Are Solar Panels a wasteful way to go green?


Last week I wrote about what I consider the main goal of green electricity efforts, namely to stop burning coal. You can do that, to some extent, by removing demand from the grid in places where the grid is coal-heavy. Even in other places, removing demand from the grid will be fairly effective at reducing the production of greenhouse gases.

Update: Since this article a flood of cheap solar panels from China has been changing some of the economics discussed here. I have not altered the article but some of its conclusions deserve adjustment.

No matter what you do -- conserve, or put up solar or wind -- your goal is to take power off the grid. Many people however, consciously or unconsciously take a different goal -- they want to feel that they are doing the green thing. They want their electricity to be clean. This is actually a dangerous idea, I believe. Electrons are electrons. In terms of reducing emissions, you get the exact same result if you put a solar panel on your house than if you put it on your neighbour's house. You even get a better result if you put it on a house that's powered by a coal plant, so long as you also reap the benefit (in dollars) of the electricity it makes.

People don't like to accept this, but it's much better to put a wind turbine somewhere windy than on your own house. Much better to put a solar panel somewhere sunny than on your own house. And much better in all cases if the power you offset is generated by more by coal than at your house.

However, the real consequences are much deeper. The following numbers reveal it is generally a bad idea to put up solar panels at all, at least right now. That's because, as you will see below, solar panels are a terrible way to spend money and time to make greener electricity. Absolutely dreadful. Their only attribute is making you feel good because they are on your roof. But you should not feel good, because you could (in theory, and I believe with not much work in practice) have made the planet much greener by using the money you spent on the panels in other ways.

The true goal is to find the method that provides the most bang per buck in removing load from the dirty grid.

Keep reading to see the math and a spreadsheet with some very surprising numbers about what techniques do that the best.

Get your cost per year

To understand this, you want to work out your cost for removing a megawatt-hour (MWH) of demand from the power grid. Today, the average U.S. power company sells a MWH for about $100 (~10 cents/kwh.) This varies a lot, though. In California, a tiered system results in high users paying as much as $310/MWH for incremental power.

You can remove a MWH from the grid in two ways. You can generate it, with solar, wind or even a generator. And you can do genuine conservation, where you take a power demand that was clearly going to continue on, but due to your efforts and money, it is reduced, ideally with minimal sacrifice. The minimal sacrifice is important, because if people don't feel happy about conservation they have done, they will feel it has an additional cost, and be ready to not do it without much provocation.

Whichever you do, it's going to involve spending time and money. I will focus on money here, but all the projects involve time as well. (If you want to put solar panels on your roof, you will certainly spend a lot of personal time on the project. And staring at the meter going backwards.)

You must calculate the cost of the project, and how many years it will last. If it's more than a couple of years, it is vital you now consider the time value of money. Money in the future is worth a great deal less than money today. Energy delivered in the future is different, too.

You should also not include any rebates, tax credits or other subsidies on your project. There are various rebates for solar and other clean energy generation methods, but we want to know the real cost to get a MHW off the grid, not a subsidized one. That's because once we learn what methods are the most cost effective, we should then hope that these are the ones that are subsidized. It's the most bang for the subsidy buck. It would be a mistake to create a world where people put in solar because it is subsidized and don't put in wind because it isn't, if it turns out the wind is more cost effective.

Just like a home mortgage, you will need to consider an interest rate. Today rates are low, perhaps 6 to 7%. Even if you are going to do the project with cash, you must consider the cost of instead using the money to pay down your mortgage (in effect earning your mortgage's interest rate) or putting it into investments. While today in 2009 the stock market is poor, historically putting money into good long term investments has returned even better than 7%.

Figure out how many MWH generated/saved per year

Once you have a cost per year, you must work out how many KWH or MWH it will generate or save. From there you can calculate the base cost of taking a MWH off the grid. At today's prices in sunny California, that's about $300 for solar (PV) panel systems.

However, one important factor is left out, and requires a second number. You, or whoever is involved will not have to pay the power company for this saved MWH. So that cost, at your local grid rate, should be subtracted to produce the net cost/MWH. With the best systems, this net cost is negative -- you actually save money with the technique.

Now everybody should need little incentive to do the methods that save money. However, it turns out they often need a push for a variety of reasons. One is that getting the money for a capital intensive project like PV solar or even a new fridge can be difficult. People are notoriously bad at figuring out they should spend $800 on a new fridge that saves them $80 per year in electricity. The payback is there, but too slow for them. Another is simply a question of time. It takes time and will to make these things happen.

However, some of the methods you can use are so good that one of the most effective things you might do with your own money and time is convince others to do them, even if you don't reap any of the savings.

As I noted above, solar panels can take a MWH off the grid for about $300. At a typical grid price, that MWH cost $100 so the cost to you of doing it is $200, plus your time.

Look at the alternatives, like CFL for others

Instead consider efficient lighting, such as CFL (Compact fluorescent) bulbs. A 23w CFL can be had for about $3, and should last about 8 years. Replacing a 100w tungsten bulb it saves a lot of electricity, just like it says on the package. If put in place of a regularly used light bulb, it takes just $6 to remove a MWH from the grid when you don't count the savings. If you do count them (ie. it's in your own house) it means a "cost" (really a benefit) of -$100 per MWH.

Look at those numbers again. The solar panel costs you $200 for each MWH you remove, and the CFL saves (not cost) $100 for each MWH.

If you're the sort of person who would put up a solar system, you probably already switched to CFL or other efficient lighting. Good. But you could do more. If you took a few hours of your time, and just fourty of your dollars, you could walk down the street in some town giving out CFL bulbs to people who would let you take out their old bulb and install the new one, making sure the bulb will really be effective. If you manage to get just 13 bulbs replaced that are used for just 3 hours per night, you've taken as much power off the grid as a $7,800 1000 watt PV panel. All for $40 and an hour or two of your time.

Or if you prefer, you might hire somebody else to run around and give out and install bulbs. It might cost a bit more, but it will still be nowhere near the cost of that panel.

Here is a spreadsheet that works out the raw cost per MWH and the net cost, factoring in power savings. There is a row where you can plug in information you have on other forms of conservation and clean generation.

In this embedded spreadsheet you can "Click to Edit" to plug in your own numbers, or you can save it out as a file to use on your own machine.

Fridges and dryers

The bulbs stick out, but there are other things you can do. A new fridge, as noted, will save a lot of power. It's surprising how bad fridges from the 80s and 90s are compared to those of 2009. Buying one for yourself actually saves you money, but you could even go out and buy new fridges for any neighbours who have models that are 15 to 20 years old, and it would still be better than putting that solar panel on your roof.

You could also buy people new gas dryers. Because gas has gone up they don't save as much money as they used to, and they need a gas plumber to install, but they still beat the panels. Not buying one for yourself. I mean simply going to your neighbour and saying, "would you like a brand new gas dryer and the free services of a plumber at my expense to hook it up and toss out your electric one?" It's a real win if you single out people with large families and lots of laundry. You could do even better if you convinced them to use a clothesline more, but that's a harder sell. A new dryer is much easier.

So easy, in fact, that you can probably get them to give you back some of their savings. You don't have to just give it away. That could reduce your $100/MWH cost in giving away fridges quite a bit. In fact, they might be quite happy just to get a microloan, that they will pay back to you because they are getting a nicer fridge out of the deal.

The loan concept has other implications. Today, credit is hard to get. This is stopping people from putting in new wind farms and other forms of renewable energy, such as solar-thermal and geothermal.


Wind is much more efficient if you make a big turbine and farm. If green-minded people got together, they could simply donate money to put up a giant (50 megawatt) wind farm. If numbers from the wind industry are to be accepted, this would let you offset dirty grid power for about $40/MWH. Giving the power away! Of course the wind farm would sell it. If the wind farm breaks even after paying your interest, your cost to do this is net-zero. You're just tying up your money, but you get it all back, with interest. (You are also risking it, of course, but again, this is still cheaper than all he other methods except the light bulbs even if you lose all your money.) People looking to be green and ready to spend $30,000 on solar panels could instead band together and be much greener by offering financing to larger green projects that can't get it because they are not yet profitable or credit is too tight.

Ah, but...

I hear you screaming, "Ah, but...." so let me cover some of the common questions.

  • Yes, grid power may go up in price, making the solar panels more attractive. The point is, all the grid-reducing methods here get equally more attractive, per MWH, as grid power goes up in price.
  • Yes, California Tier III grid power is so expensive that solar can beat it. But the other methods beat it even more.
  • Yes, CFLs have mercury vapour in them. If they break, that's bad. But burning coal to power incandescent lights releases far more mercury into the air than is in a CFL. Whether the light breaks or not.
  • No, CFLs don't use more energy than they save being shipped overseas. Overseas shipping is much more efficient than you think. It takes about 30 minutes for them to save the energy needed to ship them from Asia.
  • Yes, solar-thermal is much more cost-effective than PV panels. This is mostly about PV panels, which are one of the most popular choices for misguided money trying to make the world greener.
  • Yes, the PV panels can raise the price of your home (as can new appliances) but that is cancelled out when you sell because you stop getting the energy from them. Leaving them behind makes your buyer greener -- but there are, I hope you have learned, much more effective ways to do that.
  • Yes, I know solar generates power during the "peak demand" period. That's great for power companies who want to even their loads, which is why they do time-of-use metering and why PV owners love that, but it does not much effect the amount of fossil fuel burned until we get really good at this.

Solar lighting

One area I don't explore because it's hard to cost out is solar lighting. One of the craziest things we sometimes see is a business putting up solar panels on its roof and using them to power lights -- even efficient fluorescent lights -- inside the building during the day. That's 3% efficient at best. Companies should work at ways to bring the natural light directly into the buildings, with appropriate redirection for the heat. This can be done with skylights and light tubes, among other things.

I predict this will be a way to get a lot of MWH/dollar, especially in business where lights are on during the day.

Cap and Trade

There is a lot of similarity to the logic in this article, and the results of a real working emissions credit trading system. If you have a true working system that rewards real generation and real conservation, the money ends up flowing to the methods that get the best bang for the buck.

The problem of course is enforcement. It's not cost effective to see if people really use their $3 CFL, and people are notorious at finding loopholes in rules enforcing conservation. Green generation is easier to track.

But if such a system did exist, it would be readily apparent what methods worked best, and people would do them.

Markets for green efforts

It would be nice if companies would spring up that would reliably, and verifiably take money and use it to cause real conservation and green generation. They would pay people to go door to door with free CFLs that they install, and they go round a year later to see if the CFLs have not been swapped back out to measure effectiveness. Companies that will arrange microloans for energy efficient appliances. Companies that will pool money and finance wind and other renewable energy methods.

Even, if you still love PV, companies that will subsidize putting up solar panels on Arizona houses (where they burn a lot of coal and have lots of sun) and give the financier the remaining money to offset their own grid costs in a less sunny place.

It won't happen because people do seem wedded to that psychological connection. They want to see that meter turn backwards even if it's a horribly money-inefficient way to make the world greener. There just is no satisfaction in putting a certificate on the wall saying "I made the world 10 times greener with my money and time." I understand why the right thing isn't very compelling. I wish we could find a way to make it so.

Other solar benefits

Many people like the fact that solar panel makes them "energy independent." If the power grid goes out, they are still getting power (at least during the day -- a normal grid tie system would have no power off-peak.)

But the reality is, grid power goes off so little that, polluting and noisy as it is, a gas generator would do this job better. In the not to distant future, it is predicted that you will be able to plug your hybrid car into your house to power your house in an outage.

Increasing demand for solar, even when it is such a poor performer, may drive down the price of solar so that it is able to compete. But that's true of all the other technologies, too.

Right now, there are just too many good opportunities available in conservation and in other alternative energies to put in PV. Down the road, as it gets cheaper, it will make sense, and deserve the money. However, a lot of evidence suggests that solar thermal power (where mirrors generate heat the spins a traditional turbine) is a big winner, in spite of its moving parts.

Solar power is distributed, which means it doesn't face losses in the power lines. That's good. But it's not that good. It, and all the other benefits don't compensate for the vastly greater cost/MWH saved.

Sadly, the only real benefit is that people feel good about it. They see the panels on the roof every day when they come home. They look at their power tracking system every day. They smile to see the meter run backwards on a sunny afternoon. That's all very nice, but is it worth hurting the planet to pretend you're doing the right thing? Harsh, but true.

Still, I don't want to sneer at the people who covered their houses with PV. Their goals were good. What's important is that they analyse what they do and look at the real numbers.


Solar can make off-grid sense, in locations where grid power can't be brought in. But there are other complexities. All off-grid systems must find a way to store excess power from the sunny periods to use when it's dark or not so sunny. The most common method is batteries. The problem is, batteries should not be kept in deep discharge. So a typical off-grid solar system is designed to get those batteries back up quickly after a night of discharge.

This means, sadly, that for much of the day, often the peak time, the batteries are not heavily discharged and simply can't accept the excess power. Many off grid systems have the panels spend their time throwing away the power because the batteries are full or nearly so. This makes the energy cost much more than it does in grid-tie.

I have even heard of people who imagine they are green by disconnecting from the grid even though they live in the city. This is terrible. Their solar systems throw away power all day which they could be putting back into the grid, stopping coal from being burned. They're doing a much worse job at greening the world than they could if they were grid-tied. Grid-tie makes sure every single watt-hour is used to offset load from the fossil fuel heavy grid.

The future

Conservation and improved energy-efficiency are fantastic and we should do all we can. But you can only conserve so much. In the end electrical generation must get cleaner.

In the future, after we've conserved everywhere we can readily conserve, and when PV comes down in price, as it will, then PV will be a competitor. In fact, when it is, PV will show up on the grid, as well as on people's homes. I am eager for that future, but it is not today.

Addendum on the Prius

Check the bottom of the spreadsheet for an example for a Hybrid car, such as a Prius. In this sheet we imagine you pay $3,000 extra to get the more efficient car, which gets 42 real world mpg, compared to 30 real world mpg for the cheaper car. You save 238 gallons of gas every year, at $2/gallon. More controversially, the DoE calculates each KWH of electricity requires about 1/12th gallon of gasoline equivalent at the power plant, so I map the cost per mwh to compare it to electricity.

Further update: I've added a line for solar panels where the panels are free but you still have to pay for other costs. Remarkably, they still don't compete. There's more on this in an update post about evangelical environmentalism.


Absolutely great article, Brad. Thanks for sharing the math.

I've invested some money in a local-ish community wind farm. The small town is putting up two wind turbines to reduce their peak consumption as well as lowering overall use (because uprating the main power line is expensive). It's a nice way to deliver low-carbon power where the generation capacity is.

Much of the low-hanging savings have already been done here - we've had energy companies giving away CFLs, water-efficient shower heads, energy audits and anything else they can come up with. We spent some time collecting those for both our place and anyone else where we could get them installed. Someone I know was unofficially doing that with any publicly accessible light bulb :) It makes sense, I did a few of the houses on my ride to work where outside lights were left on. But it's not technically legal. Our second fridge (6 member household) is a converted chest freezer that uses ~1/4 the power of the similar capacity upright next to it even though we store all the drinks in the chest one (because they're most often accessed).

The easy stuff is always going to be reduced demand. Same with water (we have a water shortage here). Also, doing the numbers based on no change in consumption is nonsense. One benefit of household solar/water tanks is that people magically reduce their demand when they have a concrete link to the source.

Some ... individual... here was claiming that to get any benefit a Melbourne house needs a 20kl tank (small garage sized) when 80% of the benefit will come from a 3kl-5kl tank (enough to run the toilets and washing machine for a month once people adjust to having the tank). Extra capacity in the tank also means significantly higher marginal costs of plumbing in harder to reach sites of consumption, and often extra filtering/treatment too as the water is stored longer or used closer to human consumption. We found that once we were running toilets on tank water our toilet water use roughly halved with no exhortations required (albeit a strong government campaign at the time "if it's yellow let it mellow, if it's brown flush it down").

How does that affect energy use? Melbiurne is about to build a desalination plant, possibly followed some time later by a wind farm to power said plant. But in the interim... burn more brown coal.

Typically I am the contrarian. This is clearly a contraian academic post which is great for getting the mind to work, but in this case bad for society. This type of argument is what I expect on talk radio - although not on this subject.

Anyway, you make numerous errors, which I am sure you are aware of - but many of your readers may not. For example, you state the reason to stop burning coal, but there are several reasons to consider panels; Green/carbon goals, ROI, independence, to name a few.

My home, in Boulder - has a 9.6 kw solar system recently installed. It also has high insulation, sky tubes, lots of CFLs, and thermal solar panels (thermal solar panels are a no brainer). My panels update an Internet site hourly with performance information Though I think I've only logged in a few times to review it.

But am I really hurting the planet? Yes, I could have paid for these panels and put them elsewhere to reduce carbon, but the logistics are totally against me. Plus the ownership of the panels on my home supposedly increases the value of my home - it should - my annual power bill is targeted to be somewhere around $500 (at current rates).

I would love to have instead put solar panels up in a more efficient place - but I didn't have that option at the sames costs. I would to do something to directly reduce the amount of coal burned in Boulder - but I don't have access to Xcel decision making. But the Government did give me a nearly 50% incentive (one of the best incentives nationally) to put some panels on my roof. The end user price of my system was about $40k - that's a lot of money - I could've bought a used Mercedes for that. But all in all, I think the panels offered a better ROI.

It was actually a tough decision. I know there is more solar R&D going on now than ever before and my panels will likely be obsolete very soon. But I factored in my personal consumption and lifestyle - the incentives, the performance of my mutual funds (this is over a year ago), the liklihood of electricity prices increasing, the fact that Boulder does burn coal (instead of hydro), the liklihood we will have a plugin electrical car in the next few years, and other factors, and decided it was a good investment. Really the panels were free - the Govt paid for 50% and I sold mutual funds which since dropped by 50%. You say I should not factor in the incentives, but that is awfully academic. When something goes on sale, I consider the new price, not the old price to determine the value.

My ROI was based on the cost of electricity going up 5% a year - but in the first year it already went up over 10%. Carbon taxes could drive those increases year over year even higher.

Mongo only pawn in game of life.

Where do you draw the line? We make the best decisions we can with the information we have and move on. Try to learn, re-evaluate as necessary. Coal is bad, solar reduces coal. Tomorrow - are you going to be part of the solution or part of the problem?

On the house remodel we went thru a lot of decisions balancing emotion with desire and cost. Wood flooring? Bamboo is considered the greenest since it is harvested instead of killed to get the product. But the product all comes from China - ocean freighters contribute more pollution than all cars and trucks, Factor in Chinese human rights and environmental impacts and it becomes pretty murky pretty fast.

Counter-tops? What can be more green than granite - pure rock - no chemicals. No granite used in counter tops comes from Colorado or even the US. It all comes from S. America, Middle East, and Asia. Complex mining/raping, and shipping (ocean, trucks, trains) of one of the heaviest materials on earth for a kitchen counter top? Which is greener, granite or local Formica? Murky. It never ends... asphalt or cement driveway? asphalt is cheaper, but uses imported oil. Paper or plastic bags? Cloth or disposable diapers? You can create your spreadsheet for any of these topics and I can recreate it with a different outcome.

The fact is on many levels you are absolutely right - and you could go much further (paralysis by analysis). But the fact is the system is rigged. Doing the right thing is very difficult. I can get subsidized panels on my home very easily and make a difference today rather than work to change the rules for tomorrow (also important). Sometimes I go for the quick fix, sometimes I don't.

I don't have a Prius, but I think they are fine cars. They get 45-55 MPG in a comfortable 5 passenger vehicle that is fun to drive and increases energy awareness. You are correct that a Honda Fit is a superior value - but I don't want a sub compact 4 seater that can hide behind a trash can. Cars are one of the most emotional purchases we make - a Ferrari has no logical value in Colorado, but tell that to the guy that owns one and you might get a black eye. In fact, it was proven that sports cars actually increase testosterone, put that in your slide rule. As much as we want to, we can't rule out emotion. If solar makes you feel good - why is that any worse than other emotional vices? A hot tub, a bottle of imported whisky, a fast car?

You say there are errors in the numbers. Please be specific if you can.

You spent $40K of your own money and $35K (I am guessing, you said almost 50% off) of tax money to put the panels on your house. What could have been accomplished with that $75K? How many kwh per year does your 9.6kw solar system generate (I will take a guess that it's around 15,000, but you may know the exact number and I am interested.)

For $75,000, could we not pay somebody to walk the streets of less green cities, find people with incandescent lights (a simple colour spectrum meter pointed at their windows and porches would tell you that right away) and give them CFLs? There are lots and lots of incandescents out there. I still have some due to the X10 systems I have. For $75,000, how many people could this person convince to switch lights? Each light would save over 60kwh per year. That's 250 bulbs to change. For $75,000 -- or $300 per bulb, perhaps $1000 per house, I think you could easily do a lot more than 250 bulbs. Ten times more, in fact, and take 10 times the power off the grid that your panels take.

Likewise that $75,000 could have paid for someobody to wander a poor area and find houses with fridges from the 70s and 80s. We could buy 100 fridges for that money that poor people would love to get, and that would take 100,000 kwh off the grid, not 15,000. That's if you give them the fridges for free, even though the new fridges save them $100 on their power bill each year. Fridges won't quite last as long as solar panels, so halve it and it's still 50 MWH vs the 15 mwh you're getting. By choosing to put up panels rather than buy new fridges for your poor neighbours (and make 100 very grateful new friends) there is 35MW of extra electricity generation each year -- that's 53,000 lbs more CO2 in the air because you spent your money (and the government's) on panels rather than new fridges for the poor, to give one example.

(I have not accounted for the power you save, or that the neighbours save, in this example. If the neighbours paid you 1/3 of the power savings they get, it would balance out, though arranging that is another issue.)

Now I'm not really calling you a villain. Not only are people not aware of these numbers, but there are no easy programs in place to let them do this, to let them give CFLs and fridges and dryers to poor people. If there were, you could do much better than panels, but I expect that even without them, it's still possible.

Also, I do believe my numbers account for the added value of your house. My spreadsheet calculates the carrying cost, per year, of the capital item, be it panels or fridge. Each year, owning the panels, like owning a house, has that carrying cost. The panels, unlike a house, depreciate rather than appreciate, and most people quote lifetimes of about 25 years, which I have used. I think it's fair that 25 year old panels won't add much value to your house after that period, both because they will be starting to fail, and because vastly better panels will be available then.

You are confusing charity with a reasonable investment in solar power.

I understand charity - and contribute to various groups or causes that I wish to support. Sometimes they even send me a bumper sticker or canvas bag - but that isn't what drives the contribution. I am as happy as the next guy to give my time or money to important causes.

The errors are that you keep looking at the total cost. Using our combined figures above - I didn't give or donate or pay $75k, I paid $40k. You can't compare the $40k to $75k in fridges. So now we are at $40k of fridges which is still a noble deed. Of course, you didn't consider the environmental costs involved in manufacturing those fridges or the fuel and pollution associated with transporting them or even the landfill impact of tossing the old ones, but why would you? That would be an academic exercise that misses the point.

So we are at $40k. The Government presented me a deal that said if I spend $40k, they will give me an additional $35k in value. That is pretty compelling. If I get enough deals like that, I really could buy a bunch of fridge's for donation. But on top of that - I get an asset that increases the value of my home (hard to say how much, but yes as with almost all home improvements it will depreciate (appliances, carpet, windows, etc.). Plus I get 80% of my electricity rebated to me and since electricity prices are more likely to increase than decrease - the value of the rebate will also increase. So it really isn't $40k at all but much less. I don't recall the official ROI which was based on these and other factors, but probably around 12 years. I know it was based on modest 5% price increases in electricity, but I expect higher - so let's call it an even 10.

So if I break even at 10 years or so, and they have a 25 year lifespan - that means that not only will they contribute to society as a whole with pollution-free energy production for 25 years, but they will actually earn me money after a decade.

So My $40k investment actually has an environmental impact and a personal ROI. The fridge's have an environmental impact and arguably a social ROI - but that's what my charitable contributions (and taxes) are for. The government also offered me a tax incentive to buy a Hummer last year, which I declined. It didn't align with my personal goals and objectives.

Hindsight is always 20/20 - in this case, I can honestly say that my $35k would have completely disappeared, and not turned into fridges or panels at all - because all of my stocks crashed.

I was also indirectly involved with a project to put solar panels up on a local humane society. The project was funded 100% by charitable contribution. These panels are also "green" and attracted donors interested in being green - but not only does the environment win, but it will re-allocate $1000 a year from utilities to animal care which is their mission. It is nice to be able to be successful in a mission and help out the environment.

It's $75K because you spent $40K and various forced convinced the government to kick in $35K because of the (arguably mistaken) belief that solar panels are the right thing to subsidize to green our electrical load.

There is $75K to spend (it doesn't matter from where) and we want it to cause the most possible reduction in grid load. What is the way to do that? It's clear that replacing lighting is the biggest winner I can see so far. Replacing fridges is also good. Putting in solar panels, in comparison, is very poor.

It has nothing to do with charity.

As to when you break even, right now solar never breaks even at the national average grid price, even with the big government assist. Tell me your numbers (grid price, grid price increase assumption, kwh generated per year by your system, rebates and tax credits, current mortgage interest rate) and we can work out when your system breaks even, if ever. Right now the typical solar cost is about 22 cents/kwh without the rebates, I believe, which is of course never pays for itself because it is much more than typical grid. With rebates I've seen it get down to 12-15 cents which almost breaks even. In California with the tiers, both these numbers can break even, so long as you don't put on so much solar that you don't even fill up tier 1.

However, again, and this seems to be hard to make clear, it doesn't matter when any of the technologies "break even." If a technology saves 1 MWH, then it returns a benefit to you of about $100 at typical retail prices, $40-$50 at typical wholesale prices. However, other than the retail/wholesale distinction, it doesn't matter, because everything that saves a megawatt returns the same savings. It's what it costs to generate that which matters. You can say that anything which returns more than it costs is justifiable, but what if you can do twice as much with the same investment of time and money?

Add this to your "Ah, but..." section: Does anybody make a CFL bulb that actually *is* as bright or brighter than the incandescent it's allegedly "equivalent to"? I think CFLs would be vastly more popular if they started out a bit rather than a bit dimmer than whatever they were being compared to. Even if that meant you only saved 50% rather than 75% on the power bill. A 23 watt CFL simply does not substitute for a 100 watt bulb. Even if it appeared comparably bright the day you bought it - which usually isn't the case - the brightness is likely to decline to an unsuitable level before long so you really need to take that into account and add an extra boost at the beginning so the reduced output level isn't a problem.

(Wikipedia says: "In one test by the US Department of Energy of 'Energy Star' products in 2003-4, one quarter of tested CFLs no longer met their rated output after 40% of their rated service life.")

Lost a word there in a last-minute edit; I meant to say: "if they started out a bit brighter rather than a bit dimmer".

In short, maybe we should be calling 50 watt CFLs "equivalent to" 100 watt tungsten bulbs. Anything much less than that guarantees some fraction of consumers are unhappy with their experience and the poor reputation of CFLs continues. Trying to skimp and recommend something that provides just barely enough light to pass, but only the moment you buy it (and only if you wait a few minutes for it to warm up), is a false economy.

I've certainly noticed warm-up time. I would be interested in more figures on the decline because that does shorten their life. Though they are so cost effective you could double the numbers and they would still be anything on the chart.

Another option is moving people to non-compact fluorescents for more interior lighting. This could be done two ways -- less compact tubes that still fit in a regular screw-in socket but require a large space, like a large lampshade or large hanging lamp, and of course buying and installing entirely new light fixtures meant for larger lamps. Again, this is so efficient that even $100 for an installed new light fixture still performs fairly well.

Do the lights take less power as they decline? In that case you could start with say 25w CFLs in your 100w locations, and after a couple of years swap them into your 60w locations when they got dimmer.

Regular fluorescents are very efficient, more than CFLs. And newer models are even better, with instant-on no-hum electronic ballasts and redder colour spectra. You need about 2' of tube, which coiled up can be made reasonably compact, though not bulb sized.

I've purchased CFLs a while back when they weren't cheaper than incandescent

I've also purchased them years later and found the boxes featuring the local power utility's logo advertising that the bulb's price was subsidized by our friendly local forward looking utility company.

I've seen huge billboards advertising the twisty bulbs encouraging people to look past the new perhaps weird look, and adopt the more energy efficient bulb.

I've heard news of banning the sale of incandescent

Seems to me CFLs had a huge marketing problem: They look weird (though it didn't bother me.. many ppl had aesthetic concerns)
They weren't cheap: Prices have come down a lot (not sure if gov't or utilities are involved in that still, but they certainly have been)... even though they might last forever compared to incandescent, people don't see the benefit of something that makes it's money back over time and has a big up front payment.
They don't even work as well: people expect light to come on at full power right when they flick the switch. That doesn't happen with many CFLs, and some of them buzz...

So.. Gov't banned incandescent, utilities subsidized CFL prices and erected billboards advertising for their usage.

Perhaps these gov't actions were totally appropriate considering the decrease in energy usage..

Or maybe I could make the argument that the technology should be perfect: no buzzing and it had better turn on instantly at full power when I flip the switch, and until CFLs can do that, government should be investing in CFL and LED R&D, not funding the adoption of a current flawed technology that needs to be subsidized and have the existing incandescent options banned from the market place before it can compete..

If I'm not mistaken, this response to the energy crisis
was really a response to Enron's shenanigans.

The biggest problems with CFLs is that most of them are crap.

Most of them have horrible warm-up time, color, etc.

A few of them are good.

I'm not the best judge of color quality, but at least in terms of warm up time which is annoying for any light which isn't on for long periods of time, the best bulbs I've bought are made by Feit Electric. I've gotten a couple different bulbs made by them and they've all been instant on (just a fraction of a delay before the starter kicks the lights on). I'd judge the initial light output to be 80-90% or barely discernible from 100% light output.

Ones that have long warmup times:

Ikea. These were enclosed bulbs with some textured glass. Look nice, but poor light output and slow warmup.
n:Vision. Got these from Home Depot. Slow warmup. Widely varying initial light output across the 4-6 bulbs I got. Work fine once warmed up, but takes a while to do so.

Would be nice to collect more input to help people avoid crappy CFLs. They give CFLs a bad name.

There is a serious problem with your economic analysis. You assume that coal prices are perfectly inelastic, or equivalently that coal supply is perfectly elastic. You also assume that electricity markets are local.

Suppose (as an unrealistic hypothetical) that the opposite is true: there is a fixed supply of coal and prices adjust only according to the demand curve. Then reducing my coal use makes zero difference to the total amount of coal consumption! It only makes coal cheaper for everyone else, whom we have assumed will increase their demand to exactly make up for the coal I'm not using.

Realistically, coal supply is neither perfectly elastic nor perfectly inelastic, and the result will be that any personal reduction in coal demand will be partially offset by the fact that it lowers prices and increases demand for others. The coefficient of elasticity has to be estimated empirically. The fact that plants are large fixed costs probably makes supply less elastic.

Similarly, if my local utility company uses less coal, then the price of coal will be lower for other states, who may increase coal use themselves.

Despite it being an imperfect solution, we should still personally conserve, but let's not lose our focus on the need for a carbon tax or cap-and-trade solution.

In fact, it is in some ways worse that you say. Right now the cheapest thing for utilities to do with reduced demand is burn less natural gas. However, with enough reduced demand they will burn less coal. And cap and trade rules at the power plant level can be done, and can be built so that effectively power companies are required to cut back on coal when demand for fuel based power is reduced, rather than cutting back on natural gas or renewables. The requirements that a minimum percentage of their power be from renewables makes the latter happen anyway.

But the core point is not wrong. Those who put in local generation, or who do conservation, are all getting the same result -- taking demand off the grid. This article works out the numbers for how good each method is at doing that. Generating a MWH or reducing somebody's consumption a MWH have the same result.

Once the power is taken off the grid, you are right that the results get complex.

Power is a strange mix of local and regional. There is no national grid, but there are 3 large regional grids. However, due to line losses, and not having to pay remote companies, there is a slight incentive to generate the power closer to where the demand is.

Brad.... I was worried when I saw your earlier post warning of an upcoming analysis critical of solar PV. I knew it would be well written. Fortunately I see some major flaws that the previous comments have not touched on.

Here goes....

My biggest concern with your article... If everybody waits for the perfect PV panel, I don't think it will ever see the light of day!!

2 years ago I listened to the CEO of Nanosolar speak at the commonwealth club along with a member from, a member of the installation community and a few others..

One of the final questions of the evening was a practical one asking what people should do if they want to "go solar" today. The Nanosolar CEO essentially said, wait..... don't go solar now... wait for my product to hit the market.. it will be here next year and it will be better than what is on the market today!

The next day, I sent that CEO an e-mail and begged him to stop telling the public to "wait and see".
If everybody waits on the sidelines in anticipation of a 30% efficient module, or a 50% efficient module, or whatever the CEO of "vaporware solar" wants them to wait for, then nobody is buying "today's PV", and the existing products on the market fail. If the existing products fail, there is no future product improvement and we never get to the 50% efficient panel. I essentially told him he was shooting himself in the foot on both the funding and demand side. If his product comes out in 3 years, the market will be all that much more ready and enthusiastic for it if the products that came before his are on the rooftops of the neighbors of his prospective customers. Simply put, a neighbor (a) is happy with 5 year old modules.. I'm going to be more comfortable to install the next gen product of what my neighbor has had a good experience with for the last 5 years.

he thanked me for my note.

Clean tech was a huge area of venture capital in 2007 and 2008 and that has fueled tremendously important R&D which could result in hundreds of viable products that could be better than today's crystalline and thin film modules. But do you think Kleiner Perkins and the rest of Sand Hill Road would have been investing money in Nanosolar or Bright Source or eSolar or Solyndra, if there was zero market demand for current products on the market!?!

I'm sure the utility scale solar PV and solar thermal power plants of Bright Source and Stirling will continue to fight it out and some of them will be built, and hopefully the utilities will meet their renewable portfolio standards that were so crucial to getting them off their carbon burning asses... Like you, I'm hopeful that the solar thermal power plants that PG&E and others have in the pipeline turn out to be a terrific success.

But I absolutely disagree with you when you suggest that people who installed solar PV on their roofs in hopes of locally powering their loads and preparing for their future hybrid or EV that they'll be plugging in are "hurting the planet" because they chose that instead of an even more effective way to spend their time for clean energy future by going door to door passing out free CFLs.

By your logic, I could make the argument that your CFL donation team is "hurting" the environment by passing out CFLs because the LED lamps use even less power...

should we have all stayed away from CFLs because LEDs were only 5-10 years away?
I think if we had ALL collectively turned up our noses at the "weird twisty" bulbs (which many did for a while!), LED lighting would still be 5-10 years off!!

Current PV products are damn good, many have argued they are already at "grid parity", if you properly account for the massive defacto subsidies that gov't gives conventional grid power. ex:

"Sadly, the only real benefit is that people feel good about it. They see the panels on the roof every day when they come home. They look at their power tracking system every day. They smile to see the meter run backwards on a sunny afternoon. That’s all very nice, but is it worth hurting the planet to pretend you’re doing the right thing? Harsh, but true."

I disagree..

1) They are not "hurting the planet" by choosing to go solar.
They would be hurting the planet if the were choosing to do neither solar nor conserve energy nor pass out CFLs and hold onto their fridge from the 80s and their "classic" heavy polluting car from the 60s.

2) I disagree that the "feel good" aspect is the only benefit.. They are also sending a clear demand side signal to business, and government that "we the people" want to solve this problem! This informs venture capital to direct their cash stash into clean tech. CEOs of companies like Google, Applied Materials, National Semiconductor and Hewlett Packard dedicated teams of smart silicon valley technologists onto clean tech projects. An installed base of rooftop PV may even have strengthened the resolve to Andrew Frank at UC Davis and forward looking auto companies to keep battling on with plugin hybrids and EVs.

As for replacing the fridge, do you know if anybody has even come close to quantifying the "cost" of throwing stuff away?
I think once our economy properly attributes value to "cradle to cradle" and cost to "throwing stuff in our water and in landfills, then maybe we can accurately talk about how much energy we saved by replacing a fridge with a new one or trashing 24 inch CRT for a 30" LCD. In case you don't the fact is most people actually increase their energy usage when they get a new fridge because they put the old one in the garage maybe only using the freezer section.

By the way, I've recently heard that a number of the video game systems PS3, XBox, and our newish TV friend the DVR now consume more power than the fridge.. So now silicon valley wins the badge of shame with computers sucking more power than our "major appliances"

I should say... I'm a BIG believer in conservation...

Buying Solar Panels emphasizes further consumption as the "fix" for too much consumption of energy.

clearly the problem is too much consumption!

so yes.. use less power.. please install CFLs.. give them out as gifts for Halloween and when Christmas caroling... get some LED Christmas lights.. or don't hang them at all :-)

I'm a power mizer jewish kid and always saw them as a lovely festive waste of energy!

What does your excel spreadsheet skills say about idle currents and smart power strips, and my AC adpater for my ipod. and the other for my bluetooth, and the other for my cell phone?

No, I'm with Martin from Nanosolar. He's right. Put your money into pushing conservation and cost-effective green power (thermal-solar, wind, geothermal, skylighting) instead of into PV panels. PV panels are going to get cheaper.

And no, I don't think that people buying them when they are not cost effective helps prove the market for cost effective panels in the future. Well, it does to some degree, because it measures how much bias people will have towards panels when all other things eventually become equal. But is demonstrating that love a good way to use such large sums of money? Surely if people were putting money into funding wind farms or solar thermal plants or buying new fridges for their neighbours, the investment community would have no doubt that these people would jump at solar panels once the panels were a more cost effective way to reach the goal.

We should put our money and time where it should do the most good. That is the thesis here. Do you disagree with that? Do you really think PV panels are where it does the most good?

I should put LEDs on the list though today they are actually a fair bit more costly than CFL. I am happy to add all sorts of things to the list. As for conservation vs. green generation: While the believers will do conservation for the sake of conservation, the public doesn't. You have to show the public why it's in their direct financial interest. A light bulb that produces the same amount of light and lasts longer is in their interest. A nice new fridge with a lower power bill is in their interest. Sacrifices they don't like, and convincing people to do sacrifices is a much harder path to success. Focus on what works best now until you can't get returns from it, then go onto the next level. I believe generation is essential because of this.

Yes, we should fix the wall warts. On other sections of this blog I have long articles about the need for a universal DC power system which, among other things, would be more efficient as you would get one power supply running everything on your desk, or even have one in your house.

Yes, recycling that fridge has costs. If you have analysis that shows it is more expensive to recycle than keep running, let me know!

The fact that coal and gas don't pay for their externalities doesn't alter this thesis at all. All the methods outlined -- PV, wind, conservation -- all factor against the same grid. If grid power is made more expensive, it does bring solar down to being competitive with it, but it makes the other methods even more competitive. The base cost/MWH of offsetting grid power does not alter with the cost of grid power. Only the net cost. I should make it clearer in the spreadsheet but the net cost is essentially always $110 less than the bulk cost when working in the home (at an 11 cent grid rate) and lower if you are a power plant selling the power at a wholesale rate.

Once again, the thesis is simple.

  1. We should focus our money on time on those methods that produce the most energy offset for the dollar
  2. Solar isn't even close to being near the top of that list

Which do you disagree with?


I was really hoping I'd have an easier time proving you at least partially wrong!
I've worked for two years for a Solar Power integrator designing and permitting PV systems for residential and commercial applications. Work consumes much of my time and energy, but I'd like to at least tell myself it's for a worthy cause!

I felt like you were a bit too quick in dismissing my argument that future improvement and product cost reduction if a function of the success of existing products on the market.

Do you think the level of venture capital investment in clean tech would have been the the same over the last few years if nobody was buying and installing the current products on the market? If Google had not done this -, if nobody was installing on their rooftops, and the multi-MW farms in Europe and all the rooftop PV in Japan did not exist, do you think investors would have the same confidence and willingness to invest in the next generation of products?

I do not have any analysis to show you about the true environmental cost of disposing a fridge. I do however think that somebody should win a nobel prize if they can figure out how to recalibrate the global economy to properly attribute a realistic cost to throwing stuff away. I was living in San Diego for the last three years where they have free trash pickup.... absolutely zero fees to throw stuff away, and zero incentive to minimize waste. Mind boggling in its social irresponsibility. It's long been considered political suicide to change.. but perhaps hell is about to freeze over:

Both of these links confirm what I said about most people moving the old efficient fridge to their garage thus increasing their energy consumption when they buy a new one, so your fridge distribution plan would definitely need to pry the old one from their wall socket.


Yes, I fully agree, you can't just give them a certificate for a new fridge. You need to do it through a company (who would gladly oblige) who delivers and picks up the old fridge.

I would go further. To facilitate this, I would start a program that told manufacturers of fridges that they had to install a watt-meter in the fridge to get with the program. This is not that expensive. The new owner could see the kwh. The watt meter would be programmed to, knowing the efficiency ratings of the fridge that was replaced, display how much power this one used compared to the other one and how many dollars it saved the owner. It would know when the door was open and closed, cooling used etc.

If you did this, you could even have the new owner pay back the donor of the new fridge with half their energy savings.

For example, we find a poor family with a fridge from 1980. Plenty of those out there. We give them the new fridge. It saves them $100 per year in electricity. We require them to send $50 of that to us. The meter in the fridge tells how much. If they don't send it, we come repo the fridge. They have to believe that the meter is correct and the figures for their old one are correct, and they have to understand this is not costing them money. They have to understand their power bill went down $100 and they paid $50 of it out.

Another way to do this would be to forget the meters and work on averages. The power company would collect the $50 by adding it to the power bill, and remit it to the fridge donor. So they would get a power bill that was $50 less than before -- and a free new fridge -- and the donor would get a $50 cheque for 20 years that pays for most of the money they donated to buy the fridge. And middlemen skim amounts to grease all this.

I've done some reporting on the questionable value of photovoltaics mounted on cars, including the optional solar roof to be offered on Toyota's 2010 Prius. See Toyota's Underwhelming Solar Prius

The value of a solar panel on a car is quite dubious. For proper performance a solar panel is aimed at the sun. If stationary, that means at least pointed south and tilted to your latitude. A horizontal panel is not nearly as good, and one that will likely often be parked in shade even worse than that.

Further, the normal goal of a Prius is to have the battery be charged when you stop, so you can start electric. But once the battery is charged, the solar panel's energy is thrown away, except on things like the cooling described. You could perhaps tell the Prius "I'm going to park you in the sun in 5 miles so stop charging the battery with gasoline so the solar panel will have a chance" but will people do that and will they do it reliably?

Just a thought, Brad. When everything goes electric maybe it will become mandatory to plug a vehicle into a socket where one is available so excess current could feed back into the system?

Just for the record - Toyota never claimed that the solar panels on the car would do anything but cool the interior on hot days. A panel the size of your car roof will generate 100-200w at most. Assuming the upper end of that and you parked in the direct sun for 8 hours, that would only produce enough power to move your car 2-5 miles.

That said - If you've ever lived in an area where your car will bake in the sun, having a solar powered fan that keeps interior car temps near ambient temps instead of 150*F+ is worth it's weight in gold. Especially if you have young kids and are worried about their safety and well being climbing into hot car seats on a summer day.

Keeping the car's interior cool will also significantly reduce the load on the A/C. It can take 10-15 minutes of 100% A/C to cool the interior of a car sitting in the sun to comfortable levels. Doing so takes a lot of extra power.

So a solar car roof may not be as green as it appears, it's certainly useful.

In that the greatest heat matches the time of solar availability.

The key thing to realize is that solar panels only work properly if you use every joule they generate. The more you throw away because batteries can't take it (or because they are in shade or not pointed right) the less productive they are, and since they are not super productive it's easy to fall off the cliff.

The Prius calculations are totally wrong.

Your average car will last 9-10 years before being junked. It said 3.
Your average call will be driven close to 15,000 miles/year. It said 5000.
Your average Prius nets 46mpg. It said 42mpg.
You claim that the non-hybrid averages 30mpg. The most fuel efficient non-hybrid family sedan I could find was the Hyndai Elantra which averages 28mpg. You have to get a compact car to average 30mpg in a non-hybrid. (See

Anyway I've corrected the numbers. Still doesn't appear to be quite as good as some of the other options, but obviously how "good" it is depends on a lot of factors.

It looks like the sheet was set by accident as publicly writable, and somebody changed it. Sorry about that.

Brad - Thanks for the very well thought out explanation of the true costs on Solar Panels. I find many articles touting School Districts and higher education campuses employing “Solar” as their product of choice to cut energy costs. These decisions are presumably made by intelligent, well informed people, so I am confused. Why it is that a person of authority would choose solar when clearly it is not the best use of their limitied funds?

I say this because my husband’s company recently expanded its consulting services to Schools Districts to include energy efficient products and services to help with the budgetary contstraints they are experiencing. The goal is to cut their energy usage and costs to free funds up for other "Educational" needs.

My husband looked at Solar but ended with the same thoughts as you in that Solar was not the way to go. He found a manufacture who, along with many other pros, provides the only UL Listed LED T8 replacement tubes in Northern America. Different models range in consumptive power from 10 - 22 Watts for the same amount of luminance as the standard 32 - 40 Watt 4’ fluorescent tubes currently used in most classrooms around the country.

They have calculator that, in its simplistic and conservative version, indicates a savings of up to 70% of energy costs with an average ROI of 3 - 5 years and an approx lamp life of 11 years. Because these claims are proven, he and I believe this is the quickest and most efficient way to save energy and costs for our schools. We are however, receiving much resistance from the districts and believe it is because of the newness of the technology. Do you have any input on this type of LED lighting and/or can you tell me smart people are making dumb choices by choosing Solar over LED?


Though I absolutely appreciate the cost effectiveness aspect of your article,
our (my families) goal is to not just be more environment friendly, but also less dependent
of "government" services, the grid.

If and when the grid goes down, if we haven't taken measures and prepared to utilize other
means of energy, water, sewage, etc., we may all find ourselves in a similar situation
as those folks in New Orleans and on the north east coast. Waiting for the government to
bail them out or help them.

Though cost is something to definitely think about and consider, over all independence is another.


Add new comment