SP500 (Calculating Returns)

Discussion in 'Trading' started by GuitarXM, Oct 27, 2018.

  1. GuitarXM

    GuitarXM

    I obtained the data below from the SP Dow Jones website and using the numbers below the math doesn't seem to match.

    For the decade of 1990-1999, I get Earnings Growth (7.8%), Dividend Yield (3.0%), Speculative Return (6.9%), Market Return (18.3%).

    In the chart from John Bogle's book, the 1990's shows Earnings Growth (7.4), Dividend Yield (3.2), Speculative Return (7.2), Market Return (17.8).
    https://hurricanecapital.files.wordpress.com/2016/04/bogle_4.png


    I also noticed that he adds Earnings Growth + Speculative Return.
    But it seems that instead of adding, it should be multiplied to include compounding effect? I'm not sure about that though.

    For example, per the data below, beginning of the year 1990 P/E = 14.53, end of year 1999 P/E = 28.43
    That is a Speculative Return of 6.9% annually and not 7.2.

    Earnings were 24.32 beginning of 1990 and 51.58 at end of 1999.
    That is an Earnings Growth of 7.8% annually and not 7.4.

    If you add the Speculative Return and Earnings Growth the way he adds them you get 6.9 + 7.8 = 14.7% Annual return

    But looking at the SP500 prices below, it doesn't seem to be correct. From the beginning of the period SP500 is at 353.40 and 1469.25 at the end of the decade. That is a 15.3% annual return and not 14.7

    So adding the annual percentages doesn't seem to be correct.

    Year Earnings Dividend SP500 Earnings Dividends
    1989 6.88% 3.32% 353.4 24.32 11.06
    1990 6.86% 3.74% 330.22 22.65 12.09
    1991 4.63% 3.11% 417.09 19.30 12.20
    1992 4.79% 2.90% 435.71 20.87 12.39
    1993 5.77% 2.72% 466.45 26.90 12.58
    1994 6.91% 2.91% 459.27 31.75 13.17
    1995 6.12% 2.30% 615.93 37.70 13.79
    1996 5.49% 2.01% 740.74 40.63 14.90
    1997 4.54% 1.60% 970.43 44.09 15.50
    1998 3.60% 1.32% 1229.23 44.27 16.20
    1999 3.52% 1.14% 1469.25 51.68 16.69

    I admit, math isn't my specialty but somehow the numbers don't add up.
    Try doing the math and see what you get for 1990 - 1999.

    I'm interested in what your numbers look like.
     
  2. jharmon

    jharmon

    Did you contact John Bogle and ask?

    Slight changes in methodology/source data can cause this too. There are at least 4 headline methods for "earnings" and a few variants on the treatment of extraordinary items between different vendors.

    Many people think that the S&P 500 Total Return Index gives them actual total returns as if they were "invested" in the S&P 500 and earning dividends. However, you can't "invest" in the same way as S&P calculates the index. You'd need to rebalance every day and "reinvest" the dividend on the ex-date (which hasn't yet been paid to you).

    The difference between 15.3% and 14.7% is negligible and shows it's good enough.
     
    Last edited: Oct 28, 2018