CEGE-
Interesting ReadingGenetic Capital, Inc.
Equity Research
Special Situation Report
The CELL GENESYS-ABGENIX Capital Market Link ? Arbitrage?
March 17, 2000
Contact:
Mark Afrasiabi, Director of Research
m_afrasiabi@geneticcapital.com
DISCLAIMER: PLEASE READ THIS BEFORE READING REPORT
This is only our Investment Opinion and nothing in this report should be construed as an offer to sell or the solicitation of an offer
to buy any securities. This opinion contains forward-looking statements about the companies mentioned; these companies are
subject to many risks and uncertainties, including, but not limited to, product development risk, technological obsolescence risk,
regulatory delays or non-approvals of potential products, intellectual property litigation risks, and general capital market risks.
Please refer to risk and uncertainty factors pertaining to the companies mentioned herein, and biotechnology firms in general, in
current SEC filings, e.g., Form 10-K.
The position we take that investing in Cell Genesys offers an arbitrage opportunity is analytically only to particular market
participants as described in the report. That is, one must read the report with care and analyze who it is that we argue face the
arbitrage investment opportunity, and why. This is critical as it is not true that all investors face this opportunity. This is a theoretical
explanation for recent market phenomenon, and must be construed as only our subjective theory for a certain capital market
phenomenon. We may be wrong on any or all of it; we simply want to disseminate it for market participants to take into account.
INTRODUCTION
In the past year, gene therapy leader Cell Genesys (NASDAQ: CEGE) spun off Abgenix (NASDAQ: ABGX), a fully-humanized
monoclonal antibody (fhMAB) firm, originally retaining a 19% minority stake; subsequent to ABGX?s IPO, ABGX increased
dramatically to a recent peak of $413 per share, or app. $8.3B in market cap, recently falling to $275/share, or roughly $5.5B.
CEGE recently sold approximately 966,000 of its 3.392 million ABGX shares at $210 in the ABGX follow-on and over-allotment
(February, 2000). This ABGX-based financing raised roughly $200M for CEGE, bringing total cash to roughly $250M, with
approximately 2.426 million net ABGX shares left.
We think an extraordinary market failure is currently taking place in the CEGE share market. Recently, Cell Genesys has been
trading in a very bizarre valuation range given its equity position in ABGX. In fact this odd phenomenon has occurred before; prior
to a Robertson Stephens analyst report in mid-February reiterating its buy rating on CEGE -- owing in part to CEGE having
?negative technology value? -- CEGE had traded at or below zero technology value for a few weeks. We explain in this report why
we think this represents arbitrage, but, more importantly, why it has managed to recur even with investment banks and
sophisticated market participants clearly aware of the linkage: the analytical explanation offered by market analysts has thus far
failed to accurately characterize what in fact is going on in the CEGE share market. Given a lack of understanding of a more subtle
phenomenon which we hope to clarify, the market failure has recurred. As a proportion of its market cap, CEGE?s technology
value has again decreased dramatically -- to the point of near worthlessness or, paradoxically, negative value at times in recent
trading (i.e., March 15 and March 16). However, there may be more than meets the eye ? it may not be that CEGE is trading at
negative technology value, but simply that CEGE holders discount ABGX.
The crux of this paper is not to tout CEGE or fhMAB firms such as ABGX as stock picks, but rather to point out and explain a very
fascinating capital market situation that we believe offers arbitrage to a particular cohort of market participants. Here, we stress to
investors that when we talk of arbitrage, we DO NOT mean that buying CEGE is riskless; investors are still exposed to both CEGE
and ABGX price risk, i.e., price drops in either market. We simply mean that, at any given moment, prices across
markets ? here, an implicit market for ABGX shares through CEGE and the ABGX share market -- may be discrepant and, as
such, offer arbitrage.
ANALYSIS
First, before discussing the current valuation anomaly and explaining why we think the arbitrage persists today, here is a recap of
some recent trading activity. On February 15, 2000, ABGX was trading at $250, while CEGE, at the same time, was at $23. The
ABGX shares at this price equal approximately $638M; CEGE?s $225M cash plus this pro rata ABGX market cap equaled $863M,
or $26.50 per share assuming 33 million shares outstanding. Nonetheless, CEGE traded at $23, or roughly $805M ? less than
the market value of the sum of its cash and ABGX holdings. Then Robertson Stephens reiterated its buy, emphasizing ?negative
technology value? and the stock appreciated from negative technology value to over $20 per share of technology value at its
recent peak of $60-plus. At the time, we thought the arbitrage had been corrected and wrote off as moot our report (planned for
release the same day Robertson Stephens reiterated). However, in the recent biotech correction, CEGE dropped from $60, while
ABGX was at $400, to $29, even as low as $25 today, while ABGX fell to $275. At $60, CEGE had a $1,980M cap, with $250M
cash and $970M in ABGX holdings, that is, $760M in technology value. However, on March 15, while ABGX was at $275, CEGE?s
market cap was $957M ($29 per share), notwithstanding cash of $250M plus ABGX shares of $667M; that is, technology value
was approximately $40M. This compares to an average gene therapy technology valuation of near $500M ? even after the
mini-crash. More precisely, CEGE?s technology value has dropped from $760M to $40M in 2 weeks -- a 95% drop. Given March
15th?s $25 low, CEGE was trading with negative technology value. Put differently, although the average biotech stock has lost
$20-30%, CEGE has had the most startling drop of the entire 400 some public companies ? 95-100% of its technology value
wiped out! Because the market failure has recurred we think the theoretical explanation for the CEGE-ABGX capital market link
needs to be more accurately described; this report hopes to explain how the arbitrage persisted for so long and why it has
recently come back ? notwithstanding investment bank coverage emphasizing the link.
The problem is that analysts? observations and explanations so far have been simply that CEGE?s technology has traded at
negative or near worthless levels. This is indeed true using the market price of ABGX to value CEGE?s holdings, but this is
perhaps not the best way to look at the situation; that is, the explanation of this bizarre, seemingly paradoxical CEGE technology
value may lie somewhere else: it?s not that CEGE market participants value CEGE at such absurd, even negative, levels, but
instead that CEGE market participants simply do not value ABGX at its market value. This inter-market valuation discrepancy has
opened up an arbitrage to a particular group of investors, namely, those who value ABGX at its market price.
We refer to this phenomenon as ?holdup-based arbitrage? because the arbitrage stems from the fact that one group is ?forced? to
hold an asset due to their holding in another asset despite the fact that they value the forced position below its market value. To
the market participants in the market for the asset that is discounted in the held up market, this can offer a way to buy the asset at
below market prices ? that is, to arbitrage a price discrepancy between the direct asset market and the indirect market for the
asset through the held-up asset?s market. This is akin to an externality imposed by one market onto another ? inhibiting value
maximization; on a similar note, the purpose and effect of tracking stock is based on freeing similarly ?locked up? value ? freed
up when a unit is valued separately from the market that previously discounted its worth. Unbundling of assets and allowing two
groups of market participants with differing preferences about the two assets is how value may be freed up (see Conclusion and
Recommendation, for how to potentially resolve this capital market inefficiency by restructuring CEGE?s ownership of ABGX
shares).
There?s more here than meets the eye ? perspective is key
The seemingly paradoxical valuation can be explained in two ways, and we believe our new perspective offers the correct
explanation for why this capital market failure has occurred and persists. One way to characterize what is going on is how analysts
have for the past few weeks: that CEGE?s net technology value has been valued lately at or below zero. More interestingly, and we
think more accurately, rather than framing the arbitrage as one where CEGE?s technology value is at negative or near worthless
levels, the arbitrage argument can be framed another way: if the CEGE market consensus believes CEGE?s technology value is
(as we think it should be) substantially more than, e.g., $0-4/share, then what is actually going on is that CEGE participants are
discounting ABGX?s value: this means one can purchase ABGX at a discount to its market value, i.e., buy ABGX indirectly through
CEGE shares at less than what ABGX trades for. For example, if investors believe that CEGE, exclusive of ABGX, is worth, say,
$450M ($250M cash + $200M technology value), or $13.63/share, then the CEGE stock price can be separated into $13.63 for
CEGE alone, and the residual $15.37 ($29 market price - $13.63 net-ABGX-valuation) for CEGE?s ABGX shares. The problem is
as follows: $15.37/share * 33 million shares equals $507M in CEGE market cap for 2.426 million ABGX shares, i.e., $209/share;
however, 2.426 million ABGX shares are worth $667M at the actual ABGX market price of $275. In other words, so long as
CEGE truly does have positive technology value, CEGE?s ABGX shares must, by definition, be at a discounted value vis-a-vis
their actual market value. We think this is what is going on: it?s not that CEGE market participants value CEGE?s technology at $0
or $80M, but rather they don?t value ABGX at its market consensus valuation. A formula can be developed for this implicit discount
of CEGE?s ABGX shares: [Price(CEGE) - Price(CEGE net-of-ABGX value) * 33 million shares] / 2.426 million shares = value of
ABGX shares in the CEGE-based ?implicit market? for ABGX shares. For example, if market consensus valuation for CEGE?s
gene therapy business was truly $960M fully diluted ($250M cash, $710M technology value) or $29 per share, then the above
formula gives us a value of zero for CEGE?s ABGX shares.
Crucially, it is impossible to know the price of CEGE excluding ABGX in the above formula (e.g., is it $300M, $500M, $700M?).
That is, we can?t discern what CEGE shareholders actually value CEGE at if indeed our ABGX discount theory is taking place: one
can?t look at the CEGE market cap and know what the CEGE participants are valuing CEGE at, or, put differently, one cannot know
the degree of the discount on ABGX?s shares by CEGE holders forced into a long position they evidently don?t want to be in. In
other words, the spot market for CEGE has been commingled with the ABGX stock where valuations differ from the actual ABGX
market, thereby distorting the signaling function of a stock?s market price.
The point is, you can look at the situation in two ways: (1) negative or zero CEGE value; or (2) positive CEGE value but
below-market ABGX value. Because things generally cannot have negative market value ? especially a gene therapy business
with broad intellectual property claims to potentially the most efficient vector ? we think there is something more subtle going on
that can explain the apparent paradox. That is, we believe CEGE participants wouldn?t value CEGE?s technology at $0 or $80M,
but rather simply do not value ABGX at its market price. Either way however, market failure exists, and, in our opinion, an
extremely rare arbitrage opportunity exists.
Valuation discrepancy between CEGE and ABGX market participants as to the value of ABGX -- the case for a spin-off of
CEGE?s ABGX shares into a tracking subsidiary so that the currently bundled assets can achieve their true valuations
Thus, we believe the market failure is due to the interesting phenomenon referred to above of CEGE shareholders not valuing
ABGX at its market value, that is, although ABGX is worth $X in the ABGX share market, CEGE market participants ? forced into
an ABGX position via their ownership of CEGE ? do not value it at such levels. Put differently, there is valuation discrepancy
between CEGE market participants and ABGX participants with respect to the value of ABGX. Interestingly, this type of valuation
discrepancy across people is normally taken care of by supply and demand; the problem here is with the CEGE-ABGX capital
market link: CEGE shareholders cannot sell their pro rata ownership of ABGX even if they value it far less than ABGX?s trading
price. That is, the average CEGE shareholder evidently is not a buyer of ABGX, and, if anything, is a seller of ABGX at these
levels; because their consensus valuation for ABGX is below ABGX?s market valuation, they discount the value of CEGE?s
holdings. Put differently, they are implicitly short-selling ABGX by valuing ABGX below its market value in their CEGE valuation
calculus. This implicit short position -- unapparent at first glance -- is why CEGE has traded near or below the value of cash plus
the ABGX position. In other words, it?s not that CEGE participants value CEGE as worthless or near worthless, but rather they
simply don?t buy the going ABGX price. This is akin to Disney investors not valuing an internet business, e.g., GoNetwork, on the
same level as internet investors but owning it nonetheless through their Disney shares: so long as the internet asset is bundled
with Disney, it will be imbedded in Disney?s market cap at less than its true value, that is, the value it would achieve if internet
investors could invest in it on a stand-alone basis. Thus, the argument for unbundling of assets used in tracking stock cases is
compelling here.
The reason we find this implicit ABGX discount theory to be the only plausible explanation for the apparent paradox is really by
necessary logical deduction: if CEGE participants actually valued ABGX at its market value, then CEGE would have zero to
negative value -- as analysts recently have termed what?s going on. We refuse to believe that this is what?s going on, especially in
light of other gene therapy firms? valuations and pre-ABGX CEGE trading ranges over the past few years. Hence, we are left with
one conclusion: CEGE market participants have a different consensus valuation for ABGX than ABGX market participants.
So what does this mean? Does it mean there is no arbitrage? Absolutely not. We believe that the inter-market valuation
discrepancy (or implicit ABGX short position imbedded in CEGE?s market cap) is a market failure and will be arbitraged away by
speculators and ABGX market participants once they realize they can buy ABGX through CEGE at less than ABGX?s market
price. Evidently, given the recent return of the anomaly, this arbitrage has not been properly picked up on by arbs or ABGX
participants.
Because of this discount to ABGX shares implicit in the CEGE share market, CEGE has become an anomalous risk-return
investment to market participants who value ABGX at the current market price. In other words, ABGX market participants can?t
deny that buying CEGE is a good idea as they can buy ABGX for what it?s worth in the ABGX market and get the CEGE gene
therapy upside for free -- CEGE?s firm-specific downside cannot exist to them because they, by definition, value CEGE?s ABGX
holdings at more than the implied discount valuation ascribed to ABGX by the CEGE market, that is, they value it at the consensus
ABGX valuation in the ABGX share market. Essentially, CEGE?s ABGX shares, valued through the CEGE share market, have
made CEGE like a tracking stock for ABGX shares (2.426 million of them), but the problem lately is that the ?tracking stock? is
trading at a huge discount to ABGX?s market price. The actual percent discount depends on the formula referred to earlier,
specifically, to the ?net-of-ABGX? CEGE market value -- which of course is hard to discern from CEGE?s market cap so long as
the ABGX shares are bundled up with CEGE .
To sum up and clarify the argument for an arbitrage, so long as investors can buy CEGE for less than its cash plus ABGX market
value, they can arbitrage as follows: (1) buy ABGX through CEGE at below market prices in the event market participants ascribe
some positive value to CEGE?s technology such that the spread makes CEGE?s ABGX shares worth less than their market price;
or, alternatively, (2) buy ABGX at its market value through CEGE, but get CEGE?s gene therapy business for free with no
CEGE-specific downside risk. To ABGX market participants and any other investors who value ABGX at its market value, this is
tantamount to getting a free call option on CEGE with no maturity. Moreover, because things generally cannot have negative value,
the aforementioned situation where CEGE trades at negative technology value (as it has today on March 16) is more accurately
characterized as getting (1) CEGE for free and (2) ABGX at below market prices -- a double arbitrage in our eyes. This makes no
sense.
CONCLUSION AND RECOMMENDATION
We believe that although not everyone, and apparently not CEGE investors, value ABGX at its market price, those that do, e.g.,
average ABGX market participants, will eventually spot this capital market link and buy ABGX through CEGE. This is theoretically
how the discrepancy will be eliminated. But the crucial fact here is that it is difficult to know how deep the discount is: ABGX
holders can without question buy ABGX below its market price if CEGE is at zero or negative technology value, but how will they
know when to stop buying when CEGE trades at a positive technology value, that is, how can they tell what the net-of-ABGX
CEGE market value is? The true CEGE market cap is masked by the ABGX position so long as the gene therapy asset is
bundled with the MAB asset, i.e., so long as the CEGE-ABGX ?blob? trades as one security. Because we can?t discern what
CEGE?s net-of-ABGX market cap is, we won?t know what CEGE investors value the gene therapy business at until we separate
the ABGX shares from the CEGE business. Logically, we think the only way to correct the arbitrage is either, (1) for ABGX market
participants to see that they can buy ABGX shares at a discount to market through CEGE and somehow estimate the imbedded
discount ? a difficult feat, or, more practically and efficiently, (2) for CEGE to spin out its ABGX shares into a separately traded
tracking subsidiary so that (a) CEGE?s true consensus technology valuation can be observed when CEGE is traded on its own
and (b) the ABGX shares can be properly valued, i.e., at their market value, by eliminating the discount by gene therapy investors
who evidently do not value ABGX?s MAB technology at the same level as ABGX investors. This latter recommendation could be
executed and we estimate that it will free up roughly $400M in market value that is currently precluded due to asset bundling with
concomitant inter-market valuation discrepancy. More precisely, given relative gene therapy valuations and analyst reports
pounding the table on CEGE to no avail, we estimate that CEGE shareholders actually value CEGE?s business, net-of-ABGX, in
the $650 range; given the current $28 share price ($925M market cap), the residual $275M of market cap is the therefore the
value of CEGE?s 2.426 million shares. This translates into a discounted price of $113 per share, roughly $160, or 60%, below
ABGX?s market price.
Intuitively, this makes sense as gene therapy investors do not necessarily value a MAB firm at the same levels as the MAB
investors, and vice versa. This clientele effect creates an implicit, indirect market for ABGX shares through CEGE
|
