A Well-Designed Feed-In Tariff Can Drive Renewables in California

This FiT should be more end-user focused. There should be a <20 kW tier for residential (an essential tier), a 20 to 100 kW tier for small businesses, 100 to 500 kW, etc. Each tier has slightly different scale economies and associated profit potentials that you’d want to tune. You’d also want to tune the incentives by benefits produced. Why does this program fail to do this? This seems so obvious after the success rooftop programs in the two largest PV markets on the planet. Everything that is said about the Merit Order Effect, Substitution Effect and Locational Benefits are amplified at the rooftop level and they should be valued accordingly. You also have the potential for some price sensitive demand responsiveness at the end-user end which generation only sites can’t offer.

-Consideration should be given to some sort of experimental self-consumption incentive - something that provides a cost/benefit trade-off. It looks like the FiT they propose is under the retail rate so you already have a self-consumption incentive as long as you aren’t forced to sell at the FiT and purchase at the retail rate.

-There should be no yearly cap. You’ll get a fits and starts market from that sort of setup.

-Revisit rates every two years? I think that’s a poorly thought out plan. The ratepayers would get fleeced. You’d get another Spain2008 or Germany2010 if the FiT gets set too high. Put in capacity triggered degressions that have some discretionary latitude. For example, for each 1 GW installed the FiT would go down by 10% (+/- 2% by CPUC vote). You’d want the degressions to be tied into the tiers so that the MW sized sites don’t get all the incentive at the expense of the smaller systems. 

-“The policy should address interconnection barriers by requiring utilities to publish capacity information and upgrade plans for distribution circuits.” This is an excellent requirement. It ties into placing a value on PV if its deployment can delay or offset distribution circuit upgrades and visa versa.

-A FiT will blow right passed the RPS requirement. No need to tie the FiT to an RPS in any way. The article talks about 4 GW per year as though that’s a big deal? It’s not – at all. California’s urban sprawl is a perfect canvas for PV. Germany’s expected 8 GWs in 2010 will easily be bested by a well running market in California. Where’s the ambition? Where’s the attitude?

CORRECTION: A capacity triggered degression would be more like 2.5% (+/- .5% by CPUC vote).

Agrred - every x year price revisions are sloppy thinking, lead to expensive policies, and have a 100% failure rate in the real world so far (price overruns resulting in an unscheduled adjustment, or < 1 day program availability.)  Capacity - triggered degression is the only smart way to go.

And - no *yearly* cap, but a *program* cap tied to those digressions is equally key.  Programs with ” no cap” get scored high by some, but it’s unrealistic to expect a public subsidy to have no cap - all it means is you never know when the bill will get too high, and the program get pulled out from under you.

Finally - self consumption incentive (like, say, net metering plus 20 year PBI?  Sound familiar?)  Decreases ratepayer impact, as the net metered part of savings would not be subject to income tax.

This is just the sort of innovation that will move California even one step closer to balancing out the impacts of its huge population and help maintain our environment and economy.
-California Solar Engineering
http://www.calsolareng.com

JoeJoe, I am a co-author of this bill and wanted to respond to some of your comments. Here’s the rationale for the support in the bill for 20 MW and below solar: this scale represents a “programmatic gap” in California’s set of policies. Small-scale solar already gets a lot of support from state rebates (CSI) and federal tax credits. Larger-scale solar is less expensive due to economies of scale (at least for generation costs independent of transmission costs) and also receives federal tax credits and depreciation. Medium-scale (wholesale distributed generation) solar does receive state rebates. Power is sold directly to the grid, so to make up for the lack of a state rebate, we are seeking a slightly higher payment for power received. But this payment is still lower than the value to ratepayers from the solar power received on the grid, as demonstrated by a recent detailed report for CalSEIA.

REESA is in no way designed to exclude smaller systems. It’s a question of economics. If the CPUC sets prices high enough to justify smaller systems being built as feed-in tariff projects we are fully supportive.

It’s important, however, to distinguish between net-metered systems, which receive state rebates and credit for excess power at the retail rate (a type of subsidy) and feed-in tariff projects, which don’t get a state rebate and don’t get paid at the retail rate.

So your statement above mixes apples and oranges. Net-metering is very different than direct sale feed-in tariff arrangements. They can be mixed in a hybrid policy, but it gets complex and may not be worth the trouble.

There’s a typo in my comment above: I meant to write “Medium-scale (wholesale distributed generation) solar does NOT receive state rebates.”

Tam

No worries about the typo. I’m bad about this myself. Some damn good jokes and witty remarks have been ruined in the process. But as the French say… oh well.

It breaks down like this. We don’t need to add anymore subsidies to the subsidy stack. Very soon now we’ll need to start the process of dismantling the existing subsidies and filling their place with smarter policies, better value judgements and competition.

The FiT Coalition literature mentions many good things about PV and points out some areas for improvement as well. Things I like include:

1.  The paperwork problem needs to be addressed directly and brought under control. 
2.  PV drives wholesale prices down through the merit order effect.
3.  PV substitutes for nat gas. This lowers nat gas prices, removes pollution and mitigates price fluctuations.
4.  We need to critically examine how PV displaces transmission and distribution costs
5.  We need to incorporate points 2, 3, 4 in our Value analysis of PV.

The two recurring statements I don’t like are:

1.  “RDG does not drive volume, nor does it satisfy RPS requirements”
2.    “WDG market size100 times larger than RDG market size”

These last two statements are more than just wrong – they’re deviously wrong. There’s no excuse for this sort of junk from people who so keenly see the current state of affairs. I could suppose there’s some misplaced passion at work – maybe, but that doesn’t make sense. I could suppose you have a financial angle, a project development business for example – you might, but this still doesn’t justify these statements either. This makes my head ring… but I digress… On to the argument….

The statistics out of Germany are clear. Small scale PV scales up nicely – their market doubled in 2009 and it will probably double again in 2010. Here’s a graph that shows the distribution of system sizes in Germany for the last 15 months of growth.

http://www.polderpv.nl/PV_weltmeister_2010_prequel.htm

From eyeballing the graph, I’d say it looks like around 90% of the systems (by cumulative MW) are under 1 MW. This data indicates to me that sub-MW systems can drive serious volume.

As far as the second part of the first statement goes, I think I’ve read you say it’s goofy that tradable RECs qualify for the RPS credit but internal RDG generation doesn’t. If we’re going to give extra credit to renewable energy production, we should give it to rooftop PV. Let the utilities keep this credit, whatever, I don’t care. The FiT Coalition shouldn’t be pointing out that RDG doesn’t contribute to the RPS requirement. You should be pointing out that RDG should contribute to the RPS requirement.

We can actually go further and recognize that the RPS requirement is irrelevant. By irrelevant I mean that a healthy PV market will blow right past the 33% RPS without a problem. It won’t need an arbitrarily legislated goal to do this. This goal is irrelevant because it is easily surpass-able. I look at it as another layer of well intentioned paperwork to deal with.

I don’t know where to start with the “100 times larger” statement. If find it ludicrous. If the Californian market for RDG was 1 GW you’d have a WDG market of 100 GW. That’s twice as large as California’s peak load. Does that make any sense? I don’t think it does.

On a technical level, RDG and WDG can both provide a substantial percentage of our residential, commercial and agricultural energy needs - probably over a fifth of total consumption in these areas before any dedicated storage is required. But this technical potentials will not be unlocked unless the economics justify doing so.

Continued on next slide…

When I look at the economics of RDG vs. WDG here are the basic arguments that come to mind.

1. WDG competes against the wholesale rate. This price is currently averaging around 5 cents/kWh in California.

2. End-user PV competes against the retail rate. The baseline retail rate for PG&E customers is only 12 cents/kWh but it’s not too hard to jump up to the 30 cents/kWh range. In PG&E’s Baseline Territory X near you, over 40% of customers regularly go over the 130% of baseline usage where they pay 29 cents/kWh. Over 20% of the customers in Territory X go over 200% of baseline usage and pay 40 cents/kWh.

3. When you look at project IRRs, WDG clearly has the advantage of lower installed costs but the higher costs of capital and land costs tend to cancel out some of this advantage. When I consider the similar LEC values between RDG and WDG along with points 1 & 2 above I conclude that the profitability and by extension “market size” of RDG is actually much larger than WDG.

I agree it’s a question of economics but I think RDG blows WDG away in this department. End-user PV has the natural inside track – I don’t see any way for WDG to counter this advantage. We still have things to work out with transitioning away from net-metering but I think these issues are solvable. A first step would be to figure out how to properly value excess PV from end-user systems and then move to a more competitive arrangement where excess feed is paid a market based PPA based on this value. Everything has a legitimate price signal at that point. A second step is for FERC and the PUCs to work out the jurisdictional issues that a large deployment of end-user PV raise. A third step would be addressing the stranded cost issues that PV raises. Maybe you see other issues that need to be addressed but so far I haven’t’ found any deal breakers.

My feeling is that the FiT Coalition is crowding the field with a bad idea. I think this should be easy enough to see. If it’s money you’re after, there’s plenty to be made with end-user PV. If the environment is your concern, there’s plenty to be saved with end-user PV. End-user PV is simply a better strategy. It’s more competitive and inherently more sustainable. It’s more, better, faster Tam. You should see this.

California wholesale rates

http://www.eia.doe.gov/cneaf/electricity/wholesale/wholesale.html

PG&E retail rates

http://www.pge.com/tariffs/tm2/pdf/ELEC_SCHEDS_E-1.pdf

Some good quotes from Adam Browning are in the following link. Why the Vote Solar guys support front loaded subsidies is beyond me.

http://earth2tech.com/2009/06/18/why-california-doesnt-have-a-german-style-feed-in-tariff/

I should not have referenced the Average Wholesale Electricty prices. My bad… Rookie mistake. My Spidey sense was tiggling too. Oh well.

The time-weighted wholesale reference price should be between 10 and 15 cents/kWh.

http://www.ucei.berkeley.edu/PDF/csemwp176.pdf

And BTW this entire concept of having a Commission-set price is broken.  Look at LBNL’s “tracking the sun” report on prices of PV - the “bell curve” of PV prices per Watt at any system size is huge.  The concept of an adminsitratively-established “average” or “fair” price for a PV system makes no sense - you’re picking one intersection on a price curve that varies by $2 or so.  There are practically guaranteed to be a non-finite number of projects that can come in cheaper - they all get a windfall, and the ratepayers take the price.

And since all the stakeholders in the process have poor information, the price has always historically been set too high.  That’s why so many mfgs push for these administratively-set FITs; not because it’s good policy, but because it lets them reap fat profits from lazy price-setting an uncompetitive market.  A fatal administrative FIT price setting error that’s been successfully concealed in some places by multi-billion-dollar annual spending, and has been shown in immediate program shutdowns everywhere else.

It’s thinking inherited from old-school utility ratemaking, where you had one plant that you could establish actual costs for - not a portfolio of hundreds of potential plants you’re throwing a dart at.

You need either MW-based degression (for small systems that cannot bear transactional costs) or an action model (for systems > 500 kW that certainly can) to get an *actual* price - not the bloated, uncompetitive, unhoned prices that we’ve seen in wholesale FITs so far. 

And as to the industry whining about the transactional costs of participating in an auction process - below 250 kW?  I’m listening. Let’s set your incentive with something less complex than an auction.  But if you can’t manage an RFP pricing process for a $1.5 million project, you need to get out of the business.

Second - “System prices should get a lot more competitive (i.e. tighter) as the market matures and you clean out all the front loaded subsidies.”

Theoretically.  And inf act go look at the bell curves in “Tracking the Sun” - the bell curves *do* get tighter each year.  But only a little - systems still are a bit different for each other. they never get to a vertical line as you’re pretending they do with an admin price, and they never stay in the same place.

Your hypothetical scenario sounds like exactly like what I’m pushing for, and is close to what they came to finally in Oregon (though with an overly high initial incentive.) When I say “net metering” especially for commercial, I mean “usage offset”, I think export is a chimera whose economics are never going to make much sense.

The “generic PPA contract” you talk about seems like an attempt to have an administratively set price but have it set to be for best-case systems, not average ones.  (e.g. to shoot for a lower part of the stillw-ide curve, but still trying to get the price exactly right.)  You, the administrator, will still have bad, old, padded information compared to what the market has - and will be able to adjust the price too slowly when big factors in the market change.  Why cling to an admin-set price, (which has no discernible upside and a long history of failure) when capacity-based degressions provide equivalent transparency and predictability and have worked elsewhere?  Germany did it, Spain did it, many people end up there after a bruising experience with trying to have an enlightened administrator get the Magic Price on the Magic Day…