Monday, November 25, 2013

A Stock Market Bubble?

There has been a lot of talk of a stock bubble in the financial media. One of the cover story for the weekly Barron's magazine was "Bubble Trouble?" (subscription required to read). Also, there was another good WSJ story that used Robert Shiller's favoured Cyclical Adjusted P/E (CAPE) ratio to argue stock valuations are getting frothy. The CAPE, which divides the S&P500 price by a 10-year average real earnings number,  is near 25.1 compared to the 130 year average of 16.  With the S&P500 rallying over 27% this year, NASDAQ over 30% and Dow Jones Industrial 22%, it definitely looks like a stock market bubble is building. However, I should caution investors not to call a top based on the YTD performance or the fact that the S&P500 is up over 170% since its low in March, 2009. 

The Frothy Side:

Let's not forget the cause of the market rally this year. The key reason is that low interest rates, caused by ultra loose monetary policies like QE, pushed many investors up the risk curve. Some fixed income yields, namely short term treasuries, were negative after adjusting for inflation (real yield) which forced investors to seek higher yielding assets in either corporate bonds, high yields or even conservative blue chip stocks. The key measure that determined whether investors prefer bonds or stocks is spread between the earnings yield (E/P or inverse of the P/E ratio) and the bond yield. At the start of the year, the 10-year US  treasury yield was near 1.8% and the S&P500 earnings yield was 7.5% or almost 4 times higher! It's a no-brainer why stocks rallied. The fact pension funds and mutual funds had low equity exposure also fueled the equity rally as those "big players" re-entered the market. 

Despite a small correction in the summer due to the taper tantrum, stocks kept on rallying. The current earnings yield of the S&P500 is about 6% while the 10-year US treasury is yielding near 3%, which reduced the attractiveness of stocks by 50%. The S&P500 is trading at 16.5X trailing earnings and 15.1X forward earnings. Those measures are slightly higher than historical averages but well short of the P/E north of 20X in the dotcom boom and 17-18X at the height of the 2007 bull market. This means that stocks can still rally further. Mutual fund asset levels are still low compare to 2007 and many pension funds are still under-weighing equities. Therefore, don't be surprised if stocks move higher. 

However, what worries me is that the equity rally is mostly due to P/E multiple expansion. Recall changes in price is the product of changes in earnings multiplied by changes in the P/E ratio. Given the aggregate S&P500 earnings is foretasted to increase only 6% in 2013, the 27% rally in the index was caused mostly by a 20% expansion in the P/E ratio. P/E doesn't expand forever and the fact earnings growth is unlikely to pick up is a cause for concern. 

The Problem of Calling a Top: 

In December 1996, former Fed Chair Alan Greenspan made his famous "irrational exuberance" speech questioning the valuation of equities at that time. Some investors agreed with Greenspan and started shorting equities. The fact the stock market advanced 31% in 1997, 26% in 1998, and 20% in 1999 proved many of those skeptics wrong. The point I want to make here is that calling a top is extremely hard. Many should not even waste the time or effort to because it is hard to get both the timing and thesis correct. 

So, What to Do? 

Bonds are still yielding close to nothing (after adjusting for inflation) while stocks may be in bubble territory. My opinion is that stocks should be preferred to bonds.


  • Earnings yield of 6% is still 2 times higher than long term treasury yield and higher than most fixed income yields (including junk bonds that are yielding near 5%) 
  • Stocks provide investors with great long term performance despite short term troubles. Let's not forget that in the 20th century, the Dow Jones returned 5.1% compounded annually. For those who are not aware, the market experienced several large drops greater than 50% with the Dow losing 90% from 1929-1933. That period also included 2 World Wars, the Great Depression, a Cold War, a decade of super high inflation etc. 
For the average Joe investor, a bubbly market should not deter them from a conservative dollar cost averaging strategy where one sets aside a fixed amount to put into an index fund every month or quarter. For more active investors, there are still bargain stocks available especially in the mining space. Trying to call a market top is foolish because no one has a perfect crystal ball. Investors should adhere to strict investment rules such as buying stocks at reasonable valuations (i.e. significantly below intrinsic value) in order to provide them with a margin of safety. Timing the market appears to be simple and easiest strategy to make money but its application is hard in practice. 

In short, although I worry about a stock market bubble, I'm still long stocks in my portfolio because they still provide the best risk-adjusted potential return. I use the margin of safety principle advocated by Ben Graham to help protect my downside by buying stocks with low valuations, healthy balance sheets and high future earnings power. I tried calling market tops and bottoms in the past but learned it's better to admit you can't. Many times, you get the thesis right, but the wrong timing can kill you. As Keynes famously said: "The market can stay irrational longer than you can solvent". 

Wednesday, November 6, 2013

Are Share Buybacks Really Meaningful?

After my article on stock splits, I believe I should discuss my thoughts on share buybacks, another important topic in corporate finance. I will write about other topics such as dividend payout policy, M&A, executive compensation etc in the future.

What are Share Buybacks?

Share buybacks, also know as share repurchases, are methods of returning cash back to shareholders. The net effect of repurchasing shares will result in less shares outstanding, which increase the ownership of shareholders who decide to maintain their stake. Companies usually purchase their shares on the open-market, although a tender offer can also be used. In Canada, an open market share buyback program is called Normal Course Issuer Bid (NCIB).

Rationale Behind Share Repurchases: Returning Free Cash Flow

Basically there are 3 main uses of free cash flow (often defined as operating cash flow minus capital expenditures): (1) pay down debt (2) make an acquisition (3) return it back to shareholders via dividends or share buybacks. Most executives hate option 1 because they want to maintain an "optimal structure" of debt and equity. In my own opinion,  it may be prudent to reduce debt levels when excessive free cash flow exists since high cash levels usually build near the height of an economic boom, which implies a recession may be on the horizon. Reducing debt levels in good times prepares the company for bad times such as a recession or economic slowdown. Also, it leaves excessive capacity to take on more debt in recessionary times when interest rates are much lower. Option 2 was popular but CEOs are now more cautions when it comes to acquisitions especially after the disastrous outcomes of serial acquirers like Tyco, Worldcom, or Citigroup. Accounting treatment is also less favourable with FASB and IASB abolishing the pooling of interests method. Option 3 is the most commonly used method of returning free cash flow and most executives prefer stock repurchases just because it is "tax-efficient" over dividends. Although the argument is valid, purchasing shares above intrinsic value is value destroying (see example below) despite being a little tax efficient  If the shares do trade above intrinsic value and the company has excessive cash, it should consider boosting its dividend instead.


Stock X is trading at $150, has 100 million shares outstanding, and earns $500 million a year. This implies a market capitalization of $15 billion and EPS of $5. Let's assume that the true intrinsic value (hidden from the public but can be roughly estimated by the intelligent investor) is only $10 billion or $100 per share, which implies the stock is 50% overvalued. 

If the company decides to repurchase 5% of its outstanding shares (5 million shares) at $150 and still makes $500 million a year, EPS would increase from $5 to $5.26, a magical 5.2% increase. However, if the share price corrects to the actual $10 billion (in the long run, the market is a weighing machine and will be efficient) , the high purchases made at $150 are clearly value destroying. If the company did not conduct the share repurchase, intrinsic value would still be $100 per share (10 billion value divided by 100 million shares). If they did the share repurchase, they would destroy $250 million of value by buying shares for $750 million when intrinsic value of those shares is only $500 million. Thus, intrinsic value of the company after the repurchase is 9.25 billion or $97.37 per share. ([$10 billion - 0.75 billion]/ 95 million shares). Well congratulations shareholders of company X, it just destroyed $2.63 of value on a per share basis. The company may had good intentions, such as returning cash to shareholders, but buying back shares above intrinsic value is dumb. Many would argue against me by stating there is no exact method of determining the intrinsic value. Nonetheless, I believe experienced managers who is familiar with industry trends and the company's financials should have a rough idea on the company's intrinsic value. Academic studies have shown that following the insiders (after they disclose a purchase or sale) can earn abnormal returns. 

If the shares was 300% overvalued, this transaction would destroy over $11.50 of value on a per share basis. The prior examples are typical in the real world. Why? This is because corporations usually have large cash hoards in boom times or just before a recession hits, which also corresponds to when their share prices are often overvalued in the market. Many corporate executives are often pressured to return cash to shareholders and believe buybacks are the most efficient method. 

Rationale Behind Share Repurchases: Growing EPS or Offset Share Dilution 

Short term earnings accretion get the attention of analysts and the financial press, but prudent investors should look for underlying value creation, not earnings accretion. As the example above shows, there is earnings accretion with EPS growing 5.2%. However, the actual intrinsic value of the business, on a per share basis, is lower as a result of repurchasing shares above intrinsic value. 

Another common argument is that companies often issue stock options and exercising those options could dilute existing common shareholders. The problem of offsetting that dilution with share repurchases is that buying shares above intrinsic value would destroy more value than merely letting its shares outstanding increase. Let's go back to the prior example. If there is 5 million of options, the shares outstanding would increase to 105 million and EPS decreases to $4.76. Most executives doesn't like that lower EPS so they decide to repurchase shares. If the company doesn't do anything, intrinsic value reduces to $95.24 ($10 billion/ 105 million shares) or $4.76 lower. If the company decides to repurchases 5 million shares at $150 (greater than $100 intrinsic value) to offset the 5 million share dilution, intrinsic value is reduced to $92.50  ([10 billion - 750 million] / 100 million shares)  or $7.50 lower. Clearly $95.24 > $92.50, so repurchasing shares when the shares are overvalued is a dumb idea. Buying loonies for tonnies is plain silly (For non-Canadian readers, loonies = $1 while tonnies = $2) 

Share Buyback Lesson from the Oracle of Omaha:

We pay so much tuition to learn finance in University but the best source, Buffett's Annual Letters to shareholders, is available on the Berkshire Hathway's site for free. Regarding share buybacks, Buffett's clearly explained his position in his 2011 Annual Letter (see the section on "Share Repurchases" on page 6-7). Here is a quote taken from the letter:

"Charlie and I favor repurchases when two conditions are met: first, a company has ample funds to take care of the operational and liquidity needs of its business; second, its stock is selling at a material discount to the company’s intrinsic business value, conservatively calculated"

Buffett explains how CEOs, even the honest and hardworking ones, fail the second test miserably. A CEO may think his company's share price is cheap no matter what price it is selling at. Even the honest executive is led to believe that returning excessive free cash flow is necessary. Buffett argues that this is not true. There is no point of a share repurchase UNLESS the you can repurchase the shares below intrinsic value, hopefully conservatively calculated as Buffett suggests. Notice Buffett never mentions earnings accretion or returning excessive free cash flow as criteria when deciding a share repurchase.

Here is another interesting thought from Buffett: "When Berkshire buys stock in a company that is repurchasing shares, we hope for two events: First, we have the normal hope that earnings of the business will increase at a good clip for a long time to come; and second, we also hope that the stock UNDERPERFORMS in the market for a long time as well ." 

Wait, what? Did Buffett just say he hopes the share price would underperform the market for a long time? This is not a typo. The logic is a simple one, but often not well understood. If the share price underperforms, the company can purchase more shares and the remaining shareholders would own more of the company (and receive more share of the company's long term earnings as well). If IBM's share price remains at $200 for the next 5 years, Berkshire's 5.5% stake in the company would indirectly increase to 7% through the ~$50 billion share buyback program. But if IBM averaged $300 in the same period, then Berkshire's stake would only increase to 6.5%. The 0.5% may seem small but it makes a big difference in the long run. If net income is $20 billion for IBM per year, then Berkshire will get an additional $100 million share of that income. IBM's stock price is in fact languishing right now as expected and sell-side analysts are cutting their price targets like crazy. While analysts criticize IBM for poor share performance, Buffett must love the lower share price right now. It's hard to think long term but let's not forget the moral behind the turquoise vs. hare story. 

If the company is pursuing a large buyback program, shareholders should cheer when the stock price languishes. As Buffett says: "The logic is simple: If you are going to be a net buyer of stocks in the future, either directly with your own money or indirectly (through your ownership of a company that is repurchasing shares), you are hurt when stocks rise. You benefit when stocks swoon. Emotions, however, too often complicate the matter: Most people, including those who will be net buyers in the future, take comfort in seeing stock prices advance. These shareholders resemble a commuter who rejoices after the price of gas increases, simply because his tank contains a day’s supply." Very well said Mr.Buffett, I thank you for sharing this point many people miss. 


To answer the question in the title, share buybacks is definitely meaningful method of returning capital to shareholder only if the company can repurchase shares under intrinsic value, conservatively calculated. Investors should be cautious when CEOs announce buybacks because their intent is to offset share dilution or increase EPS growth. If CEOs announce buybacks because they want to return free cash flow, carefully assess the company's intrinsic value. If the company's shares are trading below intrinsic value, you should applaud the move and give a big thumbs up. If it is not (especially if it is significantly above your estimate of intrinsic value), you should  re-consider your investment thesis in the company. No matter how great a business may be, investors should not invest in the company if the CEO makes poor capital allocation decisions because those decisions can lead to a lot of value destruction. (An exception would a net-net special situation but I'll again save this topic for next time)

Tuesday, November 5, 2013

Some Real Thoughts On Stock Splits

Another post inspired by a WSJ CFO Journal article. (The article is found here. My prior article inspired by WSJ was on non-GAAP measures, which is here). The WSJ article discuss how corporations should consider revisiting stock splits. The author claims that academic studies have shown that the companies who split their stock will outperform the market. It is clearly the author supports the idea that a stock split will create value for its shareholders. However, my own opinion on share splits is quite the opposite: It's distracting tactic that management use to boost the stock price in the short term and does not create any value at all. Why would management want to boost short term share performance? The answer is simple: stock options grants (I'll save the topic of executive compensation for another day).

For those who are not familiar with the concept, a stock-split is a transaction that increases the company's shares outstanding by a specific multiple (usually 2 times or 3 times) and reduces the trading price of the stock by the same multiple. Here is an example to illustrate a stock-split: stock X is trading at $100 and has 1 million shares outstanding.  If it announces a 2:1 stock-split, the number of shares outstanding  doubles to 2 million and the stock's trading price would halved to $50. The company's market capitalization remains at $100 million (market capitalization is the market value of equity calculated by multiplying shares outstanding by the share price). No matter how you slice and dice it - i.e. $50 x 2 million, $33.33 x 3 million, $20 x 5 million -, the market capitalization is still $100 million. The famous Yogi Bear once said to the waiter after he ordered a pie: "Please cut it in 4 pieces instead of 8 because I can't eat 8 pieces". It doesn't matter how the pie is cut, Yogi Bear will still eat one whole pie. The fact he thinks he is eating less because he ate less pieces is plain silly.

This quote from the article made me laugh: "Kenneth Fisher, Noble Energy’s CFO, says the company’s research indicates companies that split outperform their peers by 2% to 3% in the near term.". The emphasis in that quote should be on the last two words in his sentence: NEAR TERM. 

If stock splits do create value, then companies with the high trading prices should be less attractive to investors. History suggest otherwise. Lets take the three highest priced stocks in the S&P500 which had no stock split in the last 10-years and look at their performance. The three companies with the highest trading prices are Berkshire Hathaway (at $171,700), (at $1080), and Google (at $1020). In the past 10 years (November 2003-present), the S&P500 only returned 60%. What about those stocks with the highest trading prices? Berkshire returned 120%, Priceline returned 5000% and Google returned 800% in the same period. Granted, although Priceline and Google were trading at prices below $100 in 2003, their executives never considered a stock split when their shares soared above $100 like most other S&P500 chiefs. Thus, the evidence suggests that high trading prices do not hurt share performance at all. So why do most CEOs want a stock split? 

Most S&P500 CEOs usually do a  2:1 or 3:1 split when the price of the company's shares soar above $100 because they believe they can create value by lowering the share's trading price. By lowering the trading price, CEOs think they can psychologically impact investors' perception of value just because the price per share is lower or liquidity in the stock has increased. While I agree most companies may experience short-term outperformance after a share split, the long term costs of the share split is negative for shareholders. Stock splits increase administrative and market listing costs because the shares traded more often (more turnover). Also, stock-splits generally attract more short-term oriented investors and increase management's incentive to try to meet quarterly earnings forecasts as the shareholder base becomes increasingly short-term oriented. There is nothing more value destroying in the long run than management trying to play the earnings beat game. Management may try various schemes to meet short term earnings expectations such as cutting expenses (instead of growing revenue), focusing on short term payback projects (instead of projects with the highest NPV), or doing ridiculously high priced acquisitions (which maximize short term earnings accretion but hurt long term value because they often overpay or issue undervalued stock to finance such acquisitions).

Therefore, stock-splits add zero value and it's just a short term tactic aimed at boosting investor's short term enthusiasm. If you hear a CEO talk about the benefits of a stock split, ask him/her whether 1 x 10 =  2  x 5  If he/she says yes, then you should reply why consider a stock split in the first place when the left hand side of the prior equation is equal to the right hand side. Actually the equation should be 2 x 5 - costs < 1 x 10  because stock splits will increase costs. Long term shareholder value can only be created if management focuses on building the underlying business (invest in positive NPV projects, return excessive free cash flow, and allocate capital efficiently) and generate a high return on capital. As Buffett famously said: "If the business does well, the stock will follow". If a CEO thinks his/her company should consider a stock split because the trading price is too high, he/she is spending way too much time looking at the company's stock price and not enough time building the underlying business.