"Investor's Manifesto:" Chapter 6 Notes

Continuing our detailed review of "The Investor's Manifesto" by William J. Bernstein, we're going to look at Chapter 6, "Building Your Portfolio."

This chapter boils down to one sentence of advice from Mr. Bernstein: "Save as much as you can, start as early as you can, and do not ever stop."

How Much Do You Need to Retire?

Mr. Bernstein advises looking at fixed annuities if you face the possibility of running out of money before you die. How to judge if you're going to run out of money?

My rule of thumb is that if you spend 2 percent of your nest egg per year, adjusted upward for the cost of living, you are as secure as possible; at 3 percent, you are probably safe; at 4 percent, you are taking real risks; and at 5 percent, you had better like cat food and vacations very close to home. For example, if, in addition to Social Security and pensions, you spend \\(50,000 per year in living expenses, that means you will need \\)2.5 million to be perfectly safe, and \\(1.67 million to be fairly secure. If you have “only” \\)1.25 million, you are taking chances; if you are starting with \$1 million, there is a good chance you will eventually run out of money.

That's grim news for anyone with less than 1 million in savings. And frankly, that's almost everyone. Luckily, annuities provide some help. When you pay into an annuity, you get some percentage (6 to 8% per Bernstein) annually, and you're guaranteed[1. Assuming, of course, that the insurance company doesn't go broke.] that you'll receive that money each year until you die. Once you die, the money stops, and you never get your principal back, so it's a poor choice for those who want to leave a legacy for their children

Investing and Dollar Cost vs Value Averaging

On to investing. Bernstein suggests avoiding ETFs in favor of traditional open-ended mutual funds. He includes a table of suggested index funds (and a few ETFs) at Kindle location 2261.

Next he discusses how to save -- Dollar Cost Averaging vs. Value Averaging.

Dollar Cost Averaging is when you put the same amount each month into your investments. (Which is, of course, what you should do when you're young and saving.)

Value Averaging is more appropriate to when someone has a large sum of money that they want to deploy into their investments. Instead of buying in one lump purchase, they buy roughly the same amount each month -- but adjusted so that you're buying more of falling funds and buying less of rising ones.

Frankly, Bernstein gives very little detail about DCA, I think because he's gun-shy about putting math in his books, and so we'll have to look to Edleson's Value Averaging book for more details. It's on our short list of future reading.

Sample Allocations

Now to a very fun part of the chapter. Bernstein gives sample allocations for four prototypical investors.

Young Yvonne (No assets, just starting to save)

  • 25% Vanguard Total Stock Market Index
  • 25% Vanguard Total International Index
  • 50% Vanguard Short-Term Bond Index

This gives a 50% allocation to bonds, which is high for someone of Yvonne's age, at least according to the "keep your bond percentage equal to your age" school of thought. This is because Yvonne is a "market virgin," and has never experienced a down market.

Bernstein emphasizes that the bond portion of a portfolio is for safety and security, not really for asset growth, and advises short-term, high-quality bonds.

Sheltered Sam (All assets in retirement accounts)

Sam has little savings outside his retirement accounts, where he as \$500,000. Since his assets are in tax-advantaged savings, he can rebalance without worrying about tax implications, and he can own any asset class without worrying about its tax efficiency.

Sam holds the following funds, all of which (except for the final one, his money market account) are in his IRA:

  • 10% Vanguard Large-Cap Index
  • 12% Vanguard Value Index
  • 3% Vanguard Small-Cap Index
  • 8% Vanguard Small-Cap Value Index
  • 4% Vanguard REIT Index
  • 2% Vanguard European Stock Index
  • 2% Vanguard Pacific Stock Index
  • 3% Vanguard Emerging Markets Index
  • 3% iShares MCSI Value Index
  • 25% Vanguard Short-Term Investment Grade Bond
  • 20% Vanguard Inflation-Protected Securities
  • 5% Money Market (taxable)

Sam follows the "bond percentage equal to age" rule, having 45% of his portfolio in bonds.

Taxable Ted (All assets in a taxable account)

Ted had no savings, then came into a seven-figure windfall. Because of tax considerations (which Bernstein sadly does not go into), Sam is limited to three stock asset classes -- large-cap US markets, large-cap foreign markets, and US small-cap stocks. Because Ted already has enough to retire on, he doesn't need to take many risks -- he's going with a 40/60 stock/bond split.

Because he's a California resident, he's using a tax-exempt California bond fund, but because he's concerned about California's financial future, he's diversifying his bond holdings.

  • 16% Vanguard Total Stock Market Index
  • 10% Vanguard Tax-Managed Small-Cap
  • 8% Vanguard Tax-Managed International
  • 2% Vanguard Emerging Markets
  • 4% Vanguard REIT (Variable Annuity)
  • 15% Prime Money Market
  • 15% California Intermediate-Term Tax Exempt
  • 15% Limited-Term Tax Exempt
  • 15% Short-Term Investment Grade

In-Between Ida (Assets Divided Between Retirement and Taxable Accounts)

Ida is in theory a good candidate for an inflation-adjusted annuity, but Bernstein discounts this -- he's not confident enough in the financial health of insurance companies to advise someone to put their next egg in an annuity right now.

Ida is highly risk-tolerant, and will go for an aggressive (for a 70-year-old) 50/50 stock/bond split.

  • 10% Vanguard Total Stock Market (Taxable)
  • 8% Vanguard Value Index (IRA)
  • 3% Vanguard Tax-Managed Small Cap (Taxable)
  • 6% Vanguard Small-Cap Value (IRA)
  • 5% Vanguard REIT (IRA)
  • 5% Vanguard Tax-Managed International (Taxable)
  • 6% iShares MCSI Value Index (IRA)
  • 3% Vanguard Emerging Markets (Taxable)
  • 4% Vanguard International Small-Cap Index (IRA)
  • 20% Vanguard Short-Term Investment-Grade Bond (IRA)
  • 15% Vanguard Limited-Term Tax-Exempt Bond (Taxable)
  • 10% Vanguard Ohio Tax-Exempt Bond (Taxable)
  • 5% Money Market (4% in taxable, 1% in IRA)


Bernstein talks more about the tax implications of rebalancing:

Investors rebalance portfolios for two reasons: to enhance return and to reduce risk. The excess returns generated by rebalancing are not large, usually no greater than 1 percent per year, which is much smaller than the capital gains taxes you will realize on most sales. So purely from a returns point of view, you should never sell stocks to rebalance inside a taxable portfolio. Buying is fine, of course, and you can also use fund distributions—the capital gains, dividends, and interest the funds throw off—to rebalance as well.

Bernstein advises rebalancing tax-sheltered accounts every 1 or 2 years, and taxable accounts every 2 or 3 years.

I've been reading about Opportunistic Rebalancing1 and I found it fascinating. I had forgotten that Bernstein talked about it here, and was fascinated to see his take on it:

Some asset classes are more volatile than others; set a 20 percent threshold for U.S. large cap stocks, and you will hit it every year or two. With emerging markets, you will be trading much more frequently. Finally, the thresholds will have to be portfolio-specific; using a fixed threshold in an all-stock portfolio will result in far fewer rebalances because all stock asset classes move together than in a nearly all-bond portfolio, where the stock moves relative to the bonds will be much larger.

Which is better, calendar or threshold rebalancing? It is impossible to determine, since the benefits of rebalancing are not much greater than 1 percent per year; proving that one method was better than another would mean statistically powering the test to detect differences in return of perhaps one or two dozen basis points, which would probably take hundreds or even thousands of years of data.

Because of these complexities, I recommend that beginners stick to rebalancing by the calendar once every few years or so. If at some point you do decide to switch to the threshold technique, you will need to develop individual rebalancing parameters that are not only asset-class specific, but also portfolio-specific as well. Threshold rebalancing for most practitioners tends to be a work in progress. Rebalancing a given asset class too often or too infrequently is usually a signal to adjust the threshold up or down, respectively.

  1. See also this Bogleheads Discussion.