Is Mission 300 about creating jobs or not??

5 big questions for the World Bank to help Ajay Banga succeed

Imagine the US government launched an AI initiative to produce more engineers to close the nation’s competition gap with China and then decided to track progress by counting 2nd grade reading skills. Such a metric is utterly divorced from the stated goal. Yet whenever anyone complained about it, they’d be dismissed because “everyone should be able to read” and “reading must come before an engineering degree.” True and true. But the steps are too far apart. And using a 2nd grade literacy metric is not quite the formula for a winning initiative to counter China.

Alas, Mission 300 risks a similar fate. It’s an incredibly ambitious effort. The World Bank and African Development Bank aim to connect 300 million people to electricity by 2030. On one level, I love it. Every person on earth must have electricity. 300 million is a lot of people. And 2030 is, like, tomorrow. I want M300 to be a smashing success.

But there’s a problem. As Devex reports;

Mission 300… was repeatedly cited by [World Bank president Ajay] Banga as an example of a project with tremendous job potential. He also waved off criticism that M300 should focus on electrifying businesses over homes. Banga called that “bunk.” The goal of M300 is not about “one light and one fan in the roof, but productive capacity electricity, which allows them to be capable of generating something economically,” Banga said.

That all sounds great. M300 should absolutely be about productive capacity for jobs. Jobs are what development is all about. And creating jobs is what Banga is betting the World Bank’s future on. But the World Bank just released its primary progress indicator: new connections.

Here’s the World Bank’s new M300 Progress Dashboard. It shows 32 million connected so far (since July 2023) across 84 Bank-financed projects and reports on track to hit 189 million.

Pretty good. And new connections is a perfectly fine metric if you care about giving extremely poor people their first lights.

But it’s completely divorced from jobs. It’s the energy equivalent of tracking 2nd grade reading as your advanced engineering goal.

A cynic could wave this off as just a disconnect between the Bank leadership’s talking points and the technical teams reporting the data. It could be also reflective of massive confusion in the energy sector that so often conflates first connections for the poor with large scale energy for growth.

Yet this disconnect between connections and jobs is more than a miscommunication problem because Banga is fighting for the Bank’s existential survival in a high-risk political environment. He has to show relevance and that the organization can deliver. I don’t think it’s overstated to say that his legacy – and perhaps the future of the World Bank itself – could depend on what happens with M300.

That’s why the progress metric has real consequences. And why I have 5 big questions.

Quick primer on how “energy access” has multiple meanings

Before I get to my skepticism, a little background on definitions. (If you know all about this already, jump to the next header.)

A “connection” could be a physical wire linking a building to the national grid or throwing a solar panel on a roof or even just (according to some) owning a solar lantern.

These are all different and provide different services. IMO, a better approach than the physical hardware is a minimum level of actual electricity consumption. How much depends who you ask.

• Historically, 100 kWh per person. The standard definition from the International Energy Agency for many years has been 100 kWh per person per year in cities and 50 kWh in rural areas.

• The IEA now has 3 levels: The IEA’s methodology (latest version; see page 99) now uses roughly 10, 100, and 250 kWh per person per year for what they call basic, essential, and extended bundles. IEA explains:

“Access to electricity entails a household having initial access to sufficient electricity to power a basic bundle of energy services – at the minimum, several lightbulbs, phone charging, a radio and potentially a fan or television which is broadly equivalent to a range of around 50-75 kWh per household per year… the “essential bundle” includes four light bulbs for four hours per day, a fan for three hours per day, and a television for two hours per day, which equates to roughly 500 kWh per household per year…. The “extended bundle” implies a refrigerator, four hours for lighting, four hours for TV, and six hours for a fan, which equates to roughly 1,250 kWh per household per year.” [Note: per capita figures assume ~5 people per household.]

• The World Bank says let’s use 5 Tiers. About a decade ago, a team at the World Bank proposed a “multi tier framework” with 5 levels covering capacity, duration, affordability, reliability, and other characteristics. Each tier has an implied minimum annual per person consumption (using the minimum daily household capacity times 365 divided by 5 people) of 0.9, 15, 73, 250, and 600 kWh.

• The Modern Energy Minimum is 1,000 kWh. Colleagues and I have proposed as a new metric a 250 kWh per capita residential threshold (equivalent to the IEA’s extended bundle or the Bank’s Tier 4), plus a non-residential average of 750 kWh covering industry, commerce, and agriculture. We think this is a good metric for tracking a second step after initial access because it’s both higher and wider.

• For reference, Americans use 12,700 kWh per person each year, of which 38% is residential so an average of about 4,800 kWh at home. So, more than 5000x the World Bank’s Tier 1 access. (You can guess why I’m making this comparison….)

For ease, here are all the access levels in one little table.

My questions for the M300 progress tracker

Given that initial access is a very important goal in itself (yes, everyone should have electricity at home, even if it starts tiny), but the World Bank keeps insisting that M300 is really about jobs, jobs, jobs. Here are my questions:

1. Is the World Bank counting Tier 1 as a connection?

Delivering a solar lantern to a poor household is not the same as a reliable grid connection. Both can be useful. But they are not the same. Some parts of the World Bank want to count any initial electricity as a connection for M300, while others want some slightly higher level of minimum service. The specific debate has been between Tier 1 versus Tier 3 (or 0.9 vs 73 kWh).

This is not trivial. Picking Tier 1 will make it much easier to reach that huge number of 300 million people. Opting for Tier 3 will mean far fewer connections but greater impact for each one.

The progress tracker does not specify, but I understand that it is counting all tiers as a new connection. In other words, Tier 1 is the minimum standard for M300.

I fully understand the impulse to set the bar low. No one wants to fail on a high-profile flagship effort with the target number literally in the name of the initiative. Picking the lowest possible threshold makes it much easier to declare success in five years.

Yet Tier 1 is not even 1 kWh per day. It’s 0.02% of what a typical American uses at home. If they are really using this standard, the Bank should be transparent and fully own its decision by explaining clearly what it means: lighting for a few hours a day and not much more. And definitely not jobs.

Setting the bar so low may have headline benefits but comes with political risk too. Most people (and politicians) have a common sense understanding of what a “connection” usually entails. So when an org sets the bar too low, they invite blowback. Just ask Power Africa when they faced a similar backlash after counting lanterns as connections.

2. Why not track residential electricity use at all five tiers?

If the World Bank already has its own system for identifying five tiers of service, why not just track them all and be transparent that some projects provide very basic connections while others are helping move people up the energy ladder? This would mean counting all new and improved connections. That’s slightly different but entirely in the spirit of the M300 – and closer to the impact on employment.

And here’s a gift: tracking all 5 tiers can be done for free. My team has already built a platform that can estimate electricity usage in every building on the continent, so we have the data. Here’s a sample of what it looks like for Senegal and Zambia in 2020:

3. Will the Bank report grid versus offgrid?

Energy has unfortunately gotten sucked into the culture wars. Until recently, energy access has, for many constituencies, meant delivering offgrid solar to the poor. This makes sense for some services for some people. But the push for universal electrification from a purely development perspective should be agnostic to both delivery system and fuel.

Most people in the rich world have no clue whatsoever where their electrons come from. They just know it works when they push a button.

The same should apply to poor people in poor countries. If you care about just getting basic service to very poor people, then offgrid solar is going to win on cost and speed. If you care about raising livelihoods and creating decent jobs, then you need sufficient reliable electricity at scale. But then you need grids and utilities, while accepting that diesel and/or large-scale gas may be the best option in some places for some uses.

Of course the World Bank board does not agree. Some (mainly European) shareholders want to use energy policy for climate mitigation goals and thus want absolutely no fossil fuels, with a big preference for offgrid. The US and a few others want the Bank to focus more on growth outcomes and go back to being ‘all of the above’ on fuels. The World Bank management is trying to split the difference so has created convoluted rules on gas. M300 has so far largely avoided the fuels debate but has said it will be half on-grid and half offgrid. For transparency, the tracker should at least show this split.

4. Where is M300’s progress indicator that matters for firms & jobs?

The debate over whether to prioritize families or firms is healthy IMO. The traditional access community targets households. Moussa Blimpo and Ken Opalo have weighed in decisively on the side of firms. The usual answer I get is “why not do both’? Yes, sure, I agree. Let’s do both. But if we are doing both then we need to measure both.

If M300 is really about jobs, then why is there no progress indicator tied to how electricity affects employment or firm productivity? New household connections are not a proxy. Even “new business connections” are really not right either since the problem facing business is not the initial connection but the cost and reliability of power.

I keep hearing that the Bank is working on a metric but no practical ideas have surfaced. I see (h/t Katie Auth) that the AfDB has a new electricity project in Niger with the dual goals of boosting (a) the access rate and (b) the manufacturing sector’s contribution to GDP. At least that‘s headed in the right direction. This kind of approach can work in individual projects and country plans.

M300 could apply a simpler metric across all countries that ties directly to firm productivity and jobs: reliability. The excellent study by the World Bank’s Justice Mensah shows outages are a job killer. So, M300 could track reliability in each country and set a broad improvement goal, like a 50% reduction of outages in urban areas. That would be a game-changer.

5. Why, why, why, oh why, is the World Bank M300 tracker only counting its own projects?

I’ll say it: Donor attribution is stupid.

If you care about development outcomes, it should not matter who gets credit or bean-counting the source of each dollar. Maybe it helps if you are trying to justify spending to a skeptical parliament. Or if you’re a skeptical NGO and your real goal is to hold the World Bank’s feet to the fire.

But from a development perspective, attribution is not just a futile exercise, but it creates all kinds of negative ripple effects by stripping agency away from the governments who should be accountable to their citizens, not an international organization.

The whole idea is an accounting fiction anyway. All “World Bank projects” involve host governments, contractors, co-financiers, and especially the cooperation of local actors. “World Bank money” comes from repayments from its borrowing members plus a subsidy from other members to subsidize lower interest rates. Many energy projects also involve “World Bank guarantees” which are also a complete fiction since (dirty little secret!) those guarantees are ultimately backed by the host governments. If an energy project goes bad and a guarantee is called, it’s just added to the country’s national debt. (Ask Zambia.) The World Bank is not spending its own money or taking any risk itself in any meaningful sense. It’s just acting as a loan agent and debt collector for others. So attribution for connections is misguided in both practice and theory.

Worse, the World Bank taking public credit for specific connections undercuts the entire idea of M300’s country ownership. A well-known tenet of development is that countries must lead the way with outsiders providing follow-on technical and financial backing. That’s why M300 asked countries to write national compacts. (Each of these are more tailored to each country, but they follow a set template and their topline goals are all identical: universal household access plus a target for clean cooking, renewables, and finance raised.) So if M300 is about the World Bank, AfDB, and all its partners working with host governments to deliver connections, why do we need to attribute each new connection to one financier? We don’t.

If Senegal or Tanzania successfully electrifies all their citizens, isn’t it in everyone’s best interest that the Senegalese and Tanzania governments fully own that outcome?

A better way: just track outcomes and collectively share credit.

Adding it all up provides an easy agenda to help M300 achieve its bigger goals

Taking each of my 5 questions leads to a simple 5-steps to help fix the big disconnect:

1. Clarify the methodology of what counts.

2. Track new and improved connections (using our platform or something better if they have it).

3. Report on-grid versus off-grid.

4. Add an indicator (any indicator!) that actually matters to firms, such as reliability or share of industrial energy.

5. Forget the folly of donor attribution and just report on what actually matters: final outcomes.

The World Bank is wonderful & maddening

Ajay Banga’s instinct to make M300 about job creation is exactly right. His organization is filled with some of the smartest and most dedicated people. Its mission to end global poverty is amazing. The world is clearly better with the World Bank than without it.

But the World Bank is also maddening. It can be insular, hubristic, and resistant to change. It’s often blatantly focused more on its own survival than on achieving its mission. It’s too often more worried about hitting its own corporate indicators than delivering development.

M300 is a chance for the organization to show it can accelerate electrification in Africa and prove its continued importance to creating jobs. But I think it can only succeed if it’s far more transparent — and gets the progress metrics right.