With the transformation that Berkshire has gone through over the past few years, I think we're missing the big picture if we focus only on Warren's theory on investing. Today, Berkshire's insurance businesses make up around 70 percent of its revenue. 2. "From time to time we will have such transactions in the future and we will post you of their effects upon our current earnings. When priced properly, this pain-today, gain-tomorrow business is welcomed by us and benefits our shareholders." (Selected highlights Q3 2000) 3. "BHRG is actively negotiating additional retroactive reinsurance contracts, which, if such transactions are consummated, will produce additional premiums in excess of $3 billion. It is not anticipated that these contracts will be finalized until 2001." (10-Q Q3, 2000)
I want to put myself in a position where I can better understand the things that Warren has to say about our insurance business. With the annual report coming up next month, I thought it would be interesting to take a look at some of Warren's recent comments. In particular, I hope to gain a better understanding of the following three statements: (Bold emphasis is mine.)
1. "Due to the large retroactive reinsurance contracts entered into during 1999, deferred charges increased significantly. Consequently, the periodic amortization and therefore, underwriting losses are expected to increase in future periods." (1999 Annual Report)
Important issues. Therefore, over the past few days I have tried to educate myself.... and I made some notes along the way.
Preliminary Information: (Those who want to cut to the chase - skip down to "Pain Today - Gain Tomorrow")
Float: "Unless you understand this subject, it will be impossible for you to make an informed judgment about Berkshire's intrinsic value. Float is money we hold but don't own. In an insurance operation, float arises because premiums are received before losses are paid, an interval that sometimes extends over many years. During that time, the insurer invests the money. Typically, this pleasant activity carries with it a downside: The premiums that an insurer takes in usually do not cover the losses and expenses it eventually must pay. That leaves it running an "underwriting loss," which is the cost of float. An insurance business has value if its cost of float over time is less than the cost the company would otherwise incur to obtain funds. But the business is a lemon if its cost of float is higher than market rates for money." (1997 Chairman's Letter)
An insurance company's value is determined by the amount of float that it generates and, more importantly, the cost of that float. Float is like debt in the fact that float provides Berkshire with money. Berkshire earns investment income on this float until the claim is paid. Like debt, float carries a financing cost. Unlike debt, the cost of float is never certain. The spread between the amount of investment income that is earned on the float - and the cost of obtaining the float - is what determines the value of the float.
"The combined ratio represents total insurance costs (losses incurred plus expenses) compared to revenue from premiums: A ratio below 100 indicates an underwriting profit, and one above 100 indicates a loss. The higher the ratio, the worse the year. When the investment income that an insurer earns from holding policyholders' funds ("the float") is taken into account, a combined ratio in the 106 - 110 range typically produces an overall break-even result, exclusive of earnings on the funds provided by shareholders." (1992 Chairman's Letter)
Loss Ratio = Incurred Claims plus Increase in Reserves / Premiums Earned
Expense Ratio = Incurred Expenses / Premiums Written
Combined Ratio = Loss Ratio Plus Expense Ratio
Over the past 30 years, Berkshire's cost of float has been less than 2%. Less than the government borrowing rate. Keep in mind that, in those years when there is an underwriting profit, Berkshire gets "paid" for holding "other peoples money".
Warren provided an "Approximate Cost of Funds" table (1967 to 1997) in the 1997 Annual Report. It's interesting to note that the approximate cost of funds was "Less than Zero" in 17 out of 31 years. Of the remaining 14 years, there were only 5 times when the cost of funds was higher than the long-term bond rate.
Super-Cat Insurance: "We sell policies that insurance and reinsurance companies purchase in order to limit their losses when mega-catastrophes strike. Berkshire is the preferred market for sophisticated buyers: When the "big one" hits, the financial strength of super-cat writers will be tested, and Berkshire has no peer in this respect."
"Since truly major catastrophes are rare occurrences, our super-cat business can be expected to show large profits in most years -- and to record a huge loss occasionally. In other words, the attractiveness of our super-cat business will take a great many years to measure. What you must understand, however, is that a truly terrible year in the super-cat business is not a possibility -- it's a certainty. The only question is when it will come." (1997 Chairman's Letter)
Insurance is a commodity: There are hundreds of insurance companies each selling similar products at independently established prices. This is usually a recipe for a less than exciting business.
The only time that insurance is an interesting business is when there is a shortage of capacity. A shortage of capacity only occurs when others are running from the business.
In the insurance business, Berkshire's primary sustainable competitive advantages are the fact that we: 1. Will only enter into policies which are appropriately priced for the risk incurred (Disciplined underwriting skills). 2. "We are creating Fort Knox". Remember that Warren Buffett quote from the Gen Re meeting? If appropriately priced, Berkshire has a strong desire, and ability, to write a lot of very large insurance policies. And, we have the reputation of promptly writing the check when a claim is made. We are "Fort Knox". Name brand recognition. I think these are some of the reasons why Warren said (at the Gen Re meeting) that "We should be number one - by a significant margin - in the world reinsurance market in ten to 15 years".
If Berkshire is "Fort Knox" then, in theory, any Super-Cat that results in Berkshire having to write a large check could also be the same Super-Cat that ends up being devastating for many other insurance companies to try to handle. Berkshire would suffer a temporary setback while other companies may not survive. When other insurance companies are on the roaps and looking for help, Berkshire's superior capital strength will allow Warren and Ajit to write exceptionally large amounts of risks. Warren has built a better mousetrap.
Wealth Effect: Insurance is a lot like investing. After a few years of good returns from the market, many people will start feeling very wealthy. They might take more expensive vacations than normal, buy a bigger house, a new car etc.. With insurance companies, a few good years of of investing coupled with good underwriting performance will often times lead to a "soft" market. Companies, feeling the "wealth effect", willingly decrease prices in an attempt to capture more market share and/or enter new lines of business. Just like the stock market, there's simply too much capital chasing too few "opportunities".
The well-managed insurance company is much like the successful investor. The well-managed insurance company will simply sit on the sideline during times like this. They know that, when the catalyst for improved pricing comes, those companies who have been undisciplined with their underwriting skills will have to pay the piper.
Unfortunately, the catalyst which will once again improve pricing will usually come in the form of a "catastrophe", or a prolonged drop in the bond or stock market. Those companies who have managed their risk appropriately will be in a prime position to take full advantage of the attractive pricing environment which will usually follow this "catalyst". Weaker companies will suffer. Despite the higher prices, the weaker companies will have to turn away business. "Fort Knox" wins by default.
Was Gen Re a mistake?: Here's a link to Warren's list of "Synergies" concerning the Gen Re merger. Keep in mind that, prior to this, Warren had never used the word "synergy" when listing the reasons for a Berkshire acquisition.
Many people think that Gen Re was a mistake. I don't. Here's why:
1. At the Gen Re meeting, Warren warned that Gen Re would not contribute much for at least two years. He also mentioned that, due to the soft market, it would be very difficult for Gen Re to write at a combined ratio of 100. However, don't lose sight of the fact that the insurance industry is cyclical and soft markets don't last forever. Gen Re has a historical combined ratio of 100 (101 accounting for options).
2. It took some time to get Gen Re on the same page with Berkshire's strategic plan: "At both General Re and its Cologne subsidiary, incentive compensation plans are now directly tied to the variables of float growth and cost of float, the same variables that determine value for owners." (1999 Chairman's Letter)
3. Alice Schroeder estimated that for each 1% Change in the Cost of Gen Re Float, you should adjust IV by $2,923.00 per share. Fortunately, underwriting results are a lagging indicator and, hopefully, Gen Re has already turned the corner.
4. Berkshire got a good deal. I have in my notes that Berkshire paid 22 times TTM EPS, but only 0.74 times adjusted book value.
5. Timing of the purchase. Berkshire purchased Gen Re in December, 1998. A point when overcapacity made for a very soft market. By the 3rd quarter of 2000, the insurance cycle had started to turn and we began to see an upturn in rates. Under Berkshire, Gen Re now enters this improved underwriting environment in the enviable position of having unlimited access to capital, and shareholders who don't care about smooth quarterly earnings.
Pain Today - Gain Tomorrow
Berkshire continues to write large reinsurance contracts that will increase float, but sacrifice near term earnings. These are contracts where risk is transferred from one insurer to another. Berkshire agrees to indemnify the ceding insurer for all or part of the claim liabilities under policies issued by the ceding insurer. In return, the ceding insurer pays Berkshire a premium for this coverage.
Essentially, Berkshire enters into a contract with another insurance company in order to assume the risk of a large, unexpected, payout from a contract which the other company is burdened with. Berkshire will accept these deals when they have a good understanding of the odds concerning the risk associated with the ultimate payout, and the margin of safety makes the contract attractive. The ceding company pays Berkshire X amount of dollars to assume losses of XX amount of dollars. The premiums charged for these policies are based on time discounting of loss payments and, as you might expect, Berkshire calculates this with a margin of safety built in. How much of a profit margin is built in, I don't know. This type of business is attractive due to the large amount of float that it generates.
In return, the ceding company gets to report smoother earning to Wall Street. The ceding company's earnings look good.... Wall Street likes this, and rewards this. Berkshire's earnings look bad.... but the long term economic gain belongs to Berkshire. Reason being - Berkshire is able to invest the large premium they received in exchange for this contract. If the contract was priced appropriately, and the premium is invested wisely, the investment earnings over the long term will more than offset the maximum exposure incurred with this contract.
This is a risk versus reward scenario. In this scenario, Ajit's (and Warren's) probability analysis determines the risk of the contract, and Warren's investing skills determine the degree of the reward. This, once again, highlights the importance of Warren's eventual successor.
Maximum Exposure is Predetermined: Other insurance companies willingly pay Berkshire a large premium to assume their risk. The other insurance companies are willing to do this in order to smooth out earnings and please their shareholders. Their shareholders should be aware of the fact that this contract could still come back to haunt them. Berkshire agrees to a predetermined maximum exposure with each contract they enter into. Once this cap is met (if it is met), the contract then reverts back to the original insurer for the remainder of the claim. This could leave the original insurance company's shareholders singing that old country song "I went to bed at 2 with a "10", and woke up at 10 with a "2".
"From time to time we will have such transactions in the future and we will post you of their effects upon our current earnings. When priced properly, this pain-today, gain-tomorrow business is welcomed by us and benefits our shareholders." (Selected highlights Q3 2000)
The accounting for these contracts can be difficult to understand. There seems to be two different methods of accounting for the underwriting loss associated with these contracts. That's what I'm going to try to get into next. I will add Bold emphasis to the following quotes in an attempt to highlight some of the key elements of each type contract.
Retroactive Insurance: "Generally, retroactive reinsurance contracts indemnify the ceding company, subject to aggregate loss limits, with respect to past loss events that were insured by the counterparty. It is generally expected that losses ultimately paid under these arrangements will exceed the premiums received, possibly by a wide margin. Premiums are based in part on time-value-of-money concepts because loss payments are expected to occur over lengthy time periods. However, retroactive contracts do not significantly impact earnings in the year of inception. Consistent with Berkshire's accounting policy, the excess of the estimated ultimate losses payable over the premiums received is established as a deferred charge and amortized against income over the estimated future claim settlement periods." (1999 Annual Report)
This type coverage is designed to protect an insurer's current year results from the impact of a severe loss, or to help an insurer through a period where there has been an abnormal loss frequency.
Example: Let's say that Berkshire is paid a large premium ($1B) to assume another insurance company's liability - with a maximum exposure ($2B). Berkshire then establishes a deferred charge for this contract ($1B) that will be amortized over the term (many years) of the contract ($X Million per year). Now, with Retroactive Insurance accounting, the $X Million will present as an underwriting loss each year for the term of the contract.
This underwriting loss can be misleading. The fact is, Berkshire benefits from this contract in the long run if 1. they achieve the expected earnings from investments related to the premium for this contract 2. Total claims on the contract are less than anticipated. The float from these contracts will be beneficial to Berkshire's future by generating investment income for years to come.
Non-Cat Reinsurance: "As with retroactive reinsurance contracts, the premiums established for non-catastrophe reinsurance contracts are based on time-value-of-money concepts because loss payments are expected to occur over lengthy time periods. Loss reserves for this business are established without such time discounting but, unlike retroactive reinsurance contracts, no deferred charges are established. Consequently, significant underwriting losses result. This business is accepted because of the large amounts of investable policyholder funds ("float") that is produced. It is anticipated that Berkshire will derive significant economic benefits over the lengthy period of time that the float will be available for investment." (1999 Annual Report)
Example: Same scenario as with Retroactive Insurance, EXCEPT for the fact that Berkshire does NOT establish a deferred charge and amortize over the term of this contract. The accounting requirements for Non-Cat Reinsurance contracts require a one time, "lump sum", underwriting loss. Makes for lumpy quarterly earnings.
As with retroactive insurance, Berkshire benefits from the large amount of float produced which will generate greater earnings from investments. More fuel for the perpetual growth machine.
Since we are "Fort Knox", I think we can expect to see more of these type contracts in the future. (Companies struggling with financial difficulty, mergers which require reserve guarantees, etc.) Actually, expect to see them in the near future:
"BHRG is actively negotiating additional retroactive reinsurance contracts, which, if such transactions are consummated, will produce additional premiums in excess of $3 billion. It is not anticipated that these contracts will be finalized until 2001." (10-Q Q3,2000)
Expect to see them in the near future.
"Adjusted" Cost of Float: Let's take a look at the adjustments we might make to account for our "welcomed" loss:
1. Q3 2000 Berkshire reports a 5.1% cost of float.
2. Q3 2000 Berkshire reports an underwriting loss of $347M.
3. A Gen Re contract with another insurance company resulted in an underwriting loss of $135M. ("Welcomed")
4. "Adjusted" underwriting loss = $212M = 3.1% cost of float.
Warren is really quite a guy. Last year, when he was being accused of allowing time to pass him by ("Is Buffett too old fashioned to prosper in high-tech times?"), Warren adhered to fundamental business practice. The following quote indicates that he took a hit to earnings in order to produce long term economic growth for his shareholders:
"In 1999, however, we incurred a $1.4 billion underwriting loss that left us with float cost of 5.8%. One mildly mitigating factor: We enthusiastically welcomed $400 million of the loss because it stems from business that will deliver us exceptional float over the next decade." (1999 Chairman's Letter)
1. Q4 1999 Berkshire reports a 5.8% cost of float.
2. 1999 Annual reports an underwriting loss of $1,394M
3. Warren (Barely) mentions a "Welcomed" $400M loss.
4. "Adjusted" underwriting loss = $994M = 4.1% cost of float.
I should warn that this "adjusted cost of float" is an adjustment that seems reasonable to me, but it is Not something that I have ever heard or seen Warren mention. Therefore, take it for what it's worth.
Conclusions (to this point in my education):
1. "Enthusiastically Welcomed" Loss: Berkshire shareholders are going to have to work a little harder if they want to understand the "obscure" financial accounting involved with insurance contracts. In order to account for "welcomed losses", a new metric may be needed to determine the true cost of float.
2. Necessity for Strong Succession Plan:
Alice Schroeder estimated that for each 1% Investing Performance over the risk free rate (6.2% at the time) you should adjust IV estimate by $9,255.00 per share. Don't forget, this is based on the conservative estimate that Berkshire invests at the risk free rate.
(Side Note: When Alice determines an estimate for Berkshire's IV, she discounts at the risk free rate. This IV estimate is the equivalent to a risk free bond. Therefore, an investor would want to buy when the stock is priced at a discount to Alice's estimate of IV.)
Obviously, we can't expect Warren's successor to allocate capital as successfully as Warren has. Due to the large amounts of money that Berkshire is working with today, Warren even doubts that Warren will be able to achieve similar returns in the future as he has in the past. However, after having reviewed Alice's estimates, you can see why Berkshire shareholder's need to be comfortable with the idea that Warren has been able to identify a successor who has an exceptional mind for business.
The allocation of capital is a key ingredient in Berkshire's success. Warren has invested increasing amounts of capital very wisely. Unfortunately, Warren can't be cloned.
So, who will it be? Lou Simpson? I suppose it depends on the timing. However, I doubt that Lou would want the job today, unless he were needed to take over on an urgent basis. Lou is what, mid-60's? I'm not sure that Lou will want to work at Berkshire as late in his life as Warren is planning on doing with his.
Other than Lou, I don't know who we have - with the necessary skills - to become the executive who is responsible for investments.
On the other hand: "I think this place would continue to have very respectable prospects even if our top 25 managers were all to drop dead at once." - Charlie Munger
Who's going to argue with Charlie?
3. A question concerning Berkshire's maximum exposure might lead to an interesting discussion at the annual meeting. As far as I know, our most significant exposure is with the California Earthquake Authority for ~$600M (after tax). I believe this contract expires in March 2001. I wonder if our "worst case loss" exposure has increased, or decreased since last mentioned.
4. What's It All Mean?: This "Pain Today - Gain Tomorrow" philosophy is very similar to what most people expect from a college education.
Pain Today = 4(+) Years tuition expense. (Up front cost)
Gain Tomorrow = (Potential for) A lifetime of higher annual income.
Unfortunately, I don't think Mr. Market cares. Not yet. He's not going to look hard enough to discover the "adjusted" cost of float. Not yet. Berkshire shareholders are going to have to be patient and wait until Wall Street begins to appropriately value earnings again.
This pain today - gain tomorrow approach toward building long term growth is a philosophy I very much admire. One day, in the not too distant future, we will reap the rewards for the "pain" that we willingly put up with today. So will your kids.... and their kids....
"Someone's sitting in the shade today because someone planted a tree a long time ago." -Warren Buffett
Disclaimer: I am a Novice. Everything I know about insurance (very little) came from the annual reports/meetings. Being that insurance is now such a large part of Berkshire, I feel that I need to learn as much as I can in order to understand what it is that I own. For that reason, I chose to post this message with the hopes that it might stimulate an interesting conversation which I (and hopefully others) could benefit from. Please feel free to tell me what I'm missing, and what I have completely misunderstood. I hope I have made enough mistakes to force my friend EliasFardo out of "retirement"....
With the transformation that Berkshire has gone through over the past few years, I think we're missing the big picture if we focus only on Warren's theory on investing. Today, Berkshire's insurance businesses make up around 70 percent of its revenue.
2. "From time to time we will have such transactions in the future and we will post you of their effects upon our current earnings. When priced properly, this pain-today, gain-tomorrow business is welcomed by us and benefits our shareholders." (Selected highlights Q3 2000)
3. "BHRG is actively negotiating additional retroactive reinsurance contracts, which, if such transactions are consummated, will produce additional premiums in excess of $3 billion. It is not anticipated that these contracts will be finalized until 2001." (10-Q Q3, 2000)