Friday, December 7, 2018

Is there a signal in the noise? Yield Curves, Economic Growth and Stock Prices!

The title of this post is not original and draws from Nate Silver's book on why so many predictions in politics, sports and economics fail. It reflects the skepticism with which I view many 'can't fail" predictors of economic growth or stock markets, since they tend to have horrendous track records. Over the last few weeks, as markets have gyrated, market commentators have been hard pressed to explain day-to-day swings, but that has not stopped them from trying. The explanations have shifted and morphed, often in contradictory ways, but few of them have had staying power. On Tuesday (December 4), as the Dow dropped 800 points, following a 300-point up day on Monday, the experts found a new reason for the market drop, in the yield curve, with an "inverted yield curve", or at least a portion of one, predicting an imminent recession. As with all market rules of thumb, there is some basis for the rule, but there are shades of gray that can be seen only by looking at all of the data.

Yield Curves over time
The yield curve is a simple device, plotting yields across bonds with different maturities for a given issuing entity. US treasuries, historically viewed as close to default free, provide the cleanest measure of the yield curve,  and the graph below compares the US treasury yield curve at the start of every year from 2009 to 2018, i.e., the post-crisis years:
The yield curve has been upward sloping, with yields on longer term maturities higher than yields on short term maturities, every year, but it has flattened out the last two years. On December 4, 2018, the yields on treasuries of different maturities were as follows:
The market freak out is in the highlighted portion, with 5-year rates being lower (by 0.01-0.02%) than 2-year or 3-year rates, creating an inverted portion of the yield curve. 

Yield Curves and Economic Growth: Intuition
To understand yield curves, let's start with a simple economic proposition. Embedded in every treasury rate are expectations of expected inflation and expected real real interest rates, and the latter
Interest Rate = Expected Inflation Rate + Expected Real Interest Rate
Over much of the last century, the US treasury yield curve has been upward sloping, and the standard economic rationalization for it is a simple one. In a market where expectations of inflation are similar for the short term and the long term, investors will demand a "maturity premium" (or a higher real interest rate) for buying longer term bonds, thus causing the upward tilt in the yield curve.  That said, there have been periods where the yield curve slopes downwards, and to understand why this may have a link with future economic growth, let's focus on the mechanics of yield curve inversions. Almost every single yield curve inversion historically, in the US,  has come from the short end of the curve rising significantly, not a big drop in long term rates. Digging deeper, in almost every single instance of this occurring, short term rates have risen because central banks have hit the brakes on money, either in response to higher inflation or an overheated economy. You can see this in the chart below, where the Fed Funds rate (the Fed's primary mechanism for signaling tight or loose money) is graphed with the 3 month, 2 year and 10 year rates:
Interest Rate Raw Data
As you can see in this graph, the rises in short term rates that give rise to each of the inverted yield curve episodes are accompanied by increases in the Fed Funds rate. To the extent that the Fed's monetary policy action (of raising the Fed funds rate) accomplishes its objective of slowing down growth, the yield slope metric becomes a stand-in for the Fed effect on the economy, with a more positive slope associated with easier monetary policy. You may or may not find any of these hypotheses to be convincing, but the proof is in the pudding, and the graph below, excerpted from a recent Fed study, seems to indicate that there has been a Fed effect in the US economy, and that the slope of the yield curve has operated as proxy for that effect:
Federal Reserve of San Francisco
The track record of the inverted yield curve as a predictor of recessions is impressive, since it has preceded the last eight recessions, with only only one false signal in the mid-sixties. If this graph holds, and December 4 was the opening salvo in a full fledged yield curve invasion, the US economy is headed into rough waters in the next year.

Yield Curves and Economic Growth: The Data
The fact that every inversion in the last few decades has been followed by a recession will strike fear into the hearts of investors, but is it that fool proof a predictor? Perhaps, but given that the yield curve slope metrics and economic growth are continuous, not discrete, variables, a more complete assessment of the yield curve's predictive power for the economy would require that we look at the strength of the link between the slope of the yield curve (and not just whether it is inverted or not) and the level of economic growth (and not just whether it is positive or negative).

To begin this assessment, I looked at the rates on  three-month and one-year T.Bills and the two, five and ten-year treasury bonds at the end of every quarter from 1962 through the third quarter of 2018:
Following up, I look at five yield curve metrics (1 year versus 3 month, 2 year versus 3 month, 5 year versus 2 year, 10 year versus 2 year and 10 year versus 3 month), on a quarterly basis from 1962 through 2018, with an updated number for December 4, 2018. 
For the most part, the yield curve metrics move together, albeit at different rates. I picked four measures of the spread, one short term (1 year versus 3 month), one medium term (5 year versus 2 year) and two long term (10 year versus 2 year, 10 year versus 3 month) and plotted them against GDP growth in the next quarter and the year after. 
Interest Rate Raw Data
The graph does back up what the earlier Fed study showed, i.e., that negatively sloped yield curves have preceded recessions, but even a cursory glance indicates that the relationship is weak. Not only does there seem to be no relationship between how downwardly sloped the yield curve is and the depth of the recessions that follow, but in periods where the yield curve is flat or mildly positive, subsequent economic growth is unpredictable. To get a little more precision into the analysis, I computed the correlations between the different yield curve slope metrics and GDP growth:

Taking a closer look at the data, here is what I see;
  1. It is the short end that has predictive power for the economy: Over the entire time period (1962-2018), the slope of the short end of the yield curve is positively related with economic growth, with more upward sloping yield curves connected to higher economic growth in subsequent time periods. The slope at the long end of the yield curve, including the widely used differential between the 10-year and 2-year rate not only is close to uncorrelated with economic growth (the correlation is very mildly negative).
  2. Even that predictive power is muted: Over the entire time period, even for the most strongly linked metric (which is the 2 year versus 1 year), the correlation is only 29%, for GDP growth over the next year, suggesting that there is significant noise in the prediction. 
  3. And 2008 may have been a structural break: Looking only at the last ten years, the relationship seems to have reversed sign, with flatter yield curves, even at the short end, associated with higher real growth. This may be a hangover from the slow economic growth in the years after the crisis, but it does raise red flags about using this indicator today.
How do you reconcile these findings with both the conventional wisdom that inverted yield curves are negative indicators of future growth and the empirical evidence that almost every inversion is followed by a recession? It is possible that it is the moment of inversion that is significant, perhaps as a sign of the Fed's conviction, and that while the slope of the yield curve itself may not be predictive, that moment that the yield curve inverts remains a strong indicator. 

Yield Curves and Stock Returns
As investors, your focus is often less on the economy, and more on stock prices. After all, strong economies don't always deliver superior stock returns, and weak ones can often be accompanied by strong market performance. From that perspective, the question becomes what the slope of the yield curve and inverted yield curves tell you about future stock returns,  not economic growth. I begin the analysis by looking at yield curve metrics over time, graphed against return on US stocks in the next quarter and the next year:
If you see a pattern here, you are a much better chart reader than I am. I therefore followed up the analysis by replicating the correlation table that I reported in the economic growth section, but looking at stock returns in subsequent periods, rather than real GDP growth:
As with the economic growth numbers, if there is any predictive power in the yield curve slope, it is at the short end of the curve and not the long end. And as with the growth numbers, the post-2008 time period is a clear break from the overall numbers.

What does all of this mean for investors today? I think that we may be making two mistakes. One is to take a blip on a day (the inversion in the 2 and 5 year bonds on December 4) and read too much into it, as we are apt to do when we are confused or scared. It is true that a portion of the yield curve inverted, but if history is any guide, its predictive power for the economy is weak and for the market, even weaker. The other is that we are taking rules of thumb developed in the US in the last century and assuming that they still work in a  vastly different economic environment. 

Bottom Line
There is information in the slope of the US treasury yield curve, but I think that we need to use it with caution. In my view, the flattening of the yield curve in the last two years has been more good news than bad, an indication that we are coming out of the low growth mindset of the post-2008 crisis years. However, I also think that the stalling of the US 10-year treasury bond rate at 3% or less is sobering, a warning that investors are scaling back growth expectations for both the global and US economies, going into 2019. The key tests for stocks lie in whether they can not only sustain earnings growth, in the face of slower economic growth and without the tailwind of a tax cut (like they did last year), but also in whether they can continue to return cash at the rates that they have for the last few years.

YouTube Video


  1. Raw data on US treasury rates, GDP growth and Stock Returns

Monday, December 3, 2018

Investing Whiplash: Looking for Closure with Apple and Amazon!

In September, I took a look, in a series of posts, at two companies that had crested the trillion dollar market cap mark, Apple and Amazon, and concluded that series with a post where I argued that both companies were over valued. I also mentioned that I was selling short on both stocks, Amazon for the first time in 22 years of tracking the company, and Apple at a limit price of $230. Two months later, both stocks have taken serious hits in the market, down almost 25% apiece, and one of my short sales has been covered and the other is still looking profitable. It is always nice to have happy endings to my investment stories, but rather than use this as vindication of my valuation or timing skills, I will argue that I just got lucky in terms of timing. That said, given how much these stocks have dropped over the last two months, it is an opportunity to not just revisit my valuations and investment judgments, but also to draw some general lessons about intrinsic valuation and pricing.

My September Valuations: A Look Back
In September, I valued Apple and Amazon and arrived at a value per share of roughly $200 for Apple and $1255 for Amazon, well below their prevailing stock prices of $220 (Apple) and $1950 (Amazon). I was also open about the fact that my valuations reflected my stories for the companies, and that my assumptions were open for debate. In fact, I estimated value distributions for both companies and noted that not only did I face more uncertainty in my Amazon valuation, but also that there was a significant probability in both companies that my assessment (that the stocks were over valued) was wrong. I summarized my results in a table that I reproduce below:
Apple Valuation & Amazon Valuation in September 2018
I did follow through on my judgments, albeit with some trepidation, selling short on Amazon at the prevailing market price (about $1950) and putting in a limit short sell at $230, which was fulfilled on October 3, as the stock opened above $230. With both stocks, I also put in open orders to cover my short sales at the 60th percentile of my value distributions, i.e., $205 at Apple and $1412 at Amazon, not expecting either to happen in the near term. (Why 60%? Read on...) Over the years, I have learned that investment stories and theses, no matter how well thought out and reasoned, don't always have happy endings, but this one did, and at a speed which I did not expect:
My Apple short sale which was initiated on October 3 was closed out on November 5 at $205, while Amazon got tantalizingly close to my trigger price for covering of $1412 (with a low of $1420 on November 20), before rebounding. 

Intrinsic Value Lessons
Every investment, whether it is a winner or a loser, carries investment lessons, and here are mine from my AAPL/AMZN experiences, at least so far:
  1. Auto pilot rules to fight behavioral minefields: If you are wondering why I put in limit orders on both my Apple short sale and my covering trades on both stocks, it is because I know my weaknesses and left to my own biases, the havoc that they can wreak on my investment actions. I have never hidden the fact that I love Apple as a company and I was worried that if I did not put in my limit short sell order at $230, and the stock rose to that level, I would find a way to justify not doing it. For the limit buys to cover my short sales, I used the 60th percentile of the value distribution, because my trigger for buying a stock is that it be at least at the 40th percentile of its value distribution and to be consistent, my trigger for selling is set at the 60th percentile. It is my version of margin of safety, with the caveat being that for stocks like Amazon, where uncertainty abounds, this rule can translate into a much bigger percentage price difference than for a stock like Apple, where there is less uncertainty. (The price difference between the 60th and 90th percentile for Apple was just over 10%, whereas the price difference between those same percentiles was 35% for Amazon, in September 2018.)
  2. Intrinsic value changes over time: Among some value investors, there is a misplaced belief that intrinsic value is a timeless constant, and that it is the market that is subject to wild swings, driven by changes in mood and momentum. That is not true, since not only do the determinants of value (cash flows, growth and risk) change over time, but so does the price of risk (default spreads, equity risk premiums) in the market. The former occurs every time a company has a financial disclosure, which is one reason that I revalue companies just after earnings reports, or a major news story (acquisition, divestiture, new CEO),  and the latter is driven by macro forces. That sounds abstract, but I can use Apple and Amazon to illustrate my point. Since my September valuations for both companies occurred after their most recent earnings reports, there have been no new financial disclosures from either company. There have been a few news stories and we can argue about their consequentiality for future cash flows and growth, but the big change has been in the market. Since September 21, the date of my valuation, equity markets have been in turmoil, with the S&P 500 dropping about 5.5% (through November 30) and the US 10-year treasury bond rate have dropped slightly from 3.07%  to 3.01%, over the same period. If you are wondering why this should affect terminal value, it is worth remembering that the price of risk (risk premium) is set by the market, and the mechanism it has for adjusting this price is the level of stock prices, with a higher equity risk premium leading to lower stock prices. In my post at the end of a turbulent October, I traced the change in equity risk premiums, by day, through October and noted that equity risk premiums at the end of the month were up about 0.38% from the start of the month and almost 0.72% higher than they were at the start of September 2018. In contrast, November saw less change in the ERP, with the ERP adjusting to 5.68% at the end of the month.
    Plugging in the higher equity risk premium and the slightly lower risk free rate into my Apple valuation, leaving the rest of my inputs unchanged, yields a value of $197 for the company, about 1.5% less than my $200 estimate on September 21. With Amazon, the effect is slightly larger, with the value per share dropping from $1255 per share to $1212, about 3.5%. Those changes may seem trivial but if the market correction had been larger and the treasury rate had changed more, the value effect would have been larger.
  3. But price changes even more: If the fact that value changes over time, even in the absence of company-specific information, makes you uncomfortable, keep in mind that the market price usually changes even more. In the case of Apple and Amazon, this is illustrated in the graph below, where I compare value to price on September 21 and November 30 for both companies:
    In just over two months, Apple's value has declined from $201 to $196, but the stock price has dropped from $220 to $179, shifting it from being overvalued by 9.54% to undervalued by 9.14%. Amazon has become less over valued over time, with the percentage over valuation dropping from  55.38% to 39.44%. I have watched Apple's value dance with its price for  much of this decade and the graph below provides the highlights:
    From my perspective, the story for Apple has remained largely the same for the last eight years, a slow-growth, cash machine that gets the bulk of its profits from one product: the iPhone. However, at regular intervals, usually around a new iPhone model, the market becomes either giddily optimistic about it becoming a growth company (and pushes up the price) or overly pessimistic about the end of the iPhone cash franchise (and pushes the price down too much). In the face of this market  bipolarity, this is my fourth round of holding Apple in the last seven years, and I have a feeling that it will not be the last one.
  4. Act with no regrets:  I did cover my short sale, by buying back Apple at $205, but the stock continued to slide, dropping below $175 early last week. I almost covered my Amazon position at $1412, but since the price dropped only as low as $1420, my limit buy was not triggered, and the stock price is back up to almost $1700. Am I regretful that I closed too early with Apple and did not close out early enough with Amazon? I am not, because if there is one thing I have learned in my years as an investor, it is that you have stay true to your investment philosophy, even if it means that you leave profits on the table sometimes, and lose money at other times. I have faith in value, and that faith requires me to act consistently. I will continue to value Amazon at regular intervals, and it is entirely possible that I missed my moment to sell, but if so, it is a price that I am willing to pay.
  5. And flexible time horizons: A contrast that is often drawn between investors and traders is that to be an investor, you need to have a long time horizon, whereas traders operate with windows measured in months, weeks, days or even hours. In fact, one widely quoted precept in value investing is that you should buy good companies and hold them forever. Buy and hold is not a bad strategy, since it minimizes transactions costs, taxes and impulsive actions, but I hope that my Apple analysis leads you to at least question its wisdom. My short sale on Apple was predicated on value, but it lasted only a month and four days, before being unwound. In fact, early last week, I bought Apple at $175, because I believe that it under valued today, giving me a serious case of investing whiplash. I am willing to wait a long time for Apple's price to adjust to value, but I am not required to do so. If the price adjusts quickly to value and then moves upwards, I have to be willing to sell, even if that is only a few weeks from today. In my version of value investing, investors have to be ready to hold for long periods, but also be willing to close out positions sooner, either because their theses have been vindicated (by the market price moving towards value) or because their theses have broken down (in which case they need to revisit their valuations).
Bottom Line
As investors, we are often quick to claim credit for our successes and equally quick to blame others for our failures, and I am no exception. While I am sorely tempted to view what has happened at Apple and Amazon as vindication of my value judgments, I know better. I got lucky in terms of timing, catching a market correction and one targeted at tech stocks, and I am inclined to believe that  is the main reason why my Apple and Amazon positions have made me money in the last two months. With Amazon, in particular, there is little that has happened in the last two months that would represent the catalysts that I saw in my initial analysis, since it was government actions and regulatory pushback that I saw as the likely triggers for a correction. With Apple, I do have a longer history and a better basis for believing that this is market bipolarity at play, with the stock price over shooting its value, after good news, and over correcting after bad news, but nothing that has happened  to the company in the last two month would explain the correction. Needless to say, I will bank my profits, even if they are entirely fortuitous, but I will not delude myself into chalking this up to my investing skills. It is better to be lucky than good!

YouTube Video

Blog Posts
  1. Apple and Amazon at a Trillion $: A Follow-up on Uncertainty and Catalysts (September 2018)
  2. An October Surprise: Making Sense of Market Mayhem (October 2018)

Wednesday, November 14, 2018

The GE End Game: Bataan Death March or Turnaround Play?

It seems like ancient history, but it was just 2001, when GE was the most valuable company in the world, commanding a market capitalization in excess of $500 billion. The quintessential conglomerate, with a presence in almost every part of the global economy, it seemed to have been built to withstand economic shocks and was the choice for conservative investors, scared of the short life cycles and the volatile fortunes of its tech challengers. Unlike other aging companies like Sears that have decayed gradually over decades, GE's fall from grace has been precipitous , with the rate of decline accelerating the in the last two years. As a new CEO is brought in, with hopes that he will be a savior, it is the right time to both look back and look forward at one of the globe's most iconic companies.

GE: A Compressed History
GE's roots can be traced back to Thomas Edison and his invention of the light bulb. The company that Edison founded in 1878, Edison General Electric, was combined with two other electric companies to create General Electric in 1892. The company established its first industrial lab in 1900 and it would not be an exaggeration to say that it revolutionized not just the American home, with its appliances, but changed the way Americans live. For much of of the twentieth century, though, GE remained an appliance company, though it made forays into other businesses. It was in 1980, when Jack Welch became the CEO of the company, that the company started its march towards what it has become today.

The Market History
The first place to start, when looking at GE, is to see how markets have viewed it, over its life. Skipping over the first half of GE's life, the graph below looks at the growth (and recent decline) of GE's market capitalization over time:

As you can see, GE was a solid but unspectacular investment from 1950 to 1980, and exploded in value in the 1980s and 1990s, with Jack Welch at its helm, and reached its most valuable company in the world status in 2001. Under Jeff Immelt, his successor, the stock continued to do well, but it dropped with the rest of the market as the dot com bubble burst, but then recovered leaving into the 2008 crisis. That crisis was devastating for the company and while it did recover somewhat in the years after, the bottom has clearly dropped out in the last two years, with Jeff Flannery at the top of the company.

The Operating History
To get operating perspective on how the company has evolved over time, we looked at how GE"s key operating metrics (revenues, EBITDA, net income) have evolved since 1950:

In keeping with our earlier market cap assessment, between 1950 and 1970, GE was a good but not exceptional company, delivering solid revenue growth and decent margins. Under two CEOS, Reginald Jones in the 1970s and Jack Welch in the two decades thereafter, the company transformed itself. Jones helped the company navigate through the turbulent period of high inflation and oil prices, holding margins steady and delivering double digit revenue growth. Welch made himself the stuff of legend, by doubling margins and pushing the company to the top of the market cap ranks by the time he left the firm. His successor Jeff Immelt faced the unenviable task of following Welch, but managed to keep revenues growing and delivered high margins until 2008, when the bottom fell out for the company. 

The Business Mix Shift
To understand GE's current plight, we have to go back to Welch's tenure as CEO, when he remade the firm, by moving it away from its domestic and manufacturing roots and giving it a global and multi-business focus. GE's biggest leap during that period was into the financial services business, and one reason Welch was attracted to the financial services business was its capacity to generate high profits with relatively little investment. By the late 1990s, GE Capital was the engine driving GE's growth, accounting for almost 48% of revenues in 1998 and as you can see in the graph below, it continued to do so for much of the first decade of Immelt's stewardship:

In 2008, when the crisis hit financial service firms had, GE was significantly exposed, and in the years since, GE has retreated not just from the financial services business but also from its entertainment bets (with the sale of NBC to Comcast) and from the appliance business (now owned by Haier). GE's current business mix, broken down into more detail, is shown in the pie chart below:
GE Annual Report for 2017 (Invested Capital, allocated based upon assets by business)
Today, GE is in three businesses (aviation, healthcare and transportation) that have low growth and high profitability (margins and returns on capital), in three energy-related businesses (power, renewable energy and oil) with higher growth but low profitability (margins & returns on capital), one business (lighting) that is fading quickly and one (capital) that is declining, but dragging value down with it. Note also that the collective profits reported across businesses is  before corporate expenses and eliminations of $3.83 billion (not counting a one-time restructuring charge of $4.1 billion) that effectively wipe out about half of the operating profits. When computing return on capital, I allocated these expenses to the businesses, based upon revenues, and used a 25% effective tax rate, and while GE as a whole did not deliver a return that meets its cost of capital requirements in 2017, aviation, healthcare and transportation clear their hurdle rates by plenty. Replacing 2017 income in each business with a normalized value (computed using the average margins in each business between 2013 and 2017) improves the return on capital at the power and renewable energy businesses, but the overall conclusion remains the same. GE, as a company, does not look good, but it does have significant value creating businesses.

Corporate Life Cycle
While there are different ways of framing GE's current standing, I will use the corporate life cycle, since it encapsulates the challenges facing the company.

GE's light bulb moment might have been in Thomas Edison's lab in 1878, but at an official corporate age of 126 years, GE is an ancient company and its problems reflect its age. Other than renewable energy, all of GE's businesses are mature or declining, and by the laws of mathematics, GE itself is a mature to declining company.  Any story that you tell about GE going forward has to reflect this reality, and there are three possible ones that can lead to different values.
1. Break it up: If GE at its peak represented the glory of conglomerates, its current plight is a sign of how far conglomerates have fallen in the world. Across the world, multi business companies are finding themselves under pressure to break up and in many cases, their stockholders will be better off if they do. To gain from a break up, though, here are some of the things that have to be true. 
  • Separable businesses: The different businesses have to be separable, since leakages and synergies across businesses can make it more difficult to cleave off pieces to sell or spin off. On this count, GE is probably on safe ground, since its businesses (other than GE Capital) are self standing, for the most part, with little in terms of cross business effects. 
  • Willing buyers: There have to be potential buyers who are willing to pay prices for the pieces that exceed what they will generate as value for the holding company, as going concerns, and those higher prices either have to come from potential synergies or changed management. None of GE's businesses seem alluring enough to attract multiple bidders, willing to pay premium prices, and given GE's shaky bargaining position, it is more likely than not that a rush to unload businesses will do more harm than good. 
  • Corporate Waste (at HQ):  A large chunk of the corporate overhead has to viewed as wasteful, with a big drop in corporate expenses accompanying the breakup. How much of the corporate expense of $3.8 billion that GE reported in 2017 is wasteful and could be eliminated with targeted cost cuts? Looking at the breakdown of these expenses, just about $2.2 billion in for covering pension obligations and breaking up the company will not relieve the company of its contractual obligations. Some of the remaining $1.6 billion may be fat that can be cut, but even cutting the entire amount (which would be a tall order) will not turn the company around.
Since GE will be trying to sell these businesses to buyers today, this is a pricing and not a valuation exercise, and I have estimate a pricing for GE's businesses below, using an EBITDA recomputed using the average operating margin in each business over the last five years to compute operating income and allocating corporate expenses to the divisions, based upon revenues. To convert the EBITDA to an estimated value, I used the EV/EBITDA multiples of the peer group:
Download spreadsheet
If GE is able to get buyers to pay industry-level multiples of EBITDA for each of these businesses, it will be able to net about $103 billion for its equity investors, higher than the market capitalization on November 14 of $72 billion. The problem, though, is that fire sales of entire companies almost never deliver the expected proceeds, as buyers, recognizing desperation, hold back. In fact, GE's attempts to extricate itself from a portion of its Baker-Hughes investment in the last few days show that these sales will occur at a discount.

2. Retrench and Reshape: The second choice for GE is to retrench and perhaps renew itself, not as a growth company, but as a stable, high margin company in businesses where it has a competitive advantage. In broad terms, the roadmap for GE to succeed in this path is a simple one,  shrinking or selling off pieces of its low-margin businesses, exiting the capital business and consolidating its presence in the aviation, healthcare and transportation businesses. To get a better sense of what the businesses would be worth, as continuing operations, I valued each of GE's business, using simplistic assumptions: I used the sector cost of capital for each business, set growth in the next five years equal to revenue growth in each of GE's businesses in the last five years and normalized operating income based upon the average operating margin that each of GE's businesses have delivered over the last five years:
Download spreadsheet
The value that I derive for equity is lower than the $103 billion that I estimated in the last section, but it does not require any near term fire sales at discounts. There are two big challenges that GE will face along the way. The first is that GE is saddled with a significant debt obligation, a legacy of GE Capital, that will not fade away quickly, and the debt obligations represent a clear and present danger to the firm.  One reason for the rapid drop in GE's stock price in the last few weeks has been the deterioration in the company's credit standing, as can be seen in the rising default spreads for the company in the CDS market.

The reason that GE is trying to sell some of its stake in Baker and Hughes to pay down debt, but bond markets are skeptical, with good reason. The second is that GE Capital is now more burden than benefit to investors. In the valuation table, note that the value that I have estimated for GE Capital's operations ($27 billion) is much lower than GE Capital debt ($51 billion); in fact, I derive very similar results in the pricing. Put differently, in my valuation, I foresee the cost of exiting GE capital to be $24 billion in today's terms, but spread out over time.  If GE can navigate its way through its debt payments to becoming a more focused company, with constrained ambitions, it could survive and reclaim its place as a holding for a conservative value investor.

3. Reincarnate (or the Bataan Death March): There is a third option that GE shareholders have to hope and pray that GE does not take, where the company tries to recapture its old glory, throwing caution to the winds and reinvesting large amounts in new businesses, or worse still, large acquisitions. While there is no indication that Larry Culp, GE's new CEO, has grandiose plans for the company, that may be because the company is in crisis today. If as the crisis passes, Culp is tempted to make himself the second coming of Jack Welch, the company will follow the path of other aging companies that refuse to act their age, spending billions on cosmetic surgery (acquisitions) before finally capitulating. If there is a role model that Mr. Culp should follow, it is less that of Steve the Visionary, and more that of Larry the Liquidator

General Lessons
Given its age, it should come as no surprise that GE has been the subject of more case studies than perhaps any other company in the world. In its earlier days, it was used as an example of professional management, and during Jack Welch's years, it was held up as an illustration of how aging manufacturing companies can remake themselves, with enlightened management at the top. Now that it is in trouble, I think that we look back at the last four decades and draw a different set of lessons:
  1. Conglomeration was, is and always will be a bad idea: I never understood the allure of conglomerates, even in their heyday. Only a corporate strategist could argue that combining companies in different businesses under one corporate umbrella, paying hefty premiums along the way to acquire these holdings, creates value, ignoring the logic that you and I as stockholders can create our own diversified and customized portfolios, without paying the same premium. If there is a lesson to learn from GE's fall from grace, it is that even the best conglomerates are built on foundations of sand. Note, though, that while this lesson may be learned for the moment, it will be forgotten soon, as are most other business lessons are, and we will surely repeat the cycle again in the future.
  2. Complexity has a cost: As I was going through GE's annual report, I was reminded again of why I have always described my vision of hell as having to value GE over and over and over again, for eternity. This company, through its actions and by design, made itself into one of the most complex companies in history, operating in dozens of businesses and across the world, with GE Capital acting as the cherry on the complexity cake, a gigantic financial service firm embedded in a large conglomerate. While that complexity served GE well in its glory days, allowing it to hide mistakes from sloppy acquisition practices and bets gone bad, it has bedeviled the company since 2008. Investors trying to navigate their way through the company's financials often give up and move on to easier prey. It may be too late for GE to do much about this problem, but as Asian companies rise in market capitalization, you are seeing new complex behemoths coming into play across the world.
  3. Easy money has a catch: I know that 20/20 hindsight is both easy and unfair, but GE's experiences with GE Capital bring home an age-old business truth that when a business looks like it can make you easy money, there is always a catch. Jack Welch initial foray into and subsequent expansion of GE Capital was built on the allure that it was a lot easier to make money in financial services than in manufacturing. From the perspective of having limited capital investment and growing quickly, that was true, but financial service firms through history have always had periods of plenty interspersed with bouts of gut-wrenching and intense pain, when borrowers start defaulting and capital markets freeze up. By making GE Capital such a big part of GE, Welch bet the farm on its continued success, and that bet went sour in 2008.
  4. The Savior CEO is a myth: I come to neither bury nor praise Jack Welch, but notwithstanding the fact that he has been gone almost two decades from the firm, GE remains the house that Jack built. Since Welch got the glory that came from GE's rise in the last twenty years of the last century, he deserves a portion of the blame for what has happened since. Don't get me wrong! Jack Welch was an inspirational top manager, a man with vision and drive, but he was also an imperial CEO, who made his board of directors a rubber stamp for his actions. As we look at a new generation of successful companies, this time in the technology space (the FANG stocks and the Chinese giants), with visionary founders at the top, it is worth remembering that power left unchecked in any person (no matter how smart and visionary) is dangerous.
The Bottom Line
As many of you know, I believe that every valuation has to have a story. With some companies, like Amazon and Google, the story is uplifting and optimistic, and the valuations follow, but they still might not be good investments, since their prices may be even higher. My story for GE is not an upbeat one, but if it (and its management) acts its age, accepts that slower or no growth is what lies in the future and does not over reach, it is a good investment. I believe that the market has over corrected for GE's many faults, and at the current stock price, that it is significantly under valued. I will buy GE, but I will do so with open eyes, not expecting (or wanting) dividends to be paid until the debt gets paid down and the company exits the capital business with as much grace (and as few costs) as it can muster. 

YouTube Video


Monday, October 29, 2018

An October Surprise? Making Sense of the Market Mayhem!

I don't know what it is about October that spooks markets, but it certainly feels like big market corrections happen in the month. As stocks have gone through contortions this month, more down than up, like many of you, I have been looking at my portfolio, wondering whether this is the crash that the market bears have been warning me about since 2012, just a pause in a continuing bull market or perhaps a prognosticator of economic troubles to come. If you are expecting me to give you the answer in this post, I would stop reading, since reading market tea leaves is not my strength. That said,  I have been wrestling with what, if anything, I should be doing, as an investor, in response to the market movements. As in previous market crises, I find myself going back to a four-step process that I hope gets me through with my sanity intact. 

Step 1: Hit the pause button
The first casualty of crisis is good sense, as I mistake my panic response for instinct, and almost every action that I feel the urge to take in the heat of the moment is driven by fear and greed, not reason. No amount of rational thinking or studying behavioral finance will cure me of this affliction, since it is part of my make up as a human being, but there are three things that I find help me manage my reactions:
  1. Take a breath:  When faced with fast-moving markets, I have to force myself to consciously slow down. It helps that I don't work as a trader or a portfolio manager, since part of your job is to look like you are in control, even when you are not. 
  2. Turn off the noise: Turn back the clock about four decades and assume that you were a doctor, a lawyer or a factory worker with much of your wealth invested in stocks. If markets were having a bad day, odds were that you would not even have heard about it until you got home and turned on the news, and even then, you would have been fed scraps of information about Dow, perhaps a 2-minute discussion with a market expert, and you would have then turned on your favorite sitcom. Today, not only can you monitor your stocks every moment of your working day, you can trade on your lunch break and stream CNBC on to your desktop. That may make you a more informed investor, or at least an investor with more information, but I am not sure that constant feedback is healthy for your portfolio, especially in periods like this one. I don't have a Bloomberg terminal on my desk, a ticker tape running on  my computer or stock apps on my phone, and I am happy that I don't during periods like this month.
  3. Don't play the blame game: Every market correction has its villains, and investors like to tag them. Central banks and governments are always good targets, since they have few defenders and have a history of triggering market meltdowns. The problem that I find with assigning blame to others is that it then relieves me of any responsibility, even for own mistakes, and thus makes it impossible to learn from them and take corrective action.
Step 2: Assess the damage
In an age of instant analysis and expert opinion, it is easy to get a skewed view of not only what is causing the market damage, but also where that damage is greatest. In my (limited) reading of market analyses during the last four weeks, I have seen at least a half a dozen hypotheses about the stock swoon, from it being the Fed's fault (as usual) to a long overdue tech company correction to it being a response to global crises (in Italy and Saudi Arabia). In keeping with the old adage of "trust, but verify", I decided to take a look at the data to see if there are answers in it to these questions. 

1.The Fed's fault? 
As those of you who read my blog know well, I believe that the Fed has far less power than we think it does to set interest rates, but it is a convenient bogeyman. One explanation for the stock drop that has been making the rounds is that it is fear that Fed will raise rates too quickly in the future, that is causing stocks to swoon. Is that a plausible story? Yes, but if it is the reason for the market decline, you would have a difficult time explaining the movement in interest rates during October 2018:
Source: Federal Reserve (FRED)
As stocks have gone through their pains since October 1, treasury bill and bond rates have remained steady, which would make little sense if the expectation is that they will rise in the near future. After all, if investors expect rates to rise soon, those rates will start going up now and not on cue, when the Fed acts.

There is the possibility that this could be a delayed reaction to rates having gone up over the year already, with the 10-year treasury bond rate moving from 2.41% at the start of the year to 3.06% at the start of October 2018 and to a flattening yield curve (which has historically been a precursor to slower economic growth). Note though, that much of this movement in interest rates happened in the first six months of the year and you would need a reason for why stock prices would be moving four months later.

2. A Tech Meltdown?
My view, based upon what I had been hearing and reading, and before I looked at the data, was that the October 2018 stock drop was being caused by tech companies, in general, and the large tech companies, especially the FANG+Apple combination, specifically. To see if this is true, I looked at the returns on all US stocks, classified by sectors (as defined by S&P), in October, in the year to date and for 1-year and 5-year time periods.
US Sector Market Cap Change. Source: S&P Capital IQ
I know that the S&P sector classifications are imperfect, but my priors seem to be wrong. While information technology, as a group, lost 8.76% of aggregate market capitalization in October 2018, the three worst sectors in the US market were energy, industrials and materials, all of which lost much more, in percentage terms, than technology. In fact, the two sectors that did the best were consumer staples and utilities, with the latter's performance also providing evidence that it is not interest rate fears that are primarily driving this market correction. 

I have argued that, unlike two decades ago, technology companies now are now a diverse group, and many of them don't fit the "high growth, high risk" profile that people seem to still automatically give all tech companies. Using the terminology of corporate life cycles, tech companies  run the gamut from old tech to middle-aged tech to young tech, and I have looked at how tech companies in each age grouping in the graph below (age is defined, relative to year of founding):

The median percentage change, in both October 2018 and YTD 2018,  in market capitalization was greatest at the youngest tech companies. The median percentage change becomes smaller for older tech companies, in October 2018, but the effect for the four highest age classes is more mixed for the YTD numbers. That said, a much smaller median percentage change at the largest tech companies has a much biggest effect on the market, because of the market capitalization of these companies. That is the reason I look at the FANG stocks and Apple in the table below:
While the percentage change in stock prices at these companies is in line with the market drop, if Apple is included in the mix, the five companies collectively lost a staggering $276 billion in market capitalization between October 1 and October 26. accounting for almost 11.7% of the overall drop in market capitalization of US stocks. While investors in these stocks may feel merited in complaining about their losses, I would draw their attention to the third column, where I look at what these stocks have done since January 1, 2018, with the losses in October incorporated. Collectively, these five companies have added almost $521 billion in market capitalization since the start of the year, and without them, the overall market would have been down substantially.

3. A Correction in Overvalued Stocks?
For some value investors who have argued that investors were pushing up some stocks to unsustainable levels, the market correction has been vindication, a sign that the market is correcting its pricing mistakes and marking down the stocks that it had over priced the most. That may be plausible, and to see if it holds, I broke all US stocks, at the start of October, based upon PE ratios into six groupings (low to high PE and a separate one for negative earnings companies):
PE Ratio at start of October 2018, using trailing 12 month earnings
If the selective correction argument is correct, you should expect to see the highest PE ratio and negative earnings companies drop the most in value and the companies with the lowest PE ratios be less affected. While negative earnings stocks have seen the market correction, during October 2018, there is no pattern across the other PE classes. In fact, the lowest PE ratio companies had the second worst record, in terms of price performance, among the groupings. 

4. A US Problem?

One of the lessons of the last decade is that much as countries would like to disconnect from the rest of the world and chart their own pathway to economic prosperity, they are joined at the hip by globalization, with crisis in one part of the world quickly affecting economies and markets in other parts. In October 2018, we had our share of global shocks, with the standoff between Italy and the EU and Saudi Arabia's Khashoggi problem taking top billing. To see how the market correction has played out in world markets, I broke global markets down into broad regional groupings and arrived at the following:
Source: S&P Capital IQ, based upon headquarters geography
Note that these returns are all in US dollars, reflecting both the performance of the market and the currencies of each region. Asia seems to have been hit the worst this month, with China, Small Asia (South East Asia, Pakistan, Bangladesh) and Japan all seeing double digit declines in aggregate market capitalization. Latin America has had the best performance of the regional groupings, with the election surprise in Brazil driving its markets upwards during the month.  The year-to-date numbers do tell a bigger story that has been glossed over in analysis. For much of 2018, the US market & economy has diverged from the rest of the globe, posting solid numbers (prior to October) whereas the rest of the world was struggling. It is possible that we are seeing an end to that divergence, suggesting that the US markets will move more closely with the other global markets going forward.

5. Panic Attack?

One of the more striking features of the markets during October 2018 has been that the stock market retreat, while substantial, has, for the most part, been orderly. In a panic-driven stock market sell off, you usually see a surge in government bond prices (and a drop in rates), a general flight to quality (US $ and safer companies) and a rise in the price of gold. As we noted in the earlier section, the market drop does not seem to be smaller at larger and more profitable companies, and government bond rates have not dropped. In addition, while the US dollar has had a strong year so far, especially against emerging market currencies, it generally did not see a flight to it in October 2018:

The dollar strengthened mildly against almost every currency during the month, and the only currency where there was a big move was against the Brazilian Reai, where it weakened, again on political news in Brazil. Note again that the market correction may be, at least partly, a delayed reaction to the strength of the US dollar leading into October, but the timing is still difficult to explain. Finally, I looked at gold prices in October 2018, in conjunction with bitcoin, since the latter has been promoted as millennial gold:
It has been a good month for gold, with prices up 4.44%, though there is little sign of panic buying pushing up prices. It may be a little unfair to be passing judgment on Bitcoin, after one crisis, but if it is millennial gold, either millennials are unaware that there is a stock market sell off or they do not care. 

Step 3: Review the fundamentals
With the assessment of market pain behind us, we can turn to looking at the fundamentals, again looking for clues in why stocks have had such a tough month. While almost every factor affects stock prices, the effects have to show up in one of four places for fundamental value to change significantly: a shock to base year earnings or cash flows, a change in expected earnings/cash flow growth, a increase in the risk free rate or a change in the price of risk:

Since treasury bond rates have been stable through much of the month, I am going to look at one of the other three variables as the potential culprit.
  1. Base Year Earnings/Cashflows: The earnings reports that have come out for companies in diverse sectors in the last two weeks seem to reinforce the strong earnings story. While there were a few like Caterpillar and 3M that reported headwinds from a stronger dollar, both companies also conveyed the message that they were able to pass the higher costs through to the customers.
    On the cash flow front, there were no high profile cessations of buybacks or dividends, and all signs point to the market delivering and perhaps beating the earnings and cash flows that we have estimated for 2018.
  2. Earnings Growth: This is a trickier component, since it is driven as much by actual data, as it is by perception. At the start of the year, the expectation that earnings growth would be strong for this year, helped both the tax law changes of last year and a strong economy. That growth has been delivered, but it is possible that investors are now doubtful about the sustainability of that earnings growth. That has not shown up yet in forecasted growth for next year, but it bears watching.
  3. Price of Equity Risk (Equity Risk Premium): If you have been reading my blog for a while, you are probably aware of my implied equity risk premium calculation, one that backs out a price of equity risk (equity risk premium) from the level of the index, expected cash flows and a growth rate. Holding cash flows and growth rate fixed for October, I have computed the implied equity risk premium by day. 

End of DayUS 10-yr T.BondS&P 500Implied ERPSpreadsheet
If cash flows and expected growth have not changed over the month, the price of equity risk has jumped from 5.38% at the start of the month to the 5.89% on October 26, putting it at the high end of equity risk premiums in the last decade.

You could attribute the higher equity risk premiums to global crises (in Italy and Saudi Arabia) but that would be a reach since the increase in risk premiums predates both crises. If you do lower expected earnings growth going forward, perhaps reflecting a delayed response to the stronger dollar and higher rates, the equity risk premium will drop. In fact, halving the expected growth rate from 2019 on from the current estimate of 7.29% to 4.71% (the compounded average annual earnings growth rate over the last 10 years) reduces the equity risk premium to 5.28%, but even that number is a healthy one, relative to historic norms. The bottom line is that, at least by my calculations, I am estimating an equity risk premium that seems fair, given macro and micro fundamentals and my risk preferences.

Step 4: Investment Action
One of the biggest perils of being reactive in a  crisis is that it can knock you off your investment game and cause you to abandon your core philosophy. I don't believe that there is one investment philosophy that is right for every one, but I do believe that there is one that is right for you, and shifting away from it is a recipe for bad results. I am a “value” investor, though my definition of value is different from old-time value investing in two ways:
  1. Under valued stocks can be found across sectors and the life cycle: I believe that we should try to assess fair value, not a conservative estimate of value, and that the value should include expected value added from future growth. To the critique that this is speculative, my answer is that everything other than cash-in-hand requires making assumptions about the future, and I am willing to go the distance. That is why, at different points in time, you have seen Twitter and Facebook in my portfolio in the past and may well see Netflix and Tesla in the future (just not now).
  2. Intrinsic value can change over time: I believe that intrinsic value is a dynamic number that changes over time, not only because new information may come out about a company. but also because the price of equity risk can change over time. That said, intrinsic values generally change less than market prices do, as mood and momentum shift. This has been a month of significant price drops in many companies, but assuming that they are therefore more likely to be under valued is a mistake, since the intrinsic values of these companies have also changed, because the ERP that I will be using to value the stocks on October 26, 2018, will be 5.89%, much higher than the 5.38% at the start of the month.
Given my philosophy and a reading of the data, here is what I plan to do.

  1. No change in asset allocation:  I am not changing my asset allocation mix in significant ways, since I don't see a fundamental reason to do so. 
  2. Revisit existing holdings: I normally revalue every company in my portfolio at least once a year, but after a month like this one, I will have to accelerate the process. Put simply, I have to make sure that at the current price for equity risk, and given expected cash flows, that my buys still remain buys and the sells remain sells.
  3. Bonus from short sales: I do have a portion of my portfolio that benefits from a sell off, primarily in short sales and those have provided partial offsets to my losses. I did sell short on Amazon and Apple at the start of the month, and while I would like to claim prescience, it was pure luck on timing, and the market downdraft during the month has helped me. 
  4. Check out the biggest market losers: I plan to take a closer look at the stocks that have been pummeled the most during the month, including 3M and Caterpillar, to see if they are cheap at October 26 prices, and using an October 26 ERP in my valuation. 
Please note that this is not meant to be investment advice and your path back to investment serenity may be very different from mine! 

YouTube Video