GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Rooftop solar breaks the buck a watt barrier!

Dana1 | Posted in Energy Efficiency and Durability on

Recent pricing for a ~5kw rooftop system in Australia have recently been hitting under $1.40 AUD (=$1 USD at today’s exhange rate.) Even pipsqueak sized 1.5kw systems are costing only a buck-fifty USD per watt.

In the US the average price of rooftop solar is running $3.40/watt, more than THREE TIMES the cost of solar in Australia, using the same panels, inverters, and racking systems, with comparably priced labor markets:

It’s only a matter of time before the US hits that price point.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. GBA Editor
    Martin Holladay | | #1

    It's frustrating to live in a country (the U.S.) with high PV costs -- but at least the costs continue to drop.

  2. Expert Member
    Dana Dorsett | | #2

    It's mostly a matter of volume, and the maturity of the market. Australia over-built their fossil-fired generation capacity and transmission grid capacity in the late 1990s and early 2000s in anticipation of steeply increasing demand, demand that never materialized in part due to the high capital cost of those capacity expansions. An aggressive feed-in tariff for all renewables under a more left-leaning government (prior to the right leaning Liberal party under Abbott) kick started that market in high gear, a bit less than a decade ago, just as logjam from the raw silicon pricing that had temporarily slowed the price of PV panels broke free. Between 30 cent electricity pricing, the free-falling price of PV components, the high early FIT, and the very favorable insolation levels, the Australian market soared and matured very quickly into a highly competitive market, making it one of the cheapest retail solar markets in the developed world, right up there with Germany.

    Investors in Australian fossil fired generation and the over-capacity grid have been fighting for their economic lives through legislative and regulation maneuvers (supported by the Liberal party during Abbott's tenure) and solar owners do NOT enjoy net-metering at retail the way it is for much of the US, paying nothing or next-to-nothing for power exported to the grid, attaching high fixed fees (a popular strategy in the the US). The paltry remuneration for exported PV power has created a pent up demand for smarter controls to divert excess PV power above the home's immediate loads to things like hot water heating rather than donating that power to the grid, but now that small scale power storage in batteries has passed some price/performance milestones, there is a HUGE market for behind-the-meter batteries to support the existing high penetration of rooftop PV, which has also reinvigorated the already competitive PV market there.

    At this point high fixed monthly fees by utilities can no longer work as a strategy for recovering their sunk costs for the capacity that will never be needed. With buck-a-watt solar it's financially rational for Australians to quit the grid entirely (which is actually a very rare occurrence), or never hook up at all, which is becoming a standard alternative for more remote areas near the grid edge. Even though the capital cost for the last couple hundred meters isn't large, the financial math indicates it isn't always worth it, especially in areas with high monthly fixed charged, and a regulatory environment that is currently considering allowing large fees for disconnecting from the grid. (You can quit the grid, but it'll cost you more than it initially cost to hook up- WAY more.)

    What's happening in Australia is a hint for what's likely to be happening across the US in a few years. The recently passed extension of the income tax subsidy for solar and production tax credits for wind are long enough that the ramping up of the US market will continue to be exponential. GreenTechMedia's policy wonk number crunchers suggest that the extension will result in a total capacity in 2022 that's fully 54% higher than what it would have been if the subsidy had been allowed to lapse at the end of 2016, and the market then will be many times the size of the market in 2015. (See: ) This increased volume will help drive down costs, and by 2022 it's almost unimaginable that PV won't have a lower lifecycle cost of energy well below grid-retail for nearly all US ratepayers.

    And with all of the zero marginal cost PV & wind power going onto the US grids, the wholesale market for fossil power will have to be restructured to keep it from collapsing entirely. It's already a terrible time to be in the fast-ramping peaking generator business (that business is done- stick a fork in it) but by 2022 it will be affecting the margins of all power generators that have a fuel cost component. The wind industry folks are suggesting 20% of all power going onto the grid in the US will be from wind by 2030 (I'd hazard that to be a much too conservative number), but the exponential growth of PV may well eclipse wind for total market share, mostly on a total cost basis. Wind installations have an economy of scale, but have to be fairly large to be competitive, and that takes a lot more grid infrastructure than PV that can be much more readily sized for and located near the loads. The combined power + grid-infrastructure costs of distributed PV will end up well below that of wind (sooner than most of us think), which is already beating that of centralized fossil fired generation.

    So, while the US is currently a moderately high-priced PV market in Q4 2015, that's changing at an accelerating pace.

  3. charlie_sullivan | | #3

    I just took a quick look at grid tie solar hardware package prices online in the US vs. Australia. For a 5 kW system, one US price is $7450 . An Australian example is $7390 for 5 kW That sounds the same until you covert to $US: $5318. That's US$1.06 /W, 27% lower than the US kit price. But it looks like most of the difference is in the installation, permitting, etc.

    No implied endorsement or either package or retailer, and I don't know whether they are equivalent quality, etc.

  4. Expert Member
    Dana Dorsett | | #4

    In the US a large chunk of the cost falls under the heading "Customer acquisition", which includes the advertising, marketing, bidding etc. In a mature market where the customers already understand the product those costs come down.

    Another big cost in the US is the multiplicity and lack of consistency in the permitting & inspection process. This too drops dramatically in locations where getting a permit to install & hook up to the grid is easier. There are places in the US where it's dead-easy, and the installed cost is in the $2/watt range, but most places are still north of $3, some by quite a bit.

    Kit prices aren't a good indicator of the installed cost, other than to point out that if it's more expensive than the going rate to have a pro install it the kits are harder to sell. I'm not sure how many of those kits in Australia will be sold at $1.06 USD/watt, given that in more active areas in Australia a pro will do it for less than that. But there are some underserved more rural parts of Australia where it's might be viable. A pro isn't going to spend the drive time to head out to some remote ranch to eyeball quote & install a 5kw system when they can install three systems that size more locally in the same amount of calendar time it would take to quote & finish the project.

  5. Expert Member
    ARMANDO COBO | | #5

    I have two houses in the early stages of construction, both with identical 30 SunPower 327, 9.81kW, 2 SMA inverters, peripheral systems, and yet both have different prices. One is $41K, with rib clip mounts on a 4/12 pitched metal roof and the other is $43K.on a flat roof on racking system.

    So when y’all are talking about for $3.40/watt, I really would like to know what is included in that price, as y’all know, there are huge differences in quality, warranties and efficiency of all PV systems. It feels we are only getting half of the picture. I do know most of the foreign systems in our market I would not consider using just because they are chip.

  6. exeric | | #6

    Wow, it sounds like Australian fossil fuel power generation is in a heap of trouble with it s ultra high fixed costs. I wonder at what point the consequences of the bad decisions of the few will stop being paid by the many. I'm thinking specifically of the utilities thinking of implementing a high cost to DISCONNECT from the grid. That seems like it would be a breaking point to most people and the probable cause of a major revolt by the populace.

    An interesting sidelight is that the bottleneck to disconnecting completely from the grid is the poor current battery technology. They are high cost, have low power density compared to petro fuels, and need to be replaced about every 7 years at major expense. There seems to be progress lately in lithium air batteries. They have power densities 90% of petro fuels and if the problems are ultimately resolved they would be about 10 times better than current lithium ion battery technology. That's because they don't have to carry the oxygen in the battery but instead just extract it from the air.

    It's a big "if" that they will be able to do that. They've overcome many of the initial problems lately.
    But there are still many problems to be overcome, including contamination by carbon dioxide if pure oxygen isn't used in the current stage of development. Some people working on it estimate that it will only take 10 years to work out the kinks. If it doesn't turn out to be a pipedream like fusion energy has been then get ready to rumble!

    See here:

  7. Expert Member
    Dana Dorsett | | #7

    The Australian utilities have been kind whip-sawed by the rapid policy changes between administrations with diametrically opposed directions. The response by most (not all) have been counter-productive toward their longer term interests. It's hard to conceive of a business model that will recover their sunk costs on stranded grid & generation assets that really works. Most will eventually have to write some of that stuff off and are just stalling, milking as much revue as they can on the way down. Utility bankruptcies are not out of the question, and are even anticipated.

    And it's not just in Australia. These movies will be playing at a utility near you at some point in the next decade (in Hawaii right now), as costs for both PV and batteries decline.

    The energy DENSITY of batteries is irrelevant for stationary batteries. Ambri (a startup making scalable molten-metal grid batteries out of cheap stuff: ) has referred to their comparatively high mass / low energy density technology as an "anti-theft feature". Energy density is only relevant when you have to haul that battery around, such as in an electric vehicle. If the battery out back is a couple cubic meters in size and weighs 8 tons and runs at internal temperatures of 500C, it's fine, as long as it has good longevity, sufficient capacity for the application, and reasonable round-trip efficiency. (Abmbri is convinced that PV + Ambri can beat current grid costs in Hawaii right now, based on field testing in Hawaii: )

    Flow batteries have great potential for storing large amounts of grid power on the cheap too, but it's unlikely they'll be commercializing a home-scale version any time soon, and the molten metal batteries are better suited to micro grids than single family homes, so for how lithium ion is the most popular & flexible technology at that scale. There are many mid-sized battery technologies available and suitable for commercial ratepayers looking to minimize demand-charges, where demand charges are assessed.

    The much ballyhooed Tesla PowerWall happens to be lithium ion technology, but the energy density aspect is a "don't-care", since it's just hanging on the wall. There are a handful of competitors working the small-scale PV behind the meter game in Australian and German markets right now, and the PowerWall is being promoted by at least one utility in Vermont right now. Solar City successfully pushed to get regulations changed in CA to allow behind the meter residential batteries with 2-way power flow over a year ago, and has been selling battery + PV in drips & drabs there ever since, but expect to be selling "batteries included" in most of their installations well before 2020. The obstacles aren't technological, they're utility-regulatory, it's all about who is allowed to install what on the grid, and who gets to control it. And if controlled by the utility to provide ancillary services, how much is the battery owner to be compensated for that value?

    There is no need for the "ultimate battery", only a "good enough and cheap enough" battery for grid applications large & small, and they're mostly there already. Batteries + PV are already cheaper than diesel fired backup everywhere, and cheaper than gas-fired peakers in many areas. Unlike fossil fuels, which are a commodity with high volatility in the pricing, batteries & PV are manufacturing technology, both on a logarithmic price decay (called the "learning curve" in business economist circles) against linear increases in production volumes. The key to making them cheaper is deployment, not a technology breakthrough. For small scale PV installation enhancement the current technology already is "good enough". While technology improvements are welcome, they are not necessary for battery technology to disrupt the business-as-20th-century-usual utility models.

    For electric cars too, the current lithium ion technology dominates, and is making incremental technology improvements, but much bigger incremental price improvements. When Tesla's factory hit's full production in 2017 the world price for car-type lithium ion batteries it likely to take quantum step down. A lithium air battery would be great, but the lack thereof isn't going to forestall electrification of the transportation sector, even though a breakthrough on that front might accelerate it.

  8. exeric | | #8

    Hi Dana,
    I knew I was punching above my weight class when I brought the subject of batteries up. Of course, you are right that pure power density doesn't hurt or help batteries in stationary applications. That's because power density is not the same as cost per kW. I was conflating things there. But the reason I was conflating the two is because there is an indirect relationship in cost between two battery technologies that have the same power output but one technology weighs 1/10 the other.

    All else being equal (which really isn't possible in real life) the lighter technology will be substantially cheaper. One can see that in all kinds of anecdotal evidence just by looking around one's environment.. It's a statistical fact and its not close to a direct correlation between weight and price but its definitely there. I wouldn't be surprised if a battery weighing 1/10 another battery putting out the same power wouldn't be priced 1/3 the cost of the heavier battery at similar points in their development. That's just because people would desire the lighter battery more than the heavier and that in itself would lead to natural economies of scale. Of course, a utility or government entity can put their finger on the scale of an inferior technology but that doesn't really negate the point. We were talking everything else being equal. Which they never are.

  9. charlie_sullivan | | #9

    Yes, the specs for an ideal stationary battery are very different from the specs for an ideal battery for EVs or portable electronics. Another company that is making stationary batteries from low-cost materials (like Ambri) is Aquion. I like the concepts of both companies, but Aquion is already shipping whereas I don't think Ambri is yet.

  10. exeric | | #10

    Hi Charlie,
    I just looked into Aquion. Here's an interesting link: They quote a price of $250 per kilowatt-hour. That compares very favorably to Tesla's Powerwall. The smaller Powerwall priced in that Vermont news item costs $6500 for 7 kilowatt-hours. Subtracting $2000 for the SolarEdge inverter included in the price leaves a price of approximately $650 per kilowatt-hour. It seems one is paying heavily for the sizzle behind the Tesla name.

  11. Expert Member
    Dana Dorsett | | #11

    Yep- there is a lot of companies getting into the act, now that California has mandated grid storage, as well as parts of Europe. With volume comes price drops.

    Meanwhile,the Nevada PUC may have just opened up a potential market for behind the meter batteries by retroactively screwing PV owners, eliminating net metering entirely for new PV, ramping down net metering on existing PV to wholesale price for export to the grid over the next 4 years, including third party ownership / PPA model situations! If this stands without revision it's going to be a HUGE legal fight! It goes into effect on January Fools Day!

    Who knew that Nevada was going to become part of Australia? :-)

    It's only a matter of time before the learning curves of PV & batteries cross the retail price thresholds in Nevada, or any other US state.

  12. JC72 | | #12

    Has anyone been able to recycle these batteries (ie. Tesla Powerwall)?

  13. Expert Member
    Dana Dorsett | | #13

    Lithium ion batteries can be recycled- it's a budding industry with multiple companies getting into the game eg:

    I doubt any Tesla PowerWalls have been recycled to date- they're all still new enough to be under warranty.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |