The total cost of distribution comprises four elements: the expenses of the distributing units, the cost of delivery from the central procurement agency to the pharmacies, the cost of losses through theft or expiry, and the monetary value of the stock.

Reducing the cost of losses is in theory the easiest. Stock levels should be carefully monitored to find both misappropriations and surpluses, as the latter are more likely to pass their expiry dates.

Cash costs appear only when the dispensing unit borrows against interest or must repay the owners of capital. Public health services and non-profit-making organizations rarely need to do this. Even so, those units may still reduce their cash too far and thereby reduce the quantities of drugs they can purchase. Requirements for cash should in any case be limited. Apart from reducing deadlines for payment by debtors and extending deadlines for the payment of suppliers, the best way of limiting cash requirements is to reduce stock levels by increasing their rate of turnover. Stock turnover is the number of times stock is acquired and released in the course of a year. For a given quantity of drugs distributed, the more rapid the stock turnover, the less cash will be needed (Table 12). Smaller stocks also require less storage space. However, rapid stock turnover may increase total transaction costs if each turnover involves a separate transaction to replenish stocks.

The cost of delivery from the central procurement agency to the pharmacies obviously depends on distance, terrain, existing infrastructure and transport, and on the type of delivery route and carrier.

Table 12. Influence of stock turnover on cash costs (example)

There are three types of delivery route:

• circuit delivery (one vehicle supplies several consignees);
• linear delivery (one vehicle supplies only one consignee);
• star delivery (parcels are dispatched to various consignees by hired carriers).

Three types of carrier may be used:

• the vehicles of the central procurement agency;
• the vehicles of the recipient pharmacy or health service;
• independent carriers.

Independent carriers are quite suitable for delivering small quantities at frequent intervals to widely separated destinations in different directions.

Linear bulk delivery to a single destination is always the cheapest. It appears even less expensive for public services that use their own vehicles because only the fuel costs show in the accounts. Neither the cost of vehicles, bought by the government itself or with foreign aid, nor drivers’ wages paid by the government, figure directly in the operating budget. Independent carriers may in fact cost less overall as vehicles may carry a return load and both vehicles and drivers are employed full-time with no idle periods. Nevertheless, if the health services have drivers and vehicles for which they have already paid, using them (if they are available) may be the most cost-effective solution. Linear delivery is profitable only for full loads that deliver a large quantity at once. This is an incentive to deliver fewer, larger loads to regional depots. On the other hand, widely spaced deliveries increase the risk of stock shortages, and regional depots raise cash requirements and management costs by increasing service charges.

Circuit delivery is of interest only if the destinations are not very far apart. It is used mainly in towns or in densely populated regions with many pharmacies. The dispatcher’s vehicles are used as it is too complicated to organize this through an independent carrier.

These delivery methods may be used in combination and in conjunction with regional depots. Deliveries can first be made to the regional depots, then onward. Some countries have intermediate depots at regional, subregional and district levels. This is extremely costly because it increases the value of stock (and cash requirements) and management costs (logging stock in and out, ordering and paying, service charges, packaging in parcels). It may guard against shortages by keeping drug stocks close to the consumers, which is a real benefit in regions where transport is very difficult. As a rule, this system of depots at different levels is not economically feasible unless many costs-cash flow, buildings, personnel, vehicles and losses - are not counted.

The recurrent expenditure of distribution units can be substantially reduced if resources such as personnel and buildings are minimized. In the public services, the number of staff frequently exceeds the number required to do the work. The workload and minimum recurring costs of distribution units relate more to the number of items checked in and out of stores than to the value of the drugs distributed. Rather than reduce staff it may be more rational to monitor stock more thoroughly and to order and deliver stock more frequently. Both of these actions help to reduce losses and increase stock turnover. It is thus possible not only to increase distribution using existing resources but also to reduce the relative cost of distribution.

To all these costs could be added the cost of lost revenue owing to drugs being out of stock. However, this cost differs from the costs described above. The cost of stock shortages to the distribution unit depends on whether other units are able to distribute the drugs in question. The cost to patients is of a different nature. They may look for another pharmacy or go without treatment, with widely varying repercussions. These costs cannot be measured precisely, and their implications differ greatly from the implications of the cost of distribution. Distribution costs should therefore not be compared with the costs of stock shortages where the aim is to minimize shortages. The first objective can only be to eliminate shortages, and after that to minimize costs without causing further shortages.