Al Marshall's  Instablog

Al Marshall
Send Message
Individual Investor in stocks as well as an investor and adviser to start-ups. Most of my background is in marketing and product management with some experience as a management consultant and US Navy intelligence officer. MBA and BA degrees from the University of Chicago.
  • Axion Power International PbC Cost Estimating Spreadsheet 20 comments
    May 1, 2013 10:53 AM | about stocks: AXPW

    In an effort to provide Axion Power International (AXPW.ob) investors some diversion from the tedium of waiting for financing news, I've built a spreadsheet with my attempt at an algorithm for estimating the cost of PbC batteries. I won't be able to do that with precision, but hopefully it'll be a useful tool that you can edit to arrive at your own estimates. Generally speaking, I think these numbers are conservative. I fully expect that the real total cost number for a 30HT-sized PbC is less than my $269 estimate.

    John Petersen is kindly hosting the spreadsheet, which you can download at:

    The spreadsheet has a lot of explanatory info in it, including John's comments in yellow and references to discussions in past Axion concentrators.

    The spreadsheet begins with my estimate of the economics of an AGM battery. I started with my assumptions for the selling price, gross margin, and weight/cost of lead to solve for all other costs referred to as "non-lead costs".

    In addition to providing a basis for comparison, I carry the non-lead cost down into the PbC section to provide a baseline PbC cost.

    I add to that baseline the cost of the lead in the PbC. Note, I'm assuming the lead content of a PbC is 70% of that of an AGM not our "traditional" 60% number. John Petersen recommended this change. Additionally, drawing from numbers mentioned in various concentrators, I put together fixed cost allocations for the carbon sheeting and electrode lines using a five year assumption and an annual production volume of 50,000 units.

    The 50k unit volume is much less than our current estimate (90k) of Axion's PbC production capacity. I did this to be conservative because the smaller the production volume number, the higher the per-unit cost allocation and of course the higher the total cost and, presumably, sale price numbers must be. As you can see, this analysis represents an exercise in forward pricing and this number is the key lever for that analysis.

    In addition to the fixed cost allocation for the two lines, I also came up with operating cost assumptions of $150k and $600k for the Carbon Sheeting and Electrode lines respectively. The former is for a single shift while the latter assumes two shifts. I also made the latter much larger because in addition to being more expensive, it also seems to be much more complicated. Still, I acknowledge that there is a lot of WAG there.

    Side note: If you don't want to use forward pricing then use 2,000 units as the production volume assumption and cut the electrode operating cost number to reflect a single shift.

    At the bottom of the spreadsheet I arrive at a total PbC cost and provide a place for the user to enter a price point. The spreadsheet then calculates the gross margin as well as the gross profit assuming 50k units and the current state $8m annual burn rate.

    A $360 price point gets you to the 25% margin that we think is the norm for AGM in the industry today. Axion would still be losing money ($3.4m net loss) if it sold 50k units at that price.

    Eventually, if not sooner, Axion may have higher margins due to its unique and patented technology. For example, if Axion continues at the Norfolk Southern price of $463 then it would have a 42% margin and $1.7m net income on 50,000 units.

    Disclosure: I am long AXPW.

    Additional disclosure: I have a long position in Axion stock but consider it an extremely risky investment not suitable for most people.

    Stocks: AXPW
Back To Al Marshall's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (20)
Track new comments
  • John Petersen
    , contributor
    Comments (30632) | Send Message
    I think you did a fine job on the Instablog and spreadsheet Albert. The numbers are probably not 'right,' but they're certainly within the realm of reason based on what we know about the industry and Axion's cost structure. It's certainly a far better analysis than we frequently see from starry-eyed optimists who start with silliness and then factor in undefined "economies of scale" assumptions.
    1 May 2013, 11:32 AM Reply Like
  • SMaturin
    , contributor
    Comments (2268) | Send Message
    Very elegant, thanks Al!


    I like the comment annotations from you and JP that explain the assumptions.
    1 May 2013, 02:05 PM Reply Like
  • D Lane
    , contributor
    Comments (1708) | Send Message
    Albert, thanks for sharing!
    1 May 2013, 11:55 AM Reply Like
  • H. T. Love
    , contributor
    Comments (19569) | Send Message
    Albert: Great stuff that you did there. The neat thing is that as quarterly reports are posted, the spreadsheet will migrate to greater and greater accuracy as certain assumptions will be replaced by real-world results.


    This, of course, permits greater forecasting accuracy!


    Thanks to you for sharing and John for hosting!


    1 May 2013, 11:58 AM Reply Like
  • D-inv
    , contributor
    Comments (4877) | Send Message
    Strong effort, Al. Thanks for sharing.


    FWIW, I derived a PbC gross profit margin of 35% on 2011-2012 sales using data JP shared in his spreadsheet. That derivation depended on a "legacy" battery revenue volume of $100K - $150K IIRC as well as sales and costs observed pre-continuous roll carbon sheeting process.
    1 May 2013, 12:39 PM Reply Like
  • Al Marshall
    , contributor
    Comments (633) | Send Message
    Author’s reply » D-inv: Very good point although I'm not sure what it means.


    What margin were you assuming for the contract business and the legacy battery business that led to the 35% number for PbC? When I tried to do that analysis I think I came up with a similar number.


    Back to my question. If that is the right number what would that indicate about PbC pricing and margin targets post carbon-sheeting? I'm not sure.


    One thing, when you are selling batteries for test purposes, the information derived from the margin would seem to be less important. Margins are high because customers don't really care that much about a handful of batteries or the price reflects expected support requirements? Margins are low because management wants to encourage testing and offer hope for reasonable mass production pricing?


    I'm most interested in getting a sense of how aggressive Axion will be in terms of forward pricing. On the concentrator, I've expressed concern that the dearth of sales indicates Axion isn't being aggressive enough.
    1 May 2013, 02:47 PM Reply Like
  • D-inv
    , contributor
    Comments (4877) | Send Message
    I'm thinking that since earlier estimates were derived on much lower sales volume and higher production cost, there is ample latitude for your 42% profit margin on lower sales price is in the realm of possibility.


    I have yet to delve into your spreadsheet and look forward to doing so.
    1 May 2013, 04:26 PM Reply Like
  • Pztrick44
    , contributor
    Comments (80) | Send Message
    Very nice.


    After pulling out the non-cash items (amortization of electrode, carbon sheeting lines) as they've been paid for, the break-even cash flow PbC price appears to be ~$410.59. i.e. the price point to keep us alive at 50,000 units, though at an "accounting loss".


    Am I correct to assume the $8,000,000 in D59 ("Net income/loss at 50k units") is your estimate of Axion annual G&A expenses/overhead? Just to be clear, does your AGM sale price NOT INCLUDE G&A costs for conventional AGM vendors? (e.g. "Non-lead cost") If that is the actual retail sticker price, and AGM companies are profitable (i.e. selling AGM batteries at $225 and covering their overhead), shouldn't our non-lead cost ALREADY reflect the assumed G&A/overhead of $8,000,000? And perhaps that doesn't need to be included in cell D59? In other words, I think "Non-lead AGM Costs" D47 may be double counting with the -8,000,000 in D59 -- if my assumptions/comprehension of your sheet is correct.
    1 May 2013, 02:30 PM Reply Like
  • Al Marshall
    , contributor
    Comments (633) | Send Message
    Author’s reply » Pz: Roughly speaking, Axion is burning/losing about $2m per quarter so that's not specifically an overhead figure. What I tried to do was overlay incremental PbC sales on the current business. Not technically correct, but I'm hoping that the 20% of the work I've done get's us to 80% accuracy.


    I'm not an accountant but the COGS estimate for standard AGM doesn't include operating expenses (Enersys being the strawman) which I believe includes G&A. The other AGM costs represents manufacturing costs and materials, which with the exception of the electrode, should be very similar to PbC.


    Does that answer your question?
    1 May 2013, 06:12 PM Reply Like
  • Pztrick44
    , contributor
    Comments (80) | Send Message
    Yeah, it wasn't clear on my first read if the AGM price was COGS or retail. COGS makes perfect sense :) Thanks
    1 May 2013, 08:44 PM Reply Like
  • Tampa Ted
    , contributor
    Comments (2652) | Send Message
    ap - thanks for the spreadsheet. It appears to be a very useful tool for running different scenarios.
    1 May 2013, 08:09 PM Reply Like
  • amishelvis
    , contributor
    Comments (143) | Send Message
    Hi Al,, Thank you , and nice work. My only thoughts are,,, that the $30 per hr is too high. I am thinking $14. plus 8-10 in bennies,,lets round up to $25. This particular area is very short of jobs..and this is a real job!
    Other than that,, hope all is well. matt
    2 May 2013, 07:03 AM Reply Like
  • Al Marshall
    , contributor
    Comments (633) | Send Message
    Author’s reply » Matt: Let's discuss at the annual meeting this summer. As far as the direct labor cost, I think the spreadsheet shows the impact of the 95% reduction enabled by the automated carbon sheeting process. If we go with $25/hour as the direct labor hourly cost that lowers the total cost of a battery by $.50.


    Now, the biggest cost factor appears to be that corrosion barrier although it is also arguably the cost factor that we have the least capability of estimating. I'm hoping that my wag is an over-estimate. Maybe there's a conference call question that could be formulated that would elicit a sufficient morsel of information from Tom Granville to narrow down that number a bit.
    3 May 2013, 03:42 PM Reply Like
  • John Petersen
    , contributor
    Comments (30632) | Send Message
    I'll bet a nickel that Tom won't get anywhere close to disclosing per unit costs for any component.
    3 May 2013, 04:10 PM Reply Like
  • nakedjaybird
    , contributor
    Comments (2853) | Send Message
    apm - some of us are familiar with (obsolete/traditional) measurements of product cost analysis which you may want to consider in your presentation of facts:


    1) For each separate cost element, indicate it's % of total.


    2) Consider a cost breakdown for Direct Product Cost and Indirect Product Cost, for a Total Product Cost analysis which includes Labor, Materials and Expense (using the appropriate elements of each category as you wish for lead, active materials, electrolyte, carbon, separators, case, power, wipes, electrode processing, purchased items, etc,).


    3) Then consider the rest of the cost elements for SG&A, shipping, warranty, and overheads maybe with or without cost of capital as that may depend on many options (grassroots factory for PbC lines as well as Pb, or PbC adder lines, and of course per PbC adder line in the case of multiples or just one), etc.


    Don't know if you have enough data for this, including a guess at selling price which could be somewhat independent of product cost as experienced in today's market place of true and blue Pb acid competition? (Harvard U or Madison Ave marketing: whatever the market will stand - which is the new way of doing things as compared to the traditional cost-based pricing used to determine if one could actually competitively enter a market and make a profit (and how much) or loss based on best estimates of bottoms-up product cost estimating and a feel for the profitability risk based costs buildup, instead of setting a hopeful selling price and determining how low costs have to be to make a profit (of course it never hurts to make these calculations from both directions as an Einstein evaluation (how wrong could we be?).
    4 May 2013, 11:33 AM Reply Like
  • Al Marshall
    , contributor
    Comments (633) | Send Message
    Author’s reply » NJB: Yes, I could add the percentages and breakdowns that you suggest but I'll admit that it's been twenty years since I took cost accounting in B-school and I probably wouldn't get the breakdowns right.


    I have made a few minor changes and anticipate making more as we learn more and will on occasion post an updated spreadsheet. I can also add additional tabs with other folks' versions since a big part of my purpose in building the spreadsheet and sharing it was so that others would pick it apart and build upon it.


    However, I believe the level of detail I used was the right amount given the information that I have at my disposal. I just don't have that more detailed information you mention. I would also argue that a greater breakdown on AGM battery production and the elements common with PbC is unnecessary since we know that they are common. Sort of like an algebra equation, we can cancel out like terms. That commonality is what enabled me to even attempt this analysis.


    I also agree that absolutely, Axion should use cost plus, competitive, and market based methods to identify potential price points and I do agree that maybe I should look at these other methods as well.


    I actually think Tom Granville is using competitive pricing. After Vani Dantam was hired, the company seemed to go on the offensive against lithium-ion, pointing out the problems with LI and how the PbC was better. Previously, I think Axion had focused more on comparing the PbC to standard lead-acid and AGM.


    Subsequent to the start of that offensive it would seem that coincidentally, Lithium Ion technology has taken a lot of hits, particularly in terms of its safety for use in cars and airplanes. If I'm right about TG's strategy, you'd think these recent events would only serve to increase his confidence that the PbC is superior to Lithium ion and thus should be priced at a premium to LI.


    I can understand that approach. I don't agree with it because I think market-based pricing is more appropriate because:
    A. The applications for which PbC is superior to LI are not yet economically meaningful while the shares TG has to issue to stay alive waiting for these applications to become meaningful are very economically important.
    B. High price points that suck too much of the economic gain from a given application will greatly slow the evolution of that application.
    C. Lithium Ion is still the technology that the government has designated as the winning technology. It's going to take a lot more bad publicity related to safety and failed government grants and loans to turn that battleship.


    In summary, I think the carbon sheeting break-though should cause TG to change his thinking about pricing. Instead of competitive pricing, he should set low prices for early adopters to create extremely favorable economics in key applications because this will most rapidly create markets for the PbC.
    5 May 2013, 01:38 AM Reply Like
  • nakedjaybird
    , contributor
    Comments (2853) | Send Message
    apm - based on your various comments about the competition and market place (vis battery cost) maybe you should put your time into building a sales/profitability/stock price model for the various applications where PbC will win in loco's, OTR trucking, hybrids, remote and off-grid power, near-grid regulation, etc., etc. etc.


    Now that has a more promising future with more reward than taking the PbC costs to some lower number which only makes for more profitability and maybe more markets, while understanding fully that the product cost nor product price matters as much as what can be done that nothing else can do; and that is PbC's future: doing what none other can do (as effectively, as long, and as (yes) cheaply).


    I do not believe that product cost is the opportunity; but we should know it. Well, at least TG should know it, and I'm sure he does and continues to work on it. But, to my knowledge, that end of the equation is pretty well knocked. The market and sales end ...... that's limitless, so to speak.
    5 May 2013, 11:23 AM Reply Like
  • Al Marshall
    , contributor
    Comments (633) | Send Message
    Author’s reply » I agree that setting up a simple model for the economics of the various applications would be feasible and helpful. I'll do that as my time and the available information permits.
    5 May 2013, 03:30 PM Reply Like
  • Amouna
    , contributor
    Comments (2003) | Send Message
    I just went through your spreadsheet and I really appreciate the effort. It helps breakdown the overall costs into subcategories while shedding the light on each and every process of the calculation


    Thank you!
    5 May 2013, 07:55 AM Reply Like
  • Stilldazed
    , contributor
    Comments (4120) | Send Message
    Hi AP,
    Wow, what a lot of work. Being math challenged myself I can appreciate all the effort and time devoted to this project. Thanks.
    5 May 2013, 11:13 AM Reply Like
Full index of posts »
Latest Followers


More »

Latest Comments

Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.