9/10/98
Hi, FMK & Robert,
Back on August 29, FMK (#4000) asked for my "balanced opinion" in rating Valence's chances for success. Well, here you have it: my long-winded explanation of my subjective probabilities (4027 & 4031) based on what we all pretty much know. Sorry this has been delayed; I've been busy. Worse, it seems dated because of Archibald's sudden departure. His unexplained departure at this time must be considered a strong negative. Accordingly, my new, revised subjective probabilities raise the likelihood of failure to 40%, and hence reduce the likelihoods of decent, good, and great success to 24, 24, and 12% respectively. (This reduces the E(v) to $34.20; see below.) I still hope the following can be useful as a means for helping organize our thinking about the myriad circumstances surrounding Valence's possible future. After considering my numbers and arguments, I hope you and others will tear them apart and correct them in your own analyses which you then post.
**********
This post is my "balanced opinion" as to how I rate Valence's chances for success. I will discuss its chances for survival over the next few months and its chances for success over the next several years. I want to bring into our general discussion of Valence all of the factors which might affect its prospects within a reasonable investment time-frame. No estimates of Valence's chances for success (beyond mere near-term survival) can be made by treating it in a vacuum. If it becomes a production company, it will exist in a market with other players -- customers and competitors -- and many outcomes are possible. I will also distinguish between success for Valence as a company and success for Valence's current shareholders. In the following, I will present a crude framework for thinking about these matters.
Since "success" and "failure" are awfully broad, coarse terms, I think it is useful to refine them a bit by making some distinctions. "Failure" could take a catastrophic form where new problems on the way to production and contracts doom the company to bankruptcy and complete loss for us current shareholders. Failure strikes the company and shareholders alike. Alternatively, "failure" could be of a lessor form where slow progress causes it to miss the financing progress points, leading to a collapsed stock price and massive dilution. With new financing, the company could finally succeed and benefit future shareholders, but we current shareholders would still consider it a failure. Failure strikes the shareholders but not the company. "Success," on the other hand, would benefit both the company and the shareholders, but success and its benefits could be anything from minuscule to immense. We shareholders would welcome big success, perhaps be satisfied with moderate success, and probably walk away if we thought success would only be small. At this point, we're taking much too much risk to be happy with a small reward.
Posed this way, the problem is to try to gauge the likelihood of these different possibilities in light of the circumstances facing Valence now. Circumstances have changed materially since, say, May, so it is useful to review the basic Valence story to help us to decide whether to increase, decrease, or hold our current positions.
To get a handle on the complexities of all the different possibilities, I will simplify the matter by considering the 4 most basic alternative outcomes: failure, "decent" (small) success, "good" (moderate) success, and "great" success. The three distinct success alternatives will allow focus on the assets Valence will have against the differing amounts of competition it will plausibly face depending on when it begins production.
I will begin by discussing the possibility of failure. Then I will lay out the three success scenarios. Afterward, I will discuss the background factors which differentiate them. Using these background factors, I will explain my subjective assignment of their probabilities as a way of summarizing my take on their relative likelihoods. Finally, for our general amusement I will calculate the "expected value" they imply. There is always a place for misplaced precision in these things.
**********
Of most interest to us all is the likely fate of Valence over the next five months because of the terms of the recent financial rescue package. As I read all information to date (the confirmable facts, the alleged facts, the rumors, the reasoned speculation, the press stories, and so forth), it seems to me there is perhaps a 3 in 4 chance Valence will succeed in the sense of meeting the requirements of its finance agreements. This may be optimistic, but then I'm an optimist.
I give Valence's likelihood of avoiding failure such a high subjective probability because I perceive Valence to be making satisfactory progress on the timetable for production laid out by Dawson, and to have sufficient cash (now and promised) to achieve production and to survive until revenues arrive. If the measure of progress is (re)started with the March (?) conference call when Dawson focused on the goo problem, then it seems to me that the news since then (the conference calls and press releases, as opposed to the disclaimers in the various SEC documents) indicates that as of now progress is on course. The goo problem is solved, various production equipment is in place and is being debugged and integrated, staffs are being hired (and presumably trained), the product is (evidently) finalized, potential customers are being serviced, and so forth. While Dawson will naturally choose the positive spin of describing the glass as half full rather than half empty, I believe he has every incentive to avoid outright misleading statements to the public. I also believe the progress dates laid out in the financing progress were thought reasonable by all parties to the contract. My "balanced opinion" is that the weight of evidence favors Valence being on track to success within the next five months.
On the other side of the matter, I see a 1 in 4 chance that Valence simply won't make it within the 5-month time frame from the standpoint of current shareholders. There is always the possibility that tricky unforeseen technical problems will arise to delay production or make bad yields or unacceptable product. There is always the possibility that purchase orders will not be forthcoming, whether because of extended testing, or unacceptable product or delivery schedule or price, or changes in customer plans or the emergence of a preferred competitor or whatever. There is always the possibility that the July 28 financing package will be canceled if only because it is conditioned explicitly on Lev Dawson remaining on the job. Other less obvious and probably less likely problems cannot be dismissed entirely: lawsuits, equipment fires, etc. In brief, if anything goes wrong in this time period, for current shareholders Valence will be in deep trouble. All the prospectus-type caveats of the S3 and the 10K's and Q's apply. We longs are speculating not just in a risky stock; we are speculating that from now on absolutely nothing will go wrong for the company in what has already proven to be a very difficult setting.
As a long, I personally have every available body part crossed, and I am burning incense. I implore all other longs to do likewise, or do whatever permitted by their anatomy and religion.
**********
Of less immediate importance, but (hopefully) more significant in the long run are the questions of the degree of success and who benefits. For now, mere survival is "success," but all that means is that Valence can proceed into the future as a going firm with production and customers. But mere survival is no guarantee of the ultimate success we all hope for. We hope Valence will emerge as the preeminent supplier of leading-edge batteries in a large and rapidly growing market. To get a sense of the likelihood of the form of success we hope for, we must partition the 75% likelihood of success into the three basic forms success could take. I will begin by considering great success. Then I will progressively degrade the conditions behind great success to consider good and decent success. What we find along the way will be the basis for partitioning the 75% probability.
To make the three forms of success more concrete and relevant to Valence, I will illustrate them with numbers (fortified by a whole hatful of assumptions intended to keep things manageable). The numbers are only roughly correct, but are in the general ballpark. The only outrageous assumption is that battery price will be unaffected regardless of the degree of competition. The numbers can be played with, and no doubt improved (or even made dramatic with more heroic assumptions -- e.g. higher product prices, margins, and P/E's), but doing so would not alter the differences in their background circumstances, which is our focus.
Scenario 1: Valence has "great" success as a company. Suppose Valence enters production and finds itself essentially the sole producer because, say, it has critical patents which lock out most others. To keep things simple, disregard profits from licensing possibilities, cell phone batteries, and so forth. Such complications can always be added later if wanted. I will also assume laptop demand is fixed (ugh ugh ugh). Suppose that Valence produces only laptop batteries, that the battery market is for 15 million annually, that each will bring $75 revenue, and that net margins are 20%. If Valence has 90% of the market, its annual revenue would be $1.013b and its profit would be $203m. With a P/E of 20, the market cap would be $4.05b; at 3 times annual sales, the market cap would be $3.04b. If there are 30 million shares outstanding, the share price would be $135 and $101 respectively. (A 4x sales multiple is roughly equivalent to the 20 P/E.) Of course in recent times both multiples have been far more generous in the hi-tech market where profits and rapid growth have been highly valued, but the market right now seems to be getting much less generous. A stock price between $100 and $135 seems appropriately conservative in light of emerging stock market and world economic conditions.
Scenario 2: Valence has "good" success as a company. Suppose Valence enters production and finds itself the dominant player among several other competitors. Suppose the same background numbers, except that now it gets only 50% of the market. Now its revenue would be $562m, its profit $113m, its market cap between $2.25 and $1.69b, and its share price between $75 and $56.
Scenario 3: Valence has "decent" success as a company. Suppose Valence enters production and finds itself merely a player among many. After all, everybody and his brother is supposedly trying to get into this lucrative new market and some have been working long and hard and/or have a lot of money behind them. Again, suppose the same background numbers, except that it gets only 10% of the market. That translates into $112m revenue, $22m profit, market cap between $450 and $337m, and a share price between $15 and $11.
**********
Now, in thinking about FMK's question about rating the company's chances for success, we have to try to decide which of these outcomes is likely to emerge. (Obviously, I am restricting myself to a short time frame -- three years, say. I ain't about to get sucked into trying to "forecast" the more distant future. I'm in deep enough already.) It seems to me that the outcome will be a function of Valence's assets and the competitive environment in which it finds itself. In brief, as more and more time slips by before production, Valence will find itself in an increasingly difficult competitive environment with decreasingly valuable assets with which to compete.
To achieve and hold onto great success, it seems to me a number of things must occur for Valence.
First, Valence must get into the laptop market, as a poster has said, "fustus with the mostus." It must establish dominant market position early to start building long-term relations with customers before competitors get to them. To succeed in this, it must be able to supply large amounts of high quality product at attractive prices from the beginning.
Second, Valence must be able to defend its position with patents, sufficient investment in plant to meet increasing demand (ie, to avoid putting customers on allocation or being forced to turn down orders), timely new or improved products, declining unit costs from improving production techniques and economies of scale, and declining prices. It wants its customers to have no reason to abandon it, and it wants to discourage potential competitors by raising the likelihood that any investment they make will prove uneconomic.
Third, Valence must extend its product line outside the laptop battery realm to prevent competitors from using financial strength and production experience gained elsewhere to encroach on Valence's laptop turf. Although presumably margins are best in the laptop market, grabbing significant share in other markets should not only prove profitable, but the profits would be denied to the competition. Even a minor share, however, should help keep Valence better abreast of what other suppliers are doing and what their customers want.
Fourth, it must have ready access to the resources it will need for expansion: management talent, skilled labor, and (cheap) financial capital.
Fifth and most important, it must be lucky. Luckier that its competition.
Many things, then, must fall into place if Valence is to achieve and maintain near-monopoly status within the laptop market and a stock price north of $100, and doubtlessly Valence has done much to get most of these things to happen. By all appearances it is putting into place a large production capacity for both laptop and cell phone batteries, and it seems capable of expanding its production capacity quickly by replicating the machinery. This large initial production capacity is perhaps a little audacious and certainly more than a little risky, but if production can be started well before its competition, it will pay off big in lower unit costs, attractive prices, and large revenues. It seems intent on making sure it can supply large contracts from the beginning despite the pressure this puts on its current cash balances. Likewise, it has a large and growing patent portfolio. I have no way of gauging the strength of the individual patents, but I have to think their sheer quantity will prove a deterrent to Bellcore-based competitors if only because some of the patents will have to be designed around at the cost of time and money (unless licensed from Valence). The patent portfolio and Dawson's recent comments also indicate Valence is making good progress toward improving its product. Although costly, these are potentially valuable assets.
While much of Valence's behavior to date, then, promises to attain our "great" outcome, still that outcome is in serious doubt because Valence has lost much time and may not get there "fustus with the mostest" in any significant sense. For getting there "fustus with the mostest" to really succeed, it must get there far enough ahead of the competition to have time to get itself deeply established as the primary supplier. Arriving a few months ahead of another competitor also having substantial production capacity does not guarantee a 90% market share. First to production is at best a milestone; it is not the end of the race.
Competition is inevitable. The potential market is considered huge by everyone. The li-poly product threatens, sooner or later and with varying severity, existing rechargeable products all the way from li-ion to lead-acid. Apparently even the primary battery makers are looking over their shoulders. Some will want into this lucrative new market; others will want to protect their existing markets. Some years ago Bellcore licensed its technology to some dozen companies including Valence, and it is reasonable to expect at least some of them have been quietly making progress toward production. Various news stories from around the world posted on this thread suggest as much. Of course, some potential competitors have already been making noises: ULBI has been hand-assembling cells and promises automated production soon, Toshiba (?) seems about to produce a cell phone using its own li-poly battery, and LITH claims to have its own technology which does not depend on the Bellcore process and which will be in production soon.
While much of this noise (and the press stories) reminds me of "vaporware" announcements, still it is reasonable to think that where there is a lot of smoke there will be some fire. The only consolation to me is that the problems plaguing Valence are likely to plague the competitors too. That doesn't mean, however, that they will all be equally slow as Valence in solving the problems. I have to allow that someone could be more clever, or luckier, or better financed. I have to think that the months and even years of delay in getting into production has given competitors a huge opportunity to get into the market soon after Valence, if not before. Certainly I have to think Valence's window of opportunity to thrive in the market before the arrival of competition has shrunk considerably. More than likely, delays have considerably diminished Valence's chances for great success.
While the rising prospects of early competition is discouraging for the possibility of great success, if Valence does not fail, or suffer further delays that are tantamount to failure for us longs, then it is possible Valence will be early enough to be among just a few competitors. Everybody and his brother may want to get into the market, but Valence has demonstrated that automated production of the damn battery is not easy, and ULBI has demonstrated production by hand is not economic. If Valence can get into production by, say, the end of the first quarter of 1999, these problems should keep early competition down to a handful. In this case, Valence will still bring some substantial assets to the competitive fray with the promise of good success. Regardless of what the competition does, Valence would have a finalized product, large-scale production capability, patents, product improvements in view, etc. Its capacity to produce large amounts early on, and to ramp up further production capacity quickly to capture new customers as li-poly becomes "de rigeur," might indeed leave it the dominant supplier with 50% of the laptop market. In time, it's patent portfolio might provide licensing revenues, revenues which might compensate for price weakness due to competition (and thereby permit my fiction that prices are $75 per unit regardless of the amount of competition).
On the other hand, delays which carry Valence to next summer or beyond (and serious if not massive dilution for us shareholders) will provide another year for the competitors to get their own production under way. In this case, some of the "assets" Valence would bring face serious decay in value. It's large scale production might not find sufficient customers at the prices presumed, it may not be able to expand production enough to gain further economies of scale, its improved product may face similar improvements by others, and so forth. It's patent portfolio might be its only remaining advantage over its competitors, and time erodes even this advantage by giving the competitors more opportunity to design around the relevant patents. Licensing revenues could be almost anything. (Hopefully, however, patent-based licensing revenues would still be sufficient to preserve my fiction of a $75 price unaffected by competition.) A 10% market share may be its fate.
**********
Given these four general scenarios -- failure and 3 distinct levels of competition -- my problem as an investor is to try to weigh the likelihood of each. Doing so produces "expected values" for the various scenarios, and hence allows me to derive a very crude "expected value" for the investment. If the expected value is very high relative to the stock price, then I might invest; if it is low, then I would avoid the investment; if the probability of losing the investment is high, I may not invest regardless of the expected value. (All this has direct analogies within the lottery world. Here in Virginia, the expected value of a lottery number is sometimes many times the $1 cost of entering, but the probabilities of losing are extremely high, so I don't "invest" in a ticket despite the favorable expected value. On the other hand, if I could buy -- invest in -- all the numbers, I'd win big, subject to sharing the prize with some lucky bozo who happened to guess right.)
Assigning the relative likelihoods to the four scenarios is highly subjective, but reasonable estimates can still be made. If each outcome were equally likely, then the probability of each would be .25 (25%). I have already mentioned that I think the likelihood of failure is .25. I think so because it seems to me every major thing must go right for Valence over the next 5 months, and I certainly cannot rule out something will go wrong. However, as an optimist I think (hope) things will go right. I think management is on track with everything under its control, I think the customers will want the product within this time frame because of the competitive advantages it will convey, and I think Castle Creek expects to make its real money as an investor and not as a short-seller.
Now, a 25% likelihood of failure implies a 75% likelihood of "success," and that 75% must be apportioned among the three success scenarios. Of course each could be considered equally likely and hence have a 25% likelihood, but that seems implausible to me. As production and contract announcements are delayed and delayed, the likelihood of great success seems ever less relative to good or decent success because of the ever-increasing likelihood of competition. As more time passes, it is ever less likely that Valence will enter the market as the sole major supplier.
Conversely, it seems to me the likelihoods of decent and good success are about equal because the difficulties facing Valence are probably facing all the potential competitors. In the absence of any information about the status of its competitors, it is almost impossible to guess whether a year from now Valence will face a few or many competitors. In this vacuum, we could argue that it seems more likely that it would face a few because of the production difficulties faced by all, and hence shave our weights in favor of the good outcome. On the other hand, we could argue that by then enough time would have passed for many of the (luckier) competitors to have overcome their own production problems, making us shave our weights in favor of the decent outcome. Splitting the difference between these two possibilities leads me to assign them equal probabilities.
Such considerations as these, then, lead me to assign failure a probability of 25%, equal probabilities for decent and good success, and the probability of great success of just half that of decent or good success -- which is to say, 25% failure, 30% decent, 30% good, and 15% great success.
Now that we're in for a dime, we might as well go in for a dollar. Applying these weights to the "average" share prices derived in the three scenarios (failure here means a share price of 0; the equation reads from failure to great success) yields the following "expected value:"
E(v) = .25(0) + .3($12.50) + .3($65) + .15($130) = $42.75/share.
Or, for Robert and the fun of it:
E(v) = .25(0) + .33($12.50) + .31($65) + .12($130) = $39.88/share.
Given a current share price of hardly more than $3, this expected value suggests a 12 - 14 fold return on the investment, making it attractive indeed.
On the other hand, a reasonable man could say that a 25% likelihood of losing everything is too great a gamble regardless of the potential payoff. A reasonable man, reading the same information, could also say that a likelihood of failure of only 25% in light of the necessity that everything goes right is much too low. A reasonable man might conclude it better to wait for production, even though that would mean missing the initial spike. A reasonable man could change this all around every which way and draw all sorts of reasonable conclusions. That's what makes a market!
Obviously all this is subjective, no more no less, and the misplaced precision is of course a bit absurd. Robert should be amused. It is obviously dependent on structuring the model to allow only 4 mutually-exclusive outcomes, and it obviously depends on all the assumptions and on my subjective weighing of the relative importance of all the different circumstances I consider important. It also leaves out all sorts of stuff -- like the price elasticity of demand, the state of the world economy over the next several years, the effects of Valence's participation in other markets such as the cell phone market, and on and on and on.
Nevertheless, despite its deficiencies, such an analysis as this provides us with a model with some guideposts with which to judge our investment and which may be developed in any direction we want. Don't get hung up on the "precise" numbers, though, for that would be to miss the point. The point is the reasoning behind them. The numbers merely serve to give an idea of where different lines of reasoning lead. I invite you to do a "sensitivity analysis" by playing with the numbers, the assumptions, the structure of the model -- everything. I am curious as to what you might find.
EOM !
Regards, lws |
