Dr Matthew Stern and Yash Ramkolowan

One of the fundamental reasons for the creation of a customs union is to facilitate intra-union trade, specifically through the removal of border controls on goods between member countries. The removal of border posts not only facilitates trade but can catalyse the development of regional value chains and unlock new industries as the costs of transport and logistics fall. The oldest customs union in the world, SACU, has nevertheless made no progress in dismantling border controls between its member countries over a period of 100 years. To the contrary, national customs authorities are currently investing in strengthening existing facilities and procedures to the detriment of both freight and people.

In SACU, border controls are generally justified for the following three reasons:

1. Firstly, for VAT collection (and refund) purposes, it is essential to record the value of all goods as they move between SACU member states; and also, to ensure that all goods declared for export outside of SACU, actually leave the customs union.

2. Secondly, for data collection purposes, it has become important to record all direct trade between SACU members. This is useful for balance of payments reporting purposes, but it is required largely for the calculation of the customs revenue shares to be distributed between all SACU member states.

3. Finally, for health, safety and other regulatory purposes, border controls are needed to prevent trade in illegal, illicit or diseased goods and to monitor the movement of people.

The extent to which all or some of these controls are needed is discussed in more detail below.

The border as an instrument for collecting VAT

Lesotho, Swaziland and South Africa have already implemented a unified VAT reconciliation service at the border post, and discussions between other SACU members are underway. The implementation of such a system not only serves to lift the burden of VAT compliance from the trader to the state (and the appointed agency), but in doing so, it also eliminates the possible private benefits from fraudulent VAT declarations. Once this system is in place, there is no need to inspect and assess the value of bilateral trade for VAT purposes, and a single and simple declaration should suffice. Moreover, for formal traders, who are already required to submit such declarations prior to departure and on-line, there should be no need for any further declaration at the border post; except for some recording of the fact that the shipment has actually moved from one country to the other.

The only potential exception is the case of goods that are transhipped through SACU countries, either for export or import purposes. Controls are clearly needed to prevent these goods from leaking VAT-free into a neighbouring market, but this could simply be achieved through the establishment of a dry port within SACU countries (and in special cases, within specific export orientated firms). There would then be no need for seals to be inspected at the border as long as they are inspected upon arrival at the SACU port of final destination.

The border as an instrument for collecting trade data

If formal traders are required to report electronically on all exports to other SACU member states, then there should be no additional need to collect documents and data at the border post. Similarly, for informal trade, the VAT reconciliation agency at the border could be tasked to only collect and report on transactions above a reasonable threshold.

The main obstacle to the achievement of a streamlined data collection process arises from the apparent mistrust between the revenue authorities of SACU member states over the quality of data currently collected, and the disproportionate importance of this data for the calculation of customs revenue shares under the current SACU revenue sharing arrangement. Resolving this problem should not entail the imposition of more rigorous data systems at the border; rather, it requires a complete re-think of the design and calculation of the revenue sharing formula.

Enough has been written on the problems with the current revenue sharing formula (see for example Flatters and Stern 2005). It is however worth re-emphasising that this arrangement has contributed to increased barriers at SACU borders and has become a source of conflict for SACU Governments and a cost for traders. It follows that any change to this formula should not only seek to achieve improved equity and stability in the distribution of payments, but should also serve to facilitate trade and economic cooperation across the customs union.

If and when a new revenue sharing is in place, there are numerous different ways in which trade data can be collected for policy and balance of payments purposes. For example, the EU has a similar need for data on intra-regional trade for economic policy and infrastructure planning purposes. But unlike SACU, trade that moves between EU member countries is not subjected to customs controls and there is therefore no way nor intention to measure such flows at internal borders. Instead, the EU has established a clear, common and reasonably comprehensive way of collecting basic data on intra-regional trade in a way that does not rely on border controls and does not encumber trade (the Intrastat system). A similar system could be reproduced within SACU through any VAT reconciliation system that may eventually be applied.

The border as an instrument of regulation

Perhaps the greatest obstacle to the removal or reduction of existing border controls arises from South Africa’s concern about the possible flow of illegal or illicit imports (and people) through SACU member states. A large part of South Africa’s trade policy, as described in the Industrial Policy Action Plan III and the New Economic Growth Path, is dedicated to the strengthening of domestic standards and security in order to inhibit such imports.

While South Africa might have good reason to increase the policing of its borders, it seems very unlikely that SACU member states would be the main source of illicit imports into the customs union. And if this is the case, given the landlocked nature of Botswana, Lesotho and Swaziland, goods can only be entering BLNS countries through a limited number of ports and borders. Rather than fixate on an internal SACU border, it would make much more sense for South Africa and BLNS countries to concentrate their joint resources on securing the more porous border points between the customs union and other countries. Such an initiative would serve to protect the integrity of the Customs Union as a whole and not just South Africa.

With regards to any plant or animal health concerns, there should be no reason to assume ex ante that produce from BLNS countries are no more or less diseased than goods from parts of South Africa. If and when evidence emerges of any specific problems or if the BLNS countries’ SPS systems prove unreliable, then specific controls (and support) may be required. But such controls should be well-targeted and sufficient to address the underlying problem; they should not get in the way of trade in any unrelated produce or goods.

Finally, the border restricts (or at least records) the temporary movement of people between SACU countries. Although it would desirable to eliminate all barriers to the movement of people throughout the Customs Union, this is unlikely to happen anytime soon. That said, the existing system could be greatly streamlined, especially for residents of the customs union. To require residents and visitors to queue and be processed twice upon moving between members of a customs union, is excessive, and for transport companies this imposes a further delay and cost in getting their trucks across the border.

Ideally, a more flexible mechanism should be agreed and at the very least, some form of separate or fast-track arrangement should be established for professionals and registered transport drivers. This could be extended over time to all residents, and here SACU can learn from the experience of the Asian-Pacific Economic Cooperation (APEC) and Business Travel Card (ABTC). The ABTC allows business travellers pre-cleared, facilitated short-term entry to participating member economies with card holders also benefitting from faster immigration processing on arrival.

But why have a border post at all?

There has been much discussion in the region around the concept of a one-stop border post. In general, this involves replacing duplicate systems that are applied by duplicate revenue and customs authorities, with a single but connected system. In principle, such a system would greatly reduce the number of procedures and amount of time at the border post, and thereby facilitate the easier movement of goods and people between neighbouring countries. In practice, SACU and some SADC member states have been engaged in discussions around such an approach for almost two decades, but there does not seem to be sufficient will or trust to make this happen.

What these discussions seem to overlook, is why a border post which controls the flow of goods should exist within a customs union in the first place. The overriding purpose of a customs union is to allow goods and (and ideally services) to move freely between its members. A customs border post between members is a clear obstacle to this goal, and in the case of SACU, greatly reduces the benefits of the union to its members. SACU’s ambition should not be to establish more effective internal border posts; it should be to eradicate them in their entirety. The above comment and the international experience suggest that this can be done. But to do so requires a change in mind set and the right political will.