SDMB Investing pros: Help my select my asset allocation plan!

In My Humble Opinion
1

Here’s the deal–I’m a long-term investor–this is all retirement money. I’m 33 years old and fairly risk tolerant. I can absorb a 30% loss in any given year and not worry about it too much. I’ve been gleefully dollar cost averaging into the market for the last two years, maxing a Roth IRA and my 401(k).

I’m roughly hip to Modern Portfolio Theory. I’m hip to Markowitz. I’m hip to William Bernstein, author of The Intelligent Asset Allocator. (interested lurkers, check out his website, www.efficientfrontier.com. These ideas really turned my head!)

I want to be surfing the efficient frontier. I want a rational portfolio of stocks, bonds, and other asset classes who are not highly correlated with one another, so their volatility tends to cancel out. I don’t want to try to select hot active money managers–I don’t believe I can reliably identify IN ADVANCE money managers who will outperform their benchmarks. I don’t think too many other people can, either.

I might have to go with active management in my 401(k) plan, though, which is with Fidelity, simply because some asset classes will not be represented by indexes therein. My Roth IRA is with Vanguard.

I am subject to the following restrictions:

All told I have about 14,000 dollars right now between the two accounts to play with. $4,500 in the Vanguard account–all in the Vanguard Total Stock Market Index. Vanguard has a $1,000 account minimum in retirement accounts, so I cannot asset allocate below that number

The 401 account is now 40% in the S&P 500 index, 30% in domestic midcap stocks, and 30% in domestic small (value) (Actually, in FLPSX, which is getting midcappy.)

I have zero exposure to bonds at this point (yields are so low, I’m not in a rush.) Zero to REITS. Zero to international markets, although I’m very interested in gaining exposure there, both in Europe and Pacific Rim, and in emerging markets.

My question, fellow dopers:

 What combination of asset classes is optimal over the long term, and what indexes would you use to track them? Anyone have access to an optimizer?

Having selected the optimal portfolio, how shall I balance it with a risk free investment to nail down an expected standard deviation of, say, 13 to 15 (somewhat less volatile than the S&P 500.

What is the optimal period for rebalancing?

Opinions? Theories? And are you doing it with your own money? Your clients’ money?

2

Do a Warren Buffet.

3

Which means what? Pick stocks that will go up? Be my own active manager? Easier said than done, isn’t it?

Nope, statistically, that doesn’t work. There’s only one Warren Buffett. I have tremendous respect for the man, but for all I know he’s little more than the winner of a coin flipping contest. There are tens of millions of investors out there. Someone is going to flip ‘heads’ 30 times in a row.

Why should I believe I can do the same thing? I don’t have his time nor expertise.

How can I, with just 14 grand, build an adequately diversified portfolio, representing a variety of asset classes, and push it out to the efficient frontier? As a small investor, I can’t trade in bulk. Fees would kill me.

Nah.

Read one book by John C. Bogle and one by William Bernstein and one by Charles Ellis and call me in the morning. :slight_smile:

Next?

4

The problems you run into when you attempt to design a portfolio to have a specific standard deviation are that

  1. Standard deviation alone does not provide a complete picture of risk (e.g. value stocks generally have a lower standard deviation than growth stocks, but carry a higher risk–hence the value premium) and
  2. The distribution of outcomes in the financial markets is not mesokurtic (normal) but rather is leptokurtic, meaning that the “bell curve” has “fat tails”. This means that events that would appear rare using a simple statistical analysis can occur much more frequently:

Efficient frontier models are constructed using historic data of returns, correlations, and standard deviations for various asset classes. The results obtained are extremely sensitive to these inputs, and history has shown there are long periods of time where the past is not prologue (e.g. the 16-year period where the S&P 500 returned –0.9% annualized, inflation-adjusted). While I believe MVO’s are a useful planning tool, I wouldn’t put too much faith in these models.

For Vanguard funds, Adviseronline.com has a correlation tool that may be helpful to you.

As for my portfolio, it is 100% passively managed, mostly through Vanguard index funds. I’ve got around 30% S&P 500 Index (401(k)), 25% S&P SmallCap 600/BARRA Value Index (Roth IRA), 25% CRSP Decile 10 Ultra-Small Company Index (taxable), 10% MSCI Pacific Free Index, and 10% MSCI Emerging Markets Free Index. If I had better options in my 401(k) this might look very different, as I am trying to find ways to increase my value/international exposure. But I have an unusually high tolerance for risk.

Theoretically, there should be an optimum rebalancing interval because rebalancing always involves buying low/selling high, but asset class price movements do carry some momentum which would mean selling a winner/buying a loser. I don’t know of any studies that are conclusive, so I would probably do it about once a year, if for nothing more than convenience.

5

Hickory6 I’m strictly an amatuer, but you seem to have a pretty good grounding in theory. Your mention of books by Bogle, Berstein, and Ellis would lead me to believe that you might be a devotee of Bob Brinker’s.

I think your self-proclaimed ability to accept a risk factor of 30% down in one year doesn’t bode well for the future, although you seem to have everything else under control.

I’m surprised that you don’t want some bonds in your diversified portfolio. Strictly Gov’t backed stuff, of course.

My personal opinion is that the Dow(and also the NASDQ, etc) is gonna bottom out about 20-30% lower than they are now. This may or may not happen this year. If I’m right, then you would be wrong to park your money in anything other than a fixed instrument, no matter what yield it has. If, at the end of the coming recession/deflation, you have your principal intact, you will be ahead of those who stayed in equities. But that may just be my risk-intolerance talking.

6

I’m familiar with Bob Brinker’s MarketTimer newsletter, although I don’t want to do much market timing myself in the active sense. I just don’t feel I can guess right 30 times in a row.

I do plan on moving into bonds, soon, as well, too. But with yields as low as they are, I’m just not in a rush to do so. When prospects for equities and bonds even out a bit more, say, when I can get 7% out of investment-grade paper, I may move to something like an 80-20 or 70-30 split for now. That’s my target application. I’m just not willing to walk into a buzzsaw to do it. Ok, that’s market-timing, too.

When I do go into bonds, if I have enough money to split it up, I’m willing to have exposure to foreign government debt and U.S. corporate debt --both investment grade and high-yield, as well as Treasuries. For Treasuries I like TIPS, conceptually. If I have to choose, then I will chose the asset class with the lower correlation to the rest of my portfolio.

Anyone have experience with DFA funds? Cynic, is that where you have your CRSP Decile 10? (I love that indexing concept, BTW. I wish Vanguard did something similar. Maybe big fund companies can’t pull that off, though. Too many assets to sling around.

And is there a reason you skipped over Domestic midcaps? Was that strictly a correlation decision? And you don’t mention bonds. Are you thinking the same way I am about that?

7

At only $14k, I would watch how diversified you get, especially considering costs and your ‘risk tolerance’. If you are looking at this as retirement money, and you have your tax bases covered, then take some larger bets in fewer positions. If you really mean that this money is 30 year plus money, and you have tolerance for risk- buy straight stocks with future money. Are you allowed to buy straight stocks with your 401 or Vanguard? I deal with international investing, I have no clue how my home country’s system works! I’m not allowed to open an IRA since I don’t have US income…friggin fraggin…grrr…

Don’t try rebalancing everything and shifting money around too much, that is costly. The logic behind this is simple- there are a lot of fixed costs and $14k isn’t that much…don’t eat things up with costs (and I’m talking all costs- entry/exits fees, manager fees, the $25 wire transfer, the FedEx, long-distance telephone calls, etc.). Keep it simple when dealing with small amounts.

If you can risk losing 30% a year and have a long term horizon…sheez! NOW is the time to be buying stocks. Buy when blood is running in the streets! Just a few, a little bit each month for the next 6-12 months. No one knows what will happen in 30 years, but if history is a guide, stocks will outperform bonds.

I’m not 100% positive, but I am pretty darned sure that IBM, Microsoft, McD’s, Disney, GM, ??? will still be around in 30 years. If you put $500-$1000 a month into a few stocks over the next 6 months buying the dips (to spread your risk in a bear market), in 30 years you’d be fine. In that time 1 will bomb, 2 will do OK, and 1 will do great. Your total costs for that whole time? $7.50 per trade plus taxes. No yearly fees or management costs, just sign up at Etrade or something and do it yourself.

But really, whatever you do, don’t get too fancy. You start juggling too much and costs are going to eat into your profits. My small office does the consulting for a South African company on just these things. We are on the leading edge of ‘Strategic Asset Allocation’ - it’s like portfolio optimization on speed. The main thing we struggle with? Costs. And we get fee reductions because we’re managing $1b. There is a point at which costs eat into your profits more than diversification will save you. You got $14k…watch your costs.

-Tcat

8

Bridgeway has the only Ultra-Small Company Index fund that I’m aware of. This fund is truly unique: CRSP 10 tracks the smallest 10% of all publicly traded U.S. stocks, which only account for about 0.4% of the market capitalization of the entire stock market. It has an R-squared of about 0.5 with the S&P 500, so it’s a good diversifier. Although the turnover is rather high for an index fund, you don’t need to hold it in a tax-deferred account because it is tax-managed.

Here are the historic (1926-1996) returns for the CRSP Deciles 1 and 10:

CRSP Decile 1: 11.56%
CRSP Decile 10: 19.82%

Of course, this fund has very high risk/high volatility, and therefore is not suitable for money that would be withdrawn in less than 10 years or so.

As far as mid-caps are concerned, I really don’t see any need for them in a portfolio that is already diversified with both large- and small-caps. Their returns almost always fall somewhere in between those of large- and small-caps for periods of three years or longer, so you’re better off without them (more diversification) as long as you rebalance periodically.

I like TIPS too, but I don’t have any room for them in my tax-deferred accounts. I could hold Series I or Series EE bonds, but with yields so low right now I’m not all that enthusiastic. But then again, valuations in the stock market aren’t exactly cheap, either. Eventually I’d like to have about 20%-30% in bonds, maybe half in TIPS and half in short-term corporate. But I haven’t yet worked out all of the details.

9

I forgot to mention that these are the arithmetic returns, which can be a little misleading. Here are the geometric returns (again from 1926-1996), taken from Jeremy Siegel’s Stocks For The Long Run:


**
Decile     Return  Standard Deviation

CRSP 1     9.84%       18.9%
CRSP 2     11.06%      22.4%
CRSP 3     11.49%      24.2%
CRSP 4     11.63%      26.7%
CRSP 5     12.16%      27.5%
CRSP 6     11.82%      28.5%
CRSP 7     11.88%      31.0%
CRSP 8     12.15%      34.8%
CRSP 9     12.25%      27.3%
CRSP 10    13.83%      46.5%
**

The CRSP Decile 1 is very similar to the S&P 500 Index, so it appears that small-cap premium has been just under 4% for CRSP 10.

10

Tomcat’s ideas are closest to my own experiences - buying a selection of shares now and sitting on them. With the amount of money you’re suggesting, you’d have no trouble worrying about not being diversified enough and could probably include one or two foreign companies to diversify even more.

I’m a similar age to you, so probably have a tolerance of risk that’s not too different either. I got interested in buying shares directly once I started looking closely at the fees that Fund Managers charge.

11

With 14,000 and dollar cost averaging again with my meager pittance of a wage, I cannot efficiently pay commissions to buy a selection of single stocks.

Again, I also do not want to be in the business of researching individual companies. Most pros can’t do that and beat the market by more than their costs–why should I? I have another job I have to do all day. I have other things to do than be a full time stock researcher, and I have a better understanding of funds and fund selection than I do of their underlying securities.

And it is precisely BECAUSE I have an understanding of fund selection that I do not believe I can reliably predict, in advance, fund managers which will beat the market. Past returns are not reliable predictors. Past alphas are not even predictably reliable, after accounting for the streakiness of indexes.

I would much rather pick asset classes and keep fees low. Everything I do is no load, and with index funds, low recurring expenses. Might be able to go even lower eventually with ETFs and similar vehicles.

I generally have no exchange fees, so I don’t have to worry about the trading costs of rebalancing annually. Sure there are hidden costs, but as a fund shareholder, I pay them whether I rebalance or not. But those costs are minimal, anyway, within the same fund company, as the funds will often transfer assets between themselves rather than go through a broker.

12

I like Bridgeway, and John Montgomery’s great.

Unfortunately, Bridgeway Ultra-Small is closed to new investors.

I do like Dimensional Fund Advisors, though, too, which has a similar U.S. Micro Cap fund–which buys the smallest 4% of companies based on market cap. Turnover is always high down here, but in an IRA that doesn’t matter to me. Trading costs matter, though. And in this one, I’d need to go through an advisor, so there would be some fees somewhere. But the .56% expense ratio’s great!

13

Oh, R-squared to the S&P is just 28 for the DFA Micro cap fund. So it’s a fine diversifier.

I sound like an ad for DFA funds, here. <G>