This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.


Off-Grid Solar Power for Rural India: What Say the Customers?

Johannes Urpelainen's picture
Johns Hopkins School of Advanced International Studies

Johannes Urpelainen is the Prince Sultan bin Abdulaziz Professor of Energy, Resources and Environment at Johns Hopkins SAIS and the Founding Director of ISEP. He received his Ph.D. in Political...

  • Member since 2018
  • 45 items added with 30,488 views
  • Jan 7, 2015

solar rural india

In the past five years, off-grid solar power as a strategy of rural electrification has become a popular idea — and for a good reason. More than one billion people remain without electricity today, many of them in remote rural communities far from the national electric grid. At the same time, the rapid decrease in the price of solar panels has created new opportunities for commercial off-grid electrification.

While off-grid solar power is far from replacing grid extension as the main mode of rural electrification, it can no longer be dismissed as a trivial strategy. In Bangladesh, for example, more than three million solar home systems have been installed for residential use. Companies such as OMC Power in India and Off-Grid:Electric in Tanzania have secured major commercial investments into their businesses.

But what are these companies delivering? As a first step to answering this question, I and my collaborators (Aklin, Bayer, Harish) conducted a survey of the rural customers of Mera Gao Power (MGP) in the Barabanki district of the state of Uttar Pradesh, India.

MGP is a company that provides villagers with two bright LED lights and a mobile charger in exchange for a monthly payment of 100 rupees. In a randomized controlled trial funded by the International Growth Centre and SPEED India, we evaluate the demand for MGP service and its socio-economic effects. The customer survey is an initial effort to understand what MGP is achieving through its work.

The survey report is available on my website. The most important observations about the consumer experience can be summarized as follows:

• By far the most common reason for subscription was the improved quality of lighting. In contrast, mobile charging and fuel expenditures were much less important.

• The most common use of MGP lights was outdoor lighting. In a typical household, one of the two lights was placed outside at night.

• Overall, households were satisfied with MGP service. The main complaint was the behavior of MGP staff, suggesting that further training could improve customer experience

• Willingness to pay for additional services, such as a fan or a television, was low.

These findings are interesting for several reasons. First, they show that the frequently cited benefit of reduced kerosene expenditure is not a major reason for adopting solar power. Instead, rural customers want improved lighting. This is important because much of the discussion on the benefits of solar power has focused on fuel savings. Our experience with the MGP service suggests, instead, that the key advantage of solar power is simply better lighting.

Second, MGP seems to have adopted the optimal service package. There is little demand for additional services even among customers. While this result may initially appear surprising, it is important to remember that MGP customers are among the most deprived households in rural India. Wealthier rural households, who would be able to pay for a fan, tend to live in already electrified habitations.

A challenge for expanding MGP’s impact is demand. In villages approached by MGP, only one-fifth of households adopted solar technology. Reducing the price further either through government subsidy or less expensive technology would probably help, as cost and the lack of disposable income were common explanations for not subscribing to the service among the survey respondents.

Overall, however, the results from this survey are encouraging. They show that, even in the poorest villages, many households are willing to pay for good lighting. Companies such as MGP are providing a valuable lighting solution to replace kerosene in communities where grid electricity is not available.

We’ll be conducting plenty of additional analysis in the coming months — stay tuned for updates!

Photo Credit: Energy Access and Rural India/shutterstock

Johannes Urpelainen's picture
Thank Johannes for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member
Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Bas Gresnigt's picture
Bas Gresnigt on Jan 7, 2015

Johannes, thanks.

…even in the poorest villages, many households are willing to pay for good lighting…
…costs … were common explanations for not subscribing..
These fit with my own obervations when I cycled in the mountains of Kashmir.
Costs was the answer when I asked why not a larger solar panel than the one they had (size ~17″ monitor).

Bas Gresnigt's picture
Bas Gresnigt on Jan 7, 2015

A ‘large’ solar panel in Pang(Kashmir). Owned by the woman on the picture. She operates hotel ‘New York’, being the tent with a kitchen and ‘separated’ bedroom in it (dinner+bed+breakfast ~$9).

With a good research effort, the costs of such installation can go down a factor 4 or so. So these people can then afford a real solar installation.

Josh Nilsen's picture
Josh Nilsen on Jan 7, 2015

Many of these villages have to stop all active business when the sun goes down.  Outside LED lighting allows the small businesses in these villages (a small food cart for example) to stay open many additional hours.  The ability to allow productive businesses to operate at night is something the first world takes for granted.  Or just imagine trying to walk down a block in a bad neighborhood and there was no public lightning at all.  In the 3rd world, everything outside is a ‘bad neighborhood’.

Nathan Wilson's picture
Nathan Wilson on Jan 8, 2015

It seems to me that the main innovation in these micro-grids is the micro-power product they are selling.  For the 100 rupees/month payment (about $1.60 US) they are getting only a tiny amount of energy, but it is useful, and the installation and appliance costs are spread over their monthly payments.  (The service seems to allocate a few Watts of power to each house, whereas US homes are wired to accomodate up to around 12,000 Watts.)

Based on the images in the video on their website, it looks like each LED light is only a Watt or so, and the wiring appears to be strung with much less diligence and regard for electrical building codes than the typical American do-it-yourself-er uses when installing a 12V yard-lighting system (no doubt this is adequate because these low power 12V systems aren’t very hazardous).

Another interesting twist is that the systems are installed for groups of 10 households, who all share the power from one solar panel and battery set.  No metering is provided (apparently), so if one guy hogs the daily shared energy ration by adding other loads (or a free-loader surreptitiously jacks into the microgrid), his neighbors get less, thus they have a stake in preventing this.  The other implication is that if one guy doesn’t pay his share of the bill, they could all lose their service.

As to the scale of the system, one of the videos mentioned a plan to reach 3000 households in 40 villages in the current year, which amounts to an average of 75 homes per village.

I don’t think micro-power products like these will lift anyone out of poverty (only jobs in the high energy economy can do that).   But they do provide some comforts at an affordable price point.  Also, they allow the working poor to buy electricity, rather than merely hoping someone gives it to them for free.

Nathan Wilson's picture
Nathan Wilson on Jan 8, 2015

These systems are being built by people whose ideology says that a distributed solar micro-grid is the ultimate solution.  But the micro-power product can also be provided via a centralized village scale mini-grid, with a fuel powered generator to make it reliable (and get rid of the toxic and expensive lead-acid battery).

I’m imagining the person that runs the general store starts a mini-power company with a 1 kWatt generator (which alternatively could be a fuel cell running on propane, kerosene, methanol, or ammonia), distributes power as 240V AC to nearby streets, and hangs a step-down transformer on a wall or fencepost to convert it to 12VDC for a home or cluster of homes, with each home getting 5-100 Watts.  At this low power level, a very cheap smart-meter could be provided in the transformer which could implement rolling blackouts as needed, and provide measured rate service or shut-off power daily as each customer’s fixed daily rate plan was used up.  The 240 V AC power distribution system allows 270 Watts to be transmitted for half a mile w/ 10% loss using 16 guage wire (same as a typical extension cord).

With 240 VAC distributed to streets, less impoverished customers could also upgrade to 120VAC service (“real grid power”), without paying the upfront cost or the maintenance head-aches of having their own off-grid power generation system.

Bas Gresnigt's picture
Bas Gresnigt on Jan 8, 2015


If a research boost developes PV-panels that are 10 times cheaper, these people can pay for much more! Then even the poor villagers will have light.

Bas Gresnigt's picture
Bas Gresnigt on Jan 8, 2015

Some issues with your solution:
– As I experienced, those fuel powered generators make a terrible noise in the evening (and often you smell the exhaust gasses);

– The generator will be switched off at ~23hours or so. That is nice, unless you are not finished. You have no control;

– The generator is far more difficult to maintain than a PV-solar installation.
20 years ago I was in a village with a generator. It broke down some weeks ago, repair date unclear.

– Metering, fuses, transformes make the whole (too) expensive. And invite for stealing electricity.
Now people connect using some wires. They know how it operates because it is so simpel, and can adapt themselves (or some teenager around does it).
Our electricity installation standards are excellent, but too expensive for them. Those come gradually.

Nathan Wilson's picture
Nathan Wilson on Jan 9, 2015

Funny how you judge technologies that fit your ideology as likely to become very cheap with sufficient investment, yet you assume other technologies must remain too expensive.

The solar power system sold by the Mera Gao Power company is cheap because the small system is shared by 10 nearby households, and the power allocated to each household is only a few Watts.  

  • Based on SolarBuzz data, batteries cost $0.21/Wh.  A system that can supply 10 Watts per household for 15 hours, with 50% depth of discharge, will cost $0.21*10*15/.5= $63/ household.
  • The SEIA says solar PV modules cost about $0.75/Watt.  So to deliver 10 Watts per household assuming 25% capacity factor on sunny days, and 70% battery efficiency, cost is $0.75/.25/.7*10= $42.9 per household.
  • A transformer like this one is rated for 80 Watts and costs $16, therefore with 5/8 derating, it can supply 5 households with 12V power for $3.2/household.  A smart meter that can measure power and turn-on/off each of 5 outputs for 5 households, and communicate with the power company over a short power line would be similar.
  • A 250 foot long spool of wire like this one (14-gauge, 2 conductors plus ground, outdoor) costs $84; it can supply several households, depending on how close they are together and to the source (total 1.2 kW at 2.5% loss for 250 ft. and 240V).  I’ll assume a cluster of 10 households requires twice this length on average, for $16.8 per household.

So basically, for the solar powered system, the battery and solar panel are so much more expensive than everything else, that it makes little difference if we centralize the equipment at the end of a long wire.  In fact, it would save money, because less system over-sizing is needed: each household would get exactly the energy they pay for, with no issues due to the neighbors using too much.  The centralized system is easier to roll out too, since the original system needs 10 adjacent households to agree to sign the installation contract all at once; this one has a low cost of wiring that goes in first, but the expensive parts are added only after customers sign-on.

Regarding whether fuses and quality wiring are too expensive:  fuses (or other fault protection) are a must with batteries – their power output during faults is much too high for safety.  The smart meter would provide self-resetting electronic fault protection which is very safe, and very user friendly (no fuse to replace).  In both the distributed and centralized cases, the wiring on the customer premises is 12V, and is very safe and simple for the custom to install themselves.

Regarding, electricity theft: probably this is less a problem for small business owners who know their customers, than big power companies.  The very low cost of the micro-power electricity makes it less of a theft target.  Also with smart meters, the owner could tell how much electricity was going down each wire, so he could quickly see theft, and go look for the tap.  

Regarding whether the centralized solution should be solar+batteries or fuel powered:

  • I agree that generators are much too noising, unreliable, and difficult to maintain if every group of 1-10 households were to share one; however this is not necessarily the case if 100 households share a generator.
  • If the generator goes off for a few hours each night, and batteries are used then, the total cost of batteries is still around a third of what is needed for battery+solar only.
  • I suspect that fuel cells could be made very cheap (Lazard says $3.8-7.5/Watt today which already beats solar+batteries capital cost) and reliable (able to run 24 hours/day for years) with volume production.  However this likely requires sizes over 1kWatt, so centralized is necessary.
  • The solar+battery system will certainly be notoriously unreliable due to clouds, especially during long winter nights!  The likely fix, increasing the size of the battery and solar panel is very expensive.
Bas Gresnigt's picture
Bas Gresnigt on Jan 9, 2015

Agree sharing may be cheaper. So we saw that a few houses shared the solar panel.

What you think is cheap, is expensive for the poor people in such a village.
Those live on $1/day. So $40,= equals more than a month salary.
Worse: You can save 10% of your salary so you have that amount after a year. But these people need all the money for the basic things, such as food.

So we really need the research boost, so the price for solar decreases soon with at least a factor 4.

Then the real poor can also pay it, and the less poor can buy bigger capacities. So they can connect equipment which saves time, as well as a pc, etc.
So they can move up the social ladder.

The poor people in the world live almost always at low lattitudes, so little winter.
Fuel cells have their own problems, such as the fuel needed, etc.

Nathan Wilson's picture
Nathan Wilson on Jan 11, 2015

Putting aside the question of off-grid villages and thinking about the urban poor who live near established grids, I think micro-power service should be free.  And I don’t even believe it would cost the public or utilities any thing to provide!

Again, I think of micro-power service as 10 Watts peak per household, delivered as 12VDC (or 12VAC?).  Assuming an average load of 5W and electricity market value of 10 ¢/kWh, this electricity would be worth only 36 ¢ per month.

In places like India, this “investment” would likely pay for itself in reduced electricity theft (why steal power when a small but useful amount comes to your house for free?), and reduced power infrastructure vandalism associated with such theft.

In the US, it would be another form of safety net that would allow the working poor live on lower average salaries, to fare better in the event of job loss, and return to the workforce more quickly.

In all markets, a full price upgrade to conventional (120/240 VAC) grid power would still offer many advantages, so the utilities would still have compelling product to sell.  For example, labor saving appliances, tools, and refrigerators start at 1/4 horse power, or about 200 Watts, and electric cooking devices start at around 600 Watts; everything gets bigger and better as the power goes up.

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »