POST OF THE DAY
Berkshire Hathaway
Berkshire EV / Investments

Format for Printing

Format for printing

Request Reprints

Reuse/Reprint

By howardroark
February 9, 2006

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

Thought I'd offer a look at one way I think about Berkshire's valuation. As a warning, this approach is not qualitatively different than the many credible back of the envelope valuations that this board has seen over the years. The sameness shouldn't be all that surprising; but it means this post might not be particularly helpful, either. In fact, the only pretense of a twist in this approach is that as opposed to valuing the company whole hog, it just tries to isolate Berkshire's market valuation compared to its gross investments.

The reason I do this is because Berkshire's enterprise value to investments ratio should be somewhat comparable over time. Yes, things might change enough to justify substantially different fair multiples over different periods. But to me this has always been a reasonable point of departure before going on to think about the impact of any such changes.

My method is simple. Start with 10X the operating income of the MSR businesses, which equates to roughly 15-16 times earnings. I think this is a fair multiple on balance given the constituent businesses, but a point or two in either direction doesn't move the needle enough to agonize over. Note that I'm excluding Finance (including Clayton), Mid-American, and all Insurance underwriting from segment of the analysis. If MSR generated $2.71b in 2005 EBIT, then the value of this group is $27.1b.

Next I add the value of Berkshire's stake in Mid-American. For the purposes of this post, I'm excluding any additional (or destroyed) value from the Pacificorp acquisition. I am likewise giving Mid-American a 10X EBIT (meaning after-project related interest costs but before debt payments owed to Berkshire) of $800m estimated for 2005. Berkshire's 81% fully diluted (83% before dilution) equity interest is thus worth $6.4b plus another $1.4m in owned debt, leaving total Mid-American value of $7.8b. You might argue that cyclically high HomeServices results are inflating this valuation, but this would be an incredibly small adjustment and could be argued to be offset by other under appreciated Mid-American strengths.

Then I add the value of the Financial Products business. Given Clayton's economics and Buffett's historical trading results, I think a 1.5X book value is reasonably conservative here, though you could credibly argue for anywhere between 1X and 3X book. At $4.2b in book value, this segment adds $6.3b to the overall valuation.

To recap so far, you have $27.1b from the (MSR) operating businesses, $7.8b from Mid-American, and $6.3b from Finance, for a total of $41.2b. To give some context, this comes to roughly 30% of Berkshire's current market cap.

Then I try to value GEICO separately, because it is a unique business with significantly different economics than the rest of Berkshire's insurance business. Valuing GEICO isn't easy. When Berkshire paid $70 a share for the half of GEICO it didn't own over ten years ago, it valued the whole at around $4.6 billion -- which Buffett then called a steep price. To offer some imperfect metrics, that value came to about 1.5X premiums earned and 1.7X GEICO float at the time. Since then, GEICO has been an incredible performer, consistently growing premiums earned in the double digits while still generating unprecedented underwriting profits. Buffett used to say that 4% underwriting margins were "normal" for GEICO, and that was in a time when growth marketing expenditures weren't nearly as high relative to maintenance marketing -- the recent underwriting cycle has seen margins edge into the double digits. 1.5X current year premiums would implies a $15B valuation.

Another option is to look at PGR, which appears to valued at around 1.6X EV/premiums earned, similar to the 1995 Buffett purchase of GEICO. You could argue that PGR is overcapitalized relative to a hypothetical GEICO valuation which implicitly leaves any excess capital to be valued as general Berkshire investments (but of course includes GEICO float). At 1.6X run rate premiums, GEICO's valuation would be $16 billion. From a bottom-up basis, this initially looks conservative against the current fantastic underwriting environment. For example, if you subtract GEICO estimated year end float of $6.5b at 80%, this valuation implies only 9X 2005e pre-tax profits, an inappropriately low multiple on normal earnings given GEICO's incredible growth and returns. But if you assume GEICO's current margins are frothy, which may be supported by recent rate reductions, than the $16b valuation looks a little more reasonable. For example, at 6% pretax margins and valuing float at face, it implies 17X pretax income (this again assumes float valued at 80% of face). As I mentioned, GEICO is an extremely tough company to value -- especially when you are taking its capitalization as kind of an abstraction, considering the cyclicality underwriting profits and the expensive of growth related marketing costs -- but $16b seems like a reasonably conservative guess. Gun to my head, I would bet it turns out to be worth at least that.

Adding that to the other operating businesses, you now have $57.2b in operating value. This comes to 42-43% of Berkshire's current market cap. You are left with Berkshire's investments -- both excess capital and insurance float -- and the value or cost of the underwriting that produces its float. Berkshire ended Q305 with a little over $113b in investments. Less GEICO float (already counted in the GEICO valuation), investments drop to about $107b. I like to then subtract 30% of Berkshire's deferred tax liability ($11.4b * .3 = $3.4b) as a fairly arbitrary estimate of the PV cost of that liability. That leaves $103.4b in net investments.

As mentioned, I've valued Berkshire's operating businesses at $57.2b. Berkshire's current Enterprise Value is $135b plus $3.5b in parent company notes payable (finance debt already subtracted by virtue of using a book value based valuation for the finance segment), for a total of $138.5b. Subtracting the operating business valuation above leaves $81.3b in adjusted enterprise value. By adjusted enterprise value, I simply mean an isolated estimate of how the market is valuing Berkshire's investments and the source of those investments (insurance). This compares to the above figure of $103.4b of adjusted net investments, leaving and EV per investment of 0.79X. Thus, if you buy the above estimates, the market is valuing Berkshire's investments at around 79 cents on the dollar.

It's not possible to go through a wholly identical process to make historical comparisons, as the components and features of Berkshire's operating businesses have changed over time. But I think you can come reasonably close using similar heuristics to come to informative comparisons over Berkshire's EV / Investment over a few key points in time.

In 1995 (meaning, at the time of the 1995 report or early 1996), Berkshire was valued at around $1.60 per $1 in investments, given a 10X EBIT multiple on its operating business after adjusting for its deferred tax liability. This is when Buffett wrote that he would not consider buying Berkshire at its current price. The next year, in 1996, Buffett said he thought the price was more appropriate. The implied EV / Investment multiple had fallen to $1.20 to $1, using similar assumptions.

At $45K per A share in early 2000, Buffett made his well-known repurchase offer. At the time, if you apply the same multiples to operating earnings (10x) and a similar thought process to GEICO's valuation (which would value it at $6-$7b and imply 16% CAGR in value since then) given its business at the time, I think the EV / Investment multiple was around $0.81 to $1 (this includes similar adjustments such as deferred taxes and financial products). Of course, Berkshire's major equity stakes were a little more richly valued back then.

That's not to say that that the stock is definitively cheaper today at EV/Inv of $0.79 as it was when Buffett was offering to repurchase it in 2000. As mentioned, many things can alter what defines a fair EV/Inv multiple. The key factors, of course, are the cost of float, the prospects for float growth or retraction, and investment prospects (including the attractiveness of existing investments). I think you can make the argument that Berkshire's prospects in these areas are not on balance worse than they have been in the recent past. It's true that GenRe has a lot to prove over the cycle, and that the market is softening with some real risk of float contraction for Berkshire; but it's also probably true that its investment portfolio and opportunities are better than they have been in some time, and legacy GenRe exposure is lower than it has been.

From an absolute valuation perspective, the exercise gets a little more difficult. If you want to walk around and value a dollar of float at a dollar as a general rule, you'd leverage your house to own pretty much every public reinsurance company, thinking you got them at pennies on the dollar. Of course, Berkshire is massively overcapitalized, and only $43b of its $103b in net investments is non-GEICO float. So one approach is to take the $70b in non-float investments and value it at face. I think this is pretty conservative as even given legitimate concerns about age and Berkshire's huge capital base, it's a pretty good deal to get Buffett as your investment manager at par. Taking that $70b away from the above estimated $81.3b market valuation of Berkshire's total investments, you are left with an $11.3b market value of $43b in non-GEICO float. Making this further adjustment (valuing investments in excess of float at face) for early 2000 at $45K per A share, Berkshire's $22b in non-GEICO float was coincidentally also value at around $11b using the many assumptions identified above.


Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.