Transforming Supply Logistics for Health Commodity Security in Africa

Reconfiguring Public-Sector Monopoly

Public-sector logistics systems in African countries can be described as monopolies because they handle a predominant share of health commodity supplies without the risk of losing business to competitors providing similar services.³ Contrary to what one would expect of a monopoly, public-sector supply logistics also lack adequate capital and human investments to provide uninterrupted supplies of health commodities.⁵ Nevertheless, most attempts at improving how public-sector logistics monopolies work have centered on the inefficiencies of having multitiered logistics systems and the underinvestment problems created by spreading limited public financing over multiple tiers or echelons.

The Table shows different multitiered hub-and-spoke arrangements in select African countries based on published data.^5–12 The 1-tier and 2-tier arrangements can be considered variations of the 3-tier design in which health commodities are stored and distributed from central medical stores (CMSs) to regional medical stores and then to district medical stores before delivery to health facilities.⁵ These designs are different solutions that countries have pursued or adopted to improve the performance of public-sector supply logistics. Therefore, these designs should not be seen as a prescription for all African countries; if anything, they can be improved. As Clarke and Scarf argued,¹³ the optimal solution (defined in terms of order quantities, output levels, and total costs) for a logistics system with a certain number of tiers or echelons can be replicated by an alternative system with a lower number of tiers or echelons, what is generally referred to as the “system design approach.”¹⁴ As Fritz et al. reported,¹⁵ even a simple system-design change, such as switching from manual to electronic logistics management information systems by reducing the frequency of stock-outs of lifesaving health commodities, lowered the mortality rate of newborns and children aged younger than 5 years in Ethiopia, Mozambique, and Tanzania (a reduction of 1.6% to 4.1%). As much as system-design change is desirable, health planners may have to settle for second-best options (and accept efficiency losses) where change is not feasible or there are limits on what can be implemented.

For example, in attempts to redesign a 3-tier public-sector logistics system that is structured to match or reflect the tiers or echelons of governance or public administration, an efficiency loss will have to be accepted if ignoring or working around administrative boundaries is not politically feasible. This is most likely in decentralized or conflict-prone environments. Efficiency losses can appear even in situations in which administrative boundaries can be ignored. The problem is that health planners typically do not have ex ante knowledge about the long-run average cost curves for their logistics systems, nor do they know whether these long-run cost curves are continuous or discontinuous. When such prior knowledge exists, it can be used in planning. However, when that knowledge is not available, it is advisable for planners to conduct baseline costing exercises to gain a deeper understanding of logistics costs and the key drivers. These exercises should be followed by continuous experimentation (trial and error) to determine the optimal scale and scope of operations. Experimentation could start by evaluating several discrete design sizes, such as direct delivery from the CMS to health facilities or a 1-stop (2-tier) delivery strategy. This recommendation fits with the results of a randomized trial conducted in Zambia¹⁶ that showed direct delivery and a 1-stop (2-tier) delivery strategy that involved a commodity planner who coordinated supply logistics at the district level performed better in reducing the frequency and duration of stock-outs than a 2-tier design that did not employ a commodity planner. In this study, direct delivery to health facilities from Zambia’s Medical Stores Limited—a parastatal or quasi-government CMS—used the district medical stores as a transit (cross-docking) point but not for storage, picking, and packing.¹⁶

Conducting randomized trials in each low- and middle-income country to identify the optimal system design is a rather costly venture. In fact, some might argue that randomized trials are not required because there are alternative ways of experimenting and searching for the optimal system design. The first approach applies a multicriteria decision-making process (analytic hierarchy process) to rank alternative logistics arrangements based on the priorities and preferences of relevant stakeholders. This approach has been used in Morocco.¹¹ A second less sophisticated approach constructs composite indices of volumes of throughput, size of populations served, storage capacity, and distance to service delivery points to determine tiers, distribution centers, or warehouses that need to be eliminated.¹⁷ A third approach uses mathematical programming or simulation modeling to assess differences in outputs and costs associated with different system designs. This requires prior estimates of logistics costs and an understanding of cost drivers. Activity-based costing, which involves measuring the costs of all activities and steps that comprise a task, will be particularly useful in generating credible parameter values for mathematical programs and simulation models. An example is the simulation model by Lee et al.¹⁸ Using a sample of 35 countries with 3-tier arrangements (or 4-tiers if including health facilities), the study showed that a simplified 2-tier arrangement with 7 storage hubs or depots could lower operating costs to the extent that, even if larger capital expenditures are required, this design change could still have lower logistics costs per dose of vaccine delivered. Note that this result does not include savings in capital expenditures achievable through redeployment of nonspecific (storage, warehousing, and transportation) assets during the system-design change.

Increasingly, results of mathematical programming or simulation modeling of a finite number of scenarios are used to inform pilot projects before nationwide scale-up, as in Benin, Mozambique, and Niger. Application of these techniques helps reduce the amount of resources expended on experimentation. The scenarios analyzed do not have to be restricted to considering a reduction in the number of tiers; other configuration changes, such as delivery frequency, transportation modes, and routes, can be considered simultaneously.^19–21

Next, I argue that although recent experimentation and pilot projects to improve public-sector CMS logistics are a step in the right direction, these efforts are not enough. System-design change should be defined by a continuous search for the optimal scale and scope of operations that ensures 100% availability of essential health commodities in all health facilities most of the time. Do public-sector CMS monopolies have adequate incentives to continuously search for ways of reducing costs and improving outputs?

Introducing Competition and Self-Financing

As with private wholesalers, well-functioning and efficient public-sector logistics systems have to run an appropriate scale of operations. Over time, they have to secure a sequence of short-run prices for inventory management (i.e., distribution markups) that allows full recovery of operating costs plus the costs of all continuing and repeated investments made to ensure commodity security. If the costs cannot be recovered in the long run through a series of short-run prices, public-sector logistics institutions will have to find alternative financing (e.g., taxes and loans) to cover all the costs of ensuring commodity security. The alternative is injecting public financing to cover costs, but this weakens incentives to pursue efficiency, as public-sector institutions simply spend the money they receive. Although activity-based costing can be used to estimate inventory management costs and identify pockets of inefficiencies within the logistics system, public-sector logistics systems rarely conduct or are rarely required to carry out such costing exercises. In contrast, the quest for survival and/or the competitive pursuit of profits by private wholesalers generates strong incentives to conduct such costing exercises, lower costs, and improve outputs.⁴ Therefore, public-sector logistics systems will require more than just extra funding and resources to improve outputs to desired levels (e.g., ensuring near 100% order fill rates, shorter supply lead times, less frequent stock-outs, shorter spells of stock-outs, and near zero stock expired or damaged).

Public-sector logistics systems (which are more or less public wholesalers) should be allowed to compete with alternative producers of the bundle of services they provide if health planners want to generate strong incentives for the continuous search for new ways of reducing costs and improving service levels (with or without the aid of mathematical programs and simulation models). The “new ways” could be identifying the optimal location, size, and number of warehouses; providing extensive staff training and supervision; using electronic information management systems for more routine data collection; using logistics data for decision-making and evaluation; and automating and experimenting with alternative inventory control policies, transportation modes, distribution routes, and delivery schedules. However, the mere presence of competing alternatives will not put adequate competitive pressure on a public CMS monopoly. In the absence of repeated requests or calls for last-mile delivery and shorter delivery times or generally, if health care purchasers and providers are not sensitive to prices and/or outputs, there will be little or no reason to lower costs and/or upgrade logistics infrastructure for higher service levels even if the money and resources needed were available. Given the emergence of competitors, health care purchasers and providers will have to be sensitive to both price and output to break the dominance of a CMS monopoly.⁴

Subjecting public-sector logistics systems to competition is not the only way to improve logistics performance. Results-based financing (RBF) of public-sector logistics systems can improve performance if these institutions have full discretion as to how performance-tied incentive payments are used. The additional funding, shared with logistics staff and invested in logistics infrastructure, provides strong incentives to improve specific, verifiable, achievable, and time-bound performance indicators. As Spisak et al. reported,²² a 1-year pilot RBF of the CMS in Mozambique improved the number of reports submitted on time and the percentage of inventory records that match physical inventory counts, reduced the time from receipt of orders to completion of distribution planning and from completion of distribution planning to delivery of commodities, and improved the percentage of health commodities delivered that match the packing list in distribution plans. Thus, through RBF of the CMS monopoly, a lot can be achieved without alternative competitors. So, why opt for competition?

In contrast to RBF or performance-based contracting with a single CMS entity, competition requires health planners to enter into a number of performance-based contracts with a multiplicity of logistics providers. To encourage continued efficiency gains, contracted logistics providers who cannot ensure health commodity security and/or charge lower prices for inventory management should lose business to those who can. The difference is the risk-reduction benefits of having more than 1 logistics unit. When a fire crippled Ghana’s CMS, supply disruptions were mitigated by international (nongovernmental) agencies building a parallel logistics system that used a private-sector logistics provider to maintain continuity in health commodities supply.¹² Having 2 or more logistics institutions capable of full-line stocking and last-mile delivery is an insurance policy against unexpected catastrophic supply disruptions. In contrast, the CMS monopoly model offers no scope for hedging supply disruption risks but ample room for deviations from best standard practices in frantic attempts to reverse the negative impacts of disruptions when they occur.⁴

To promote choice and competition, health planners may add another public-sector logistics system to the existing CMS monopoly. In the Democratic Republic of the Congo and Nigeria, where there are 2 or more CMSs (Table), it might be possible to carve out at least 2 competing public-sector logistics systems from the existing infrastructure. However, given current underfunding problems, creating 2 or more public-sector logistics systems is questionable, especially when similar logistics systems exist or can be built in private markets. An alternative solution is introducing competition between segments of public-sector logistics systems. For example, Watson and McCord²³ discussed alternative arrangements to a CMS monopoly that included having a competing or complementary parallel CMS. Putting aside concerns about redundant capacity and nonoperational supply lines, that arrangement only offers scope for competition between the central warehouses with little or no effect on supply disruptions that occur at tiers below central warehouses. Such downstream disruptions can be minimized by reducing or removing the tiers below the CMS level, but this will not be enough if disruptions affect all central warehouses or medical stores. Instead of increasing the number of subunits within (tiers of) an existing CMS monopoly, a separate competing logistic pipeline is needed.

There is a pro-competition solution that maintains adequate protection against supply disruptions. The first step is transforming a public-sector CMS monopoly into a full-line national entity capable of supplying all health commodities needed to all health facilities in private, government, and nongovernmental sectors. Indeed, the current wave of integrating public-sector logistics pipelines for vaccines and non-vaccines and pharmaceuticals and non-pharmaceuticals in African states and other low- and middle-income countries²⁴ is a trend toward full-line supply. Where such public-sector CMS offering integrated, full-line logistics exist, they should be reconfigured to provide national coverage by serving all government, private, and nongovernmental health facilities located in different geographical areas. This requirement should not be viewed as attempting to force 3 sectors into a single system. Despite their commonalities, objective functions, institutional structures (for procurement), and incentives differ across these sectors. The requirement only ensures a set of competing logistics providers have access to the largest possible country demands for health commodities by serving public, private, and nongovernmental sectors. As one of these logistics providers, a transformed public-sector CMS should be able to secure a sequence of prices for the services offered, such that in the long run, they can cover their operating costs and capital investments made without public financing. Self-financing full-line national supply logistics systems may not be feasible in all country contexts. But where it is feasible, health planners should consider turning the transformed public-sector CMS into a fully private or quasi-government unit by ceding management to private agents or through establishing public-private ownership. The choice between transformations of a CMS monopoly into a fully private or quasi-governmental entity will depend on technical, administrative, and political feasibility. For example, having a quasi-government status will be a more workable solution in the face of opposition from proponents of government ownership of supply logistics.

An example of the solution proposed is Zambia’s Medical Stores Limited, which has a mandate to provide last-mile delivery and move to a “fully commercial, self-financing model.”¹² More recently, in June 2020, Benin’s Central Purchasing Office for Essential Medicines and Medical Consumables underwent a transformation into a public limited company, the Societé Béninoise pour l’Approvisionnement en Produits de Santé (SoBAPS, Benin Society for Health Product Supply).²⁵ Currently, there is no evidence on the performance of the SoBAPS, and Zambia’s Medical Stores Limited has been the subject of a number of system-design changes.¹⁶ Extending these examples to other African countries will require cautious implementation through pilots and careful modifications to reflect differences in local contexts.

An equally important point is that simply privatizing a public-sector CMS monopoly is not the answer. Transforming a public-sector CMS monopoly into a full-line national private (quasi-government) entity should be integrated with attempts to turn existing logistics institutions in private markets and nongovernmental voluntary sectors into credible competitors. Because the competing private or quasi-government logistics units are meant to be self-financing, health planners need to create a level playing field for all competitors, whether private or quasi-government. For example, a quasi-government wholesaler should not benefit from any preferential tax rules unless similar provisions are made for their private competitors.

An alternative strategy is to skip the step of transforming a public-sector CMS and instead focus solely on developing a set of competing full-line national logistics providers in private markets. Existing public-sector logistics infrastructure will then have to be decommissioned and nonspecific assets deployed elsewhere. Because decommissioning a public-sector CMS is unlikely to receive strong political support, the views expressed in this article favor transforming a public-sector CMS monopoly into a full-line national logistics provider and building competing alternatives to the transformed CMS from private markets.

Private Markets in Africa Failing to Ensure Commodity Security

In Africa, underfunded public-sector logistics systems sit next to small-scale, short-line, subnational private wholesalers who do not serve rural or remote areas and survive on high margins on commodity acquisition prices. According to the World Health Organization,⁵ Uganda has 100 officially registered wholesalers, with 12–14 being the “big players”; Nigeria has 292 importer-wholesalers; and Zambia has 30 licensed wholesalers, with business volumes concentrated among 5 or 6 wholesalers. In Francophone Africa, the average is 5 wholesalers/importers per country, with the exception of Burundi (14) and Rwanda (32). In Kyrgyzstan, there are 200 private wholesalers, with only 10 considered “volume leaders.” In contrast, the average across Organisation for Economic Co-operation and Development countries is 3–5 large private wholesalers who can deliver medicines several times a day to ensure health commodity security.⁵ According to Palafox et al.,²⁶ the multiplicity of formal and informal private antimalarial wholesalers in Benin, Cambodia, Democratic Republic of the Congo, Nigeria, Uganda, and Zambia is such that health commodities go through 4–6 steps (2–3 steps in most cases) in moving from manufacturer to retailers. Intermediate private wholesalers selling health commodities to each other before selling to end users is evidence that they are mostly short-line and/or subnational entities. Although this can be described as multiplicity, it does not encourage competition in ensuring health commodity security or minimizing supply disruption risk (Figure 1).⁴

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1

Transactions of Health Commodities Between Short-Line Subnational Private Wholesalers Operating Within Regions of a Given Low- and Middle-Income Country in Africa^a

Source: Author; Yadav et al.,⁵ Palafox et al.,²⁶ Rosen et al.,²⁷ McCabe et al.²⁷

The system of numerous middlemen or brokers undermines health commodity security in 2 ways. First, serial markups levied on manufacturer selling prices make end-user prices unaffordable. Private wholesaler markups in African countries range from 2% to 380% such that manufacturer selling prices could represent as little as 8% of final patient price.²⁷ Second is unreliable service levels. For example, private wholesalers in urban areas in Ghana receive orders from health facilities, but those facilities in periurban and rural areas (in the Northern regions) have to travel to commercial areas or hubs in the capital and other major cities.²⁸ As Palafox et al. reported,²⁶ the practice of “way billing” (i.e., billing for transportation) in which health commodity orders/supplies are placed on bus or mass transport routes for customer pickup at regional bus or taxi stations is quite common in African countries. The small scale of operations of private wholesalers in Africa is such that they do not usually offer last-mile delivery services, so their customers have to pick up the commodities themselves. The adage “if the drugs [health commodities] won’t come to you, you go for the drugs [health commodities]” is a sign and symptom of gaps in commodity security. These gaps often invite individual salesmen to fill their vans with health commodities, traveling countrywide to sell these commodities to rural pharmacies and chemical sellers.²⁸

Some might argue that numerous differentiated short-line private wholesalers are more suited for the supply of different health commodities with different storage requirements (some need cold-chain equipment and others do not). But given a choice between full-line national wholesaling (a set of long-distance distributors running up and down the country) and short-line subnational wholesaling (a set of short-distance distributors operating locally or regionally), the latter arrangement is more likely to result in gaps in commodity security. Indeed, 1 private-market response to these gaps is the emergence of “one-stop-shop” middlemen who recognized that the numerous short-line importers and wholesalers do not offer the wide range of health commodities needed to meet all demands in different geographical locations.²⁸

Steering Private Markets Toward Public Interests

There is some evidence that private wholesaling in Africa can move from what is shown in Figure 1 toward a state of prudent multiplicity without government intervention. In an attempt to broaden its product portfolio and scale of operations, a company called Gokals (Laborex Ghana) merged with or acquired several small-scale private wholesalers, demonstrating that it is possible to consolidate small-scale private wholesalers across different regions of the country into national full-line entities or at least logistics institutions with a broader portfolio of health commodities and a wider geographical coverage. That said, Gokals holds exclusive distribution rights over selected health commodities sourced through Laborex, raising some anticompetitive concerns. These exclusionary arrangements may be a response to low expected business volumes and/or attempts to cover the additional (quasi-fixed) costs of getting health commodities in-country within 24 hours. Supply lead times of small-scale private wholesalers are a lot longer as they rely mostly on sea routes in sourcing their products.²⁸ However, the emergence of similar large-scale private wholesalers in Ghana will require revoking any existing exclusive (geographic) distribution rights for health commodities because it works against competition and acts as a deterrent to market entry and growth of other private wholesalers aiming to transform their businesses into full-line national logistics units.

Generally speaking, government intervention (legislation, policy, and regulations) will be needed to initiate the process of transforming a public-sector CMS monopoly and steering private wholesaling toward the societal goal of ensuring health commodity security. Governments should at least accelerate efforts made by private wholesalers on their own toward full-line stocking and national coverage. For example, the emergence of a full-line national self-financing quasi-government wholesaler may create incentives for private wholesalers to transform themselves into full-line national units. That process could be hastened by removing legislation that requires exclusive use of a public-sector CMS for inventory management and government regulations that require (new) logistics providers to be full-line national carriers, making it impossible for any private competitor to focus on a smaller range of health commodities, serve only selected geographical regions, or serve a subset of health facilities. Figure 2 shows how competing full-line national logistics institutions can be formed out of the short-line subnational private wholesaling arrangements depicted in Figure 1. In Figure 2, short-line subnational private wholesalers collaborate or cooperate with each other to form a minimum of 2 or more full-line national logistics providers (Unit I and Unit II). The separate full-line national logistics units created compete to deliver all essential health commodities demanded nationwide. Short-line subnational entities that are members or affiliates of the same full-line national unit do not compete with each other. These private entities could be either pure wholesalers (who only manage inventory) or wholesalers who are also importers and procurement agents. The key requirement is that their combined capacities should allow the delivery of a full range of inventory to all health facilities in different geographical regions.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2

Transformation of the Multiplicity of Short-Line Subnational Private Wholesalers to Form Multiple Competing Full-Line National Logistics Units in a Low- and Middle-Income Country in Africa

For example, in Benin, there are 5 formal private wholesalers (4 before 2014) that are by law required to import and distribute the same pharmaceutical products: at least nine-tenths of all registered products in the country.^25,29 However, these wholesalers are only allowed to supply pharmacies, public health centers, and “other pharmaceutical establishments” but not private clinics or warehouses (depots) in rural areas, a restriction that appears to have created a sizable and thriving informal drug market. In fact, adding informal wholesalers supplying registered and unregistered antimalarials to the 5 formal private wholesalers increased the number of “wholesalers” in Benin to more than 200 without a corresponding improvement in health commodity security.²⁶ I will argue that the requirement in Benin for wholesalers to supply at least 90% of all registered medicines should be maintained, as it ensures any new entrant into the formal Beninese wholesaling market will at least be a full-line logistics institution. In addition to this measure, regulations should be relaxed to allow formal private wholesalers to serve all governmental, private, and nongovernmental health facilities in different geographical locations. After adding SoBAPS (the newly created public limited company with the Beninese government as the sole shareholder), it is possible to have 6 full-line national logistics institutions competing to ensure uninterrupted supplies of health commodities in Benin.

It is worth mentioning that in Benin, as in most Francophone Africa, there is no wholesaler markup competition. Private wholesaler markups are fixed by regulation,²⁷ which might hinder attempts to build competing full-line national private wholesalers in Benin. This is because the free setting of wholesaler markups and consumer prices, as well as the removal of any monopoly licenses or exclusive distribution rights over specific products, expands the ability that private wholesalers have to recover any continuing or repeated capital investments that are made to ensure commodity security (e.g., providing just-in-time deliveries even in rural areas). Private wholesalers will be more willing to make such investments if the associated costs can be recouped through flexible prices charged for their services. Competition driven by health care payer and provider price sensitivity should then keep prices for inventory management low. End-user prices can equally be kept low if there is also competition between manufacturers to reduce procurement costs and competition to lower retail pharmacy margins.