Is the HS still fit for purpose?

20 June 2018
By Gael Grooby, Deputy Director, Tariff and Trade Affairs Directorate, WCO

As the WCO celebrates the 30th anniversary, in 2018, of the Harmonized Commodity Description and Coding System, more popularly known as the Harmonized System or the HS, the question may be asked whether the time has come to contemplate a major overhaul of the system. This article looks at the structure of the HS, its changing uses, whether it is still a sufficient solution, how it can be improved, and whether a structural revision is now necessary.

What is the HS

The HS is a list of duties, right? Wrong! In many people’s minds, the HS is so closely tied to duties that it colours their perception of what is the fundamental purpose of the HS. In reality, the HS is a taxonomic system for goods: it provides a taxonomic hierarchy (the arrangement of various categories in successive levels of classification) and a nomenclature under which goods can be classified. So, what does this mean?

The group of ‘things that are traded’ is vast and chaotic – treated as individual items and given common names they would overwhelm our ability to know what goods are crossing national borders, or how they should be treated. Is a crown something that goes on your head or on your tooth? Or, maybe it is a part for a hat, or an old British coin?

As a taxonomic hierarchy, the HS creates an ordered system for the classification of goods, where more general classes of goods contain more specific classes of goods. Let’s look at a pair of 30 denier (a unit of measurement used to identify the fibre thickness of individual threads in cloth) nylon pantyhose. The hierarchy of classes for these goods in the HS is as follows:

  • Articles of apparel and clothing accessories, knitted or crocheted
    • Pantyhose, tights, stockings, socks and other hosiery, including graduated compression hosiery (for example, stockings for varicose veins) and footwear without applied soles, knitted or crocheted
      • Other pantyhose and tights (i.e. other than graduated compression hosiery)
        • Of synthetic fibres, measuring per single yarn less than 67 decitex

To specify the group of the descriptive words above would make for very long reports. So, the ‘nomenclature’ part comes into play and the above-mentioned categories are given numbers as names:

  • Chapter 61
    • 61.15
      • 6115.2
        • 6115.21

The HS classified the commodity into a group and gave it a name. A product named ‘6115.21,’ assuming correct classification, will be pantyhose or tights, not being graduated compression hosiery, which are made of synthetic fibres with a yarn weight less than 67 decitex.

The changing uses of the HS

The HS causes individual goods to be put into classes so that administrations can decide what to do about these goods as a group: charge a particular rate of duty, require an import permit, check for insect contamination, or whatever else is appropriate. It also provides a range of data on the trade in these groups, which can be turned into highly useful information. The uses of the HS stem from these two outcomes.

The first outcome, the actual grouping (i.e. the classification of goods), allows for differential treatment, and is particularly used for:

  • revenue collection – the assigning of duties and taxes;
  • support measures – allowing certain needed inputs, essential supplies or other goods to come in without the usual duties or taxes, or with a facilitated clearance;
  • restrictions – putting controls or requirements on goods, including health and safety, security and environmental controls;
  • trade protection – the application of dumping or countervailing duties, quotas or other measures to protect domestic industry;
  • trade agreements – agreed favourable treatment of goods of particular classifications;
  • trade sanctions – preventing the movement of certain classes of goods to or from a particular place;
  • assessing the level of transformation of a good as it changes classification – this is a common method of determining the origin of a good.

The second outcome, the collection of data, is just as important. For those countries with low levels of revenue collection, it can be even more important.

Trade data is used by governmental agencies, non-government organizations, intergovernmental organizations, businesses, journalists, academics, researchers, the media, politicians, and anyone else who has an interest in trade. It is used as an input into the compilation of balance of payments statistics and national accounts, and in the formation of trade and economic policy (including fiscal, monetary, structural and sectoral policies) as well as in multilateral and bilateral negotiations.

Such data is also used for other purposes such as modelling and forecasting, price indices, business cycle analysis, market identification, export performance, commodity performance, market share calculations, and the identification of trade patterns that might indicate illicit activity such as duty evasion, money laundering and the importation of illegal goods. To list the full range of uses that trade data can be put to is beyond the scope of this article. However, its numerous uses need accurate data on what is being traded – i.e. accurate classification. This is what the HS was designed to deliver.

Is the current HS still a sufficient solution?

This is a complex question to answer. The HS works. This is demonstrated by its success. It is used in more than 200 economies and has provided a level of knowledge of trade that has enabled good outcomes for Customs administrations and sophisticated trade analysis. As such, the HS is truly invaluable.

However, there are two principal aspects which indicate that its current format should be deeply revised. One is the ability to achieve consistency of decisions, and the other is adaptability to changing needs. In fact, the HS is a difficult system under which to achieve consistent classification, particularly with the increasing complexity of products. Thomas Reuters and KPGM International have undertaken annual surveys of trade professionals for the last three years. In their 2016 survey, 91% of their respondents found classification a challenge, and this remained true in 2017 too.[1]

Misclassification undermines the effectiveness of border controls, reduces the reliability of statistics, and often results in incorrect duty and tax payments. This is a problem that is not restricted to under‑resourced developing countries: misclassification has been a challenging problem to address globally.

For example, in 2010, the Auditor General of Canada reported that “According to studies undertaken since 2005, it is estimated that importers (…) have misclassified between 17 and 30 percent of releases.”[2] By the 2017 report, the Auditor General reported that “Over the last 15 years, the [Canada Border Services] Agency’s compliance verifications on specific goods revealed that importers misclassified imported goods more than 20 percent of the time.”[3] Audit results show it can be substantially higher for problematic classifications, and it appears that this is similar to results in other countries.

This problem is partly unavoidable, given the range and complexity of the world’s products. Yet it is worsened by the HS’s chain of development from the Geneva Nomenclature, itself derived from national tariffs, through to the Brussels Nomenclature, and into the current HS. At each step, most of the previously existing product definitions, grouping, structure and set-up were inherited.

For simplicity and consistency, a classification system would ideally have clear criteria for deciding on the membership of a group that resolves down to, as far as possible, yes/no questions on easily observable characteristics. This is particularly so for a system, like the HS, that is used by a vast number of private users, with different languages, backgrounds, and capabilities.

Instead, the HS relies heavily on classification by name, type or similarity. It adds another layer of definition and direction through legally binding Notes, and then wraps it in a set of General Interpretive Rules (GIRs) that use subjective considerations such as specificity and essential character. But, what does this mean? Take the example of heading 73.10: “Tanks, casks, drums, cans, boxes and similar containers, for any material (other than compressed or liquefied gas), of iron or steel, of a capacity not exceeding 300 l, whether or not lined or heat-insulated, but not fitted with mechanical or thermal equipment.”

The terms “of iron or steel” and “of a capacity not exceeding 300 l” are simple yes/no criteria, while the terms “for any material (other than compressed or liquefied gas)” and “whether or not lined or heat-insulated” clarify scope in a simple manner. The terms “Tanks, casks, drums, cans, boxes and similar containers” and “but not fitted with mechanical or thermal equipment” however, require identification against a name, or the assessment of similarity. While this type of requirement is not unusual for complex, real world goods classification, it does reduce the ability to ensure that the same decision will always be made. So, for example, questions like “is a biscuit tin a similar container” or “is a tap considered mechanical equipment” may be answered differently by different people.

The HS also mixes types of criteria. So, faced with the classification of an imported “container,” one has multiple headings to consider each based on different factors or combinations of factors: constituent material, capacity, function, fitted equipment, sphere of use (e.g., household or not), associated industry/activity, ability to be fitted to different types of vehicles, ornamentation, status as a part or accessory, and so on. Some of these choices are resolved by the HS Notes, others are guided by the applicable GIRs, but many require outside help from extrinsic aids.

Systems like this tend to rely heavily on extrinsic aids and previous decisions. In the case of the HS, this includes its Explanatory Notes and its Compendium of Classification Opinions compiled by the WCO, various other explanatory materials and rulings from national administrations, and review or judicial rulings from tribunals and courts. The end result of complex criteria and the heavy reliance on explanatory material is that the level of accuracy and consistency achieved in classification tends to be relatively low.

In terms of product definitions/groupings, many were developed during the first half of the last century. This leaves us with dated terminology (e.g., “automatic data processing machines” rather than “computers”), and a slant towards basic commodities and finished manufactured products. This makes it difficult to integrate both intermediate manufactures and multifunctional, composite goods that have become increasingly important in world trade over the last 20 years.

Goods such as ready prepared meals, medicalized foods (e.g., spreads with added cholesterol lowering phytosterols), wearable electronics, sub‑assemblies for global value chain manufacturing, and electronic devices such as smartphones, which combine multiple functions are all examples of goods that do not fit easily into the old headings. Developing good structures for such complex goods will always be difficult. Fitting new groupings for these types of goods into the existing structure greatly increases that difficulty. Not only must the grouping be defined, it must also be differentiated from existing groupings that work on the basis of different criteria.

Can we improve the HS?

Yes. After 30 years of negotiating harmonization and working to adapt the HS for different uses, we have a far clearer idea of where the system is under pressure. In the intervening years between the launch of the HS and today, there have been other attempts to develop a better classification system by both international statistical bodies and national governments. Thus, we have new models to study, which can give new insights into different ways of seeking to solve the same problem: categorizing trade.

It is possible to revise the structure and some design elements of the HS while still retaining the value and integrity of the system. The current HS is like a penny farthing bicycle, one can argue it has a certain elegance, but it is still not the ideal solution for the cycling commuter today. A modern bicycle may look different from the penny farthing, but the important design revolutions, such as a seat, wheels, peddles and handlebars, are still all there: it simply improves its operation. In the same way, we can modernize the HS by making it simpler to use, and able to better handle the changing trade environment.

Should we consider a structural revision of the HS?

The current HS has been in place for three decades. Simply because it has flaws does not automatically mean that it is necessary, or even desirable, to undertake major changes. It is deeply embedded into systems and procedures globally. The work in changing laws, software, documentation, databases, and training would be immense and disruptive.

Think of the replacement of imperial measurements with the metric system. Few would disagree that it is simpler to use a standard decimal system across all measurements than having, for example, 1,760 yards to the mile and 16 ounces to the pound. But, while simpler, changing is difficult and expensive, and requires people to ‘relearn’ parts of their basic skill set. In fact, it is so difficult that seven countries have not yet officially switched to the metric system.

The situation is similar for product classification. While there may be better ways to categorize electric toothbrushes than as “Electro-mechanical domestic appliances, with self-contained electric motor, other than vacuum cleaners of heading 85.08: Other,” that category is in our laws, in our databases, and in the memories of classifiers: current classification terminology is entrenched. Therefore, the question of whether the gains would be sufficient for the global community to accept the cost and disruption is not an easy one.

It is also important to remember that while a revised system may be an improvement, no system is perfect and it would not suit everyone or solve all problems. Under the HS Convention, changes require acceptance by all Contracting Parties to the Convention, so to make major changes would not be an easy path. Close cooperation between Customs administrations, other users and the WCO Secretariat would be vital to achieving an acceptable outcome. Based on past experience, the possibility is high that after much work, acceptance would still not be gained for the whole package.

Despite all of this, we should give serious consideration to initiating such a project. The world has had the HS for 30 years and yet, it remains difficult to use and the source of continued contention for traders, administrations, courts and countries. A deep and innovative revision that succeeded in making a generational change to the HS would be one of the most important global reforms to trade in decades. As it affects the everyday reality for traders and Customs administrations globally, increased ease of use of the HS would have a major impact on their operations.

To achieve major reform would require input from user groups. While the WCO’s HS Review Sub-Committee and HS Committee are excellent at looking at issues around specific classifications or products, they only meet twice a year respectively, and are not a forum in which a major holistic review can be realistically undertaken. Such an undertaking would require a working group, with varied representation, to examine the fundamentals of the system. This was the approach with the original development of the structure and content of the HS.

One potentially valuable starting point would be the consideration of commodity groups or HS areas where the resolution of classification disagreements between Contracting Parties is frequently prolonged or requires legal changes. Chapter 29 of the HS is an example of this. There are over seven million organic compounds to classify, and a compound can have multiple plausible classifications. The current structure of the chemical chapters uses diverse methods on which to base headings, such as:

  • common functional groups;
  • constituents or chemical classes;
  • carbon range numbers;
  • chain-length categories;
  • common precursors and/or breakdown products, via physical or biological processes (e.g., acid/ester/salt);
  • the activity/function of a compound.

So, a single compound may fit into the terms of different headings based on different aspects of its identity and the general rules of specificity or essential character often have little meaning in this context. Over the various revisions of the HS, ‘add-on’ changes have been made to prioritize one heading over others for various specific chemicals or groups of chemicals, but it still causes confusion for the trading community.

The underlying assumptions about the role of parts and accessories is also worthy of re-examination. In the context of global supply chains and multi-use components, a lot of work and effort goes into determining the future goods something may be part of, or used with, rather than what it is when actually crossing the border.

There is also the future question of goods whose value and identity is in the embedded information, and how to provide parity of treatment between virtual and physical delivery. Does a novel have a truly different identity depending on whether it is presented as a printed copy or an electronic copy? What about the difference between an imported sculpture and the same piece downloaded as a single use file and 3D printed by the buyer?

We know the difficulties that people face with the HS and believe that we would be able to make it more user-friendly. Changes ranging from the simple, such as grouping notes that have nomenclature wide application, to the fundamental, such as a glossary with agreed language translations for terminology along the lines of the International Electrotechnical Commission’s ‘Electropedia’ could be considered.

It is time to consider adapting the HS for the 21st Century.

More information

[1] Managing Global Trade: A Look Beyond the Surface, retrieved on 22 February 2018 from the online version at

[2] Auditor General of Canada 2010, ‘Report of the Auditor General of Canada – Fall 2010: Chapter 8, Facilitating the Flow of Imported Commercial Goods – Canada Border Services Agency

[3] Auditor General of Canada 2017, ‘2017 Spring Reports of the Auditor General of Canada to the Parliament of Canada: Report 2—Customs Duties