My not so profound valuations and observations on Markets
My not so profound valuations and observations on Markets
First, what is DeFi?
DeFi is financial services on the blockchain. Where you can do transactions such as borrow/lending or exchanging assets in a decentralized manner. As this is all decentralized, there is no KYC. All transactions are permissionless and trustless. Where code is the law. So you might be asking how is this done? Well, through protocols.
What are protocols?
To put it simply, protocols are just a set of instructions on the blockchain. After conditions are met, the instructions are executed. Since these instructions are on the blockchain, once executed, they are immutable.
Type of protocols:
Here you could go to https://defillama.com/, and you can see all of the most popular protocols. You can go around on the site while reading to reinforce what you read.
Borrow/Lending:
On defillama, go to the lending section on the left and see all these borrow and lending protocols. Here you could go to the first site – aave.com - and see what are the borrow and lending rates right now. This type of protocol is like a bank. You can take your crypto, such as ETH, and deposit in these protocols to earn interest. Similarly, to a bank, if you deposit money, you will receive interest. Conversely, you could also borrow, where you would have to pay interest on the borrowed amount. Now you may be wondering, if a bank needs to do a credit check to give you a loan, how can these protocols give you a loan without any information? This is all done with over-collateralization – where the collateral is worth more than what you borrow. For example, say you deposit 100$ worth of ETH, you can borrow up to 70$ worth of BTC. Eventually, when you want your collateral back (your 100$ of ETH), you have to pay back the $70 of BTC plus interest.
Now the risk of the protocol is that you could get liquidated (fined a fee). Liquidation would happen, if the collateral drops in value or the borrowed assets go up in value past a certain point. The documentation of avee does a good job explaining the liquidation process (https://docs.aave.com/faq/liquidations).
Decentralized Exchanges:
The same thing as before, go to the Dexes section on defillama to follow along. Here you can think of it as the same thing as a normal centralized cryptocurrency exchange such as Coinbase, but a key difference is that it is decentralized. In a centralized exchange, it operates through a limit order book- where each buyer and seller puts in their bid and ask price, and the exchange connects the two. Now there are decentralized exchanges that use a limit order book, but they aren’t as popular and thus not going to be the focus in this post. More popular decentralized exchange, instead of using a limit order book to match you, a dexs is an AMM (Automated Market Maker).
Here an AMM doesn’t have to wait until a buyer, and a seller has a matching order to execute the trade, and AMM can always trade whatever it is. It does this most commonly through the product constant formula.\
Where the multiplication of the quantity of the two assets must remain constant. In this one equation, the ratio of x and y also set the price for the two. So, for example, if you trade assets x for y, the price of y would go up, and the price of x would go down.
Now for someone to trade between assets, someone must be providing liquidity. Liquidity providers deposit their token in a “pool”, and every time someone trade within the pool, they get a percentage of the fees incurred. Another way for liquidity providers to generate yield is when these Dexes give out an incentive token. For example, Bob goes on to deposit BTC and ETH on Curve, then Kate trades 100$ ETH to 100$ BTC for a 0.5$ fee. Since Bob has deposited into the BTC-ETH pool and Kate traded in the pool, then Bob would get a portion of the 0.5$fee depending on the percentage to the total pool he deposited in. So if there was 900$ ETH and 900$ BTC into the pool, he would own 1% of the pool and thus get 1% of 0.5$. To incentive people to provide liquidity, the dexes would give them their own governance token. In the case of Curve, it would be CRV. So Kate would also get a small percentage of CRV.
Yield:
I hope you know where to find the yield section on defilama. Here these yield protocols take the incentive tokens from the liquidity pools and sell them for whatever was in the pool and then automatically redeposit to the pool. For example, going back to the example of the BTC-ETH pool, since Bob is getting CRV as a reward, he could sell the CRV for BTC and ETH himself and redeposit the BTC and ETH back to the BTC-ETH pool. Or he could just deposit his LP on these yield sites, and it will automatically do it for him.
Other types:
Now there are many more types of protocols, but these three listed above are the main ones. Other protocols will include insurance, derivatives, staking, asset management, and many more. All the others are more a bit more advance and will deserve their own post.
Layer 1 vs Layer 2
On Defillama, you can go to the Chain section to explore. Layer 1 is a blockchain platform for protocols to be built on top of, such as ETH. Whereas layer 2 is built on top of layer 1 for faster and cheaper transactions. Now a key thing to understand is that these protocols are on multiple chains, with some protocols on one chain and others on many (cross-chain). Since the transaction cost to be on the Ethereum network is very expensive, with every single transaction costing more than 100$. Thus, unless you have a large sum of money to work with (more than 100k USD), it is advised to be on other chains, where transaction costs are as low as 0.01$.
Examples of a strategy:
A potential strategy might be to deposit your ETH into a lending protocol, use that as collateral and borrow USDC (a stable coin), then deposit your USDC in a liquidity pool, and then lastly, deposit the LP token to an auto compounding protocol.
Risk:
Smart contract risk – this is an inherent risk when dealing with defi protocols. This is when the protocol you are using gets hacked. Since this is on the blockchain, the action is immutable, and nothing can really be done about it unless you bought insurance for it.
Impermeant loss – this is the risk when providing to a liquidity pool. Since you deposit two assets, if one of the assets goes up by a lot more than the other - by the design of the constant product AMM – you would end up with less of the assets that appreciated and more with the assets that didn’t do as well. This can be thought of as the opportunity cost of providing to the liquidity pool. For example, if you deposited 50$ of BTC and ETH, and ETH goes up to 100$ and BTC stayed at 50$, if you haven’t deposited, the value would be 150$. But since you provided liquidity, you would be getting a value of 141.42$. This can be calculated by (https://baller.mechanaut.xyz/).
Coin risk – if you are getting a high APY, you might be thinking that you are getting a great deal. But if the underlying coin drops massively, then the APY doesn’t mean as much. For example, if you have 1 coin that is worth 1$ that is earning an APR of 100%. After a year, you have 2 coins, but if the value drops to 0.5$, you still have a value of 1$.
Regulatory Risk - since DeFi started in the summer of 2020, the government hasn't yet regulated the DeFi space. The outlook on this risk is far and wide, from people saying that the government is going to shut everything down to people saying that it is a run away train. This might be the key reason to why traditional institutions hasn't entered this space. The US government is already starting to scrutinize stable coins.
Vocab:
TVL – Total locked value, the amount of value locked in these protocols.
Stable Coin – A coin that is pegged to fiat currency, most of the time USD. Such as USDT and USDC.
Bridge – transferring assets from one chain to another.
ERC-20 – this can be thought of as a token/contract that is compatible with all other ERC-20 tokens/contracts. For earlier examples, I kept using BTC as an example, but BTC can’t interact with defi protocols since it is not an ERC-20 token. So for BTC to be used, it must be converted to an ERC-20 token, such as WBTC.
Wallet – This is the address where you send your coins.
Hot wallet – this is a wallet that is connected to the internet.
Gas – the transaction cost. For every interaction with a protocol, there is a transaction, and each transaction costs gas. Make sure you have enough money in your wallet to cover the gas because once you run out, you can’t make any more transactions.
APR - these are Annaual Percentage Rate. Often showed to attract liquditiy providers, but for most protocols, these APR are variable, calcualted based on past performances. So if yout see a APR at 250%, dont be surprised a week later it becomes 80%.
How to get started:
Here are some quick ways to get started in defi, even if you have never interacted with crypto.
Closing thoughts:
As of now DeFi feels like a very self referencing system. Where cryptocurrency are moved around to gain the highest yield - where is the value created? Is the value just so that people earn money? To be honest, I am struggling with that question. However, I do know that this space is evolving at an extreme speed, and innovation is happening at a breakneck speed.
Maybe DeFi creates the financial backbone for the metaverse to exist, or it get heavily regulated and becomes a fad? Who knows?
Resources:
Tips-
Make sure to read the documentation for each protocol to understand what they are doing.
Join discord groups.
Join crypto Twitter, follow founders and developers of protocols.
Google is your fren.
Academic –
https://berkeley-defi.github.io/f21
Websites –
Youtubers –
https://www.youtube.com/c/JustinBram
https://www.youtube.com/user/TheTaikster
https://www.youtube.com/user/williamhenkel
Podcast -
https://www.youtube.com/c/Bankless
During the March 2020 draw-down, I remember looking at the premarket returns to see whether the day is an up or a down day. While I was panicking, I was wondering if the premarket returns would tell us anything about the present day returns. I decided to find out, so I download the historical SPY price data from 1993 to 2021.
The US stock market starts trading from 9:30AM to 4:00 PM, this is when most people trade. Then since 1991 some stock exchanges started to allow institutional investors to trade after 4:00PM this is called after hours trading, now days after hours can run as late as 8:00PM. Then there is premarket trading, where trading occurs from 8:00 AM to 9:30 AM.
I first choose to use Close instead of Adj.Close because Adj.Close accounts for dividends and I decided that since I just wanted to compare the return during the night and day, so dividends shouldn't be included.
"Day returns" refer to the returns made during the day from 9:30 AM to 4:00 PM, and it is calculated by market close divided by market open so 4:00 PM price divided by 9:30 AM price.
Then on line 20, in day Returns I got rid of the first entry just so I can match with premarket returns. So the first day return is 1993 Feb 2nd.
In my code when I use "night returns" it is referring to "after hours and premarket returns". After hours and premarket returns is defined as 9:30 AM price divided by yesterday 4:00 PM price, so day open divided by previous day close. When I say "if you saw the premarket to be up", this premarket has already incorporated the returns of yesterday's after hours.
Lastly, "whole" refers to day close divided by previous day close, so today's 4:00 PM price divided by yesterday's 4:00 PM price.
I then combined all returns in one data-frame, and call it both (I know, I should probably work on my naming skills).
Just to show how the returns have been distributed throughout time.
Here we can see that during crashes there are huge movements in both after hours/premarket and day returns such as 2001, 2008 and 2020.
Then to see the cumulative returns of after hours/premarket returns, every day from 1993 to 2021, it would just be a product of all the "night Returns". Since if you were to buy at market close and sell at market open you would only get the after hours/premarket returns and all the after hours/premarket returns must be multiplied to reflect the compounding from 1993 to 2021. Same thing if you were going to buy at market open and sell at market close (the day returns).
So cumulative returns of after hours/premarket is 975%.
While cumulative return of day return is actually negative 9.6%. So you would actually lose money if you bought at open and sold at close every day, from 1993 to 2021.
There seems to be no real explanation with this huge difference, theory range from institution knows of this difference, and so they buy at close and sell at open or earnings always drops outside of trading hours, so most gain are from after hours or as simple to there are more hours at night than during the day (hahaha). This data suggests that buy and hold to be a better strategy than trading during the day. There is a whole academic paper that is written about how this behaviour is present from individual stocks to futures. Here we can confirm that we got the same results as what was given in the paper.
To continue with this analysis, I wanted to answer whether it was statistically significant that the mean of the day returns were smaller than the after hours/premarket returns.
So first I did a T-test:
Null hypothesis: Mean of Day Returns = Mean of after hours/premarket returns
Alternative Hypothesis : Mean of Day Returns < mean of after hours/premarket returns
From the T-test we can see that since the significance level is lower than 0.05, so we reject the Null hypothesis. From the given data we can support the alternative hypothesis. Here we can see that the mean of the day returns is + 0.0033% while the after hours/premarket returns is +0.034%, which make sense given our previous statement saying that the cumulative night returns are much larger than day returns.
But I realized that this data wasn't normally distributed, since if the market was actually normally distributed it would be practically impossible to have a + or - 10% in a single day. Which renders the T-test kind of useless for more analysis, since the T-test assumes the data are normally distributed.
As we can see from the QQplot both the night and day returns are far from being normally distributed. If they were, then all the points would have to be distributed on the line, here we can see that both after hours/premarket and day returns has fat tails, so it is not a normal distribution.
Since we can't use the mean to compare day and night return, we can however, use the median. In our case of "If you saw the premarket to be up, when the market opens, do you expect the market to keep going up and close higher?" it actually makes more sense to use the median rather than the mean, since you should be wondering whether most of the time if the premarket returns is up would the market keep going up?
Here I first selected all the data that has after hours/premarket return greater than 1 and called it above1. Because to answer the question "If you saw the premarket to be up, when the market opens, do you expect the market to keep going up and close higher?", I have only selected the times when the after hours/premarket is up along with the corresponding day returns.
Then I used an R-package made by my professor Ken Butler. In this package one can use the sign test instead of the T test, since the sign test doesn't assume the data to be normally distributed.
Null Hypothesis: Median of Day Returns = 1
Alternative Hypothesis: Median of Day Returns > 1 (I made this 1 because if you think the market is going to keep going up because the after hours/premarket was up, then day return should be greater than 1)
To use this test, we first find "upper" in the table, and get the p-value. We can see that this p-value is very small, much smaller than 0.05, so we can reject the Null hypothesis. And conclude that the median night returns are larger than the median day returns.
So to answer our question "If you saw the after hours/premarket to be up, when the market opens, do you expect the market to keep going up and close higher?", the answer should be a yes. Since the median value of day returns is larger than 1, when the premarket/after hours is up. In a laymen example, if the market closed yesterday at $10, and opened at $15 the market is going to close most of the time higher than $15.
However, from the previous point that most gains coming from after hours/premarket and not from day, we can confirm that the median after hours/premarket gains are greater than median day gains.
Here Null Hypothesis: Median of Day Returns = Median after hours/premarket Returns
Alternative Hypothesis: Median of Day Returns < Median after hours/premarket Returns
Here looking at the "lower" alternative the significance level is below 0.05, so we can reject the Null hypothesis, and we can conclude that the median day returns are smaller than the median after hours/premarket returns. Which means that even though the market keeps going up after seeing the after hours/premarket gains, most of the gain has already been taken during the after hours/premarket. Buy and hold investors should be happy reading this.
Now you might be wondering "If you saw the premarket to be down, when the market opens, do you expect the market to keep going down and close lower?"
Here I did the same thing as above, the only change we have to do is having different hypothesis.
Null Hypothesis: Median of Day Returns = 1
Alternative Hypothesis: Median of Day Returns > 1
Here instead of using "lower" we use "upper", here we can see that the p-value is very small, much smaller than 0.05. So we reject the Null Hypothesis. So we accept that the median day returns are positive when the premarket/after hours is down.
To answer the question "If you saw the premarket to be down, when the market opens, do you expect the market to keep going down and close lower?", the answer is no. This means that if the premarket is down, the market has already dropped and is on the way back up. In a laymen example, this means that if yesterday the market closed at $10 and opened at $5, the market will close higher than $5.
However, similar to how the premarket/after hours take most of the gains, it takes most of the losses too.
Here we use "whole", which is the return of "day close divided by previous day close". We use this because we want to find out whether the market will close lower today than it did yesterday, when after hours/premarket is down.
Here is the null hypothesis: Median return of whole = 1
Alternative hypothesis: Median return of whole < 1
From the data, since we want to see if it's lower than 1, we use "lower", and here we see that the significance level for the null hypothesis is much smaller than 0.05, so we reject the null hypothesis, and accept the alternative. This means that even though the median day return is positive, the median whole day return isn't. Combining with our previous finding, an example would be, if the market closed yesterday at $10, and opened today at $5, the market will close today higher than $5 but lower than $10. This could mean that during the after hours/premarket, they tend to overreact to bad news and push it lower than it needs to be, but the premarket gets the direction correct, and the present day closing price will still be lower than yesterday's closing price.
From this exercise we can conclude,
This shouldn't be a suggestion to actually buy at close and sell at market open every day, since I didn't facotor in will be taxes and transaction fees, but it is nonetheless interesting to see the dynamic of the market. Also note that this analysis is just using the price data of SPY from 1993 to 2021, and might not reflect the true nature of every other stock and ETF across different regions and time period. The academic paper that I linked above went into more details exploring this phenomenon.
Schrodinger develops a platform for drug discovery, either collaborates or services with all top 20 pharmaceutical company, while also developing their own internal drug programs. They generate revenue through two streams, “software” and “drug discovery”.
Before diving into the specifics one need to understand the process of drug discovery. The illustration below does a good job explaining.
Drug discovery is a length process and one that takes a long time and mostly ends with failure. 66% of all programs never succeed in delivering an IND (Investigational New Drug, when FDA gives permission to start human trials). An average of 13.8% succession rate in the clinical trials, while oncology succession rate is 3.4%. Here we are going to do a crash course for the Drug Discovery part of the illustration above since this is the major differentiator with Schrodinger.
In the Target Discovery known also as “Hit Discovery”, is to find the biological origin of a disease, and finding a target (ex a protein) that is “drugable”. Then one need to demonstrate that the “drugable” effects on the target can actually be a therapeutic. Then Lead Discovery also known as “Hit to Lead”, where the target is gone through processes to identify molecules that can interact with the target to produce desired biological effects. Then lastly with Lead Optimization, the leads previously found are synthesized and improve potency, reduced off target variables… Then voila you have a drug development candidate. This typical process using traditional drug design takes around 4 to 6 years, with each step taking from 2 to 3 years.
A very misleading fact got me very excited at first, it was the spending of R&D in pharmaceutical industry is 188 billion in 2020 and forecasted to be 233 billion in 2026, CAGR of 3%1. Seeing this huge market potential with their current revenue of around 80 million, I thought there was going to be huge growth potential. I was wrong. Upon researching I found this site https://www.baybridgebio.com/drug_pricing_calc.html which uses average R&D cost to determine price of a drug. He uses , Paul et al Nature Reviews Drug Discovery 2010 and DiMasi et al, Journal of Health Economics 2016 to calculate average cost of R&D, then adjusting difference from 2010 to 2019. Using his averages, the average cost of Drug Discovery is just 28.4 Million, where the total average cost of Discovery and Development is 562 Million, since later in the R&D process the more labour and capital intensive it is. This would effectively mean that the drug discovery part of total R&D spending is just 5%. However, I have also found an industry overview by Dr Gerhard Goldbeck, titled “The scientific software industry: a general overview” which he states that the scientific software industry only gets to 0.001% of all R&D spending. This overview was published in 2017, and stating the current condition, as he said the total market is 100M but the fact is Schrodinger just had 80M in software revenue, and they don’t not have 80% of the market, so I think it is safe to say that his doesn’t reflect the future. I don’t believe that 0.001% is the TAM, so I decided to take the average of the two and the revised percentage being 2.5%.
Starting with the software part of Schrodinger’s business. It is a platform that help pharmaceutical companies do drug discovery. Compared to manual drug discovery this method is cheaper and faster.
Due to this technology not employed long enough, there isn’t enough data to show statistical significance that the drug discovery has a higher chance of success. But I do believe with the nature of ML and a progressively larger library of data that there is indeed a higher chance of success. That is why all top 20 pharmaceutical companies all use Schrodinger and contract growth rate is increasing since 2013 at CAGR of 16%. I believe, conservatively, as this technology gets proven over time Schrodinger can mange to have a 1B revenue in in the software segment alone. Which translate to a total market share of just 15% percent of the market. I prefer to have a conservative estimate, as I am not in this field and I am uncertain of the market size. Another interesting thing to note of this software segment, is that there are more then 1350 academic institution that uses them to do research. This can cause graduate and PhD student to be familiar with the software and if they enter the work force they would have a bias towards this software. Schrodinger also uses its software in material design, but this is only small portion of revenues so I am not going to focus on them too much.
Considering that the software is the bread and butter of Schrodinger, then Drug Discovery is the growth engine. This is where Schrodinger uses their own software, to discover their own preclinical drug and collaborate with other pharmaceutical companies. For collaborations Schrodinger gets paid single digit royalty for every drug that gets to commercialization and gets paid if reaches certain milestones. As an average cancer drug would produce 12.3B over its lifetime, and they have already 20 collaborative pipeline with other drug manufactures. Which I can take the expected value of this transaction, using the probability of success given the stage at the drug development process and using 5% as the amount of royalties. Over these 20 collaborative pipeline the expected value of the royalties discounted back at the cost of capital, I get a 536M. This method is not the cleanest, but this is a way to estimate royalties. Note that if Schrodinger’s method actually improves the success rate of these drugs, then it would also increase this expected value. For the 5 pipeline that are wholly owned by Schrodinger, will either produce it themselves, or out licencing it. This drug discovery part starts to become a flywheel, where internal pipeline will help them refine the software, and the more success in the internal and collaborative pipeline will attract more customers, and with more customer the more data and feedback so the software gets better. This signals the asymmetric upside.
Other things to note, Bill gates and David Shaw is one of the original investors in this company. Everyone knows Bill Gates but might not know David Shaw. David Shaw was a computer scientist, got a PhD in AI from Stanford, started one of the first quant hedge fund ever, and now turned to full-time scientific research in computational biochemistry. This is the definition of smart money. Another point more then half of their employees have a PhD. All the mean while Schrodinger has a Glassdoor rating of 4.6, while the average rating for a company on the Glassdoor “Best Places to Work” is 4.3. The average overall rating for Glassdoor companies is 3.3. This shows that the company has a good culture.
Finally, this brings us to the valuation. Where I used valuation by parts breaking down the software and drug discovery segment. Where I projected that by 2030 the software business would generate close to 1B in revenue, and would have an operating margin of 20%, with a cost of capital 6.34% and a sales to capital ratio of 1.1.
For the drug discovery segment, I found the expected value of the 20 collaborative pipeline, totalling to 536M. With the five internally-owned pipeline, I did the same expected value discounted back at cost of capital, and subtracting the cost of development, resulting in 1.3B. Adding all this up we get to a price of 62 USD.
Couple important note on this valuation. First, the only “forecast” I did was for the software side, so this valuation doesn’t incorporate future growth in the drug discovery segment. This can be considered as a call option. Especially with the flywheel effect that Schrodinger has, I believe the true intrinsic value to be much higher. Second, I do realize that this is a very rough valuation, with multiple short cuts with expected value, but as the saying go, you don’t have to be right, you just have to be less wrong. I entered in the position at 56 USD.
1 https://www.statista.com/statistics/309466/global-r-and-d-expenditure-for-pharmaceuticals/
MAXR operates in two segments, Earth Intelligence (They take pictures of the world, and do data processing and sell them to government and commercial industry. The other called space infrastructure ( satellite manufacturing and space robotics )
Earth Intelligence: This their bread and butter where they have a high margin business as they are currently the world leaders in this space, the google map satellite picture you see it's them that took it. Their main customer is primarily the government but they are currently diversifying their revenue stream to more commercial players. They are going to launch the WorldView Legion in 2021 where they would have a higher revisit rate and much better resolution. This would cause a much lower CapEx and a higher growth rate. Compare to low earth orbit (LEO) satellites they have much better resolution and a comparable revisit rate. They have an option to expand, with all these proprietory data they can add a lot of value with software eg automated cars to map out a better version of the map, they can help farmers track vegetation growth, calculate CO2 emissions...
Space Infrastructure: This is the money-losing side of the company, currently LEO satellite is all the rage due to reduced cost for space travel and large investments backed by VC and the lower cost, so as a company that specializes in geosynchronous (GEO) satellite they have suffered quite a bit as you see below.
However as the CEO said in a 2Q19 Investor call, they are currently reshaping space solutions to be able to break even with just 2 GEO contracts per year. As the GEO orders have bottomed I believe if there is ever a chance with GEO coming back to favor, they are in the perfect position to benefit from the trend. As LEOs have not exactly been fully proven as seen from OneWeb, although Starlink could change that, only time will tell. The FCC actions to clear the C-Band spectrum for future 5G applications in August will require new GEO satellites to cover U.S. territories. MAXR has already won 5 out of 7 contracts. The full life cycle of a GEO is 15 years so as GEO starts to be worn out, there will be a chance of replacement and hence more orders. MAXR has also poured a lot of R&D into LEOs to try to compete with Telesat LEO Constellation, however, they have dropped out of the competition against Thales and Airbus. That doesn't mean they will give up on LEOs as the CEO said in a recent interview, he is trying to play both sides, so it doesn't matter which way the market leans towards.
There is risk attributed to the launch of WorldView legion, the risk of Starlink taking over the GEO market, difficulties with financing new debt.
I have valued this company at 40$ so I entered in at 27.7$.
MMAC is a financial service company that mainly focuses on lending to late-stage solar infrastructure projects. These loans are an average of $19MM. With an average coupon around 10% to 12%. It takes years to complete and get approval for these solar projects so they are heavily incentives to finish these projects. As solar gets cheaper there is the headwind for more installations. They have grown their book value and earning proportionally as the solar construction lending business scales. They also have racked up 400M of NOL from their previous business in real estate and after the GFC.
MMAC is trading under its book value. Its book value of 218 M while it is trading at 148 M at a 32% discount. Since this is a financial service company, the book value would be the most accurate measure of the company. Since finding the book value of loans is much easier than find the book value of a tech company. A reason for this to be trading under book value would happen if the return on equity (ROE) is below the cost of equity (COE). However from my calculation for the past five years has an average ROE of 31% and the current COE is 11%. So I don't see the reason for the undervaluation. As they continue to buy back stock and continue to grow both book value and earning I am confident they would come back to trade near book value. At the end of my valuation, I have estimated to trade close to its book value at around 34$, so I've entered the position at 24.4$
My not so profound valuations and observations on Markets