P/B ratio = market capitalization / book value of equity

Market capitalization = shares outstanding * market price per share

Book value of equity = book value of assets - book value of liabilities

So therefore, P/B = market cap / (BV of assets - BV of liabilities)

• Assets are overstated on the balance sheet. In this case, we should avoid the company, because it may be destroying shareholder value. Ford (NYSE: F) is a good example. According to MSN Money Central, Ford's P/B was 0.72 in 1997, and its book value per share was $25.54. Today, Ford's P/B is 1.07, and its book value per share is down to $7.33. During that time, the share price has fallen from nearly $50 to $8. Clearly, Ford had other problems, but the low P/B certainly did not indicate value. At Inside Value, we generally look for companies that have been increasing book value per share over a number of years, because -- as Ford's plight shows -- the share price often follows the book value per share.
  
  
• The company will generally have a poor return on equity (ROE) and poor return on assets (ROA). If earnings are negative, there will be a negative ROE and ROA. Of course, if a solid management team is turning around the company's fortunes, then we may have an interesting value proposition. When I first recommended MCI for Inside Value subscribers, it certainly fit that mold.
  
  
• The industry at large has a low P/B. Certain industries have low P/B ratios, generally because they are cyclical or because the companies generate relatively low ROE. Hurricane Katrina reminded us that insurance companies typically have low P/B ratios because of the cyclicality of that industry. "Hard insurance markets" (those with higher premiums) form after a major disaster. This attracts new capital in the short term, when investment returns can be very good. After a while, however, competition increases and the market softens. It's important to find insurance companies that maintain discipline in a soft market. Otherwise, they'll take on risks that are not adequately covered in the premiums. Eventually, claims will roll in, along with underwriting losses, and the book value will be reduced.

• Companies that have very long-lived assets (like real estate) still on the balance sheet at original cost (i.e., the book value) will have understated assets and, therefore, an understated book value (remember, book value of equity = assets - liabilities). All other things being equal, the effect will be an increase in the P/B ratio, as the reported "B" is lower than the real value of the equity. In this case, you might miss an undervalued company by simply looking for low P/B ratios.
  
  
• Service companies and those that rely on intellectual property (IP) are not capital-intensive, and they do not have significant assets recorded on the balance sheet. A good example is Microsoft (Nasdaq: MSFT). At the end of its 2006 financial year, Microsoft had $69.6 billion in assets, with a full $31 billion in cash and equivalents, $18 billion in other current assets, and only $20.5 billion in longer-term assets. Nearly all Microsoft's IP was developed in-house and is not capitalized on the balance sheet -- and it can't be, because R&D must be expensed on the income statement under U.S. generally accepted accounting principle rules.
  
  
• Recent acquisitions will generally increase the book value and lower the P/B, because the new assets go on the balance sheet at the full price paid. In the unlikely event that Microsoft acquired Oracle (Nasdaq: ORCL) and all of its IP for $81 billion in shares, the full $81 billion would be recorded on the balance sheet as an asset. The amount in excess of Oracle's $15 billion book value ($81 - $15 = $66) would show as goodwill.
  
  
• A serial acquirer of other companies will almost always have a high book value, which may artificially lower P/B. However, a huge part of the book value will be in goodwill or intangibles. In this case, it's prudent to subtract goodwill from book value, resulting in a "tangible book value." We can then calculate the more meaningful "price-to-tangible-BV ratio."
  
  
• Recent write-offs will reduce the book value of equity. Companies typically say that this is a non-cash charge, yet in a real sense, it reduces the value of shareholder equity. It may not show up in the P/B ratio, because as companies reduce the "B" via write-offs, investors reduce the "P" via a lower stock price.
  
  
• A company with a history of large share buybacks (beyond covering dilution from employee stock options) will have a lower book value. All of the shares bought back go into what is called treasury stock at the full buyback price, and these are subtracted from book value. The original shares are recorded at par value, which is usually as low as $0.10 to $1.00 per share. A good example is Inside Value newsletter pick Anheuser-Busch (NYSE: BUD), which bought back 140 million (15%) of its shares in the past four and a half years. Book value is just $3.8 billion, yet the company has almost $15.7 billion in treasury shares. Without any buybacks, the book value would be around $18 billion. A really good indicator here is that both the P/B and the ROE will be extraordinarily high. For Anheuser-Busch, P/B is 9.6 and ROE is 56, which are extraordinarily high and of little use in assessing the value of the company. To overcome this, we can use a BV adjusted for share buybacks. This would reduce the P/B to 2.0 and the ROE to around 13.