Defend Truth


Participation Rules: Our crisis, and our hope, runs deeper than December

Luke Jordan is the CEO of Grassroot, a community organizing tech start-up he founded in 2015. He worked at the World Bank in India from 2011-14, and at McKinsey in China from 2005-10. He writes in his personal capacity, and the views expressed here do not reflect those of Grassroot or any other organization with which he may be affiliated.

All eyes are on the ANC’s elective conference. We are obsessed with the idea that a decade ago, in another December, bad leaders taking power started us on the road to crisis. Our only road to a better life, we think, is a change of leaders in a month’s time or 2019. But our problems are deeper than that, and we have more means to tackle them.

Our policy system is broken, even when run by “the good guys”, and our people have no sense of ownership in our country. On the ground and on the law books, we have the resources to build a participative, learning state. If we don’t start to use them – now – then neither next month nor 2019 will matter very much.

Details, broken everywhere

Take policy and learning first. From among many examples, I’ll begin with scientific research. This is of enormous importance to our long-term future. It is barely touched by the Zuma faction, run for years by the technocratic part of the ANC, regularly graded an ‘A’ or ‘A+’ by the press.

It also has a funding system that is broken, not because of implementation, but because of a mistake in the detail of policy. Most of our researchers are funded by grants from the National Research Foundation (NRF), which provides such grants on the basis of papers published. But the formula the NRF uses does not count papers accepted for international conferences, no matter how prestigious. Worse, it divides any award by the number of co-authors on the paper, including the number of foreign co-authors.

It is hard to overstate how much damage this does in advanced science. The rules create a massive disincentive for local researchers to collaborate, nationally and internationally. Add a collaborator to a paper, and never mind if the idea gets better, you will lose funding. Publish at leading conferences, and you won’t get a grant, no matter how important to your field that conference is. The funding rules incentivise our best researchers to cut themselves off from scientific collaboration, just as science at the frontier is built on collaboration.

Some of the world’s best researchers in artificial intelligence are South African. They will tell you that we have the ingredients to be leaders in the field. But that will come to nothing if we do not engage in the global networks that define the field, and our researchers will hesitate to do so as long as they are penalised for it. When I asked a scientist from DeepMind what the principal barrier was to us competing in the field, he pointed immediately to the NRF’s grant rules.

Those rules predate our current leadership. The members of our elite have barely advocated for their change, despite endless references to AI and automation (or, in a bad Davos slogan, the “fourth industrial revolution”). The rules just sit there, quietly wreaking havoc on our future.

Examples like this can be multiplied with unfortunate ease. Among the most important is procurement, or how government buys things. Since the early 2000s, that has been governed by a set of rules formulated by Trevor Manuel’s Treasury and reissued by Pravin Gordhan’s. The rules state that in any government procurement the technical scores for bids will be disregarded in the final stage of evaluation. Bids do have to cross a technical threshold, but after that, technical quality does not matter. The lowest price, or combination of price and BEE score (with price having an 80% or 90% weight), wins.

This is a bad rule. It means that honest officials have no way to stop low-balled, low-quality bids from winning except by setting a very high threshold – and then risking no bids meeting that threshold, and having to start again. Dishonest officials can rig a process by setting a low threshold, making sure their friends low-ball on price, and then rely on renegotiation later.

The fix is to make the final score a weighted blend of price, BEE and technical scores. Internal and external observers have pointed this out for more than a decade, yet nothing has been done. The rule just sits there.

Or take housing and credit. The National Credit Act of 2005 enforces little distinction in debt work-outs between loans secured against assets or for consumption. That is one more factor skewing lending towards furniture and away from houses. Again, it requires a change in a detailed clause to fix this, but no such change is even debated. We have a programme for first-time home buyers, FLISP, which gives a cash subsidy to people who already have a mortgage – and hence merely inflates house prices without expanding access. It should be converted to a credit loss backstop, absorbing the first portion of losses if a loan goes bad. Not only has it not been fixed, mere discussion of fixing it is non-existent.

An upside-down policy system

This is not intended to be one more screed about government incompetence. Public policy involves enormous systems of extreme complexity. Mistakes are inevitable. Sometimes they are a result of corruption, and sometimes what one party considers a “mistake” another considers a necessity. But a large proportion of mistakes are the result of human beings not being perfect.

What matters is the ability to spot and fix such mistakes quickly. The ability of a policy system to do so can be called the ability of a state to learn. China’s ability to learn rapidly – in contrast, for example, to a slower-learning India – has accounted for much of its rapid growth. Its policies tend not to be better than others’ at the start, but it rapidly finds and fixes flaws.

That is where we are unusually and deeply flawed. Our mistakes just sit there. No internal process within government catches them. In part, that is because we have relegated almost all review processes to the Auditor General. Auditors are important, but as process specialists they are not qualified to diagnose or engage on detailed substantive errors.

Other causes are cultural. We seem to have a deep fear of admitting to mistakes. Our policy writing is shallow, even in elite publications, and we always personalise. A problem is not a problem until it involves the actors in our national drama. I almost hope a Gupta nephew applies to the NRF, because that seems to be the only way anyone will talk about it.

The institutional structure of our government is top-heavy. Policy can change on the whim of top policy-makers. The fetishisation of leadership means almost all of them believe they know better than their officials. That top-heavy design was purposeful. It flowed in part from legitimate fears that a hold-over bureaucracy would block transformation post-1994; in part from managerialist ideas common in the 1990s and 2000s, and in part from egos and a desire for control. It now means details are almost impossible to fix unless a senior policy-maker agrees, and, whether because of demands on time or size of ego, they rarely deal in details.

This upside down structure, in which broad goals change fast and important details change slowly, is not a feature of the ANC alone. This year, the City of Jo’burg dissolved City Power’s corporate structure and turned it back into a city department because Mayor Herman Mashaba had tantrums with its board. That was precisely the opposite of privatisation. If the DA stands for anything, it is privatisation. But the Mayor wanted more power, so he got it.

Fixing these institutional and cultural flaws will be hard. It is true that our current leaders stand in the way. But removing them will not, on its own, fix much. Doing so requires creative institution-building. It requires surrendering the feeling of power. None of our current cast of would-be leaders seems capable of that. None seems to have realised it is even a problem.

We do have some resources available. The Constitution’s requirement for the “progressive realisation” of socio-economic rights implicitly demands state learning. Our governance structures are replete with commissions, councils and delivery bodies that could be repurposed for state learning – if we exert the pressure necessary to turn them from ignored talk shops into places of constructive and deliberative change. Of course that will be difficult. But there are no shortcuts to a state that progressively realises a better life.


So we need a state that learns. But that will not be enough. If that is the change we need from above, from the apex institutions of the state down, the change we need from below is to close our democratic deficit. That requires meaningful participation in governance.

Participation is sometimes confused with something weaker, “consultation”. But they are worlds apart. Participation is something you do; consultation is something done to you. When your parents decide to move to a new house, they may consult you. As an adult if your partner decides to move, you expect to participate. We are not children in relation to our government. We expect to participate, not be consulted – and in many cases, not even that is done.

The idea of participation is not new. It is already in our law. The Municipal Systems Act defines how local government runs. It devotes a chapter to participation, stating that cities must “complement formal representative government with a system of participatory governance”. It goes on to require participation in not only planning, but budget setting and, most important, performance management.

Performance management means the setting of annual targets by a department and then measurement against those targets to determine bonuses and salary increases. It is rare to find someone who wakes up worrying about what the plan for their department says. It is common to find people who wake up worrying whether they will hit their targets. It is the heart of power, and our legislation says we can and must participate in it.

What happens in reality is quite different. Every year cities “consult” on their targets. The targets are printed near the back of the Integrated Development Plan, which are drafted in a six-month internal city process and then handed out at consultation meetings. There is one meeting per ward, when each ward has 20,000 people or more. That meeting has two hours to cover a document of a few hundred pages and deliberate on the needs of 20,000 people.

If this sounds like a farce, it is, and deliberately. It allows officials to set whatever targets they want, while claiming consultation. It results in things like a target for “informal settlement upgrading” with no definition of “upgrading”. Or a target for housing that turned into “housing opportunities”. Or kilometres of road resurfaced, not kilometres of road that don’t need resurfacing because they were done right the first time. The same farce repeats itself when these targets come to be measured. The auditor asks for receipts for the roads being surfaced or the “upgrade” and no one asks the people if it was done and how it was done.

Changing this in practice will be difficult. Participation is messy. There are many meetings, and they are hard to manage. It requires practice, and skill. But when it is done right, in the long term it changes the limits of the possible.

In Europe and Latin America, participative governance is spreading quickly. The more surprising cases come from East Asia. In Japan, the early 1960s were a period of student riots and brawls in parliament. But a series of remarkable mayors built participation from the ground up. In Yokohama, they called it “the citizen designs the future of the city”. Whole departments were built to make participation happen. Deep democratic practice ensured top-level project continuity over decades, even as 80% of the plans were revised in implementation.

Or take China. Yes, China. It is undoubtedly an authoritarian state. But the city of Chongqing, with a population of 30-million, started a little experiment a few years ago. Every year it did a survey of 1%-2% of its citizens, several hundred thousand people, with a sample skewed to the poor. It asked just one simple question: “Did your life get better last year?” Forty percent of the bonus and salary increase for every official was tied to the answer in their ward.

So we have a revolution in the practice of our government, already on our law books. It can close our democratic deficit and reinforce our state and society. None of our leaders – past, present or future – so much as talk about it, except for the odd banality about listening to the people. What is at stake is self-image. Real participation would require them to give up the self-gratification of being someone who knows best. But not just them. It would also require all our legion of commentators and self-appointed champions of the poor to do the same. Which perhaps accounts for the silence.

Participation and learning

To conclude, it may be asked: why treat these together. This is a long article. But I recently met someone who was there when participative democracy was written into our legislation. I asked him what happened. “Well,” he said, “a couple of cities tried, but it was just hard, and it didn’t feel like it was working as we hoped, so people gave up quickly”. A participative state that does not learn is likely to lapse back into empty consultation, or worse.

The reverse holds too. A learning state may be able to function without participation. Technical people sitting in rooms can self-correct, if there is a strong culture of doing so. But we do not have that culture. So we need participation to correct our tendency to self-flattery, a tendency that all of us share. Learning requires information, and information on the condition of our society lies both in data and in our people. Most of all, state learning is not learning for its own sake – it must have a goal. That goal must come from the people.

A participative, learning state, whose goals are set by the people and that learns to deliver those goals – this may sound beyond reach. But much is already done. Participation is in the law. Our institutions are littered with committees and reviews that, with a tweak here and a charge there, could be the instruments of learning.

The Constitutional Court has gone even further. It has repeatedly emphasised the importance of “progressive realisation” and “meaningful engagement” in its rulings on socio-economic rights. The court may have put participation and learning in the very foundations of the state created by our Constitution – if that is how we choose to leverage those rulings.

Our current leaders stand in the way. But not only them. Nothing in our media today will get us where we need to go. It’s time to turn away from those who’ve had nothing but dead ideas for decades. Perhaps we should turn off talk radio, uninstall Facebook, and attend a public meeting, read a policy document or make a submission to a hearing.

Finally, we need less leadership, not more. We need officials who will yield the floor, and who will admit to mistakes. Who do not yearn for people to just listen to them, to just let them lead. We need people who are not afraid to be both more and much, much less than what we think of as leaders. We need that to start with ourselves, whatever happens in December. DM

Ideas for this piece have derived in part from a monograph written at the World Bank but also from the challenges and privileges of managing Grassroot


Please peer review 3 community comments before your comment can be posted