Tuesday, May 18, 2010

How should John Doe (an ordinary, unsophisticated saver) invest his/her money?

From: VEGA
Subject: Long-term portfolio for unsophisticated investors

What do you think? The average investor should hold the following depending on their liquidity needs. The premise is that trying to get more specific is more likely to cause harm then help. Following the plan below also minimizes advisor fees, churn (both tax consequences and transaction costs), and mental energy

The basket:
VEU = whole world ex-US. Very low expense ratio of .25%. A little too much emphasis on Japan and old Europe for my taste.
EEM = emerging markets
LQD = investment grade corporate bond fund
HYG = high yield corporate debt

20+ years:
10% LQD, 10% HYG, 30% EEM, 25% SPY, and 20% VEU, 5% physical gold

10-20 years:
20% LQD, 15% HYG, 25% EEM, 20% SPY, 15% VEU, 5% physical gold

5-10 years:
25% LQD, 30% HYG, 15% EEM, 15% SPY, 10% VEU, 5% physical gold

2-5 years:
40% LQD, 25% HYG, 10% EEM, 10% SPY, 5% VEU, 10% physical gold

<2 years:
65% LQD, 25% cash, 10% physical gold

Subject: RE 1: Long-term portfolio for unsophisticated investors

I think of it differently, but I agree that one must try to keep it simpler for the "average investor" or saver. Some general themes are to be global and watch for inflation (one must have a personal view on the likelihood and/or timing of inflation).

1) Safe Assets: The bond allocation should equal the investor's age. So a 60-year old should have 60% bonds and cash.
ETFs choices for that would be: SHY (short-duration US Treasuries, 1-3 yr), TIP (US inflation protected bonds), EMB (Emerging market bonds), HYG/JNK (High yield corp. bonds)
Cash should be enough to meet the next 1-3 year's spending needs, and should be split between 2-3 currencies (dollar, euro, gold, with my personal preference being some local currency mixed with a large gold holding over the next 5 years). If one has extreme views about inflation (above 5%), up to half the safe assets allocation could be in gold. If one has extreme views about deflation, long-term USTs or IGs such as LQD should work.

2) Risk Assets: Stocks and a real estate allocation should be the rest. One needs to have a sense of over and under valuation to invest in risk assets. I would never give a generic recommendation (passive holders make stocks go to dumb prices). Generally, fair valuation for US stocks occurs when the P/E is 15 (a 6.7% yield), and for real estate when REITs yield 5-6%. But this all depends on what the benchmark 10-year UST and German Bund rates are. For risk assets, one should try to buy low, sell high. Right now, some sectors of the stock market seem expensive (financials) whereas others seem cheap (energy and mining). So while I may suggest some SPY and EEM for diversification, I would feel better about suggesting IXC, XLE, XES, XOP. For REITs, I would consider RWO, VNQ, and so on.

Other factors to consider:

Risk preferences (willingness to take risk) - people willing to take risk could have a larger risk assets allocation

Large wealth (ability to take risk) - ditto for people with large amounts of wealth and small commitments/liabilities - if one is frugal, has a large income and large wealth base, that person can own more risk assets

Correlation to day job - people with safer/stable day jobs (dentists, doctors, accountants, etc.) can afford to take more risk compared to others (bankers, traders, real estate developers, etc.). For example, the billionaire Donald Bren, one of the world's best developers, doesn't own stocks or other risk assets. 90% of his wealth is in real estate which fluctuates a lot and which he personally oversees (control matters!), and the remaining 10% is in ultra-safe USTs (I wouldn't be surprised if he has millions in gold stocked away too).

From: VEGA
Subject: RE 2: Long-term portfolio for unsophisticated investors

The whole idea is to take “personal view” out of the average investor’s decisions, since their personal view is by definition very likely to be wrong. When the average investor is most bearish on anything, that’s the best time to buy. By creating a mechanical system, we protect the investor from themselves. Also, the purpose is to create a template that will be generally sustainable for 20+ years, similar to what Graham did in Intelligent Investor. Recommendations of specific sectors will be outdated within 4 years, and more importantly, its fundamentally subjective. With a good template, advisors might argue over appropriateness, but not the investment recommendations themselves.

For most investors, I don’t think SHY provides any benefit over a savings account or money market account. The transaction costs are likely to outweigh the slight interest.

The target audience that I mean when I write “unsophisticated investor” is someone who would likely believe that PEs were over 200 back in March of 2009. PEs were indeed over 200 by some calculations but that was because of one-time writedowns. Sifting through the various PE calculations requires some sophistication.

During bottoms there are a ton of smart analysts who write convincing articles about how PEs are rich, and the opposite is true at peaks.

Graham came up with mechanical ways to buy low and sell high, like automatically rebalancing when stocks become too big a portion of the portfolio. I think that’s reasonable, but I wanted to keep my template as simple as possible, and I don’t think all that much is lost (after tax considerations and transaction costs) by not rebalancing.

An even simpler form (comparable to old adage of 50% bonds 50% stocks) might look like this:

Putting starting prices and short-term economic views aside, gold is the least attractive asset (compared to stocks and bonds). This has been true over the last thousand years, 150 years, and 50 years, and I would wager it would be true over the next 50 as well. It should be viewed as an alternative currency and used for diversification and protection again hyperinflation risk. It is not a good hedge against moderate inflation, since stocks and high yield bonds tend to outperform gold under those conditions (both empirically and conceptually).

What template would you create if you had to present it to the world and then go into a 100 year coma, and 75% of all investors would follow it exactly for 100 years?

Subject: RE 3: Long-term portfolio for unsophisticated investors

I guess I don't think complete dunces should be "investing", just as they shouldn't be practicing medicine on themselves (it's as dangerous to one's financial health as it is to operate on one's own kidneys or heart). I don't think one needs to follow markets and economies daily, but without basic views/knowledge about the world, it's perilous to invest without an advisor (a good advisor is worth it). The world and markets are too dynamic for a 100-year mechanical allocation (even Graham's rules are obsolete and have been for 30 years).

Also, your 100 year hypo is an unrealistically long time horizon. Over the last two hundred year periods (1800-1900, 1900-2000), a diversified basket of equities in the safe, English-speaking countries (US, Canada, UK, Australia) did the best by far. If I were forced to, I would do an 90/10 stocks/bond basket in "safe" countries for a 100 year horizon (the 10% bonds are "what-if" safe money). But if a longer sample of history is any guide (500BC to 2000AD, the recorded history of the West), then the lesson is that everything is in flux, and no rule can work. You have to adapt and move to the countries where economic growth and opportunities are, following booms and busts with care. So there is no 100-year buy and hold - as your time horizon gets longer, everyone becomes a trader.

Markets are humbling. There is no such thing as a "smart" analyst. If you're right 60% of the time, that's good enough to make money. 70% (the Soros/Buffett/Klarman league) is superhuman.

From: VEGA
Subject: RE 4: Long-term portfolio for unsophisticated investors

I think it may actually be impossible, since we simply don’t have access to advisor track records. Even of institutions, its hard to overcome reporting bias (e.g. a mutual fund family is constantly closing unsuccessful funds), lottery sale bias (i.e. a manager can generate an impressive track record selling downside puts). If someone generates accurate macroeconomic predictions over 5 years, odds are they just flipped heads a few times. If they do it over 20 years, we couldn’t afford their services.

If it is at all possible to identify a good advisor, I think it would require such skills and knowledge that the advisor would be superfluous. Investing with the smartest guy you know doesn’t work, nor does investing with the most respected.

Subject: RE 5: Long-term portfolio for unsophisticated investors

I agree. Tough to do. Talent evaluation is very hard. Yet many people with common sense (Ann Landers) bought BRK shares in the 1970s and 1980s and did quite well. Bill Ruane and other Buffett/Graham associates have beaten the benchmarks. I would bet on El-Erian and Berkowitz outperforming their benchmarks over the next 20 years, assuming their fund sizes don't balloon too much (like Gross's Total Return Fund). Yet talent can get de-motivated by wealth, cars, toys, mistresses, divorces, a fawning press, etc. You're right that it takes a great manager to recognize other great ones (e.g. Steindhart, Greenblatt, and Julian Robertson are good pickers of talent).

The qualities I look for:
-an even and contrarian temperament and open mind willing to change quickly
-smarts but more specifically a shrewd common sense that understands human nature and institutional logic/constraints
-mathematical fluency and the ability to deal with opposing probabilistic views, tail probabilities, and so on
-frugality in life and a sense of purpose/values (heavy drinkers/drug users/stressed people don't last) AND
-finally a deep understanding of economies, markets, businesses, and financial statements (usu. people only consider the last element).


No comments:

Post a Comment