It is generally agreed that a key element in the transformation of India is the creation of a large number of good jobs. While micro and small enterprises provide lots of jobs, consistent with their low productivity, they pay relatively low wages.

For example, according to recent research by Rana Hasan and Nidhi Kapoor of the Asian Development Bank (ADB), manufacturing firms with less than 20 workers each employed 73% of manufacturing workforce but produced only 12% of manufacturing output in 2010-11, the latest year for which such data are available. With such a large share in employment but small share in output, these firms are able to pay only a fraction of the average wage paid by larger firms, which is itself low in India when seen in international context.

The position of the small firms within manufacturing is not unlike that of agriculture in the economy as a whole. As many observers have noted, agriculture had 49% share in the workforce and 15% in the GDP in 2011-12.

There is compelling evidence that wages rise with the size of the enterprise. This wage pattern mirrors the pattern of average labour productivity, which shows a rising trend with the size of the enterprise. High productivity of large firms partially results from their ability to effectively exploit scale economies. Moreover, with the domestic market often small, they also overwhelmingly operate in the world market where competition is intense. Therefore, they must continuously innovate and adopt cost-saving technologies and management practices.

The presence of large, export-oriented firms also fosters a highly competitive environment in regions of their location. In so far as small and medium firms either become ancillaries of large firms or must compete against them, they too are compelled to strive for efficiency. Therefore, substantial presence of large firms combined with an outward-oriented trade policy fosters high overall productivity. Conversely, the absence of large firms is often associated with low average productivity.

India’s own experience is consistent with these observations. In apparel, where we lack substantial presence of large firms, average labour productivity is low. This in turn translates into meagre exports in relation to the total employment in the sector. Our apparel exports are less than one-tenth those by China and less in absolute terms than those by much smaller Bangladesh and Vietnam. In contrast, in software, we have significant presence of large firms and this sector exhibits high productivity and a large volume of exports.

Lack of substantial presence of large firms in India has impacted average labour productivity in two ways. First, the level of productivity in micro, small and medium firms is low compared with their counterparts in countries such as China. And, second, a disproportionately large volume of the workforce is employed in these low productivity firms.

According to a 2009 ADB study, only 10.5% of manufacturing workforce in India was employed in firms larger than 200 workers compared to China’s 51.8% in 2005. At the other extreme, 84% of India’s manufacturing workforce was in firms with less than 50 workers compared to China’s 24.8%. These differences translate into substantially lower average labour productivity and wages in India than China.

Unfortunately, large firms are missing in India in precisely the sectors in which they are needed the most: employment-intensive sectors such as apparel, footwear, electronic and electrical products and host of other light manufactures. These are products in which China has done well thereby generating a large volume of good jobs for its workers. In 2014, the country exported $56 billion worth of footwear compared with $3 billion by India and $782 billion worth of electrical and electronic goods compared with $9 billion by India.

The single most important key to China’s success in manufactures has been its decision to go for the large world markets in preference to its much smaller domestic market.

In 1980 when China’s GDP was less than $500 billion at today’s prices and exchange rate, it began by establishing four very large Special Economic Zones (SEZs) along its southeast coast. These zones were located directly across from Taiwan and Hong Kong, which then faced the prospect of being priced out of the world market due to their high wages.

Shenzhen, one of these four SEZs, was then at best semi-urban with a population of 300,000. Attracted by low wages and business- and foreign-investment-friendly environment, investors from Hong Kong immediately flocked to this SEZ. Later, investors from Taiwan, Japan, United States and other countries followed as well. Coastal location allowed these firms to operate in the world markets unhindered by the poor infrastructure in the hinterland, especially in the early years. They could import inputs from and export outputs to foreign destinations. Employment opportunities for Chinese workers multiplied.

Today, Shenzhen has a population of 11 million and it boasts of gross city product of $265 billion. Though originally Cantonese, it speaks Mandarin because the bulk of its population migrated from other parts of China. Most of the major multinational firms have a presence in Shenzhen.

Having risen at the rate of 10% a year in real terms since at least 2007, average annual manufacturing wages in China today stand above Rs. 5 lakh per year. Due to demographic transition, the country also faces worker shortage that would only get worse in the years to come. When asked in surveys, Chinese firms today point to labour costs as the most important barrier to their development. Already, many multinational firms are looking for alternative locations where they can find abundant supply of workers.

So far the firms exiting China have gravitated more towards countries such as Vietnam and Malaysia. But with its large labour force, India is well positioned to take advantage of the opportunity. What is needed to convert this opportunity into reality is a business friendly ecosystem in regions that can serve as export bases of the migrating firms. Given our relatively weak internal infrastructure links, coastal regions adjacent to deep-draft ports are the best candidates for such bases.

Happily, this opportunity coincides with the launch of Sagarmala project of Prime Minister Narendra Modi. Building on the Gujarat experience of the Prime Minister, this project seeks to unleash port-led development in the country. The impressive success of Gujarat during the Prime Minister’s tenure as Chief Minister had been partially built on a port-led-development strategy.

In 2013-14, Kandla distinguished itself among the major port for carrying the most cargo. At the same time, non-major ports of Gujarat jointly carried three times the cargo carried by Kandla. The SEZs in Gujarat also accounted for a hefty 45% of exports by all SEZs in India in 2013-14.

Therefore, in the context of Sagarmala project, India could begin by creating one Shenzhen-style Coastal Economic Zone (CEZ) on its western coast and another on the eastern coast near deep-draft ports capable of accommodating very large and heavily loaded ships.

To be successful, these zones would have to cover a large area (Shenzhen covers 2,050 square kilometres) and would have to have some existing infrastructure and economic activity. They would need to must provide a business friendly ecosystem including ease of doing business, especially, ease of exporting and importing, swift decisions on applications for environmental clearances and speedy water and electricity connections.

Apart from conventional infrastructure, the zones would need to create urban spaces to house local resident workforce. For firms that create a threshold level of direct employment (e.g., 50,000 jobs), a tax holiday for a pre-specified period may be considered. To incentivize early investments in the zones, the tax holiday might be limited to investments made in the first three or four years of the creation of the zones.

An important advantage of locating the zones near the coast is that they would attract large firms interested in serving the export markets. These firms would bring with them technology, capital, good management and links to the world markets. They would help create an ecosystem around them in which productive small and medium firms would emerge and flourish.

It may make sense to initially limit the number of zones to a few, perhaps two or three. This would help ensure that many sector-specific zones and clusters emerge within each CEZ to fully exploit economies of scale and agglomeration. Simultaneous creation of too many zones would spread the available public resources thinly while also diffusing economic activities with potential synergies. As initial zones succeed, more may be subsequently launched. This is not unlike the software industry, which initially concentrated in Bangalore but subsequently spread to other towns. (Of course, since software travels on the wire, this industry did not require location near a coast.)

There remain two final questions. First, why can we not rely on a protected domestic market to attract investment? The answer is that the domestic market still remains small and fragmented so that it will not give rise to genuinely large firms. For example, home market in electronic goods is $65 billion of which $26 billion is already supplied by domestic firms. In comparison, the world market in electronic goods is $2 trillion. Domestic market can serve as an attractive complement; it cannot substitute for the large world market.

Second, with the export market growing slowly, can we rely on an export-oriented strategy? The answer is in the affirmative. At $18 trillion, the world export pie is extremely large. Even if this pie is not growing, our current share of it at 1.7 per cent leaves us considerable scope for expansion. We should remember that during 1995 to 2013 when China grew at 10%, the OECD [Organization for Economic Cooperation and Development] countries grew only 1.4% annually. China succeeded by taking an ever-larger slice of the world market: it expanded its share in the world exports from 2.9% in 1995 to 12.3% in 2014.