Policy Research Institute - PRI Bangladesh

The Policy Research Institute of Bangladesh (PRI) is a private, nonprofit, nonpartisan research organization dedicated to promoting a greater understanding of the Bangladesh economy, its key policy challenges, domestically, and in a rapidly integrating global marketplace.

Putting tariff anomalies in proper context

Published: Tuesday, Oct 07, 2008


Putting tariff anomalies in proper context

Zaidi Sattar
October 07, 2008

The other day a close friend was whopping mad at National Board of Revenue (NBR) for having imposed the highest duty rate (25%) on the principal imported ingredient used in his state-of-the-art production facility that manufactures a leading brand-name food product. With inputs costs up substantially, his modern production facility lost the competitive edge due to local processing, and his final product lost significant market share. This might or might not be an instance of tariff anomaly. Let me explain.

Starting with the budget of June 1992, NBR embarked on a mission of tariff reform with the advice of international experts. A good part of that mission was to remove the many knots and wrinkles in the tariff structure in order to make it consistent, more uniform, and transparent. That, said experts, would contribute to efficiency and go a long way in minimizing malfeasance in customs administration. In those days, a frequent refrain that was heard from the business community ran as follows: there were too many "tariff anomalies", with input tariffs often higher than tariffs on the final product. True, that was illogical and undermined profitability. Oftentimes, these anomalies were created by acceding to the demands of one enterprise at the expense of another. In this brief note, let me try and put all this in proper context.

Up until 1991, as much as 40% of all importables were subject to some form of ban or quantitative restrictions for trade as well as non-trade reasons; nearly half the tariff lines had tariff rates of 100% or more, yielding an average rate of about 90%; and there were no less than 20 tariff bands. Moreover, there was no systematic treatment of raw materials, intermediate goods, and final consumer goods for the purpose of tariff imposition, though a general notion of high protective tariff on anything that was produced domestically was accepted as the norm. With an eye on revenue, high tariffs were frequently imposed on high volume imports regardless of whether the commodity was an industrial input or output. The more cogent reason for high tariffs was to protect domestic industries. Needless to say, in those conditions, tariff anomalies - where input tariffs exceeded output tariffs -- would not be hard to find. Actually, revenue suffered often due to tariffs being so prohibitive as to restrict imports substantially.

We are aware that the budget announced in June 1992 was a watershed in the history of policymaking in Bangladesh. Drastic changes were announced in the country's trade and exchange rate policy as the economy moved to embrace greater trade openness. Tariffs were cut sharply while many bans and quantitative restrictions on imports were lifted.

We have come a long way since those days of prohibitively high tariffs. Now, the top general rate is down to 25% (plus selective imposition of supplementary duties), non-zero tariff bands are down to only four (3, 7, 12, 25), and, as a general rule, each band is targeted to a certain classification of imported commodities: 25% for final consumer goods, 12% for intermediate goods, 7.0% for basic raw materials, and 3.0% for capital machinery. This step-wise imposition of duties incrementally at every stage of processing is described as tariff escalation, though economists will tell you that this is not necessarily the most desirable structure of tariffs. The idea is to offer protection to every stage of processing. Basic raw materials, at the first stage of production, have the lowest rate of duty, followed by 12% rate on intermediate inputs which fall in the category of semi-processed materials. At the final stage of processing are final consumer goods subject to the highest rate of 25%.

The problem arises when this general scheme is deviated from. In particular, a tariff anomaly would occur if there is an attempt to protect a product that is semi-processed and falls in the intermediate goods category. Occasionally, high tariffs might also be imposed on high volume imports with a view to raising revenues. It turns out that in the present tariff structure, roughly a quarter of all imports of such commodities, which are basically industrial inputs, have duties of 25% (with occasional supplementary duties added). I suspect, protection rather than revenue is the predominant motive behind such deviation from the norm. Examples of such semi-processed intermediate goods would include gray and Portland cement, spun yarn, wire, glass, wide range of industrial chemicals, products of steel re-rolling mills and the cable industry, etc. The upshot of this scheme is that many downstream industries have to face high cost of their inputs which undermine their profitability.Complaints of tariff anomalies are the predictable result.

In this discussion, I have avoided using the popular terms, "raw materials" and "finished goods", following a more technical specification instead by categorizing products in terms of their stage of processing for the simple reason that what is "finished product" for one industry might be "raw material" for another.

What is relevant for the entrepreneur is effective rather than nominal protection. Effective protection is the final outcome of tariffs imposed on outputs and inputs. All enterprises know that higher the tariff on inputs they use lower the effective protection, hence profitability, enjoyed by them. Thus imposing protective tariffs on an "intermediate goods" industry undermines protection to the producer of final consumer goods giving rise to complaintsof tariff anomaly. This deserves some elaboration.

In developing countries, stylized facts indicate that, at the first stage of industrialization, the typical manufacturing base is dominated by a range of consumer goods production - so-called import-substitutes -- often with the help of protective tariffs and other import restrictions. It is at the next stage that one sees growth of intermediate goods industries which supply inputs to producers of consumer goods - a sort of backward linkage for the economy as a whole. When entrepreneurs in this sector call out for the same kind of protection, the conflict begins. You cannot protect intermediate and final goods production without pitting one against the other. If revenue objective is the source of the anomaly, it can be remedied at the expense of revenue. But conflicts between protection objectives are difficult to reconcile. In Bangladesh, consumer goods production dominate manufacturing. Hence, their complaints are loud when faced with occasional high protective tariffs on intermediate inputs.

In Bangladesh, the intermediate goods segment of industry is growing. In countries like India and China, which have large industrial base with equally large intermediate goods segments, they have resolved the issue by moving towards uniform tariffs. With few exceptions, the top general rate for industrial tariffs in both countries is around 10%, with intermediate and final goods tariffs close to each other. Thus the so-called spread between input and output tariffs has been squeezed down to a minimum or is non-existent. This is described in trade policy literature as the "concertina" approach. Bangladesh, in contrast, has been moving away from such a strategy. The spread between intermediate and final goods tariffs has recently increased, not decreased. Consequently, effective tariff protection generally has gone up - a sop to domestic industry. Consumers end up being taxed for their inefficiency.

There is another potential source of tariff anomaly - or, is it? The Harmonized System nomenclature, an internationally standardized system for classifying traded products, is logically structured by economic activity or component material. There are over 1200 four-digit headings describing distinct products under this system. A change of heading implies substantial product differentiation or transformation. Within a heading, however, items are similar and might differ by degree of processing only. For instance, HS-4 heading 1507 lists Soya bean oil, both crude and refined. The processing done in a refinery does not significantly transform the product so as to be listed in a different HS-4 heading. Likewise, many food preparations involve processing of ingredients that might be listed under the same HS-4 heading.

Bangladesh's tariff structure has been moving in the direction of uniformity of rates since 1992, by reducing the number of tariff bands and avoiding multiplicity of rates within each tariff heading. Consequently, it is quite possible that, in many food processing industries, such as edible oil, milk or malt-based drinks, tariffs on inputs and outputs could be identical. That does not mean zero protection, as seems to be the popular notion. One cardinal principle of protection is that equivalent tariffs on input and output result in effective protection being equal to nominal protection. To exemplify: if tariff on both input and output is 25%, effective protection is also 25%, not zero. The fact that in such cases there is also a supplementary duty applied on top of customs duty indicates more protection (over 25%), not less.

It is practically impossible to protect the group of products described as basic raw materials, except when they are agricultural produce. In the current scheme of tariffs, average agricultural tariffs tend to be higher than industrial tariffs (28% versus 18%). The principal agricultural activity in the country - cereal production - receives no tariff protection as pro-poor policy warrants maintaining zero or negligible tariffs on rice and wheat. Livestock, fisheries, poultry and forestry end up getting most of the protection in agriculture. If our farmers had the lobbying power of our business chambers, or the political clout of their brethren in Europe and America, they would have screamed tariff anomaly till the heavens came down!

Finally, it is pertinent to point out the implications of maintaining a low concessional rate (presently 3%) on capital machinery in order to lower the capital cost of a wide range of investments. Products of the indigenous light engineering industry located in Zinjira and across the country fall in this group. As such they could complain of tariff anomaly as most of their inputs are subject to higher rates of tariffs.

The long and short of this story is that under the current tariff regime of few tariff bands and assignment of tariffs generally by product categories, incidents of tariff anomalies where input tariffs exceed output tariffs are rare. This might happen when an intermediate goods industry is being protected or a high volume import is subjected to high tariffs with a revenue objective. Tariff anomaly of the second kind where input and output tariffs are equal might occur in cases where the degree of processing is not sufficient to transform the product enough to move it to another tariff heading (change of heading). Tariff anomalies do exist in respect of agriculture and light engineering sectors, but one does not hear much about them.

Complaints of tariff anomalies will wither away once we move to a low but uniform tariff regime. I would like to believe that day is not far off.

(Dr. Sattar, a former civil servant and World Bank economist, is Vice Chancellor, The Millennium University)