GEMExpert: Is better governance enough to secure access to water?

We often hear that “good institutions” would be enough to secure access to water. Your research shows it is more complicated than that: why is “decent” governance not enough?

Thomas Bolognesi: Talking about “good institutions” generally refers to the vocabulary of “good governance,” that is, the broad principles of so-called effective governance. These include transparency, accountability of decision-makers, citizen participation and cost-effectiveness. However, these principles are only effective depending on the context in which they are applied. “Good governance” corresponds to different realities from one territory to another.

Moreover, these “good governance” principles do not work in isolation; they are interdependent. Improving just one of them may bring no gain in coordination effectiveness if the others are lacking. For example, strengthening transparency and information sharing has little effect if citizens lack the means to participate, and can neither take ownership of decisions nor adapt them to their own context.

At what point does the accumulation of norms, instruments and reforms stop improving the effectiveness of governance? In other words, where are the limits of the quest for “good governance”?

T.B.: “Good governance” depends on the problem to be solved. Water security covers five main issues: access to water for industry, for agriculture and for ecosystems, flood and drought risks, and (chemical) water quality.

Faced with increasingly complex problems, humans often respond with specialization. For governance, this mainly translates into producing new norms and regulations, each targeting a specific problem but poorly coordinated, or not coordinated at all, with the others. As a result, the effectiveness of new norms and regulations tends to decline, because they end up interfering with one another through indirect effects.

This summer, for example, water regulation for agriculture showed that it also weighed on water intended for industry, households or energy. These regulations, which had clearly been designed in separate silos, generate institutional complexity, which reduces the effectiveness of coordination. One way to address this is to produce regulation that puts the existing rules in order, that is, to prioritize, plan and organize the whole, moving from a proliferation of rules to a far more coordinated whole. Simplification does not necessarily mean abolishing existing rules; it also means making them coherent with one another.

One of the study’s key findings is that water security can be measured in two very different ways, yielding two completely opposite diagnoses for the same countries. Can you explain this difference?

T.B.: In science as in politics, the ability to measure is essential to decision-making, and water security is a widely used indicator (World Bank, OECD, states). Yet the way a figure is manufactured is rarely discussed, even though how it is built inevitably shapes what we see. Here, we started from the same raw observations of water security, measured through about a hundred sub-indicators, and showed that the way they are aggregated into a single index ultimately changes the decision.

What we did was propose metrics that link the three pillars of water security ever more closely: risk management (floods, droughts), access to water (households, industry, agriculture) and the environmental dimension (the functioning quality of rivers, lakes and ecosystems). The stronger this interdependence, the further the diagnosis departs from that of a simple average, which can mask an imbalance: a territory can have excellent access to water and good risk management while having a degraded environment. A simple average would be fairly good, but taking interdependencies into account would revise the overall score downward. We thus move from a logic in which being good on a few targets was enough to a logic in which overall balance is what counts.

The first metric, known as weak sustainability, treats the main components relatively separately and gives a snapshot of the situation at a given moment. The second, known as strong sustainability, tells us about the dynamics of the system and the problems to come. By giving more weight to the environment, and on the principle that societies are embedded in their environment, it integrates today the future effects of a degraded environment on the quality of society, which makes it possible to say, in the strict sense, whether or not a system is sustainable.

Concretely, which countries or regions flip from a “reassuring” to a “critical” diagnosis depending on the method used?

T.B.: These are mostly developed countries, such as Western Europe or the southwestern United States (Colorado, for example). With a weak sustainability metric, they get a good score, because their infrastructure facilitates access to water and limits risks. With a strong sustainability metric, which increases the interdependence between pillars and the weight of the environment, we realize that the diagnosis is reversed and that their water security is not satisfactory, which changes how we read their future problems. Conversely, some countries in Sub-Saharan Africa remain in a difficult situation but appear slightly less poorly ranked under a strong sustainability metric, as their environment is less degraded. In these countries, the right to water remains a persistent problem, but perhaps they are closer to a positive tipping point than previously thought, which increases the value of international aid.

Weak sustainability metrics reward human development: where there are more political, economic and infrastructural means, risk management and access to water improve and can offset environmental degradation. Countries with lower human development often have two out of three pillars in difficulty, which pulls their overall score down. When the interdependence between pillars is strengthened, their score is penalized less, because they are often less badly placed on the environmental side.

Back to governance: you show that the quality of institutions (rule of law, citizen participation) has a positive effect on water security only beyond a very high threshold. What does that mean in practice for a country that is only “moderately” well governed?

T.B.: The rule of law and citizen participation are two aspects of the institutional environment, that is, of the general functioning of the state and of regulation, and they have spillover effects on all public policies. Here, we hypothesized that this had a positive effect, as already documented in the literature. Our study covers the entire globe (at a 50 km by 50 km resolution) and therefore presents a very wide variety of situations. We ranked all our territories according to their degree of institutional quality (from weakest to strongest) and estimated the relationship with the level of water security for each degree of institutional quality. Positive and significant effects appear only in countries where institutional quality is very high, a minority of countries worldwide (about 20%). Below that, the phenomenon is not observed, which means that a country with average institutional quality, which is nonetheless the global norm, does not benefit from it.

Why? Because a substantial critical mass of institutional quality is needed to produce positive effects. The entire institutional design must already have been shaped into something well developed in order to know how it will steer behavior. This requires substantial economic, political and institutional resources. As long as the state is not credible (corruption, unpredictable justice, uncertain sanctions), there remains a strong possibility of undesirable behavior, and behavior is hard to anticipate. You cannot be half credible: a country with average institutional quality therefore gains little from a marginal improvement; a qualitative leap is needed.

For decision-makers, this means that a water policy can fail if the fundamentals are missing. It is, in a way, the image of the champion with feet of clay. Institutional quality must be consolidated first, and its indirect benefits will be strong once the critical mass is reached, in water as well as in other sectors such as education, energy or industry. Public action gains credibility, and behavior becomes more predictable.

If the indicators used today tend to be too optimistic, what is the concrete risk for the public decision-makers who rely on them?

T.B.: This study has an indirect implication, which applies to public decision-makers but also to other organizations. We invite people to question how figures are manufactured. A figure is reassuring because it gives an appearance of solidity and rationality to a decision, but we often forget how it was produced. Here, the comparison between weak and strong sustainability shows that the former overestimates water security compared with the latter. The latter, which takes into account the interdependence of the pillars and restores weight to the environment, gives a more faithful picture of the system’s dynamics. It does not predict how things will unfold; that is not the point. What it does allow us to say is that, despite good results on two pillars, environmental weakness will sooner or later translate into a backlash against the development model. This matters because it is intrinsically more realistic. For planning and risk management, it helps avoid unanticipated future costs. In this case, overly optimistic metrics mask an unsustainable dynamic while the situation looks satisfactory today.

What do your results change, in concrete terms, about how a country, a water agency or a local authority should assess its own situation?

T.B.: Question how figures are manufactured, and change the way indicators are assembled. Simply moving from an arithmetic mean, where you add, to a geometric mean, where you multiply, makes it possible to reflect interdependencies. This second approach rewards a more balanced system.

Take the example of a high-mountain guide: we expect them to be able to climb, to be fit, and to assess risks well. A very strong climber who is indifferent to danger puts others at risk. A mountain expert who is out of shape will eventually run out of energy and make poor decisions. We are looking for a well-rounded, relatively balanced guide, and that is the idea behind this metric.

Beyond the choice of metric, it can be healthy to consider alternative measurement methods and see whether the results change drastically. This is what we call sensitivity analysis in science: changing the way we observe and checking whether the results remain stable.

If decision-makers and citizens were to retain just one idea from your study, what would it be?

T.B.: Actually, there are two main ones! First, we need to change the way we do our calculations in order to better assess how critical the water situation is, and to compare results and situations. Tipping points and interdependencies are numerous in complex systems. Thinking in a linear way is very likely to lead to a suboptimal, even harmful, decision.

Second, on the construction of governance itself: rather than staying in silos, it is often far more effective to rethink the coordination of the system as a whole, and not to assume that reducing the number of rules is necessarily a simplification. What counts is coherence.

Water security and human development: The role of size, footprint and governance

Thomas Bolognesi, Giulia Marti et Gregory Giuliani.

https://www.sciencedirect.com/science/article/pii/S0921800926001230