It's been a big week of events: the Federal Reserve on Wednesday (8/1) downgraded its view on the economy, but did not initiate QE or changes to the Fed Funds interest rate (which will remain between 0 and .25). Then, on Thursday, stocks continued a fourth straight decline on the heels of ECB President Mario Draghi's disappointing comments. Draghi set expectations high last week, noting that the ECB would "do whatever it takes" to save the euro, and disappointed the markets after announcing that the ECB would purchase Italian and Spanish bonds on the open market, but only after other European governments utilized bailout funds to do the same. Friday also yields the non-farm U.S. payrolls report for July, with ADP's prior estimate at +163,000 jobs and Marketwatch notes that nonfarm payrolls should increase by 100,000 with unemployment remaining at 8.2%. In the midst of it all: Knight Capital Group (KCG) announced a $440 million loss after a trading algorithm malfunctioned and created abnormal trading volumes in 148 stocks. With such a drastic drop in price (from nearly $10/ share three days ago to now a mere $3.50), is KCG a contrarian value play?
My idea here is that market cap generally reflects the market's idea for the overall valuation of a company. There are two things to consider in the face of Knight's recent blunder: will it survive in the short-term, and if it does, how much revenue/ earnings, and thereby market cap erosion will it face due to it over the coming years? This analysis assumes, for some simplicity, that market cap loss should be 1:1 with potential liabilities and earnings loss moving forward.
Things to Consider:
The dramatic drop in share price changed the market cap from $982 million (stock at approximately $10) to $244.5 million (stock at $2.49). This represents a drop in market cap of $737.56 million, nearly $300 million dollars more than the initial announced loss. What else could contribute to the loss in market cap aside from the initial loss? Seeking Alpha contributor Adam Gefvert notes three other items: loss in customer confidence, potential lawsuits, and the need to fix technology issues while it loses revenue.
A loss in customer confidence is very real, especially in the face of MF Global's losses and eventual bankruptcy. TD Ameritrade has already noted that it is running tests before routing through Knight, but notes that if Knight remains in good standing with exchanges, there is no reason for TD Ameritrade to stop routing trade through it. E*Trade did not route trades through Knight the day after Knights' trading system disintegrated. CME group, which was burned by MF Global and also regulates KCG's newly acquired futures brokerage, is "monitoring" Knight Capital. KCG is certainly walking a tight rope with regulators and retaining customer confidence - who would want to trade with a firm who's trading software just cost them $440 million? To further affect the company: the need to fix the technology issues will be costly and needs to be fairly immediate. The software has been taken down and is being analyzed, and it is unclear to what extent customers have avoided routing trade through the company.
That being said, the potential for lawsuits here is also real. The NYSE has canceled trades on six stocks outside of a 30% range between 9:30 am and 10:15 am; but, as the article mentions, Knight will likely be directly liable for any losses incurred by traders from the opening price through the 30% breakage threshold. Knight will most likely also be liable for trades in any of the other 140 stocks that experienced abnormal volumes and strange spike ranges.
The WSJ's streaming markets pulse notes that "the question for investors is whether the Knight Capital Group is in a death spiral or if it can find a white knight to help weather the storm". The company, realizing the need to shore up its current capital structure, has reported to the Wall Street Journal that it is in talks with Silver Lake Partners-backed trading firm Virtu Financial about a possible deal or capital infusion. This is extremely important: seeking a suitor or a capital infusion screams "we need cash". Before doing any type of valuation, investors must ask themselves if Knight can simply survive the short-term. Survival will almost certainly require finding a helping hand. Backing from an outside firm will help to keep KCG financed through the turmoil, prevent further degradation of the stock, and maybe restore some confidence from customers.
Estimating the Loss, and a Potential Contrarian Buy-In Price:
If investors believe that KCG will weather the storm, then the loss must then be estimated to find a buy-in price. I believe that after this one time event, the revenue/ earning loss should be reflected in a similar market cap loss.
At an initial loss of $440 million, the additional losses must be determined in terms of loss of revenue due to short-term and long-term losses in customers and costs to quickly fix the technology. Taking the most recent quarterly income statements from Knight shows that in trailing twelve month period, earnings were $103 million. Dividing this number by the amount of trading days (251) yields earnings of approximately $410,000 a day. Let's assume, albeit arbitrarily, that in the short term, for the next 10 days, Knight loses 95% of its revenues and that trickles down to losses of 95% of earnings on average. (This is ignoring how margins change revenues into earnings. It is not unreasonable that a 1:1 revenue loss to earnings loss in the short term will exist; Knight sells a service, not a product, and SG&A is fixed in the short term). Each day, Knight would lose $389,841 dollars or $3.90 million in a two-week period. Moving forward, we'll assume that the firm bleeds 50% of earnings each day because of this gaffe for the rest of the year. With a little more than half the year left, we can assume that there is approximately 120 trading days left. This amounts to an earnings loss of $205,000 a day and a total of $24.6 million total. Stemming solely from this single event, I'll estimate that the market generally forgives Knight if it survives. For next year, Knight loses 10% of earnings. This means for another 251 trading days, Knight loses $41,000 a day or $10.29 million. Together, these amount to approximately $39.40 million. Tack on technology expenses and rounding, and I estimate earnings losses and a resultant market cap drop due to customer loss and technology at approximately $43 million. This estimate puts Knight losing 42% of its 2011 earnings, if it survives.
The biggest estimate here is the liabilities on stock losses. With Knight most likely directly liable for any losses in the 30% band on the six stocks that the NYSE cancelled orders on, and with volumes at eight times average for WZE and even higher than that for others, the liabilities could be very high but are amorphous at best. This is any investors guess, but I believe that a fair estimate of $300 million in liabilities, and a similar loss in market cap, is reasonable. This would put KCG having five times liability of what the NASDAQ has set aside for the Facebook IPO, $62 million in cash. (For a slightly longer reasoning, see the addendum).
To be transparent, I would like to remind readers that the above loss estimates are very arbitrary estimates; there is no way to definitively estimate customer confidence and customer loss in such a short time frame following the automated trading incident, or how extensive lawsuit liabilities will end up being (the lawsuit estimates are just that: estimates). Many of these numbers are totally contingent upon estimates.
With that stated, I think that Knight is setting itself up for a contrarian play - but don't start buying yet. Investors must ask themselves both if Knight can find a suitor or capital infusion and if, with the capital infusion (or without it), can Knight whether the storm and avoid bankruptcy? How much will the trading loss actually cost the company? Beyond this, investors must also consider that Knight only has $555 million in cash and short-term investments as per its most recent balance sheet (without considering debt, acquired through Fidelity). It's worth repeating: the most important thing is in the short-term, can Knight regain its footing and avoid bankruptcy? It clearly must do if investors are to be rewarded. All told, I would be interested in a small position at or below $2.01, which represents a further 20% drop from current prices, $0.5 drop from the current price of $2.49 a share, and a total market cap loss (from approximately $10 a share) of approximately $785 million. This $785 million loss is inclusive of $440 million upfront loss, $43 million loss over a year and a half, and a cushion for $302 million in lawsuit loss. It also assumes that market cap loss should mirror losses the company faces going forward on a 1:1 basis. That said, I plan on waiting even if the stock reaches this number in the coming days. Waiting to see if a bottom forms, if Knight can find a "white knight" to infuse it with capital, and if bankruptcy can be avoided is advisable for those interested in playing this as a contrarian.
- Wizard Software (WZE): traded approximately 450,000 shares, but many of these trades were cancelled. Trades above $4.68 were cancelled, and the stock opened at $3.50. Looking at the volumes (rough estimates, based on "eye-balled" volumes from stock charts on freestockcharts.com), I would estimate that no more than 100,000 shares were traded below this level. The liability for straight losses (barring legal fees, etc,) is estimated at most $118,000 (100,000 * 1.18 = $118,000).
- China Cord Blood Corp (CO): Trades were cancelled above $3.22, and the stock opened at about $2.43. Total volume for the day was at about 1,889,000 according to Yahoo finance. However, intraday volumes appear to indicate that no more than 250,000 of these were traded below $3.22. I estimate the liability for only losses at most $197,500 (250,000 * 0.79).
- E House China Holding Limitied (EJ): Trades were cancelled below $3.36 and the stock opened at $4.75. Total volumes were 4,859,400, and I would estimate that 1/5 of these at most were traded above the level. I estimate the liability for only losses at most $1,350,000 (1,600,000 * 1.39 = $1.35 million).
- American Reprographics Company (ARC): Trades were cancelled above $5.71 and the stock opened at $4.41. Total volumes were 3,550,300, and I would estimate that about 1/4 of these at most were traded above the level. I estimate the liability for only losses at most $1,153,848.
- Quicksilver Resources, Inc (KWK): Trades were cancelled above $5.91 and the stock opened at $4.49. Total volumes were at 10,104,600, and I would estimate that about ½ of the shares traded below that level. I estimate the liability for only losses at most $7,174,266.
Totaling these gives me a total of about $10 million liability in losses for 5 stocks. Considering that these were the ones that were cancelled out, without much further investigation, I assume that liabilities for the other 143 stocks should be less than these liabilities' average, since they most likely did not move more than 30%. (Note: the liability range is wide from the above five stocks, so take it with a grain of salt; also, upon inspection of other stocks listed by NYSE, many moved within a closer range of 15%, so liabilities should be less on average). In an effort to develop any kind of estimate, I take the average and multiply by 148. This comes out to $300 million. This number should be high considering that this average should be high because I am taking the worst of the potential liabilities, developing liability estimates, and then multiplying the average out. I reconcile this number by deciding that this average of $300 million also includes legal fees and processes (to mitigate what should be a higher average). Even ignoring these estimates, $300 million is a very fair cushion for any liabilities.