Does book value even matter?

Book value is praised as the one true metric that matters to investing, and derided as an accounting fiction.  Book value is fascinating, on one hand it's an accounting creation, yet on the other hand it's a very rough estimate for the tangible value of the company.

At the most basic level book value is the sum of a company's assets minus all liabilities.  What's left over is termed equity, or a company's book value.  A company's book value isn't anything specific, it's just the remainder from subtracting two values, assets and liabilities.

Value investors place a lot of emphasis on book value, for most book value represents a tangible high water mark for an investment.  What's that you say, you are a value investor but don't care about book value and only growth?  My friend you are a value GARP investor, you're looking for cheap growth, you're in good company, many "value" investors are really trying to buy growth cheap.  Read on to discover how those of us with plaid sport coats and elbow pads view book value.

Book value is at best an estimate, an estimate of what equity holders tangibly own in a company.  In theory if a company were to pay off all liabilities and distribute the remaining assets investors would receive book value.  In reality this never happens, and when it does it's a messy affair, at times what's distributed is far in excess of book value, and other times far below.

Two companies could report $100m in equity, at the first company the value might be realistic, and at the second nothing more than a faint hope.  A company's book value is determined by the composition of its assets and liabilities.  I have a friend who runs a factoring business, he sent me an email recently discussing how many client receivables are hopelessly overstated and will most likely never be collected.  He brought up a good point, a point that Graham discussed in Security Analysis.  In all companies liabilities are real, whomever a liability is owed will ask for it to be settled at some point, but assets are never sure, cash is sure, but the value of receivables, inventory and property is often a best guess.  While inventory might be overstated, a debt on the balance sheet is never wrong.  I have yet to read a financial statement with the note "We borrowed $10m, but the lender decided they only wanted $9.5m back."

A company's book value is not precise.  The balance sheet is merely a snapshot in time, but thankfully most items captured on the balance sheet are slow to change.  A company's earnings might fluctuate quarter to quarter or year to year, sometimes slowly, and sometimes dramatically.  Book value rarely fluctuates quickly, the value it captures is slow moving and rarely changing.  Cash is valued the same five years ago as it is today and as it will be in twenty years.  A manufacturing company's building is probably worth the same today as it was ten years ago.  Until a company decides to sell their assets and determine what the market will pay book value is the best estimate we have.

Some items captured in book value are clearly understated.  Companies are required to depreciate property, plant and equipment and scheduled intervals.  This depreciation is supposed to capture the fact that property and machinery decline in value as they are used.  Depreciation is a rough metric, at times it's close, but with certain industries it clearly misses the mark.

My brother works for a manufacturing company whose machinery is from the 1930s and 1940s.  The company's needs are simple and the machinery is essentially bulletproof.  The machinery has long been depreciated, yet continues to have years of useful service left.  I have worked in a number of manufacturing facilities where I've seen similar things, although not to the same level.  Machinery manufactured in the 1960s and 1970s is commonplace in many American factories.

A company I own was an example of this recently, Bowl America announced that they had sold an underperforming facility for more than $2m.  They subsequently stated that the facility was held on the books for $230,000, quite a difference!

What I find very interesting is that while equity investors give book value a passing thought bond investors and bankers pay the utmost attention to book value.  A creditor has a lien on the company's income stream, but if for some reason a company can't pay they need to determine if they'll be able to collect on the money they lent.  In a bankruptcy or distressed situation a creditor might force a company to sell off some assets and return the cash to creditors.  An asset sale moves theoretical values to the tangible column.  A sale shows the world what the company thought a building was worth, and what others thought it was worth.

I like to see companies where book value is growing over the years.  When book value grows equity investors own more of something each year.  A high growth company could earn a lot of money, but need to re-invest all of it for continued growth.  If the company fails to invest growth slows, and often investors flee.  An investor in such a company is betting that growth will continue, but they don't own anything more than they did a year before.

When a company grows book value they show that they earn enough to cover costs, and have something left over.  Companies with growing book values selling at discounts are often incredible investments.

The one unstated item in this post is the quality of assets, I briefly mentioned it above, but it deserves further discussion.  Let me start with two examples.  Value investors are often afraid of banks, they seem complex and mysterious.  A claim I've heard is that bank investors don't actually have a claim on a bank's equity because it's held for regulatory purposes.

I would argue that banks are the ultimate liquid and commodity investment.  There are over 7000 banks in the US, banks routinely buy and sell loan portfolios to each other, as well as branches.  A bank with a good loan book and good branches could sell their loans and branches at par fairly quickly.  They could then take those proceeds and distribute them to depositors leaving equity holders owning the company's book value.  Due to regulation book value consists of cash and securities, highly liquid salable items.  Whereas on the surface a bank might be confusing, a bank's equity capital is very simple, and often very liquid.

The second example is a company floating around the blogosphere, Premier Exhibitions.  The company owns the Titanic artifacts, which a judge determined were worth $190m.  The company sells at a wide discount to the artifact values.  When looking at the book value of Premier it consists of the Titanic artifacts and not much else.  Clearly equity holders are betting that Premier will be able to sell their artifacts at close to their valuation.  They might be, but I am positive the market for $190m worth of Titanic artifacts is much smaller than the market for commodity items such as office buildings, or loan portfolios, or bank branches.

For book value investors the question to ask is what does book value consist of?  If a company is selling at a discount to book value is it because book value is a hard to value or a hard to sell asset?  Is it because the market doesn't understand how to value the assets?

Talk to Nate about book value

Disclosure: Don't own anything mentioned specifically, generically I own a number of bank stocks.


  1. Nate, in this post you remind me of Marty Whitman! He has written a lot about this - that the apparently illiquid parts of the balance sheet, such as high quality office buildings, are often the ones easiest to turn into cash.

    His books are worth a read, if you haven't already.


    1. I have read The Aggressive Conservative investor, it was good, but I have trouble with Whitman's writing style. It's really his obsessive use of abbreviations that gets in the way.

    2. His newer two books are written with a co-author and so I think some issues are toned down.

  2. One problem with book value is that it doesn't capture the value of the business as a running concern, that is it's not answering the question "how much would it cost to start a similar business from scratch?" which is pretty critical to ultimate value, and extremely variable.

    1. It actually does, a company that is generating value compounds book value. An exception to this is a company that pays out all earnings as dividends.

      A company that earns a lot, or generates a lot of cash, but requires the same amount of cash to continue operations isn't generating value.

  3. What I'm understanding from this post is that the market value of the assets matters and that book value doesn't matter. I agree with that. That's why growth investments at the cost of capital increase book value but don't increase value -- that's why " don't own anything more than they did a year before".

    But there's another problem with book value: it implicitly assumes (and retail equity investors are prone to believe) that, in the final analysis, book value belongs to equity. Sometimes it does, and sometimes it doesn't. The quickest way to shake one's confidence in book value is to take a tour of bankruptcy filings and see how often book value provides protection from total loss.

    Community banks, of course, are a different kettle of fish, and the different species of insurance companies are different still.

    1. I agree, but disagree, part of my point that was probably unclear is that book value is the best estimate we have of the market value in most cases. In some cases like companies heavy in real estate a market value is easily determined, but for most companies we don't have that.

      Bankruptcy is where all things are laid bare, and in most cases (heck almost all) what is owned is indeed worth less than book value. One of the reasons I prefer companies without debt is because the risk of bankruptcy is significantly reduced, a company without debt can survive much longer in a downturn, especially one with reserves. I don't want a court to determine the value of assets for me.

  4. To me, book value matters most when it involves "regulated" companies whereby the returns they are statutorily-able to reap is governed by book value. So insurance companies, banks, utilities.

    Now, I know, I know, an InsuranceCo's "Statutory Surplus", a Bank's "Tier I Capital", and a Utility's "Regulatory Equity" are not equal to GAAP Book Value, but my point is that the "Book" is what determines how much levered free cash flow can ultimately be handed over to the owners each year. No other industries are like that.

    I'd normally say I don't even bother looking at GAAP book on companies outside the above 3 industries, but it drove me crazy when it's used to justify or NOT justify something that is plainly not true.

    Let me give you an example -- a company whose stock I recently owned underwent an MBO at <0.9x P/B. The proxy statement's fairness opinion stated that book value was a meaningless, antiquated relic of accounting fiction. Like hell it was! The Company had *just* dumped $50m into buying PP&E just one year ago, and was now claiming that this machinery was worth 60 cents on the dollar (mind you, this machinery could easily be operable for 10+ years).

  5. Another way to look at it is that the assets are worth what they earn over the long term (and why the earn it) and nothing more or less than that.

    Of course, some companies earn less than their potential because of rubbish managerial practices, in which case one has to imagine what some other management team could do with those same assets. (That's why GP/Assets is becoming an increasingly popular metric and, if one traces the activities of someone like Icahn one can see that exploiting that wrinkle is one of his specialties).

    So, if it's the case that true asset value is determines what those assets can earn, then owner earnings yield is a priori a better metric than P/B. Hence "the intrinsic value of a business is the discounted sum of all the cash flows that can be taken out of the business during its remaining life".

    I take your point about avoiding companies with debt. On the other hand, I have seen companies with no debt taken down by the current liabilities account and the equity rendered worthless (Trident Microsystems, for example, was a net-net with no financial debt before its bk filing; its equity was virtually worthless at the end of the process).

    As always, Nate, thanks for your blog & for the opportunity to discuss these value investing ideas.

    1. I have a lot of caveats with earnings, I guess here are a few thoughts.

      -The first is for a simple business where all assets are deployed and utilized with an average management then yes, earnings are a better metric of value. This probably captures most companies in the market.
      -For a business with bad management or under utilized assets then book value is probably better metric of value, market value is obviously the best.
      -A lot of assets might not generate earnings, but that doesn't mean they don't have value, cash clearly has a lot of value, but in almost all cases it just sits there. The same is true for an underutilized building, or empty land in a prime location. The converse is true as well, not all assets have value, some like inventory or receivables might be overstated, this is a huge blind spot for many investors.

      I've seen companies go into bk without debt as well, usually the result of some legal liability. I guess the essence is all of these are technically off balance sheet obligations, if disclosed these are the things I try to avoid the most.

      Thanks for the comments, they're very thoughtful!

    2. Current liabilities are debt, so I don't understand your Trident Microsystems example. I assume you mean long-term debt, but a prudent investor will account for all debt, not simply long-term debt.

    3. Intrinsic value is the discounted weighted probability of all future cash flows regardless of whether they are from operations or sale of assets.

  6. One thing I find many people overlook is the fact that although a companies book value may indicate it is worth something, it is of less value to a shareholder if that is tied up in operating the business.

    For example banks, which are required to hold ever increasing capital ratios. They frequently trade below book value but that's because their book value although it may be real and properly valued, isn't really worth as much to shareholders as they cant access it and its not really providing decent returns.

    Don't get me wrong, I love finding a company with plenty of assets on the balance sheet, but only if they are non-operating and not too hard to value.

    1. What do you mean by access it? Technically shareholders can't access the equity of any company. With your line of thinking the only value a company has is the dividends they pay to shareholders correct?

    2. I think the point is that a company's book value stands on its own. If the assets are tied up in operations they have value, this value might be determined by the earnings they throw off, but there's clearly value.

      If the assets don't throw off earnings then they might have an independent value.

      Would you say that a manufacturing company's property and plant have no value because they're tied up in the business?

      In a lot of ways a bank's equity is more "true" than any other company. A regulator ensures that a certain amount of cash and securities exist and are held in reserve. In an unregulated industry we don't actually know the value of many assets, we just have a guess.

    3. I think I understand what Sidekick is saying. Let's take two fictional banks.

      One bank is trading at 90% of book but has generated only 3% book value growth annually over the past 5 years. A second bank is trading at 110% of book but has generated 12% book value growth annually over the past 5 years. I would argue that owning the second bank is a superior investment.

      Now if the first bank (trading at 90% of book) has some non-operated assets they could sell quickly to generate better returns, then the better buy might be the first bank. Hence this is where activists come in.

    4. Josh,

      I understand your point, I'm not sure that's what sidekick was aiming for or not. The problem with the example is that bank 1 might actually be worth more than bank 2. If bank 1's assets are underutilized, then they might have more value to a buyer. Maybe bank 2 is in an area where loans have been easy to come buy, they could grow book value easily with a very low quality portfolio.

      I guess nothing is every easy, it always pays to dig below the surface.


  7. (I'm having trouble replying in the proper place, so this is in response to 10:38)

    Yes, cash is often worth what it says on the balance sheet.

    FFWIW, I use [Normalized forward operating profit]/[Total Enterprise Value] as my preferred metric for estimating the relationship between price and value. Which means that I bypass everything on the asset side of the balance sheet except for cash. The liabilities side of the balance sheet is important and add the off-balance sheet liabilities.

    That a[n industrial] business is yielding 25% on a normalized profit/TEV basis imparts very much more information than the fact that it is trading at P/B of less than, say, 0.3.

    Buffett was a much better cigar butt investor than Graham. Take away GEICO and G-N's annual returns were something like 9% a year. I suspect that Buffett's approach in his early phase, like yours, was a great deal more sophisticated than just buying up assets at way less than book in that he asked not only "will it live?" but also "will it prosper?". (I believe Geoff Gannon has written about this before).

    The one thing that I have learned in investing is that one can always find a good business selling at the price of a bad one -- i.e. find a bad business selling at 25% yield and you'll always find a good business at that same price or below.

    Anyway, I've probably taken up too much space in your comments section so I'll leave it at that!

    1. This is an old post, and probably nobody is going to read this comment, but I happened to feel the need to respond to one thing in Red's comment above.

      First off, it's an interesting discussion regarding book value and earning power. Red and Nate, I've enjoyed these two posts and the subsequent comments.

      But one thing I had to correct: Graham's returns were far better than 9% ex-GEICO. Graham didn't buy GEICO until the late 40's, and soon after he purchased it had to distribute it to his partners due to regulations against an investment company owning an insurance company. True, GEICO helped make his partners rich, but Graham's results were still stellar regardless.

      I have Graham's partner letters from the 1940's and 1950's and it's possible to calculate the gross returns before fees. There was about a 12 year period from the inception of this fund in 1936 to the time he bought GEICO. In one letter Graham summarized the previous 10 year period, and during that 10 year period (before GEICO) he achieved 17% annual returns vs 10% for the Dow. His returns got better after GEICO. But even if you exclude GEICO from the equation, he still achieved around 20% annual returns before fees on a portfolio that included 50-100 positions at times.

    2. That is the beauty of using an RSS reader (I use free version of netvibes). You can track all new posts as well as new comments. Do you have the Graham partner letters in email-able format? Or can you scan them in? I am sure many people would love to read them.

    3. Josh,

      You can find the letters here:

    4. John,

      Thanks for the comment, surprisingly people do read these old posts. Some weeks I'll have 50-100 reads on some post I did a year or two ago. It's strange, almost random.

      Great information, I never thought to calculate Graham's returns from his partnership letters. It just goes to show that he really did do well practicing what he preached in Security Analysis.

  8. I don't think it's wise to think of BV independently of the earnings those assets produce. There are exceptions, but generally speaking it's not the best way to value/think about it.

    1. I actually disagree, assets have value regardless of whether they are generating income or not. A pile of assets might be worthless with one management team, but generate considerable earnings with a second. So are those assets worthless?

    2. Sorry for the dalay, was traveling for work.

      If you'd read my original comment again you'd see I never said assets are worthless without earnings. I said in most situations, it isn't wise to think of either BV or earnings independently of each other.

      Unfortunately I don't have time to get into it at the moment but I'll send you an email sometime this weekend.

  9. @panther1158

    No, I meant payables and accrued expenses.

    1. @red

      I see your point, but anyone buying a company that trades at a discount to book with negative working capital is playing with fire.

      Net-net investing typically takes current assets and nets out all liabilities including current liabilities. This is the original Ben Graham approach.

      Usually you get in trouble with companies that are burning tons of cash and have bad management teams. Working capital was positive but ends up going negative because of cash-burn. These are not good situations.

  10. Nate,

    I'm sure you already know this, but just to be clear: there's a difference between "book value" and "tangible book value". Several times in this post you describe book value as being "tangible", which is of course not always the case (e.g., goodwill for acquisitions).

    - aagold

    1. Yes, thanks, I meant tangible book value. There's often a lot of value in intangibles, but that's a lot harder to get a grasp on. I should have been clearer.

  11. I'll try and explain myself better, but I think Josh got an understanding of what I was getting at.

    If we take an extreme hypothetical example, lets say we have a locked box with $100 cash in it. The $100 represents the assets of the business, so it could represent plants, receivables etc.

    This locked box may provide a return (by magic!?) of $1 per annum and is available for sale for $20 say in the market, even though we know it has $100 in it.

    Now when you buy the box you can open it and take your $100 and lose the $1 per annum, or try and infuse a bit more magic to make it produce more money, which may or may not work.

    The problem is that you dont have the key, management do, and they dont want to open the box. Now there is a chance in say 5 years that management will open the box and just give you your $100 but you dont really have any way of knowing for sure.

    Under the scenario that management dont open the box, you as a shareholder may own $100, but it doesn't bring you any benefit. All you personally gain is $1 per year, so in my opinion the business should be valued based on these cash flows alone.

    But if they open the box in 5 years, that doesn't mean I will pay $100 today for $100 in 5 years, I'll obviously discount it back to today, and add extra on for the $1 a year I get in the interim.

    The chance of the box being unlocked (i.e. management disposing of underperforming assets) is subjective and up to the investor to determine. But if that chance is slim then I personally will just assume it wont happen to be cautious.

    Thats what I mean when I say the operating assets of a company aren't really worth anything to a shareholder (besides the cash flow they produce) unless management is prepared to liquidate them if they dont perform to shareholders expectations.

    I'll freely admit though that I am not the typical Ben Graham type investor though, I am much more inclined towards a Buffett style, or more GARP investing but am always on the lookout for companies with lots of 'non-core' assets they will be prepared to liquidate.

  12. With regards to premier exhibitions isn't this speculation rather than investment, and would they be able to dispose of an historical collection?