Monday, September 1, 2014

The Tax Dance: To Pass Through or Not to Pass Through Income?

I started last month by looking at US tax law and how it induces bad corporate behavior  and in this one, I want to expand the discussion to look at how the tax structuring of a business can affect its value. In particular, I would like to look at the differences between taxable entities (public corporations, private C-Corps) and pass-through entities (MLPs, REITs and private S-Corps), both on taxes and other aspects of doing business, and the trade off that determines why companies in one group may try to move to the other. I use the framework to look at Kinder Morgan’s decision to bring its master limited partnerships under the corporate umbrella and the value effects of that decision.

The Evolution of Different Tax Entities
For most of the last century, publicly listed firms in the United States followed the corporate model, where the earnings made by a company were first taxed at the corporate level, and investors in the company were then taxed again, often at different rates depending on whether the income was paid out as dividend or allowed to accumulate in the firm to generate capital gains. In its spasmodic attempts to use the tax code to encourage the "right kind" of investments, Congressional legislation created two entities that were allowed to escape entity-level taxes: real estate investment trusts (REITs) in 1960 and master limited partnerships (MLPS) in 1987, the former obviously for investments in real estate and the latter directed towards energy investments.

Across the privately owned business spectrum, the vast majority of businesses are structured as sole proprietorships, but among private businesses that choose to incorporate (to get the benefits of limited liability), the choice is between C-Corps, which resemble publicly traded companies in tax treatment and S-Corps, which are pass-through entities. While the original rationale for these pass-through structures may have been long forgotten, a CBO report in 2012 noted an undeniable trend towards pass-through entities in the last three decades:
Source: Congressional Budget Office
Not only has the percentage of income reported by taxable corporations dropped from almost 90% in 1981 to less than 60% in 2007, but the proportion of firms organized a pass-through entities has climbed from 83% to 94%. While the CBO report stopped in 2007, there are signs that the movement towards pass-through entities has continued in the years since. In particular, not only have we seen dozens of MLPs go public in the last few years, driven by the energy boom in the United States, but we have also seen increasingly creative REIT issuances, such as the one by Windstream, a telecom company, which classified its phone lines as real estate. In the private business space, the S-Corp form, which was generally used by smaller firms, has been embraced by some larger entities as well; Bechtel is one example of a large private business structured as an S-Corp.

The Trade Off
As pass-through entities grow in popularity, we are not only faced with the question of why this shift is occurring but also pragmatic questions about how best to value or price these entities. In this section, I will start with a discussion of the tax differences that are not unexpectedly at the center of the story and then look at the rest of the story (which does muddle the trade off somewhat), with the intent of answering these questions.

The Tax Story
Not surprisingly, most of the news stories that I have read about the growth in pass-through entities emphasize the tax angle. They argue that replacing the double taxation that is inherent in a conventional corporation (C-Corp), where the entity first pays taxes and equity investors are then taxed again, with a pass-through model, where income is taxed only at the investor level only, saves taxes. That may be true, in a general sense, but the story is a little more complicated. To understand the difference, I have outlined how taxes work at taxable corporations versus a pass-through entity (S-Corp or MLP) in the figure below:

Pre and Post tax Income: The Tax Effect of Pass Through Entities
Using a company with $100 in pre-tax operating earnings, in conjunction with the top statutory tax rates on corporate income (35%), personal income (40%), dividends (20%) and capital gains (20%) today, and assuming that 40% of corporate earnings get paid out as dividends, the pass through entity delivers higher post-tax income to the investor.
Statutory tax rates and income
In particular, starting with pre-tax income of $100, the pass-through entity delivers after-tax income of $60, $8 higher than $52 you receive as an investor in the corporate entity, translating into an earnings and value premium of about 15.38%:
Earnings/Value Premium for pass-through = ($60/$52) -1 = .1538 or 15.38%
The story gets complicated when you consider two additional factors. The first is that corporations get more wiggle room that individuals do when it comes to taxes, both in terms of tax deductions/deferral and how foreign income is taxed, allowing them to lower the effective tax rate that they pay. The second is that the tax due on the capital gains portion of corporate income can be deferred by investors to future periods, since it is not collected until the stock is sold. In the example above, for instance, replacing the corporate statutory tax rate of 35% with the average effective tax rate of 27% (the weighted average tax rate across money-making US corporations in 2013) and allowing for the fact that about 25% of investors are tax exempt (thus reducing the tax rate that investors pay on dividends/capital gains to 15%), we get the result that  the collective taxes paid by a corporate entity is lower than that paid by a pass-through.
Actual taxes and Earnings
The pass through entity will trade at a discount of 3.30% on the taxable entity in this case and this example, though simplistic, illustrates the trade off between taxable and pass through entities depends on the following factors:
  1. The differential between corporate and personal taxes: The greater the the tax rate on pass-through investor income, relative to the corporate tax rate, the less incentive there should be to shift to a pass-through entity form. This may explain why more companies have shifted to the pass-through structure since 1986, when the individual tax rate was brought down to match the corporate tax rate.
  2. The proportion of earnings that gets paid out as dividends: The returns to investors from holding stock in a corporate entity can take the form of dividends or price appreciation. Until 2003, dividends were taxed at a much higher rate than capital gains, and the effective tax paid by investors on corporate income was therefore greater in high dividend paying stocks. While the tax rates have converged since, capital gains retain a tax-timing advantage, since investors don't have to pay until they sell the stock. 
  3. The make up of investors in the entity: As noted in the section above, investors in publicly traded companies can have very different tax profiles, from wealthy individuals who pay taxes on dividends/capital gains to corporations that are allowed to exempt 70-80% of their dividend income from taxes to tax-exempt investors (pension funds) that pay no taxes on any corporate income. Furthermore, the taxes paid by taxable investors on capital gains can vary depending on how long they hold the stock, with short term investors paying much higher taxes than investors who hold for longer than a year. In summary, companies with primarily wealthy, short-term, individual investors holding their shares have a greater benefit from shifting to pass-through status than companies with primarily tax-exempt or long term investors, holding their shares.
The Rest of the Story
If the only difference between pass-through entities and taxable entities were on taxes, the assessment that mattered for value would be whether you paid more or less in taxes with one form over the other. However, the laws that created the pass-through entities also created restrictions on other aspects of business behavior including where and how these businesses invest, how much they have to pay out in dividends and how they finance their operations. In the picture below, I have categorized business decisions into investing, financing and dividend pieces, to illustrate the restrictions that you face with the pass-through entities.

Investment Policy
Financing Policy
Dividend Policy
Pass Through
Pass Through
Pass Through
There are generally no investment constraints.REITs and MLPs are restricted in the businesses that they can invest in, the former in real estate and the latter in energy.
S-Corps face    no explicit investment constraints.
Can claim interest tax deduction on debt, giving it tax benefits from borrowing.No direct benefit from debt.
MLPs and REITs can issue new shares, but S-Corps cannot have more than 100 shareholders.

Managers have the discretion to set dividends, reinvest earnings in projects or just hold on to cash.
Both REITs and MLPs are required to pay out 90% of their income as dividends, on which investors have to pay the ordinary income tax rate (rather than the dividend tax rate).
S-Corps do not have explicit dividend payout requirements.

In summary, choosing a pass-through tax-status for your business will narrow your investment choices (to real estate if you are a REIT, and to energy, if you are an MLP), require you to return much or all of your earnings as dividends and reduce your financing options (by restricting your capacity to attract new stockholders if you are a sub-chapter S corporation and by making the tax benefits of debt indirect). 

Will these restrictions make you a less valuable business? If you are a mature business with few growth opportunities and no desire to stray from your investment focus, the costs imposed by the pass through entity are minimal. In fact, the discipline of having to pay the cash out to investors may reduce the chances of hubris-driven growth and wasteful investment. If you are a business with good growth opportunities, the restrictions can reduce your value, either because you are unable or unwilling to issue new shares to cover your investment needs, or because you cannot take direct advantage of the tax subsidies offered by debt (to traditional corporations).

The valuation of pass through entities
Most of valuation practice and theory has been built around valuing publicly traded companies. In a typical public company valuation, we generally estimate cash flows after corporate but before personal taxes and discount it back at a cost of capital that is post-corporate taxes (with a tax benefit for debt) and pre-personal taxes. While the principles of valuation don't change when you value pass-through entities, it is easy for inconsistencies to enter valuations, especially if some of the inputs (margins, betas, costs of equity) come from looking at publicly traded, taxable entities. The process enters the danger zone, when appraisers create arbitrary rules of thumb (and legal systems enforce them), and attach premiums or discounts are attached to public company values (obtained either in intrinsic valuations or from multiples/comparable). To value pass through entities consistently, I would suggest the following steps:

Step 1: Take a post-personal tax view on cash flows
Broadly speaking, you can value a passthrough entity either on a pre-tax basis or on the basis of post-tax cash flows. If you decide to do your valuation on a  pre-tax basis, you estimate the cash flows generated by the entity, whereas on a post-tax basis, you will have to net out the taxes that investors have to pay on these cash flows. The rest of the inputs into cash flows, including growth and reinvestment, then have to be tailored appropriately. This choice, though, has value consequences and I would argue that, if valuing a pass through entity, you will get a fairer estimate of value using post-personal tax numbers for two reasons:

  1. If the rationale for shifting from one tax form to another is to save on taxes, it seems incongruous to be using pre-tax numbers. After all, it is the fact that you get to have higher cash flows, after personal taxes, that causes the shift in tax status in the first place.
  2. Some practitioners use the argument that if you are consistent, it should not matter whether you look at pre-tax or post-tax numbers, but that holds only if there is no growth in perpetuity in your entity earnings. If you introduce growth into your valuation, you do start to see a benefit that shows up only in the post-tax numbers and there is an intuitive explanation for that. The expected growth rate in an intrinsic valuation model is a measure of the value appreciation in the business, i.e., the capital gains component of return. Even though the income in a pass through entity is taxed in the year in which it is earned at the personal tax rate, the increase in value of a pass through entity (MLP, Subchapter S, REIT) is not taxed until the business is sold and qualifies for capital gains taxes, thus creating the premium in the post-tax value. It is precisely to counter this tax benefit that pass through structures are required to pay almost of their earnings out as dividends. As a result, the growth in earnings in a pass through entity has to be funded either with new equity issues (whose prices will reflect the value of growth) or new debt (without the direct tax benefit from interest expenses) and that growth has much smaller or no price appreciation effect.
Step 2: Estimate a tax-consistent discount rate
This is a key step, where the discount rate has to be estimated consistently with the cash flows, with pre-tax discount rates for pre-tax cash flows, and post-personal tax discount rates for post-tax cash flows. It is at this stage that inconsistencies can enter easily, since most of the public data that we have and use in estimating discount rates comes from taxable entities (publicly traded companies) and is post-corporate, but pre-personal taxes. Consider, for instance, the estimation of the cost of equity for a publicly traded real estate development company, where we use a risk free rate, a beta (say for real estate as a business) and an equity risk premium (either historical or implied). Using a 10-year T.Bond rate of 2.5% as the risk free rate in US dollars, a beta of 0.99 for real estate development and an equity risk premium of 5.5%, the cost of equity we obtain for this company is 7.95%:
Cost of equity of publicly traded real estate development firm = 2.5% + 0.99 (5.5%) = 7.95%
Using the average debt to capital ratio of 19.94% and an after-tax cost of debt of 3.30%, we estimate a cost of capital of 7.02% for the company:
Cost of debt = 5.50% (1-.40) = 3.30%
Cost of capital = 7.95% (.8006) + 3.30% (.1994) = 7.02%
Note, though, that this is a cost of equity and capital for a public company, post-corporate taxes and pre-personal tax.

To convert these numbers into pass through discount rates, you first have to take the tax benefit of debt out of the equation. Consequently, it is safest to work with an unleveled beta/cost of equity, on the assumption that the pass through does not use or at least does not benefit from the use of debt and then bring in the effect of personal taxes. The steps involved are as follows:
1. We start with the unlevered beta of 0.85 for a real estate development company and compute a cost of equity based not that beta.
Unlevered cost of equity = 2.5% + 0.85 (5.5%) = 7.18%
2. To convert this number into a pre-tax cost of equity that you can use to discount pre-tax cash flows on a pass-through real estate development company, you will need two additional inputs. The first is the tax rate that investors in the publicly traded entity pay on corporate returns (dividends & capital gains) and the second is the tax rate that investors in the passthrough entity pay on their income. If you assume that the investor tax rate on corporate income is 15% and the investor tax rate on pass-through income is 40%, the pre-tax cost of equity for a real estate development company is 10.17%.

The after-tax cost of equity will then be 6.10%, computed as follows:
Pass through aftertax cost of equity = 10.17% (1-.4) = 6.10%
Implicitly, we are assuming that investors demand the same return after personal taxes of 6.10% on an investment in real estate development, no matter how the entity is structured for tax purposes.

At the start of every year, I estimate costs of equity for publicly traded companies, classified by sector, on my website. I have updated that table to yield pre-tax and post-personal-tax costs of equity for pass through entities in each sector, using a tax rate of 15% for investors on corporate income and 40% for investors on passthrough income, but I give you the option of changing those numbers, if you feel that they are unrealistic for the sector that you are working with.

Step 3: Build in the constraints that come with pass-through form
As a final step in the valuation, you should bring in the effects of the constraints that come with pass-through entities. One reason that I scaled the cost of equity of publicly traded companies for the tax effects, rather than the cost of capital, is because taxable companies get a direct tax benefit from borrowing money (thus lowering cost of capital) and pass through entities do not get this debt benefit.  To capture the investment constraints, coupled with and perhaps caused by the forced dividend payout requirements, you have to either assume a lower growth in income (if the company chooses not to issue or cannot issue new equity) or equity dilution in future periods.

Step 4: Do the valuation
With pretax (post tax) cash flows matched up to pretax (post tax) discount rates, you can now complete the valuation of the pass through entity with that of an equivalent corporate entity. If you are consistent about following this process, the value of a business can go up, down or remain unchanged, when it shifts from a taxable form to a pass through entity, depending in large part on the tax characteristics of the investors involved and the growth potential of the company. In particular, you will be trading off any tax savings that may accrue from the shift to a pass through status against the lost value from the investment and financing constraints that accompany a pass through structure. Thus, the notions that a S-Corp is always worth more than a C-Corp or that converting to a MLP is always beneficial to investors are both fanciful and untrue.

To illustrate this with a simple example, assume that you have a real estate development business that generated pretax operating income of $100 million last year, on invested capital of $400 million, and expects this income to grow at 2.5% a year, in perpetuity. Assume that you are considering whether to  incorporate as a taxable corporation or as a pass through entity and that you are provided the corporate and investor tax rates on both corporate and pass through earnings. In the picture below, I value the effect for a given set of inputs.

In this case, the pass through entity has a slightly higher value than the taxable form, but reducing the corporate effective tax rate to 25% tips the scale and makes the taxable entity more valuable. In fact, using the average effective tax rate of 19.34% that was paid by companies in the real estate development sector last year gives the taxable form a decided benefit. You can use the spreadsheet yourself and change the inputs, to see the effects on value.

The Kinder Morgan Conversion
With the framework from the last two sections in place, let us look at the recent news out of Kinder Morgan. The company (KMI), one of the lead players in the MLP game, with its pipelines constructed as MLPs, announced recently that it planned to consolidate three of these MLPs (KMP, KMR and EPB) into its corporate structure, thus shifting them back from pass through entities to more traditional taxable form. 

The market reaction to the consolidation has been positive for all of these stocks, but is the market right? And if yes, where is the additional value coming from? Like other MLPs, the Kinder entities were trapped in a vicious payout cycle, where investors expected them to pay out ever increasing amounts of cash, which they funded with debt, on which they get no direct tax benefit (though their investors indirectly benefit). Thus, converting back to corporate form will release them from the vice grip of dividends, partly because they will not be obligated by law to pay out their earnings in dividends and partly because their investor base will shift. That release presumably will allow them to both pursue more growth and perhaps fund it more sensibly with equity (retained earnings) and tax-subsidized debt. This value creation story rests on the company being able to find value-enhancing growth investments and on good corporate stewardship.  

There is one final aspect of the Kinder Morgan deal that suggests that the net effect of this deal will be much more negative for partnership unit holders than it is for parent company stockholders. The partnership has deferred taxes on its income that will come due on the consolidation, estimated at $12 to $18 per partnership unit, depending on the investor's tax rate and how long he or she has held the unit, which is higher than the $10.77/unit that will be paid out as a distribution on the conversion. 

Since the prevailing wisdom seems to be that corporations and wealthy investors evade or avoid taxes, it is not surprising that any story on conversions to or from a pass through tax status becomes one about tax avoidance. Thus, we are told by journalists and analysts that the broad shift of businesses to pass through status (MLPs) is all about saving taxes and we are also then told that Kinder Morgan's conversion back from an MLP set up to a corporate entity is also about saving taxes. This borders dangerously close to  journalistic and analyst malpractice for two reasons. The first is that both pass through and taxable entities pay taxes and the tax savings are never as large as either critics or promoters of pass through entities make them out to be. The second is that if you compare the tax structures of traditional corporations and pass through entities, it strikes me that it is the former with its multiple layers of taxes (corporate, dividend, capital gains) that is convoluted, complex, and ripe for manipulation, and not the latter. In fact, if you wanted to make one tax system your standard one, it is the pass through version that seems to offer more promise.


Friday, August 8, 2014

Reacting to Earnings Reports: Let's get real!

In my last two posts, I considered how earnings reports can generate narrative shifts or changes, thus affecting value, and pricing effects, when companies trail or beat investors’ estimates on metrics (earnings per share, revenues, user numbers etc.). In this one, I intend to apply the lessons in those posts to three companies that I have been working with over the last couple of years: Apple, Facebook and Twitter. In particular, I would like to look at the most recent earnings report for each company, the news each report contained, the distractions in each one and the effect on stock prices. I would also like to look at the information in past earnings reports for each company, over the entire (limited) histories for Facebook and Twitter’s, and the last two years of reports for Apple, with the intent of incorporating what I have learned into updating my narrative for each company.

Apple Earnings Reports: The Meh Chronicles

I looked at Apple in detail a few months ago, chronicling my estimates of value for the company and stock price movements starting in 2011 and going through April 2014. The graph below reproduces my findings (with prices and values per share adjusted for the recent seven to one stock split), with an update through August 2014:
Apple: Price versus Value (My Estimates)
Note that while stock prices have ranged from $45 to close to $100 over this period, my value estimates have had a much tighter range, reflecting my largely unchanged story line for the company, over the period. Starting in 2011, my narrative for Apple has been that it is a mature company, with limited growth potential (revenue growth rates< 5%) and sustained profitability, albeit with downward pressure on margins, as its core businesses becomes more competitive. I allowed for only a small probability that the company would introduce another disruptive product to follow up its trifecta from the prior decade (the iPod, the iPhone and the iPad), partly because of its large market cap and partly because I thought it had used up its disruption karma over recent years. 

Looking at the earnings reports from the company over the last nine quarters, it is remarkable how little that narrative has changed. In the first two sets of columns, I report on Apple’s revenues and earnings per share and contrast the actual numbers with the consensus estimate for these numbers in each quarter. For much of the time period, Apple has matched or beaten revenue and earnings estimates, albeit by small amounts, but the market has been unimpressed, with stock prices down on six of the nine post-report days and seven of the nine post-report weeks. 
Apple: Earnings Reports from July 2012 to July 2014
Note that after controlling for the quarterly variations, revenues have been flat or have had only mild growth and operating margins have been on a mild downward trend.  With Apple, the other focus in the earnings reports has been on how iPhone and iPad sales are doing and the table below reports on the unit sales that Apple reported each quarter, with the growth rates over the same quarter’s sales in the prior year. In the last two columns, I report Apple’s global market share in the smartphone and tablet markets, by quarter. 

Apple: Unit Sales for iPhone & iPad, with global market share
While the market fixation with Apple’s iPhone and iPad sales may be disconcerting to some, it makes sense for two reasons. First, it reflects the fact that Apple derives most of its revenues from smartphones/tablets and that the growth in unit sales and change in market share therefore becomes a proxy for future revenue growth. Second, Apple’s earnings are being sustained by its impressive profit margins in the smartphone and table businesses and looking at how well it is doing in these markets becomes a stand-in for how sustainable the company’s margins (and earnings) will be in the future. Each quarter, there are rumors of another Apple disruption in the works, but each time the promises of an iWatch or an iTV don’t pan out, investor expectations that Apple will pull another rabbit out of its hat have eased. 

The most recent earnings report seems to reflect this period of stability, temporary though it may be, for Apple, where investor expectations have moderated and the company is being measured for what it really is: an extraordinarily profitable company, with the most valuable franchise in the world. It seems to have stabilized its position in the smart phone world, is seeing its tablet market shrink and its personal computer business is being treated as a rump business. In effect, analysts are treating it as a mature company that is being powered by the iPhone money machine, where margins are declining only gradually. Since that is the narrative that I have using all along in my valuations, I see little change in my assessment of intrinsic value for Apple. Allowing for the stock split, the value per share that I assess for the company, with the information in the new earnings report incorporated into my estimates, is $96.55, almost unchanged from my estimate of $96.43 in April 2014. With the new iPhone 6 launch just a few months away, I am sure that the distractions will start anew, and I think it is prudent for me as an investor to map out an exit plan, if the stock price rises to $100 or higher. If it does not, I will happily continue to hold Apple, collect my dividends, and hope for a disruption down the road.

Facebook: Bigger than Google?
I valued Facebook just before its IPO in this post, and argued that the stock was being over priced at $38 for the offering. The tepid response to the offering price made me look right, but for all the wrong reasons. The botched IPO was not because the stock was over priced or because the market attached a lower value to the stock but largely due to the hubris of Facebook’s investment bankers who seemed to not only think that the stock would sell itself but actively worked against setting a narrative for the company. My initial valuation, though it will look conservative in hindsight, was based upon the belief that Facebook would be as successful as Google in its growth in the online advertising business, while maintaining its sky-high profit margins. 

Looking at Facebook’s earnings reports since its IPO, there have been nine reports and the market reaction has shifted significantly over the period.
Facebook: Earnings Reports & Price Reaction
I think that the botched public offering colored the market response to the very first earnings report, with the stock down almost 25%. In fact, I revalued Facebook after this report, when the stock price plunged below $20, and wrote this post, arguing that there was nothing in the report that changed the narrative and that the company looked under valued to me. I was lucky enough to catch it at its low point, since the company turned the corner with the market by the next quarter and the stock price more than doubled over the following year. I revisited the valuation after the August 2013 earnings report, and chose not to change my narrative, leaving me with the conclusion that the stock was fully priced at $45 and that it was prudent to sell.  Looking at the earnings numbers over the nine quarters, it is clear that has Facebook has mastered the analyst expectations game, delivering better-than-expected numbers for both revenues and earnings per share for each of the last seven quarters. 

With Facebook, the market has also paid attention to the size and growth of its user base as well as the company’s success at growing its mobile revenues. In the table below, I list these numbers as well as Facebook’s invested capital each quarter (computed by adding the book values of debt and equity and netting out cash) and a measure of capital efficiency (sales as a proportion of invested capital):
Facebook: User Numbers, Revenue Breakdown & Invested Capital
This table captures the heart of the Facebook success story: a continued growth rate in a user base that is already immense, a dramatic surge in both online users and advertising and improving capital efficiency (note the increasing sales to capital ratio).  The most recent earnings report provided more of the same: continued user growth, increased revenues from mobile advertising and improved profitability, both relative to revenues and invested capital. Looking at the last report, I have to conclude that I was wrong about Facebook’s narrative remaining unchanged, for the following reasons: 
  1. While my initial reaction to Facebook’s success on the mobile front was that it was what it needed to do to sustain its narrative as a successful online advertising company, the rate at which it has grown in the mobile market has been staggering. In fact, I think that there is now a very real possibility that Facebook will supplant Google as the online advertising king and continue to maintain its profitability. That is a narrative shift, which will translate into a larger market share of the online advertising market, higher revenue growth and perhaps more sustainable operating margins (than I had forecast). 
  2. The inexorable growth in the user base, astonishing given how large the base already is, has also been surprising. That remains Facebook’s biggest asset and a platform that they can try to use to enter new markets and sell new products/services. Facebook has also shown a willingness to spend large amounts of money on acquiring the pieces that it needs to keep increasing its user base and build on it. The downside of this strategy is that growth has been costly (though the costs are hidden for the moment in the financials), but the upside is that is putting in place the pieces it needs to monetize its user base. While the revenue breakdown does not reflect this business expansion yet, I think that Facebook is better positioned for a narrative change now than it was a year or two ago. 
My updated valuation for Facebook reflects these adjustments. Incorporating a higher revenue target ($90 billion, rather than $60 billion) and more sustained margins (40% instead of 35%) , I estimate a value per share of $63 today. For those of you who have been taking me to task for selling at $45 in September 2013, I commend you for your foresight in holding on to the stock, but I am at peace with decision for two reasons. First, given what I knew in September 2013, I did what I had to do, given my investment philosophy, and second guessing it now is an exercise in futility. Second, if the biggest regret I have in my investing life is that I sold a stock to make a 150% return rather than holding on to it to make a 300% return, I would consider myself to be truly blessed. I may be compounding my mistake here, but at $72, I don't see it as a bargain, and I am in no hurry to buy the stock now. For those of you who are Facebook stockholders, though, this may be one of those companies where the value could chase the price for years, as the company finds way to turn the possible into the plausible and the plausible into the possible, and I wish you only positive returns.

Twitter:  The Early Returns
In the weeks leading up to the Twitter IPO, I wrestled with valuing the company. In the valuation that I did in the week before the IPO, the narrative I offered was of a company that would become a significant but not a dominant player in the online advertising business. I argued Twitter’s strength (the 140 character limit on messages) would also be its weakness, and that businesses would be loath to make Twitter their primary advertising platform. My targeted revenues in ten years were still substantial, and in conjunction with a healthy profit margin of 25%, yielded a value per share of $18, well below the offering price of $26 and even further below the opening day price of $46. 

In the months since, Twitter has had three earnings reports, and the accounting results are summarized below, with analyst expectations and the stock price reaction to each report. 
Twitter: Earnings Reports and Price Reactions
Twitter’s first two earnings reports were received badly by the market, though the company beat revenue forecasts on both, partly because the company continues to lose money. The most recent earnings report received a rapturous response immediately after it came out, though some of the rapture seems to have eased in the days after.  Even more so than Facebook, the market has focused on secondary numbers at Twitter, with particular attention being paid to the growth in the user base. In the table below, I list these other numbers: 
Twitter: The Other Numbers
The negative reaction to the second earnings report was partly due to the low growth (relative to expectations) that Twitter reported in the number of users and the positive reaction to the last report seems to be traceable to Twitter beating analyst expectations for user growth in the most recent quarter.  Looking at both the accounting and user numbers, what is striking about Twitter is how little the company has changed over the period that it has been in the market. The proportion of revenues it receives from advertising has remained around 90%, its revenues from international sales have increased only marginally and its mobile advertising has stayed at a high percentage of revenues (which is not surprising given that its compact format travels well to mobile devices). Its use of invested capital has not become more efficient and while you may argue that this is early in the game, contrast Twitter's evolution with Facebook's over the first few quarters.

There is nothing that I see (and I may be missing some key component) in these reports that would lead me to reassess my initial narrative, i.e., that Twitter will be a successful, but not dominant online advertising company, and the last earnings report only reinforced that view. There was some good news in the report, especially on the revenue front, but there was some game playing that was needless, in my view. First, as this article points out, the user numbers includes those who are not Twitter users in the conventional sense but are exposed to tweets in the context of news stories. Since these indirect users will not get to see (and therefor click on) the sponsored tweets that are the company’s advertising mainstay, I think that including them with total users is a little misleading, though the company may not have intended to be deceptive. Second, and this is not a problem specific to only Twitter, is the claim that the company actually made money, if you do not count stock-based compensation as an expense. As I argued in this post, this is nonsense, but I blame the analysts and investors who buy into this fiction just as much as I do the companies that feed them the fiction. In fact, as I listened in on the Twitter earnings call, I was left with the uneasy feeling that this was an earnings report produced for equity research analysts, by an equity research analyst, in terms of the numbers it emphasized. It may be pure coincidence that Twitter acquired a new CFO between its last report and this one, and that in addition to being a banker who led their public offering, he was an equity research analyst in a prior incarnation, but I don't think so. 

Incorporating everything that I have learned from these reports into my valuation, I see little movement in my intrinsic value. Even allowing for a much more efficient use of capital in the future, my estimate of value per share is $22.53. It is still early in Twitter’s corporate life and like Facebook, and I did see this news story about Twitter perhaps entering the online retailing world, and while it may be just as much a sign of desperation as hope, it is true that young company narratives can change quickly. There is a lot more about Twitter (and its business model) that I do not know than I do, and I would like to see Twitter come through on their promise of better metrics of user engagement with their business model. 
  • I am curious about how many users actually click on the sponsored tweets. I don't like to extrapolate from personal experience but I not only have never clicked on one (or even been tempted to do so) but I find myself irritated to see tweets in my timeline from businesses trying to sell me their products and services. For Twitter's sake, I hope that I am an outlier.  
  • I am also curious about how much it is costing Twitter to get new business (how much does it cost to add an advertiser) and what their pricing system for ads is. After all, surging revenues don't have much value, if your costs to deliver those revenues surge even more.
As someone who uses Twitter a lot more than Facebook, I would like to see the company succeed, but as an investor, I remain a skeptic.

The Bottom Line 
I could go on with other companies but I think I will outlive my welcome. With just these three companies, I hope that I have been able to bring home two salient points about earnings reports. The first is that while it is always the most recent earnings report that people tend to focus on, there is value in looking at a time series of reports, since there are patterns that may emerge from that series. The second is that the patterns you observe should feed back into your narrative and valuations, reinforcing your existing views in some cases, changing them in small ways in other cases and shifting them dramatically in still others. The earnings report trail is leading me to different destinations: with Apple, to an exit point, with Facebook, to a shifting of perception on what the company is worth (though not to the point of being a buyer at its current price) and with Twitter, to no real change in my perception that while the company has promise but is over priced.

Wednesday, August 6, 2014

Reacting to Earnings Reports: Pricing Metrics and Market Reactions

In my last post, I looked at how earnings reports and other news stories about a company contain information that can lead you to reassess your narrative and consequently the value that you attach to that company. If you are an investor or analyst who is familiar with how markets react to earnings reports, you can legitimately argue that I am making this process more complicated than it has to be and that what you see at the time of the earnings report is a market reaction to how well or badly companies deliver on a specific pricing metric, relative to expectations. You are right and in this post, I will look at why investors often focus on these simple (and sometimes simplistic) metrics, how these metrics can change as a function of where a company is in its life cycle and  the dangers of focusing on metrics rather than value.

The Allure of Price Metrics
As you watch investors react, sometime violently, to a company reporting earnings per share that are a few cents below or above expectations, you may wonder why so much attention is being paid to a single metric and so little to the rest of the news in the earnings report. While I think the practice is dangerous, it can be explained with the following:
  1. It is easier to focus on a single number or metric than it is to develop a narrative and a valuation for a company. Consequently, investors and many analysts prefer to spend almost all of their time in coming up with estimates for that number, on the assumption that if they are right about those estimates, neither narratives nor valuations matter.
  2. In several earlier posts, including these on Apple and Twitter, I have drawn a distinction between value and price and argued that while investors care about the former, traders are much interested in the latter. If you are a trader, it makes complete sense to not only find the metric that other traders are using to judge companies (even if that metric is a weak measure of value and subject to manipulation) but to make estimating that metric the center of your investment strategy.
It is for this reason that I call this the "earnings game", where analysts and investors form expectations about a metric (earnings per share, revenues, number of users/subscribers), companies try to deliver actual numbers that beat these expectations and markets then react to the reports. I describe the process in detail in this post, with the aggregate evidence on both the market reaction to earnings reports as well as the post-report price adjustment process.

Pricing Metrics and the Corporate Life Cycle
While it is true that the predominant metric used to judge a company is earnings per share, investors sometimes use other metrics, rewarding Twitter for increasing its user base more than expected, pushing up Facebook's stock price for its success in mobile advertising and leaving Apple's stock price unchanged on the news that it sold more iPhones than expected, but fewer iPads. Given that investors choose one or two metrics on which they judge company performance, how do they decide which metric to use for a company? Using the narrative adjustment categorization that I introduced in my last post can provide some perspective:
  • For firms where the prime concern that investors have is about narrative breaks, the pricing metric reflects that concern. Thus, with young start-ups, investors may focus on cash in hand or access to capital, as proxies for survival risk. For distressed companies, the pricing metric may be linked to the company's capacity to service debt, interest coverage ratios or debt payments coming due.
  • For firms that have well established narratives, i.e., firms that have established business models in clearly defined markets, the focus will be on narrative shifts, and small ones at that. Given that the market size is stable and market shares are sticky, you can see why investors pay attention to earnings per share and react to surprises on that score. 
  • For firms where the narrative is still taking form, investor will shift away from earnings not only to top-line numbers (like revenues) but to other measures that are related to narrative change. With social media companies, for instance, that explains why investors pay attention to the number of users or measures of user intensity, on the assumption that companies can exploit larger values for either to enter new markets. 
Thus, the pricing metrics (and multiples) that investors focus on will vary across the life cycle of a company and this is the point that I hope to bring through in the picture below:
Life Cycle, Pricing Metrics & Multiples
Thus, early in the process, it is not irrational to focus on market potential or users, but as a company matures, the attention will inevitably turn to profitability and cash flows. Using real-world examples to illustrate this shift, a company like Yo, priced recently at $10 million by investors, even though it really has no product or service to speak of at the moment, is being priced entirely on market potential (all those people with smart phones whose notification center is accessible to apps like Yo). Moving further up the life cycle, consider Snapchat, a company that has a product with tens of millions of users but has not figured out a way to monetize them, it is the number of users and the frequency of their use that drives its pricing (with Alibaba willing to pay $10 billion for them). With Twitter, Linkedin and Yelp, investors seem to be making a transition, where revenues are clearly part of the equation, even though the number of users is still a key number. With Google and Apple, companies with established business models, revenue growth is a consideration but earnings clearly dominate.   

The Dangers of Pricing Metrics
While it is easy to see why investors focus on one or two metrics, when measuring company performance, the dangers of using these short cuts are manifold:
  1. Missing the rest of the story: The value of a business is driven by many determinants, including its capacity to generate profits (and cash flows) from existing assets, the expected growth rate in these earnings & the efficiency with which this growth is delivered and the risk in future cash flows. No single metric will ever capture all of these factors, and in using any metric (number of users, revenues, earnings), you are in effect assuming that everything else that drives value remains unchanged. To illustrate, a company that reports higher earnings per share but does so because it has entered riskier businesses, may see its stock price jump on the earnings report, even though its value has dropped.
  2. Tunnel vision: It is natural to develop tunnel vision, when your focus narrows. Investors and analysts who spend all of their time and resources refining their estimates of one or two metrics, whether they be revenues or number of users, will soon care just about those numbers and ignore the rest of the data (and story). 
  3. Game Playing: Once companies recognize that investors are focused on one or two metrics, it is natural for them to play the game as well. Thus, if analysts are unduly focused on earnings per share, companies will play accounting games to make their earnings per share look better than they should. I argued that the market's focus on the number of users at social media companies best explained why Facebook would pay $19 billion for Whatsapp
  4. Transition Phases: As the life cycle picture illustrated, investor focus does change as a company moves through the life cycle. It is therefore a given that at some point in time, as social media companies evolve and grow, investors will stop looking at the user base and focus first on user engagement with business models (revenues) and then on the profitability of these models (earnings). However, these transitions are unpredictable, and when they do occur, companies that think that they are playing by market rules (delivering more users than markets expected) may be shocked to discover that the rules have changed (and that they are being punished for losing money). 
The Bottom Line
To play the earnings report game, if you are a trader, you have to be able to both pinpoint the metric that markets will react to (which may not always be the metric that analysts following the company are focusing on) and be better at estimating how the company will do on that metric than other investors. If you can play that game well, it is a lucrative one but it is also risky, since seismic shifts can occur from quarter to quarter. That is why it may behoove traders to understand the narrative/value process described in my last post, even if they choose not to invest based upon that process.

  1. Earnings Surprises, Price Reaction and Value
  2. Winning (losing) by losing (winning): The Power of Expectations
  3. Reacting to Earnings Reports: Narrative Adjustments to Value
  4. Reacting to Earnings Reports: Pricing Metrics and Market Reaction
  5. Reacting to Earnings Reports: Let's get real!

Reacting to Earnings Reports: Narrative Adjustments and Value Effects

Reality shows seem to have taken over much of television, but I am not a fan for two reasons. The first is that I don't enjoy watching people who are either so psychologically damaged that they like living their lives in a fishbowl or are so economically desperate that they do not have a choice. The second is that while I recognize the draw of these shows comes from their unscripted nature, I prefer getting my reality jolts from two other forums. The first is live sports, where the allure is that no matter how scripted a sport is, there are those moments of magic, where anything can happen. The second is financial markets, which delight in bringing investors and especially market experts to their knees by behaving in unpredictable ways. Just as ratings week is when television shows are made or ended, earnings season is when the markets deliver their biggest surprises. Building on a theme I introduced about narrative and numbers in an earlier post, I would argue that earnings reports are the vehicles that we should use to confirm, reject or modify narratives (and thus value).

Narrative Adjustments: The Real World Intrudes
In my post on narrative and numbers, I argued that valuation acts as a bridge between the story tellers and number-crunchers and that in a good valuation, every number should be part of a story and that the story has to be checked for viability against history, common sense and data. I also argued that big differences in value can be attributed to differences in narratives rather than differences in numerical assumptions.

So, let’s say that you have taken this message to heart, constructed a defensible narrative and converted that narrative into a valuation. That narrative, and the valuation that is built on it, cannot be etched in stone, since the real world will deliver surprises, positive as well as negative, that should lead you to revisit your narrative. While some of these narrative-changers can come from macro economic developments and occasional news stories about the company, earnings reports remain the primary mechanism for delivering news about companies. While much of the focus in these earnings reports remains on the bottom line, often defined as earnings per share, and investors react to whether that number comes in above or below expectations, it is also true that these reports can contain news that should lead you to revisit your narrative and change your valuation. I would broadly classify these narrative effects into three categories:
  1. Narrative breaks/ends: The most dire scenario is news that leads you to conclude that your existing narrative for the firm is no longer operative, rendering your original valuation moot. Narrative breaks are almost always bad news and can be caused by legal events (e.g., the Supreme Court decision that brought Aereo's disruptive efforts to a halt), economic events (e.g., the Argentine government default and its effect on the valuation of any Argentine company), government actions (the nationalization of a company or the removal of protection from competition) or credit events (a company's failure to make a debt payment, resulting in bankruptcy). 
  2. Narrative shifts: In most cases, earnings reports don’t deliver large surprises about a company’s business model or direction, and this is especially the case for mature companies. Instead, you get is information that lead you reassess your narrative and extend that reassessment into new estimates for the company’s future revenues, earnings and cash flows, shifting value. This is the exercise, for instance, that I chronicled in a post about Apple's earnings report on April 23, 2013, where I deconstructed the information in that report and looked at its impact on key inputs into Apple's valuation. Note that narrative shifts, especially if they are consistently positive or negative can lead to major changes in company value over time. This has been the case for a company like Google, which in spite of all its innovations and new services, derives almost all of its revenues still from online advertising but has done it so well that it has managed to expand both the size of the overall online advertising market and its share of it over the last decade.
  3. Narrative changes (expansion/contraction): In some cases, earnings reports deliver news that may be peripheral in terms of the impact it has on operating numbers (revenues, earnings) but are significant because they signal that the company’s business model is changing in ways that you had not anticipated in your original narrative. No company has epitomized this process better than Amazon, a company that I have valued multiple times since 1998. In my very first valuation of Amazon, I valued it as a book retailer, but in subsequent valuations, I have seen it evolve first into a specialty retailer, then become a general retailer, and in recent years, make forays into the media, entertainment and cloud storage businesses. I am sure that there are people more prescient than me who saw all of this coming in 2000, but I sure did not, though notwithstanding that vision failure, I still found Amazon to be cheap (and a good investment) at least four times in the last decade.
Narrative Shifts versus Narrative Changes: Shades of Gray
This distinction between narrative breaks, shifts and changes is a good one to think about when you look at earnings reports, but it is not alway easy to make. In fact, it is entirely possible that you and I could look at the same earnings report and come to very different conclusions about its impact on narrative for three reasons:
  1. Your classification (break, shift or change) will depend upon your initial narrative: The more expansive your initial narrative, the more likely it is that you will see narrative shifts, rather than radical changes. Let me take Uber as an example, and use the contrast of my narrative with Bill Gurley's  to illustrate this point. Uber is still private but any information that I receive about Uber’s success in suburban markets will be a narrative change for me, since my base valuation is built on the presumption that Uber will be successsful as a urban car service company. For Bill Gurley, whose base narrative already incorporates expectations of sucess in suburban markets, this news will be more of a narrative shift than a change.
  2. The lines between the categories can become fuzzy: Even for a given narrative, information in an earnings report or news story can fall in gray areas and be tough to categorize. For instance, Facebook’s better than expected performance in the mobile advertising market in its last few quarters may be viewed by some as just a narrative shift (giving them a larger market share of the online advertising business) and by others as a narrative change (with the mobile users giving them a platform that they can use to enter other online businesses), with very different implications for Facebook's value.
  3. Narrative adjustments can vary across time, for the same company: As a company reports earnings over many periods, you can see narrative shifts in some periods and narrative changes in others, good news in some and bad news in others. Staying with Amazon, a company that I used as my example of a successful narrative changer, the market reaction to the last two earnings reports has been brutal, as markets seem less  focused on revenue growth (which continues to be extraordinary) and more on profit margins (which have been abysmal). A narrative shift may be occurring, where investors are reassessing Amazon’s potential profitability in steady state and concluding that it will make less money than they thought it would.
Narrative Adjustments: Reactive and Proactive Valuation Responses
How do we deal with these narrative adjustments in conventional valuation? Very badly, I am afraid. If your valuation is a rigid discounted cash flow valuation, your response to narratives breaking, shifting or changing is denial. In that static world, the narrative remains constant, your valuation inputs stay the same, intrinsic value never changes and it is the market that is viewed as being at fault for its volatility. Even those who claim to use more dynamic processes for valuation often stay within the traditional framework, trying to increase discount rates to reflect potential narrative breaks and growth rates to capture narrative changes.  Finally, even the best among us tend to be more reactive than proactive, adjusting value for narrative adjustments that have already occurred, but not making any attempt for the potential for adjustments in the future.

By definition, since you cannot anticipate the unexpected, you have to draw on the full arsenal of valuation tools to both react to narrative adjustments as they happen and to proactively incorporate the possibility of future adjustments into value.  While the reactive effects of narrative adjustments on value are straight forward, incorporating expectations of future narrative adjustments into current value is much more difficult to do and the table below lists some of the tools that we have available:
  1. When valuing companies where the possibility of a narrative break is high, either because they are young, start-ups or debt-ridden, distressed companies, you have to bring in the likelihood of the narrative ending, explicitly as a probability. (See my papers on valuing young companies and declining, distressed companies for estimation tools that you can use to make this judgment)
  2. For narrative shifts, where the effects play out as better-than or worse-than expected revenue growth and profit margin numbers, the tool that works best is a simulation, where you use probability distributions for the inputs into your valuation rather than just your base case numbers and estimate a distribution of values. (See my paper on probabilistic approaches in valuation)
  3. For narrative changes, which, by definition, are unanticipated and unexpected, you have to treat them as real options and value them as such. (See my paper on the promise and peril of real options)
In the table below, I summarize this discussion of narrative adjustments, how they affect value when they do occur and how you can proactively bring them into your valuations:

The Bottom Line
Earnings reports remain a company's key delivery mechanism for news about both its operations and plans of the future, but they are filled with distractions. Paraphrasing Nate Silver, it is important that we separate the signal from the noise and use these reports to revisit our narratives and valuations. If you are an avid market watcher, you may feel that I am over analyzing the earnings process and that the market reaction to an earnings report has little to do with narrative shifts or long term value, and more to do with meeting investor expectations on key numbers (earnings per share, revenues, number of users etc.). I don't disagree with you and in my companion post, I will focus on the metrics that investors use to judge earnings reports, why these metrics might vary across companies and over time and the potential danger of letting these metrics determine investment decisions.