Friday, August 11, 2017

A Tesla 2017 Update: A Disruptive Force and a Debt Puzzle!

These are certainly exciting times for Tesla. The first production version of the Tesla 3 was unveiled on July 28, with few surprises on the details, but plenty of good reviews. Elon Musk was his usual self, alternating between celebrating success and warning investors in the stock that the company was approaching "manufacturing hell", as it ramps up its production schedule to meet its target of producing 10,000 cars a week. It is perhaps to cover the cash burn in manufacturing hell that Tesla also announced that it planned to raise $1.5 billion in a junk bond offering. Investors continued to be unfazed by the negative and lapped up the positive, as the stock price soared to $365 at close of trading on August 9, 2017. With all of this happening, it is time for me to revisit my Tesla valuation, last updated in July 2016, and incorporate, as best as I can, what I have learned about the company since then.

Tesla: The Story Stock
I have been following Tesla for a few years and rather than revisit the entire history, let me go back to just my most recent post on the company in July 2016, where I called Tesla the ultimate story stock. I argued that wide differences between investors on what Tesla is worth can be traced to divergent story lines on the stock. I used the picture below to illustrate the story choices when it comes to Tesla, and how those choices affected the inputs into the valuation.

In that post, I also traced out the effect of story choices on value, by estimating how the numbers vary, depending upon the business, focus and competitive edge that you saw Tesla having in the future:

With my base case story of Tesla being an auto/tech company with revenues pushing towards mass market levels and margins resembling those of tech companies, I estimated a value of about $151 a share for the company and my best case estimate of value was $316.46.

Tesla: Operating Update
If you are invested in or have been following Tesla for the last year, you are certainly aware that the market has blown through my best case scenario, with the stock trading on August 9, at $365 a share, completing a triumphant year in markets:

As Tesla's stock price rose, it broke through milestones that guaranteed it publicity along the way. It's market capitalization exceeded that of Ford and General Motors in April 2016, and in June 2016, Tesla leapfrogged BMW to become the fourth largest market cap automaker in the world, though it has dropped back a little since. It now ranks fifth, in market capitalization, among global automobile companies:
Largest Auto Companies (Market Capitalization) on August 9, 2017
While Tesla's market cap has caught up with larger and more established auto makers, its production and revenues are a fraction of theirs, leading some to use metrics like enterprise value per car sold to conclude that Tesla is massively over valued. I don't have much faith in these pricing metrics to begin with, but even less so when comparing a company with massive potential to companies that are in decline, as I think many of the conventional auto companies in this table are currently.

As I noted at the start of the post, it has been an eventful year for Tesla, with the completion of the Solar City acquisition, and the Tesla 3 dominating news, and its financial results reflect its changes as a company. In the twelve months ended June 30, 2017, Tesla's revenues hit $10.07 billion, up from $7 billion in its most recent fiscal year, which ended on December 31, 3016; on an annualized basis, that translates into a revenue growth rate of 107%. That positive news, though, has to be offset at least partially with the bad news, which is that the company continues to lose money, reporting an operating loss of $638 million in the most recent 12 months, with R&D expensed, and a loss of $103 million, with capitalized R&D. The growth in the company can be seen by looking at how quickly its operations have scaled up, over the last few years:

Tesla's growth has not just been in the operating numbers but in its influence on the automobile sector. While it was initially dismissed by the other automobile companies as a newcomer that would learn the facts of life in the sector, as it aged, the reverse has occurred. It is the conventional automobile companies that are, slowly but surely, coming to the recognition that Tesla has changed their long-standing business. Volvo, a Swedish automaker not known for its flair, announced recently that all of its cars would be either electric or hybrid by 2019, and Ford's CEO was displaced for not being more future oriented. A little more than a decade after it burst on to the scene, it is a testimonial to Elon Musk that he has started the disruption of one of the most tradition-bound sectors in business.

Tesla: Valuation Update
The production hiccups notwithstanding, the company continues to move towards production of the Tesla 3, with the delivery of the handful to start the process. There is much that needs to be done, but I consider it a good sign that the company sees a manufacturing crunch approaching, since I would be concerned if they were to claim that they could ramp up production from 94,000 to 500,000 cars effortlessly.  My updated story for Tesla is close to the story that I was telling in July 2016, with two minor changes. The first is that the production models of the Tesla 3 confirm that the company is capable of delivering a car that can appeal to a much broader market than prior models, putting it on a  pathway to higher revenues. My expected revenues for Tesla in ten years are close to $93 billion, a nine-fold increase from last year's revenues and a higher target than the $81 billion that I projected in my July 2016 valuation. Second, the operating margins, while still negative, have become less so in the most recent period, reducing reinvestment needs for funding growth. The free cash flows are still negative for the next seven years, a cash burn that will require about $15.5 billion in new capital infusions over that period. With those changes, the value per share that I estimate is about $192/share, about 20% higher than my $151 estimate a year ago, but well below the current price per share of $365.
Download spreadsheet
As with every Tesla valuation that I have done, I am sure (and I hope) that you will disagree with me, with some finding me way too pessimistic about Tesla's future, and others, much too optimistic. As always, rather than tell me what you think I am getting wrong, I would encourage you to download the spreadsheet and replace my assumption with yours. I think I am being clear eyed about the challenges that Tesla will face along the way and here are the top three: 
  1. Can Tesla sell millions of cars? One of Tesla's accomplishments has been exposing the potential of the hybrid/electric car market, even in an era of restrained fuel prices. That is good news for Tesla, but it has also woken up the established automobile companies, as is evidenced by not only the news from Volvo and Ford, but also in increased activity on this front at the other automobile companies. In my valuation, the revenues that I project in 2027 will require Tesla to sell close to 2 million cars, in the face of increased competition.
  2. Can it make millions of cars? Tesla's current production capacity is constrained and there are two production tests that Tesla has to meet. The first is timing, since the Tesla 3 deliveries have been promised for the middle of 2018, and the assembly lines have to be humming by then. The second is cost, since a subtext of the Tesla story, reinforced by hints from Elon Musk, is that the company has found new and innovative ways of scaling up production quickly and at much lower costs than conventional automobile companies. 
  3. Can it generate double digit margins? In my valuation, I assume an operating margin of 12% for Tesla, almost double the average of 6.33% for global auto companies. For Tesla to generate this higher margin, it has to be able to keep production costs low at its existing and new assembly plants and to be able to charge a premium price for its automobiles, perhaps because of its brand name. 
Tesla has shown a capacity to attract and keep customers and I think it is more than capable of meeting the first challenge, i.e., sell millions of cars, especially since its competition is saddled with legacy costs and image problems. It is the production challenge that is the more daunting one, simply because this has always been Tesla's weakest link. Over the last few years, Tesla has consistently had trouble meeting logistical and delivery targets it has set for itself, and those targets will only get more daunting in the years to come. Furthermore, if its production costs run above expectations, it will be unable to deliver on higher margins. To succeed, Tesla will require vision, focus and operating discipline. With Elon Musk at its helm, the company will never lack vision, but as I argued in my July 2016 post, Mr. Musk may need a chief operating officer at his side to take care of delivery deadlines and supply chains. 

Financing Cash Burn: Tesla's Odd Choice
There is much to admire in the Tesla story but there is one aspect of the story that I find puzzling, and if I were an equity investor, troubling. It is the way in which Tesla has chosen to, and continues to, finance itself. Over the last decade, as Tesla has grown, it has needed substantial capital to finance its growth. That is neither surprising nor unexpected, since cash burn is part of the pathway to glory for companies like Tesla. However, Tesla has chosen to fund its growth with large debt issues, as can be seen in the graph below:

That debt load, already high, given Tesla’s operating cash flows is likely to get even bigger if Tesla succeeds in its newest debt issue of $1.5 billion, which it is hoping to place with an interest rate of 5.25%, trying to woo bond buyers with the same pitch of growth and hope that has been so attractive to equity markets. That suggests that those making the pitch either do not understand how bonds work (that bondholders don't get to share much in upside but share fully in the downside) or are convinced that there are enough naive bond buyers out there, who think that interest payments can be made with potential and promise.

But setting aside concerns about bondholders, the debt issuance makes even less sense from Tesla's perspective. Unlike some, I don’t have a kneejerk opposition to the use of debt. In fact, given that the tax code is tilted to benefit debt, it does make sense for many companies to use debt instead of equity. The trade off, though, is a simple one:

If you look at the trade off, you can see quickly that Tesla is singularly unsuited to using debt. It is a company that is not only still losing money but has carried forward losses of close to $4.3 billion, effectively nullifying any tax benefits from debt for the near future (by my estimates, at least seven years). With Elon Musk, the largest stockholder at the company, at the helm, there is no basis for the argument that debt will make managers more disciplined in their investment decisions. While the benefits from debt are low to non-existent, the costs are immense. The company is still young and losing money, and adding a contractual commitment to make interest payments on top of all of the other capital needs that the company has, strikes me as imprudent, with the possibility that one bad year could put its promise at risk. Finally, in a company like Tesla, making large and risky bets in new businesses, the chasm between lenders and equity investors is wide, and lenders will either impose restrictions on the company or price in their fears (as higher interest rates). So, why is Tesla borrowing money? I can think of two reasons and neither reflects well on the finance group at Tesla or the bankers who are providing it with advice.
  1. The Dilution Bogeyman: The first is that the company or its investment bankers are so terrified of dilution, that a stock issue is not even on the table. Once the dilution bogeyman enters the decision process, any increase in share count for a company is viewed as bad, and you will do everything in your power to prevent that from happening, even if it means driving the company into bankruptcy. 
  2. Inertia: Auto companies have generally borrowed money to fund assembly plants and the bankers may be reading the capital raising recipe from that same cookbook for Tesla. That is incongruent with Elon Musk’s own story of Tesla as a company that is more technology than automobile and one that plans to change the way the auto business is run.
Tesla’s strengths are vision and potential and while equity investors will accept these as down payments for cash flows in the future, lenders will not and should not. In fact, I cannot think of a better case of a company that is positioned to raise fresh equity to fund growth than Tesla, a company that equity investors love and have shown that love by pushing stock prices to record highs. Issuing shares to fund investment needs will increase the share count at Tesla by about 3-4% (which is what you would expect to see with a $1.5 billion equity issue) but that is a far better choice than borrowing the money and binding yourself to make interest payments.  There will be a time and a place for Tesla to borrow money, later in its life cycle, but that time and place is not now. If Tesla is dead set on not raising its share count, there is perhaps one way in which Tesla may be able to eat its cake and have it too, and that is to exploit the dilution bogeyman's blind spot, which is a willingness to overlook potential dilution (from the issuance of convertibles and options). In fact, why not issue long term, really low coupon convertible bonds, very similar to this one from 2014, a bond only in name since almost all of its value came from the conversion option (which is equity with delayed dilution)?


The Tesla story continues to evolve, and there is much in the story that I like. It is changing the automobile business, a feat in itself, and it is starting to deliver on its production promises. The next year may be manufacturing hell, but if the company can make its through that hell and find ways to deliver the tens of thousands of Tesla 3s that it has committed to delivering, it will be well on its way. I still find the stock to be too richly priced, even given its promise and potential, for my liking, but I understand that many of you may disagree. That said, though, I do think that the company's decision to use debt to fund its operations makes no sense, given where it is in the life cycle.

YouTube Video

Previous Blog Posts
  1. Tesla: It's a story stock, but what's the story? (July 2016)

Spreadsheet Attachments
  1. Tesla Valuation: August 2017
  2. Tesla Valuation: July 2016

Tuesday, August 1, 2017

The Crypto Currency Debate: Future of Money or Speculative Hype?

When it comes to any finance-related questions, I am fair game, and those questions usually span the spectrum, from what I think about Warren Buffett (or why I don't agree with everything he says) to whether tech stocks are in a bubble (a perennial question for worry warts). In the last few months, though, I have noticed that I have been getting more and more questions about crypto currencies, especially Bitcoin and Ether, and whether the price surges we have seen in these currencies are merited. While I have an old post on bitcoin, I have generally held back from talking about crypto currencies in this blog or in my other teaching for two reasons. First, I find that any conversation about bitcoin quickly devolves into an argument rather than a discussion, since both proponents and critics tend to hold strong views on its use (or uselessness). Second, I find that some of the technical underpinnings of bitcoin, ether and other cryptocurrencies are beyond my limited understanding of block chains and technology and I risk saying something incredibly ill informed. While both reasons still persist, I am going to throw caution to the winds and put down my thoughts about the rise, the mechanics and the future, at least as I see it, of crypto currencies in this post.

The Market Boom
Any discussion of crypto currencies has to start with the recognition that the experiment is still young.  Satoshi Nakamoto's paper on bitcoin was made public in October 2008 and implemented as open source in January 2009. Less than ten years later, the market capitalization of bitcoin alone is in excess of $40 billion and the success story, at least in terms of bitcoin as an investment, can be seen in the graph below:

The initial rise could have been a flash in the pan, a fad attracting speculators, but in the last two years, Bitcoin seems to have found new fans, as can be seen below:

Bitcoin's success, at least in the financial markets, has attracted a host of competitors, with Ethereum (Ether) being the most successful. Ether's rise in market price, since its introduction in 2015 has been even more precipitous that Bitcoin's, though it has pulled back in recent weeks:

The list of crypto currencies gets added to, by the day, with a complete list available here, with the market caps of each (in US dollars) listed. At least from a market perspective, there is no doubting the fact that crypto currencies have arrived, and enriched a lot of people along the way.

The Mechanics 
While the crypto currencies emphasize their differences, the most successful ones share a base architecture, the block chain. A block chain is a shared digital ledger of transactions in an asset where the validation of transactions is decentralized. I know that sounds mystical, but the picture below (using bitcoin to illustrate) should provide a better sense of what's involved:

The key features of a block chain are:
  1. Decentralized verification: The validation and verification of a transaction is sourced to members, called miners in the crypto currency world. Verification usually involves trying different algorithms (hashes) to find the unique one that matches the transaction block, and the successful miner is rewarded, currently with the crypto currency. At least, as I understand it, this process requires more brute force (powerful processors trying different algorithms before you find a match) than intellectual firepower.
  2. Complete and open records: Every transaction, once validated and verified, is converted into a block of data that is recorded in the block chain ledger, which is accessible to everyone in the network. If you are worried about privacy, the transaction records do not include personal data but take the form of encrypted data (hashes).
  3. Incorruptible: A block chain, once recorded and shared, cannot be changed since those changes are visible to everyone in the network and are quickly tagged as fraudulent. Thus, the ledger, once created, becomes almost incorruptible.
In effect, a block chain is a digital intermediation process where transactions are checked by members of the network, and recorded, and once that is done, cannot be altered fraudulently. As you can see from its description, the block chain technology is about far more than crypto currencies. It can be used to record transactions in any asset, from securities in financial markets to physical assets like houses, and do so in a way that replaces the existing intermediaries with decentralized models. It should come as no surprise that banks and stock exchanges, which make the bulk of their money from intermediation, not only see block chains as a threat to their existence but have been early investors in the technology, hoping to co-opt it to their own needs.  

The Currency Question
If you define success as a rise in market capitalization and popular interest, crypto currencies have clearly succeeded, perhaps more quickly than its original proponents ever expected it to. But the long term success of any crypto currency has to answer a different question, which is whether it is a "good" currency.  Harking back to Money 101, you measure a currency's standing by looking at how well it delivers on its three purposes:
  1. Unit of account: A key role for a currency is to operate as a unit of account, allowing you to value not just assets and liabilities, but also goods and services. To be effective as a unit of account, a currency has to be fungible (one unit of the currency is identical to any other unit), divisible and countable. 
  2. Medium of exchange: Currencies exist to make transactions possible, and this is best accomplished if the currency in question is easily accessible and transportable, and is accepted by buyers and sellers as legal tender. The latter will occur only if people trust that the currency will maintain its value and if transactions costs are low.
  3. Store of value: To the extent that you hold some or all of your wealth in a currency, you want to feel secure about leaving it in that currency, knowing that it will not lose its buying power while stored.  
Given these requirements, you can see why there are no perfect currencies and why every currency has to measured on a  continuum from good to bad. Broadly speaking, currencies can take one of three forms, a physical asset (gold, silver, diamonds, shells), a fiat currency (usually taking the form of paper and coins, backed by a government) and crypto currencies. Gold's long tenure as a currency can be attributed to its strength as a store of value, arising from its natural scarcity and durability, though it falls short of fiat currencies, in terms of convenience and acceptance, both as a unit of account and as medium of exchanges. Fiat currencies are backed by sovereign governments and consequently can vary in quality as currencies, depending upon the trust that we have in the issuing governments. Without trust, fiat currency is just paper, and there are some fiat currencies where that paper can become close to worthless.  For crypto currencies, the question then becomes how well they deliver on each of the purposes. As units of account, there is no reason to doubt that they can function, since they are fungible, divisible and countable. The weakest link in crypto currencies has been their failure to make deeper inroads as mediums of exchange or as stores of value. Using Bitcoin, to illustrate, it is disappointing that so few retailers still accept it as payment for goods and services. Even the much hyped successes, such as Overstock and Microsoft accepting Bitcoin is illusory, since they do so on limited items, and only with an intermediary who converts the bitcoin into US dollars for them. I certainly would not embark on a long or short trip away from home today, with just bitcoins in my pocket, nor would I be willing to convert all of my liquid savings into bitcoin or any other crypto currency. Would you?

So, why has crypto currency not seen wider acceptance in transactions? There are a few reasons, some of which are more benign than others:
  1. Inertia: Fiat currencies have a had a long run, and it is not surprising that for many people, currency is physical and takes the form of government issued paper and coins. While people may use credit cards and Apple Pay, their thinking is still framed by the past, and it may take a while, especially for older consumers and retailers, to accept a digital currency. That said, the speed with which consumers have adapted to ride sharing services and taken to social media suggests that inertia cannot be the dominant reason holding back the acceptance of crypto currencies.
  2. Price volatility: Crypto currencies have seen and continue to see wild swings in prices, not a bad characteristic in a traded asset but definitely not a good one in a currency. A retailer or  service provider who prices his or her goods and services in bitcoin will constantly have to reset the price and consumers have little certitude of how much the bitcoin in their wallers will buy a few hours from now.
  3. Competing crypto currencies: The crypto currency game is still young and the competing players each claim to have found the "magic bullet" for eventual acceptance. As technologies and tastes evolve, you will see a thinning of the herd, where buyers and sellers will pick  winners, perhaps from the current list or maybe something new. It is possible that until this happens, transactors will hold up, for fear of backing the wrong horse in the race.
Ultimately, though, I lay some of the blame on the creators of the crypto currencies, for their failure, at least so far, on the transactions front. As I look at the design and listen to the debate about the future of crypto currencies, it seems to me that the focus on marketing crypto currencies has not been on transactors, but on traders in the currency, and it remains an unpleasant reality that what makes crypto currencies so attractive to traders (the wild swings in price, the unpredictability, the excitement) make them unacceptable to transactors. 

The Disconnect
You can see the disconnect in how crypto currencies have been greeted, by contrasting the rousing reception that markets have given them with the arms length at which they have been held by merchandisers and consumers. In the graph below, I focus on the divergence between the market price rise of bitcoin and the increase in the number of transactions involving bitcoin:

While the price of bitcoin has increase more than a thousand fold, since the start of 2012, the number of transactions involving bitcoin was only about thirty two times larger in July 2017 than what it was at the start of 2012. In my view, there are three possible explanations for the divergence, and they are not mutually exclusive:
  1. Markets are forward looking: If you are a believer in crypto currencies, the most optimistic explanation is that markets are forward looking and that the rise in the prices of Bitcoin and Ether reflects market expectations that they will succeed as currencies, if not right away, in the near future. 
  2. Speculative asset: I am second to none in having faith in markets, but there is a simpler and perhaps better explanation for the frenzied price movements in crypto currencies. I have long drawn a distinction between the value game (where you try to attach a value to an asset based upon fundamentals) and the pricing game, where mood and momentum drive the process. I would argue, based upon my limited observations of the crypto currency markets, that these are pure pricing games, where fundamentals have been long since forgotten. If you don't believe me, visit one of the forums where traders in these markets converse and take note of how little talk there is about fundamentals and how much there is about trading indicators.
  3. Loss of trust in centralized authorities (governments & central banks): There can be no denying that the creators of Bitcoin and Ether were trying to draw as much inspiration for their design from gold, as they were from fiat currencies. Thus, you have miners in crypto currency markets who do their own version of prospecting when validating transactions and are rewarded with the currency in question. For ages, gold has held a special place in the currency continuum, often being the asset of last resort for people who have lost faith in fiat currencies, either because they don't trust the governments backing them or because of debasement (high inflation). While gold will continue to play this role, I believe that for some people (especially younger and more technologically inclined), bitcoin and ether are playing the same role. As surveys continue to show depleting trust in centralized authorities (governments and central banks), you may see more money flow into crypto currencies. 
The analogy between gold and crypto currency has one weak link. Gold has held its value through the centuries and is a physical asset. For better or worse, it is unlikely that we will decide a few years from now that gold is worthless. A crypto currency that few people use as currency ultimately will not be able to sustain itself, as shiner and newer versions of it pop up. Ironically, if traders in bitcoin and ether want their investments in the crypto currencies to hold their value, the currencies have to become less exciting and lucrative as investments, and become more accepted as currencies. Since that will not happen by accident, I would suggest that the winning crypto currency or currencies will share the following characteristics;
  1. Transaction, not trading, talk: From creators and proponents of the currency, you will hear less talk about how much money you would make by buying and selling the currency and more on its efficacy in transactions.
  2. Transaction, not trading, features: The design of the crypto currency will focus on creating features that make it attractive as a currency (for transactions), not as investments. Thus, if you are going to impose a cap (either rigid like Bitcoin or more flexible, as with other currencies), you need to explain to transactors, not traders, why the cap makes sense. 
  3. Trust in something: I know that we live in an age where trust is a scarce resource and I argued that that the growth in crypto currencies can be attributed, at least partly, to this loss of trust. That said, to be effective as a currency, you do need to be able to trust in something and perhaps accept compromises on privacy and centralized authority (at least on some dimensions of the currency). 
It is also worth noting that the real tests for crypto currencies will occur when they reach their caps (fixed or flexible). After all, bitcoin and ether miners have been willing to put in the effort to validate transactions because they are rewarded with issues of the currency, feasible now because there is slack in the currency (the current number is below the cap). As the cap becomes a binding constraint, the rewards from miners have to come from transactions costs and serious thought has to go into currency design to keep these costs low. Hand waving and claiming that technological advances will allow this happen are not enough. I know that there are many in the crypto currency world who recognize this challenge, but for the moment, their voices are being drowned out by traders in the currency and that is not a good sign.

If you expected a valuation of bitcoin or ether in this post, you are probably disappointed by it, but here is a simple metric that you could use to determine whether the prices for crypto currencies are "fair". Currencies are priced relative to each other (exchange rates) and there is no reason why the rules that apply to fiat currencies cannot be extended to crypto currencies. A fair exchange rate between two fiat currencies will be on that equalizes their purchasing power, an old, imperfect and powerful theorem. Consequently, the question that you would need to address, if you are paying $2,775 for a bitcoin on August 1, 2017, is whether you can (or even will be able to) but $2,775 worth of goods and services with that bitcoin. If you believe that bitcoin will eventually get wide acceptance as a digital currency, you may be able to justify that price, especially because there is a hard cap on bitcoin, but if you don't believe that bitcoin will ever acquire wide acceptance in transactions, it is time that you were honest with yourself and recognized that is just a lucrative, but dangerous, pricing game with no good ending.

Crypto currencies, with bitcoin and ether leading the pack, have succeeded in financial markets by attracting investors, and in the public discourse by garnering attention, but they have not succeeded (yet) as currencies. I believe that there will be one or more digital currencies competing with fiat currencies for transactions, sooner rather than later, but I am hard pressed to find a winner on the current list, right now, but that could change if the proponents and designers of one of the currencies starts thinking less about it as a speculative asset and more as a transaction medium, and acting accordingly. If that does not happen, we will have to wait for a fresh entrant and the most enduring part of this phase in markets may be the block chain and not the currencies themselves.

YouTube Video

Thursday, July 20, 2017

Online Teaching: Promise, Pitfalls and Potential!

I am a teacher. That is how I describe myself to anyone who chooses to ask me what I do for a living. I am not a professor (sounds pedantic and pompous), definitely not an academic (how boring is that..) and don't consider myself anything more than a dilettante on almost every topic that I hold forth on. It is in pursuit of my teaching mission that I have put my regular classes online for most of the last two decades, though technology has made that sharing easier. For those of you who have read my postings before, I usually announce a few weeks ahead of every semester, the classes that I will be teaching at Stern, what each class is about and how you can access it, as I did in January with my Spring 2017 valuation and corporate finance classes. As September 2018 approaches, I was going to skip that ritual, since I will be on sabbatical next year (and if you have no idea what a sabbatical is, more on that later..) but I will be teaching, nevertheless, during the year.

Online Education
I still remember the first semester that I shared a class with an online audience was in the 1990s, when the internet was still in its infancy, we were still using dial-up modems and phones were connected to landlines. I recorded my regular classes using a VHS camcorder onto tapes, and then converted the tapes into videos of woeful quality, but with passable audio. I posted these online, but with only minimal additional material, since sharing was both time consuming and difficult to do. Needless to say, the internet has grown up and made sharing much easier, with class recordings now being made with built-in cameras in classrooms and converted to high quality videos quickly, to be watched on tablets on smart phones. Here, for instance, is my entire Spring 2017 valuation class, with links to the videos as well as almost every scrap of material that I provide for the class and even the emails I sent to the class.

I have long believed that the traditional university model not only is ripe for, but is deserving of , disruption, saddled with legacy costs and a muddled mission. That said, the attempts by online education to upend the university model have, for the most part, had only marginal success and it is in trying to answer why that I started thinking about how we teach, and learn online. In particular, online classes have proved a imperfect substitutes for regular classes due to three shortcomings:
  1. No personal touch: This may be a reflection of my age, but there is a difference between being in a live  and watching a video of the same class, no matter how well it is recorded and presented. 
  2. No interaction: We forget how much of the learning in a classroom comes, not from lectures, but from interaction, not just between the teacher and students but between students, often in informal and serendipitous exchanges. With online education, the interaction, if it exists, is highly formalized and there is less learning.
  3. Tough to stay disciplined: When you were in college, and enrolled for an 8.30 am class, did you feel like not going to class? I certainly did, but what kept me going was the fact that my absence would be noticed, not just by the professor, but by other students in the class. In fact, it is that group pressure and class structure that keeps us focused on project deadlines and exam dates, with regular classes. With online classes, that discipline has to come from within, and it should be therefore no surprised that most people who start online classes never finish them
It is perhaps easiest to see the challenges and limits of online teaching by looking at what it is that makes for a good class, in person or online. In my view, the measure of good teaching is that students don't get just content (tools, techniques, models) but that they learn how to create their own content, i.e., the capacity to devise their own tools to meet their needs. In the context of a regular class, you use readings, problem sets, quizzes and exams to deliver the former (content) but the latter (learning) requires a more complex mix of classroom and informal interaction, real life projects and intellectual curiosity (and I believe that it is partly a teacher's responsibility to evoke that). The time schedule of a regular class also puts limits on how much students can procrastinate, and peer pressure, from others taking the class or working with you on assignments, serves to keep most on task. 

With this framework, the challenges of teaching online become clear. You have to find ways to keep students engaged, disciplined and interactive, and you have to do it online. While there are technical solutions to each one of these challenges (great videos for engagement, a time schedule and online exams for discipline, and discussion boards for interaction), and we have come a long way in the last few years, there is still a great deal of work to be done.  

Online Classes: My learning curve
My search for a better way of delivering what I teach online started about five years ago, with a simple first step. I decided to try to take each of my regular lectures, which go for 80 minutes, and see if I could compress it into a 10-12 minute slot and the results were both revealing and humbling. It was not that difficult to compress my classes, a testimonial to how much buffer I build into my regular classes to ramble and pontificate. (If you have been in one of my regular classes or watched one, you probably know that there is nothing I enjoy more than going off on a riff on a topic or news story and I think you need a few of these in a 80-minute class to keep your class engaged.) I also started developing short post-session quizzes with solutions that someone watching the class could take, to check on whether they were "getting" the session material. I organized and sequenced the sessions and you can click to see the online versions of my corporate finance, valuation and investment philosophy classes. 

I was under no illusions that I had unlocked the key to online learning with these classes, and these classes had significant limitations. First, packing material densely into 10-12 minute chunks can make watching even these short sessions taxing. Second, the videos that I made (with the help of a friend who was a camera man) were lacking in bells and whistles, basic talking-head videos with slides in the background. Third, there is no personal touch or interaction, since the videos are recorded. Finally, given the number of people in each of these classes, there was no way for me to give and grade exams, look over valuations or corporate financial analysis (a key ingredients of my regular classes) or provide certification that someone had taken the class.

Valuation Certificate Class
Just over a year ago, the Stern School of Business, which is where I teach, asked me whether I would be willing to teach an online certificate class. My initial response was to say no for two reasons. First, universities always seem to operate at deficits, no matter how much revenue they collect from tuition, and I knew that Stern would extract its pound of flesh from those who took the certificate. Second, I was concerned that if I did do a certificate class, and it became a money generator, that I would be asked to remove my free online classes. Stern must have wanted to do this certificate really badly since they offered to leave my online material untouched, if I agreed to work on the certificate course. It was this assurance, in conjunction with the opportunity to have videos shot in a studio, a platform that would allow me to offer exams and quizzes and discussion boards that finally led me to yes. 

So, what makes the valuation certificate class different from the free online version? It is certainly not the content, since everything I teach in the certificate class is available on my website in multiple forms, but here are a few of the primary differences:
  1. Studio-shot videos: A studio, with professionals manning cameras, sound and lights, does allow for much better videos. With the help of a talented group that knows a lot more about editing and animation than I ever will, the final versions of the online classes are better than my online videos. There are, in all, 28 video sessions, with two sessions each week, over a 14 week time period. 
  2. Supporting material: In addition to the post class tests and the supporting slides, I have links to papers, spreadsheets, data, YouTube videos and blog posts that go with each session. While I am a realist and know that much of this additional material will go untouched, having it accessible will make it easier for you to use it, if you feel the urge.
  3. Live Webex sessions: Every two weeks, through the semester, we will have a live webex session, where you (if you are enrolled in the class) can ask questions, not just about material covered in the previous week's sessions but news stories and happenings. I know it is not much, but it is a step in the right direction.
  4. Announcements and outreach: I contact the students in my regular classes about once a day, but I will spare you that level of harassment. You will hear from me a couple of times every week, checking in on how you are doing and keeping you updated on the course. 
  5. Exams/Quizzes: There will be three quizzes and a final exam for the class. While they will  be scheduled on specific dates, you can take them any time during a 24-hour time period and if you miss a quiz, the points will be moved to the remaining quizzes. So, if life gets in the way and you are unable to take a quiz, it is not the end of the world.
  6. Valuation Project: Each person in the class can pick any company he or she want to value and value it, over the course of the class. Midway through the semester, I will offer feedback, if you want,  to allow you to tweak your valuation, and at the end of the semester, it will become a significant part of your overall grade.
  7. Certificate: After the final exam and valuation are graded, you will receive a certificate for the class, if you complete the requirements. If you do exceptionally well (and you will have to leave that judgment to me), your certificate will come "with honors".
There were 66 people who signed up for the pilot version of the class, which started in January 2017 and 39 completed the class in May 2017. I learned as much from my students as I hope they learned from me, and here are a few lessons. First, I discovered that the discussion boards were effective at creative interactive discussions, among the students, if I did my job and organized the boards by topic. Second, in perhaps the most rewarding part of the class, a few students, who found the material both interesting and easy to grasp, took on the role of teachers helping others deal with mechanical and conceptual questions. Since the most effective way to learn something is to explain it to someone who does not quite "get" it, I restrained myself from jumping into the discussion boards, unless absolutely necessary. Third, I was impressed with both the work that was put into and the quality of the valuations that were turned in by those who finished the class. Of the 39 who were certified at the end of the class, about a third did well enough to get "with honors" attached to the certificate. I would have been proud with any of these students in my regular classes.

This fall, Stern will be offering the valuation certificate class to a bigger audience, with a class of several hundred. The good news is that the class will be tweaked to reflect the lessons learnt from the pilot class. I will continue to do what I did for the pilot, with my webex sessions, and provide feedback and grades not only for your exams but on the companies that you choose to value. The bad news is that Stern will charge "university level" prices for the class and I will not try to tell you that it is "worth it", since that depends on your circumstances. It is entirely possible that you will decide that the price charged is too much for a certificate, that you cannot afford it, or that you are more interested in the learning than in the certification, and if so, I hope that you give the free online version of the class a shot. If you are interested in enrolling in the class, the webpage where you can start the process is here. Incidentally, a pilot version of my corporate finance class, also offered as a certificate class, will be run in Spring 2018, and if you are interested, here is that link.

My Sabbatical
I mentioned, at the start of this post, that I would be on sabbatical, and at the risk of evoking envy, I will tell you what that involves. I am taking the 2017-18 academic year (September 2017- September 2018) off from my regular teaching, as I am allowed to do every seventh year. It is an entitlement that people in most other professions don't have and I recognize how incredibly lucky I am to be able to take a paid break from work. I do have a few odds and ends to take care off during the year, including teaching the certificate classes that I just listed and writing the third edition of The Dark Side of Valuation, but I plan to spend much of the year idling my time away, thinking about nothing in particular. That may sound wasteful, but I have discovered that my mind is most productive, when I am not trying too hard to be insightful. At least, that's my hope and if it does happen, that would be great. I  But then again, if I don't have a single creative thought all year, that too was meant to be! 

YouTube Video

My Free Online Classes
  1. Corporate Finance (YouTube Playlist version)
  2. Valuation (YouTube Playlist version)
  3. Investment Philosophies (YouTube Playlist version) (New version will be out at the start of 2018)
Stern Certificate Classes

Thursday, July 13, 2017

The Dark Side of Globalization: An Update on Country Risk!

The inexorable push towards globalization has stalled in the last few years, but the change it has created is irreversible. The largest companies in the world are multinationals, deriving large portions of  their revenues from outside domestic markets, and even the most inward looking investors are dependent upon global economies for their returns. As a consequence, measuring and incorporating country risk into decision making is a requirement in both corporate finance and valuation. It is in pursuit of that objective that I revisit the country risk issue twice every year, once at the start of the year and once mid-year, at which time I also update a paper that I have on the topic, that you are welcome to read or browse or ignore.

The Globalization of Companies
There are some investors, especially in the United States, who feel that they can avoid dealing with risk in other countries, by investing in just US stocks. That is a delusion, though, because a company that is incorporated and traded in the United States can derive a significant portion of its revenues and earnings from outside the country. In 2015, the companies in the S&P 500, the largest market cap stocks in the US, derived approximately 44% of its revenues from foreign markets, down from 48% in the prior year.
Source: S&P
The composition of foreign sales is also changing, though gradually, over time, shifting away from the UK and Europe to emerging markets, as evidenced in the graph below:
Source: S&P
Lest you feel that this graph is skewed by the biggest companies in the index, 239 of the 500 companies in the index reported that foreign sales represented between 15% and 85% of their total sales and 13 companies reported that more than 85% of their sales came from outside the US. In 2014, two companies, Accenture and Seagate Technology, reported that all their sales were foreign, making them US companies only in name. (Many of you have pointed out that Accenture has significant US sales and that is true. I am just excerpting from the S&P report, which should lead you to question how S&P classifies foreign sales.)  This phenomenon is not restricted to US companies, as the largest companies in most markets exhibit similar characteristics. While we can debate whether these trend lines are good or bad for consumers and investors, the consequences are real:
  1. Fraying link to domestic economies: For decades, the conventional wisdom has been that the stock market in a country is closely tied to how well the economy of that country is doing. That relationship has been weakened by globalization and equity market performance around the world is disconnecting from domestic economic growth. Taking the US as an example, consider that equity markets in the US have been on a bull run, with indices up 170% to 200%, cumulatively since 2009, even as the US economy has been posting anemic growth.
  2. Central Banking power is diluted: In the decades since the great depression, we have to come to accept that central banks can use the policy levers that they have at their disposal to move long term interest rates and to strongly influence overall economic growth, but that power too has been reduced by globalization and its unpredictable flows. It should come as no surprise then that the frantic efforts of central banks\ in the US, Europe and Japan, in the last decade, to use the interest rate lever to pump up economic growth or to alter the trajectory of long term interest rates have failed.
  3. Taxing questions: When writing tax code, governments have generally assumed that companies incorporated in their domiciles have little choice but to accede to tax laws eventually and pay their share of taxes. While companies have historically played the tax game by delaying and deferring taxes due, their global reach now seems to have shifted the balance of power in their direction. In the United States, in particular, where the government has tried to tax companies on their global income, this push back has taken the form of trapped cash, as companies hold trillions of dollars of cash on foreign shores, and inversions, where some US companies have chosen to move their home base to more favorable tax locales.
  4. Declining cross-market correlations: As companies globalize, it should come as no surprise that the correlations across global equity markets have climbed, with two immediate consequences. The first is that global crises are now an almost annual occurrence rather than uncommon surprises, as pain in one market quickly spreads across the world. The second is that the salve of geographic diversification, long touted as protection against domestic market shocks, provides far less protection than it used to.
The bottom line is that there is no place to hide from country risk, and as with any other type of risk, it is best to face up to it and deal with it explicitly.

Country Risk - Default Risk Measures
The simplest and most easily measured country risk is the risk of sovereign default. When countries default on their obligations, it is not just the government that feels the pain but companies, consumers and investors do, as well.

Sovereign Default: Frequency and Consequences
Governments borrow money, both from their own citizens and from foreign entities, and they sometimes borrow too much. Some of these government default, not only on their foreign currency debt but also on their local currency debt, with the latter having become more common over time:
Source: Fitch Ratings
You may be puzzled by local currency debt defaults, since governments do have the capacity to print more of their own currency, but faced with a choice between defaulting or debasing their currencies, many governments choose the latter. When default occurs, the immediate pain is felt by the government and lenders, the former because it loses the capacity to borrow more, and the latter because they don't get paid., but there is collateral damage:
  1. Capital Market Turmoil: Liquidity dries up, as investors withdraw from equity and bond markets, making it more difficult for private enterprises in the defaulting country to raise funds for projects and resulting in sharp price drops in both bond and stock markets.
  2. Real growth: Sovereign defaults are generally followed by economic recessions, as consumers hold back on spending and firms are reluctant to commit resources to long-term investments.
  3. Political Instability: Default can also strike a blow to the national psyche, which in turn can put the leadership class at risk. The wave of defaults that swept through Europe in the 1930s, with Germany, Austria, Hungary and Italy all falling victims, allowed for the rise of the Nazis and set the stage for the Second World War. In Latin America, defaults and coups have gone hand in hand for much of the last two centuries.
Sovereign Ratings
The most accessible measures of sovereign default risk are sovereign ratings, with S&P, Moody's and Fitch all providing both local currency and foreign currency ratings for most countries around the world. While there are many who mistrust these ratings, they are widely used as proxies of country risk and changes in ratings, especially down grades, are news worthy and affect markets. The process and metrics used to arrive at the ratings are described more fully here and here but the picture below summarizes the sovereign ratings assigned to countries in July 2017 and the data can be downloaded at this link:
Link for live map
The last decade has turned the spotlight on both the pluses and minuses of ratings. On the plus side, as the ratings agencies are quick to point out, ratings and default spreads are highly correlated. On the minus side, ratings agencies seem to have regional biases (under rating emerging markets and over rating developed markets) and are slow to change ratings. 

Sovereign CDS Spreads
In the last decade, we have seen the growth of a market-based measure of default risk in the Credit Default Swap (CDS) market, where you can buy insurance against sovereign default by buying a sovereign CDS. Since the insurance is priced on annual basis, the price of a sovereign CDS becomes a market measure of the default spread for that country. In July 2017, there were 68 countries with sovereign CDS and the picture below captures the pricing (with the data available for download at this link). One of the limitations of the CDS market is that there is still credit risk in the market and to allow for the upward bias this creates in the spreads, I compute a netted version of the spread, where I net out the US sovereign CDS spread of 0.34% from each country's CDS spread. 
Link for live map
To provide a comparison between the CDS and sovereign rating measures of default risk, let me offer two example. The sovereign CDS for Brazil on July 1, 2017, was 3.46%. On the same day, Moody rated Brazil at Ba2, with an estimated default spread of 3.17%, close to the CDS value. For India, the sovereign CDS spread on July 1, 2017, was 2.42%, very close to the default spread of 2.32% that would have been assigned to it based upon its Baa3 rating.

Country Risk - Institutional Risk
When investing in a company, the sovereign default risk is just one of many risks that you have to factor into your decision making. In fact, default risk may pale in comparison to risks you face because of the institutional structure, or lack of it, in a country. At the risk of picking at scabs, here is my shot at assessing some of these risks.
1. Corruption
Much as we like to inveigh against its consequences, corruption is not just part and parcel of operating in some parts of the world, but it takes on the role of an implicit tax, one that is paid to free agents, acting in their own interests, rather than to governments. Transparency International, an entity that measures corruption risk around the world, estimates corruption scores for individual countries and heir findings for 2016 are summarized in the picture below. To see where a country falls on the corruption continuum, you can either click on the live link below the picture or download the data by country by clicking here.

Link to live map
While it is easy to fall back on cultural stereotypes to explain differences across countries, there is a high correlation between economic well being and corruption. Thus, while much of Latin America scores low on the corruption, Chile and Uruguay rank much higher, as do South Korea and Japan in Asia.

2. Legal Protections
Even the very best investments are only as good as the legal protections that you have as an investor, against expropriation or theft, which is why the property right protections rank high on investor wish lists. To measure the strength of property rights, I turned to the International Property Rights Index (IPRI), and report the scores they assigned in their most recent update in 2016, to countries in the picture below. You can click on the live link below the picture or download the data here.

Link to live map
Europe, North America, Japan and Australia all score high on property rights, but the hopeful sign is that index itself has seen increasing respect for property rights across time and Venezuela and Myanmar are now more the exception, than the rule.

3. Risk of violence
It is difficult to do business, when you have bullets whizzing by and bombs going off around you. Holding all else constant, you would prefer to operate in parts of the world that are safer rather than riskier. To measure exposure to violence, I again turn to an external entity, Vision of Humanity, and reproduce their Global Peace Index in the picture below (with link to live map and to data):
Link to live map
In keeping with the adage that when it rains, it pours, the countries that are most susceptible to corruption and have weak property rights also seem to be most exposed to physical violence.

Country Risk - Equity Risk
As you can see, there are multiple dimensions on which you can measure country risk, leading to different scores and rankings. As an investor in the country, you are exposed to all of these risks, albeit to varying degrees, and you have to consider all these risks in making decisions. Consequently, you would like (a) a composite measure of risk that (b) you can convert into a metric that easily fits into your investment framework.

1. Country Risk Scores
There are several services that provide composite measures of country risk, including the Economist, Euromoney and Political Risk Services (PRS). These country risk measures take the form of numerical scores, and in the heat map below, I report the change in the PRS country risk score between July 2016 and July 2017 and categorize countries based on the direction and magnitude of the change. Here, as in the prior pictures, you can see the PRS scores and the change, by country, by either clicking on the live map link below the picture or download the data by clicking here). 
Link for live map
Based on the PRS scores, the vast majority of emerging markets became safer during the time period between July 2016 and July 2017, with the biggest improvements in Latin America and Asia. The North American countries saw risk go up, as did pockets of Africa and South East Asia. The problem with country risk scores, no matter how well they are measured, is that they do not fit a standardized framework. Just to provide an illustration, PRS scores are low for risky countries and high for safe countries,  whereas the Economist risk scores are high for risky countries and low for safe countries.

3. Equity Risk Premiums
To incorporate and adjust for country risk into investing and valuation, I try to estimate the equity risk premiums for country, with riskier countries having higher equity risk premiums. I start with the implied equity risk premium for the US, which I estimate to be 5.13% at the start of July 2017 as my mature market premium and add to it a scaled up version of the default spread (based upon the rating); the scaling factor of 1.15 is based upon the relative volatility of emerging market equities versus bonds. You can see a more detailed description of the process in the paper that is linked at the end of this post. You can look up the equity risk premium for an individual country by clicking on the live map link or download the data by clicking here.
Link for live map
These equity risk premiums are central to how I deal with country risk in valuation, as I will explain in the last section of this post.

Closing the Loop
When valuing companies that have substantial exposure to country risk, it is easy to get overwhelmed by the variety of risks. To keep the process under your control, you should start by breaking country risk into three buckets: risk that is specific just to that country, risk that is macro/global and discrete risks that are potentially catastrophic (such as nationalization or terrorism). Each has a place in valuation, with country specific risks incorporated into expected cash flows, macro economic risks in the discount rate and discrete risks in a post-valuation adjustment. 

1. Adjusting discount rates
The key to a clean country risk adjustment, when estimating discount rates, is to make sure that you do not double or even triple count it. With the cost of equity for a company, for instance, where there are only three inputs that drive the cost, it is only the equity risk premium that should be conduit for country risk (hence explaining my earlier focus on equity risk premiums, by country). The risk free rate is a function of the currency that you choose to do your valuation in and the relative risk measure (or beta, if that is how you choose to measure it) should be determined by the business or businesses that the company operates in. 

If you are discounting the composite cash flows of a multinational company, the equity risk premium should be a weighted average of the equity risk premiums of the countries that the company operates in, with the weights based on revenues or operating assets. If you are valuing just the operations in one country, you would use the equity risk premium just for that country.

2. Expected cash flows
With risks that are specific to a country, it is better to incorporate the risks into the expected cash flows. Thus, if a country is rife with corruption, you could treat the resulting costs as part of operating expenses, reducing profits and cash flows. When legal and regulatory delays are a feature of business in a country, you can build in the delay as lags between investing and operations. When violence (from terrorism or war) is part and parcel of operations, you may want to include a cost of insuring against the risk in your cash flows. 

None of these adjustments are easy to make, but it is worth remembering that incorporating the risk into your cash flows is not risk adjusting the cash flow, since the latter requires replacing the expected cash flow with a certainty equivalent one.  Where does currency risk play out? When converting cash flows from one currency (foreign) to another (domestic), you should bring in expected devaluation or revaluation into expected exchange rates. If you want to hedge exchange rate risk, you can incorporate the cost of heeding into your cash flows but it is not clear that you should be adjusting discount rates for that risk, since investors can diversify it away.

3. Post-Valuation Adjustment
There are some risks that are rare, but if they occur, can be devastating, at least for investors in a business. Included in this grouping would be the risk of nationalization and terrorism. These risks cannot be incorporated easily into discount rates and adjusting expected cash flows in a going concern valuation (DCF) for risk that a company will be nationalized or will not survive is messy. 

Thus, to estimate the effect that nationalization risk will have on the value of a business, you will have to assess the probability that the business will be nationalized and the value that you will receive as owners of the business, in the event of nationalization.

Danger and Opportunity
One of my favorite definitions of risk is the Chinese symbol for crisis, a combination of the symbols for danger and opportunity.
With risky emerging markets, this comes into , I am reminded that to have one (opportunity), I have to be willing to live the other (danger). Blindly ignoring these markets, as some conservative developed market companies are inclined to do, because there is danger will lead to stagnation, but blindly jumping into them, drawn by opportunity, will cause implosions. The essence of risk management is to measure the danger in markets and then gauge whether the opportunities are sufficient to compensate you for the dangers. That is what I hope that I have laid the foundations for, in this post.

YouTube Video

  1. Country Risk: Determinants, Measures and Implications - The 2017 Edition
Data Links