South African “Free Basic Water” policy: from progressive to regressive agenda?

The problems described in South Africa do not belong to them alone. In the United States we have thousand upon thousands of mutual domestic rural water solutions. These systems struggle with politics, funding, regulation, vested interests, and self-dealing that virtually prevent maintenance and the future sustenance of their communities. The South African move toward decentralization is a convenient way to pass the buck. We find it in California where State Government passes everything to the lowest unit of government which may not even be aware of state laws regarding them. Having created the system one can not easily convert to a better system because of political inertia to change. It took revolution in South Africa out of which new water statutes were born. I fear it will take enormous suffering to implement improvement. Having worked all over Africa and Latin America one of the finest organizations to emerge is the Rural Water Supply Network (RWSN) based in St. Gallen, Switzerland. It is an international organizations of all manner of water supply professional who gather together to share with others what programs work and what programs do not. Bill TUrner

Without a balanced financial budget a municipality or water utility cannot undertake the necessary maintenance and ensure the water provision. The question is how to balance a budget in a context of poverty?
While central government support is needed essentially during the investment phase, covering the annual cots should come from households payments as well as philanthropy and donations. The experience in Oman Rural Water is that philanthropy played a major role, in the past, in supporting the maintenance of water infrastructure. Several farmers have allocated the returns of their land (Land rent) exclusively to the maintenance of water infrastructure ensuring a sustainable income over the years. The international aid can be used to buy assets to generate a continuous flow of financial resources for such a purpose. During exceptional events such as COVID, which exacerbated unemployment, an extra effort from the government as well as money collection through charity events is needed. Covering the basic water needs for the poor is much cheaper than dealing with the health consequences related to poor sanitary services. Funds from water users-poor households, municipalities and central governments of developing countries will not be enough. Local, national and international solidarity are in deer need.

Before commenting, I should declare extensive involvement in SA’s Free Basic Water process, from initial policy formulation and implementation (within government) as well as a critical analyst (from an academic perspective albeit, in anthropological terms, as a participant observer). However, because of this prior involvement, I have been encouraged to contribute.

It is difficult, however, to enter a discussion without some agreement on the verifiable parameters that underpin the assumptions and analysis on which the debate and underlying paper is based.
To start, the paper suggests that the current decentralised model of local government was a choice of the post-apartheid government. That’s ahistorical. The model was prescribed by the 1996 Constitution, a compromise between the primary negotiating partners, the incoming African National Congress and the outgoing apartheid National Party. It can only be changed by Constitutional Amendment.

Then it is stated that:-

"A foundational assumption of the municipal operating model is the view that local governments can simultaneously offer affordable services and generate their own revenue through service tariffs and property taxes, to remain financially viable."

Not true. The Constitution explicitly requires the transfer to local government of “an equitable share of revenue raised nationally to enable it to provide basic services and perform the functions allocated to it”.
Failure to acknowledge the 'foundational assumption' of the purpose of the Constitutional equitable share reflects the article's focus on Cape Town (arguably the richest metropolitan community in South Africa) for which the equitable share is a small proportion of the municipality's total revenue.

But Cape Town does not provide a framework for analysis of the generic problems of service provision in the majority of municipalities where the equitable share is the largest - and, in some, almost the only - source of finance. This implicit extrapolation from a rich city to, more typical, poor small towns and rural municipalities is unhelpful and inaccurate and leads to some egregious errors.

Thus it is stated that:-

".... municipalities have increasingly prioritised cost recovery on basic service provision (electricity, water and sanitation) because these make up one of the main sources of own revenue ..."

Yet the evidence is to the contrary. Nation-wide household surveys report that over 50% of households with access to piped water ‘do not pay for water’ a proportion that grew substantially over the past decade. In this context, the much smaller numbers reported with formal access to ‘free basic water’ are almost irrelevant.

This detracts from a more important issue: Failure to manage access to water, which includes monitoring its use and requiring payment for use of excess, contributes to poor services. Supplies to households at the end of pipelines often fail because of uncontrolled usage by those upstream. This failure is systematically ignored by a particular cohort of critics, perhaps because it presents a challenge: How to achieve collective responsibility for the public service?

There are other equally obvious and significant errors. It is stated that:- “the Municipal Infrastructure Grant (MIG) has been replaced”. Yet it is well-documented that the MIG has been and continues to be the primary (and growing) source of capital investment for basic water supplies. Since there is copious, easily available budgetary information, failure to report data accurately further weakens the paper's credibility and value.

As a minimum, an author who publishes analytical work on access to and the finances of local government water services should be familiar with the formal documentation of the sector and present it (with critique where appropriate) as a starting point. Otherwise, the impression is that what is offered is simply poorly founded polemic. That does not provide the basis for serious discussion.

My original plan was to focus on the Free Basic Water supply policy itself, because based on Suraya Sheba’s essay, it has not been implemented in the spirit in which it was intended. It was the creation of Ronnie Kasrils when he was minister, and it was not universally favored by the department officials. When I lived in Pretoria (2000-2011), the policy was that everyone qualified for the FBW allotment of 6,000 liters/family/month. This was reflected on my water bill. However, the graduated rate for additional water was steep enough to more than finance providing free water to poor households. FBW was never intended to be aimed only at the poor, as Sheba describes, and never intended be enforced with water supply devices.

As Mike Muller notes, relatively wealthy cities like Cape Town and Pretoria are not the same as poor communities. In the latter case, the allocation for the central government was intended to cover the cost of providing FBW since they did not have the potential to recover from wealthier citizens. Nevertheless, the very critical comments by Muller notwithstanding, I think Sheba’s essay contains a lot of important insights. But in my view, both the essay and Muller’s comments miss the deeper cause of the problem.

The reason why most municipal governments have failed to provide equitable, sustainable water services (among other services) is the lack of accountability of government officials managing the services. Local councilors and mayors are not directly elected by the citizens of the municipality; rather, they are appointed by the party in power. The same is true at provincial and national levels: members of the provincial and national assembly are not elected directly by their constituents. South Africa has chosen a proportional representation system in which parties are the decision-makers: party cadre members are appointed to positions based primarily on party loyalty. It is a form of indirect democracy. Even the President is not chosen by people voting in a general election; he or she is appointed by the party controlling the national assembly (i.e., structurally he is both a prime minster and a president).

This contrasts with systems in which local officials are directly elected by the citizens of the city or county. Aspiring mayors or city council members must appeal directly to the citizenry for their votes. A recent article in The Economist contrasted a British city and a Swedish city: the Swedish city is rapidly creating modern “green” infrastructure and offering services that the British city can only dream about. Both are at a similar level of income; the difference is the accountability of local politicians to the citizens they serve, their clear mandate to manage multiple services, plus a far more rational model for financing these services.

I am not glorifying directly elected officials; this system has its own problems. But in the long run, there is a greater likelihood that politicians will be accountable to their citizens and strive to provide the services people want and are willing to pay for.

Global climate change is taxing municipal water agencies more than ever before, yet many are dealing with climate change crises poorly. Decentralization, a major tenet of contemporary good water governance ideas, does not perform well in many situations as the case of South Africa illustrates. Without local agency capacity, including scientific, engineering and management expertise as well as money, goals, however lofty, are not well served. Instead, national governments are avoiding taking responsibility and money does not follow responsibility down to lower level governments. This case is especially tragic as human rights and welfare were explicitly recognized in South Africa. Lack of local agencies' capacity, as well as prioritizing finances over equity should not be interpreted as dooming policies that set high equity goals.

I think a serious and critical engagement with FBW is much needed after twenty years of (mostly poor) implementation - so thank you, Suraya Scheba, for initiating this discussion.
Let me start my response by stressing that the human right to water, as a derivative right, has always been made compatible with water commodification - it is not a right to any water, but only to 'basic', 'affordable' amounts (see Bakker 2007). Thus, leaving aside for a moment the way in which municipalities have operationalized it, a fundamental problem with FBW is that it legitimizes and normalizes inequality in water access - the fact that the poor/'indigent' do not pay for water (something that the PARI report actually questions) seems to make OK the fact that they are left with only the bare minimum to survive.
Seen from the perspective of under-resourced rural municipalities, where I have conducted research, things become more complicated perhaps (just think of the lack of water redistribution due to the extremely slow pace of the water reform, but also the increase in water demand sometimes related to farm evictions). Yet, one thing I would like to recall here is that in rural areas FBW is mostly delivered by means of 'basic' water infrastructure (communal standpipes) and upgrades appear very unlikely - something that fails to increase living standards and livelihood opportunities, yet is very effective in disciplining (fiscally and materially) water consumption by the rural poor.

It’s really sad to see what has happened in South Africa. The Water Code was a remarkable feat. As far as I know, it was the first of its kind globally. It took off nicely, and the number of people served rose rapidly. However, there was a parallel trend within the water administration, a kind of reverse apartheid, which squeezed many white officers out. This, of course, had some consequences concerning the management capacity of that administration. Nonetheless, it was my impression that the water coverage was getting closer and closer to 100 %. But then, it was understood that sanitation was essential complement to ensure the desired health benefits. Thus, an additional effort concerning sanitation was deemed necessary which, of course, increased the financial burden. As I visited Pretoria in 2009, I was told that there was a third campaign. This time about shelter for all. However, the rollout of this campaign was slower, because of financial constraints. A couple of years ago, I was dismayed to learn about the problem with water supply in South Africa. Not only had it stopped expanding, but the systems had started to fall apart, due to neglected maintenance. As far as I have understood, ANC is no longer what it was. It seems to have been coopted by people with another agenda. Corruption seems to have spread like wildfires. This reminds me of what happened in the Sovjet Union as Stalin came into power. Thus, I think that the water problems are caused more by a breakdown of the democratic system than by inadequate capacity within the water governance system, even if the latter does suffer from some problems.

The author raises important points on how a measure aimed at making water supply services affordable (even free) to the poor ended up being regressive. The experiences from South Africa don’t stand on their own. A World Bank report from 2019, which reviewed subsidies for affordability across 10 countries, found that some 56% of subsidies end up with the richest 20% of the population, and only 6% of subsidies find their way to the poorest 20%. This is not necessarily due to a regressive 'agenda', as the title of the original post suggested; but a combination of poor targeting, bias towards users of networked services, lack of data and insight into who the poor are and where they live. The same report also conceded that subsides are an important tool to ensure affordable services to the poorest is complex, often requiring a combination of demand- and supply-side measures, that requires careful finetuning and targeting, so that multiple policy objectives can be met.

What resonates with me from that report is that what looks like progressive at one level of scale, becomes regressive at another level. In the remainder of my comment, I will try and illustrate that referring to other countries than South Africa.

The first place at which measures to ensure affordability are put in place is usually within the service area of a single service provider. At that level, the measure that can be taken is one of cross-subsidy, whereby the tariff is lowered for those users who cannot afford (or brought down to zero as in the case of FBW), and increased for those who can. There are two approaches to establish such cross-subsidies: 1) based on wealth group, or 2) based on consumption. There is ample experience that the latter - often formalised in rising block tariffs - ends up being regressive. The World Bank report uses that as a classical example, whereby lower-income households tend to be bigger and end up paying more. The former has more potential to be progressive, and is applied for example in Colombia. There all households are classified on basis of their wealth status, and utilities charge a higher tariff rate to the households in highest wealth classes, which cross subsidize the lower wealth classes. This can work generally in larger cities, such as Cape Town, where there are enough better-off users who can cross-subsidize with low income households, and where typically there are stronger municipalities and utilities who can administer such a system.

This approach may not work in smaller towns or rural areas, where there are insufficient users who have a higher income as compared to the low income households who otherwise cannot afford it. In such cases, national government can transfer public finance to the utility or to the municipality. This is then a wealth distribution from richer parts of the country to the poorer, and could be considered a progressive manner. From what I understand from the comments by Mike Muller and Doug Merrey this is what the equitable share aims to do. However, there is a regressive element to it. It means that in effect the State is subsidizing water use of those low-income households who have access, whereas those funds could have also been used for infrastructure development to create access for those who are completely unserved.

That is why the WB calls for demand-side subsidies (i.e. geared towards lowering the costs to the users) and supply-side (i.e. supporting the service providers to extend to the unserved). That is why the financial flows in South Africa combine the MIG (for new infrastructure) and the equitable share (to support the consumption). One would have to analyse the extent to which these two financial flows are sufficiently well targeted to 1) the unserved, and the poorest among them, and 2) the lowest income households among the served. That this is not easy is again shown in the case of Colombia. It has a very similar affordability framework (though for the lowest incomes it is not free). The auditor general in a detailed review concluded that over time public financial flows had become more progressive, particular ensuring a more equitable distribution between richer and poorer municipalities, but not yet reaching better equity within territorial units. So, after much fine-tuning, the financial flows were not yet fully progressive. It doesn’t mean it is an instrument, or agenda, of regressiveness. It is just very difficult to target subsidies and financial flows across and within territorial units, like municipalities.

A final level is to look at how subsidies for public services fit into the broader framework of social protection measures, such as social grants and subsidies. Sevel countries argue that it is not very efficient nor effective to have separate subsidies for public services like water (or energy) and have these administered by local governments or service providers. They argue that it is more efficient to increase unemployment grants, social grants and pensions so that low-income households have a slightly higher income, but then let all households pay similar tariff rates. Several countries in Europe for example explicitly don't provide affordability measures that are specific to water supply and sanitation, but rather include that in their overall social protection framework. The advantage is that the existing systems for social protection measures can be used, and that service providers don't have to access data on low-income households and set up separate systems for administering what are in effect relatively small amounts, compared to the overall size of social grants.

I particularly like this document by UNECE (United Nations Economic Commission for Europe), which reviews the range of measures applied across the countries of Europe, including both the high income ones, but also some of the lower- and middle-income ones. These illustrate the progressiveness of various measures, but also some of the risks of regressiveness. I hope that these examples, as well as some of the ones from the World Bank study help informing the further development of FBW. Because it is clear that strong affordability measures are needed, within cities, between cities and smaller towns and rural areas, and as part of the overall social protection system.

I agree with Mike Muller, the picture is very different with smaller cities/towns or rural areas. When it comes to access to water for the poor, the most pressing issue seems to me to rethink the way water services authorities get appointed in South Africa. To grant competence for water service delivery to local governments when we know that some municipalities were created ex nihilo in the Post-Apartheid context, was a bit too optimistic, in my opinion, with respect to the capacity of some of these municipalities to be fit for the task. I suspect some of them might never be…