Substack

Thursday, October 15, 2020

India's curious export performance story of the last three decades

Excellent paper by Arvind Subramanian and Shoumitro Chatterjee pointing to the important role played by exports in India's growth performance over the last three decades, the rising protectionist trends since 2014, and how this could harm India's economic growth prospects. 

This is perhaps one of the more relevant macro research papers on the Indian economy of recent times, not because it is presenting something completely new but it dissects one of the most important contributors to India's growth and presents some very important insights. An example of first rate research. 

They highlight the paradox of the country being the third best export performer over the last three decades but also being a abysmal failure in massive under-exploitation of its most abundant factor of production, unskilled labour. They show how India's growth since 1990s has been no less export-led than those of East Asian neighbours, and that India's export performance has been driven by high-skill goods and not low-skill goods (textiles, apparel, footwear etc) as would have been predicted.

India produces roughly $34 billion in textiles and clothing. If it were to produce in line with its labor force, the size of its domestic textile and clothing sector should be $174 billion. In other words, India’s missing production in the key low-skill textiles and clothing sector amounts to $140 billion, which is about 5 percent of India’s GDP... India’s underperformance in low-skill manufacturing is a longstanding structural problem that goes back decades... Bangladesh and Vietnam export roughly in line with their endowments and have been doing so at least since 1995. China started as a normal exporting country, exporting in line with endowments; it then became a massive outlier on the positive side, exporting much more than its endowments. India, in contrast, has been a steady, consistent, and massive under-performer throughout the last 25 years, exporting about 15 percentage points less than what would be implied by its abundance of unskilled labor... Between 1995 and 2018, the share of low-skill intensive exports declined from about 34 percent to 20 percent. In contrast, skill-based exports have risen from 42 percent in 1995 to 65 percent in 2018.  
Reflecting the export composition, the authors find that the skill-intensity of aggregate exports (obtained by weighting the skill intensity index for each sector by it export share) is highest for India.
What is striking about the chart is both the level and change in this low- skill intensity export index. For its level of income, India’s index is well above that of China, Bangladesh, and Vietnam, and over time, India’s index rises and remains well above that of other countries whose exports have grown more rapidly. Therefore, in terms of is level and change, India’s skilled export intensity is too high.
Further underlining this point, they find that India's performance in low-skill manufacturing and especially in the four important ones (textiles, apparel, leather, and footwear) to be the poorest. 
The authors argue that the sectoral composition trends of India's exports point to two vulnerabilities, 
The first vulnerability is India’s inability to emulate the performance of China, Vietnam, and Bangladesh to become a powerhouse exporter of goods from relatively low-skill manufacturing sectors such as clothing and footwear. India’s second vulnerability lies in its over-reliance on skill-based exports—the underlying supply of skilled labor is limited, which makes reliance unsustainable over the medium term.

Besides, they also point to the possibility of scarcity of skilled labour becoming a constraint to the growth of skilled services exports.  

In light of this, as the authors argue in this oped, the recent trends of rising protectionism should count as a matter of deep concern.

Between 1991 and 2014, average tariffs declined from 125 per cent to 13 per cent. However, since 2014, there have been tariff increases in 3,200 out of 5,300 product categories, affecting about $300 billion or 70 per cent of total imports. The average tariff increased from 13 per cent in 2014 to nearly 18 per cent. The largest increases occurred in 2018 when tariffs for nearly 2,500 product categories were increased, amounting to nearly 4 percentage points. Tariff increases have been greatest in low-skill manufactured imports and cell-phone assembly, amounting to 10-15 percentage points... In each of the three decades since the 1990s, exports contributed about one-third of overall growth. As a result, India’s export-GDP ratio is currently 20 per cent, more than twice as high as in the early 1990s, despite the post-global financial crisis (GFC) slowdown. Thus, an export slowdown today is likely to have a more consequential impact on the overall economy. Every 5 per cent of the export growth foregone will shave off 1 per cent in overall GDP growth.

In a companion paper they seek to dispel some myths about the rationale behind the Atmanirbharata campaign,

The first is that India’s domestic market is large and buoyant. In fact, the domestic market is still quite small, and likely to remain small over the medium-term, since domestic demand will be weighed down by heavy debts across the economic horizon—in firms, households, and the government. The second myth is that India’s growth has been driven by domestic demand, not by exports, and definitely not manufacturing exports. Our evidence illustrates the opposite, namely that for three decades a stellar export performance has played a critical role in India’s overall growth. The third myth is that India’s exports prospects are dim. We show to the contrary, that India has considerable room to increase exports by increasing its market share. India has been gaining market share for several decades now, even during the difficult global conditions of the past decade. Moreover, India’s past under-performance on low-skill exports creates a significant opportunity at a time when China is vacating export space in such products.

Given these conditions, we argue that an inward turn could doom India to a trajectory of mediocrity. A strategy of abandoning exports to focus on the domestic market will give up a valuable—perhaps the most valuable—opportunity for growth.

I am strongly inclined to agree with all these. They show that India's gain from the rebalancing of global trade away from China in the aftermath of the global financial crisis of 2008 has been limited.

This paper, while being very valuable in dispelling the myths, is disappointing flawed misleading in its conclusions. The one big problem with the paper is with its central conceptual foundation which leads it to make this unqualified advocacy of trade liberalisation,

To the contrary, the big export opportunity for India—low-skill manufacturing—can only be exploited through less protection and more outward orientation because this industry is import-intensive, as China and the other successful low-skill exporters have shown.

This is pure ideology. It is problematic at multiple levels and there is nothing in the paper by way of evidence which warrants any such gung-ho advocacy of tariff liberalisation. There are several problems with the above claim.

One, as several researchers, from Dani Rodrik to Ha Joon Chang, have conclusively demonstrated, "China and other successful low-skill exporters" grew their export industries behind the walls of high tariff and non-tariff barriers. In fact, such infant industry protection has been a feature of the historical development trajectory of all developed countries. However, as Joe Studwell has shown, such protection has to be  calibrated towards productivity improvements through the likes of export competition and not become opportunities for crony capitalism or protection of vested interests. 

Two, such blanket advocacy for tariff liberalisation overlooks the significant heterogeneity in global market conditions and stage of domestic industrial development for different sectors. In fact, if India had pursued this strategy on a blanket basis, it would not have been able to reap perhaps its only manufacturing success of the last five years, the apparent success of the mobile phones industry, where tariffs on assembly of mobile phones increased by 20 percentage points in the 2014-20 period. It would negate the policy of dynamic calibration of tariffs on inputs to facilitate movement along the value-chain.  

Three, even at a conceptual level, the comparison with China and others runs into problems. For example, as Dani Rodrik again has argued, it is one thing to liberalise when tariffs are very high and an altogether different thing to continue doing so when tariffs have fallen to reasonable or even low levels. In fact, as Rodrik has written, this may end up creating more problems. Extending the same logic, one could argue that India's gains from the post-liberalisation success with exports owes more to the starting point of liberalisation than an ideological case that tariff liberalisation is always good. 

Four, the bold and unqualified recommendation to "eliminate all tariffs on inputs" in clothing, based on India's foreign import share of clothing and leather exports being just 16% to China and Vietnam's 40% share, can be argued to be a case of confusing correlation to causation. India's general integration with the global value chain is and has always been far less than that of China or Vietnam. The recommendation assumes that input tariff liberalisation and not establishment of scale manufacturing eco-systems followed by movement up the clothing and textiles value-chain is what led to this condition of high import share of exports. I am inclined to the latter.  

In fact, further tariff liberalisation, without accompanying measures to relax the as yet unknown constraints to scale manufacturing in all these sectors, is most likely to weaken and further immiserate the local manufacturing base in these low-skill sectors. 

So, instead, I would suggest this non-ideological and cautious conclusion,

To the contrary, the big export opportunity for India—low-skill manufacturing—can only be exploited by maintaining the general outward orientation, even deepening it where required, but also carefully deploying tariffs and other industrial policy levers in a calibrated and incentive compatible manner, as China and the other successful low-skill exporters have shown.
It is one thing to criticise the government for raising tariffs across the board, it's an altogether different thing to go beyond that and advocate further tariff liberalisation, and that too in an unqualified and across the board manner. It does no good to substitute one bad (a reversal of openness) with another bad (further blanket tariff liberalisation).

Some other observations:

1. The clear stratification of export segments and their associated labour markets points to a dual-nature of the larger Indian economy itself - the co-existence of a modern and productive economy with a pre-modern and unproductive economy. Given the country's overall size, despite the former being small in relative terms, it is a significant presence and therefore its volume of exports is understandable. This stratification cascades all through the different parts of the economy and society. 

It is not inaccurate to say that the Indian economy is broadly an ocean of mediocrity with islands of productive excellence.

2. India's consumer market too reflects this stratification. The vast majority of the Indian middle class are most likely to be the skilled workforce that inhabit the high-skill manufacturing segment. While the paper does not point to, the reality is that the employment share of high-skill manufacturing is tiny. Therefore, like with the relatively small size of the high-skilled labour force, the Indian middle class too is expectedly tiny. 

The corollary being that without low-skill manufacturing becoming productive and prospering, broad-based economic growth will remain elusive.

3. Any attempt to address the weaknesses with low-skill manufacturing will have to overcome the headwinds of automation which are likely stronger in these sectors. 

4. Finally, while the paper has outlined the problem, the challenge of boosting productivity and production in the low-skill manufacturing sectors remains an elusive quest. Over the years, multiple governments have tried every thing possible to address this problem. While they are important reforms, it is hard to believe that the recent labour market reforms by themselves can contribute much to overcoming the problem. As I have blogged illustrating the case of footwear, the constraints bind at several levels. 

The Indian manufacturing sector appears to be stuck in a very bad equilibrium, which becomes galling and frustrating when we compare with the progress being made by the likes of Bangladesh. The question then is how/why did we get there. And is the answer then only to try out everything and hope that the dynamics of growth will gradually relax these numerous constraints!

1 comment:

Unknown said...

"Extending the same logic, one could argue that India's gains from the post-liberalisation success with exports owes more to the starting point of liberalisation than an ideological case that tariff liberalisation is always good."

This is an important point.

(2) "And is the answer then only to try out everything and hope that the dynamics of growth will gradually relax these numerous constraints!"

Is there an alternative?