|Home||Introductory Topics||Intermediate Topics||Commentary Archive||Calculators|
By now you've probably heard of Warren Buffett, who became a billionaire by investing in the stock market. He runs a company, Berkshire Hathaway, which used to be primarily an investment vehicle and is now a diversified conglomerate. Between 1965 and 2003, the company's book value has grown by 259,485%, while the S&P (with dividends reinvested) as returned 4,783%.
Berkshire clearly will not be able to duplicate these results. The company has a huge amount of money to invest (almost $100 billion, and growing) which limits Buffett's investment opportunities; it makes little sense for him to invest in small companies, as he wouldn't be able to invest enough money in them to be worthwhile. But Buffett hopes to "modestly" outperform the S&P 500.
So let's start valuing the company. Berkshire is a diversified conglomerate, and as a result, it can't be measured as most companies usually are. We must break the company into several different parts and analyze each part.
First, you'll need to go to berkshirehathaway.com. We'll be getting our data from two sources: The most recent interim shareholder report, and Berkshire Hathaway's letter to shareholders, from its 2003 Annual Report.
I like to break the company into three parts: Insurance float, investments, and wholly owned companies.
1) Insurance float. When customers pay premiums to an insurance company, it holds on to that money until it has to pay claims. In the mean time, it can invest this money (which is known as "float") and earn interest on it. Usually, insurance companies take in less in premiums than they pay in claims; in other words, they operate at an "underwriting loss." They make their profit from the interest on their float.
Berkshire's total insurance float is about $44 billion, and Geico's float represents $5 billion of this.
Float may be worth more or less than cash, depending on the circumstances. Most of Berkshire's insurance businesses operate at an underwriting loss, meaning, in effect, that their "float" is like "borrowing" money at a certain interest rate. According to Berkshire Hathaway's vice president Charlie Munger, this rate is around 3%. This float will be invested mostly by Warren Buffett, although a guy named Lou Simpson manages a small percentage of it. Both of these investors are very good at earning decent returns. Even better, this float is growing, and as it grows, its value increases. I might value the non-Geico float at 90% of total float value, which is total float minus $3.9 billion.
Here's a possible calculation of the float's value: Assume that you are using a discount rate of 10%. Further assume that the float earns a return of 8% (I assume that Berkshire's float will always partially consist of short term bonds, which usually generate low returns. So I would expect the return on Berkshire's float to be a little bit lower than the return on is other investments).
Berkshire also has investments that are outside of its insurance float, and we must consider two cases here: either these non-float investments will earn a return of about 10% (equal to our discount rate) or it will earn a different return, probably less. This is important because an insurance company grows, it typically has to retain part of its earnings and add it to its equity, in order to ensure that, in adverse conditions, the company will have enough assets to be able to pay out claims. If the insurance company earns a good return on these investments, the cash has as much value in the company's hands as it does in yours. But if the company earns a low return on these investments, then they are worth less than their cash value. In this case, it would be more difficult to value the company.
For Berkshire Hathaway, let's assume that non-float cash will earn a return of 10% (equal to our discount rate).
Assume that the cost of the float is 3%. Then, 8% of the float is being earned as investment income, and 3% of it is going towards expenses. So, the return Berkshire will earn on the float is 8% - 3%, which is 5%. Add a tax rate of 15%, and the company is earning 4.25% on the float, after tax. If this float grows at 3%, we essentially have an income stream paying 4.25% of float value and growing at 3%. Suppose we call X (our unknown variable) the float value. The float is growing at 3%, so if Berkshire earns a 7% return on our float value X, the income stream will return 10% (a 7% income return + 3% growth=a 10% total return.) Algebraically, we get:
Float value * 4.25%=X * 7%. So,
X=float * 4.25 / 7
or about 0.61 * float.
Now assume that Berkshire returns 9% on its float, and the float grows at 5%, indefinitely. Then our income stream is paying 6% on the float, which is 5.1% after taxes. So our new equation is:
Float value * 5.1%=X * 5%.
So X is roughly equal to float, meaning that the float is worth about as much as cash. If you believe these two events have an equal probability of occurring, you could average these and get a float value of 80% of cash.
Now let's look at Geico's float. Auto insurance companies generally earn an underwriting profit. So they are, in effect, paid to hold others' money. It's reasonable to expect that Geico's pre-tax underwriting profit (100 - its "combined ratio") will average about 4%. Assuming a 35% tax rate, we get an underwriting profit of 2.6%. And since Geico's float is only a about 2/3 of its revenues, we get an underwriting profit as a percentage of float of 2.6% divided by 2/3, which is about 4%. So Geico is being paid 4% interest to hold other people's money (imagine borrowing $200,000 from the bank to buy a house, where you don't have to make any mortgage payments, and in fact, the bank pays *you* every month). This leads to the conclusion that Geico's float is worth more than cash.
Furthermore, Geico is growing at a substantial rate. As a direct insurer, they don't have to pay agents and are, overall, able to charge lower prices than their competitors (in some areas, however, their rates are not the lowest). As a result, this float should continue to grow at a reasonably high rate, which adds additional value to the company. I might value this float at two times the float's value, or total float plus $5.2 billion.
So the value of the insurance float would be 44 billion - 3.9 billion + 5.2 billion, or float + 1.3 billion. The actual float is represented by investments on the balance sheet, so we merely need to add $1.3 billion to the total cash and investments that Berkshire holds. You don't want to add this float to the total value of Berkshire's investments, because then you'd be counting it twice.
2) Investments that are not part of the company's insurance float. Looking at the company's latest quarterly report, at http://www.berkshirehathaway.com/qtrly/interim.html, we see that they have, in billions of dollars:
3) Wholly owned companies.
From Berkshire Hathaway's annual report, the Pre-Tax Earnings (in $ millions) for Berkshire's subsidiaries are:
*From date of acquisition, May 23, 2003
Values I might chose for these companies are:
Building Products - $4500
Shaw Industries - $3500
Apparel - $2300
Retail - $1800
Flight Services - $2500
McLane - $2000
(note: these values include a margin of safety.)
For the others, I might take the value of Berkshire's investment in Mid-American Energy from Berkshire's balance sheet, where it is listed at 3.8 billion, and I might add the net asset value of Berkshire's finance operations, which is 7 billion. There are a couple of other businesses which are probably worth around $1-2 billion or so. We get a total value of about $29 billion.
Now you just need to add the values of these three components, and you have your buy price.