3 ms·

This line of reasoning assumes the future looks like the past. Which is generally a solid mode of thinking. But, if the future looks like the past, shouldn't i

by sfblah 6y ago

This line of reasoning assumes the future looks like the past. Which is generally a solid mode of thinking. But, if the future looks like the past, shouldn't it give you pause that on many metrics markets are substantially more richly valued than at any peak in the last 100 years? You can't have it both ways. Either the past is useful or it's not. If it's useful, then you have two conflicting data points -- market timing doesn't work, and we're at the top of a big bubble. If the past isn't useful, then neither data point is valid. Now, you'll probably argue that even if it is a big bubble, the past suggests it always comes back. Setting aside the obvious fact that it's mathematically better to buy after the bubble bursts than before, you're in the same conundrum. Does the past matter or not? If it does, then yes you should expect to recoup the money eventually from buying at the top (though, if you bought at the top in 1929, you had to wait until the 1950s to get back to even). If the past doesn't matter and this time is different, then you can't say anything productive about the future.

throw0101a 6y ago
> If it's useful, then you have two conflicting data points -- market timing doesn't work, and we're at the top of a big bubble. Market timing does not work over the long-term, and even if we are at the top of the bubble, you can't know we're at the top, but more importantly: it doesn't matter. * https://awealthofcommonsense.com/2014/02/worlds-worst-market-timer/ https://awealthofcommonsense.com/2014/02/worlds-worst-market... > If it does, then yes you should expect to recoup the money eventually from buying at the top (though, if you bought at the top in 1929, you had to wait until the 1950s to get back to even). And what would your returns if you kept doing DCA every paycheque from the 1929 top, on the ride all the way done, and then little by little through the 1930s, 1940s, and 1950s? Turns out, not horrifically: > I wanted to show how market conditions can affect the end results of an investor who periodically invests in the stock market over time. Leaving aside taxes, costs, inflation, etc., I ran the numbers by decade going back to the 1930s to see how much money an investor would have ended up with by investing $10,000 each year on a monthly basis (or $833/month) in the S&P 500. * https://awealthofcommonsense.com/2018/04/the-luck-of-the-draw-when-diversifying-across-time/ https://awealthofcommonsense.com/2018/04/the-luck-of-the-dra... Especially if you have some bonds to rebalance with: > For instance, investors earned a real 2% average annual return on their equity investments during the 1930s, according to Ibbotson Associates, a market research firm. But they pocketed a real 7.1% on their government bonds and 2.7% on short-term Treasury bills. Yet during the 1950s investors earned a real average annual return of 16.8% on stocks, while losing 2.2% on bonds and 0.3% on bills. * https://www.marketplace.org/2009/01/05/history-rewards-stalwart-investor/ https://www.marketplace.org/2009/01/05/history-rewards-stalw... Bonds and rebalancing also would have saved a portfolio with the S&P 500 in the 2000s: * https://www.forbes.com/sites/investor/2010/12/17/the-lost-decade-was-a-golden-age-if-you-rebalanced/ https://www.forbes.com/sites/investor/2010/12/17/the-lost-de...
- sfblah 6y ago
  Cool. Now do the Japan bubble in the late 1980s.
  - throw0101a 6y ago
    Already did in another comment: * https://news.ycombinator.com/item?id=26138600 https://news.ycombinator.com/item?id=26138600
    - sfblah 6y ago
      Thanks. It sounds like you flipped the script and basically endorsed market timing based on the CAPE in that scenario. Maybe I’m misreading.
      - throw0101a 6y ago
        Asset allocation and portfolio management is about managing risk for the returns that are desired and/or needed. If I can achieve my (retirement) goals with only 4% returns, then why would I take on extra risk to get 6%? If I have 60% equities and 40% bonds to achieve my 4% returns, and equities go crazy, such that I've earned a pile of money in that portion of the portfolio such that they now make up (e.g.) 70%, then selling some off to rebalance is not about market timing, but rather about managing risk and not having too many financial eggs in a particular basket. The Nikkei 225 was 6700 in January 1980: * https://fred.stlouisfed.org/series/NIKKEI225 https://fred.stlouisfed.org/series/NIKKEI225 It had doubled by February 1986: you should have been taking money off the table as it went up. Take your profits and put it in other places: bonds, international. Further, as the valuations/CAPE went up, you should be expecting future returns to go down. By the time the Nikkei hit a CAPE of 50, that means expected returns would be 2%: is there any place where one could have put one's money that earned >2%? When the Nikkei hit a CAPE of 94, that means expected returns were ~1%. Japan 10-year bond were making 5% at the same time: * https://fred.stlouisfed.org/series/IRLTLT01JPM156N https://fred.stlouisfed.org/series/IRLTLT01JPM156N A bunch of money invested in Japan, in 1990, with a 30% JP equity allocation, a 40% JP bond allocation, and a 30% international equity allocation, would have funded a thirty year retirement using the 4% rule just fine: * https://www.gocurrycracker.com/lessons-from-japans-lost-decades/ https://www.gocurrycracker.com/lessons-from-japans-lost-deca... Why would anyone be investing 100% anywhere? Regardless of it being the Japan, US (S&P 500, NASDAQ, Russell 3000, etc), or any other other area/country? Harry Markowitz, who won the Economics Nobel for portfolio theory, said that diversification is the only free lunch in finance. * https://en.wikipedia.org/wiki/Harry_Markowitz https://en.wikipedia.org/wiki/Harry_Markowitz And answer to the "but Japan" quip is: diversification.

by sfblah 6y ago

throw0101a 6y ago

> If it's useful, then you have two conflicting data points -- market timing doesn't work, and we're at the top of a big bubble. Market timing does not work over the long-term, and even if we are at the top of the bubble, you can't know we're at the top, but more importantly: it doesn't matter. * https://awealthofcommonsense.com/2014/02/worlds-worst-market-timer/ https://awealthofcommonsense.com/2014/02/worlds-worst-market... > If it does, then yes you should expect to recoup the money eventually from buying at the top (though, if you bought at the top in 1929, you had to wait until the 1950s to get back to even). And what would your returns if you kept doing DCA every paycheque from the 1929 top, on the ride all the way done, and then little by little through the 1930s, 1940s, and 1950s? Turns out, not horrifically: > I wanted to show how market conditions can affect the end results of an investor who periodically invests in the stock market over time. Leaving aside taxes, costs, inflation, etc., I ran the numbers by decade going back to the 1930s to see how much money an investor would have ended up with by investing $10,000 each year on a monthly basis (or $833/month) in the S&P 500. * https://awealthofcommonsense.com/2018/04/the-luck-of-the-draw-when-diversifying-across-time/ https://awealthofcommonsense.com/2018/04/the-luck-of-the-dra... Especially if you have some bonds to rebalance with: > For instance, investors earned a real 2% average annual return on their equity investments during the 1930s, according to Ibbotson Associates, a market research firm. But they pocketed a real 7.1% on their government bonds and 2.7% on short-term Treasury bills. Yet during the 1950s investors earned a real average annual return of 16.8% on stocks, while losing 2.2% on bonds and 0.3% on bills. * https://www.marketplace.org/2009/01/05/history-rewards-stalwart-investor/ https://www.marketplace.org/2009/01/05/history-rewards-stalw... Bonds and rebalancing also would have saved a portfolio with the S&P 500 in the 2000s: * https://www.forbes.com/sites/investor/2010/12/17/the-lost-decade-was-a-golden-age-if-you-rebalanced/ https://www.forbes.com/sites/investor/2010/12/17/the-lost-de...

sfblah 6y ago
Cool. Now do the Japan bubble in the late 1980s.
- throw0101a 6y ago
  Already did in another comment: * https://news.ycombinator.com/item?id=26138600 https://news.ycombinator.com/item?id=26138600
  - sfblah 6y ago
    Thanks. It sounds like you flipped the script and basically endorsed market timing based on the CAPE in that scenario. Maybe I’m misreading.
    - throw0101a 6y ago
      Asset allocation and portfolio management is about managing risk for the returns that are desired and/or needed. If I can achieve my (retirement) goals with only 4% returns, then why would I take on extra risk to get 6%? If I have 60% equities and 40% bonds to achieve my 4% returns, and equities go crazy, such that I've earned a pile of money in that portion of the portfolio such that they now make up (e.g.) 70%, then selling some off to rebalance is not about market timing, but rather about managing risk and not having too many financial eggs in a particular basket. The Nikkei 225 was 6700 in January 1980: * https://fred.stlouisfed.org/series/NIKKEI225 https://fred.stlouisfed.org/series/NIKKEI225 It had doubled by February 1986: you should have been taking money off the table as it went up. Take your profits and put it in other places: bonds, international. Further, as the valuations/CAPE went up, you should be expecting future returns to go down. By the time the Nikkei hit a CAPE of 50, that means expected returns would be 2%: is there any place where one could have put one's money that earned >2%? When the Nikkei hit a CAPE of 94, that means expected returns were ~1%. Japan 10-year bond were making 5% at the same time: * https://fred.stlouisfed.org/series/IRLTLT01JPM156N https://fred.stlouisfed.org/series/IRLTLT01JPM156N A bunch of money invested in Japan, in 1990, with a 30% JP equity allocation, a 40% JP bond allocation, and a 30% international equity allocation, would have funded a thirty year retirement using the 4% rule just fine: * https://www.gocurrycracker.com/lessons-from-japans-lost-decades/ https://www.gocurrycracker.com/lessons-from-japans-lost-deca... Why would anyone be investing 100% anywhere? Regardless of it being the Japan, US (S&P 500, NASDAQ, Russell 3000, etc), or any other other area/country? Harry Markowitz, who won the Economics Nobel for portfolio theory, said that diversification is the only free lunch in finance. * https://en.wikipedia.org/wiki/Harry_Markowitz https://en.wikipedia.org/wiki/Harry_Markowitz And answer to the "but Japan" quip is: diversification.