It's a common financial advice to buy an index and hold it forever. Particularly, John Bogle endorsed buying an S&P 500 index and just hold it / buy more of it during bear markets so you get rewarded when the market recovers.

However, if someone applies the same principle to the Nikkei 225, 21 years later, the market still hasn't recovered. Maybe, a 21-year time frame is still too short but for an average investor's lifetime, it's already quite long.

If such is the case, how should one structure a portfolio given such possibility, however remote, can happen to any US total stock market index?


4 Answers 4


Here are a few things that you could try. But note that they are all capable of failing. They will just reduce the chance of you personally having a lost decade. First a quibble: John Bogle advocates a total stock market index (something like Vanguard's VTSMX) instead of an S&P fund, as the latter represents "only" 85% or so of the US market's total capitalization. Smaller companies behave slightly differently than members of the S&P, so this might provide a small help. Bogle also advocates holding some bonds in addition to equities. I'll expand on that below.

Account for dividends. Just because the value of the index is the same as its value 10 or 20 years ago doesn't necessarily mean that decade was lost. The companies in the S&P are currently paying out an annualized dividend of about 2%. Even if the actual value of the index doesn't change, you're still getting that 2% per year.

Include bonds. As I mentioned above, Bogle recommends holding some bonds. I have seen two common rules. One is to never have less than 20% of your total holdings in bonds, and never have more than 80% of your total holdings in bonds. The other popular rule is to hold your age in bonds. For example, I'm about 30, so I should keep about 30% of my holdings in bonds. Regardless of the split, rebalance periodically to keep yourself at that split. What effect would holding bonds have on a lost decade? To make the math easy, let's say you split your holdings evenly between an S&P fund and 10 year Treasuries. Coincidentally, 10 year T-notes have the same 2% yield as the S&P dividends. If you're getting that on half your holdings, and nothing on the other half, you're netting 1% per year. Not great, but not totally lost. To illustrate the effect of rebalancing, use my example of a 70/30 stock/bond split. The S&P lost about 50% of its value from its peak to the bottom of the market in early 2008. If you only held stock, you would need the market to increase in value by 100% in order for you to recover that value. If 30% of your holdings are in bonds, and you rebalance at exactly the bottom of the stock market, you only need the stock index to increase in value by about 80% from the bottom in order to make you whole again.

I mention those two to emphasize that your investment return is not just a function of the price of a stock index.

Dollar cost average. It's rare that you will actually face the situation of putting (say) $100,000 into the market all at once, let it sit for 10-20 years, then take it all out at once. The situation you face is closer to putting about $1000 into the market every month for 100 months. If you do that, then you're getting a different price for each purchase you make. Your actual return will be a weighted average of the return from each of those purchases. But note that this could help or hurt you. Using the chart Victor showed in his answer, if your lost decade is from one peak to the next peak, your average price will be below the price you would have entered and left at. So this helps. But if your lost decade is from trough to trough, then your average price is higher than the start and end price, so this has hurt you. Those are the two extreme cases, and the general case will be somewhere in between. And you can use these regular purchases to help you carry out your regular rebalancing.

Foreign equities. Since you mention the S&P500 specifically, I assume that you are in the United States. The US equities is approximately 45% of the world equities market. So even if the S&P500 has a lost decade, it's unlikely that the rest of the world will also have a lost decade at the same time. For comparison, the Tokyo Stock Exchange is the third largest in the world (behind the US's NYSE and NASDAQ); the market cap of the TSE is less than 20% that of the combined market cap of the NYSE and the NASDAQ, which puts it at about 10% of the world's market cap. When the Nikkei had its lost decades, no one else had a lost decade. Note that buying foreign equities is more expensive than buying domestic, and it exposes you to fluctuations in the exchange rate of the currencies. But the benefit of diversification probably outweighs those downsides. And obviously it's easier to diversify away from Japan than it is to diversify away from the United States. But there are people who advocate holding exactly the market weight of every country in the world.

  • +1 nice discussion. Wouldn't the 70/30 mix be reallocated each year? So money is shifted into sticks after each relatively (vs bonds) bad year, and out after each good? I'd imagine that, plus the DCA of new money would make the decade look far better for those still depositing. Jun 22, 2013 at 15:35
  • @JoeTaxpayer, thanks for the reminder. I have edited to mention rebalancing. Jun 23, 2013 at 2:26
  • 1
    I would like to add that collecting time premium on covered calls can also improve your returns, regardless of what the index does. A covered call limits your gain during a market rally, but you still make a gain in 3 out of 4 scenarios of price action (up, down to an extent, sideways.. but not further down) , where someone that has only bought the index only makes a gain in one scenario (up)
    – CQM
    Jun 24, 2013 at 7:18

John Bogle never said only buy the S&P 500 or any single index

Q:Do you think the average person could safely invest for retirement and other goals without expert advice -- just by indexing?

A: Yes, there is a rule of thumb I add to that. You should start out heavily invested in equities. Hold some bond index funds as well as stock index funds. By the time you get closer to retirement or into your retirement, you should have a significant position in bond index funds as well as stock index funds. As we get older, we have less time to recoup. We have more money to protect and our nervousness increases with age. We get a little bit worried about that nest egg when it's large and we have little time to recoup it, so we pay too much attention to the fluctuations in the market, which in the long run mean nothing.

How much to pay

Q: What's the highest expense ratio that one should pay for a domestic equity fund?

A: I'd say three-quarters of 1 percent maybe.

Q: For an international fund?

A: I'd say three-quarters of 1 percent.

Q: For a bond fund?

A: One-half of 1 percent. But I'd shave that a little bit. For example, if you can buy a no-load bond fund or a no-load stock fund, you can afford a little more expense ratio, because you're not paying any commission. You've eliminated cost No. 2....

  • 1
    If we have a 21 year flat market as Japan has had, how will this answer help? Jun 18, 2013 at 2:12
  • 1
    The question, at least the edited question, implied that Bogle said invest in the S&P 500 and sit back. You would be invested in an international fund, domestic equities, bond fund...The diversification would be how you structure a portfolio, plus re balancing and changing how you invest over time. Jun 18, 2013 at 2:28

Generally, you need something that goes up over time during periods of index decline, but otherwise holds some value. Historically, people tend to use gold for that purpose. But with gold also set up for possible declines, that raises questions. Silver has dropped a bit more than gold in terms of percentages.

If you think the downward motion will be in the form of sudden jumps, you can look at putting some of your money in puts away from the current price, but you can easily wind up paying too much for this protection.

In the case of a deflation, most things lose value vs. money, and you want all cash.

These things might already be obvious. I don't think there is a clear answer to your question. But if the future were clear, the present market could possibly anticipate and adjust... one reason the future of the market always seems a bit murky.


Common financial advice is just that - it is common and general in nature and not specific for your financial needs, your goals and your risk tolerance.

Regarding the possibility of a US market not going anywhere over a long period of time, well it is not a possibility, it has happened. See chart below:

S&P 500 Monthly Chart over 20 years

It took 13 years for the S&P 500 to break through 1550, a level first reached in March 2000, tested in October 2007 (just before the GFC) and finally broken through in March 2013. If you had bought in early 2000 you would still be behind when you take inflation into account.

If you took the strategy of dollar cost averaging and bought the same dollar value (say $10,000) of the index every six months (beginning of each January and each July) starting from the start of 2000 and bought your last portion in January 2013, you would have a return of about 35% over 13.5 years (or an average of 2.6% per year).

Now lets look at the same chart below, but this time add some trend lines.

S&P 500 Monthly Chart over 20 years with Trendlines

If we instead bought whenever the price crossed above the downtrend-line and sold whenever the price crossed below the uptrend-line (with the first purchase at the start of January 2000), we would have a return of 93% over the 13.5 years (or an average of 6.9% per year).

Another more aggressive option (but manageable if you incorporate a risk management strategy) is to buy long when the price crosses the downtrend-line and sell your existing long position and sell short when the price drops below the uptrend-line. That is profiting both up-trending and down-trending markets. Again we start our buying at the start of January 2000. By shorting the index when the market is in a down-trend you could increase the above returns of 93% by another 54%, for a total return of 147% over 13.5 years (or an average of 10.9% per year).

To conclude, using a simple long term strategy to time the markets may result in considerably higher returns than dollar cost averaging over the medium to long term, and I know which strategy would help me sleep better at night.

  • 2
    Nice charts. How might dividends change the scenarios above? Jun 18, 2013 at 23:20
  • @ChrisW.Rea - I don't usually buy Index ETFs so am not sure if you get dividends from these. If dividends were received as well you might want to add say about 3% to 4% per year to the returns of dollar cost averaging scenario, and maybe half that for the timing the market (long only) scenario (as you would not be in the market all the time). The main point is that the US market has gone nowhere in terms of growth for 13 years.
    – Victor
    Jun 18, 2013 at 23:56
  • 4
    Hindsight being 20-20, one can fit very good uptrend and downtrend lines after the fact. While the market is trading, mis-calculating the trend and missing major trend changes not suprisingly will cause more loss in the more aggressive strategies. How do you expect to deal with the dynamic nature of this trend estimation problem?
    – Paul
    Jun 19, 2013 at 3:04
  • 2
    Emphasis: You'd be well behind if you bought in early 2000 and took inflation into account. As in more than 20% behind.
    – mbhunter
    Jun 22, 2013 at 20:23
  • 1
    @mbhunter - moneychimp.com/features/market_cagr.htm shows Jan'01 - Dec'10 return was 14%. Not great, but not negative with dividends. -9% for the decade including inflation. I've not run the numbers to calculate return either DCA'd over decade or re-balanced every year with 80/20 or 70/30 split. I suspect those go positive. Jun 23, 2013 at 2:13

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .