Sunday, November 4, 2012

為何我們不願再平衡?

Rebalance,即投資組合再平衡,是指投資組合歷經一段時間的市場波動後,將失衡的各個資產配置調回原始比例的操作。研究顯示,再平衡具有明顯的長期利益;然而,絕大多數投資人並不會將他們的投資組合重新調整,也就是低價買入、高價出售。實際上,許多投資者的行為卻是反其道而行,這個行為令專家充滿疑惑。

投資人為何難以對投資組合再平衡呢?原因並非出在不理性行為,而是有些理智的投資者,更關心那些將投資報酬最大化的其他事情上。因為投資者表現出風險趨避的心態,使他們很難再平衡投入到高報酬的資產上,也就是那些價格已經跌深的資產上。投資顧問和金融學者發現︰不願意低買高賣不僅是一般人投資人的特點,就算是經驗豐富的機構投資者,也容易犯同樣的錯誤。
為何要再平衡?
金融研究的一個重要發現顯示︰資產價格的長期走勢會出現均值回歸(mean-reversion)。因此,當一個資產的價格下跌,使資產的估值相對于歷史水準更具吸引力,也就是未來將有更高的投報率。例如︰當S&P500指數價格下跌,股息收益率增加;就經驗而言,S&P500指數在接下來五年內的報酬將高於平均值。相同地,當公司債的價格隨著信貸利差(credit spreads)驟增而下降,則公司債未來的投報率將增加。從資產投報率的價格均值回歸顯示︰從長遠來看,對資產配置的估值變化採取嚴謹的再平衡方法,將可提高投資組合的投報率。
如果再平衡有如此優良的投資優勢,為何投資人不願意採取這個低買高賣的投資策略?研究顯示︰我們看待風險的態度在某種程度上會受到我們財富水準的影響,並因此影響我們對於再平衡的決策。確切地說,當投資組合財富減少時,投資者會愈來愈懼怕風險,那麼儘管價格很低他們都不願再冒險;反之當投資組合財富增加時,投資者愈來愈喜歡冒險,那麼即使價格再高他們也很樂意冒險投機。
下面將更加具體地解釋這種行為。設想一個有穩定收入和財富的家庭,收入及財富狀況決定了這個家庭的生活水準與消費模式。當經濟出現正面的刺激,例如GDP激增或公司盈利增加,則股票價格上漲;一系列正面的經濟因素形成股票牛市,投資者的財富亦隨之增長。市場環境也呈現低失業率、強勁的薪資增長、豐厚的獎金和房地產增值等特徵。
投資獲利及穩定富足的收入,提供這個家庭計畫外的度假、換購大房子以及/或提早退休等選擇。家庭收入變多及大幅增加的財富,允許他們在不影響生活水準和退休計畫的情況下,能夠承擔更大的投資風險。投資者將新獲得的財富視為意外之財,如同在賭場贏得的賭資。通常,投資者會將此意外之財進行投機;這與一個賭徒在一場瘋狂的賭注中成為贏家,並開始大肆揮霍沒有什麼不同。賭資效應可以解釋,為何投資人在股票價格大漲後,並不急於再平衡的原因,儘管價格升高將明顯地降低未來的投報率。
相反地,當市場遭遇負面的經濟襲擊,所產生的熊市會摧毀許多家庭的財富。熊市通常伴隨著高失業率、低薪資、無分紅和不景氣,並使得房價下跌。收入減少和財富損失,對家庭生活造成重大威脅。的確,投資組合價值的任何損失,對一個家庭的生活水準和退休計畫,都將產生永久的、令人不悅的影響。財富驟降可能意味著投資者再也無法負擔孩子的大學教育費用,更糟的是要賣掉自己的房子,或是把退休計畫延後幾年。無庸置疑地,投資人開始厭惡風險,不管資產價格有多誘人也不願再冒投資風險。

機構投資受託人(institutional investment fiduciary)也有類似的行為。在市場大跌後(通常伴隨著較低的利率,用以提高退休基金資產的市場價值),退休基金很可能出現資金短缺;退休金積蓄率(funding ratio)更進一步的下跌趨勢,將會引發強制性供款(mandatory contribution)或者其他不利投資人的監管機制。此外,人們在艱困時期更加容易責怪他人,個人財富的巨大損失讓工作上壓力更加沉重。當下的環境不能容忍任何短期的負面結果,即使這種負面結果是因長期高報酬的明智投資所造成的。因此,投資受託人在經歷了熊市後更加害怕承擔風險;即使有更好的投資報酬,但因為股價劇跌後,他們無意願再平衡至風險性資產。
風險溢酬或是免費午餐?
強烈的證據顯示,天真的投資人會追逐趨勢,例如︰他們對最近一連串好或壞的消息過度地推斷,並且在資訊不足的情況下追價,導致不理智的買賣行為;這使得價格脫離它們合理的價值,也同時為別人創造超額收益的機會。事實上,那些被認為經驗老道的投資者,其實也好不了多少,他們獲取免費午餐的能力並未特別獨到。隨著時間變化的風險規避,是交易週期和股票牛熊市週期所形成的自然結果。理智的投資者,在股市下跌時懼怕風險,而不敢購入那些價格吸引人的股票;在股市上揚時過度地追求風險,而不會減少持有那些未來投報率較不吸引人的股票。結果,股票(以及其他更廣的資產類別)的價格均值回歸,並未讓理性的投資者套取利益。
價格均值回歸所產生的投資報酬,經常被稱為再平衡報酬。當投資者再平衡至價格下跌的資產,並遠離安全的資產;也就是他們再平衡至未來高報酬的資產,並遠離未來低報酬的資產。久而久之會產生較佳的投資組合績效,並優於買入並持有策略(buy-and-hold approach)。因為天真投資者的過度論斷,以及理智投資者的隨著時間變化之風險規避,這兩種特性使得再平衡的報酬,被視為免費午餐風險溢價的一部分。免費午餐是來自沒有經驗投資者的錯誤交易行為,風險溢價是來自於承受更多無法忍受的市場風險。
你應該再平衡嗎?
如果說為何我們不再平衡?這個問題很難回答,那麼你應該再平衡嗎?這個問題就更難回答。就統計上,若你對重大的價格變動做出相應的再平衡,長遠看來你很有可能績效優異。然而,當你在經濟不景氣時購買了風險資產,在這段過渡期間有很高的機率,你的投資組合很可能遭受比不做再平衡更慘烈的損失。在短期看來,當你進行再平衡,身為投資受託人的你被開除的機率、被客戶責怪的機率以及導致夫妻間爭執的機率,都會變得較高。
所以,誰應該再平衡呢?如果與你的消費需求比起來,你是過份地儲蓄,那麼再平衡及相關的短期風險並不會威脅到你的生活水準,再平衡對於喜歡儲蓄的人來說是個極好的策略。如果你在一個隻關注長期投資績效、不在意短期波動的機構工作,那麼再平衡對你來說是個很好策略。此外,你最好要記住凱恩斯的偉大見解︰市場處於不理性的時間,比你我擁有工作與幸福婚姻的時間還來的長。
因此,即使投資者確信資產報酬中存在著價格均值回歸,他也不一定會進行再平衡。歐洲股票,特別是金融類股票的價格被大打折扣──為投資者提供了前所未有的收益及較高的預期報酬。你是否會懷著堅定的信念,將再平衡至歐洲股票?或金融類股票?我們經歷了與投資相關的恐懼不安,我們不敢買進相對便宜的歐股,並非因為我們有較好的能力去預測歐盟的失序解體。畢竟資產的價格,在很大程度上受華爾街操盤手和避險基金經理人的規範,反映了所有與風險相關的情境和機率。在訊息不對秤的競爭下,我們並沒有特別的能力去預測未來股價的波動;我們不敢再平衡,是因為我們並非是刻版追求投資報酬率最大化的投資人。
因為我們只是凡夫俗子,我們在做投資並尋求良好的長期報酬之前,會再三考慮其他各種與投資無關的個人風險因素,這也就是為何我們不做再平衡。或許這就是為何學者專家們現在會大力推薦制度化再平衡(institutionalized rebalancing)”,也就是要讓再平衡變成退休金管理的制度的一部分,而不是讓投資官員和基金受託人做決定。雖然人類擁有任何電腦都不及的智慧與學習的能力,但遺憾的是,我們卻比一套簡單的定期再平衡的程式還不善於長期報酬的管理,因為我們的風險意識不僅多變,且經常是受到與投資無關的外在環境影響而更令其捉摸不定。

作者: 許仲翔 

上海證券報: http://finance.sina.com.cn/stock/t/20121015/012613364066.shtml
  

The Role of Risk in Asset Allocation

The traditional asset allocation framework, unsurprisingly, starts with assets. It is a tradition based on convenience and, perhaps, an implicit assumption that key asset classes match well to the important risk exposures. The more modern asset allocation and analytic framework anchors, instead, on “risks.”1  While the two frameworks may lead to similar outcomes, the risk-based approach can often offer greater simplicity and allow for more natural asset allocation intuition. In this article, I explain the benefits of the risk-based approach relative to the asset-based approach. Additionally, I introduce simplifying analogies, which facilitate building intuition on the differences between the two approaches. Toward the end of the article, I also offer three applications of the risk-based framework to demonstrate investment issues, which, otherwise, would not be apparent in an asset-based analytical framework. However, a complete description on how to implement a risk-based approach is outside of the scope of this article. 


Asset Classes vs. Risk Exposures
In the asset-based framework, the allocation process involves assigning weights to the various asset classes available to the investor (e.g., equities, bonds, commodities, real estate, etc.). Asset classes are captured by their corresponding market indexes. Each specific major asset category is split across finer asset classes such as U.S., international, and emerging markets for equities, and U.S. Treasuries, sovereigns, and corporates for bonds. In this framework, assets are investment vehicles for “owning” risk exposures; so the “asset-based” approach is, essentially, an “investment product-based” approach.

The more modern analytical framework is a risk-based approach, which makes a strong distinction between investment vehicles and risk exposures. In this framework, the allocation process involves assigning weights to a set of risk exposures rather than assets. The allocation process first determines the “risks” that an investor wants to hold, taking into account how the risks interact with each other and the premia they generate. Then, the investor can construct his preferred combination of “assets” to achieve his desired risk exposures, taking into account the valuation levels attached to assets. Typically, the investor will have a preference for using “attractively priced” assets to access the desired risk exposure.2

The standard criticism of the traditional asset-based approach is that it leads to portfolios that are dominated by equity-like risk, even though portfolios appear to be well diversified.3 This occurs, in part, because very different assets can often contain significant exposure to equity-like risk. Generally, most researchers agree that there are a few primary economic risk exposures: shocks to economic growth, shocks to inflation, and shocks to credit availability, among others. Many assets, if not most, contain multiple risk exposures. For example, corporate bonds are exposed to all three of the above risks. Similarly, high yielding stocks can also have significant exposure to all three risks. Therefore, adding high yield bonds to a portfolio of high yielding stocks wouldn’t necessarily improve the portfolio’s risk diversification, despite the increase in asset class diversification.

Nutrients are to Foods as Risks are to Assets
The risk-based approach, with its associated technical jargon such as “risk factor loadings,” can seem unintuitive to many investors. I find the following food analogy to be very effective at illustrating the risk-based framework.4 It is often convenient to think of risks as nutrients, assets as foods, and portfolios as meals. People need to consume a mix of nutrients, which vary by individual circumstances. Because nutrients come bundled in various foods—dairy, grains, meats, for example—people must combine foods to create a meal that supplies them with the desired nutrition. However, it is likely that many different meals would provide comparable nutrition. Thus, personal taste and food prices often dictate the preferred meal.5 

In asset allocation language, individual asset classes contain different risk exposures. A desired combination of risks can be achieved with different asset allocation portfolios. Ultimately, prices, costs, and investment governance will dictate the preferred portfolio.

The food analogy is also helpful for understanding tactical asset allocation (TAA). For example, when food prices change, we can choose to consume the same nutrients at a lower cost by eating a different meal consisting of different food ingredients. In the risk framework, TAA can be understood as tactically rebalancing toward out-of-favor assets that provide “cheaper” access to a set of underlying economic risks and away from the “expensive” assets offering the same risk exposures.

Applications of the Risk-Based Framework
We illustrate the risk-based framework with the following three applications. These applications are meant to illustrate investment insights, which would not be available through the traditional asset-based analysis.

Application 1: Re-thinking “rebalancing and the strategic portfolio weights”
n the asset-based framework, the stocks (proxied by the S&P 500 Index) and bonds (proxied by the BarCap Agg Index) are viewed as fundamental portfolio building blocks. U.S. investors generally hold large (and often static) strategic allocations tied to the two benchmarks, with a 60% equity/40% bond strategic allocation as the traditional “norm.”

It is dangerous, however, to assume that the S&P 500 or the BarCap Agg6 are assets with static risk exposures over time. In 1995, technology stocks comprised 9.4% of the S&P 500. The index had a P/E ratio of 17.4 and a dividend yield of 2.2%. In 2000, technology stocks became 21.2% of the S&P 500, pushing the index volatility from its historical average of 15% to 24%, the P/E ratio to 24.4, and the dividend yield to 1.2%. Similarly, in 2000 the BarCap Agg had a 4.5 year duration, while yielding 6.4%. Today, the BarCap Agg has duration risk of 5 years, while yield fell to an abysmal 1.6%. Clearly, a disciplined rebalance back toward the 60/40 allocation over this period would have produced a portfolio with wildly fluctuating underlying risk exposures!

Using the food analogy again, it is instructive to think of the BarCap Agg as a hamburger. As America demanded more “manly” beef patties, fast food restaurants moved to double patties, often with bacon to boot. The proteins, not to mention the calories and fat, of today’s gourmet burgers are significantly higher than the burgers of the past (333 calories for an average burger 20 years ago vs. 590 calories today). Given the Agg’s significant increase in duration risk, not to mention the lower yield—is it wise to still insist on a hamburger combo meal? In fact, would it not be better to change our meal completely and source our proteins and calories from cheaper ingredients?

Application 2: Interpreting hedge fund performances
From the asset-based framework, hedge funds are particularly difficult to examine. Many hedge funds trade exotic and illiquid assets. The hedge funds, which hold conventional securities, would often apply complex strategies involving leverage and shorting. The complexity has sometimes led investors to treat hedge funds as a separate asset class, to which the cynics retort that the only shared characteristics for entrees in the asset class are opacity and high fees.

Much of the black-box complexity can be unraveled in the risk-based space, providing some useful insights into hedge fund strategies. It turns out that many hedge fund strategies can be mimicked using more liquid and traditional assets. This is because many hedge funds, despite their exotic holdings and strategies, actually (probably unintentionally) end up owning fairly commonplace risk exposures. Further, for the average funds, there is often little evidence that accessing standard risks through more exotic assets or using complex trading strategies has led to superior returns.7  To be fair, some hedge funds may provide exotic risk exposures that are not found in conventional assets or strategies. For example, earning returns from exposures to extreme economic shocks by writing options is an innovation that expands the investment frontier.

Using our nutrient analogy, hedge fund providers argue that their products provide exclusive nutritional compounds in the form of “alphas” and rare nutrients in the form of “exotic betas.” Hard-to-get nutrients and exclusive health compounds are necessarily expensive. We now know that the average hedge fund actually provides nutrients that can be found, readily, in standard assets; only a small fraction of hedge funds truly provide the hard-to-get “exotic betas” and even fewer provide proprietary “alpha.” In this context, most hedge funds are more like foo-foo health foods, such as bird nest and shark fin, which, at hundreds to thousands of dollars per pound, are advertised to combat aging and cancer, but actually contain nothing more than garden variety vitamins and proteins.

Application 3: Risk parity
Risk parity is an asset allocation portfolio heuristic that attempts to provide a diversified portfolio of risk exposures. Specifically, it seeks to overcome the heavy dependence on equities in the conventional 60/40 allocation portfolio. The implementation of the concept is often in the “asset” space. This means there would be parity in the assets’ contribution to the overall portfolio volatility, but no parity in the underlying economic risk exposures.

The popular and standard risk parity solution is based on volatility weighting of “distinct” asset classes. As with a naïve reliance on the 60/40 allocation, a naïve asset-based approach to risk parity is also sub-optimal, because asset classes can often appear distinct but actually contain similar risks.8 A seemingly diversified risk parity portfolio, constructed from equities, commodities, high yield credit, real estate, and bonds, is like a mixed grill of beef, pork, lamb, and chicken with a small side salad—i.e., not a balanced meal nutritionally. This risk parity portfolio probably provides no better diversification than a simple 60/40 equity/bond portfolio.

Conclusion
When investors analyze choices in the asset-based framework, the large variety of different yet related assets can make the analysis extremely complex; naïve investors can often mistake the asset diversity in their portfolios for adequate risk diversification. Further, because assets contain both risks and valuation in the same bundle, it would lead to easier analyses if we unbundle the two components. The risk-based approach to asset allocation allows us to separate the two, leading to more intuitive and perhaps more sensible portfolio solutions. Despite the technical jargon and the seemingly more abstract framework, the risk-based approach has a lot to offer investors—particularly in a world where investment options and strategies are becoming exponentially more complex.

References
Agarwal, Vikas, and Narayan Y. Naik. 2000. “Multi-Period Performance Persistence Analysis of Hedge Funds.” Journal of Financial and Quantitative Analysis, vol. 35, no. 3 (September):327–342.

Bhansali, Vineer, Josh Davis, Graham Rennison, Jason Hsu, and Feifei Li. 2012. “The Risk in Risk Parity: A Factor-Based Analysis of Asset-Based Risk Parity.” Journal of Investing, vol. 21, no. 3 (Fall):102–110.

Chaves, Denis, Jason Hsu, Feifei Li, and Omid Shakernia. 2012.”Efficient Algorithms for Computing Risk Parity Portfolio Weights.” Journal of Investing, vol. 21, no. 3 (Fall):150–163.

Chen, Nai-Fu, Richard Roll, and Stephen Ross. 1986. “Economic Forces and the Stock Market.” Journal of Business, vol. 59, no. 3 (July):383–403.

Eling, Martin. 2008. “Does Hedge Fund Performance Persist? Overview and New Empirical Evidence.” Working Paper No. 37, University of St. Gallen Law & Economics.

Ennis, Richard M., and Michael D. Sebastian. 2003. “A Critical Look at the Case for Hedge Funds.”  Journal of Portfolio Management, vol. 29, no. 4 (Summer):103–112.

Fung, William, and David A. Hsieh. 1997a. “Empirical Characteristics of Dynamic Trading Strategies: The Case of Hedge Funds.” Review of Financial Studies, vol. 10, no. 2 (Summer):275–302.

———. 1997b. “Survivorship Bias and Investment Style in the Returns of CTAs.” Journal of Portfolio Management, vol. 24, no. 1 (Fall):30–41.

Fung, W. and Hsieh, D. (2004) ‘Hedge fund benchmarks: A risk based approach’, Financial Analyst Journal, 60(5), 65-80.

———. 2004. “Hedge Fund Benchmarks: A Risk-Based Approach.” Financial Analysts Journal, vol. 60, no. 5 (September/October):65–80.

Endnotes
1 The modern approach has grown out of the literature on APT (see Ross, 1976) and the subsequent refinement of the risk factors into meaningful economic risk exposure (see Chen, Roll, and Ross, 1986).

2 Note that this “unbundling” of risk and valuation decision allows us to think carefully about what (beta) risks we are willing to take to earn returns and to examine how diversified our sources of “beta” risks are. The valuation question enters next. For many investors, who believe that assets can be mispriced relative to their risk exposures, this offers an opportunity for asset allocation “alpha” through selecting cheaper assets to gain the desired risk exposures.

3 Note that the classic pension portfolio, structured from the 60/40 equity/bond construct, has 90% of its total portfolio variance driven by equity risk. See Bhansali, Davis, Hsu, Li, and Rennison (2012) for a review of the risk concentration issue commonly found in asset-based asset allocation approaches.

4 The nutrient vs. food analogy is not original; it has been used previously by Professor John Cochrane at the University of Chicago and Professor Andrew Ang at Columbia University.

5 Also important is that some assets provide access to a particular risk without introducing other unwanted risks. For example, chicken breasts provide protein more effectively than rib-eye steaks, which are both more expensive and contain more artery-clogging saturated fat.

6 BarCap Agg is the Barclays Capital Aggregate Bond Index, which is one of the most commonly used bond indices.  It contains almost all of the U.S. investment grade bonds, including Treasury, agency, mortgage, and corporate bonds; the weights are based on market capitalization of the bond issues.  The index is generally dominated by Treasury bonds due to the issuance size of U.S. Treasuries relative to other bonds.

7 See Fung and Hsieh (1997a,b, 2004), Aggrawal and Naik (2000), Ennis and Sebastian (2003), and Hasanhodzic and Lo (2007). For a comprehensive survey review of the literature on hedge fund performance, see Eling (2008).

8 See Chaves, Hsu, Li, and Shakernia (2012) and Bhansali, Davis, Hsu, Li, and Rennison (2012).


jason c hsu

the article has been reprinted at the following websites:

http://www.rallc.com/ideas/pdf/simply_stated/Simply_Stated_Oct_2012_The_Role_of_Risk_in_Asset_Allocation.pdf