by Rakesh Mohan
However, in both these Budgets, tariffs were reduced for a large number of products.17 While several tariffs were reduced, some items were also subjected to higher tariffs in line with the finance minister’s statements: ‘The gradual reduction of import duties in the past few years has resulted in certain distortions and anomalies. My proposals seek to correct them as far as found feasible without causing abrupt disruption in the duty structure.’18
A noteworthy feature of India’s tariff peaks and averages in the early 1990s was that though they were reduced from high levels, they were high (particularly for non-agriculture) compared to most tariffs prevailing in economies with low tariffs, e.g. the United States.19
In 1992–93, India’s maximum tariff rate was reduced to 110 per cent, with the exception of alcoholic beverages and baggage.20 This trend continued and the peak customs duty in India was brought down to 50 per cent in 1995–96. The reduction in peak tariffs continued again every year since 1999–2000 to reach 10 per cent in 2007–08 (see Table 3 and Annex II).
Interestingly, in the early part of the previous decade, more than one finance minister signalled the intention to continue decreasing tariffs. The Budget speech in 2004 stated that India would continue to decrease its tariffs in a measured way, with the intention to align India’s tariff structure with that of ASEAN countries. This view was reiterated in 2005, the year when tariffs on 217 items covered under the WTO’s Information Technology Agreement (ITA) were reduced to zero.
Two interesting features of the reduction in India’s peak tariffs are:
these peak rates are not necessarily the maximum tariffs prevailing, but the highest rate for a large proportion of tariff lines (say, 70 per cent or more lines); and
these rates do not apply to agricultural items.21
These two aspects are shown in certain parts of Annex II, when the finance ministers mentioned tariffs higher than the peak rate and also that the peak-rate reduction applies only to industrial products.
(b) Reduction in Tariffs Peaks Prepare India’s Pathway to being a Low-tariff Economy
The sustained focus on reducing tariff peaks has contributed to India moving towards being a low-tariff economy. Table 4 shows that as a result of the tariff changes introduced since the tariff reform in 1991, India’s simple average tariff in 2015–16 was about one-tenth of that in 1990–91.
On the basis of these estimates, with about 13 per cent simple average applied MFN (Most-Favoured Nation) tariffs, India today is close to the upper part of the range of tariffs in ASEAN economies.22 If we consider only the MFN tariffs, then it would appear that considerable further tariff reduction would be required if India aims to reach the middle part of the ASEAN tariff structure, with the average applied MFN tariffs of Indonesia, Malaysia, Myanmar and the Philippines ranging between 5.6 to 6.9 per cent.
However, an important point to bear in mind is that though the tariff averages for India based on MFN tariffs and certain adjustments show relatively high tariffs, the actual applied average tariffs, which encompass all concessions and exemptions, are shown by estimating the customs tariffs per unit imports. We can see the difference between the usual estimates of tariffs and those based on customs duties from the information provided in Tables 4 and 5.
Thus, if we alternatively consider the weighted average tariff based on customs revenues for India, the picture is very different. In terms of its weighted average tariff, India is already one of the low-tariff economies in the world.
(c) India: An Economy with Very Low Average Tariffs
Table 5 shows India’s trade-weighted tariff average, based on the customs revenue collected divided by corresponding year’s merchandise imports.
Table 5 shows two estimates relating to customs revenues as a proportion of import value in that year. ‘Total customs revenue’ includes both tariffs and the countervailing duties (CVD) on imports, the latter being imposed on imports in lieu of domestic excise tax. ‘Basic customs revenue’ shows the revenue from only tariffs, i.e., it excludes revenues due to CVD (see section (h) below for discussion of CVD). To examine the average level for tariffs alone, we should consider the estimates based on basic customs revenues.
We can see that after 1991, a large fall in weighted average applied tariffs (without CVD) took place till end 1990s, reaching about 13 per cent of a tariff level of almost 50 per cent. From 2005–06 onwards, India’s average tariffs based on total customs duties collected was mostly in single digits; average tariffs in terms of basic customs duties were in single digits from 2002–03 itself. It is noteworthy that the average customs duty corrected for CVD is below 2 per cent in the years from 2012–13 to 2014–15.23
The difference between CVD and actual tariff rates started increasing in the late 1990s and has continued till today, showing a preponderance of CVD in total customs duties of India.
It should be noted that the estimates of tariffs given in other Tables are not strictly comparable to those calculated based on customs revenue shown in Table 5. The other tariff estimates are based on MFN tariffs or adjustments made to the MFN tariffs for concessions. The estimates in Table 5 are a composite estimate taking into account all aspects of concessions in an aggregate way.
The customs-revenue–based weighted average tariff for the US is 1.5 per cent, computed on the basis of dividing the US customs revenue by its merchandise imports.24 The estimate for India is not too different from that of the US, which is one of the low average-tariff economies.
This shows that India is far more open than what has normally been perceived till now, in terms of tariffs.
(d) Does the Picture Change Due to a Downward Bias of Trade-weighted Average Tariff?
The weighted average tariff estimates in Table 5 are downward biased because of relatively lower imports of products with higher tariffs. This estimate does, nonetheless, provide a reasonable indication of the overall qualitative low tariff level, particularly if most of the tariff lines are in the lower range of tariffs. This is the situation for India, which has a large share of its tariff lines with low tariffs.
Table 6 shows that 13 per cent of agriculture tariff lines and 90 per cent of non-agriculture tariff lines are with tariffs of 10 per cent or below.25 Internationally, most of the focus of tariff liberalization is towards non-agriculture products, which in India’s case have an overwhelming share of their tariffs between 0 and 10 per cent. For example, even within the TPP, immediate tariff elimination applied only to 70 per cent (Vietnam) to 100 per cent (Singapore) of non-agriculture tariff lines, but in the case of agriculture tariff lines, this ratio was 42.6 (Vietnam) to 100 (Singapore) per cent; in the case of Japan and the US, immediate elimination under TPP was agreed only for, respectively, 51.3 and 55.5 per cent of their agriculture tariff lines.26
Annex Table 1 shows the relatively few non-agriculture tariff lines in India that are above 10 per cent. Annex Table 2 suggests that most of these non-agricultural products with tariffs above 10 per cent are not likely to be among the major traded items for India. We see from Annex Table 2 that the overall import share of non-agricultural products with tariffs above 10 per cent in total imports is 4.54 per cent. Of these, iron and steel and articles of iron and steel (HS categories 72 and 73, with tariff of 15 per cent) account for 3.93 per cent. The rest of the products account for only about 0.6 per cent of total imports. The combined impact of a reduction in tariffs for these items and the consequent rise in their imports is unlikely to contribute to a significant increase in India’s weighted average tariff.27
Therefore, it would be reasonable to consider that the low average-weighted tariff shown in Table 5 gives a reasonably correct message about India’s effective tariffs, namely that on average, India is a low-tariff economy similar to some of the developed economies that are seen as more open in terms of their low tariffs.
An important message from this estimate is that we need to revise the general impression that average tariffs in India are significantly higher compared to those in
other large countries. The average tariffs of India are close to those prevailing amongst the economies, which are considered to be the lowest-tariff economies.
(e) India: Different Tariff Patterns for Agriculture and Non-Agriculture Products
The profile of tariff distribution across agriculture and non-agriculture products (see Table 6) show the greater tariff protection provided by India to its agriculture. This is not unusual, as can be seen for several developed economies as well, including the United States and very much the EU.28 India’s agriculture tariffs were affected by two very important events at the WTO, one that increased the extent of tariffs and another that reduced them.
The first was the fact that under the Uruguay Round results, all non-tariff measures of agriculture, for all WTO members, were converted into tariffs. The WTO Agreement allowed this for certain developing countries (including India) under the so-called ‘ceiling binding’ scheme under the Agreement on Agriculture concluded in the Uruguay Round.29 This implied that the prevailing tariffs for agriculture in India were increased at the end of the Uruguay Round to replace the prevailing quantitative restrictions by tariffs.
The second was a significant development that led to a further liberalization of India’s tariff and non-tariff regime, namely, the balance of payments (BOP) related dispute, which India lost at the WTO. The loss in that dispute meant India did not have the possibility to claim under WTO that it was entitled to implement higher trade restrictions than those allowed in WTO under normal circumstances, on the grounds that it faced a BOP problem.30 Though the focus of the dispute brought by the United States was on quantitative restrictions, it also led to a reduction in a number of tariffs.31 This added to the momentum of changes brought about through a general emphasis on reducing tariffs by successive finance ministers.
The combination of the ‘ceiling binding’ and concern to provide high-tariff protection to agriculture has led to relatively higher tariffs for these products compared to non-agriculture ones.
Thus, a closer look at India’s tariff structure shows that India’s overall average tariffs are low primarily because of the low tariffs for non-agricultural products and their large share in total imports (see Table 6). The weighted average tariff for agriculture is increased32 both due to a preponderance of significant share of agricultural tariff lines with higher tariffs and the larger share of imports subject to these higher tariffs (see Table 6).
These estimates show that the conventional view that India is a high-tariff economy is incorrect even for most tariff lines with MFN tariffs. The tariff estimates in Table 6 suggest that Indian MFN average tariffs for non-agriculture products too are likely to be close to that of economies that are considered very open in terms of their tariff regime.
(f) Comparison of India’s Average MFN Tariffs with Tariffs of APEC Economies
Based on the information in Table 6, we compare India’s weighted average MFN tariff for agriculture and non-agricultural products, with those of APEC economies (see Table 7). An interesting picture emerges here. India’s non-agricultural tariffs are close to certain APEC economies, and is in fact less than the average for some APEC economies (Chile, Russia and Vietnam). In contrast, India’s average agriculture tariffs are higher than all APEC economies, except for Republic of Korea. Therefore, even for MFN tariffs, India’s tariffs are within the range covered by APEC economies.
This shows us that India has travelled a long way from the days when tariff reform was initiated in 1991. Today, its level of average MFN tariffs for non-agriculture are comparable to several economies that are conventionally considered to have carried out significant tariff reform and opening up of their markets. Further, as shown by Table 5, its weighted average tariffs are amongst the group of countries which are considered to have the lowest tariffs in the world.
(g) The Spread of Basic Tariff Rates
In 2015–16, the range of maximum MFN tariffs for the different chapters covering agriculture33 is from 30 per cent to 150 per cent. Within this group, the most common maximum tariff is either 30 per cent or 100 per cent. For the non-agriculture products,34 Annex Table 1 shows that the highest maximum MFN tariffs apply to motor vehicles (125 per cent).
For most tariff lines, a maximum or peak MFN applied tariff of 10 per cent is the most common rate. As shown by Table 6, about 78 per cent of tariff lines for non-agriculture products are between 5 to 10 per cent; 90 per cent of India’s tariff lines on non-agriculture products are in the range of 0 to 10 per cent.
Annex Table 1 shows a large dispersion amongst the relatively higher tariffs for non-agricultural items, with ten different rates ranging from 12.5 to 125 per cent (without considering the rates for personal baggage). This indicates the possibility of rationalizing this tariff structure to both reduce the number of tariffs rates and to simplify the tariff regime. In this context, it is worth recalling the Chelliah Committee Report of 1993, which had recommended a smaller number of basic tariff rates even in the early phase of reform, in comparison to the present spread of tariffs.
The Chelliah Committee Report stated:
A large number of rates not only create administrative problems, but leads to non-transparency in the degrees of effective protection given to different products as well as to distortions in the allocation of resources. By keeping only a limited number of rates, we shall render administration of the tariff a fairly easy task, and by limiting the spread, we shall minimize distortions.… We are recommending that by 1997 or 1998 March end, the structure of ad valorem rates of duties in place should be: 5, 10, 15, 20, 25 and 30. In addition, when non-essential consumer goods are allowed to be imported, there should be another ‘slot’ for them, namely, 50 per cent.35
The number of basic tariff rates have come down in a major way, from twenty-two different basic tariff rates in 1991–92, to only four by 2001–02.36 We can see from Annex Table 1 that the number of these different basic tariff rates have gone up since the early 2000s.
This suggests that the current spread of tariff rates is an obvious area to focus for tariff reform.
(h) Auxiliary Duties, Additional Duties, Special Additional Duties, and Surcharges
Before 1991, India imposed high auxiliary duties on imports. These were merged with basic tariffs and the overall combined tariff level (basic rate plus auxiliary, inclusive of specific duties) was capped at the rate specified in the Budget: 150 per cent in 1991, 110 per cent in 1992, and so on as discussed earlier. In the 1993 Budget, the government merged the auxiliary duty with the basic duty rate, thus doing away with the auxiliary duty.
In addition to basic customs duty, all goods imported into India are subject to additional duty of customs, popularly known in India as ‘CVD’.37 This additional duty of customs is equivalent to excise duty,38 and is refundable through the credit provided under the CENVAT scheme.39 Further, imported goods are also subject to special additional duty of customs, popularly known in India as ‘SAD’. This SAD is levied in lieu of state taxes, local taxes or any other charges.40 The additional duty and special additional duty on imports have continued to be part of the import-duty regime in different years. This is supposed to be in lieu of internal tax, in order to provide a level playing field for domestic producers. The tax system allows crediting these taxes back if the product is used as a part of the value chain and subsequently enters domestic sales as an input.
From time to time, the government has changed the additional and special additional duties, and imposed surcharges. For example, a special duty of 2 per cent was imposed in the 1996–97 Budget,41 which was raised to 5 per cent in September 1997. This was discontinued in the Budget of 1999–2000. Nonetheless, a uniform surcharge of 10 per cent of the basic duty was imposed on many products. The proportion of additional and special additional duties in total charges on imports at the border was small when tariffs were high. With a decrease in tariffs where bulk of them are not over 10 per cent, the significance of CVD and SAD has increased in the total charges at the
border.
(i) Devaluation/Depreciation of the Rupee
An important point to note is that increased protection is provided to domestic production because of high tariffs as well as depreciation of the currency. The value of the Indian rupee fell in comparison to the US dollar, at the same time when India’s average tariffs declined (see Chart 1). In 1991, the reduction in tariffs announced through the Budget was cushioned due to a devaluation of the Indian rupee, with a two-step downward exchange-rate adjustment by 9 per cent and 11 per cent between 1–3 July 1991 in order to counter the massive draw down in the foreign-exchange reserves, and to instil confidence in the investors and to improve domestic competitiveness.42 In March 1992, India established the ‘Liberalized Exchange Rate Management System’ (LERMS) with a dual exchange-rate system for one year.
Chart 1 shows that the period till 1992–93 saw a devaluation of the rupee, which more than compensated for the reduction in India’s trade-weighted average tariff reduction.
A unified exchange-rate system was established in March 1993, with the exchange-rate system moving from an adjustable peg to a market-based exchange rate for trading. As stated in the Budget speech of 1993–94:
… last year … a dual exchange-rate regime was introduced, and import licensing was eliminated on most items of capital goods, raw materials, intermediates and components. These items became freely importable against foreign exchange purchased in the market. The system has worked fairly well and the market exchange rate has been remarkably stable. The existence of a dual rate, however, hurts exporters and other foreign exchange earners who have to surrender 40 per cent of their earnings at the official rate, getting the benefit of the higher market rate on only 60 per cent. Many exporters have said that this amounts to a tax on exporters of goods and services whose continuation cannot be justified at a time when exports must receive our fullest support. There is merit in this point of view, and the experience of the past year gives ground for confidence that we can unify the exchange rate and still manage the balance of payments with a reasonable degree of stability in the exchange rate. The government has therefore decided to eliminate the dual rate arrangement. All exporters, as well as other foreign exchange earners such as our workers abroad, will henceforth be allowed to convert 100 per cent of their earnings at the market rate. All imports will also henceforth have to be paid for at the market rate. (para 26)
