BUFFETT PARTNERSHIP. LTD.
610 KIEWIT PLAZA
OMAHA, NEBRASKA 68131
TELEPHONE 042-4110
January 25, 1967
The First Decade
The Partnership had its tenth anniversary during 1966. The celebration was appropriate -an all-time record (both past and future) was established for our performance margin relative to the Dow. Our advantage was 36 points which resulted from a plus 20.4% for the Partnership and a minus 15.6% for the Dow.
This pleasant but non-repeatable experience was partially due to a lackluster performance by the Dow. Virtually all investment managers outperformed it during the year. The Dow is weighted by the dollar price of the thirty stocks involved. Several of the highest priced components, which thereby carry disproportionate weight (Dupont, General Motors), were particularly poor performers in 1966. This, coupled with the general aversion to conventional blue chips, caused the Dow to suffer relative to general investment experience, particularly during the last quarter.
The following summarizes the year-by-year performance of the Dow, the performance of the Partnership before allocation (one quarter of the excess over 6%) to the general partner, and the results for limited partners:
| Year | Overall Results From Dow (1) | Partnership Results (2) | Limited Partners' Results (3) |
| 1957 | -8.4% | 10.4% | 9.3% |
| 1958 | 38.5% | 40.9% | 32.2% |
| 1959 | 20.0% | 25.9% | 20.9% |
| 1960 | -6.2% | 22.8% | 18.6% |
| 1961 | 22.4% | 45.9% | 35.9% |
| 1962 | -7.6% | 13.9% | 11.9% |
| 1963 | 20.6% | 38.7% | 30.5% |
| 1964 | 18.7% | 27.8% | 22.3% |
| 1965 | 14.2% | 47.2% | 36.9% |
| 1966 | -15.6% | 20.4% | 16.8% |
| Cumulative Results | 141.1% | 1028.7% | 641.5% |
| Annual Compounded Rate | 9.7% | 29.0% | 23.5% |
(1) Based on yearly changes in the value of the Dow plus dividends that would have been received through ownership of the Dow during that year. The table includes all complete years of partnership activity.
(2) For 1957-61 consists of combined results of all predecessor limited partnerships operating throughout the entire year after all expenses, but before distributions to partners or allocations to the general partner.
(3) For 1957-61 computed on the basis of the preceding column of partnership results allowing for allocation to the general partner based upon the present partnership agreement. but before monthly withdrawals by limited partners.
On a cumulative or compounded basis, the results are:
| Year | Overall Results From Dow | Partnership Results | Limited Partners’ Results |
| 1957 | -8.4% | 10.4% | 9.3% |
| 1957 – 58 | 26.9% | 55.6% | 44.5% |
| 1957 – 59 | 52.3% | 95.9% | 74.7% |
| 1957 – 60 | 42.9% | 140.6% | 107.2% |
| 1957 – 61 | 74.9% | 251.0% | 181.6% |
| 1957 – 62 | 61.6% | 299.8% | 215.1% |
| 1957 – 63 | 95.1% | 454.5% | 311.2% |
| 1957 – 64 | 131.3% | 608.7% | 402.9% |
| 1957 – 65 | 164.1% | 943.2% | 588.5% |
| 1957 – 66 | 122.9% | 1156.0% | 704.2% |
| Annual Compounded Rate | 11.4% | 29.8% | 23.9% |
Investment Companies
On the following page is the usual tabulation showing the results of the two largest open-end investment companies (mutual funds) that follow a policy of being, typically, 95-100% invested in common stocks, and the two largest diversified closed-end investment companies.
| Year | Mass. Inv. Trust (1) | Investors Stock (1) | Lehman (2) | Tri-Cont (2) | Dow | Limited Partners |
| 1957 | -11.4% | -12.4% | -11.4% | -2.4% | -8.4% | 9.3% |
| 1958 | 42.7% | 47.5% | 40.8% | 33.2% | 38.5% | 32.2% |
| 1959 | 9.0% | 10.3% | 8.1% | 8.4% | 20.0% | 20.9% |
| 1960 | -1.0% | -0.6% | 2.5% | 2.8% | -6.2% | 18.6% |
| 1961 | 25.6% | 24.9% | 23.6% | 22.5% | 22.4% | 35.9% |
| 1962 | -9.8% | -13.4% | -14.4% | -10.0% | -7.6% | 11.9% |
| 1963 | 20.0% | 16.5% | 23.7% | 18.3% | 20.6% | 30.5% |
| 1964 | 15.9% | 14.3% | 13.6% | 12.6% | 18.7% | 22.3% |
| 1965 | 10.2% | 9.8% | 19.0% | 10.7% | 14.2% | 36.9% |
| 1966 | -7.7% | -10.0% | -2.6% | -6.9% | -15.6% | 16.8% |
| Cumulative Results | 118.1% | 106.3% | 142.8% | 126.9% | 141.1% | 641.5% |
| Annual Compounded Rate | 8.6% | 7.9% | 9.8% | 9.0% | 9.7% | 23.5% |
(1) Computed from changes in asset value plus any distributions to holders of record during year.
(2) From 1966 Moody's Bank & Finance Manual for 1957-1965. Estimated for 1966.
These investment company performance figures have been regularly reported here to show that the now is no patsy as an investment standard. It should again be emphasized that the companies were not selected on the basis of comparability to Buffett Partnership, Ltd. There are important differences including: (1) investment companies operate under both internally and externally imposed restrictions on their investment actions that are not applicable to us; (2) investment companies diversify far more than we do and, in all probability, thereby
have less chance for a really bad performance relative to the now in a single year; and (3) their managers have considerably less incentive for abnormal performance and greater incentive for conventionality.
However, the records above do reveal what well-regarded, highly paid, full-time professional investment managers have been able to accomplish while working with common stocks. These managers have been favorites of American investors (more than 600,000) making free choices among many alternatives in the investment management field. It is probable that their results are typical of the overwhelming majority of professional investment managers.
It is not true, however, that these are the best records achieved in the investment field. A few mutual funds and some private investment operations have compiled records vastly superior to the Dow and, in some cases, substantially superior to Buffett Partnership, Ltd. Their investment techniques are usually very dissimilar to ours and not within my capabilities. However, they are generally managed by very bright, motivated people and it is only fair that I mention the existence of such superior results in this general discussion of the record of professional investment management.
Trends in Our Business
A keen mind working diligently at interpreting the figures on page one could come to a lot of wrong conclusions.
The results of the first ten years have absolutely no chance of being duplicated or even remotely approximated during the next decade. They may well be achieved by some hungry twenty-five year old working with \$105,100 initial partnership capital and operating during a ten year business and market environment which is frequently conducive to successful implementation of his investment philosophy.
They will not be achieved by a better fed thirty-six year old working with our \$54,065,345 current partnership capital who presently finds perhaps one-fifth to one-tenth as many really good ideas as previously to implement his investment philosophy.
Buffett Associates. Ltd. (predecessor to Buffett Partnership. Ltd.) was founded on the west banks of the Missouri. May 5. 1956 by a hardy little band consisting of four family members, three close friends and \$105,100. (I tried to find some brilliant flash of insight regarding our future or present conditions from my first page and a half annual letter of January, 1957 to insert as a quote here. However, someone evidently doctored my file copy so as to remove the perceptive remarks I must have made.)
At that time, and for some years subsequently, there were substantial numbers of securities selling at well below the "value to a private owner" criterion we utilized for selection of general market investments. We also experienced a flow of “workout” opportunities where the percentages were very much to our liking. The problem was always which, not what. Accordingly, we were able to own fifteen to twenty-five issues and be enthusiastic about the probabilities inherent in all holdings.
In the last few years this situation has changed dramatically. We now find very few securities that are understandable to me, available in decent size, and which offer the expectation of investment performance meeting our yardstick of ten percentage points per annum superior to the Dow. In the last three years we have come up with only two or three new ideas a year that have had such an expectancy of superior performance. Fortunately, in some cases, we have made the most of them. However, in earlier years, a lesser effort produced literally dozens of comparable opportunities. It is difficult to be objective about the causes for such diminution of one's own productivity. Three factors that seem apparent are: (1) a somewhat changed market environment; (2) our increased size; and (3) substantially more competition.
It is obvious that a business based upon only a trickle of fine ideas has poorer prospects than one based upon a steady flow of such ideas. To date the trickle has provided as much financial nourishment as the flow. This is true because there is only so much one can digest (million dollar ideas are of no great benefit to thousand dollar bank accounts - this was impressed on me in my early days) and because a limited number of ideas causes one to utilize those available more intensively. The latter factor has definitely been operative with us in recent years. However, a trickle has considerably more chance of drying up completely than a flow.
These conditions will not cause me to attempt investment decisions outside my sphere of understanding (I don't go for the "If you can't lick 'em, join 'em” philosophy - my own leaning is toward "If you can't join ‘em, lick 'em”). We will not go into businesses where technology which is away over my head is crucial to the investment decision. I know about as much about semi-conductors or integrated circuits as I do of the mating habits of the chrzaszcz. (That's a Polish May bug, students - if you have trouble pronouncing it, rhyme it with thrzaszcz.)
Furthermore, we will not follow the frequently prevalent approach of investing in securities where an attempt to anticipate market action overrides business valuations. Such so-called "fashion" investing has frequently produced very substantial and quick profits in recent years (and currently as I write this in January). It represents an investment technique whose soundness I can neither affirm nor deny. It does not completely satisfy my intellect (or perhaps my prejudices), and most definitely does not fit my temperament. I will not invest my own money based upon such an approach hence, I will most certainly not do so with your money.
Finally, we will not seek out activity in investment operations, even if offering splendid profit expectations, where major human problems appear to have a substantial chance of developing.
What I do promise you, as partners, is that I will work hard to maintain the trickle of ideas and try to get the most out of it that is possible – but if it should dry up completely, you will be informed honestly and promptly so that we may all take alternative action.
Analysis of 1966 Results
All four main categories of our investment operation worked out well in 1966. Specifically, we had a total overall gain of \$8,906,701 derived as follows:
| Category | Average Investment | Overall Gain |
| Controls | $17,259,342 | $1,566,302 |
| Generals – Private Owner | $1,359,340 | $1,004,362 |
| Generals – Relatively | $21,847,045 | $5,124,254 |
| Undervalued | ||
| Workouts | $7,666,314 | $1,714,181 |
| Miscellaneous, including US Treasury Bills | $1,332,609 | $(18,422) |
| Total Income | $9,390,677 | |
| Less: General Expense | $483,976 | |
| Overall Gain | $8,906,701 |
A few caveats are necessary before we get on with the main discussion:
-
An explanation of the various categories listed above was made in the January 18, 1965 letter. If your memory needs refreshing and your favorite newsstand does not have the pocketbook edition. we'll be glad to give you a copy.
-
The classifications are not iron-clad. Nothing is changed retroactively but the initial decision as to category is sometimes arbitrary.
- Percentage returns calculated on the average investment base by category would be understated relative to partnership percentage returns which are calculated on a beginning investment base. In the above figures, a security purchased by us at 100 on January 1 which appreciated at an even rate to 150 on December 31 would have an average investment of 125 producing a 40% result contrasted to a 50% result by the customary approach. In other words, the above figures use a monthly average of market values in calculating the average investment.
- All results are based on a 100% ownership, non-leverage, basis. Interest and other general expenses are deducted from total performance and not segregated by category. Expenses directly related to specific investment operations, such as dividends paid on short stock, are deducted by category. When securities are borrowed directly and sold short, the net investment (longs minus shorts) is shown for the applicable average investment category.
The above table has only limited use. The results applicable to each category are dominated by one or two investments. They do not represent a collection of great quantities of stable data (mortality rates of all American males or something of the sort) from which conclusions can be drawn and projections made. Instead, they represent infrequent, non-homogeneous phenomena leading to very tentative suggestions regarding various courses of action and are so used by us. - Finally, these calculations are not made with the same loving care we apply to counting the money and are subject to possible clerical or mathematical error since they are not entirely self-checking.
Controls
There were three main sources of gain during 1966 in respect to controlled companies. These arose through: (1) retained business earnings applicable to our holdings in 1966; (2) open market purchases of additional stock below our controlling interest valuation and; (3) unrealized appreciation in marketable securities held by the controlled companies. The total of all positive items came to \$2,600,838 in 1966.
However, due to factors mentioned in my November 1, 1966 letter, specific industry conditions, and other relevant valuation items, this gain was reduced by \$1,034,780 in arriving at our fair valuation applicable to controlling interests as of December 31, 1966. Thus the overall gain in the control category was reduced to \$1,566,058 for the year.
We were undoubtedly fortunate that we had a relatively high percentage of net assets invested in businesses and not stocks during 1966. The same money in general market holdings would probably have produced a loss, perhaps substantial, during the year. This was not planned and if the stock market had advanced substantially during the year, this category would have been an important drag on overall performance. The same situation will prevail during 1967.
Generals -Private Owner
Our performance here falls in the "twenty-one dollars a day, once a month" category. In the middle of 1965 we started purchasing a very attractive widely held security which was selling far below its value to a private owner. Our hope was that over a two or three year period we could get \$10 million or more invested at the favorable prices prevailing. The various businesses that the company operated were understandable and we could check out competitive strengths and weaknesses thoroughly with competitors, distributors, customers, suppliers, ex-employees, etc. Market conditions peculiar to the stock gave us hope that, with patience, we could buy substantial quantities of the stock without disturbing the price.
At yearend 1965 we had invested \$1,956,980 and the market value of our holding was \$2,358,412 so that \$401,432 was contributed to performance luring 1965. We would have preferred, of course, to have seen the market below cost since our interest was in additional buying, not in selling. This would have dampened Buffett Partnerships Ltd.’s 1965 performance and perhaps reduced the euphoria experienced by limited partners (psychically, the net result to all partners would have been a standoff since the general partner would have been floating) but would have enhanced long term performance. The fact that the stock had risen somewhat above our cost had already slowed down our buying program and thereby reduced ultimate profit.
An even more dramatic example of the conflict between short term performance and the maximization of long term results occurred in 1966. Another party, previously completely unknown to me, issued a tender offer which foreclosed opportunities for future advantageous buying. I made the decision that the wisest course (it may not have been) for us to follow was to dispose of our holdings and we thus realized a total profit of \$1,269,181 in February, of which \$867,749 was applicable to 1966.
While any gains looked particularly good in the market environment that intimately developed in 1966, you can be sure I don't delight in going round making molehills out of mountains. The molehill, of course, was reflected in 1966 results. However, we would have been much better off from a long range standpoint if 1966 results had been five percentage points worse and we were continuing to buy substantial quantities of the stock at the depressed prices that might have been expected to prevail in this year's market environment.
Good ideas were a dime a dozen, such a premature ending would not be unpleasant. There is something to be said, of course, for a business operation where some of the failures produce moderate profits. However, you can see how hard it is to develop replacement ideas by examining our average investment in the Private Owner category - we came up with nothing during the remainder of the year despite lower stock prices, which should have been conducive to finding such opportunities.
Generals - Relatively Undervalued
Our relative performance in this category was the best we have ever had - due to one holding which was our largest investment at yearend 1965 and also yearend 1966. This investment has substantially out-performed the general market for us during each year (1964, 1965, 1966) that we have held it. While any single year's performance can be quite erratic, we think the probabilities are highly favorable for superior future performance over a three or four year period. The attractiveness and relative certainty of this particular security are what caused me to introduce Ground Rule 7 in November, 1965 to allow individual holdings of up to 40% of our net assets. We spend considerable effort continuously evaluating every facet of the company and constantly testing our hypothesis that this security is superior to alternative investment choices. Such constant evaluation and comparison at shifting prices is absolutely essential to our investment operation.
It would be much more pleasant (and indicate a more favorable future) to report that our results in the Generals -Relatively Undervalued category represented fifteen securities in ten industries, practically all of which outperformed the market. We simply don't have that many good ideas. As mentioned above, new ideas are continually measured against present ideas and we will not make shifts if the effect is to downgrade expectable performance. This policy has resulted in limited activity in recent years when we have felt so strongly about the relative merits of our largest holding. Such a condition has meant that realized gains have been a much smaller portion of total performance than in earlier years when the flow of good ideas was more substantial.
The sort of concentration we have in this category is bound to produce wide swings in short term performance – some, most certainly, unpleasant. There have already been some of these applicable to shorter time spans than I use in reporting to partners. This is one reason I think frequent reporting to be foolish and potentially misleading in a long term oriented business such as ours.
Personally, within the limits expressed in last year's letter on diversification, I am willing to trade the pains (forget about the pleasures) of substantial short term variance in exchange for maximization of long term performance. However, I am not willing to incur risk of substantial permanent capital loss in seeking to better long term performance. To be perfectly clear - under our policy of concentration of holdings, partners should be completely prepared for periods of substantial underperformance (far more likely in sharply rising markets) to offset the occasional over performance such as we have experienced in 1965 and 1966, and as a price we pay for hoped-for good long term performance.
All this talk about the long pull has caused one partner to observe that “even five minutes is a long time if one's head is being held under water." This is the reason, of course, that we use borrowed money very sparingly in our operation. Average bank borrowings during 1966 were well under 10% of average net worth.
One final word about the Generals - Relatively Undervalued category. In this section we also had an experience which helped results in 1966 but hurt our long term prospects. We had just one really important new idea in this category in 1966. Our purchasing started in late spring but had only come to about \$1.6 million (it could be bought steadily but at only a moderate pace) when outside conditions drove the stock price up to a point where it was not relatively attractive. Though our overall gain was \$728,141 on an average holding period of six and a half months in 1966, it would have been much more desirable had the stock done nothing for a long period of time while we accumulated a really substantial position.
Workouts
In last year's letter I forecast reduced importance for workouts. While they were not of the importance of some past years. I was pleasantly surprised by our experience in 1966 during which we kept an average of \$7,666,314 employed in this category. Furthermore, we tend to ascribe borrowings to the workout section so that our net equity capital employed was really something under this figure and our return was somewhat better than the 22.4% indicated on page six. Here, too, we ran into substantial variation. At June 30, our overall profit on this category was \$16,112 on an average investment of \$7,870,151 so that we really had a case of an extraordinarily good second half offsetting a poor first half.
In past years, sometimes as much as 30-40% of our net worth has been invested in workouts, but it is highly unlikely that this condition will prevail in the future. Nevertheless, they may continue to produce some decent returns on the moderate amount of capital employed.
Miscellaneous
Operationally, we continue to function well above rated capacity with Bill, John, Elizabeth and Donna all contributing excellent performances. At Buffett Partnership. Ltd. we have never had to divert investment effort to offset organizational shortcomings and this has been an important ingredient in the performance over the years.
Peat, Marwick, Mitchell & Co., aided for the second year by their computer, turned in the usual speedy, efficient and comprehensive job.
We all continue to maintain more than an academic interest in the Partnership. The employees and I, our spouses and children, have a total of over \$10 million invested at January 1, 1967. In the case of my family, our Buffett Partnership, Ltd. investment represents well over 90% of our net worth.
Within the coming two weeks you will receive:
- A tax letter giving you all BPL information needed for your 1966 federal income tax return. This letter is the only item that counts for tax purposes.
- An audit from Peat, Marwick, Mitchell & Co. for 1966, setting forth the operations and financial position of BPL, as well as your own capital account.
- A letter signed by me setting forth the status of your BPL interest on January 1, 1967. This is identical with the figures developed in the audit.
Let me know if anything in this letter or that occurs during the year needs clarifying. My next letter will be about July 15 summarizing the first half of this year.
Cordially,
Warren E. Buffett
WEB eh
BUFFETT PARTNERSHIP. LTD.
610 KIEWIT PLAZA
OMAHA, NEBRASKA 68131
电话 042-4110
1967年1月25日
第一个十年
合伙企业于1966年迎来了十周年。庆祝恰如其分——我们相对于道指的业绩优势创下了空前（未来也将绝后）的纪录。我们的领先优势是36个百分点，来自合伙企业+20.4%的收益率和道指-15.6%的收益率。
这一令人愉快但不可复制的经历，部分归因于道指表现平平。实际上，当年所有投资管理人的业绩都跑赢了道指。道指由其包含的三十只股票的美元价格加权计算。几只价格最高的成分股（杜邦、通用汽车）在1966年表现尤其糟糕，从而带偏了指数权重。再加上市场对传统蓝筹股的普遍回避，导致道指相对整体投资体验表现逊色，尤其是在第四季度。
以下汇总了道指逐年业绩、合伙企业分配前业绩（超过6%部分的四分之一归普通合伙人），以及有限合伙人的结果：
| 年份 | 道指整体结果 (1) | 合伙企业结果 (2) | 有限合伙人结果 (3) |
| 1957 | -8.4% | 10.4% | 9.3% |
| 1958 | 38.5% | 40.9% | 32.2% |
| 1959 | 20.0% | 25.9% | 20.9% |
| 1960 | -6.2% | 22.8% | 18.6% |
| 1961 | 22.4% | 45.9% | 35.9% |
| 1962 | -7.6% | 13.9% | 11.9% |
| 1963 | 20.6% | 38.7% | 30.5% |
| 1964 | 18.7% | 27.8% | 22.3% |
| 1965 | 14.2% | 47.2% | 36.9% |
| 1966 | -15.6% | 20.4% | 16.8% |
| 累计结果 | 141.1% | 1028.7% | 641.5% |
| 年复利收益率 | 9.7% | 29.0% | 23.5% |
(1) 基于道指每年变动值加上当年持有道指可获得的股息。该表包含合伙企业所有完整会计年度。
(2) 1957-1961年包含所有前身有限合伙企业在整个年度内的合并业绩，已扣除所有费用，但未扣除分配给合伙人的款项或分配给普通合伙人的部分。
(3) 1957-1961年基于前一列合伙企业结果，按现行合伙协议对普通合伙人进行分配，但未扣除有限合伙人的月度提款。
按累计或复利计算，结果如下：
| 年份 | 道指整体结果 | 合伙企业结果 | 有限合伙人结果 |
| 1957 | -8.4% | 10.4% | 9.3% |
| 1957 – 58 | 26.9% | 55.6% | 44.5% |
| 1957 – 59 | 52.3% | 95.9% | 74.7% |
| 1957 – 60 | 42.9% | 140.6% | 107.2% |
| 1957 – 61 | 74.9% | 251.0% | 181.6% |
| 1957 – 62 | 61.6% | 299.8% | 215.1% |
| 1957 – 63 | 95.1% | 454.5% | 311.2% |
| 1957 – 64 | 131.3% | 608.7% | 402.9% |
| 1957 – 65 | 164.1% | 943.2% | 588.5% |
| 1957 – 66 | 122.9% | 1156.0% | 704.2% |
| 年复利收益率 | 11.4% | 29.8% | 23.9% |
投资公司
下一页是惯常的表格，列出两只规模最大的开放式投资公司（共同基金）以及两只规模最大的多元化封闭式投资公司的业绩，这些基金通常遵循95-100%投资于普通股的政策。
| 年份 | Mass. Inv. Trust (1) | Investors Stock (1) | Lehman (2) | Tri-Cont (2) | 道指 | 有限合伙人 |
| 1957 | -11.4% | -12.4% | -11.4% | -2.4% | -8.4% | 9.3% |
| 1958 | 42.7% | 47.5% | 40.8% | 33.2% | 38.5% | 32.2% |
| 1959 | 9.0% | 10.3% | 8.1% | 8.4% | 20.0% | 20.9% |
| 1960 | -1.0% | -0.6% | 2.5% | 2.8% | -6.2% | 18.6% |
| 1961 | 25.6% | 24.9% | 23.6% | 22.5% | 22.4% | 35.9% |
| 1962 | -9.8% | -13.4% | -14.4% | -10.0% | -7.6% | 11.9% |
| 1963 | 20.0% | 16.5% | 23.7% | 18.3% | 20.6% | 30.5% |
| 1964 | 15.9% | 14.3% | 13.6% | 12.6% | 18.7% | 22.3% |
| 1965 | 10.2% | 9.8% | 19.0% | 10.7% | 14.2% | 36.9% |
| 1966 | -7.7% | -10.0% | -2.6% | -6.9% | -15.6% | 16.8% |
| 累计结果 | 118.1% | 106.3% | 142.8% | 126.9% | 141.1% | 641.5% |
| 年复利收益率 | 8.6% | 7.9% | 9.8% | 9.0% | 9.7% | 23.5% |
(1) 根据资产价值变化加上当年向登记持有人分配的任何款项计算。
(2) 1957-1965年数据来自1966年穆迪银行与金融手册。1966年为估计值。
这些投资公司的业绩数据定期在此报告，以说明道指作为投资基准并非易与之辈。需要再次强调，这些公司并非基于与Buffett Partnership, Ltd.的可比性而选择。存在重要差异，包括：(1) 投资公司在投资行为上受内外部限制，这些限制不适用于我们；(2) 投资公司的多元化程度远高于我们，因此它们在单一年度相对道指表现糟糕的可能性很可能更低；(3) 其管理人追求非正常业绩的动机要小得多，而追求从众性的动机更大。
然而，上述记录确实揭示了备受尊敬、薪酬优厚、全职专业的投资管理人在普通股投资中能取得怎样的成果。这些管理人是美国投资者（超过60万人）在投资管理领域众多选择中自由选出的宠儿。他们的业绩很可能代表了绝大多数专业投资管理人的典型水平。
但这并不意味着这些就是投资领域的最佳记录。少数共同基金和一些私人投资操作取得了远优于道指的记录，在某些情况下甚至大幅超越Buffett Partnership, Ltd.。他们的投资技术通常与我们的截然不同，也不在我能力范围之内。不过，这些管理人通常非常聪明、有动力，在对专业投资管理业绩的总体讨论中，我理应提及这类优异结果的存在。
我们业务的趋势
一个敏锐的头脑若勤勉地解读第一页的数据，可能会得出许多错误结论。
头十年的结果绝对没有可能在下一个十年被复制，甚至无法接近。这些结果或许可以由某个饥肠辘辘的25岁年轻人实现，他以105,100美元初始合伙资本起步，并在一个十年期的商业和市场环境中运营，该环境常常有利于他成功实施自己的投资哲学。
但这些结果不会由我这个营养更良好、现年36岁、管理着54,065,345美元合伙资本的人实现——因为我目前发现的真正好主意数量，可能只有以前的五分之一到十分之一，用来落实投资哲学。
Buffett Associates, Ltd.（Buffett Partnership, Ltd.的前身）于1956年5月5日在密苏里河西岸成立，创始成员是一支由四名家庭成员、三名密友和105,100美元组成的坚韧小团队。（我曾想从我1957年1月那份一页半的年度信里找些关于我们未来或现状的闪亮洞察，引在这里作为语录。然而，显然有人篡改了我的存档副本，把那些我肯定写过的精辟言论删掉了。）
那时及随后几年，有大量证券以远低于我们用于选择低估类投资的市场股票的"私人持有者价值"标准出售。我们还遇到了大量"套利类"机会，其概率非常合我们心意。问题总是选哪个，而不是选什么。因此，我们能持有15到25个标的，并对所有持仓的内在概率充满热情。
最近几年，这种情况发生了巨大变化。我们现在发现，很少有证券能让我理解、规模合理，并且预期业绩能达到我们每年领先道指十个百分点这个标尺。过去三年，我们每年只找到两三个有如此超额预期的新主意。幸运的是，在某些情况下，我们充分利用了它们。然而，在早年，付出更少的努力就能产生十多个类似的机会。对于自身生产力下降的原因，很难做到客观。三个显而易见的因素是：(1) 市场环境有所改变；(2) 我们的规模增大；(3) 竞争显著加剧。
显然，一个仅靠少量好主意维持的业务，前景不如一个有好主意持续涌现的业务。到目前为止，那涓涓细流提供的财务滋养与当初的洪流一样多。这是因为一个人能消化的东西有限（百万美元的主意对于千美元银行账户没有多大好处——这在我早年就深有体会），也因为主意有限会促使你更深入地利用已有的那些。近年来，后一个因素确实在我们身上发挥了作用。然而，涓涓细流完全干涸的可能性远大于洪流。
这些情况不会导致我试图在自己理解范围之外做出投资决策（我不信奉"打不过就加入"的理念——我更倾向于"加入不了就打败它"）。我们不会投资那些技术远在我理解能力之外、却对投资决策至关重要的企业。我对半导体或集成电路的了解，并不比我对chrzaszcz（波兰语中的五月金龟子——同学们，如果你发音困难，就把它和thrzaszcz押韵）的交配习性了解更多。
此外，我们不会追随那种流行的投资方法，即试图预测市场走势而非进行企业估值。这种所谓的"时尚"投资近年来（以及我写这封信的1月份当下）频繁产生了非常丰厚且快速的利润。这是一种我既无法肯定也无法否认其合理性的投资技术。它不完全符合我的智力偏好（或许是我的偏见），而且绝对不符合我的性格。我不会用这种思路投资自己的钱，所以当然也不会用你们的钱。
最后，我们不会在投资操作中寻求活跃，即使它提供丰厚的利润预期，但似乎很有可能引发重大的人际问题。
作为合伙人，我向你们承诺的是：我会努力工作以维持这涓涓细流，并尽力从中获得最大收益——但如果它完全干涸，我会诚实、迅速地通知你们，以便我们都能采取替代行动。
1966年业绩分析
1966年，我们投资操作的四个主要类别都表现良好。具体来说，我们的总收益为8,906,701美元，来源如下：
| 类别 | 平均投资额 | 总收益 |
| 控制类投资 | $17,259,342 | $1,566,302 |
| 低估类投资——私人持有者估值 | $1,359,340 | $1,004,362 |
| 低估类投资——相对低估 | $21,847,045 | $5,124,254 |
| 套利类投资 | $7,666,314 | $1,714,181 |
| 杂项（包括美国国库券） | $1,332,609 | $(18,422) |
| 总收入 | $9,390,677 | |
| 减：一般费用 | $483,976 | |
| 总收益 | $8,906,701 |
在进入主要讨论之前，需要几点说明：
- 上述各类别的解释已在1965年1月18日的信中说明。如果您需要重温，而您最喜欢的报刊亭没有口袋版，我们乐意提供一份副本。
- 分类并非铁板钉钉。没有事后的修正，但最初的类别判断有时是主观的。
- 按类别平均投资基数计算的百分比收益率，相对于按初始投资基数计算的合伙企业百分比收益率，会被低估。在上表中，如果一只证券于1月1日以100买入，到12月31日均匀升值至150，则平均投资为125，产生40%的收益率，而按常规方法为50%。换句话说，上述数字在计算平均投资时使用了市场价值的月度平均值。
- 所有结果均基于100%持股、无杠杆的基础。利息和其他一般费用从总业绩中扣除，不按类别划分。与特定投资操作直接相关的费用（如做空股票支付的股息）按类别扣除。当直接借入证券并卖空时，净投资（多头减空头）显示在相应的平均投资类别中。
- 上表用途有限。每个类别的结果主要由一两个投资主导。它们不代表大量稳定数据（如所有美国男性的死亡率之类）的集合，从中可以得出结论并进行预测。相反，它们代表不频繁、非同质化的现象，只能对我们各种行动方案提供非常初步的提示，我们正是这样使用的。
- 最后，这些计算并非像我们数钱那样精心细致，可能包含文书或数学错误，因为它们并非完全自检。
控制类投资
1966年控制类公司的收益有三个主要来源： (1) 1966年我们持股对应的留存企业利润；(2) 在公开市场以低于我们控制权估值价格增持股票；(3) 控制类公司所持可交易证券的未实现增值。1966年所有正项合计为2,600,838美元。
然而，由于我在1966年11月1日信中提到的因素、特定行业状况以及其他相关估值项目，在确定我们截至1966年12月31日适用于控制权益的公允价值时，这一收益减少了1,034,780美元。因此，该年度控制类别的总收益降至1,566,058美元。
毫无疑问，我们很幸运在1966年有相对较高比例的净资产投资于企业而非股票。同样数额的资金若投资于一般市场持仓，当年很可能产生损失，甚至可能是重大损失。这并非事先计划，如果当年股市大幅上涨，这个类别将对整体业绩构成重要拖累。1967年情况类似。
低估类投资——私人持有者估值
我们在这方面的表现属于"一个月一次，每次21美元"的类别。1965年中，我们开始买入一只非常有吸引力、广泛持有的证券，其价格远低于私人持有者价值。我们希望在两三年内，能以当时有利的价格投入1000万美元或更多。该公司运营的各项业务可以理解，我们可以通过竞争对手、分销商、客户、供应商、前雇员等渠道彻底核实其竞争优势和劣势。该股特有的市场状况让我们有希望，只要耐心，就能在不影响价格的情况下大量购入。
1965年底，我们已投资1,956,980美元，持仓市值为2,358,412美元，因此1965年有401,432美元贡献给业绩。当然，我们更希望看到市价低于成本，因为我们感兴趣的是继续买入而非卖出。这可能会拖累Buffett Partnership Ltd. 1965年的业绩，并可能减少有限合伙人体验到的 euphoria（心理上，对所有合伙人来说净结果可能是扯平，因为普通合伙人会浮起来），但会提升长期业绩。该股价格已较成本有所上涨，这已经减缓了我们的买入计划，从而减少了最终利润。
另一个更戏剧性的例子，展示了短期业绩与长期结果最大化之间的冲突，发生在1966年。另一个我此前完全不认识的参与方发出了要约收购，从而扼杀了未来有利买入的机会。我判断最明智的做法（可能未必是）是清仓我们的持仓，于是我们在2月实现了总计1,269,181美元的利润，其中867,749美元归属于1966年。
虽然任何收益在1966年最终形成的市场环境中都显得特别好看，但你可以肯定，我一点不喜欢把大山变成小土丘。小土丘当然反映在1966年的结果里。然而，从长期角度看，如果1966年业绩差五个百分点，而我们继续以今年市场环境中可能出现的低价大量买入该股，情况会好得多。
好主意一毛钱一打时，这种过早的结局并不令人不快。当然，在某些业务操作中，部分失败能产生适度利润，这也有其好处。但你可以从我们私人持有者类别平均投资额中看出，要找到替代主意有多难——尽管当年股价更低，理应更有利于发现此类机会，但我们在余下时间里一无所获。
低估类投资——相对低估
我们在这个类别的相对业绩是有史以来最好的——归功于一个持仓，它同时是我们1965年底和1966年底的最大投资。在我们持有它的每一年（1964、1965、1966），这只投资都大幅跑赢了大盘。虽然任何单一年度的业绩可能相当不稳定，但我们认为，在三四年期内，其未来表现的优越概率非常有利。正是考虑到这只特定证券的吸引力和相对确定性，我才在1965年11月引入了第七条基本规则，允许单一持仓最高可达净资产的40%。我们持续投入大量精力评估该公司的方方面面，并不断检验我们的假设，即这只证券优于其他投资选择。这种在价格变动中持续的评估和比较，对我们的投资操作至关重要。
如果报告说我们在低估类投资——相对低估类别中的业绩来自十五只证券、十个行业，并且几乎所有都跑赢了市场，那会愉快得多（并预示更美好的未来）。但我们就是没有那么多好主意。如上所述，新主意不断与现有主意进行比较，如果换仓会降低预期业绩，我们就不会变动。这一政策导致近年来活动有限，因为我们对最大持仓的相对优势如此强烈。这种情况意味着，已实现收益占总业绩的比例远小于早年好主意源源不断的时候。
我们在这个类别中的这种集中度，必然导致短期业绩的大幅波动——有些波动肯定不愉快。在我向合伙人报告的时间跨度内，已经发生过一些这样的波动。这就是为什么我认为在像我们这样以长期为导向的业务中，频繁报告是愚蠢且可能产生误导的。
就个人而言，在去年信中关于分散化的限制范围内，我愿意用短期业绩的显著波动（忘掉愉悦部分）来换取长期业绩的最大化。然而，我不愿意为了提升长期业绩而冒重大永久性资本损失的风险。清楚地说——在我们集中持仓的政策下，合伙人应完全准备好经历显著跑输的时期（在急剧上涨的市场中可能性更大），以抵消偶尔的超常表现，比如我们在1965年和1966年经历的那样，这是我们为期望的良好长期业绩所付出的代价。
所有这些关于长期拉锯的讨论，让一位合伙人评论道："如果头被按在水下，五分钟也是很长的时间。" 当然，这就是我们在操作中极少使用借入资金的原因。1966年的平均银行借款远低于平均净值的10%。
关于低估类投资——相对低估类别，最后再说一句。在这个类别中，我们也经历了一次帮助1966年结果但损害长期前景的事件。1966年我们在该类别只有一个真正重要的新主意。我们的买入始于春末，但只投入了约160万美元（可以稳步买入但速度很慢），此时外部条件将股价推高到不再相对有吸引力的水平。虽然我们在1966年六个月的持有期内实现了728,141美元的总收益，但如果这只股票长期停滞不前，同时我们积累起真正可观的仓位，那会理想得多。
套利类投资
在去年的信中，我预测套利类的重要性会降低。虽然它们没有过去几年那么重要，但1966年的经历让我惊喜，当年我们在这类中平均投入了7,666,314美元。此外，我们倾向于将借款归入套利类，因此我们使用的净权益资本实际上低于这个数字，回报率也比第6页显示的22.4%略高。在这里，我们也遭遇了大幅波动。截至6月30日，我们在这类中的总利润为16,112美元，平均投资为7,870,151美元，因此下半年异常出色，抵消了上半年的糟糕表现。
过去几年，有时我们净资产的30-40%投资于套利类，但未来这种情况不太可能重现。尽管如此，在适度资本投入下，它们可能继续产生一些不错的回报。
杂项
运营方面，我们继续超额定产能运转，Bill、John、Elizabeth和Donna都表现出色。在Buffett Partnership, Ltd.，我们从未需要将投资精力用于弥补组织缺陷，这多年来一直是业绩的重要成分。
Peat, Marwick, Mitchell & Co. 在计算机的辅助下，第二年再次完成了快速、高效和全面的工作。
我们所有人继续对合伙企业保持着超越学术层面的兴趣。员工和我本人、我们的配偶及子女，在1967年1月1日总共投资了超过1000万美元。就我家而言，我们在Buffett Partnership, Ltd.的投资占我们净资产的90%以上。
未来两周内，您将收到：
- 一封税务信，提供您1966年联邦所得税申报所需的所有BPL信息。这封信是唯一用于税务目的的文件。
- 一份来自Peat, Marwick, Mitchell & Co.的1966年审计报告，说明BPL的经营和财务状况，以及您个人的资本账户。
- 一封由我签署的信函，说明您1967年1月1日BPL权益的状况。这与审计结果中的数字一致。
如果这封信或年内发生的任何事情需要澄清，请随时告知。我的下一封信将在7月15日左右发出，总结今年上半年情况。
此致，
Warren E. Buffett
WEB eh