大概率向上，小概率像1987. 下麵是我的一些討論：
We just had a down month of October, Nasdaq down 9.2%, third worst October since Reagan presidency from 1981: surpassed only by October 1987, -27.2% (7th year of Reagan presidency), and October 2008, -17.7% (8th year of Bush Presidency). In comparison, Nasdaq was only down 8.25% in October 2000 (8th year of Clinton Presidency).
What would the coming November turn out? The first two days of November turned out net positive, largely due to President TRUMP’s "nice" phone call with China President Xi and the hope for a trade deal between US and China (Nov. 1, up), and news about preparing or not preparing deal draft (Nov. 2, down). Trump and Xi are set to meet on November 29 in Brazil during G20 meeting, and the hope and progresses could run high to end of November and give investors a good November return, as pointed out earlier by 十二少等名博. However, as discussed in the following, some resemblance with Nov 1987 and some worry about election and impeachment could not be ruled out.
(1) History of November following a down October
History doesn’t repeat but it rhymes. One may look at history to find clues. Here I examined stock return data since Reagan presidency, as that’s when US started the path of increased borrowing and enjoyed looser and looser financial conditions over time.
For Nasdaq: From 1981 to 2017 (37 years), there were 15 years when Nasdaq returned negative for October. Of the 15 negative October months, 2/3 of the time was followed with a positive November, and 1/3 was followed with a negative November. For the 10 following November that returned positive, the average return was 3.54%. For the 5 following November that returned negative, the average return was -8.80%. These five November months were:
Nov. 1984, -1.86%
Nov. 1987, -5.6%
Nov. 1988, -2.88%
Nov. 2000, -22.9%
Nov. 2008, -10.77%
For SP500: From 1981 to 2017 (37 years), there were 13 years when SP500 returned negative for October. Of the 13 negative October months, 9/13 of the time was followed with a positive November, and 4/13 was followed with a negative November. For the 9 following November that returned positive, the average return was 3.43%. For the 4 following November that returned negative, the average return was -6.36%. These four November months were:
Nov. 1984, -1.51%
Nov. 1987, -8.53%
Nov. 2000, -8.01%
Nov. 2008, -7.48%
Current situations are very different from 2008: 2008 US was in recession since Nov. 2007, and Oct/Nov 2008 period was when Lehman Brothers sank into bankruptcy. Current situations are also very different from 2000: 2000 Internet bubble burst started in April 2000 and weakness in technology stocks were widely known by Oct/Nov 2000. If one removes Nov 2000 and Nov 2008 as unlikely for this coming November, one would have Nasdaq return of 1.93% in average for November following a down month of October.
(2) History of mid-term election years
Another look for Nasdaq returns in mid-term election years are even more promising: all mid-term election years since Reagan presidency had great October, except Bush senior’s 2nd year in 1990 when US was in a mild recession:
Reagan Oct 1982, 13.3%; Nov 1982, 9.26%
Reagan Oct 1986, 2.88%; Nov. 1986, -0.33%
Bush1 Oct. 1990, -4.27%; Nov. 1990, 8.88%
Clinton Oct. 1994, 1.73%; Nov. 1994, -3.49%
Clinton Oct. 1998, 4.58%; Nov. 1998, 10.06%
Bush2 Oct. 2002, 13.45%; Nov. 2002, 11.21%
Bush2 Oct. 2006, 4.79%; Nov. 2006, 2.75%
Obama Oct. 2010, 5.86%; Nov. 2010, -0.37%
Obama Oct. 2014, 3.06%; Nov. 2014, 3.47%
Trump Oct. 2018, -9.2%
Recent history thus indicated a high chance of a great November 2018: it could even be something like Nov. 1990.
(3) Other considerations
However, current situations have some resemblance with Oct/Nov 1987, when stock turned down for no clear reason, as some blamed high interest rates, some blamed worry about recession which turned out not there until July 1990, and some blamed for quantitative algorithms, etc. Currently, investors worried about increasing interest rate (inflation), tightening financial conditions, trade wars with China, high valuation for stocks, and late stage of economic expansion. In other words, a down November 2018 like 1987 could not be ruled out.
Another possibility is clear win by democrats for the House and Senate and worry of proceedings of TRUMP impeachment. The effect of impeachment proceedings can be huge if we have something like what happened to Nixon presidency in 1974 which was coupled with a weak economy. The effect of impeachment proceedings can be minimal or barely identifiable if we have something like what happened to Clinton presidency in late 1998 and early 1999 when the economy was super strong. In the end, it is the economy that’s the deciding factor for stock return. Current earnings data and next quarter guidance are decent, with over 80% firms beat earnings estimates and most guidance misses are due to head wind arising from a strong dollar, i.e. not due to a weakening economy. There are some weaknesses in housing and car sales, but jobs add are robust. It does not seem like a recession period. Effects of impeachment proceedings against President Trump, if any, would depend on how the economy goes in 2019 and 2020.
In a May 2018 Dialogue with the Fed presentation, Waller explained that the Fed is taking a capped, controlled approach to unwinding its balance sheet: letting Treasury securities “run off” at about $6 billion a month and letting mortgage-backed securities run off at about $4 billion a month. “And then it’s going to increase at every three months,” he said, “to where there’s a maximum of $30 billion a month in Treasuries running off, and $20 billion a month in mortgage-backed securities” running off.