HomeTop StoriesPoverty Tariffs - A Novel Implementation

Poverty Tariffs – A Novel Implementation

Poverty Tariffs – A Novel Implementation

The primary objective of the Poverty Tariff is to supply a preset amount of electricity to all the people of South Africa at pre-determined period frequencies, free of charge.

Most manufacturers and solution providers have been busy developing solutions for the industry in recent months. The challenge is to provide a free amount of electricity (and possibly other commodities) to all users by developing a solution which will have minimum impact on the current infrastructure, both in terms of cost and implementation time.

The constraints are varied.

  • Existing infrastructure. A large infrastructure for vending is in place throughout the country, and it would be of great benefit if the solution could use that system.
  • Group Coded Areas. The solution must be able to satisfactorily handle areas where the meters are all coded to one vending key and tokens are not meter specific.
  • Proprietary Systems. The system may not be able to function for all proprietary meters, but it should at least cover most of them.
  • The STS System. The system must work under the current STS data transport method, because of the installed base.


Arrears Recovery
One solution is simply to give consumers free tokens and then charge them at a later date for the ‘extra’ tokens they have claimed by a process of reconciliation. This has some obvious flaws and is open to abuse. If a consumer claims a token from each of, say, ten vending points, it may not be necessary for him to return to buy a token for a few months. During this time the utility is losing revenue.

Administrative Systems
This type of system is based on a paper issue token, a coupon, or a credit note, issued directly to the consumer through a clerical/administrative infrastructure. 
Although cost-effective to implement and beneficial in terms of job creation, it tends to be difficult to manage. Its implementation would, however, have minimal impact on the current vending infrastructure.

Smart Card Solution
Many solutions exist based on a smart card. Some of them require the meters to be adapted to accept the card as the token transport mechanism. These solutions are ignored for the moment because of their obvious logistical and cost implications.

Another solution is the use of a ‘transient database’. This would allow users to carry a record of their transactions with them in their smart/memory card. The free issue token would only be issued on presentation of this card and verification of previous transactions at any vending point.

The implications are the issue of the cards themselves (a cost to the industry) as well as the deployment of card readers at all vending stations. The advantage of such a system would be the introduction of new technology, possibly in line with the National Electronic Identity Card (NEIC) project for South Africa. Its aim is to provide every citizen with a multi-function smart card as an identity card.

Online Vending
From the point of view of technology, the online system is the most attractive as the longer-term solution. In the short term, however, it would be far too expensive to implement. The existing infrastructure would effectively have to be abandoned or completely re-engineered. In addition, the real issues associated with communications link failures could seriously impact on the performance, and hence acceptance, of such a system. 


The STS credit token is made up of various fields:

  • Class 
  • Sub-Class 
  • Random Pattern 
  • Token Identifier 
  • Amount 
  • CRC

The Token Identifier field’s primary use is to guarantee that a token that has already been used in a dispenser is rejected. This forms the basis for our solution.

The Token Identifier is made up of 24 bits, representing the number of minutes elapsed since the 1st January 1993, 00:00:00. The resultant cipher text uniqueness is guaranteed by the fact that the time does not repeat. We used this fact to develop the Static Token


The time of day and date are zeroed to create a token that is always at, say, 00:00:00 on the first day of the month. This leaves the month as a variable. If two tokens with the same time and day are generated, only one will be accepted by the meter. This allows this type of token to be generated at a multitude of vending points. If we assume that the period for the free commodity is monthly, the month portion of the date string is allowed to change. If the period is defined as daily, then the relevant portion of the time string would be allowed to vary, with the rest being fixed. At the period change, a new Static Token will be generated, thus allowing the user to claim only once per period.

If this seems too simple to work – it is simple and it does work! 

The Static Token will handle all proprietary systems where the token is based on a derivative of the STS system (i.e. STS with proprietary keys). It can also be applied to any token that uses the date and time as a token ID. 

The vending units of most vending systems record both the units sold and the money received for each transaction. When Static Tokens are issued the operator receives no money, so the amount is zero but the number of units recorded is non-zero. The batch totals will still match the money actually taken by the operator; and Poverty Tariff tokens can be identified from the transaction history as pre-paid sales with a zero value and non-zero kWh. 

The Static Token system requires no changes to the STS or STS-derivative proprietary electricity meters. It can be implemented relatively easily by a simple upgrade to existing vending software. An additional option in the vending process could be added whereby the operator would select a Poverty Tariff token instead of a standard token. The value would be pre-set. This system makes it possible to implement the Poverty Tariff on the existing vending infrastructure with relatively small changes to the vending units, and low cost to the system. It does not preclude the use of other solutions, however, and each should be investigated for its merits.