很多人批评对冲基金的回报不高，其实犯了概念错误。说对冲基金「赢不了大盘」，就像拿船和汽车比速度，然后抱怨船在高速上跑得慢，完全比错了对象。
购买标普 500 指数（也就是市场因子）的年度成本约为 0.09%。顶级对冲基金的年度费率在 5%-8%（2/20 收费模式加各种费用）。成本差异达到 50-80 倍。
如果两者提供的东西一样，那投资者就是傻瓜。但它们提供的确实不同，那些投入数十亿美元的机构投资者也不是傻瓜。
他们购买的是用金钱也无法复制的东西：因子中和性、高夏普比率、大规模且不相关的收入来源。明白这一点后，就会理解高昂费用的合理性，也不会再拿对冲基金和指数基金做比较了。
常见的批评是：“今年标普上涨了 17%，对冲基金只赚了 9.3%。”对大多数对冲基金来说，这种批评或许成立，毕竟很多基金只是市场波动的打包产品。
但这完全误解了像 Citadel/Millennium/Point72 这类顶级基金的产品逻辑。它们的目标不是赢得大盘，这根本不是它们的任务。用一个设计为零相关性的基金与 100% 股票基准比较，就像奇怪地指责保险单不赚钱一样毫无道理。
当你管理着数千亿养老金时，其中已经有 600 亿投资于股票。你并不缺乏股票风险敞口，反而持有过多股票。你真正需要的是在股市下跌时也能实现上涨的资产（或至少不会跟随下跌）。你需要的是风险分散。更准确地说，你想要的是无论市场如何都能上涨，表现优于现金的资产。
听起来很棒吧？感觉会很昂贵吧？没错！真正的风险分散非常昂贵，因为它非常稀缺！
The S&P 500's long-term Sharpe Ratio is around 0.35-0.5, which means that for every 1% change in volatility, you achieve a 0.35%-0.5% excess return. The Sharpe Ratio of top global hedge funds is between 1.5-2.5 or even higher.
What we are talking about is maintaining a Sharpe Ratio around 2 for decades, not only gaining returns unrelated to market fluctuations but also with much lower volatility. These companies have minimal drawdowns and quick recoveries.
Hedge funds are not just an expensive version of the same product but belong to a completely different category. Top hedge funds offer two advantages that ETFs/index products do not:
· Factor Neutrality
· High Sharpe Ratio
To understand the value of factor neutrality, look at this formula:
Return = Alpha + Beta × Factor Return + Random Error
· Alpha = Return from skill
· Beta = Exposure to systematic factors
· Factor Return = Return of market factors
· Random Error = Individual differences
The Beta part can be replicated with public factor combinations. Anything that can be replicated should only incur replication costs. Replication is inexpensive: market factors 0.03%-0.09%, style factors 0.15%-0.3%.
Alpha is what remains after deducting all replicable parts. By definition, alpha cannot be synthetically created through factor exposure. This non-replicability is the basis of premium.
Key insight: Beta is cheap because factor returns are public goods with unlimited capacity. If the market goes up by 10%, all holders earn 10%, with no exclusivity. The S&P's returns do not decrease because more people buy in.
Alpha is costly because it is a zero-sum game with limited capacity. For every $1 earned in alpha, someone else loses $1. The inefficiency creating alpha is limited, and it disappears with capital inflows. A strategy with a Sharpe of 2 at $100M might only be 0.8 at $10B because large-scale trading itself affects prices.
Factor neutrality (where all systematic exposures' beta ≈ 0) is the only truly non-replicable source of return. This is the rational basis for the premium, not the return itself, but the inability to obtain this return in any other way.
高度复利效應會隨時間顯現。兩個預期收益都是 7% 的組合，波動率不同（16% vs 10%），20 年後結果天差地別。低波動組合虧損概率減半，下行保護好得多。
對需要穩定支出的機構來說，這種可靠性值得付費。
波動率不僅影響投資體驗，在數學上還會侵蝕長期收益：
幾何平均收益 ≈ 算術平均收益 - （ 波動率²/2）
這叫「波
