The promised land of SEZs


back to issue

IN the last year, Special Economic Zones (SEZs) were a much discussed issue. Most of the discussion focused on two issues, viz. (a) the acquisition of land, rehabilitation, the consequences for farmers and agricultural output, and (b) the cost of the various tax benefits provided to developers of SEZs and the units to be located in them. While these are important issues, they address only the cost aspect of the equation. Taking another path, this paper tries to determine the expected benefits from SEZs and whether they are being achieved. For this, it relies not on aggregate data but on projections made by the developers of individual SEZs that form the basis of claims advanced by the Ministry of Commerce and Industry. What does the data say about the promised land of SEZs, the exports, the nature of investment and employment, regional dispersion and governance?

While the preamble to the SEZ Act 2005 says that SEZs have been established ‘for the promotion of exports’, section 5 of the Act says that the central government will be guided by the following principles while notifying any area as a SEZ, viz.: ‘(a) generation of additional economic activity, (b) promotion of exports of goods and services, (c) promotion of investment from domestic and foreign sources, (d) creation of employment opportunities, [and] (e) development of infrastructure facilities.’

The brief discussion in both Houses of Parliament on the SEZ Bill in May 2005 indicates that lawmakers were more concerned with investment, growth of manufacturing and employment and referred to costs only in passing. The focus was much more on the condition of labour, the role of the state governments and the private sector in the administrative structures, and most critically, employment.1 

The lawmakers saw exports not as an end in themselves but as a route to increasing employment. The Minister for Commerce and Industry, Kamal Nath stated in the Lok Sabha: ‘We no more talk of exports to earn foreign exchange. [In the] Foreign Trade Policy which was announced by the UPA government… We kept the focus on how we would generate employment. In fact, exports today represent one of the most important employment generating activities.’

The other benefit apparently expected from SEZs is inclusive development. Speaking in the Rajya Sabha the minister said: ‘A concern expressed by almost all the members was that the development which takes place, the SEZs which takes place, don’t happen in a localised area. What happened in China? Largely the development and SEZs there are highly localised. We don’t. We have always stated that our developmental process has to be all-inclusive. Not only all-inclusive but must encompass all states. That is not the Chinese belief. So, we cannot have a Chinese concept.’ So, will SEZs help increase employment growth, especially manufacturing employment and help spread development more evenly?


Discussing benefits does not mean that the issue of costs is resolved. There are serious deficiencies in our land acquisition policy. With changes in economic structure, the pattern of land use will change. If the new uses generate enough surpluses, most affected persons can be persuaded to part with land voluntarily. Unfortunately, in India, state power has been used not to overcome recalcitrant hold-outs, e.g., through super-majority provisions, but to subsidise the cost of acquisition.


Dispossessing poor farmers to subsidise SEZs is prima facie unconscionable. In the land acquisition process, it is important to ensure (a) security for the family whose livelihood is being affected, (b) fairness, i.e., enabling them to share in the gains (Box 1 addresses some of the recent suggestions in this regard), and (c) capability to take advantage of the changes in economic structure. Our current land acquisition policy ensures none of these. Even when one agrees with the objectives of SEZs, this aspect needs to be thought through much more carefully.


Sharing the Gains

One idea that has gained currency is to give affected persons a stake in the proposed project. This is not always sensible, since it ignores the possibility that the project may fail. There is little logic in making a deprived section of the population absorb the cost of failure, especially when they had little choice in initiating the project. Ideally, they should share in the gains, while their losses should be limited. One way to achieve this would be to transfer the payback from successful projects into a community fund that would go towards improving common physical and social infrastructure, like electricity, water, road connectivity, schools and hospitals. In case the project fails, the government would assure that it would finance these services instead.

The challenge is to ensure premium educational, health and physical infrastructure in the affected area so that the next generation, the teenage child of the forty year old can aspire to be and work as engineers and managers in the factory and not as unskilled workers. This is not romantic utopia. Even within our existing institutional system, our little-touted Navodaya Vidyalayas, coupled with an extensive scholarship and training programme, can make this happen.2 Sadly, any reasonable person would disbelieve the government’s assurance of basic rehabilitation, leave alone such tall promises. The basic problem is a lack of trust in the state. In the final analysis, until the state is seen not as an instrument of expropriation, but as a fair arbiter, conflict is inevitable. The way to gain such confidence is by repeated demonstration of good intent. Regretfully, there are no signs that the state wants to start on this path.

Similarly, the fiscal benefits for SEZs are ill-designed. The current structure of incentives for SEZs envisages a tax holiday for five years, then a low tax rate for the next five, and an investment tax credit for the final five. Various estimates of revenue loss as a result of these incentives have been presented but a greater cost may be induced by investment distortions. Tax holidays tend to reward the founding of a company, rather than investment in existing companies and benefits short-term investments, characterised by companies that can quickly relocate from one jurisdiction to another. This undermines the effort to attract long-term stable investments. So, if tax incentives are indeed necessary, and this in itself is debatable, an investment tax credit or a lower tax rate may be better than a blanket tax holiday.


Thus, the SEZ policy may be much more costly than necessary, but is it delivering the expected benefits? Earlier this year, the Ministry of Commerce and Industry began releasing data on the commitments made by developers in their applications for the grant of SEZ status. These promises are the basis of the ministry’s projections of employment and investment in the SEZs. A close look at this data can help us understand the promised land of SEZ.


For comparability, we focus only on the 154 SEZs that have been notified under the SEZ Act. These SEZs occupy a total of 20,388 hectares, i.e., about 204 sq km. Information about the name, location, area and type of SEZ is available for all 154 SEZs. For the purposes of analysis, the types of SEZ have been grouped under four broad heads, viz. (i) IT/ITES, which includes information technology (IT) and IT enabled services (ITES),3 (ii) Existing Strengths, which cover our current export basket, i.e., apparel, textiles, gems and jewellery, footwear and pharmaceuticals, (iii) Multi product, and (iv) Others.

Table 1 shows the number of SEZs for which data is available for an additional set of sixteen items of data. It is curious that data on items mandated in the SEZ application is not available for all the notified SEZs. Indeed, only three notified SEZs seem to have provided the necessary details. This shows the approval process in poor light.


Data Availability for Notified SEZs


No. of SEZs reporting data

Proposed Investment by Developer


Proposed Investment by Units


Proposed Indirect Employment


Proposed Direct Employment


Proposed Rupee FDI


Proposed Dollar FDI


Cost of Land


Projected Exports 2007-08


Current Investment in Land


Current Investment (non-land)


Current Investment by Units


Current Number of Units


Current Indirect Employment


Current Direct Employment


Current Rupee FDI


Current Dollar FDI


Table 2 shows the share of different types of SEZs for six different parameters, viz. (a) number of SEZs, (b) area under SEZs, (c) proposed investment by the developer, (d) proposed investment by units, (e) proposed direct employment, and (f) proposed indirect employment. This provides a measure of sectoral concentration of SEZs. As a measure of the geographical concentration for each of these six items, the table also shows the share of two states, viz. Andhra Pradesh and Gujarat and five states, viz. Andhra Pradesh, Gujarat, Karnataka, Maharashtra and Tamil Nadu, usually the top two and top five states, except as noted below.


Sectoral and Geographical Share of SEZs by Different Measures




Investment (developer)

Investment by units

Direct jobs

Indirect jobs

Existing strengths














Multi product














Share of two states







Share of five states







Note: The two states are Andhra Pradesh and Gujarat, to which Karnataka, Maharashtra and Tamil Nadu are added to make up the five states.


To begin with, consider the number of SEZs and the area under SEZs. As is clear, the IT/ITES sector dominates the number of SEZs, with almost two-thirds in this sector. However, most of these SEZs are small in size and therefore the large multi-product SEZs, though much smaller in number, dominate in size, with over half the area under the eight multi-product SEZs that have been notified so far. Geographically, more than three-fourths of the SEZs by number and 92% by area are in the five states. Of this, two-thirds of the area is in just two states. The distinction in share between number and area occurs because Gujarat has relatively few IT/ITES SEZs and more large multi-product zones.


Information about proposed investment by the developer is available for 109 SEZs, with a projected total of over Rs 100,000 crore, i.e., USD 25 billion, apparently over five years.4 Of this, 46% is in IT/ITES, another 4% is in existing strengths and 25% in multi-product SEZs, amounting to 75%. The remaining 25% is in other types of SEZs of which 17% is in port and power. Locationally, 83% of this investment is in five states with 62% in just two states. In this instance, the five states are not the top five. If one replaces Maharashtra by Kerala, which accounts for 12% of the total, the share of the top five states jumps to 89%.

Information about proposed investment by units is available for a much smaller number of SEZs, only for 63 SEZs, with a projected total of Rs 166,785 crore. For purposes of comparison, the investment in manufacturing in India in 2005-06 alone was Rs 360,000 crore. Of this, 9% is in IT/ITES, another 4% is in existing strengths and 78% in multi-product SEZ, amounting to 91%. Locationally, 92.4% of this proposed investment is in AP and Gujarat (46% each), followed by Karnataka, Punjab and Haryana, who add another 4.4%. The proposed investment in units is therefore extremely skewed, even more so than investments by developers. To illustrate, 40% of this, Rs 67,500 crore, is accounted for just by one 10 sq km multi-product SEZ in Kakinada.


The criticality of employment as an objective is well exemplified by the Minister for Commerce and Industry, Kamal Nath, who, at the end of the debate in the Lok Sabha, said: ‘With these few words, I request the support for this Bill to start a new avenue for employment generation.’

Information about proposed direct employment is available for 110 SEZs, projecting a total of 2.14 million employees. Of this, 61% is in IT/ITES and another 15% is in existing strengths with a further 21% in multi-product SEZ, amounting to 97%. It is interesting to note that the 1.25 million direct employment proposed to be created by the IT/ITES SEZs alone exceeds the current employment in that sector. Further, 85% of this proposed employment is in the five states, with 40% in Andhra Pradesh alone, of which two-thirds is from IT/ITES SEZs.

In addition to direct employment, information about proposed indirect employment is available for 82 SEZs, with a projected total of 2.94 million employees. The methodology for calculating the indirect employment is not apparent and varies widely across SEZs, even in the same sector, as noted later. Of this indirect employment too, 68% is generated by IT/ITES, another 12% is in existing strengths and 17% in multi-product SEZ, again amounting to 97%.

The five states account for three-fourths of the indirect employment generated but in this instance, if one replaces Tamil Nadu by Punjab, the share of the top five states jumps to an amazing 92%. This is because 17% of the total indirect employment, i.e., half a million jobs are generated by one IT/ITES SEZ, Quark City, in Mohali, Punjab. Even so, it is not the top job generator, which is another IT/ITES SEZ, viz. Sanghi in Andhra Pradesh which proposes to create 600,000 jobs. Of the approximately two million indirect jobs to be created by the IT/ITES SEZs, over half, i.e., 1.1 million jobs are in just two SEZs (see Box 2).


The fundamentals of SEZs5

Almost the entire indirect employment in Punjab comes from one zone, Quark City SEZ, in Mohali, which proposes to create half a million indirect jobs and directly employ 55,000 IT/ITES workers on a 13.75 hectare plot. If we could replicate Quark’s proposed employment intensity, across the approximately 20,000 hectares of SEZs that have so far been notified, we would have created more than 700 million proposed jobs! Another SEZ that would create more than half a million jobs is the Sanghi SEZ in Ranga Reddy district in AP, which proposes to create 600,000 indirect but only 1,000 direct jobs on a 200 hectare SEZ. These cases need to be studied in more detail.


Not only are SEZs located mostly in a few states, even within these states they are concentrated in a few districts. The notified SEZs are limited to only 53 districts out of 607 districts and even within these 53, they are highly concentrated. Figure 1 shows how the 154 SEZs and 20,388 hectares occupied by them are distributed across twenty districts and it shows the share of each district in the 2.1 million direct and 2.9 million indirect jobs proposed to be generated by the SEZs. These twenty, mostly urban, districts account for 71% of SEZs by number, 82% by area, 88% by number of direct jobs and 89% of the indirect jobs generated. Even within these twenty districts, the top five districts in each category account for 43% of the number of SEZs, 53% of the area, 57% of the direct jobs and an astonishing 79% of the indirect jobs generated, the last, driven by three distinctive SEZs which account for 1.5 million of the 2.9 million indirect jobs proposed to be generated by the SEZs.


Not only are the SEZs localized, they are also localized in particular types of districts. Table 3 shows how many SEZs are situated in districts that are above the national average on a variety of parameters. Only 35 and 22 SEZs are in districts with above average numbers of Scheduled Castes and Tribes respectively. In contrast, of the 154 notified SEZs, 124 are in districts with an above average urban population and 131 with an above average number of non-agricultural workers and a staggering 148 in districts with an above average level of literacy.

All the major cities except Kolkata are part of these districts, viz. Delhi (Noida and Gurgaon), Hyderabad, Bangalore, Chennai, etc. even though some of the larger proposed SEZs around Mumbai and Delhi are yet to be notified. In addition, many of the new cities, such as Pune, Vishakapatnam, Coimbatore, Indore, Ahmedabad, Mohali, Nagpur and Surat figure in the list.


Number of SEZ in Districts Above the National Average (Total = 154)

Urban Population


Non Agricultural Workers




Male Labour Force Participation







As for Foreign Direct Investment (FDI), information about proposed FDI is available for very few SEZs: 18 SEZs provide data on FDI in rupees and 15 for FDI in US dollars. These indicate that Rs 16,139 crore (roughly US$ 4 billion) is proposed to be invested, of which 44% is in IT/ITES, another 21% is in existing strengths and 8% in multi-product SEZ, amounting to 73%. Most of the remaining (26%) is in two engineering products SEZs in Gujarat. Location-wise, 82% of the FDI is in three states, viz. AP (32%), Gujarat (28%) and Tamil Nadu (22%). The US dollar data indicate that another USD 1.92 billion is proposed to be invested, of which 38% is in IT/ITES, another 42% is in existing strengths, all of it in one textile SEZ in Andhra Pradesh and 17% in multi-product SEZs, amounting to 99% of proposed dollar FDI. Over 93% of this is in AP (55%), Punjab (21%) and Maharashtra (17%). The total FDI is thus USD 6 billion.


The available information also provides data for current investment by developers, separately by investment in land (Rs 8,447 crore in 97 SEZs) and in items other than land (Rs 10,220 crore in 66 SEZs). The combined total of Rs 18,667 crore is about 19% of the total proposed investment by developers in 119 units. Of the investment in land, 44% is in IT/ITES and 21% in multi-product SEZ with power accounting for another 10%. Of the investment in non-land activities, again 44% is in IT/ITES, but 51% is in multi-product SEZs, which is almost entirely in two Gujarat SEZs. Existing strengths attract only 5% and 3% of investment respectively.

Information about current investment by units is available only for 42 SEZs, with a total of Rs 23,434 crore, i.e. about 14% of the total. Of this, 6% is in IT/ITES, another 3% is in existing strengths but most of the rest, i.e., 81% is in one multi-product SEZ in Gujarat. So far, this Rs 18,939 crore of investment has resulted in 680 jobs, including 43 indirect jobs. The data on current investments, therefore, if anything, reinforces the trend towards concentration seen in the data on proposals.


The examination of the data available from the ministry raises many questions. A few broad trends are discernible.

First, the SEZ boom is concentrated heavily in the IT/ITES sector, one that has already boomed. It may not be out of place to characterize the SEZ policy as a continuation of support for the IT/ITES sector through the back door. Nearly 75% to 80% of the proposed employment emanates from the IT/ITES and the traditional exporting sectors viz. i.e., apparel, textiles, gems and jewellery, footwear and pharmaceuticals. The multi-product SEZs, which are to be the harbinger of manufacturing growth contribute but a sixth of the projected employment growth. Neither are the projections of FDI, limited as they are to a few SEZs, very encouraging. It is thus difficult to describe the SEZ policy as one promoting manufacturing employment, even going by the official numbers. If anything, it appears to be reinforcing existing paradigms and providing the IT/ITES sector with an arguably unnecessary tax loophole.


Second, regardless of the minister’s statement in Parliament that the ‘developmental process has to be all-inclusive; not only all-inclusive but must encompass all states,’ the forces of economic agglomeration appear to have triumphed over the minister. By all measures, whether by number or area or employment, most of the SEZs are in a small number of districts in select states such as Andhra Pradesh and Gujarat. These districts are more educated as compared to the national average and more industrialized and urbanised. SEZs can therefore be expected to exacerbate regional concentration.


This is not necessarily bad, since there can be significant benefits from such agglomeration. However, the consequences of such concentration and the need to put in place policies to ensure that people from other areas also benefit from SEZs is being ignored, perhaps because of an unwillingness to accept a reality that conflicts with a stated goal. To begin with, there needs to be much more readiness to accept migration on a larger scale to the favoured districts from other areas in the state and the country. For those who think that there is already plenty of migration in India, a comparison with China is a useful cross-check. While about two million people migrate from rural to urban areas in India each year on average, the comparable figure is 14 million in China.


How fast do SEZs grow?

An example from China may help to illustrate the time frame of growth in SEZs. The Beijing Economic-Technological Development Area (BDA) was established in 1993 on 46.8 sq. km, i.e., 4680 hectares of land, a size at the upper end of our current limit and is thus a more reasonable comparator as compared to SEZs like Shenzhen, which extend over 350 sq km. This BDA area is designed to contain industrial, business and residential spaces. It is well connected by road, rail, sea and air. Investors benefit from levelled land and access to (i) roads, (ii) storm water drainage, (iii) waste water drainage, (iv) tap water, (v) natural gas, (vi) power, (vii) telecommunication, (viii) heat, and (ix) cable television. Initially, the enterprise income tax rate was 15%, reduced to 10% for enterprises that exported at least 40% of total value of output.6 By 2004, BDA was home to over 1600 companies from 30 countries employing about a 100,000 persons, with a total investment of USD 8.12 billion, of which USD 3.2 billion was FDI. Of the original 47 sq km, only 23 sq km had been developed.

Compared to this, the ministry projects that the SEZs will create two million direct and three million indirect jobs. This will be a result of investing approximately USD 70 billion in an area of 200 sq km, an investment to employment and an area to employment ratio of about five times that achieved by BDA. Furthermore, this will be achieved in five years compared to 12 years for BDA.

Third, the credibility of the projections is doubtful. Box 3 provides a Chinese perspective on this issue in terms of the kind of growth that was experienced. In contrast, the case of Quark City and Sanghi SEZ mentioned earlier are striking departures. A number of the other projections could prove equally misplaced. At this point it is useful to ask whether these commitments made by developers of SEZs are binding? Is there any enforcement mechanism? What will happen if the projections are not met? Who will be held responsible?


The answers to these questions are, as yet, unknown. The worrying part of these projections is that the Board of Approvals, composed of seventeen officers of the Government of India, one nominee of the state government (which may be a commentary on the participation of state governments), and one professor of IIM, has accepted them. Is this body applying its mind?

To answer this question, it is useful to consider a subset of 87 IT/ITES SEZs. Usually, similar projects should have similar characteristics within some range of parameters. For example, the cost of a power plant per mw (for a given type of plant, e.g., coal, gas, hydro, etc.) would lie within some bounds, as also the cost per lane kilometre of highway. Part of the appraisal and project approval process is to ensure that the project costs are within acceptable ranges.


In the case of SEZs, this does not seem to be the case. Table 4 shows the range of variation in a few basic parameters across different SEZ projects of a single type, i.e., IT/ITES projects. The parameters relate to (a) investment by developer per hectare, (b) direct employment per hectare, (c) direct employment per crore of investment, and (d) ratio of indirect to direct employees. As can be seen, the distribution is quite wide, varying from less than Rs 5 cr. per hectare to Rs 200 cr. per hectare.


Distribution of Parameters Across SEZs in the IT/ITES Sector

Investment/ha (Cr./Ha)


Employees/ hectare


Employees/ Rs. Cr. of Inv.


Indirect/ Direct Employees


Less than 5


Less than 100


Less than 10


Less than 0.10


5 - 10


100 - 250


10 - 20


0.10 - 0.25


10 - 20


250 - 500


20 - 30


0.25 - 0.50


20 - 30


500 - 750


30 - 40


0.50 - 0.75


30 - 40


750 - 1000


40 - 50


0.75 - 1.00


40 - 50


1000 - 1500


50 - 100


1.00 - 2.50


50 - 80


1500 - 2000


100 - 500


2.50 - 5.00


80 - 100


2000 - 3000


500 - 1000


5.00 - 10.00


100 - 150


3000 - 4000


1000 - 2000


10.00 - 20.00


150 - 201


4000 - 6000


2000 - 4000


20.00 - 600.00










The variation in employment ratios is even more, from less than 100 direct employees per hectare to over 5000 per hectare in direct employment, less than 10 employees per crore of developer investment to more than 2000 and less than one tenth to over 10 (indeed, one SEZ has a ratio of 600!) for the ratio of indirect to direct employees. This kind of variation for a key parameter of interest in a given sector and relatively well understood sector, i.e., IT/ITES, is difficult to explain as variation across business models. Data such as this leads one to suspect non-application of mind at the approval stage.


Not just approval, this apparent mindlessness prevails in monitoring too. Regardless of the statement by the minister in Parliament that ‘we no more talk of exports to earn foreign exchange... We kept the focus on how we would generate employment,’ the monitoring formats remain antiquated. Form I, which is supposed to monitor the activities of units in the SEZ, and is issued as part of the SEZ rules, focuses almost entirely on whether the unit is earning ‘net foreign exchange’ devoting a couple of pages to gathering the relevant details. Employment, on the other hand, merits one line. From Form I, all one can infer is the number of men and women employed, with no information about wages, quality of employment, etc.

Are these just more examples of implementation at odds with policy or lackadaisical and mindless administration?


From their location it is evident that SEZs are very much an urban phenomenon, with the formation of new cities that will be clustered around existing cities. Worse, a number of the SEZs, especially in the IT/ITES sector, are too small to be planned in an integrated manner. For example, of the 27 SEZs in Ranga Reddy district around Hyderabad, 19, all in the IT/ITES sector, are less than 50 hectares. A higher rate of urbanisation is an inevitable consequence and a necessary facilitator of rapid growth but are we prepared for this? What are the arrangements to run the SEZ cities that will emerge if the policy succeeds?

The existing SEZ Act7 mentions the word ‘urban’ in two places, once to note that the Ministry of Urban Development may form part of the Board of Approvals, and the other to provide fiscal relief in case a unit moves from an urban area to an SEZ. The word ‘plan’ or ‘planning’ occurs once in the SEZ Act, in the fiscal context mentioned above, but it does occur twice in the rules, to state that the building approval plan will have to be submitted to the Development Commissioner who shall place it before the Approval Committee for consideration and to state that the developer and co-developer should ‘abide by the local laws, rules, regulations or bye-laws in regard to area planning, sewerage disposal, pollution control,’ etc.

It is apparent that the urban aspects of the SEZs have received little thought and consideration. This is true even in the deliberations of the Parliamentary Standing Committee on Commerce, which did receive some submissions on the urban planning issue from the Ministry of Urban Development. However, the Committee’s recommendation to reduce the overall size of SEZs and increase the processing area indicates an inappropriate comparison between SEZs and industrial estates, rather than industrial townships.


It appears that the physical planning of the SEZ is at the mercy of the Approval Committee, composed of the Development Commissioner, five central government officers, two state government officers and the developer as a special invitee. The admonition to abide by local laws potentially involves the local panchayat that exercises jurisdiction over the SEZ area but it is unfortunately unlikely that they will get a role. Moreover, the capacity of the panchayat to engage in discussions with the SEZ developer on issues of area planning is questionable and significant capacity building would be needed in this area. Situations where multiple panchayats exercise jurisdiction over one SEZ can occur, and what will happen in this case is unclear. Indeed, the National Capital Region Planning Board has thus far been ignored in decisions regarding SEZs in the NCR.

So, if the SEZs were to succeed, they could well degenerate into the same kind of urban mess that we see in our cities today, for the same reason – lack of governance. There is almost criminal neglect of urban planning issues in the legislative and administrative framework for SEZs. Indeed, given the kind of location that we have indicated, the urban outgrowth from the existing cities and that from the SEZs can merge to form a large chaotic unplanned morass that will enclose the SEZ.

In contrast, the Chinese approach, e.g., in BDA, referred to earlier in Box 3, is to bring the zone under municipal management. BDA is one of the thirteen districts of Beijing Municipal Government. A similar zone in Hang Zhou, in Zhejiang province, is also under the administration of one of the municipal districts of Hang Zhou. The Chinese can do this easily because of two differences: (i) the zone land is publicly owned, and (ii) the major municipalities have substantive planning capacity, e.g., Shenyang, a city of about four million and the capital of Liaoning province has a Planning Institute with 300 professionals, including about 60 urban planners.

Since we have made it harder for ourselves by choosing privately owned SEZs, we will have to evolve alternative governance structures to address this issue. While on the issue, it is useful to clarify that there is no conflict between public ownership of land and private provision of infrastructure services. The implicit cross-subsidy from profits obtained by developing the non-processing area to overall infrastructure can be achieved with the help of a dedicated fund. All the benefits of the SEZ that are currently touted could have been achieved as easily while retaining ownership of land with the public sector. The rationale for choosing this particular development approach to SEZs has never been clearly explained.


Based on an examination of data available from the ministry’s own website, while the costs appear very real, the benefits of SEZs appear to be a mirage. If at all, the subsidies in terms of land and tax benefits extended are only helping to support the existing economic structure. More than two-thirds of the proposed employment growth is in IT and IT enabled services and almost 90% of the jobs will be available in twenty districts, which are all above average in terms of urbanisation, industrialization, and education.

Moreover, if SEZs were real, they foreshadow a promised land that we are not prepared for and appear unwilling, if not unable to manage. There has been no thinking of what would need to be done if the SEZs were to actually succeed and grow into cities. Given that many of them are close to existing urban areas, there are major implications for urban planning. Regrettably, this piece of the SEZ puzzle has been missing from inception and even today, there appears to be no recognition of the problem. It may well be that a ministry used to dealing with Export Processing Zones and Export Oriented Units is quite unsuited to the task of regulating Special Economic Zones. However, given the multi-ministerial nature of the Board of Approvals, it appears that these issues are not appreciated by others in the government either.


The failure is not so much of the SEZ concept as the fact that the existing governance of the process does not inspire confidence due to an inability to define priorities, and regulate and manage the phenomenon called SEZs. The government baldly accepts averments that half a million indirect jobs will be created from a 15 hectare zone. It appears to have no standard, even within sectors, for appraising a proposal and does not even insist on complete information being provided before an SEZ is notified. While monitoring, it ignores employment and sticks to hackneyed metrics like net foreign exchange earned.

Evidently the government believes it can abdicate governance and outsource the task of development to the private sector. It will soon learn otherwise.



1. The Parliamentary Standing Committee on Commerce, however, did address the land acquisition issue and recommended a reduction in the maximum size of the SEZs, a proposal that the government acceded to by putting a cap of 5000 hectares on the SEZ, which it now seems to be reconsidering (see ‘Govt. may ease land ceiling on multi-product SEZs’, Indian Express, 4 December 2007

2. While the Tatas have undertaken an initiative to train some local Singur youth in ITIs, this effort is limited.

3. This includes electronic hardware, which is a very small proportion of this group.

4. This is not clearly mentioned but is inferred from the fact that the SEZ Application Form mentions that the projections are over a five-year period.

5. The Quark City SEZ is promoted by Quark Inc., which has a large share of the publishing software market. A quark is a physical particle that is visible only fleetingly. They form one of the two basic constituents of matter and various species of quarks combine in specific ways to form protons and neutrons.

6. Further, high tech enterprises were exempt from income tax for the first three years with 50% reduction in taxes from the 4th to 6th year and software firms were exempt for the first two years with 50% reduction in taxes from the 3rd to 5th year. In a recent change in the law in China, these benefits have been rationalized and both Chinese and foreign enterprises now have a common tax rate of 25%.

7. The rules mention it only once, in the context of the Board of Approvals.