Market-timing:
A two-sided coin
Vanguard research
Advice to investors routinely mentions how difficult it is to successfully
time the market. Although the possibility for success at market-timing
does exist, the probability of victory is slight.1 On the other hand, a
review of historical returns shows that investors who managed to avoid
market downturns – even just a handful of the worst days – would have
seen their returns improve compared with those of a static equity
allocation. Given the 2008 equity bear market and the high market
volatility experienced in 2008 and so far in 2009, strategies that may
help to curb some of the market’s downside risk have received renewed
attention. Debates about the prudence of market-timing versus “staying
the course” have been common and, ironically, have often relied on
different sides of the same coin to support their views.
For Professional Advisers only. Not to be given to retail investors.
This document is published by the Vanguard Group Inc it’s for educational purposes
only and is not a recommendation or solicitation to buy or sell investments. It should
be noted that it is written in the context of the US market and contains data and
analysis that is specific to the US.
October 2010
Authors
Donald G. Bennyhoff,
CFA
Yan Zilbering
One side of the coin: The theoretical
benefits of perfect market-timing
On the other side: The daunting practical
challenges of market-timing
Figure 1 illustrates the price return of the Standard
& Poor’s 500 Index from 1928 through 2008,
which averaged 5.0% for the period. We used price
returns in the figure because of limitations in the
availability of daily total returns before 1980
(dividends accounted for an additional 4.28% per
year over the period). Consistent with common
sense and simple mathematics, investors would
certainly have improved their returns during this
time if they were able to sidestep the market’s
worst days. On the other hand, they would have
reduced their returns if they happened to miss out
on the market’s best days. Historical results bear
this out: Missing either the 20 worst or 20 best
trading days in the 81-year period (or 0.1% of the
20,340 total trading days) would have increased
or decreased an investor’s overall return by
approximately 50%. Although an 81-year streak
of such impeccable timing would be improbable,
to put it mildly, the magnitude of the return
differentials may seem adequate justification to
attempt this feat. As often happens, however, the
investor considers the results of only one action
or inaction. Rather than assess the risks of missing
the market’s best days, for example, an investor
will focus only on the benefits of avoiding the
worst – one-half of the results presented in
Figure 1. Investors may overlook the forest for
the trees.
Looking at the sequence, rather than the magnitude,
of best and worst days provides a more instructional
perspective on the challenge of effective markettiming: Not uncommonly, the best and worst trading
days occur within days of each other (see shaded
areas in Figure 2). Perhaps more significantly for
the average investor, the best days for the S&P
500 Index have closely followed the worst. Of the
20 best days during the period, 8 occurred within
10 trading days of one of the 20 worst days. This
nuance is important: As stated earlier, the perfect
market-timing necessary to achieve the results in
Figure 1 is improbable. Also, although some traders
might be inclined to view a significant daily decrease
as indicative of more to come, this has not
necessarily been the case. In many cases, negative
returns did linger following some of the 20 worst
days, but the weeks following those worst days
posted positive returns, on average (see Figure 3).
Figure 1.
During volatile markets, it’s easy to see how active
strategies that offer the prospect of better returns
may appear tempting. However, the prospect of
better returns does not necessarily translate into
an improved probability of achieving higher returns.
Historically, the best and worst trading days tend to
cluster in brief time periods, often during periods of
heightened uncertainty and distress, making the
prospect of successful market-timing improbable.
Missing the best or worst trading days: Price returns for the S&P 500 Index, 1928 through 2008
8%
7%
7.5%
Price return
6%
6.5%
Price return
5%
5.0%
Price return
4%
3.6%
Price return
3%
2.6%
Price return
2%
1%
0%
20 worst trading
days missed
10 worst trading
days missed
Invested for all 20,340
trading days
10 best trading
days missed
20 best trading
days missed
All investments are subject to risk. Past performance is no guarantee of future returns. The performance of an index is not an exact representation of
any particular investment, as you cannot invest directly in an index.
Source: Vanguard.
1 Philips, Christopher B., and Frank J. Ambrosio, 2009, The Case for Indexing (Valley Forge, Pa.: Vanguard Investment Counseling & Research,
The Vanguard Group).
2
Figure 2.
Twenty best and worst trading days for S&P 500 Index, 1928 through 2008 (chronological)
Percentage
change
Trading
days
10/28/1929
-6.81%
—
11/10/1932
7.51%
24
10/29/1929
-16.06%
1
3/15/1933
16.61%
77
10/30/1929
12.53%
1
4/20/1933
9.52%
25
11/6/1929
-9.92%
3
6/15/1933
-6.97%
39
11/14/1929
8.95%
6
7/20/1933
-8.88%
24
6/3/1931
7.54%
387
7/21/1933
-8.70%
1
6/22/1931
8.32%
13
7/26/1934
-7.83%
253
9/24/1931
-7.29%
67
9/5/1939
11.86%
1,279
10/6/1931
8.90%
8
5/14/1940
-10.30%
172
10/8/1931
8.59%
2
5/17/1948
7.73%
2,000
12/18/1931
8.29%
48
10/19/1987
-20.46%
9,905
2/11/1932
8.27%
37
10/21/1987
9.10%
2
5/31/1932
-7.45%
74
10/26/1987
-8.28%
3
Date
Percentage
change
Date
Trading
days
6/10/1932
7.66%
8
10/27/1997
-6.87%
2,530
8/3/1932
8.86%
37
9/29/2008
-8.79%
2,748
8/8/1932
7.79%
3
10/9/2008
-7.62%
8
8/12/1932
-8.02%
4
10/13/2008
11.58%
2
9/13/1932
-7.91%
21
10/15/2008
-9.03%
2
9/21/1932
11.81%
6
10/28/2008
10.79%
9
10/5/1932
-8.20%
10
12/1/2008
-8.93%
23
S&P 500 Index using daily price returns.
Notes: Chart includes the 20 best and 20 worst trading days. “Trading days” reflects the number of trading days since the last occurrence of those best or
worst days.
Source: Vanguard.
Figure 3.
Twenty worst days for S&P 500 Index (1928 through 2008) and returns for 20 trading days following each
Trading days
Date
Return
5
10
15
20
10/19/1987
-20.5%
1.2%
13.6%
8.1%
9.6%
10/29/1929
-16.1%
4.6%
-5.8%
5.8%
2.5%
5/14/1940
-10.3%
-11.1%
-10.6%
-11.6%
-3.8%
11/6/1929
-9.9%
-14.3%
3.2%
1.5%
8.9%
10/15/2008
-9.0%
-1.2%
2.5%
4.9%
-6.1%
12/1/2008
-8.9%
11.5%
6.4%
6.8%
9.1%
7/20/1933
-8.9%
1.0%
-2.0%
0.6%
1.8%
9/29/2008
-8.8%
-4.5%
-9.3%
-10.9%
-23.3%
9.6%
7/21/1933
-8.7%
8.1%
5.0%
9.7%
10/26/1987
-8.3%
12.3%
6.8%
8.4%
6.7%
10/5/1932
-8.2%
-9.7%
-2.2%
-6.0%
-12.3%
8/12/1932
-8.0%
7.6%
19.9%
26.3%
21.0%
9/13/1932
-7.9%
-2.3%
4.2%
3.2%
-11.8%
7/26/1934
-7.8%
7.8%
9.2%
9.0%
11.8%
10/9/2008
-7.6%
4.0%
-0.2%
4.9%
-0.6%
5/31/1932
-7.5%
8.5%
10.5%
6.7%
-0.7%
9/24/1931
-7.3%
-10.8%
-0.6%
-3.0%
1.5%
6/15/1933
-7.0%
4.9%
10.3%
22.8%
21.7%
10/27/1997
-6.9%
7.1%
5.0%
7.9%
7.9%
10/28/1929
-6.8%
-15.3%
-27.4%
-12.6%
-14.1%
Average
-9.2%
0.5%
1.9%
4.1%
2.0%
Source: Vanguard.
3
© 2010 The Vanguard Group, Inc. All rights reserved.