咖啡罐方法：为何做得更少能让你获得更多
Perspectives
Perspectives
2011 年 3 月，Legg Mason Capital Management，“咖啡罐投资法”
March 2011 Legg Mason Capital Management The coffee can approach
本信息为一般性资料，不代表任何投资产品的立场。
This information is general in nature and is not on behalf of any investment product.
所有投资均涉及风险，包括可能损失本金。
All investments involve risk, including possible loss of principal.
投资产品不受联邦存款保险公司（FDIC）承保 • 无银行担保 • 可能损失价值
INVESTMENT PRODUCTSNOT FDIC INSURED • NO BANK GUARANTEE • MAY LOSE VALUE
Batterymarch • Brandywine Global • ClearBridge Advisors • Legg Mason Capital Management Legg Mason Global Equities Group • Permal • Private Capital Management • Royce & Associates Western Asset Management
Batterymarch • Brandywine Global • ClearBridge Advisors • Legg Mason Capital Management Legg Mason Global Equities Group • Permal • Private Capital Management • Royce & Associates Western Asset Management
March 25, 2011
March 25, 2011
咖啡罐投资法：为什么做得更少，反而收获更多
The Coffee Can Approach Why Doing Less Can Leave You with More
不活动在我们看来是明智之举。
Inactivity strikes us as intelligent behavior.
沃伦·巴菲特，《伯克希尔·哈撒韦年报》，第 1 页
Warren Buffett Berkshire Hathaway Annual Report 1
什么也不做，是极其辛苦的工作。
It is awfully hard work doing nothing.
奥斯卡·王尔德，《不可儿戏：一部供正经人消遣的轻喜剧》，第 2 页
Oscar Wilde The Importance of Being Earnest: A Trivial Comedy for Serious People 2
来源：www.dreamstime.com。
Source: www.dreamstime.com.
• 投资者常常出于善意调整投资组合，但这些调整并没有创造价值。
• Investors often make changes to their portfolios—with the best of intentions—that do not add value.
• 这些错误包括将投资组合从一个资产类别重新配置到另一个资产类别，以及在同一个资产类别内从一个管理人切换到另一个管理人。
• These mistakes include reallocation of a portfolio from one asset class to another as well as switching from one manager to another within an asset class.
• 通过模拟分析表明，如果投资者将评估管理人的行业标准窗口期从 3 年延长，结果会更好。
• Analysis through simulation shows that investors would be better off extending the industry standard three-year window for manager assessment.
Brewing a Profit
Brewing a Profit
罗伯特·柯比是 Capital Guardian Trust 的创始人之一，他讲述了一个故事，主角是他从 1950 年代中期开始担任投资顾问约 10 年的一对夫妻。由于客户的首要目标是保本，柯比遵循公司的指导原则，买卖投资品，以确保投资组合既合理又定位良好。柯比主要与丈夫一起打理以妻子名义设立的投资组合。
Robert Kirby, one of the founders of Capital Guardian Trust, told a story of a couple he worked with as an investment counselor for about a decade through the mid-1950s. Since wealth preservation was the primary objective of the client, Kirby followed his firm’s guidelines and bought and sold investments to make sure that the portfolio was sensible and well-positioned. Kirby worked primarily with the husband on a portfolio in the wife’s name.
丈夫突然去世后，妻子打电话说，她继承了丈夫的遗产，并打算把他的投资组合并入她自己的。柯比查看了丈夫的投资组合，既觉得好笑，又感到震惊。好笑的是，他发现自己公司对妻子的买入建议，被丈夫悄悄地“搭了便车”。丈夫对每只股票各买了大约 5000 美元，然后把凭证扔进银行保险箱，此后便对投资不闻不问。柯比称之为“咖啡罐投资组合”，因为这让他想起从前人们常把贵重物品放进咖啡罐，然后藏在床垫下的日子。由于没有产生任何交易或管理费用，这个罐子的价值完全取决于主人当初放了什么进去。
After the husband died suddenly, the wife called to say that she had inherited his estate and was adding his investment portfolio to hers. Kirby reviewed the man’s portfolio and was amused and shocked. He was amused to see that the man had piggybacked the firm’s buy recommendations to his wife. The man purchased about $5,000 of each stock, tossed the certificates into a safe deposit box, and simply ignored the investments. Kirby called it the “coffee can portfolio” because it reminded him of a time when it was common for someone to place his valuables in a coffee can and stick it under his mattress. Since it incurred no transaction or administrative costs, the can’s value hinged solely on what the owner placed in it.
柯比震惊的是，丈夫投资组合的价值，竟然远远超过了妻子的。当然，这是一个奇怪的组合：一些持仓跌到了 2000 美元，几个大仓位超过了 10 万美元，还有一只股票的价值超过了 80 万美元。
Kirby was shocked when he saw the value of the man’s portfolio, which greatly exceeded that of his wife’s. It was an odd mix, to be sure. There were a number of holdings that had sunk to $2,000, several large positions that exceeded $100,000, and one stock with a value in excess of $800,000.
那个庞大的仓位，源于对一家名为 Haloid Photographic 公司的小额投资，这家公司后来改名为施乐。
That jumbo position was the result of a small commitment to a company called Haloid Photographic, which later changed its name to Xerox.
柯比从这个案例中学到的教训，并非投资者应该买入股票，指望找到下一个 Haloid（或谷歌、苹果）。相反，他发现，一个仅仅按照公司一半建议买入、且交易成本可忽略不计的投资组合，其表现轻松超过了柯比全心打理的另一个投资组合。买入被低估的股票，然后什么都不做，比试图在市场的起伏波动中“冲浪”效果更好。沃伦·巴菲特表达了类似的观点，他说：“近乎懒惰的昏沉，仍是我们投资风格的基石。”
The lesson that Kirby took from the episode was not that an investor should buy stocks hoping to find the next Haloid (or Google or Apple). Rather, it was that a portfolio created by acting on only half of the firm’s recommendations and with negligible costs handily outperformed the portfolio to 3 which Kirby fully attended. Buying undervalued stocks and doing nothing did better than attempting to navigate the market’s ups and downs. Warren Buffett expressed a similar point when he said, “Lethargy bordering on sloth remains the cornerstone of our investment style.” 4
我们大多数人从小就被教导：努力会带来结果。但如果把“努力”理解为“活动”，这个道理对长期投资者并不适用。这里的启示很简单：投资者常常出于善意调整投资组合，但这些调整并没有创造价值。这一点，对老练的机构投资者和懵懂的个人投资者都一样。做得更少，反而能让你收获更多。
Most of us are taught from a young age that effort leads to results. But if you take effort to mean activity, the lesson doesn’t apply for long-term investors. The message here is simple: investors often make changes to their portfolios—with the best of intentions—that do not add value. This is as true for sophisticated institutions as it is for the unsophisticated individual. Doing less can leave you with more.
我们研究两种对长期业绩有害的决策。第一种是将投资组合的权重从一个资产类别重新配置到另一个；第二种是在同一资产类别内更换主动管理人。这些错误的根源包括：使用过短的时间跨度来评估、未能认识到均值回归、为保住职位而决策、以及屈服于近因偏差——即倾向于过度重视最近发生的事情。
We examine two kinds of decisions that are deleterious to long-term results. The first is the reallocation of the weightings of the portfolio from one asset class to another. The second is the swapping of active managers within an asset class. The sources of these mistakes include applying a time horizon that is too short, failing to recognize reversion to the mean, seeking job preservation, and succumbing to recency bias—the tendency to overweight what has happened in the recent past.
大多数表明投资者减少活动会受益的研究，都依赖于反事实分析——即仔细研究“本来会怎样”。例如，这种方法会问：“如果我们当初不解雇经理 A 而改聘经理 B，我们的回报会是多少？”自然，雇佣或解雇经理这一行为本身，就帮助决定了该经理的回报，就像增加或减少对某个资产类别的敞口会影响其回报一样。对一个基金或资产类别的资金流入，会贡献于正向的相对回报，而资金流出则与负向的相对回报相关。这一观察限制了反事实分析，因为结果并非独立于行为。但即使考虑到买入有助于回报、卖出会损害回报，频繁交易带来的回报依然糟糕，这恰恰说明了投资者在做出明智决策方面有多么挣扎。
Most of the studies showing that investors would be better off with less activity rely on counterfactual analysis—a careful study of what would have been. For example, this approach would ask, “what would our returns have been had we stuck with money manager A instead of firing 5 A and hiring manager B?” Naturally, the very act of hiring or firing a manager helps determine the manager’s returns just as the act of increasing or decreasing exposure to an asset class affects its returns. Inflows for a fund or asset class contribute to positive relative returns and outflows are linked to negative relative returns. 6 This observation limits counterfactual analysis because the outcomes are not independent of the actions. But that the returns from activity are poor even after considering that the buying helps, and selling hurts, returns indicates the degree to which investors struggle to make good decisions.
资产配置决策：你最近为我做了什么？
The Asset Allocation Decision: What Have You Done for Me Lately?
研究人员已经证明，个人投资者获得的回报低于通过买入并持有策略所能获得的回报。例如，先锋集团的创始人约翰·博格尔研究了截至 2009 年年中的 5 年内交易型开放式指数基金的表现，发现投资者获得的年化回报，平均比他们所投资的 ETF 报告回报低 4.5 个百分点。原因在于资金进出 ETF 的时机。核心概念在于基金的回报与投资者的美元加权回报之间的区别。基金的回报只是每股资产净值的复利年化增长率。而美元加权回报则考虑了投资者的时机选择。由于投资者倾向于在基金表现良好之后才买入，他们错过了上涨，却要承受随后的表现不佳。此外，他们在下跌后卖出，又未能享受到随后的反弹。
Researchers have documented that individual investors earn lower returns than those achieved through a buy-and-hold strategy. 7 For example, John Bogle, founder of the Vanguard group, examined the performance of exchange-traded funds (ETFs) for five years through mid-2009 and found that investors earned annualized returns that were on average 4.5 percentage points lower than the reported returns of the ETFs they invested in. 8 The reason is the timing of the flows into and out of the ETFs. The central concept is the distinction between a fund’s return and an investor’s dollar-weighted return. The fund’s return is simply the compounded annual growth rate in net asset value per share. The dollar-weighted return considers an investor’s timing. Because investors have a tendency to buy a fund after it has done well, they miss the upside but suffer from the subsequent underperformance. Further, they sell after a drop and fail to enjoy the subsequent rebound. 9
投资公司协会保存着共同基金资金流入流出的详尽记录。总体格局一目了然，如图表 1 所示。投资者在市场表现良好时买入（参见 1990 年代末到 2000 年），在市场表现不佳时卖出（参见 2002 年和 2008 年）。这一分析并非暗示你只需要一个简单的买入并持有策略，例如，这样的方法在 2000 年代的头十年会带来负回报。然而，该分析确实凸显了买入被低估的证券并持有的优点。众所周知，投资者在这方面做得很糟。只有时间能证明 2008 年和 2009 年大规模流入债券基金、同时从股票基金大举撤离的资金会带来什么结果，但从数学上看，很难证明未来十年债券表现会更好，即使经过风险调整也是如此。
The Investment Company Institute maintains excellent records of the investments in and out of mutual funds. 10 The overall pattern is clear, as Exhibit 1 shows. Investors buy when the market has done well (see the late 1990s into 2000) and sell when the market has fared poorly (see 2002 and 2008). This analysis does not suggest that all you need is a simple buy-and-hold strategy. Such an approach would have yielded a negative return for the first decade of the 2000s, for instance. The analysis does feature the virtue of buying undervalued securities and holding them. 11 Investors are notoriously poor at doing this. Only time will tell how the massive inflows into bond funds and exodus from equity funds in 2008 and 2009 will play out, but it’s hard to make the math show that bonds will do better over the next decade, even adjusted for risk.
图表 1：股票基金资金流量与市场结果
350 50%
净新现金流 股市总回报 40%
Exhibit 1: Equity Funds Flow and Market Results $350 50% Net New Cash Flow Total Return on Equities 40%
$250 30% 20% $150
$250 30% 20% $150
十亿美元 百分比点
Billions of Dollars Percentage Points
原件此处是表格，PDF 抽取时列结构已丢失，下面只剩按列读出的数字，行列对应关系无法还原。核对数据请打开来源正文。
10% $50 0% -10% 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 -$50 -20% -30% -$150 -40% -$250 -50%
10% $50 0% -10% 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 -$50 -20% -30% -$150 -40% -$250 -50%
来源：投资公司协会。
Source: Investment Company Institute.
你可能会认为，糟糕的资产配置决策只应发生在个人投资者身上，因为他们缺乏培训和资源来做出更好的决策，而机构投资者应该对此类错误免疫。然而，研究表明，机构计划发起人，包括退休计划、工会、捐赠基金和基金会，在从一个资产类别转移到另一个时，也同样未能创造价值。
You might reason that the poor asset allocation decisions are to be expected from retail investors, who lack the training and resources to make better decisions, but that institutional investors would be immune to such mistakes. However, research shows that institutional plan sponsors, including retirement plans, unions, endowments, and foundations, also fail to add value when they move from one asset class to another.
最近一篇论文总结了一项针对计划发起人决策的研究，这些发起人控制着数千种产品和数万亿美元资产，时间跨度超过 20 年。作者的结论是：“计划发起人资金流入的产品组合，其表现逊色于资金流出的产品组合。”换句话说，如果计划发起人总体上什么也不做，结果反而会更好。这一普遍规律中的例外是全球固定收益投资者。（见图表 2。）研究人员估计，计划发起人通过买卖产品已损失了超过 1700 亿美元的价值，即便考虑到其资产规模，这也是一笔不小的数目。
A recent paper summarized a study of the decisions of plan sponsors controlling thousands of products and trillions of dollars over a span in excess of 20 years. The authors conclude that, “Portfolios of products to which they allocate money underperform compared with the products from which assets are withdrawn.” In other words, the plan sponsors would have been better off in the aggregate had they done nothing. The exception to the general pattern was global fixed income investors. (See Exhibit 2.) The researchers estimate that plan sponsors had forgone over $170 billion in value through their purchases and sales of products, a sizeable sum even considering the size of the asset base. 12
作者表明，资产配置并非价值流失的唯一来源。事实上，在获得资金流入的资产与遭受资金流出的资产之间的相对表现差距中，资产配置大约只占三分之一。另外三分之二可归因于作者所称的“产品选择”，即投资者挑选单个管理人的能力。
The authors show that asset allocation was not the only source of the value slippage. In fact, asset allocation represented only about one-third of the relative underperformance of assets getting inflows versus those seeing outflows. The other two-thirds was attributable to what the authors call “product selection,” which reflects how well investors pick individual managers.
图表 2：资金流动后加权投资组合的后续表现（3 年年化）
Exhibit 2: Post-Flow Performance of Flow-Weighted Portfolios (3-Year Annualized)
1.5%
1.5%
表现（流入组合 1.0% 0.5% 0.0%
Performance (Inflow Portfolio 1.0% 0.5% 0.0%
减去流出组合）
Less Outflow Portfolio)
-0.5% -1.0% -1.5% 国内价值 全球 国内 全球
-0.5% -1.0% -1.5% Domestic Value Global Domestic Global
股票 固定收益 来源：摘自 Scott D. Stewart, CFA, John J. Neumann, Christopher R. Knittel, and Jeffrey Heisler, CFA，“Absence of Value: An Analysis of Investment Allocation Decisions by Institutional Plan Sponsors”，《金融分析师杂志》，第 65 卷第 6 期，2009 年 11 月/12 月，第 45 页。
Equity Fixed Income Source: From Table 5 in Scott D. Stewart, CFA, John J. Neumann, Christopher R. Knittel, and Jeffrey Heisler, CFA, “Absence of Value: An Analysis of Investment Allocation Decisions by Institutional Plan Sponsors,” Financial Analysts Journal, Vol. 65, No. 6, November/December 2009, 45.
管理人选择：如果你正火爆，你很快就将不再
Manager Selection: If You’re Hot, You’ll Soon Be Not
投资者倾向于将资本配置给近期表现良好的基金。
Investors have a tendency to allocate capital to funds that have done well in the recent past.
金融学教授安德烈亚·弗拉齐尼和欧文·拉蒙特将投资者通过重新配置决策降低其已实现回报的可预测倾向，称为“傻钱”效应。研究人员通过比较实际回报与假设投资者原地不动时的回报，量化了这一效应。这一反事实分析表明，频繁交易每年使投资者的回报损失超过 1 个百分点，再加上主动管理收取的费用，共同导致了投资者相对于基准的整体不佳表现。
Andrea Frazzini and Owen Lamont, professors of finance, call the predictable propensity of investors to lower their realized returns through reallocation decisions the “dumb money” effect. 13 The researchers quantified the effect by comparing the realized returns to the returns for a portfolio assuming the investor had stayed put. This counterfactual analysis shows that activity costs investors over one percentage point a year in returns, which when added to the fees from active management, contribute to the overall underperformance of investors versus their benchmarks.
研究表明，对冲基金投资者获得的美元加权回报也远低于买入并持有的回报。
Research shows that investors in hedge funds also earn dollar-weighted returns that are much lower than buy-and-hold returns. 14
正如在资产类别之间的配置决策一样，机构也难以为管理人进行卓有成效的资金配置。教授阿米特·戈亚尔和苏尼尔·瓦哈尔仔细分析了 3400 个计划发起人的选择和解除决策，涵盖了超过 9600 个独立决策。他们得出的结论是，计划发起人的这些决策并未增加价值。例如，发起人聘用的管理人在聘用前近期表现优于市场，而被解雇的管理人平均表现不佳（尽管解雇决策很复杂）。但在随后的时期里，“被解雇公司的表现超过了被雇用公司的表现”。
Just as with allocation decisions between asset classes, institutions struggle to allocate funds to managers fruitfully. Professors Amit Goyal and Sunil Wahal did a careful analysis of the selection and termination moves of 3,400 plan sponsors, reflecting over 9,600 distinct decisions. They concluded that the moves of the plan sponsors did not add value. For example, the managers whom the sponsors hire recently outperformed the market and the managers whom they fire have underperformed on average (although the termination decisions are complex). But “the performance of the fired firms exceeds that of the hired firms” in subsequent periods. 15
苏尼尔和瓦哈尔问道：为什么计划发起人会做出事后看来很糟糕的决策？他们首先认为这是狂妄自大，是计划发起人一种毫无根据的信念，即他们能够成功地选聘和解雇。
Sunil and Wahal ask why plan sponsors make decisions that look poor in retrospect. They first suggest hubris, an unfounded belief among plan sponsors that they can hire and fire successfully.
我的经验表明，很少有计划发起人在运作时带着多少傲慢。他们把第二种解释称为“保住饭碗”。其想法是，计划发起人必须展示一些雇佣和解雇的动作，以证明自己在做有价值的事。懒散到近乎怠惰，更容易让投资委员会感到沮丧，而非留下好印象。这就引出了一个关键问题：计划发起人的首席投资官（CIO）在评估一位投资经理的回报时，应该有多大的耐心。
My experience suggests that few plan sponsors operate with much hubris. They call the second explanation “job preservation.” The idea is that plan sponsors have to show some hiring and firing activity in order to demonstrate that they are doing something of value. Lethargy bordering on sloth is much more likely to frustrate an investment committee than impress it. This raises a crucial question of how patient a chief investment officer (CIO) of a plan sponsor should be when evaluating the returns of an investment manager.
投资行业的标准评估期似乎是三年左右——考虑到大多数首席投资官的任期，这看起来是一个合理的时限。然而，一些研究方法声称，你需要远超十年的业绩数据，才能有把握地判断一位基金经理具备技能（即 95% 的置信区间）。如此长的时间跨度并不现实。问题在于，首席投资官们是过为耐心、过为急躁，还是恰到好处。
The standard evaluation period in the investment industry appears to be about three years, a seemingly sensible amount of time given the tenure of most CIOs. Yet some research methods claim that you need well in excess of a decade of results to confidently conclude that a manager 16 has skill (i.e., a confidence interval of 95 percent). A time horizon that long is impractical. The issue is whether CIOs are too patient, too impatient, or about right.
大卫·多诺霍、罗伯特·克里尼安和马修·斯坎伦在最近的一篇论文中探讨了这个问题。17 他们不是去翻查历史数据，而是用一组设定假设做了模拟。这种方法让他们能够评估不同投资时间跨度和模拟所产生的各种结果。
David Donoho, Robert Crenian, and Matthew Scanlan address this question in a recent paper. 17 Rather then scouring historical results, they did simulations using set assumptions. This approach allowed them to evaluate the results that the various investment time horizons and simulations produced.
他们的设定是这样的：有一组投资管理人，共 1,000 名，其中 10% 被研究者赋予了技能。这个技能通过夏普比率来指定——夏普比率衡量的是单位风险所获得的回报——他们为这些有技能的管理人设定了较高的夏普比率。¹⁸他们还创建了一个首席投资官群体，这些 CIO 的聘用与解雇算法体现了他们各自的个性。（见图表 3。）他们改变了各项参数，并将每组模拟运行了 1,000 次。
Their setup had a group of 1,000 investment managers, 10 percent of whom the researchers endowed with skill. They specified skill through the Sharpe ratio, a measure of return per unit of risk, and designated high ratios for the skillful managers. 18 They also created a population of CIOs who had hiring and firing algorithms that reflected their personalities. (See Exhibit 3.) They varied the parameters and ran each simulation 1,000 times.
表 3：首席投资官的性格类型及其决策算法
Exhibit 3: Chief Investment Officer Personalities and Their Decision-Making Algorithms
| 投资者类型 | 选择规则 | 淘汰规则 | 重新启用规则 |
|---|---|---|---|
| 冷酷型短期投资者 | 一年内业绩前 10% | 回撤 10% | 永不 |
| 宽恕型短期投资者 | 一年内业绩前 10% | 回撤 10% | 创新高 |
| 警惕型长期投资者 | 五年内业绩前 10% | 一年内业绩后 10% | 一年内业绩前 10% |
| 忠诚型长期投资者 | 五年内业绩前 10% | 五年内业绩后 10% | 一年内业绩前 10% |
| 买入并持有型 | 五年内业绩前 10% | 永不 | 不适用 |
Investor personality Selection Rule Firing Rule Rehiring Rule Ruthless Short-Term One Year Top 10% Drawdown 10% Never Forgiving Short-Term One Year Top 10% Drawdown 10% New High Watchful Long-Term Five Years Top 10% One Year Bottom 10% One Year Top 10% Loyal Long-Term Five Years Top 10% Five Year Bottom 10% One Year Top 10% Buy and Hold Five Years Top 10% Never Not Applicable
来源：David L. Donoho、Robert A. Crenian 与 Matthew H. Scanlan 合著，《耐心是美德吗？评估回报时采用长期视角的不带感情色彩的论证》，《投资组合管理期刊》，2010 年秋季刊，第 114 页。
Source: David L. Donoho, Robert A. Crenian, and Matthew H. Scanlan, “Is Patience a Virtue? The Unsentimental Case for the Long View in Evaluating Returns,” The Journal of Portfolio Management, Fall 2010, 114.
他们的研究揭示出至少三个重要结论。第一，对于定义为夏普比率 0.5 的熟练基金经理来说，五年模拟期显示的结果差异巨大。
Their research revealed at least three important points. The first is that the simulated results for skillful managers, defined as a Sharpe ratio of 0.5, showed wide variation for a five-year simulation.
这是波动的自然结果，但也突显了一个事实：即使能力出众的经理人，也会经历表现不佳的阶段。正如作者所强调的，预期夏普比率与实际实现的夏普比率之间存在巨大差异。对于大量基金的样本，或者对单只基金而言在很长的时间跨度内，实际实现的比率确实会显现出来，但在多年期间内，单只基金的预期夏普比率与实际实现的夏普比率之间会出现很大的偏差。
This is a natural consequence of variation, but underscores that skillful managers will have periods of underperformance. As the authors stress, there’s a big difference between the expected Sharpe ratio and the realized Sharpe ratio. The realized ratio does surface for a large sample of funds, or for a single fund over a long period of time, but individual funds do see large divergences between expected and realized Sharpe ratios over multi-year periods.
第二点是，即便是存在有能力的基金经理，短期业绩也很难区分技能与运气。例如，100 位有能力的经理中，只有 35 位会出现在一年业绩排名的前 10%。即使将时间跨度拉长到十年，不到一半的熟练经理最终能进入前 10%。换言之，大部分跻身前 10% 的基金靠的是运气。因此，尽管长期业绩是衡量能力的极佳指标——十年期业绩排名前 10% 的基金，其靠能力的概率远高于靠运气的概率——但业绩本身仍带有大量的随机性。而这一切还只是在一个模拟世界中得出的结论——在那个世界里，我们确切知道 10% 的基金经理是有能力的。
The second point is that it is difficult to sort skill and luck through short-term results, even when a subset of managers is skillful. For example, only 35 of the 100 skilled managers show up in the top decile based on one-year results. Even if you expand the horizon to ten years, less than one-half of the skillful managers end up in the top decile. Said differently, a majority of top decile funds are there as the result of luck. So while long-term results are a very good indicator of skill—the probability that a top decile fund over ten years is skillful is vastly higher than the probability that it is lucky—results reflect a large dose of randomness. And this is in a simulated world where we know that 10 percent of the fund managers are skillful.
最后一点是，采用耐心、长期决策规则的投资者，业绩优于使用短期规则的投资者。图表 4 展示了在十年跨度内，短期投资者、长期投资者和买入持有投资者的年化收益率。 19 作者写道：“这些模拟投资情景表明，长期投资者的选择规则产生了最高的年化收益率、最低的管理人更换率，以及投资于有技能管理人的时间占比最高。”作者总结道：“最有利可图的耐心程度，与当前行业实践中的做法截然不同。” 20
The final point is investors who employ patient, long-term decision rules outperform investors who use short-term rules. Exhibit 4 shows the annualized returns for short-term, long-term, and buy-and-hold investors over a ten-year span. 19 The authors write, “these simulated investment scenarios show that the selection rules of long-term investors yielded the highest annualized returns, lowest manager turnover rates, and highest proportion of time being invested in skilled managers.” The authors conclude that, “the most profitable degree of patience is very different from that found in current industry practice.” 20
附录 4：短线投机者受冷落，审慎长期投资者胜出
Exhibit 4: Short-Termers Get Short Shrift and Watchful Long-Term Investors Do Best
12.00% 10.00% 8.00% Annualized Returns 6.00%
12.00% 10.00% 8.00% Annualized Returns 6.00%
4.00%（扣除退出成本）
Net of Exit Costs 4.00%
2.00%
2.00%
0.00% 无情 宽恕 警惕 忠诚 长持 短买 短期 短期 长期 持有 来源：David L. Donoho、Robert A. Crenian 和 Matthew H. Scanlan，《耐心是美德吗？评价回报时坚持长远视角的理性依据》，《投资组合管理期刊》，2010 年秋季刊，第 115 页。
0.00% Ruthless Forgiving Watchful Loyal Long Buy and Short Term Short Term Long Term Term Hold Source: David L. Donoho, Robert A. Crenian, and Matthew H. Scanlan, “Is Patience a Virtue? The Unsentimental Case for the Long View in Evaluating Returns,” The Journal of Portfolio Management, Fall 2010, 115.
Conclusion
Conclusion
在技能与运气的连续统上，投资更靠近运气而非技能这一端。考虑到投资是一个竞争激烈的行业，投入资源巨大且进入门槛相对较低，这一点并不令人意外。不过，确实有证据表明存在具备技能的投资管理人，尽管他们所占比例不到十分之一。21 对交易活动负面效应的一种合理解释是，它反映了追逐随机性所付出的高昂代价。对长期投资者而言，主要有三条经验教训：
On a continuum of skill and luck, with pure skill on one side and pure luck on the other, investing sits a lot closer to the luck side than the skill side. This should not be too surprising considering investing is a very competitive business that employs substantial resources and has relatively low barriers to entry. Still, there is evidence of skillful investment managers, even if they do not represent one-tenth of the population. 21 A reasonable interpretation of the deleterious effect of activity is that it reflects costly randomness chasing. There are three main lessons for long-term investors:
• 警惕均值回归。几乎所有专业投资者都认为自己熟悉均值回归的概念，但总体结果却表明，他们的行为并非如此。均值回归指的是，极端结果之后，将出现一个预期值更接近均值的结果。例如，如果股市过去几年的回报率显著低于历史平均水平，那么有理由预期未来回报率将更接近平均水平。然而，这与投资者的实际行为恰恰相反。高于平均水平的回报会吸引更多资金流入，而低回报则导致资金撤出（见图表 1）。请仔细考虑你正在处理的系统所产生的结果分布，并确保在你的决策中明确纳入均值回归的考量。
• Watch out for reversion to the mean. Nearly all professional investors believe they are familiar with the concept of reversion to the mean, but the aggregate results show that they don’t behave as if they do. Reversion to the mean says that an extreme outcome will be followed by an outcome that has an expected value closer to the mean. For example, if returns from the stock market over a few years have been substantially below the historical average, it is reasonable to expect that future returns will be closer to the average. Yet this is in contrast to how investors behave. Above-average returns attract more capital and poor returns lead to withdrawals (see Exhibit 1). Consider carefully the distribution of outcomes for the system you are dealing with and make sure that you explicitly consider reversion to the mean in your decisions.
• 不过是做好本分工作。如前所述，无所作为——尤其是在短期业绩不佳时——对大多数投资委员会来说难以接受。由于在大多数领域，活动量与结果成正比，大多数人便认为在投资世界里也是如此。
• Just doing my job. As noted earlier, doing nothing—especially in the face of poor short-term results—does not sit well with most investment committees. Since activity and results are correlated in most fields, most people believe they are correlated in the world of investing,
同样，总体数据表明，频繁交易往往会削弱而非提升长期收益。很少有组织或委员会能够在争论问题后决定按兵不动。针对“为行动而行动”的解毒剂，就是不断将反事实情境纳入你的反馈机制（即：假如不做会怎样），并坦然接受在某些情况下什么都不做。
too. Here again, the aggregate data show that activity tends to diminish, not enhance, long-term results. It is the rare organization or committee that can debate the issues and resolve to do nothing. The antidote to acting for the sake of acting is to constantly include counterfactuals as part of your feedback (i.e., what would have been) and to be open-minded about doing nothing in certain situations.
• 近因偏差。人类天生就是寻找模式的动物。因此，当某个东西在上涨时，我们就预期它会继续上涨；当它下跌时，我们就预期它会继续下跌。这种过分看重近期事件并将其外推到未来的强烈倾向，是我们未能吸取均值回归教训的主要原因之一。
• Recency bias. Humans are natural pattern seekers. As a result, when something is going up, we expect it to continue going up. When it’s going down, we expect it to continue going down. This strong tendency to overweight recent events and extrapolate them into the future is one of the main reasons we fail to heed the lessons from reversion to the mean.
为了对抗这种偏差，你需要退后一步，确保自己正在考虑更大范围的案例。一个特别有效的方法是采用外部视角——仔细审视其他人处于同样境况时发生了什么。
To deal with this bias, step back and make sure you are considering a larger set of instances. One particularly effective technique is adopting the outside view—a careful 22 consideration of what happened to others when they were in the same situation.
现有证据整体表明，长期投资者从事的活动——包括在资产类别之间切换，或者从一个基金经理换到另一个——并不能增加价值。尽管投资者的行动无疑基于一种信念，即他们可以提高自己的长期回报，但心理力量却导致他们做出了错误的决策。
The balance of evidence shows that long-term investors engage in activity, including switching between asset classes and from manager to manager, that does not add value. While investors certainly act in the belief that they will enhance their long-term returns, psychological forces cause them to make the wrong decisions.
过往业绩不代表未来表现。所有投资均包含风险，可能损失本金。
Past performance is no guarantee of future results. All investments involve risk, including possible loss of principal.
文中表达的观点系作者截至 2011 年 3 月 25 日的看法，可能因市场及其他条件的变化而调整。这些观点可能与其他作者、投资组合经理或公司整体的看法存在差异，并且不构成对未来事件的预测、对未来业绩的保证或投资建议。预测和模型结果本身具有局限性，不应作为未来表现的指标加以依赖。投资者不应将本信息作为投资决策的唯一依据。
The views expressed are those of the author as of March 25, 2011 and are subject to change based on market and other conditions. These views may differ from the views of other authors, portfolio managers or the firm as a whole, and they are not intended to be a forecast of future events, a guarantee of future results, or investment advice. Forecasts and model results are inherently limited and should not be relied upon as indicators of future performance. Investors should not use this information as the sole basis for investment decisions.
所有统计数据均取自作者认为可靠的来源，但信息的准确性和完整性无法保证。本文评论中提供的信息不应被视为美盛资本管理公司或其任何关联公司对买入或卖出任何证券的建议。
Any statistics have been obtained from sources the author believed to be reliable, but the accuracy and completeness of the information cannot be guaranteed. The information provided in this commentary should not be considered a recommendation by Legg Mason Capital Management or any of its affiliates to purchase or sell any security.
1 沃伦·巴菲特，“致股东信”，伯克希尔·哈撒韦 1996 年年报。
2 奥斯卡·王尔德，《不可儿戏：一部给严肃之人的无聊喜剧》，1895 年 2 月 14 日首演于伦敦圣詹姆斯剧院。见 http://www.gutenberg.org/files/844/844-h/844-h.htm。
Endnotes 1 Warren Buffett, “Chairman’s Letter to Shareholders,” Berkshire Hathaway Annual Report, 1996. 2 Oscar Wilde, “The Importance of Being Earnest: A Trivial Comedy for Serious People,” performed first on February 14, 1895 at St. James Theatre in London. See http://www.gutenberg.org/files/844/844-h/844-h.htm.
3 罗伯特·G·柯比，“咖啡罐投资组合：被动式主动投资比主动式被动投资更能赚钱”，《投资组合管理期刊》，1984 年秋季刊，第 76-80 页。
3 Robert G. Kirby, “The Coffee Can Portfolio: You Can Make More Money Being Passively Active than Actively Passive,” The Journal of Portfolio Management, Fall 1984, 76-80.
4 沃伦·E·巴菲特，《董事长致股东的信》，伯克希尔·哈撒韦年报，1990 年。5 迈克尔·J·莫布森，《哈利·波特是不可避免的吗？累积优势、反事实与光环效应》，《莫布森论战略》，2007 年 9 月 7 日。
4 Warren E. Buffett, “Chairman’s Letter to Shareholders,” Berkshire Hathaway Annual Report, 1990. 5 Michael J. Mauboussin, “Was Harry Potter Inevitable? Cumulative Advantage, Counterfactuals, and the Halo Effect,” Mauboussin on Strategy, September 7, 2007.
6 Paul A. Gompers 和 Andrew Metrick，“机构投资者与股票价格”，《经济学季刊》，第 116 卷，第 1 期，2001 年 2 月，第 229 - 260 页。
6 Paul A. Gompers and Andrew Metrick, “Institutional Investors and Equity Prices,” The Quarterly Journal of Economics, Vol. 116, No. 1, February 2001, 229-260.
7 Ilia D. Dichev，“股票投资者的实际历史回报是什么？来自美元加权回报的证据”，《美国经济评论》第 97 卷，第 1 期，2007 年 3 月，第 386-401 页。近期媒体提及，参见 Paul Sullivan，“何时买入或卖出？别相信你的直觉”，《纽约时报》，2011 年 3 月 11 日。
7 Ilia D. Dichev, “What Are Stock Investors’ Actual Historical Returns? Evidence from Dollar Weighted Returns,” American Economic Review, Vol. 97, No. 1, March 2007, 386-401. For a recent mention in the media, see Paul Sullivan, “When to Buy or Sell? Don’t Trust Your Instincts,” The New York Times, March 11, 2011.
8 Matt Hougan，“博格尔：投资者在 ETF 中正损失惨重”，IndexUniverse，2009 年 6 月 17 日。9 教授 Ilia Dichev 提供了一个很好的例子。假设你以每股 10 美元的价格买入某公司 100 股，总投资 1000 美元。随后股价翻倍，你在年底又买入 100 股（投资 2000 美元）。第二年年底，股价回到 10 美元。买入并持有策略的回报率为零（你在 10 美元买入，股价最终回到 10 美元）。但如果你遵循上述操作模式，你就会亏钱。你的初始 1000 美元保持不变，但你的 2000 美元投资缩水到 1000 美元，这意味着你的 3000 美元投资现在只值 2000 美元。在这种情况下，你的美元加权回报率为 -26.8%，与基金回报率的零相差甚远。
8 Matt Hougan, “Bogle: Investors Are Getting Killed in ETFs,” IndexUniverse, June 17, 2009. 9 Professor Ilia Dichev provides a good example. Say you buy 100 shares of a company at $10 a share, for a total investment of $1,000. The stock then doubles and you purchase an additional 100 shares ($2,000 investment) at the end of the year. The stock then goes back to $10 at the end of year 2. The buy and hold strategy yields a return of zero percent (you bought at $10 and the stock ended at $10). But if you had followed the described pattern, you would have lost money. Your initial $1,000 remained even but your $2,000 investment sunk to $1,000, which means that your $3,000 investment is now worth $2,000. In this case, your dollar-weighted return is -26.8 percent, vastly different than the fund return’s return of zero.
10 投资公司协会，《2010 年投资公司数据手册：投资公司行业趋势与活动回顾》。参见 http://www.icifactbook.org/pdf/2010_factbook.pdf。
10 Investment Company Institute, “2010 Investment Company Fact Book: A Review of the Trends and Activity in the Investment Company Industry.” See http://www.icifactbook.org/pdf/2010_factbook.pdf.
11 家指数基金顾问公司指出，投资于 Dimensional Fund Advisors 所创指数基金的投资者，因时机把握得当，获得了超越基准的回报。参见 http://www.ifa.com/12steps/step1/step1page2.asp。
11 Index Fund Advisors notes that investors in index funds that Dimensional Fund Advisors created have earned returns above their benchmark as the result of good timing. See http://www.ifa.com/12steps/step1/step1page2.asp.
12 Scott D. Stewart, CFA, John J. Neumann, Christopher R. Knittel, 和 Jeffrey Heisler, CFA,《价值的缺席：机构计划发起人投资配置决策分析》,《金融分析师期刊》第 65 卷第 6 期，2009 年 11/12 月，第 34-51 页。
12 Scott D. Stewart, CFA, John J. Neumann, Christopher R. Knittel, and Jeffrey Heisler, CFA, “Absence of Value: An Analysis of Investment Allocation Decisions by Institutional Plan Sponsors,” Financial Analysts Journal, Vol. 65, No. 6, November/December 2009, 34-51.
13 安德烈亚·弗拉齐尼与欧文·A·拉蒙特，《愚蠢的钱：共同基金资金流与股票收益的截面分析》，《金融经济学杂志》，第 88 卷，第 2 期，2008 年 5 月，第 299-322 页。14 伊利亚·D·迪切夫与格温·余，《高风险，低回报：对冲基金投资者实际赚了多少》，《金融经济学杂志》，第 100 卷，第 2 期，2011 年 5 月，第 248-263 页。
13 Andrea Frazzini and Owen A. Lamont, “Dumb Money: Mutual Fund Flows and the Cross-Section of Stock Returns,” Journal of Financial Economics, Vol. 88, No. 2, May 2008, 299-322. 14 Ilia D. Dichev and Gwen Yu, “Higher Risk, Lower Returns: What Hedge Fund Investors Really Earn,” Journal of Financial Economics, Vol. 100, No. 2, May 2011, 248-263.
阿米特·戈亚尔（Amit Goyal）和苏尼尔·瓦哈尔（Sunil Wahal），“计划发起人对投资管理公司的选择与解聘”，《金融学刊》，第 63 卷，第 4 期，2008 年 8 月，第 1805-1847 页。 16 理查德·C·格里诺尔德（Richard C. Grinold）与罗纳德·N·卡恩（Ronald N. Kahn），《主动投资组合管理：一种产生超额收益与控制风险的量化方法（第二版）》（纽约：麦格劳-希尔，2000 年），第 480 页。
15 Amit Goyal and Sunil Wahal, “The Selection and Termination of Investment Management Firms by Plan Sponsors,” The Journal of Finance, Vol. 63, No. 4, August 2008, 1805-1847. 16 Richard C. Grinold and Ronald N. Kahn, Active Portfolio Management: A Quantitative Approach for Producing Superior Returns and Controlling Risk, Second Edition (New York: McGraw-Hill, 2000), 480.
17 David L. Donoho、Robert A. Crenian 与 Matthew H. Scanlan 合著论文《耐心是美德吗？评估回报时坚持长期视角的冷酷案例》，载《投资组合管理期刊》2010 年秋季号，第 105-120 页。
17 David L. Donoho, Robert A. Crenian, and Matthew H. Scanlan, “Is Patience a Virtue? The Unsentimental Case for the Long View in Evaluating Returns,” The Journal of Portfolio Management, Fall 2010, 105-120.
夏普比率的公式如下，其中 R 是投资组合的回报，Rf 是基准的回报，σ 是超额回报 (R – Rf) 的方差。在其他条件相同的情况下，夏普比率越高，表明能力越强。
18 The formula for the Sharpe ratio is as follows, where R is the return of the portfolio, Rf is the return of the benchmark, and σ is the variance of the excess return (R – Rf). All things being equal, higher Sharpe ratios indicate higher skill.
R − Rf E[ R − Rf ]
R − Rf E[ R − Rf ]
S= = σ variance[ R − Rf ]
S= = σ variance[ R − Rf ]
该模拟假设 10% 的经理人是熟练的（使用 0.75 的夏普比率），年化波动率 18%，且结果相互独立。这些假设没有一个是完全现实的，但核心结论很可能在更准确的假设下依然适用。
19 This simulation assumed that 10 percent of the managers were skillful (using a Sharpe ratio of 0.75), annualized volatility of 18 percent, and independent results. None of these assumptions are totally realistic, but the core finding likely applies with more accurate assumptions.
唐诺霍、克里尼安和斯坎伦，前引书，第 116 页及第 105 页。
Donoho, Crenian, and Scanlan, op. cit., 116 and 105.
21 劳伦·巴拉斯（Laurent Barras）、奥利维耶·斯卡耶（Olivier Scaillet）和拉斯·沃默斯（Russ Wermers），《共同基金业绩中的虚假发现：衡量估算阿尔法中的运气成分》，《金融学刊》，第 65 卷，第 1 期，2010 年 2 月，第 179-216 页。
21 Laurent Barras, Olivier Scaillet, and Russ Wermers, “False Discoveries in Mutual Fund Performance: Measuring Luck in Estimated Alphas,” The Journal of Finance, Vol. 65, No. 1, February 2010, 179-216.
22 丹·洛瓦洛与丹尼尔·卡尼曼，《成功的错觉》，《哈佛商业评论》，2003 年 7 月，第 56-63 页。
22 Dan Lovallo and Daniel Kahneman, “Delusions of Success,” Harvard Business Review, July 2003, 56-63.
如需更多信息，请联系：
For more information:
www.leggmason.com
www.leggmason.com
www.twitter.com/leggmason
www.twitter.com/leggmason
本文所表达的观点仅代表美盛资本管理截至 2011 年 3 月 25 日的看法，并可能随市场及其他条件变化而调整。这些观点可能与其他投资组合经理或整个公司的看法存在差异，不应被视为对未来事件的预测、未来结果的保证或投资建议。所提供的信息仅用于描述美盛资本管理的投资策略和证券筛选流程，并未顾及任何可能收到该信息的特定个人的具体投资目标、财务状况或特殊需求。
The views expressed are those of Legg Mason Capital Management as of March 25, 2011 and are subject to change based on market and other conditions. These views may differ from those of other portfolio managers or the firm as a whole, and are not intended to be a forecast of future events, a guarantee of future results, or investment advice. The information provided is intended solely to describe Legg Mason Capital Management’s investment strategies and securities selection process, and does not have regard to the specific investment objectives, financial situation and particular needs of any specific person who may receive it.
本文所表达的观点旨在阐述美盛资本管理公司（Legg Mason Capital Management）在固定收益市场上的整体策略与看法。并非所有策略和观点都适用于所有产品。这些意见不针对某一特定策略，也不代表美盛资本管理公司所提供的任何产品。分散投资并不能保证避免市场损失。
The views expressed are intended to describe Legg Mason Capital Management’s overall strategy and view concerning the fixed-income markets. Not all strategies and views may be relevant to all products. The opinions are not directed to any one strategy in particular, and are not meant to represent any product offered by Legg Mason Capital Management. Diversification does not ensure against market loss.
任何统计数据均来源于投资组合经理认为可靠的渠道，但无法保证这些信息的准确性和完整性。过往业绩不保证未来表现。未经书面许可，不得以任何形式复制本材料。欲了解更多信息，请访问我们的网站：www.leggmason.com/individualinvestors。
Any statistics have been obtained from sources the portfolio manager believed to be reliable, but the accuracy and completeness of the information cannot be guaranteed. Past performance is no guarantee of future results. This material may not be reproduced in any form without written permission. For more information, visit our Web site at www.leggmason.com/individualinvestors.
本出版物仅供信息参考，反映美盛资本管理公司当前的看法。其中所含信息被认为准确，但无法担保。所表述的观点不构成对任何证券买卖的邀约或招揽，且可能随时变更，恕不另行通知。本材料中的陈述不应被视为投资建议。美盛资本管理公司的员工和/或客户可能持有文中提及证券的头寸。本出版物在编制时未考虑您的目标、财务状况或需求。在依据此信息采取行动前，您应结合自身的目标、财务状况或需求，考量其适当性。您有责任了解并遵守您所在国家/地区的适用法律法规。
This publication is for informational purposes only and reflects the current opinions of Legg Mason Capital Management. Information con-tained herein is believed to be accurate, but cannot be guaranteed. Opinions represented are not intended as an offer or solicitation with respect to the purchase or sale of any security and are subject to change without notice. Statements in this material should not be considered investment advice. Employees and/or clients of Legg Mason Capital Management may have a position in the securities mentioned. This publication has been prepared without taking into account your objectives, financial situation or needs. Before acting on this information, you should consider its appropriateness having regard to your objectives, financial situation or needs. It is your responsibility to be aware of and observe the applicable laws and regulations of your country of residence.
Legg Mason Perspectives® 是 Legg Mason Investor Services, LLC 的注册商标。
Legg Mason Perspectives® is a registered trademark of Legg Mason Investor Services, LLC.
© 2011 年美盛投资者服务有限责任公司。美国金融业监管局（FINRA）成员，证券投资者保护公司（SIPC）成员。美盛资本管理公司（隶属于美盛公司）和美盛投资者服务有限责任公司以及上述提及的所有实体，均为美盛公司的子公司。405389 MIPX013439 3/11 FN1110656
© 2011 Legg Mason Investor Services, LLC. Member FINRA, SIPC. Legg Mason Capital Management of Legg Mason, Inc. and Legg Mason Investor Services, LLC and all entities mentioned above are subsidiaries of Legg Mason, Inc. 405389 MIPX013439 3/11 FN1110656
投资产品非 FDIC 承保 • 无银行担保 • 可能损失价值
INVESTMENT PRODUCTSNOT FDIC INSURED • NO BANK GUARANTEE • MAY LOSE VALUE