29 April 2009

Water for Sale -- The Review

Fredrik Segerfeldt originally wrote Water for Sale (2005) in Swedish. The CATO Institute has published an English-language version. So, what does a Swede have to say about the private companies selling water to the world's poor? Privatization is a good idea. [Also see this post for Tyler Cowen's opinion that water can be sold in a laissez-faire market. I disagree with him.]

Anyone who works on water policy should read this book. Segerfeldt argues sensibly, with examples, statistics and documentation. His arguments echo and reflect recurring themes (equity, poverty, development, capitalism and corruption) in water policy, and this book will be useful (barring miracles in the water sector) for the next 15-20 years. The book is short (118pp plus 25 pp of notes and references) but complete. The prose is clear, with few redundancies, inconsistencies or errors. The book's biggest strength is Segerfeldt's logical foundation (defend the poor). Its biggest weakness is the omission of the bigger picture (how does one do anything well when the government is corrupt?), but we all fail that test at some point.

Segerfeldt's brutally-effective condemnation of water policy in the developing world is as dispassionate and rational as one can be when discussing how “every minute of every day, 22 people die because they cannot get enough safe water” (p. 8). Children suffer the most -- accounting for one-quarter of the 12 million annual deaths from water shortage.

His main point -- and a good one -- is that the private provision of water by companies seeking profits can hardly make the poor worse off, since “ninety-seven percent of all water distribution in poor countries is managed by public suppliers, who are responsible for more than a billion people being without water” (p. 1).

Can we blame Nature for water shortages? No -- the wettest place on earth (Cherrapunji, India) has water shortages due to poor water policies. Although shortages are strongly correlated with a lack of economic development, development is happening too slowly to save lives. Instead, Segerfeldt suggests improving the governance of water supply, and he recommends importing good governance, via private enterprise.

Private companies don't just have the advantage of outside cash. They have the advantage of management experience and specialization, the advantage of competitive pressures, and the advantage of the profit motive. Public bureaucracies have none of these advantages. Every locality learns by doing (often repeating mistakes "learned" elsewhere), facing no competition, and catering to political whim. The worst problem is that public bureaucracies suffer no penalty when they deliver poor results. Bureaucratic rewards accrue to those who spend their budgets -- not to those who serve more people. Political forces favor rich urbanites and big farmers, not poor slum dwellers or small farmers.

Further, Segerfeldt notes that water is too cheap. Because prices are low, demand exceeds supply and shortages result. Why are prices low? Subsidies of $45 billion per year mean that water prices cover -- on average -- 30 percent of the cost of water service. What's the result? Deferred maintenance (leaky pipes, poor quality) and small service areas. While the rich get piped water, the poor pay 10 to 80 times more to get water from “pirate” vendors. (Pirate in the unregulated sense; they are providing a valuable service to grateful customers.)

What are the barriers to change? “Anti-privatization activists... are driven by an ideologically inspired aversion to enterprise, coupled with fear on the part of vested interests of losing their privileges.” He points out the cost of such ideology -- a failure to serve the supposed beneficiaries of their interest, i.e., “it would be not just a pity but quite outrageous if millions of people were to starve, fall ill, and die through water shortages brought about by the strident propaganda of vested interests and powerfully ideological movements with quite different ends in view” (pp. 4-5).

The rich and middle classes also dislike privatization because they are likely to face higher water bills. First, because private companies want to collect their money, Second, because private water companies are likely to expand service to truly poor customers who are going to use less water (thus making is hard for wealthier customers to claim that they cannot do with less). Third, because government policies designed for the “poor” are more likely to end up serving the poor.

Public sector unions want to protect their jobs, of course, but the worse perpetrators of the status quo are the politicians who use water utilities for selfish gain -- hiring relatives and cronies, diverting cash flow, and contracting with “friendly” firms. (The mayor of Cochabamba, Bolvia would not allow the city's water supply to be privatized until a dam was included in the deal. Conveniently, his friends were in charge of building that dam. The infamous failure of the Cochabamba privatization can be partially blamed on that dam.) Even more common than politicians-cum-thieves are politicians who fail to monitor public water managers. Where, after all, would fines for bad performance go, except from one pocket to the other?

What are his solutions? Private trade in water rights. After Chile “introduced private ownership of water in the 1980s... water supply has grown faster than in any other country. Thirty years ago, only 27 percent of Chileans in rural areas and 63 percent of urban communities had steady access to safe water. Today's figures are 94 and 99 percent, respectively -- the highest for all the world's medium-income countries” (p. 31). Even better, the incentives to sell conserved water increased agricultural efficiency and -- through competition -- lowered the price of water. Farmers did not suffer -- despite the lack of major infrastructure investments, the shift to higher-value crops and greater efficiency resulted in six percent annual productivity growth between 1975 and 1990. Oh, and don't forget those bureaucrats. The lack of “capricious pricing” and quotas on water use left farmers alone to do what they do best -- raise food and make money.

At the retail/urban level, Segerfeldt recommends that contracts for operation be awarded to companies through a competitive tender. He then gives several examples of privatizations that failed and succeeded, noting what went wrong and why. Disturbingly (for anti-privatization activists), it seems that many failures resulted from political failures and corruption -- not greedy capitalists. His main recommendations to avoid failure are that contracts reflect local conditions, annual price increases be capped, and alternative providers be allowed to continue operations (competition!). These common-sense ideas would deflect most concerns about privatization. (Remember that breach of contract can be remedied by re-municipalization!)

Further, Segerfeldt recommends that equity issues (“the poor will suffer with market prices”) be directly addressed through income supports or vouchers. “Cheap water” policies not only subsidize the middle class and rich, but sometimes they only subsidize those classes -- like when the poor don't even have piped water!

Bottom Line: “Keeping water distribution in the public sphere is often identified as more democratic... Using food as an analogy, we can observe that food is also essential to life. Yet in countries where food has been produced “democratically” -- that is, by the government -- there has often been neither sufficient food nor democracy. In this regard water is no different... the question that naturally comes to mind is why anti-privatization activists do not expend as much energy on accusing governments of violating the rights of the 1.1 billion people who do not have access to water as they do on trying to stop its commercialization.” (pp. 113-114).


Mel said...

Not sure I understand this paragraph:

"The rich and middle classes also dislike privatization.... because private water companies are likely to expand service to truly poor customers who are going to use less water (thus making is hard for wealthier customers to claim that they cannot do with less). Third, because government policies designed for the “poor” are more likely to end up serving the poor."

First, is it safe to assume that private companies will expand service to poor (particularly poor rural users who are located quite some distance from the main service area)? I thought Sheila Olmstead did a piece concluding that private companies were likely not to expand service under such conditions.

Secondly, is the last sentence correct? Was it meant to say that governmental policies benefit the middle class/wealthy more than the poor?

KD said...

Sounds like an interesting book and a topic near and dear to me. Some comments...

Private sector participation (PSP) was largely promoted by the World Bank and others in response to a failure of governments to provide basic services to their citizens. The Catch 22 here though, is that effective provision of these services by a private company still requires a significant degree of government responsibility and oversight - the lacking of which provided the impetus for PSP. As I see it, this is the largest hurdle to effective PSP in the water sector (particularly in developing countries).

Is there any mention of the negative consequences of Chile's unfettered regulation of water rights and trade? See NY Times article:

In response to Mel - in my experience the rich and the middle class dislike privatization simply b/c they will likely have to pay more for small/no improvement in service. Many of these people have already found ways to work around the deplorable services provided by govn't (or they live in areas the govn't treats more kindly than others), so they will not be the ones who see significant improvement in services through PSP. However, they will likely be charged more to subsidize other system improvments, potentially including the expansion of services to the poor.

Private companies will expand services to the poor (1) if it is in their interest to do so (additional connections means more money/ more paying customers versus those stealing water) OR (2)if government requires them to expand service to the poor (in which case they will likely subsidize these services through increased rates to everyone else). I'm not sure exactly which Olmstead piece you are referring to (one related to the US-Mexico border?).

Lastly, in response to the comment regarding addressing the impact of pricing policies on the poor through income supports or vouchers - I agree that this should be in the case in developed countries (as opposed to "some for free, pay for more"), but it is just not logistically feasible in developing countries. This is where "some for free, pay for more" policies are needed, such as S. Africa's "lifeline".

DW said...

So Fredrik Segerfeldt believes that faulty water delivery systems should be privitized, but that after privitization takes place, "alternative providers be allowed to continue operations". Not sure what that means. To me, privitization means that a private entity would take over management of a local water delivery system, and perhaps the local governments former water department staff who didn't go to work for the managers would keep working as contract manager for the local government, overseeing the efforts of the private water system manager.

But according to your review, Segerfeldt seems to be saying that when a private company wins a bid to take over a local municipality's water deliver system, the orginal publically managed systems should somehow continue to operate in competition with the now privately managed system. Does this mean there would be two water deliver systems built to serve the same service area?

That kind of duplication and overlap of services seems unaffordable to me, and I don't think that water customers would want to pay for two seperately operated water delivery systems just so they could say there's competition.

In San Diego, the city is trying to look at all city services, including water delivery and distribution, then force city employees to compete with private service providers to see if "managed competition" would bring down the cost of service.

Lots of politics involved with the public employee unions.

Not sure if they've found any private firms who would be interested in taking over the city's multi-billion dollar system at this point.

One question this raises is whether or not corrupt politicians and their appointees would ever be capable of overseeing the operations of a new private operator and still keep their jobs. I suspect that if they ever do try to privitize water delivery and distribution what gets into the contract for services, and how that contract is actually enforced (or not) will be key.

David Zetland said...

@Mel -- KD answers some of your questions, but -- to clarify -- yes, gov'ts in *developing* countries do favor the rich [perhaps developed too!]. Was Olmstead referring to the US or developing countries?

@KD -- Thanks for the great comment! The Chile article was after the book, but there's probably some truth to it. Nobody ever said capitalism was easy! For a form of developing country voucher, check out the bolsa familia in Brazil or the opportunidades in Mexico.

@DW -- he's talking about water delivery trucks, not side-by-side pipes. Yes, monitoring is critical. I think too little happens now -- for public workers.

joan said...

This might all be true for countries with a corrupt regime, but a good exception is for example the Netherlands (my home country): privatization was banned, people pay a cost price, water consumption is relatively very low, the distribution network is highly efficient and water quality is one of the best in the world: no chlorine is necessary. "The question should not be whether the public or the private sector performs better hypothetically, but how the actual performance regarding public goods and services can be improved."

Eric said...

Just a thought. If you want my comments, and likely the comments of many other readers, I need more than a day turn around time. As exciting as Aguanomics is, my landlord likes it when I pay the rent. My employees like getting paid on time.

FYI, my normal turn around time is 2 weeks not a few hours.

David Zetland said...

@Joan -- yes, I admire water management in NL. (I'll be in AMS from 23 - 29 Jun, so please introduce me to water folks...)

@Eric -- sorry that I am putting stuff out quickly. Time waits for no man -- and do what you can :)

Fredrik Segerfeldt said...

David: Thanks for the good review.


1. The rich and politically influential middle classes are unlikely to support price hikes in order to pay for network expansion to cut the costs for the poor. (Insider-outsider theory)

2. The problem is most acute in shanty-towns in urban and peri-urban areas. For obvious reasons that is where the health problems are most severe. (However, in Gabon, privatization has been successful also in rural areas.) The residents of such areas are economically, politically and socially remote, but not geographically so. Empirically, I don't know of any privatization case where network expansion to poor neightborhoods has not taken place.

3. You reverse the burden of proof in an unjust way. "Is it safe to say..." It is safe to say that governments have failed more than a billion poor people. In this debate, the private alternative is often compared to the perfect public alternative, with little relation to reality.

4. Yes, government water policies have benefitted the rich and hurt the poor. There is plenty of data showing how poor people and neigborhoods are overrepresented when it comes to lacking piped water. One of the reasons for this is government subsidies to the rich.


I agree. But what is better: a dysfunctional bureaucracy in a country with a dysfunctional government excluding the voice of the poor or a multinational company with a profit motive to serve the poor as customers?

You ignore the fact that the majority of the poor are not connected to any water mains at all, and pay around ten times the price of piped water for their water. Therefore, they would not be hurt by price hikes. I see vouchers as an exception, not as a generalized scheme.


Just like in all rich countries. The OECD average percentage of people with access to an improved water source is 99 (and as you know that includes countries like Mexico and Turkey). The problem is dysfunctional governments and lack of development, which in most cases are closely related.