Thursday, August 30, 2007

Tax Loss Harvesting in the Mutual Fund and ETF Context

This is to introduce the idea of tax loss harvesting with mutual funds and ETFs. I'm not a tax professional and you're fully responsible for the consequences when following this practice.

1. What is tax loss harvesting?

Basically sell your underperforming funds to realize short-term capital losses and get tax deductions in your tax return (a discount to your loss).

2. What is the risk with tax loss harvesting?

Market risk. If the fund recovers after you sell your shares, you lock in your losses and miss the gains.

3. How do you avoid the risk?

Don't lose your market position. Instead find a replacement fund to hedge the price movement risk. However, there is a catch: the IRS has a "wash sale rule" to specifically disallow fully hedged tax loss harvesting. Basically the wash sale rule says you can't claim a loss on a stock/fund sale if you bought a substantially identical stock/fund within 30 days before or after the sale. It's not clear how IRS interprets "substantially identical" in the MF/ETF context. So this is a currently a grey area.

4. How do you hedge the risk without using "substantially identical" funds?

There are a lot of funds that are not substantially identical in terms of holdings. But when it comes to price movement, they are closely clustered.

For example, are FXI/PGJ/GXC/MCHFX substantially identical? I don't think so, because their holdings are so different. Yet their price movements are almost the same.

Another example, are VFINX/VLACX/VTSMX substantially identical? This one is harder to tell, since VFINX is a dominant subset of VLACX, which is then a dominant subset of VTSMX. However, I think it's safe to say that VIVAX or VIGRX is not substantially identical to VFINX/VLACX/VTSMX, yet their price movements are similar.

5. Is tax loss harvesting worth doing?

Absolutely. This is one of the very few ways you can improve upon the market average return. Let's do some math here.

Marginal income tax rate: 28%
Long-term capital gain (LTCG) tax rate: 15%

For each $100 loss you claim, you get $28 back from tax return. If you replaced your fund with another one, you basically lowered the cost basis by $100 for your market position. So you will eventually pay, hopefully, the LTCG tax on this $100 difference, which will be $15 (assuming tax rates don't change).

The best thing is that you can defer this $15 LTCG tax as far as you want, when the currency is seriously devalued. For example, if you hold the fund for 10 years while you invest the $28 tax refund in something that gives you a 5% after-tax annual return, your $28 will grow to 28*1.05^10 = $46, and you only need to pay $15 back in the end.

The saving is even more if you currently have a high state tax (e.g. California's 9.3%) and there is chance that you will move to a low tax state in the future.

6. How to implement tax loss harvesting with MFs/ETFs?

Implementing it systematically is pretty hard. Some of the difficulties are:

1) Mutual funds don't support tax lot sales, so you can't sell your high cost basis shares only.
2) Most brokers don't let you select tax lots of ETF sales either (they usually assume FIFO), but it appears that you can do it in the tax return, which is probably very troublesome.
3) If the price does move up during the 30 day period after you swapped your funds, you're stuck with the replacement fund if you want to take advantage of the LTCG tax rate. So you have to make sure your replacement fund also matches your long term investment goals.
4) There is commission costs and bid/offer spreads if you trade ETFs.

That said, the recent market downturn provides a perfect opportunity to effectively implement tax loss harvesting, especially for these who bought new funds near the market peak (me included).

7. Conclusion

Tax loss harvesting is a useful strategy, especially in the MF/ETF context, where you can be well hedged against the price movements during the wash sale blackout period. However, one has to play the game very carefully and one mistake could wipe out the gains (e.g. you do accidentally buy a
substantially identical fund, even in your retirement account). Again, this is a grey area of IRS. Only play the game if you know what you are doing and can afford the consequences.

8. References
Google Search

Tuesday, August 28, 2007

David Swensen on Core Asset Classes vs. Non-Core Asset Classes

Over the weekend, I finished reading "Unconventional Success: A Fundamental Approach to Personal Investment" by David Swensen, the chief investment officer of the Yale University endowment. In his book, Swensen discusses asset allocation, a very important topic for long-term investors. I found his argument on the core asset classes vs. the non-core ones particularly useful. Specifically, Swensen believes the following ought to be the investor's core asset classes.

Domestic Equity
Domestic equities provide high return and good protection against inflation in the long run. The corporate management often serves shareholders' interests reasonably well.

U.S. Treasury Bonds
Treasury bonds are "risk-free" and provide reliable diversification and excellent hedge against financial crises. The government is a market-neutral player.

U.S. Treasury Inflation-Protected Securities
TIPS are also "risk-free" and provide the most reliable protection against unexpected inflation. The market size of TIPS is about one tenth of that of treasury bonds.

Foreign Developed Equity
Foreign investments have similar expected returns as U.S. equities. In the long run, foreign equities also provide good diversifying power, although in the short term the global market often reacts to financial crises synchronously.

Emerging Market Equity
Emerging markets are substantially more risky and compensate with higher expected returns (risk premium). A drawback is that corporates in emerging markets are often less well regulated and can harm foreign investors' interests.

Real Estate
Real estate investments combine bond-like returns from rental income and equity-like returns on the residual value. Real estate provides great protection against inflation due to the strong correlation between inflation and the replacement cost of real estate (intrinsic value).

Swensen recommends using the above core asset classes to construct the following example portfolio:

Domestic equity30%
Foreign developed equity15%
Emerging market equity5%
Real estate20%
U.S. Treasury bonds15%
U.S. Treasury Inflation-Protected Securities15%

On the non-core asset class part, Swensen also provides reasons why the following should not be the investor's core holdings.

Domestic Corporate Bonds
They suffer from credit risk, illiquidity and callability, as well as misaligned interests with the corporate management and shareholders.

High-Yield Bonds
These are the so-called junk bonds. They magnify corporate bond's drawbacks even further and the extra return does not provide adequate compensation.

Tax-Exempt Bonds
These municipal bonds, especially long-term ones, suffer from credit risk and call options. Their value could be heavily altered by marginal tax rate changes, either positively or negatively. However, short-term tax-exempt bond or money market funds are worth consideration.

Asset-Backed Securities
Asset-backed securities are highly complex products created by those most sophisticated Wall Street financial engineers. The complicated structures do not serve investors' interests. Investors do not receive adequate rewards for accepting credit and call risks. The government-sponsored enterprises (GSE) cause false assumptions on their credit attributes.

Foreign Bonds
Although foreign bonds might have different interest rates from U.S. bonds, the end result (of unhedged foreign bonds) is equivalent to holding U.S. bonds plus some currency exchange exposure. Foreign currency exposure, while providing some diversification, has no real returns.

Hedge Funds
Although some hedge funds produce good returns, it is too difficult to identify them beforehand. The hefty fee arrangements erode returns enormously.

Leveraged Buyouts
Leveraged buyouts incur a higher degree of risk (highly leveraged) and hefty fees (management fees and profit share) without delivering satisfactory risk-adjusted returns. The fund managers' motivation of seeking high returns diminishes when the fund becomes large.

Venture Capital
Over long periods of time, the return of venture capital investments as a whole is similar to that of the public market investment, failing to compensate the extra risk involved. The top-tier well-established firms often have superior access to deals, entrepreneurs and capital markets and provide higher returns. However, such successful firms are usually closed to new money, leaving only unattractive choices to investors.

Sunday, August 26, 2007

Vanguard's U.S. Index Funds

Vanguard is the market leader in providing index funds to retail investors. Understanding what their offerings really track is important when constructing a portfolio. Here is my summary of their U.S. index funds. Perhaps the biggest surprise is that the mid-cap index fund is actually a subset of the large-cap index fund! Also I truly wish Vanguard could offer a micro-cap index fund like DFA (Dimensional Fund Advisors) does.

500 Index Fund (VFINX)
Index: S&P 500 Index
Constituents: 500 largest companies
% of total market cap: 75%

Extended Market Index Fund (VEXMX)
Index: S&P Completion Index
Constituents: total market - 500 largest companies
% of total market cap: 25%

Large-Cap Index Fund (VLACX)
Index: MSCI US Prime Market 750 Index (MSCI US Large Cap 300 Index + Mid Cap 450 Index)
Constituents: 750 large and medium-size companies
% of total market cap: 86%

Value/Growth Index Fund (VIVAX/VIGRX)
Index: MSCI US Prime Market Value/Growth Index
Constituents: value/growth half of the 750 large and medium-size companies

Mid-Cap Index Fund (VIMSX)
Index: MSCI US Mid Cap 450 Index
Constituents: 450 medium-size companies
% of total market cap: 15%

Mid-Cap Value/Growth Index Fund (VMVIX/VMGIX)
Index: MSCI US Mid Cap Value/Growth Index
Constituents: value/growth half of the 450 medium-size companies

Small-Cap Index Fund (NAESX)
Index: MSCI US Small Cap 1750 Index
Constituents: 1750 small companies (total market - 750 large and medium-size companies - micro-cap companies)
% of total market cap: 12%

Small-Cap Value/Growth Index Fund (VISVX/VISGX)
Index: MSCI US Small Cap Value/Growth Index
Constituents: value/growth half of 1750 small companies

Total Stock Market Index Fund (VTSMX)
Index: MSCI US Broad Market Index (MSCI US Large Cap 300 Index + Mid Cap 450 Index + Small Cap 1750 Index + Macro Cap Index)
Constituents: 3000+ companies
% of total market cap: 99.5%


Saturday, August 25, 2007

Investing in China: Fund Choices

As a Chinese citizen, I have a natural tendency of investing in China. Here is my summary of the representative funds available to investors in the U.S.

FXI (iShares FTSE/Xinhua China 25 Index)
Expense Ratio: 0.74%
Turnover Rate: 45%

Index fund tracking the FTSE/Xinhua China 25 index (25 largest Chinese companies traded in Hong Kong), weighted on their float-adjusted market caps, with a cap of 10% on the largest holdings.

Index fund, lower cost than actively-managed funds. Highly popular and liquid.

Highly concentrated on state-owned giant companies. The turnover of 45% is very high for an index fund, probably because the 10% weight cap, which China Mobile is currently hitting, requires frequent rebalancing.

PGJ (PowerShares Golden Dragon Halter USX China)
Expense Ratio: 0.70%
Turnover Rate: 17%

Index fund tracking the Halter USX China index (Chinese companies listed in the U.S. as ADRs).

Index fund, lower cost than actively-managed funds. Underlying traded in the more mature U.S. market. Less risky.

Because the underlying is traded in the more mature and less risky U.S. market, its return could be lower too. Plus it's correlated to the U.S. market, so it does not provide very good diversification.

GXC (SPDR S&P China)
Expense Ratio: 0.60%
Turnover Rate: unknown, presumably very low

This is a new fund tracking the S&P Citigroup BMI China Index. This fund is the broadest so far, with about 200 underlying holdings traded either in Hong Kong or the U.S., weighted on their float-adjusted market caps. (It's not quite clear where they buy shares of these companies traded in both Hong Kong and the U.S., e.g. China Mobile.)

Broadest index fund available. Could be the best choice for indexers and the core China holding for everyone.

Although not as highly concentrated as FXI, this fund is also dominated by giant companies. For example, it has 12.5% of its asset in China Mobile. Not as liquid as FXI.

CAF (Morgan Stanley China A Share)
Expense Ratio: 1.98%
Turnover Rate: 14%

Actively-managed and closed-end fund investing in China A shares (companies listed in Shanghai and Shenzhen) through QFII (Qualified Foreign Institutional Investor) quota.

The only fund available to U.S. investors that invests directly in the China domestic market. Least correlated to the U.S. market. Excellent diversification. Low turnover.

Investing directly in China domestic market could be very risky. The market valuation is very high (P/E ratio). Being a closed-end fund could be risky too. However, the current heavy discount (15%~20%) provides some cushion for risk. High expense ratio, due to the lack of alternative funds.

MCHFX (Matthews China)
Expense Ratio: 1.26%
Turnover Rate: 12%

Actively-managed mutual fund investing in large companies in China and Hong Kong.

Low turnover, good for tax-efficiency. Not as concentrated on giant state-owned financial and energy companies as FXI.

Low turnover, which makes you wonder if the manager actually does anything.

OBCHX (Oberweis China Opportunities)
Expense Ratio: 1.91%
Annual Turnover: 53%

Actively-managed mutual fund investing in small to mid-size companies in China, Hong Kong, Taiwan and Singapore.

Exposure to small and mid-size companies. Can be used to complement other large-cap funds.

Could be more risky than large-cap funds. Never use it as a core China holding. Also, because of the lack of similar mid/small-cap indexes/funds, its performance is hard to judge. High expense ratio.

Friday, August 24, 2007

Hello World!

Hi, my name is Yi. I'm a 29-year old working in Silicon Valley, California. This is my first ever blog and I dedicate it to my investing journey which started in January 2007. You will read my investment ideas and analysis along the journey. I hope you will find useful stuff here. Enjoy!