In our view, one of the most tell-tale signs of extremely frothy conditions in the current market is the triumphant return of "quants" on Wall Street. One obvious sign is that hedge funds such as Viking Global and Third Point that have long prided themselves on fundamental long/short stock picking are now hiring quants to better deal with "factor" model algorithmic investors that have made alpha much harder to find. Bloomberg also recently profiled the "quant" strategy, referring to the quant strategy as a "craze" that is "evoking memories of trades that got too crowded in the past".
We think Universal Corp. (NYSE:UVV) presents a compelling opportunity for man to outsmart machine. Given that there has been little to no fundamental news or fundamental improvement in Universal's business over the past few months (or years for that matter), we believe that a perfect confluence of factors has resulted in momentum-loving algos piling into this stock hand over fist. Universal presents savvy traders with an opportunity to make money at the expense of algos.
We think the stock has been run up by momentum-loving algos (ones that chase both price momentum and EPS momentum). Based on management guidance, we think revenues are set to go negative in the upcoming quarter. If momentum-loving algos were able to take this stock up 50% in short order, imagine what that will look like on the way down as this trade unwinds.
We believe that the basis for the recent surge in Universal's stock will come undone as early as the next reporting quarter.
While many think of algos as the smartest money on Wall Street, when it comes to this boring tobacco stock, we are pretty sure that it's the algos that have their wires crossed.
We are recommending investors SELL UVV shares with a price target of $55/sh, or ~35% downside from current levels. We advise UVV investors run for the exits before this tobacco stock goes up in smoke.
Source: Slash Film Image
Universal is, by all standards, one of the most boring and unsexy companies you will ever come across. The company is an independent leaf tobacco merchant, effectively serving as an outsourced tobacco procurement arm for the large cigarette companies. There is no sell side coverage on the stock, which is surprising given that the company has a $2B market cap. This company has gone nowhere in the 15+ years. The company's revenues were just shy of $2.5B in 2000 and in 2016 were just under $2.2B. Operating income has followed a similar path, slowly bleeding down from slightly over $200M back in 2000 to slightly under $200M today. The company faces an obvious negative secular backdrop of stagnating tobacco consumption, as well as increased supply chain verticalization that makes Universal increasingly irrelevant as time passes.
Source: UVV IR deck
However, a curious confluence of technical factors and a lack of sell side coverage have resulted in this sleepy stock running to all-time highs and becoming a darling of retail investors that we think are setting themselves up for slaughter in the coming months.
Universal's stock now trades at an all-time high, with its dividend yield at ~2.6% versus its long-term average of between ~3.5% and 5%. Its P/E multiple is now sitting at almost ~20x versus a long-term average in the low-teens. The company generated $178M of free cash flow (using management's fallible math to calculate FCF) in FY2016, implying that on a trailing basis, the company's free cash flow yield is somewhere around 8.5% based on the company's current valuation. On our math, adjusting for the working capital movements over time, we think the stock is trading closer to a 7.5% normalized free cash flow yield. While on a headline basis this seems fairly robust, one must keep in mind that other companies that face secular pressure such as Xerox (NYSE:XRX) trade at solidly double-digit free cash flow yields (XRX trades at a ~12% FCF yield). So both on a historical backward-looking basis, as well as a relative to its own history, we believe that Universal is egregiously overvalued.
This sleepy tobacco stock up almost 30% YTD in 2017 alone, putting in as good or better a position than tech darlings like NFLX (+17%), TSLA (+23%), FB (+17%), and CRM (+17%). Even more amazing is that this stock is now about 50% higher since December 1st. This compares to cigarette name Philip Morris (NYSE:PM), which is up 17% over the same time period.
Source: Bloomberg, and Friendly Bear graphics
Given the lack of credible news flow that could have driven up Universal's stock price since December 2016, we started to wonder whether the company had secretly announced plans to move from the traditional tobacco industry into the "whacky tobacky" industry.
But alas, we now present the reasons that Universal is trading at all-time highs. We ultimately attribute the move in the stock to a strange confluence of factors that caused "momentum hunting" algorithms to plow into shares in Universal as rapidly as possible.
By understanding what happened to Universal, you too can find a way to outsmart the machines.
So how did Universal go from a ~$45-55/sh to as high as $83/sh in a ~2 month time span despite its fundamentals doing nothing? In other words, how did Universal end up travelling without moving?
We break Universal's climb into steps.
Step 1: Algos and Arbs Anticipated the Convertible Redemption and Squeezed
For many years, Universal has had a convertible security outstanding. Convertibles securities are effectively quasi-bond, quasi-stock, with the buyer having the potential right to convert his bond holdings into stock holdings should the stock perform well.
Many quant-oriented funds practice a strategy known as convertible arbitrage. We won't go into the specific details in this report, but note that one of the largest hedge funds in the world, Citadel, was founded by Ken Griffin who got his start in the convertible arbitrage business. In this strategy, the arbitrageur tries to generate a risk-free profit by buying a convertible security and simultaneously selling short shares of the company that issued the convertible security. The premise of the strategy is that the convertible security is sometimes priced inefficiently relative to the underlying stock, for reasons that range from illiquidity to market psychology. This explains why Universal has historically had such high short interest in its shares. Arbitrageurs were trying to arbitrage out the spread between the convertible security and UVV's actual stock price.
On a historical basis, Universal has had ~3M shares sold short of its stock. This represents around ~15% of the float. Given the company's boring and relatively uncontroversial nature, this level of short interest is obviously surprising. We, therefore, believe that the bulk of the short interest in Universal shares came as a result of arbitrageurs who were hedging their holdings in Universal convertible securities by shorting Universal's stock.
All convertibles come with different "trigger rules" that cause the security to "convert" from a bond-like security into stock. In the case of Universal, the convertible was set up such that if the stock traded at over 135% of its "conversion price" for at least 20 days over a 30-day period, then the convertible security would immediately (due to a mandatory conversion provision) get converted from "bond" to "stock". The conversion price was set at around $46.58 when the company first issued the convertible, meaning the actual trigger price was ~$61/sh.
There were about 107,418 shares of the convertible outstanding, and those shares had a conversion ratio of ~22:1 (meaning that once the conversion triggered, owners of the convertible were entitled to ~2.5M shares of UVV and their convertible securities would be retired in exchange for these straight common shares in UVV).
We're simplifying a bit, but this is the essence of what happened.
The massive rally in small cap stocks after Trump was elected lifted all boats - including Universal. Cigarette stocks also appear to have gotten a bit of a boost during this time period due to M&A in the space. As a result, the stock started to get bid up due to money flows coming into small caps.
As this took place, the probability of the convertible security getting pushed into a mandatory redemption increased. As UVV pushed above $60, it became increasingly clear that the convert would be more likely to trigger given the sweet spot for conversion was set at ~$61/sh. Any "quant" would have been able to measure on a daily basis the likelihood of the convertible security hitting the threshold required to convert from debt to equity.
In fact, we think quant-oriented convertible arbitrageurs likely spotted the potential for Universal's convertible to go into mandatory redemption and also spotted the massive short squeeze that would accompany that conversion.
As UVV's stock price moved higher, we think an interesting exercise in game theory took place. Assuming redemption did one day occur, the convert arb players were going to, in the future, receive ~22 shares of UVV for every 1 share of convertible they owned based on the future trading price of UVV. Therefore, the savviest players would have recognized that the liquidity-driven short squeeze from the unwinding of hedges was likely to send UVV shares higher in the future, creating a race to unwind short position hedges in UVV as quickly as possible so that their funds were exposed to the future upward movement in UVV shares.
As a reminder, the convertible arb funds likely had long positions in the convertible security that were offset by short positions in Universal stock. The arbitrageur attempts to profit off the "spread" implied between the price of the convertible security and the price of the stock. Due to liquidity issues, there is often a spread that can be captured by arbitrageurs who go long the convertible security and short the stock.
However, as Universal's stock price started moving up, the potential for the convertible to end up in mandatory redemption increased (this is effectively a reflexive exercise). If the convertible ended up getting redeemed, the short positions in UVV stock would ultimately HAVE to be unwound. The short positions, as we have already noted, were likely not tied to fundamentals, but rather to a hedging strategy. With the convert out of the picture, there would be no need for a hedging strategy and all shorts would cover (by buying UVV).
Normally, convertible arb funds are supposed to be fully hedged until the redemption takes place - i.e. they would not cover their short position in the stock until they are getting actual stock delivery because they remain long the convertible and would be exposed to movements in the stock price if they were not hedged through a short position.
However, in the case of Universal, there was a good speculative reason to actually wind back hedges well ahead of the redemption period. It has to do with the liquidity in Universal's stock versus the short interest that needed to be covered ahead of the convertible redemption.
The funds that covered earliest positioned themselves to make the most amount of money if Universal's stock ended up rising by the time the redemption period occurred, but could have also stood to lose money if Universal's stock fell.
Short interest in Universal - that we believe is largely tied to a convertible arb strategy - fell from almost 3M shares in December 2016 to what we now believe is potentially less than 1M shares.
The "new" Universal - that we believe had been invaded by momentum-loving algos and chart-loving day traders - started trading like water - with the stock trading over $100M on February 8, 2017. However, between December 1, 2015, and December 1, 2016, Universal traded an average of only 174,708 shares per day, or under $10M per day.
Source: Bloomberg, Friendly Bear analysis
So as certain convertible arb players started covering their short positions in ANTICIPATION of redemption, you had the perfect storm develop that resulted in a monster short-term short squeeze.
It became a matter of liquidity and timing. Keep in mind that almost all of the short interest in Universal had to be covered at some point (assuming they were all tied to arb strategies) because the convert was eventually going to be redeemed. Arb players do not remain short once a redemption is complete. So for the arbitrageurs, the only question they had to ponder was when to cover their hedges… either wait until the redemption period formally began… or cover early in anticipation of the convertible being redeemed, and end up naked long the stock price, effectively gambling that the stock would be worth more in the future when the physical delivery took place the stock closing higher over the redemption period.
In other words, there was pent up forced buying pressure to the tune of 17x average daily volume (3M shares required covering versus ADV of 170k shares) that would have kicked off at some point in December as arb players became increasingly convinced that the redemption was going to happen.
Algos Fooled by Reflexivity:
Source: Friendly Bear Graphics
Step 2: Redemption Happens and Remaining Arb Shorts Cover
As expected, the company came out on January 9th with an 8-K that announced formally the redemption of the convertible security. The stock had traded above the trigger threshold for the requisite number of days by this point in time. The company indicated that the redemption price would be based on the trading value of Universal between January 17th and January 30th, and would be paid out in cash on January 31st to holders of the convertible security.
As we already pointed out, certain risk-seeking convert arb players clearly covered their hedging position prior to the redemption period. They were trying to play the "reflexivity" chart above. They assumed the stock would close higher in the future due to the massive upward pressure from all of the short covering that needed to take place relative to the stock's paltry historical trading volume. Again, when this process all kicked off in early December, about 17 days' worth of volume needed to be covered in order to close out convertible hedges. So, some convert arb players simply front ran the covering, knowing that the immense buying pressure in the stock when compared to its light liquidity was going to create a monster squeeze and the potential for quick trading profits.
However, other convertible arbitrageurs who have tighter mandates would not have started to close out their hedge positions until January 17th.
We suspect some convertible arb shops have tight mandates that do NOT allow them to take naked speculative positions in stocks. So these funds waited until January 17th to begin covering, and would have likely covered based on the weighted average volume traded each day between the 17th and the 30th, to make sure their hedge was as effective as possible relative to the final redemption price.
And unsurprisingly, the stock again had a huge run between January 17th and January 30th as the remaining "short covering" from the leftover convertible arb players that remained hedged until the very last day continued:
However, this is where we think things went into crazy land. An algo must have gotten its wires crossed…
We did a quick analysis. Between 12/1/16 and 1/30/17, UVV stock went up 29 of 40 days (compared to 19 for the SPX). We think this incredible daily upward move in the stock resulted in momentum-loving algos plowing into the stock.
Keep in mind, algos are ultimately trained to recognize inflection points in stocks.
Here, we had a sudden surge in volume in the stock - with trading volumes going from the average of ~175k shares per day to well north of 500k shares per day - as well as a sudden surge in price - with the stock up 29 of 40 trading days.
What's not to love from the perspective of any algo out there. From an algo perspective, it really did look as if something had magically improved/changed at Universal... despite this being the same boring sleepy company that had existed forever.
So suddenly, Universal, a company with no sell side coverage, is now trading around 50M to 100M per day as algos and other momentum-loving day traders piled in.
Step 3: The Pullback Followed by "Momentous Earnings"
As one would expect, short sellers likely clued into the technical short squeeze and the redemption date of January 31st for the convertible security. Therefore, unsurprisingly, Universal saw a massive correction right on the day of the redemption (January 31st) as the forced buying dried up. As a reminder, by January 31st, all convertible arb players would have fully covered their short positions as their convertible securities got taken out for the cash value of the stock on January 31st, rendering their hedge unnecessary.
However, rather than the stock staying put on January 31st, something very interesting happened. The stock first saw a bit of a dead cat bounce, hovering around $70/sh after its initial January 31st correction.
This did not last for long.
As a reminder, Universal does not have any sell side coverage. Therefore, there are no estimates for revenues or earnings available to investors. This makes knowing what is "priced in" difficult. It also made for a rather entertaining situation after the company reported its most recent quarter.
In the most recently reported quarter, Universal reported revenue growth of 14% and earnings growth of 26%. Not too shabby for a boring unsexy business right?
Both of these metrics look strong on face value. Even at its current valuation of ~20x trailing earnings, these growth rates would be attractive from a high-level screening perspective.
Now here is where things get interesting and we suggest anyone playing around with Universal stock pay close attention:
From a human perspective, one would easily be able to see that Universal's 14% growth rate in the most recent quarter was up against a -23% growth rate in the comparable prior year quarter. Similarly, its 26% earnings growth rate was up against a -20% growth rate in the prior year comparable quarter. So while Universal put up strong headline growth, looking at the business on a rolling two-year period tells a very different story.
Had any sell side analyst been covering the stock, that analyst almost certainly would have baked in the easy comps when designing his/her model and would not have resulted in a beat relative to consensus estimates, which is usually what determines if a stock goes up or down on earnings.
Universal also does not provide formal guidance. However, the company did provide the following color commentary on the call regarding its outlook for 2017:
Source: Bloomberg Transcripts
Furthermore, on the call, it provided some incremental detail on Q3:
Source: Bloomberg Transcripts
As a reminder, Universal has a 3/31 year end. So here you have a company with no consensus numbers that put up a strong headline number that was driven purely by a timing shift of revenues from Q4 into Q3. Reading the 10-Q, the company explicitly states that YTD revenue growth has been driven by higher volumes and then goes on to say that volumes for the full year will be down. If that was not clear enough, it even goes so far as to explicitly state that Q4 volumes were going to be down.
Doing the math, this implies that revenues in Q4 should be down at least 15%.
Source: Friendly Bear Analysis
As for valuation, we have some thoughts - including how to think about the fact that the company put its cash to use by redeeming the convertible, an act that should be accretive.
Given Universal has not grown in the last 15 years, the dividend is a crucial component to UVV's stock price return. Over the past 15 years, the stock has traded at a similar 2.6% dividend yield exactly twice. Shorting the stock when it hit those levels led to 35% declines in the stock within the next three months.
Similarly, if UVV returns to a 4% dividend yield, you are playing for 35% downside.
Historical UVV Dividend Yield
Also, we wanted to provide the math of how the conversion of the preferred shares will impact EPS going forward. 2.4M shares of the preferred were redeemed for cash, which effectively equates to a 10% buyback. While we normally applaud returning capital to shareholders, we think UVV massively overpaid for this buyback, paying $74 per share for this buyback, 25% above our estimated fair value. When companies overpay to repurchase their shares, this is capital destructive.
The redemption of the convert does have a positive impact on EPS due to the buyback dynamic. While UVV will collect less interest income on the cash it was holding, the conversion does increase EPS by ~9% by lowering share count. Applying the historical multiple for UVV of ~12.5x, this gets you to a share price of $53.75 or down 34% from current levels.
Source: Friendly Bear analysis
Some investors we have discussed Universal with have pointed to the free cash flow yield as providing a margin of safety.
If you just use management's definition of free cash flow for FY2016 of $178M, then the stock is trading at a trailing FCF yield of 8.6%. However, for some reasons, management defines FCF as EBITDA - Capex. We know that people are excited about Trump's tax policies, but even he isn't suggesting a 0% rate.
Last time we checked, companies still had to pay taxes. UVV also has massive swings in working capital given the capital intensity of the business. FY2016 FCF was really $136M using the proper calculation of Operating cash flow less capex. With this math, its FCF yield is a paltry 6.6%.
Source: Friendly Bear analysis and UVV SEC filings
Longer term, FCF has swung around quite a bit (largely from working capital), which shows why UVV has only been able to increase its dividend by about 1 penny per year.
Source: Friendly Bear analysis of actual free cash flow over time
Conclusion: How to Think Like an Algo and Profit
Instead of thinking like human investors, let's think like algorithmic computers for a moment.
Here is the setup.
Universal's stock went from $55 to $75 on high volume simply because of a short squeeze related to a convertible hedging strategy. The algos did not know this, however. They simply perceived there to be some shift in sentiment around Universal given the rapid short covering and the heavy volume.
This was followed by a huge pull back in the shares from 75 to under 68 in the days leading up to the company reporting earnings. The company then posted earnings - that on face value looked robust - with huge top line and earnings growth.
With no sell side numbers to benchmark against, the algos saw a company that had been unfairly punished running up into earnings. Their models compared the valuation to the growth rates and stock action… and said
BUY BUY BUY
However, if the algos were able to actually parse language in the company's call transcript, they would be able to easily tell that the company simply pulled forward demand and is going to see yet another negative revenue growth quarter in the quarter ended 3/31/17.
Once the company puts up yet another quarter of same old same old, we think the obvious next step will be Universal trading back to where it began - $55/share, its historical range and in-line with its historical valuation.
That gives the shares ~35% downside from current levels.
Therefore, we have one parting thought for the day traders and opportunistic investors who have piled into Universal.
While the algos are currently saying BUY BUY BUY, tread very carefully. The company has already told you growth is going back to negative in the upcoming quarter, leading us to believe that the algos are going to go from BUY BUY BUY to BYE BYE BYE very soon.
Disclosure: I am/we are short UVV.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: I am/we are short UVV. All information for this article was derived from publicly available information. Investors are encouraged to conduct their own due diligence into these factors. Additional disclosure: This article represents the opinion of the author as of the date of this article. The information set forth in this article does not constitute a recommendation to buy or sell any security. This article contains certain "forward-looking statements," which may be identified by the use of such words as "believe," "expect," "anticipate," "should," "planned," "estimated," "potential," "outlook," "forecast," "plan" and other similar terms. All are subject to various factors, any or all of which could cause actual events to differ materially from projected events. This article is based upon information reasonably available to the author and obtained from sources the author believes to be reliable; however, such information and sources cannot be guaranteed as to their accuracy or completeness. The author makes no representation as to the accuracy or completeness of the information set forth in this article and undertakes no duty to update its contents. The author may also cover his/her short position at any point in time without providing notice. The author encourages all readers to do their own due diligence.