Prime Minister Modi’s new electrification scheme, Saubhagya (good fortune), aims to electrify all Indian households by end of 2018. Although the focus on households instead of villages is commendable, the scheme is mostly old wine in new bottles. The government should adopt an integrated strategy that puts more emphasis on distributed energy solutions, such as solar micro-grids, and commit to improving the quality of electricity service.

The new scheme aims to electrify households by offering economically disadvantaged households a free or inexpensive electricity connection. The poorest households can obtain a connection for free, and many others only have to pay 500 rupees in installments over time. While below-the-poverty-line households can already obtain connections for free, the new scheme significantly expands the number of households that benefit from connection subsidies.

The new scheme will run into a few major challenges. The most fundamental problem is the problem of the quality of electricity service. In collaboration with the Council on Energy, Environment and Water (CEEW), we conducted a large-scale energy access survey in rural India three years ago. We found that in rural Bihar and Uttar Pradesh, households typically had fewer than ten hours of supply per day. Our research shows that this lack of adequate supply can explain people’s widespread dissatisfaction with their electricity connections.

Electrifying more households does not improve the quality of service. In fact, Modi’s new scheme may actually prove counterproductive. If electricity distribution companies must serve poor rural households that consume little power at subsidized rates, the financial problems of the power sector in the states may grow worse. The government itself estimates that even if the new households only used electricity for eight hours per day, the additional generation requirement would be 28,000 megawatts. If this power is sold at heavily subsidized rates, either distribution companies face heavy losses or state governments’ subsidy expenditure further increases.

A related problem is the cost of power consumption. Reducing the cost of household connections helps, but households may hesitate to connect to the grid if they worry about electricity bills. The new scheme does not deal with the problem of paying one’s electricity bills. If households worry about the quality of service, they may be unwilling to connect and pay their bills in a timely manner.

Finally, there is the problem of distribution infrastructure. Many households cannot connect to the grid because they are too far from the nearest power pole, and the new scheme promises to install power poles to enable household connections. The total cost of electrifying households can be extremely high if the government must install a power pole to serve just a handful of households. Such investments would not be a good use of scarce resources.

To solve India’s rural electricity problem, the government should focus on crafting a robust strategy for improving rural energy access in the medium term. Modi’s Bharatiya Janata Party (BJP) is currently so dominant in the Indian political landscape that it can afford to resist the pull of flashy publicity stunts. Now is the time to look to the future and get strategic about bringing reliable power to every Indian home.

The government can improve the state of rural electricity access in a few different ways. For poor households, distributed energy solutions offer a promising alternative to grid connectivity. Solar home systems, micro-grids, and mini-grids can serve rural households’ basic electricity needs at a relatively low capital cost. The government should develop an integrated strategy that uses both grid extension and distributed alternatives to meet the basic needs of the underserved.

Even more important is a decisive approach to solve the problem of quality. In our ongoing research, we find that the low quality of electricity service is a major reason why rural households are unwilling to pay for their power. Efforts to improve the quality of rural electricity service would enable rural households to pay electricity tariffs that help distribution companies reduce their losses. The government has already signed a series of Power for All agreements with the states, but these agreements do not go far enough to link higher electricity prices to better quality of service.

What India needs is a virtuous of cycle of improved electricity service and better financial performance. If electricity distribution companies can improve their service, then they can charge higher prices without a popular backlash. Increased revenue, in turn, allows even better service over time. If Prime Minister Modi wants to achieve his dream of bringing light to every Indian home, his cabinet should firmly commit to improving the quality of rural electricity service in collaboration with state governments.



Last week, the United Nations’ Sustainable Energy for All (SEforAll) published a new report called Energizing Finance to compare actual investment in energy access with the estimated USD 45 billion a year that would achieve universal electrification by 2030.

One of the main messages of the report was that the annual investment in distributed power generation, such as solar home systems, was only USD 200 million per year – 1 percent of total investment in electrification. As Rachel Kyte, CEO of SEforAll, put it, “we urgently need targeted, refined strategies to increase investment in integrated electricity solutions.”

Why has financing for distributed power generation been such a disappointment? One explanation is that investors simply do not understand how scalable or economically viable distributed power generation is. The CEO of Practical Action, Paul Smith Lomas, said as much: “We have long argued that financial systems are skewed; the amount of finance directed towards decentralized energy technologies, like mini-grids and stand-alone systems, remains tiny in comparison to investments in national grids. This is despite decentralized energy technologies being the most economical solution to meet the needs of the majority of unconnected people by 2030.”

My own view is different; even though distributed power generation holds promise, it is a fundamentally challenging business that still faces major obstacles. There are no easy answers to the problem of financing distributed power generation. Until we find effective solutions to the core challenges of distributed power generation as a business, financing will remain a headache.

Both refereed research and practitioner experience show that affordability remains a fundamental obstacle to distributed power generation. Whether decentralized systems are more affordable than grid extension is beside the point as long as prospective customers are unable to afford the systems even considering government and other subsidies. One simple explanation for the lack of financing to distributed power generation is that few companies in this sector deliver large returns or have scaled beyond a certain extent.

Another problem is the small size of the projects. When investors do their due diligence and negotiate contracts, small projects present a challenge because their transaction costs per project are very high. Most distributed power generation do not invest billions of dollars at the time, but instead implement small projects in difficult conditions. For a financier, small projects mean lots of work for at most limited returns and a much higher perceived per capita credit risk.

While the SEforAll report is correct to point out weaknesses in existing financial models and policies, the challenge we face is much more fundamental. Is there a case for large-scale financing for distributed power generation? Does the distributed energy industry have the technology and business models to support scalable financing?

Investors do not care about nominal targets such as universal electricity access by 2030. They care about financial returns or, in the case of the public sector, social benefits. Therefore, expanding the financing of distributed power generation is ultimately a question of developing business models that generate concrete gains to investors, users, and the society more broadly.

The fundamental challenge for SEforAll and others active in this space is to demonstrate that distributed power generation can generate either large profits or social. Until someone makes this case, energizing finance shall remain an elusive goal.

India’s turn toward renewable energy has been widely hailed, but the future promises challenges that put the country’s climate leadership to test.

When Donald Trump announced that the U.S. was to withdraw from the Paris climate agreement, India’s Prime Minister, Narendra Modi, said that his country would go “above and beyond” the treaty. Modi’s statement was widely hailed as a signal of India’s emerging climate leadership.

There is something to Modi’s claims. India has set an aggressive target of achieving 175 gigawatts of renewable power capacity by 2022, and the construction of coal-fired power plants has all but collapsed because of record-low solar power costs. India has also invested in energy efficiency and pioneered the International Solar Alliance.

These achievements are truly notable for a country that suffers from widespread poverty. Anywhere between two to three hundred million Indians still have no electricity at home. For those who have access to power, outages remain common. A 2014 rural survey found that grid-connected households in Uttar Pradesh and Bihar – states with a total population of three hundred million people – had fewer than ten hours of power on  a typical day.

But while India’s actions deserve applause, Modi’s claim to climate leadership remains far from certain. The past few years have treated the prospective climate leader gently, and the future could bring many new challenges.

One reason for India’s ability to halt the construction of coal-fired power plants is that disappointing industrial growth has led to lower than expected power demand. For the past three decades, the Central Electricity Authority of India has consistently over-projected the growth of power demand. Between 2012-2017, the authorities predicted peak demand to grow 8.5% a year, but the real growth was only 4.9% per annum.

If India’s drive to industrialize grew more robust, the demand for new power generation capacity could grow much faster. Policymakers might hesitate to shut down coal-fired power plants, and India would remain dependent on coal for a long time. Modi’s government has also formulated an ambitious plan to provide every home in India with 24 hours of power on a daily basis. Any progress toward meeting this goal increases power demand and puts pressure on the government to build more power plants.

Published in June 2017, India’s Draft National Energy Policy predicts that even under ambitious energy conservation and renewable energy policy, total coal-fired power generation capacity grows to 330 gigawatts from two hundred today. Robust economic growth would contribute to a rapid growth in power demand, and coal would remain an important fuel for India’s engine. Under these conditions, India would be a major obstacle to limiting global warming under the Paris Agreement.

India is also still far from the point at which natural fluctuation – intermittency – in solar and wind power generation begins to create systemic problems. As India’s renewable generation capacity grows, grid stability becomes an increasingly important challenge. Although a recent technical study found that India’s renewable energy targets need not compromise grid stability, achieving this goal is a major challenge for India’s strained regulatory system and electric utilities. The state-owned power sector suffers from serious financial, technical, and institutional problems. Indian electric utilities will face major challenges in managing the growth of renewables, and the state governments would pay a political price for increased outages.

India’s turn toward renewable energy has earned its applause, but the fact remains that Prime Minister Modi’s climate leadership has not yet been tested. Given India’s pivotal role in global climate policy, addressing the country’s energy problems is a challenge of global importance. Indian policymakers should prepare for them well in advance, and the international community should offer their full financial and technical support to help India continue its climate leadership.

Today, India is the world’s most important country for energy and climate policy. The country’s population is huge and growing rapidly, but the vast majority of Indians still live in poverty. In global climate policy and energy markets, India’s economic growth and pattern of industrialization are critical variables with global implications.

On June 27, India’s NITI-Aayog (National Institution for Transforming India) published a Draft National Energy Policy for public comments. It is a remarkable document that sheds lots of light on India’s energy challenges for the coming decades.

The first important component of the draft is the projection of energy demand supply up to 2040. Although coal has recently suffered setbacks and solar power is thriving in India, NITI-Aayog predicts that under business as usual, coal-fired power generation capacity could reach 441 gigawatts by 2040 – more than doubling from today’s less than 200 gigawatts. Even under an ambitious scenario with energy efficiency measures, coal-fired power generation capacity would reach 330 gigawatts. Renewables would grow fast, but coal would remain a hugely important component of India’s power sector.

NITI-Aayog’s  predictions are based on scenario modeling, and it’s unclear to which extent their models capture recent decreases in the cost of renewable power. Their models also assume that India’s annual GDP growth will remain above 8% until 2040 — an optimistic scenario of robust, uninterrupted growth for a long time. It’s a heroic escape from the poverty trap, not unlike China’s.

The other remarkable feature of the NITI-Aayog policy is a frank recognition of the political sources of India’s struggling power sector:

“The fundamental cause of repeated episodes of discoms [electricity distribution companies] going bankrupt is the absence of commercial pressure on them. Political pressures to provide electricity either free or highly subsidized prices to certain segments of the society have meant that revenues of discoms often fall well short of payments due to generation companies … The ultimate solution to this problem lies in ensuring that electricity distribution is subject to commercial pressure.”

This statement is spot on, but it remains unclear whether state governments of India – who have constitutional authority over electricity distribution – are ready for reforms that would subject electricity distribution to “commercial pressure.” Such reforms would require a firm commitment against bailing out struggling distribution companies, allowing costs to reflect the real cost of power generation, and creating truly autonomous and independent regulatory commissions.

Coal and power sector reform are but two aspects of India’s complex energy challenge, but they illustrate a bigger point: India is under tremendous pressures from all directions.

The country still faces major challenges in expanding its power generation capacity to meet the needs of a growing economy and  wealthier households, but the draft national energy policy relies on coal so heavily that a global backlash is all but guaranteed.

At the same time, India’s dysfunctional power sector undermines economic growth and puts heavy fiscal pressures on state governments with recurring bailouts. To this day, India’s overall progress in rationalizing electricity pricing is very disappointing.

NITI-Aayog’s draft energy policy is a commendable document: transparent, comprehensive, and in some respects even bold. Ultimately, though, India needs a pragmatic strategy for building a coalition in favor of a rational energy policy that brings together economic, environmental, and equity considerations.

On June 26, India’s Prime Minister Narendra Modi met U.S. President Donald Trump for bilateral talks. The much-awaited official statement, Prosperity through Partnership, covered a wide range of issues – including energy and climate.

The bilateral statement is a pretty stark contrast to Prime Minister Modi’s earlier claims for India’s climate leadership and going “above and beyond” the Paris Agreement. Here’s the relevant paragraph, with commentary by yours truly:

“President Trump affirmed that the United States continues to remove barriers to energy development and investment in the United States and to U.S. energy exports so that more natural gas, clean coal, and renewable resources and technologies are available to fuel India’s economic growth and inclusive development.”

  • Here we see Trump’s influence on the statement: include “clean coal” in a statement when the cost of renewable power capacity is collapsing in India.

“Prime Minister Modi and President Trump looked forward to conclusion of contractual agreements between Westinghouse Electric Company and the Nuclear Power Corporation of India for six nuclear reactors in India and also related project financing.”

  • President Trump has advocated nuclear power, though without any real policy effort, but India has also expressed interest in expanding its nuclear power generation capacity. This part of the statement is likely of interest to both sides.

“Both leaders welcomed upcoming visits between India and the United States that will expand energy and innovation linkages across the energy sector and deepen cooperation, including on more efficient fossil fuel technologies, smart grids, and energy storage. They supported financing of energy projects, including clean coal projects, by Multilateral Development Banks to promote universal access to affordable and reliable energy.”

  • Innovation has been a cornerstone of U.S.-India energy cooperation for years, and it’s encouraging that it remains on the radar. Here, again, we see Trump’s influence in emphasis on “efficient fossil fuel technologies” and “clean coal projects.” The call for Multilateral Development Banks to consider clean coal is a little preposterous, though.

Overall, Trump appears to have had a lot of success in steering the discussion of energy cooperation. The text is pretty underwhelming compared to what Modi said about the Paris Agreement. That’s not surprising, given that the statement covers many issues that are of critical importance to India, ranging from terrorism to China.

To wrap it up, the bilateral statement is not a bad way to understand India’s budding climate leadership. India sees genuine value in promoting clean energy for energy security, electrification, and air quality. But India’s commitment to climate issues is not strong enough to stop compromises on “clean coal” and other non-starters that Trump alone values. Modi’s ready to tone down his climate rhetoric whenever he expects gains from doing so.


Last week I had the pleasure of attending the 2017 Environmental Politics and Governance Conference at Indiana University. I had not participated in the previous years, so I was quite curious to see where the study of environmental politics is going in political science.

In general, I was quite impressed with the conference. Many of the papers had good ideas on important topics and were sufficiently rigorous in execution that a meaningful discussion was possible. Such papers would have been impossible to imagine in the subfield just five years ago.

Although American politics and quantitative methods dominated, the conference did feature a wide range of studies from air pollution in China to international regimes in Antarctica. I was glad to see many graduate students presenting ambitious and interesting research.

But I also saw some pretty major challenges for the ongoing effort to mainstream environmental politics in political science.

A key issue with the conference agenda was the uneasy balance of substantive/methodological diversity and the ambition to break through to the leading journals of political science: APSR, AJPS, JOP.

Environmental issues are by their very nature interdisciplinary, and there are many good reasons to call for greater diversity in substance and methods. Alas, editors and referees of political science journals could not care less about any of this.

Breaking into our discipline’s top journals requires a relentless focus on contributing to political science and maximizing empirical rigor. I have banged my head against that wall with varying degrees of success. Mostly, I have failed.

At the conference, I saw a certain degree of wishful thinking. Scholars of environmental politics still seem to have a pretty poor understanding of modern standards in political science, and I did not see enough serious focus on perfecting papers for the brutal referee process at our leading journals.

If the authors listen to some of the more positive commentary and send their papers out too early, there will be many broken hearts and lots of frustration in the future. My comments were probably the harshest in the group, but that’s only because I wanted to make it very clear to the authors how much work remains to be done for a publication in a top journal.

I am slowly realizing that the problem with environmental politics in political science is pretty fundamental. The subject matter calls for interdisciplinary research, which is less rigorous and more confusing than disciplinary research by necessity — bringing disciplines is not much easier than convincing the GOP to embrace climate policy. But the professional standards of our discipline call for prestigious publications, and those publications require extreme obsession with disciplinary framing and research design.


The World Bank’s 2017 State of Electricity Access Report offers a number of important data updates on global progress toward universal electricity access. Among the findings, though, one is clearly the big story of the quest for universal energy access: in Sub-Saharan Africa, the off-grid population is growing.

That’s correct — the number of people without electricity in Sub-Saharan Africa is growing. Although the electrification rate of the region has grown from 26.5 to 37.5 percent between 2000-2014, this growth has not been enough to cancel out the effects of rapid population growth.  While the Bank’s estimate of the total number of people without electricity was below 500 million in 2000, it was 609 million in 2014.

No other region of the world faces a similar challenge. In 2000, South Asia had more people without electricity than Sub-Saharan Africa, but the WB estimates that this number has decreased from over 600 million to 343 million in 2014. East Asia and Latin American, on the other hand, are close to universal electrification already.

Africa’s difficulties with rural electrification do not have a single cause. Low income levels constrain investments and low population densities make grid extension very expensive. The governance of the power sector remains a major problem in many countries, and in the past civil conflict prevented improvements in electricity generation, transmission, and distribution.

There are, however, many reasons for optimism now. Of all the regions in the world, Sub-Saharan Africa has the most to gain from off-grid alternatives to grid extension. Because of low population densities across many countries, the opportunity for off-grid electrification is a real boon for the region, and many companies are now rapidly scaling up their off-grid operations.

South Asia still has a huge non-electrified or under-electrified population, but the direction is right. For anyone interested in sustainable energy access, Sub-Saharan Africa is thus the frontier of the future. The world’s ability to achieve sustainable energy for all is, first and foremost, determined in Sub-Saharan Africa.

feature_africa_energy_2016_main-760x378Credit: SolarAid Photos (Flickr / Creative Commons)


This one’s a little off topic, but it might be of interest to graduate students, post-docs, and junior faculty navigating academia.

The end of the academic year 2015-2016 was the low point of my career. By then I was certain that I was going stay at Columbia as Associate Professor of Political Science, with tenure.

Both my department and the Graduate School of Arts & Sciences had supported my case, and the statistics were clear: The Provost’s advisory committee pretty much never denies anyone who has made it that far. Confused about the long wait, my colleagues told me we should celebrate before the summer since it’s clearly a done deal.

But there it was: I got the bad news at the end of May. That itself was a source of concern, given that it was quite late for schools interested in my services to start preparing for a senior faculty search.

The practical problem at hand, of course, was the least of my concerns on that dark day. What went wrong? Would anyone actually want to hire me now?

I recovered from the worst shock pretty quickly, partly because tenure denial is a pretty minor setback compared to what I’ve had to weather in the recent past. But equally important was the flood of messages expressing compassion and, in many cases, asking if I was interested in a position at the sender’s institution. I knew that many of these questions would not turn into actual interviews, but it was still reassuring to see that so many people did not agree with the Provost’s decision.

In August and September, I actually sent out more applications than eight years ago, fresh out of grad school. The market in 2008 was awful, but this year there were jobs everywhere for someone with my profile. I had also expanded the range of my interests quite a bit.

I applied to all kinds of jobs, from quantitative international relations to vanilla comparative politics, and most importantly — energy and environment policy. I applied to political science departments and policy schools alike. I applied to North American, European, and Asian positions.

The annual meeting of the American Political Science Association was a seemingly endless series of meetings with possible employers, and everyone else I met wanted to know why Columbia had denied me. Some of the far out conspiracy theories from the rumor mill were pretty entertaining, too.

The fall semester was grueling. I gave job talks — some very formal, some very informal — around the world. I had already scheduled a large number of regular talks and run-of-the-mill workshops before the Provost gave me the boot, so finding time for everything was a royal mess. On the upside, my frequent flyer status went up one level.

Visiting Singapore for a job talk was fun, but I don’t recommend spending only 36 hours there and then heading back to teach two classes straight out of the airplane. My teaching assistant for mathematical methods deserves a medal for filling in for me so many times. As for my students, they deserve an apology for my sleep-derived rants.

My first breakthrough came early. Johns Hopkins SAIS had reached out to me pretty much within days after my tenure denial to see if I would be interested in a Full Professor’s position in Energy, Resources and Environment. After both I and them had done our background research, and very much liked what we saw, I gave a very early talk and met with everyone there.

It was a great deal of fun, and less than two weeks later the Dean called to share good news: the SAIS Academic Board had unanimously recommended an offer. I then negotiated as hard as I possibly could, knowing that I had no competitors because there was no formal search. The deal was very sweet, so my wife and I had every reason to celebrate.

I gave a number of other talks as well, but at some point fatigue kicked in and I started rejecting invitations. Unless I thought there was a chance that an institution might match the SAIS offer — all things, not just money, considered — I did not interview.

I learned that the big difference between the senior and junior markets in political science is the level of organization. Junior searches follow a reasonable schedule and the ads contain relevant information, but this is not true of senior searches.

The whole process takes a long time, and is very confusing. Several places (no names) advertised a Full Professor job, and then the search committee told me that the best they can do is Associate without tenure. Almost the same thing, yeah — and a waste of everyone’s time.

By January, it was very likely I would take the SAIS offer. However, SAIS had to complete a formal tenure review. While their leadership did an admirable job reassuring me that the decision would be positive, I ended up irrationally stressed out about the tenure review. Publication machine or not, I’m also a human being with a brain evolved to survive in the jungle, not to deal with academic bureaucracy.

Waiting for the tenure review was more stressful than waiting for the offers, especially after I began to imagine myself at SAIS, finally doing the energy/environment policy work that is my real passion.

So when I got the good news that the tenure review was over and the decision positive, I was ecstatic. The day before I signed the contract I also learned that I actually had an endowed chair waiting for me at SAIS. For a lifetime environmentalist, being the Prince Sultan bin Abdulaziz Professor is good fun for sure.

At the end of the day, I cannot say the Columbia tenure denial bothers me. I had a great time at Columbia and will miss my colleagues and students, but I’m also very excited about the opportunity at SAIS. I’ve wanted to work in a great policy school for a long time, and that’s were I’m heading now.

Dealing with the denial is also a lot easier because the decision was made at the administrative level. It would have been much more difficult if my senior colleagues, all of whom I respect and admire, had decided that I’m just not good enough to be there.

Note: this is a guest post by Nikhil Jaisinghani, a co-founder of Mera Gao Power, on our Science Advances impact evaluation. Read the study here:

I want to thank Dr. Urpelainen for all allowing me to post my thoughts on his blog. I have followed his team’s study for a number of years as the survey was planned, executed, and analyzed. The data is credible and well analyzed in the study, but I am concerned about how many readers may interpret (and a few publications already have interpreted) the results.

The study wasn’t a comparison between the 24 hour unlimited electricity that the grid promises and focused electricity services such as what MGP provides. Yet many of the newspaper and magazine headlines imply that the study did exactly that. In order to provide some level of comparison, I’d like to take a crack at highlighting what social benefits have and have not been found to be impacted by electrification. For those interested in reading more, I suggest three papers which review a breadth of data and reports on the subject:

  1. The Impact of Small Scale Electricity Systems”, The World Resources Institute
  2. The Evidence of Benefits for Poor People of Increased Renewable Energy Capacity: Literature Review”, Institute of Developed Studies
  3. Welfare Impact of Rural Electrification”, The World Bank’s Internal Evaluation Group

But in case you don’t have time to read a hundred pages of reports, let me summarize some of the themes. First, the income benefits of electrification are disproportionately enjoyed by richer households, the offered explanation being that poorer households such as the ones MGP serves cannot afford the larger appliances that use the larger power loads and instead simply use electric lighting. A second commonality across studies is that when rural communities are electrified, the poorer communities almost exclusively benefit through improved lighting. Third, electrification has not been found to spur new enterprise creation in the short term. New businesses take many years to spring up in any significant number and are more reliant on roads and customers than they are on power. Of course there are always outlier studies. I’m sure if I searched, I would find one study that links ice cream consumption with cognitive ability. That’s why macro studies are important; they pull the findings from a large number of studies to identify what findings are consistent (I’m pretty sure there aren’t multiple studies linking ice cream to cognitive ability, as much as I wish it were true).

Education, measured in the number of hours children spend studying and how many years children spend in school, was in fact the most commonly found benefit of electrification. That benefit is the outcome of electric lighting that enables students to study longer and better each night. It is important to note, however, that studying is a choice and the child and the parents must choose studying over socializing and playing. Previous studies have not found children who choose not to study with kerosene to be more likely to study with electric lighting, whether it be MGP, the grid, a solar lantern, or other. The common finding has been that those children who do choose to study – with or without electric lighting – study longer with electric lighting. Though Dr. Urpelainen’s study did not measure that variable, we did our own randomized control trial in 2012 and made our best attempt at measuring it. We aren’t survey experts or econometricians, but we did find a dramatic increase in the number of hours those children spent studying (as an aside, that RCT found consistent results as Dr. Urpelainen on enterprise creation and income).

It was because of this global evidence, coupled with my own experience living in a poor, remote off-grid village, that we created MGP with the intention of offering the electricity services that poor communities benefit from at a price that would allow them access. But actually doing that – delivering great service to customers – is actually pretty complicated (for those of you who think our jobs are easy, I invite you to come help us out). The data from this survey implies MGP was not delivering on its promise to provide 7 hours of service per night. When Dr. Aklin described the service provided as “paltry”, it initially caused me to bristle. But when I looked at the data, I realized he was right; customers responded that they were only getting 1 to 2 hours of service per night. In 2014 when this survey was conducted, we were a young company still solving many problems. Our batteries were stored in wooden cabinets in a customer’s house– we were testing an assumption that communities would protect the assets in order to maximize service. What we found was that these cabinets would often be broken into by people (not always customers) connecting additional devices directly to our battery banks. Customers also found ways of charging batteries from our system (there was a marked increase in battery ownership over the survey period, possibly because customers could now charge them in their homes). These uses were not what our micro grids were designed for, of course. Further, our construction teams would build micro grids quickly, and we had no quality control processes to ensure the micro grids were built properly. The 1 to 2 hours customers reported receiving may have been the consequence of us being at the beginning of a sharp learning curve. Since then, we have taken significant steps to better secure our equipment, control power theft, and improve our construction quality. With these improvements in place, we tested the performance of our micro grids in 2015 and measured our service delivery to customers through the foggy winter season. We found that with these improvements, we are now able to provide the promised service to customers. And the value our customers are receiving is therefore significantly greater now than it was a few years back.

There are many fair conclusions here, including the importance of quality service delivery, that electrification is not a silver bullet to end poverty, and that in addition to electrification other investments are required to help those customers in the bottom few quintiles to prosper. But one conclusion that is not justified is that the focused electricity services that we provide to customers are significantly less valuable or impactful than grid electricity.

Our customers value our service, and the lighting we aim to provide our customers is life changing; not only has the importance of lighting to poor households already been established, but we see if first hand. There are hundreds of millions of people in India still reliant on kerosene for household lighting; over a hundred million children studying by kerosene. These people cannot affordably be served by the grid. Dr. Urpelainen’s earlier publications have shown how few people are connected to the grid even when they live in grid connected villages. These people shouldn’t be left out, especially since we have an affordable option for them. Don’t dismiss us just yet; hang in there with us, be open to learning with us, and don’t be too quick to rush to conclusions. It is a long journey, and we are working hard and against great odds to make a difference. We are optimists by nature and are optimistic about what we can achieve. But we can’t do it in a world that doesn’t believe that this work is important.

Solar Power provided by Mera Gao in UP, IndiaMera Gao Power lighting in rural India.

Globally, over a billion people lack household electricity to this date, and most non-electrified households are found in Sub-Saharan Africa and South Asia. The high capital cost of extending a national or regional electric grid to remote rural communities is an important barrier to progress in household electrification.

No wonder, then, that the collapse of the cost of solar panels has created enthusiasm for off-grid electrification. Combined with ultra-efficient energy technologies, such as LED lights, less expensive solar panels have brought back the idea that countries could ‘leapfrog’ by supporting communities’ local off-grid electrification schemes instead of extending the electric grid.

Many scholars and practitioners are enthusiastic about this approach. Researchers from the University of California, Berkeley write in Nature Climate Change that “the present day is a unique moment in the history of electrification where decentralized energy networks are rapidly spreading, based on super-efficient end-use appliances and low-cost photovoltaics.” The environmental group Sierra Club states that “off-grid renewable energy technologies … are better suited to meeting the challenges of energy access than centralized coal projects and grid extension. That means we can solve energy poverty and climate change at the same time.”

An important question for the enthusiasts is how much energy is enough for off-grid electrification to make a real difference. LED lights are ultra-efficient but freezers, welding machines, and many other appliances still require substantial amounts of power. If off-grid electrification can deliver results for minimal loads of power, such as domestic lighting, the case for replacing or at least delaying grid extension is much stronger than if large loads of power are necessary for beneficial impacts.

In a recent Science Advances paper (open access), we (Aklin, Bayer, Harish, and myself) collaborated with an Indian solar microgrid company, Mera Gao Power, to assess the impact of basic energy access from a very small microgrid. In a randomized controlled trial, previously non-electrified households in Barabanki district of the state of Uttar Pradesh in India were offered, for 100 Indian rupees (about 1.5 U.S. dollars) a month, two LED lights and a mobile charger powered by a small solar panel and a battery.


Solar panels for a micro-grid that offers two bright LED lights and mobile charging to villagers in Barabanki district of the state of Uttar Pradesh, India.

We found that access to a minimal level of solar power did reduce kerosene expenditures, as households replaced their kerosene wicks with the solar-powered LED lights. The reductions were pretty substantial, too: where MGP installed solar microgrids, monthly kerosene expenditures decreased by almost fifty rupees (about 45%) relative to the control group.

This estimate is one-half of the MGP monthly fee and does not even consider that many households did not subscribe – so that many MGP customers with high previous kerosene expenditures likely reduced their expenditures even more. These reductions were achieved without government subsidies and at a low capital cost.

Now, the bad news: we found no evidence of broader economic benefits on outcomes such as savings, spending, business creation, and time spent working. The estimates were either zero or, in some cases, modestly positive but subject to large uncertainties.

While the basic energy access afforded by the MGP microgrids offered a substitute for kerosene, we did not find evidence of substantial effects on rural development. While we did not conduct a systematic study of educational (e.g., more study hours) or health benefits (e.g., less indoor air pollution), the evidence for benefits in other areas was weak.

These results do not mean off-grid solar power is useless, but they do raise two important questions. First, can larger systems that generate more energy produce better results? If households could run appliances and machines, they would be able to use solar power for a wide range of economic activities, making electricity access more appealing. However, larger systems are also more expensive and households – many of whom are poor – would have to pay higher fees to participate.

Second, could off-grid solar power furnish more benefits with complementary policies? The lack of electricity is far from the only constraint on economic growth in these communities. Access to finance, roads, water infrastructure, supply of skilled enough labor, and markets for products are all inadequate in the area that we studied. Solar power might have produced better development outcomes if some of these other problems had also been solved – be it by the government or a private company.

An important lesson from the study is how much India’s heavily subsidized kerosene hurts the growth of off-grid lighting solutions. We found that people reduced their kerosene expenditure by no longer buying kerosene from the private market, but by continuing to purchase their heavily subsidized quota from the government’s public distribution shops.

If India were to succeed in replacing traditional kerosene subsidies with cash transfers, or perhaps even vouchers for off-grid lighting solutions, the market for off-grid lighting would grow much faster. This change would not only bring environmental and health benefits, but also likely boost technology and business model innovation in off-grid electrification. Just replacing kerosene subsidies with more flexible social policy might make basic energy access through solar microgrids much more beneficial to the society.