Case Study: Accounting vs. Economic Profit at GEICO

“Our calculus is different: We simply measure whether we are creating more than a dollar of value per dollar spent — and if that calculation is favorable, the more dollars we spend the happier I am.” —Warren Buffett

Accounting vs. Economic Profit: Acquisition of New Policy Holders

Sanjay Bakshi has written about Warren Buffett’s views about the decisions relating to customer acquisition costs for GEICO. Bakshi refers to an interview with Tom Russo “where he talked about the logic of Buffett’s decision to lay out large sums of money to acquire new, likely to be profitable customers for GEICO, even though such actions would severely penalise near term earnings.”

In this interview Tom Russo gives some insight as to how to think about the short-term earnings impact from acquiring additional policy holders, versus the change in intrinsic value (emphasis added).

When Warren Buffett bought GEICO, the company had something like two million policyholders, and Berkshire’s belief was that they were the best insurer in the country. So why should they have a bigger than two percent market share? And the answer was that every new policy that they put on the books reported losses to the firm of $250 of operating losses: reported losses at least when they would sign up for those new policies. And of course at the same time because of high persistency and because of the great business model of GEICO the net present value of each new insured was $1500. So when it was a separately run public company, GEICO was constrained by not being able to take the steps to grow the business optimally because they had to report a $250 loss for every new insured and they kind of measured their growth in the modest pace. Once it came inside of Berkshire, Warren who doesn’t really care about reporting profits, he was delighted to take steps that could increase the firm’s net present value by adding new policy holders even if it meant that in the year of acquisition, he showed more losses. And you have seen as anybody who has inhabited the earth over the past two decades would have seen this extraordinary life of lizards and cavemen and other forms of spokespeople for Berkshire’s GEICO division and they cost a lot of money. His advertising budget went from $30 million to $950 million over those years. And yet the number of insured went from two million to 10 and with $1500 NPV for every new insured customer the value was around $15 billion. Berkshire did that in part because they were permitted to suffer the report the losses of the first year of and celebrate the NPV gains that they picked up. That is one example how Berkshire showed the value of this approach.

In the table below, I have visualized the two scenarios that Geico faces when deciding whether or not to acquire an additional policy holder. The short-term effect is an expense of $250, that will lower reported earnings for the fiscal year in wich the policy holder is aquired. The short-term impact is negative when looking at the accounting profit for the year. But, the long-term effect is just the opposite, since in this case adding a new policy holder results in a net present value (NPV) of $1,500, i.e. the economic profit that adds to Geico’s intrinsic value.


First-Level vs. Second-Level Thinking

When I read the text above about the short-term P&L impact from acquiring a new policy holder, versus the long-run value creation that adds to the business’s intrinsic value, I came to think of Howard Marks and his mental models first- and second-level thinking. It looks like the Geico acquisition case above fits right into these models.

Let’s refresh our minds, and start with Howard Marks description, from his book The Most Important Thing, about what separates first-level thinking from second-level thinking.

Howard Marks answers the question “What is second-level thinking?” by giving a few examples, among them this one:

• First-level thinking says, “I think the company’s earnings will fall; sell.” Second-level thinking says, “I think the company’s earnings will fall less than people expect, and the pleasant surprise will lift the stock; buy.”

Marks then goes on to explain the two different ways of thinking.

First-level thinking is simplistic and superficial, and just about everyone can do it (a bad sign for anything involving an attempt at superiority). All the first-level thinker needs is an opinion about the future, as in “The outlook for the company is favorable, meaning the stock will go up.”

Second-level thinking is deep, complex and convoluted. The second-
level thinker takes a great many things into account:

• What is the range of likely future outcomes?
• Which outcome do I think will occur?
• What’s the probability I’m right?
• What does the consensus think?
• How does my expectation differ from the consensus?
• How does the current price for the asset comport with the consensus view of the future, and with mine?
• Is the consensus psychology that’s incorporated in the price too bullish or bearish?
• What will happen to the asset’s price if the consensus turns out to be right, and what if I’m right?

Geico: From a Second-Level Thinking Perspective

From the Geico policy holder acquisition case above, we could formulate our own rules as:

• First-level thinking says, “The company’s earnings will fall; don’t acquire any/or too many new policy holders in a certain fiscal year.” Second-level thinking says, “NPV for each new policy holder signed up is $1,500; acquire additional policy holders as long as the NPV for every new policy holder is positive.”

Let’s end this post with another Buffett quote worth remembering.

“Accounting consequences do not influence our operating or capital-allocation decisions. When acquisition costs are similar, we much prefer to purchase $2 of earnings that is not reportable by us under standard accounting principles than to purchase $1 of earnings that is reportable.” —Warren Buffett


What GEICO’s Customer Acquisition and Associated Costs Taught Me about Business Economics, Management Quality, and Valuation (by Sanjay Bakshi)


The Security I Like Best: GEICO

“Of course the investor of today does not profit from yesterday’s growth. In GEICO’s case, there is reason to believe the major portion of growth lies ahead.” —Warren Buffett


Click here for the article in PDF format.


The Government Employees Insurance Co.

Full employment, boom time profits and record dividend payments do not set the stage for depressed security prices. Most industries have been riding this wave of prosperity during the past five years with few ripples to disturb the tide.

The auto insurance business has not shared in the boom. After the staggering losses of the immediate postwar period, the situation began to right itself in 1949. In 1950, stock casualty companies again took it on the chin with underwriting experience the second worst in 15 years. The recent earnings reports of casualty companies, particularly those with the bulk of writings in auto lines, have diverted bull market enthusiasm from their stocks. On the basis of normal earning power and asset factors, many of these stocks appear undervalued.

The nature of the industry is such as to ease cyclical bumps. The majority of purchasers regards auto insurance as a necessity. Contracts must be renewed yearly at rates based upon experience. The lag of rates behind costs, although detrimental in a period of rising prices as has characterized the 1945-1951 period, should prove beneficial if deflationary forces should be set in action.

Other industry advantages include lack of inventory, collection, labor and raw material problems. The hazard of product obsolescence and related equipment obsolescence is also absent.

Government Employees Insurance Corporation was organized in the mid-30’s to provide complete auto insurance on a nationwide basis to an eligible class including: (1) Federal, State and municipal government employees; (2) active and reserve commissioned officers and the first three pay grades of non-commissioned officers of the Armed Forces; (3) veterans who were eligible when on active duty; (4) former policyholders; (5) faculty members of universities, colleges and schools; (6) government contractor employees engaged in defense work exclusively, and (7) stockholders.

The company has no agents or branch offices. As a result, policyholders receive standard auto insurance policies at premium discounts running as high as 30% off manual rates. Claims are handled promptly through approximately 500 representatives throughout the country.

The term “growth company” has been applied with abandon during the past few years to companies whose sales increases represented little more than inflation of prices and general easing of business competition. GEICO qualifies as a legitimate growth company based upon the following record:

Year— Premiums Written Policy Holders
1936… $103,696.31 3,754
1940… 768,057.86 25,514
1945… 1,638,562.09 51,697
1950… 8,016,975.79 143,944

Of course the investor of today does not profit from yesterday’s growth. In GEICO’s case, there is reason to believe the major portion of growth lies ahead. Prior to 1950, the company was only licensed in 15 of 50 jurisdictions including D. C. and Hawaii. At the beginning of the year there were less than 3,000 policyholders in New York State. Yet 25% saved on an insurance bill of $125 in New York should look bigger to the prospect than the 25% saved on the $50 rate in more sparsely settled regions.

As cost competition increases in importance during times of recession, GEICO’s rate attraction should become even more effective in diverting business from the brother-in-law. With insurance rates moving higher due to inflation, the 25% spread in rates becomes wider in terms of dollars and cents.

There is no pressure from agents to accept questionable applicants or renew poor risks. In States where the rate structure is inadequate, new promotion may be halted.

Probably the biggest attraction of GEICO is the profit margin advantage it enjoys. The ratio of underwriting profit to premiums earned in 1949 was 27.5% for GEICO as compared to 3.7% for the 135 stock casualty and surety companies summarized by Best’s. As experience turned for the worse in 1950, Best’s aggregate’s profit margin dropped to 3.0% and GEICO’s dropped to 13.0%. GEICO does not write all casualty lines; however, bodily injury and property damage, both important lines for GEICO, were among the least profitable lines. GEICO also does a large amount of collision writing, which was a profitable line in 1950.

During the first half of 1951, practically all insurers operated in the red on casualty lines with bodily injury and property damage among the most unprofitable. Whereas GEICO’s profit margin was cut to slightly above 9%, Massachusett’s Bonding & Insurance showed a 26% loss, New Amsterdam Casualty an 8% loss, Standard Accident Insurance a 9% loss, etc.

Because of the rapid growth of GEICO, cash dividends have had to remain low. Stock dividends and a 25-for-1 split increased the outstanding shares from 3,000 on June 1, 1948, to 250,000 on Nov. 10, 1951. Valuable rights to subscribe to stock of affiliated companies have also been issued.

Benjamin Graham has been Chairman of the Board since his investment trust acquired and distributed a large block of the stock in 1948. Leo Goodwin, who has guided GEICO’s growth since inception, is the able President. At the end of 1950, the 10 members of the Board of Directors owned approximately one third of the outstanding stock.

Earnings in 1950 amounted to $3.92 as contrasted to $4.71 on the smaller amount of business in 1949. These figures include no allowance for the increase in the unearned premium reserve which was substantial in both years. Earnings in 1953 will be lower than 1950, but the wave of rate increases during the past summer should evidence themselves in 1952 earnings. Investment income quadrupled between 1947 and 1950, reflecting the growth of the company’s assets.

At the present price of about eight times the earnings of 1950, a poor year for the industry, it appears that no price is being paid for the tremendous growth potential of the company.


Buffett & GEICO: 1976 Letter and 1951 Research Report

“At the time I felt that GEICO possessed an extraordinary business advantage in a very large industry that was going to continue to grow. Since that time they never have lost that advantage – the ability to give the policyholder back in losses a greater percentage of the premium dollar than any other auto insurance company in the country, while still providing a profit to the company. I always have been attracted to the low cost operator in any business and, when you can find a combination of (i) an extremely large business, (ii) a more or less homogenous product, and (iii) a very large gap in operating costs between the low cost operator and all of the other companies in the industry, you have a really attractive investment situation. That situation prevailed twenty-five years ago when I first became interested in the company, and it still prevails.”
—Warren Buffett

WB11976 GEICO Letter

“Warren Buffett recently gave the WSJ a letter he wrote in 1976 to one of Berkshire’s top executives. In the letter, Warren Buffett explained his reasoning behind the GEICO purchase.” (Source:

See here for this letter about GEICO, written by Warren back in 1976.

1951 GEICO Research Report

Another good read, also about GEICO, is the reprinted research report The Security I Like the Best, written by Warren back in the end of 1951.

See here for this report that was included in the Berkshire Annual report from 2005 on page 24.

WB2Disclosure: I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company or individual mentioned in this article. This article is informational and is in my own personal opinion.