higher growth because the
growth rate crossed the 5 percent mark in the 1980s, well before the
launch of the July 1991 reforms. This is a wrong reading of the Indian
experience for two reasons. First, liberalization was already under way
during the 1980s and played a crucial role in stimulating growth during
that decade. Second, growth in the 1980s was fragile and unsustainable.
The more systematic and systemic reforms of the 1990s… gave rise to more sustainable growth.
In short, Jim’s claim is incorrect. A more accurate summary of what happened in the 1980s is: "China made big steps to liberalise its trade and was rewarded with much faster economic growth, while India opened up less and enjoyed only a modest uptick in growth."
2. Trade liberalisation is clearly not the only reform that
China and (to a lesser extent) India have undertaken, and it is not
responsible for all their growth. But it has certainly boosted both
China’s and India’s rate of growth.
The Ravallion/Chen article that Jim mentions says nothing about the impact
of trade liberalisation on growth: it is concerned with poverty and
inequality. And its findings do not contradict my contention: although
some people may lose from trade liberalisation in the short term, the
long-term productivity gains from freeing trade, which boosts
competition and innovation, raise long-term growth, thus reducing poverty.
There is a mountain of evidence to support the belief that freeing trade promotes economic growth in developing countries. For instance:
- Studies of nine countries – Chile, Colombia, Egypt, Ghana, Israel, Korea, the Philippines, Turkey, and, yes, India – directed by Anne Krueger and Jagdish Bhagwati for the National Bureau of Economic Research in the late 1970s showed that liberalising trade led to faster economic growth.
- These findings were confirmed by studies of nineteen countries – Argentina, Brazil, Chile, Columbia, Greece, Indonesia, Israel, Korea, New Zealand, Pakistan, Peru, the Philippines, Portugal, Singapore, Spain, Sri Lanka, Turkey, Uruguay, and Yugoslavia – over four decades conducted in the early 1990s by Demetris Papageorgiou, Michael Michaely, and Armeane Choksi of the World Bank.
Just look at China:
- Since 1978, China has gone from a system where trade was wholly controlled by the central government to one where a huge number of private companies engage in foreign trade, import licenses have largely been abolished, industrial tariffs have fallen to single figures and service sectors are being opened up too.
- The volume of China’s trade has risen 70-fold, trade’s share in gross domestic product has risen five-fold and the country’s share in world trade has jumped from 0.8% to 7.7%.
- Over the same period, Chinese living standards, as measured by GDP per person at purchasing power parity, have risen five-fold – and the country has witnessed the fastest fall in poverty ever recorded.
Not connected? Hardly. Indeed, Harm Zebregs finds that foreign direct investment alone boosted GDP growth by 3 percentage points a year in the 1990s, even though it accounted for only 5 percent of GDP.
3. Whatever measure of openness you use, China is more open than most developing countries.
Its weighted average industrial tariff is less than 6%, compared with a developing-country average of 8% (bound) and 12.5% (applied).
Imports account for over a quarter of GDP, a huge figure for a such a large and populous economy.
So is China a fully open economy? Clearly not. Is it much more open than before? Yes. Is it now more open than most developing countries? Most certainly.
So I repeat:
China and India are very powerful examples of the benefits of
liberalisation: before they started their reforms, growth was slow, but
as they have opened up their economies, growth has accelerated.
Philippe, thank you for such a considered response.
“In the mid-1970s, all trade was centrally controlled in both countries. But starting from December 1978, China began to open up.”
What you’re describing is a move from a situation of total central control to one where both countries were ‘only’ extremely protectionist. If you want to call that ‘significant liberalisation’, fine, but I think you’re stretching the term. It also begs the question of whether it is low tariffs (i.e. the level) you think cause higher growth or just tariff lowering (i.e. the trend). If it’s the first, then trade does not explain China (and India’s – see below) high growth in the 1980s, since both were highly protectionist at the time, but if it’s the second then trade still can’t explain India’s growth in the 1980s OR why so many countries that did lower tariffs significantly in the 1980s and 1990s actually stagnated in the period, exactly contrary to the predictions of free traders. This kind of confusion is unfortunately rampant in this subject: for example, in Dollar and Kraay’s famous “Trade, Growth, and Poverty”, ““globalizers” selected on the basis of reductions in average tariffs from the 1985-89 period to the 1995-97 actually had higher levels of average tariffs than “non-globalizers” in both the 1980s and 1990s”.(1)
“India’s trade liberalisation in the 1980s was more modest … but then so too was its economic growth … Since the economic reforms in 1991, which ended import licensing and slashed tariffs, India’s growth has averaged 5.8%.”
I don’t know what data you’re using here but according to the World Bank data (2) India’s per capita GDP growth in 1980-1989 was 3.7% per annum, virtually identical to that in 1999. And Rodrik and Subramanian (3) cite evidence to the effect that output per worker and total factor productivity actually grew faster in the 1980s. India clearly began to thrive well before significant trade liberalisation.
“The Ravallion/Chen article that Jim mentions says nothing about the impact of trade liberalisation on growth: it is concerned with poverty and inequality.”
Correct me if I’m wrong, but haven’t you argued before that freer trade was a major cause of China’s immense poverty reduction in the 1980s and 1990s? If you’re now agreeing with Ravallion and Chen that it was not, then that’s fine by me. But all that poverty reduction – which you seem to be saying wasn’t caused by freer trade – must obviously have been reflected in higher growth in the 1980s, in which case the part of growth you are able to ascribe to freer trade must be substantially lower.
“There is a mountain of evidence to support the belief that freeing trade promotes economic growth in developing countries.”
No, there really isn’t. The study from the 1970s you cite is not only a tiny sample but an ancient one, and seriously undermined by the record of the 1980s and 1990s, which saw a lot of trade liberalisation but – outside of India and East Asia – precious little poverty reduction. Similarly, that from the 1990s looks at a small sample of middle-income countries. The reference to FDI is of little relevance.
I think you should spend some time reading more recent, more comprehensive and more robust evidence, of the kind I cited in my first comment. Here they are again, with a few more besides:
http://www.pitt.edu/~ripoll/files/rstat-round3.pdf
http://www.cer.ethz.ch/resec/teaching/seminar_aussenwirtschaft_wt_04_05/yanikkaya_JDE.pdf
http://ksghome.harvard.edu/~drodrik/Rodrik%20on%20Dollar-Kraay.PDF
http://www.maketradefair.com/assets/english/finalDKcritique.pdf
http://hdr.undp.org/docs/publications/background_papers/2005/HDR2005_Samman_Emma_22.pdf
“Whatever measure of openness you use, China is more open than most developing countries.”
Apart from the measures of openness from the IMF paper I cited in my first comment. You cite one measure of trade policy, ie average industrial tariffs.
“Imports account for over a quarter of GDP, a huge figure for a such a large and populous economy.”
That is an outcome measure and not an indicator of policy. It’s common knowledge that policy is but one and in many cases a weak determinant of trade patterns.
(1) http://www.intermonoxfam.org/cms/HTML/espanol/818/Critica_Dollar_Kraay.pdf
(2) The ‘Macro Time Series’ here: http://www.nyu.edu/fas/institute/dri/global%20development%20network%20growth%20database.htm
(3) Table 1 here: http://www.imf.org/external/pubs/ft/wp/2004/wp0477.pdf
Freeing trade has indeed boosted growth for China and for India, and for a lot of other countries for that matter. A lot of reform was indeed undertaken by the 1970’s and 80’s, albeit in varying degrees, but still the growth level that was seen in that period was nothing compared to the growth that can be seen now for both countries. The main difference is the degree of liberalization that took place. Interestingly enough, the further a country opens up the faster the pace of reform, the faster the growth, the faster the need to open up, and so forth.
Evidence old and new would tell the same story, freer trade helps poorer countries. A distinction could be made with regard to national growth and the poverty levels being reduced, with a lot of commentary indicating that the latter depends highly on the internal infrastructure of that country and not merely on free trade. Admittedly, a lot of free trade advocates indicated before that free trade, by its own, could help solve a lot of a country’s ills, which apparently now is being understood to be wrong. Nevertheless, the rate of poverty reduction is certainly faster, despite a country’s faulty internal structure which could mitigate the effects of trade on poverty, if a country’s economy is opened up.
Finally, it is widely recognized other factors are indeed important to a country bettering its economy, its growth, with such factors to include stability of contracts, mitigated corruption levels, business credit, judicial effectiveness, etc. However, these factors should be taken not of the same level as opening up of trade but rather those factors upon which depends the level of effectiveness that opening up of trade can bring. To say that free trade is of little or no effect is I believe inaccurate. Freeing trade will indeed help a country but the level of which that can happen is dependent on that country.
Apologies, just noticed a potentially misleading typo in my comment above. When I wrote “India’s per capita GDP growth in 1980-1989 was 3.7% per annum, virtually identical to that in 1999”, that should have been “virtually identical to that in 1990-1999”.
Dear Jim
I must admit that I’m astonished that you do not consider China’s liberalisation in the late 1970s and 1980s “significant”.
Within the space of a decade, China went from being a centrally planned, virtually closed Communist economy – a bit like North Korea is today – to being a partly capitalist, partly open economy, with far more liberal conditions applying in the “special economic zones” (which not surprisingly recorded the fastest economic growth).
In 1978 the central government alone dictated what was imported; a decade later, import decisions were made by a variety of companies, responding to local conditions and acting on price signals.
In effect, China shifted from a system where import quotas were fixed in the five-year plan – and therefore where the tariff on imports not in the plan was infinite (they were banned at any price) – to one where the average industrial tariff was a bit over 40%.
That is a dramatic liberalisation by any standards.
I’m also somewhat perplexed by your question about whether I think tariff lowering (trade liberalisation) or low tariffs (reasonably free trade) lead to higher growth. The answer, of course, is both.
Lowering trade barriers produces one-off (misleadingly known as “static”) gains as countries specialise in what they do best and reap economies of scale. Specialisation is equivalent to a productivity increase: instead of making a particular good, an economy can obtain more of it, indirectly, by exporting something else.
The gains from shifting from a virtually closed economy to a partly open one can be huge – since trying to produce everything domestically is extremely inefficient – and lead to very fast growth as the economy adjusts.
Reasonably free trade also leads to sustained, long-term (known as “dynamic”) gains: increased foreign competition forces companies to continually innovate and become more productive, which boosts economic growth year after year.
Last but not least, trade boosts productivity through imports of capital and intermediate goods that embody superior foreign technology, for instance, high-tech car-making machinery.
There is thus no difficulty in explaining how trade liberalisation helped China grow so fast in the 1980s and why India’s more limited efforts produced much smaller gains.
As to why some countries that have liberalised their trade failed to see a boost to growth, weaknesses in the domestic economy may have prevented them reaping the new opportunities or other factors may have offset the pro-growth impact of liberalisation.
For instance, a country may lack adequate road and port facilities to enable farmers to export their produce, corrupt customs officials may prevent importers from seeing the benefit of lower tariffs, the government may have applied contractionary monetary and fiscal policies to curb inflation and public debt, or a financial crisis may have wreaked havoc with the economy.
As regards India, my data refers to GDP growth from 1981-88 (4.8%) and 1991-2003 (5.8%). You argue that “India clearly began to thrive well before significant trade liberalisation”. I would hardly describe the Indian economy in the 1980s as thriving: it grew much more slowly than China’s.
To this day, India has liberalised much less than China has, and has not enjoyed as fast growth. Whereas Indian living standards, as measured by GDP per person, have risen by 4% a year on average since 1991 – and hence by around three-quarters in total – China’s have risen by 8.5% a year – and hence more than doubled over the same period.
If you want to hold up India in the 1980s as a poster-child of how countries can grow without trade liberalisation, feel free. I don’t think the 3.7% a year rise in GDP per person between 1980 and 1989 (if your figures are correct) was a stellar performance. Nor do I think that the policies that led to the financial crisis in 1991 were exemplary. In any case, they were clearly unsustainable.
If you are trying to argue that a burst of economic growth is possible in the absence of trade liberalisation, then of course you are right. Stalin, for instance, used brutal state power to mobilise the Soviet economy’s resources and industrialise extremely rapidly. His efforts were not a path to long-term development, however.
More happily, domestic market-based reforms (such as de-collectivising agriculture) and sound government policies, such as increased spending on infrastructure and education or an anti-corruption drive, can all be good for growth.
I have never suggested that trade liberalisation is the only source of growth for developing countries – but it is certainly a necessary condition for the sustained economic growth that leads to development.
The simple truth is that whatever statistical quibbles you may come up with about the relationship between trade liberalisation and economic growth – and economics being a social science where controlled experiments are impossible, one can always find something to quibble out – there are no examples of countries that have risen in the ranks of global living standards while being less open to trade in the 1990s than in the 1960s.
Incidentally, since you do not accept that China’s economic success is due to its greater openness, I would be interested to know what you do attribute it to.
As for the Ravallion/Chen article, in your haste to score points, you have failed to carefully read what I said, namely: ‘its findings do not contradict my contention: although some people may lose from trade liberalisation in the short term, the long-term productivity gains from freeing trade, which boosts competition and innovation, raise long-term growth, thus reducing poverty.’
Ravallion and Chen look at the short-term interaction between changes in trade and changes in poverty and find little correlation. They conclude that: ‘China’s periods of more rapid expansion in external trade were not associated with more rapid poverty reduction.’
This is not surprising. As you should know, the short-term impact of trade liberalisation on the poor is ambiguous: while some gain, others lose. But longer term, faster trade-led growth reduces poverty, because the general rise in living standards swamps these one-off distributional effects. Faster growth also provides the means to spend more on pro-poor policies such as education and health.
In short, trade-led growth helps explain the dramatic downward trend of China’s poverty rate – which, as measured by Ravallion and Chen, has fallen from over half the population in the early 1980s to less than a tenth in 2001 – not every short-term shift in it.
Thank you for your reading recommendations. Please allow me to repay the favour. Since I imagine you have better things to do than reading through a mass of economic papers, I suggest you start by looking at Doug Irwin’s excellent review of the trade and development debate. This points to the overwhelming weight of evidence that freeing trade promotes economic growth in developing countries.
Incidentally, I do not accept that the NBER and World Bank country studies are out-of-date and unrepresentative. It is beyond me how you can airily dismiss the World Bank studies – which look at 19 countries over four decades and include countries such as Indonesia, Korea, Pakistan, the Philippines, Singapore, Sri Lanka that started off dirt-poor – as looking at ‘a small sample of middle-income countries’.
As for how open China now is, I’m surprised that you take the IMF’s word as gospel. In any case, the IMF trade-restrictiveness index you refer to is out of date, since China has continued to open up its markets under the terms of its WTO accession agreement: China’s average tariff has fallen again this year, and by two-fifths over the past five years alone.
I also think that it is perfectly correct to argue that, in the context of the dramatic trade liberalisation I have detailed, the five-fold rise in the share of imports in GDP in 25 years – a far faster rise than elsewhere, and which has led to an exceptionally high level of imports for such a large and populous economy – is an accurate measure of how much more open China’s economy now is.
That Shanghai’s stores are now bursting with foreign consumer products, when before only Chinese goods were on sale, is a powerful indication of how much more open (and prosperous) the new China is.
First of all, apologies for the delay in getting back to you on this.
“I would hardly describe the Indian economy in the 1980s as thriving: it grew much more slowly than China’s.”
But so did every other country in the world! I must say, I find your insistence that India did so poorly in the 1980s rather bizarre. After all, during that decade India grew faster than all but a handful of developing countries (excluding micro-states), faster than the OECD and every major geographical region except East Asia, and faster than it ever had before, with output per worker growing and poverty falling faster (1) than it did in the 1990s. And you call that ‘pretty poor’? I can only conclude that you’re keen to dismiss India’s remarkable success during that decade because it so clearly had nothing to do with trade liberalisation.
Which brings me back to China, where I have to say I think you’re still choosing to ignore the main finding of the Ravallion and Chen article I cited, which again was that trade reform has had at best “a rather minor impact on poverty” in China, contrary to the claims of free trade activists.
You ask what I attribute China’s growth to. Some of the factors you actually mention yourself, such as de-collectivisation of agriculture and high public spending on infrastructure and education. There is a lot I would add to that, such as a very equal distribution of land, manipulation of the exchange rate, easy access to credit (indeed, China has the kind of banking system that people will immediately denounce as ‘unsustainable’ if and when it ever falls apart), capital controls during the Asian financial crisis of the late 1990s, technology transfer through joint-venture requirements on foreign investment and weak enforcement of intellectual property laws, the hugely succesful institutional innovation of township and village enterprises (co-owned by local authorities and the source of most private investment in China until the mid 1990s), and so on.
But I would also stress that we shouldn’t really find China’s growth so surprising. Bear in mind that for much of human history it was the world’s economic superpower, and its decline and stagnation in the past five hundred years or was an extended aberration caused a combination of catastrophic leadership and malign interventions by outsiders. So the last 25 years have been something of a re-convergence with the rest of the rich world, and when we look at China’s geography we should expect nothing less – a very large chunk of its land area enjoys the temperate climate that accounts for 8% of world land area but 53% of world GDP, it is blessed with many more miles of sea-navigable river than all of Africa, and it has a very long coastline with some very rich neighbours, with some of whom it has extremely close cultural, historical, ethnic and financial ties, the most obvious example being Hong Kong, source of 30% of all FDI into mainland China in 2005.
With all these natural advantages, even the emergence of a government that was merely not completely insane (as Mao’s regime was) would I think have resulted in a very large boost to Chinese growth. That the regime was also smart enough to implement some of the above measures helped greatly too. If you’re right that China has finally become a relatively liberal developing country in terms of trade policy in the last few years, that is clearly a very recent development, the outcome of careful sequencing of market reforms free from the meddling of the international financial institutions, and the result of the massive strides made in the 1980s and 1990s rather than a cause of them.
Thanks for the link to the Doug Irwin document. Actually it doesn’t seem to me to add a great deal. Quite a lot of the evidence it puts forward is either fairly anecdotal or focuses on trade openness (ie trade as % of GDP), which is as I’ve said an outcome and not a policy variable. In terms of statistical evidence it seems to put quite a lot of weight on the Sachs-Warner study or work based on it, which is a shame as the Sachs-Warner research has been fairly comprehensively debunked by Rodrik and Rodriguez.
Finally,
“I have never suggested that trade liberalisation is the only source of growth for developing countries – but it is certainly a necessary condition for the sustained economic growth that leads to development.”
That clearly isn’t the case, as the example of South Korea which you describe in your previous post demonstrates. That Korea made protectionism work by clever use of market signals doesn’t actually weaken the case for selective trade protection.
To sum up, I think advocates of freer trade would do their case a big favour if they stop pretending liberalisation has been a big success for poor countries when it so clearly hasn’t, and instead started focusing on the reasons for that failure.
(1) According to the stats here, those living on less than $1 a day in India fell by 12 percentage points and 25 million people between 1981 and 1990, compared to 7 percentage points and no numeric fall between 1990 and 2001: http://www.worldbank.org/research/povmonitor/MartinPapers/How_have_the_poorest_fared_since_the_early_1980s.pdf
Dear Jim,
I’m glad you no longer appear to take issue with the significance of China’s trade liberalisation in the 1980s or the degree of its openness today.
As regards India in the 1980s, a country so poor should not only be capable of much faster growth than it recorded – because in an incredibly inefficient economy, there are so many ways of dramatically improving productivity – but indeed must register much faster growth if it is going to catch up with developed countries. According to the Penn World Tables, Indian living standards, as measured by GDP per person, rose by an average of 3.3% a year in the decade before the reforms of 1991. That adds up to a 39% rise in living standards over a decade. If sustained over 25 years – and note that it wasn’t sustainable, because it ended in financial crisis in 1991 – it would lead to a 127% rise in Indian living standards. A generation later, then, India would still be very poor.
Now compare that with the record of the countries that have caught up with developed countries. According to data from the economic historian Angus Maddison (updated to 2004), China’s GDP per person at purchasing power parity rose by 370% between 1978 and 2004, a trend rate of 6.1% a year. Taiwan’s rose by 600% between 1958 and 1987, a trend rate of 7.1%, South Korea’s by 680% between 1962 and 1990, a trend rate of 7.6%, and Japan’s GDP by 460% between 1950 and 1973, a trend rate of 8.2%.
The bottom line is that India’s growth in the 1980s was neither particularly impressive nor a sustainable path to development.
As for Ravallion and Chen, I refer you back again to what I said. They find little short-term correlation between fluctuations in trade and movements in the poverty rate, but this is neither surprising nor incompatible with the fact that trade boosted long-term productivity growth, which in turn raised living standards and reduced poverty.
I have already explained why South Korea does not show that protectionism works in an earlier post.
And I repeat: there are no examples of countries that have risen in the ranks of global living standards while being less open to trade in the 1990s than in the 1960s.
Much as I have enjoyed this exchange of views, I feel that we are starting to go round in circles – and there are many other things I want to talk about in my blog. So I suggest we just agree to disagree.
Cheers
Philippe