United States Investment Returns during
Democratic and Republican Administrations,
1928-1993
Chris R. Hensel and William T. Ziemba
A major difference in returns from large-capitalization stocks, small-capitalization stocks,
various bond indexes, and cash followed elections won by Democratic and Republican
presidents from 1928 to 1993. Small-cap stocks had significantly higher returns during
Democratic administrations than during Republican administrations, primarily reflecting a
lack of losses in the April-December period. This phenomenon was not a manifestation of the
January small-firm effect. Indeed, the results indicate a significant small-cap effect outside
January during Democratic presidencies. Large-cap stocks had statistically identical returns
under both administrations. For both Democratic and Republican administrations, small- and
large-cap stock returns were significantly higher during the last two years of the presidential
term than during the first two years. Corporate, long-term government, and intermediate
government bonds and cash had significantly higher returns during Republican administrations.
From 1937 to 1993, the simple investment strategies of investing in small-cap stocks
during Democratic administrations and either intermediate-term government bonds or
large-cap stocks during Republican administrations produced higher mean returns—with
higher standard deviations—than did investing in large-cap stocks throughout the period. The
cumulative wealth from these "politically correct" investment strategies greatly exceeded that
from other strategies—small- and large-cap stocks, all types of bonds, cash, and a 60/40
stock/bond mix.
S
everal researichers have studied the effect of
presidential elections on U.S. stock returns.
Herbst and Slinkman used data from 1926 to 1977
and found a 48-month political/economic cycle
during which returns were higher than average;
this cycle peaked in November of presidential
election years. ^ Kiley and Luksetich, and Hobbs
and Riley showed that, since 1900, positive shortterm effects follovkfed Republican victories and negative retums followed Democratic wins.^ Huang,
using data from 1832 to 1979 and for various
subperiods, found higher stock returns in the last
two years of political terms than in the first
^
Chris R. Hensel is senior research analyst at Frank Russell
Company in Tacoma, Washington. William T. Ziemba is the Alumni
Professor of Management Science in the Faculty of Commerce and
Business Administration at the University of British Columbia in
Vancouver, and a consultant to Frank Russell Company.
Financial Analysts Joumai / March-April 1995
These findings are consistent with the hypothesis
that political reelection campaigns create policies
that stimulate the economy and are positive for
stock returns.
We used data from 1928 to 1993 to test several
hypotheses concerning U.S. stock, bond, and cash
returns. We investigated three questions: Do
small- and large-capitalization stock returns differ
between Democratic and Republican administrations? Do corporate bond, intermediate- and longterm government bonds, and Treasury bill retums
differ between the two administrations? Do the
returns of various assets in the second half of each
four-year administration differ from those in the
first half?
This study in several ways expands on the
previous literature relating to presidential election
effects on asset returns. First, we examined smallcap stock and various cash and bond returns.
whereas the previous literature focused on largecap stocks. The results indicated a significant
small-cap effect during Democratic presidencies.
Not only have small-cap stocks had higher retums
during Democratic than during Republican administrations but also the small-cap minus large-cap
advantage was documented outside of the month
of January for the Democrats. We found the welldocumented small-firm effect, in which small-cap
stock retums significantly exceed those for largecap stocks in January, under both Republican and
Democratic administrations. This advantage was
slightly higher for Democrats, but the difference
was not significant. Moreover, bond and cash
returns were significantly higher during Republican compared with Democratic administrations.
Our results also extend previous findings that
stock returns have been higher in the second half
compared with the first half of presidential terms.
This finding is documented for small- and largecap stocks during both Democratic and Republican
administrations. Finally, two simple investment
strategies based on these findings yielded substantially better portfolio performance than did common alternatives during the sample period.
U.S. STOCK RETURNS AFTER PRESIDEtlTIAL
ELECTIONS
This analysis is based on monthly return data for
the 66-year period from January 1928 to Decem-
ber 1993. The large-cap data are from the S&P 500
hidex. Since March 1957, the Index has consisted
of the 500 largest stocks weighted by market value
(price times number of shares outstanding); before
then, it consisted of the 90 largest stocks. The
small-cap stocks consist of the bottom 20 percent
(by capitalization) of companies on the New York
Stock Exchange. The data are from Ibbotson Associates and comprise monthly, continuously compounded total returns, which include dividends,
for the large-cap and small-cap stocks. The return
distributions are close enough to being normally
distributed that statistical tests based on normality
are valid.
Figure 1 shows the average monthly return
differences of large- and small-cap stocks for election months and the subsequent 13 months (to the
end of the following year) minus the 1928-92
average for each month. Large-cap stocks had
higher mean retums during November and December of the election year and March, April, and
May of the following year. In all other months,
large-cap stocks had lower-than-average or average mean returns following an election. Small-cap
stocks had higher-than-average mean returns during the election month and in March, April, May,
and July.
Table 1 presents the aggregate information
from Figure 1. Both large- and small-cap stocks
Rgure 1. Stock Monthly Return Differences: Presidential Election Months and the
Subsequent 13 Months, 1928-89, minus Monthly Average for 1928-92
Nov
Dec
Feb Mar Apr May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
D Large Cap
. _ j p Small Cap
Rnancia! Analysts Joumai / March-April 1995
had lower-than-average mean returns during election years. Small-stock returns were 732 basis
points a year lower than the average, but this
result was not statistically significant.
Tabie 1. Annuai Average Equity Retums for
Presidentiai Election Months and the
Subsec|uent 13 Months, 1928-89, minus
Annuaiized iVIonthiy Averages for 19281992
Return Period
Election years
192S-92 average
Annual difference
Large Cap
Small Cap
6.55%
9.48
-2.93
4.26%
11.58
-7.32
Note: Monthly means were annuaiized by multiplying by 12,
The 1928-93 period encompassed 17 presidential elections. With the end of the Bush Republican
presidency in January 1993, there were 32 years of
Republican and a like number of Democratic administrations during this period. Table 2 lists and
compares the first-year, first-two-year, last-twoyear, and whole-term mean returns under Democratic and Republican administrations from January 1929 to December 1992 and for January 1937 to
December 1992, a period that excludes one term
for each party during the 1929 crash, subsequent
depression, and recovery period. Each term is
considered separately, so two-term presidents
have double entries. The f-values shown in Table 2
test the hypothesis that, during each of these
periods, retums did not differ between Democratic
and Republican administrations.
For 1929 to 1992, mean returns for small stocks
were statisticall)' higher during the Democratic
presidential terms than during the Republican
terms. The data confirm the advantage of smallcap over large-cap stocks under Democratic administrations—a phenomenon that was widely
discussed in the media in the aftermath of the
Clinton election victory. Small-cap stocks returned, on average, 20.54 percent a year under
Democrats compared with 1.94 percent under Republicans for the 1929-92 period. This difference,
18.59 percentage points, was highly significant.
The first-year return differences for this period
were even higher, averaging 33.33 points.
The right-hand panel of Table 2 presents the
return results after eliminating the 1929 crash, the
Depression, and the subsequent period of stock
price volatility. Removing these eight years (1929
through 1936) from the study period eliminates
Financial Analysts Joumai / Mardi-April 1995
one Democratic and one Republican administration from the data. The small-stock advantage
under the Democrats was still large (an average of
7.37 percent per four-year-term) but was no longer
statistically significant. The large-cap (S&P 500)
returns during Democratic rule were statistically
indistinguishable from the retums under Republican administrations.
For Democratic and Republican administrations, the mean small- and large-cap stock returns
were much higher in the last two years compared
with the first two years of presidential terms for
both of the time periods presented in Table 2. For
example, small-cap stocks returned 24.86 percent
during the last two years compared with 16.22
percent during the first two years for Democrats
and 10.18 percent compared with -6.29 percent
for Republicans from 1929 to 1992. Returns on
large-cap stocks increased to 16.29 percent from
Tabie 2. Annuai Average Stocic Retums for the
Rrst Year, First Two, i-ast Two, and Four
Years of Democratic and Republican
Presidencies
January 1929
to December
1992
Return Period
Large
Cap
January 1937
to December
1992
Small
Cap
Large
Cap
Small
Cap
18.88%
16.22
24.86
20.54
32.59
2.99%
5.16
13.74
9.45
17.16
8.90%
10.64
22,43
16.54
29.62
-14.45
-6.29
10.18
1.94
27.81
1.87
6.98
15.03
11.00
15.12
-6.22
1.39
16,95
9,17
19.89
33.33
22.51
14.68
18.59
1.12
-1.81
-1,28
-1,55
15.12
9.25
5.49
7.37
0.16
-0.28
-0.24
-0,36
1,39
0.98
0.62
1.14
Democratic
Average first year
8.01%
Average first two years
7.13
Average last two years
16.29
Average term
11.71
Standard deviation, term 19.18
Republican
Average first year
0.54
Average first two years
3.77
Average last two years
9.06
Average term
6,42
Standard deviation, term 21.17
Difference: Democrats
minus Republicans
First year
First two years
Last two years
Term
7.48
3.36
7.22
5.29
t-va!ues (Ho:Diff = 0)
First year
First two years
Last two years
Term
0,81
0,48
1.03
1.07
2.38*
2.18*
1.41*
2.55*
Note: Monthly means were annuaiized by multiplying by 12.
Monthly standard deviations were annuaiized by multiplying
by the square root of 12.
* = Significantly different from zero at 5 percent or lower level.
63
Table 3. Averaae iMonthiy Smali- and i^rge-Cap Stocic Retums during Democratic and Repubiican
Presidencies, January 1929-December 1992
Democratic Administrations
Month
January
February
March
April
May
June
July
August
September
October
November
December
Republican Administrations
S&P 500
Total Retum
U.S, SmaUStock Total
Retum
Small Cap
minus
Large Cap
S&P 500
Total Retum
U.S, SmallStock Total
Retum
Small Cap
minus
Large Cap
1.49%
-0.57
-0.58
2.32
0.72
1.66
1.97
1.22
0.10
0.39
1.31
1.67
6.88%
0.79
-0.96
2.79
0.90
1.93
3.07
1.29
0.37
-0.13
1.93
1.68
5.38%
1.36
-0.38
0.47
0.18
0.27
1.10
0.07
0.28
-0.51
0,61
0,01
1.65%
1.59
0.96
-0.24
-0.50
0.78
1.69
1.73
-2.87
-0.40
0.44
1.59
5.93%
2.78
1.21
-1.82
-1.52
-0,40
1,11
1.25
-3.31
-2,66
-0.53
-0.09
4.28%
1.19
0.25
-1.57
-1.02
-1.18
-0.58
-0.47
-0.45
-2.26
-0.97
-1.68
7.13 percent for Democrats and to 9.06 percent
from 3.77 percent for Republicans for the same
period. This result is consistent with the hypothesis that incumbents embark on favorable economic
policies in the last two years of their administrations to increase their reelection chances and that
the financial markets view these policies favorably.
The advantage of small stocks over large
stocks under Democratic administrations was not a
manifestation of the January small-stock effect.
Instead, as Table 3 shows, the relative advantage
of small- over large-cap stocks under Democrats
compared with that under Republicans was attributable to having fewer small-stock losses, as well
as higher mean small-stock returns, in the AprilDecember period. Under Democrats, mean retums
were positive in each of these months except
October, and the small-minus-large differential
was positive for 10 of the 12 months; under Republicans, the small-minus-large differential was negative during 9 of the 12 months.
The small-stock advantage was also evident
during the months following Democrat Clinton's
election. From November 1992 to December 1993,
the Ibbotson small-capitalization index had a total
return of 36.9 percent compared with 14.9 percent
for the S&P 500.
In response to a reviewer's comment, we
investigated how inflation varies with political
regimes. The results for the 1929-92 period, using
the Ibbotson inflation index, indicate that inflation
was significantly higher under Democrats, but this
difference was contained in the 1929-36 period.
Excluding this early period, inflation was slightly
64
higher, on average, under Democrats but not statistically different from inflation under Republicans. Inflation rates differed across the years of the
presidential terms. For example, for the 1937-92
period, in the first year of the presidential term,
inflation under the Democrats was significantly
lower than it was under the Republicans. An
analysis of the first and second two years of
administrations during this same period indicated
that inflation was higher under Democrats but the
difference was not statistically significant.
U.S. BOND RETURNS AFTER PRESiDENTIAL
ELECTIONS
The bond data are also from Ibbotson Associates
and consist of monthly, continuously compounded total returns for long-term corporate
bonds, long-term (20-year) government bonds, intermediate (5-year) government bonds, and cash
(90-day T-bills).
Figure 2 illustrates average retum differences
for bonds during election months and the subsequent 13 months (1929-89) minus each month's
1928-92 average return. Corporate, long-term government, and intermediate government bond returns were all higher than the monthly average in
the year following an election only in May, October, and November. Both government bonds also
exceeded the average in some other months.
Throughout the 14-month period from the
election month through the following year, corporate, intermediate government, and long-term
government bonds returned less than the monthly
Rnancial Analysts Joumai / March-April 1995
Figure 2. Bond Monthly Retum Diflerences: Presidential Election Months and the
Subsequent 13 Months, 1928-89, minus Monthly Average for 1928-92
-1.2
Nov
Dec
Jan
Feb
Mar
•
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Intermediate Government
Wi Long Government
•
Long Corporate
average, while cash returned more than the average (Table 4).
Table 4. Annual Average Bond Retums for
Presidentiai Election Months and the
Subseqiuent 13 Months, 1928-89, minus
AnnuaiJzed i\/lonthly Averages, 1928-92
Return Period
Election years
1928-92 average
Annual
difference
Long
Long
Intermediate
Corporate Government Government Cash
4.97%
5.28
-0.31
4.58%
4.62
-0.05
4.59%
5.04
-0.45
3.98%
3.65
0.33
As Table 5 shows, the performance of fixedincome investments differed significantly between
Democratic and Republican administrations. All
fixed-income and cash returns were significantly
higher during Republican than during Democratic
administrations during both study periods. The
high significance of the cash difference stems from
the low standard deviation over terms. The performance of fixed-income investments differed very
little between the first two years and the last two
years of presidential terms.
The distribution of Democratic and Republican administrations during the 1929-92 period
Financial Analysts Joumai / March-April 1995
plays a part in the significance of the fixed-income
and, especially, cash returns. As Table 6 indicates,
the cash returns for the first four Democratic
administrations in this period (1933-48) were very
low (0.20 percent, 0.08 percent, 0.25 percent, and
0.50 percent annually). This result helps explain
why the term cash-return differences in Table 5 are
so significant (f-value = -16.18 for 1929-92). Democratic administrations were in place for three of
the four terms during the 1941-56 period, when
government bonds had low returns. Bond returns
in the 1961-68 period (both Democratic terms) and
1977-80 period (Democratic) were also low.
SOME SIMPLE PRESIDEr^AL INVESTMEMT
STRATEGIES
Based on the results reported so far, we devised
and tested two presidential-party-based investment strategies. The first always invests in
stocks—small capitalization with Democrats and
large capitalization with Republicans; the second,
a simple alternating stock-bond investment strategy, invests 100 percent in small-cap stocks during Democratic administrations and 100 percent in
intermediate government bonds during Republican administrations. The test period was January
1937 through December 1993.
Figure 3 illustrates the growth of a $1 investment in each of these two strategies, as well as in
Table 5. Annuai Average Bond Returns for the Rrst Year First Two, Last Two, and Four Years of
Democratic and Republican Presidencies
January 1929 to December 1992
Return Period
Long
Long
Intermediate
Corporate Government Government
January 1937 to December 1992
ash
Long
Long
Intermediate
Corporate Government Government
Cash
Democratic
[
First year
First two years
Last two years
Term
Standard deviation, term
5.26
2.31%
2.85
1.00
1.92
5.62
1.57%
2.77
2.83
2.80
3.62
1.63%
1.83
2.61
2.22
0.82
5.97
8.27
7.28
7.77
8.13
6.66
8.50
6.43
7.46
9.23
7.54
8.64
6.20
7.42
5.00
6.48
5.88
4.2
5.08
0.90
-2.41
-4.42
-5.40
-4.91
-4.34
-5.65
-5.43
-5.54
-5.97
-5.87
-3.37
-4.62
-0.64
-1.78
-2.30*
-2.98*
-1.01
-1.95
-2.14*
-2.92*
-2.64*
-3.86*
-2.13'
-4.21*
,3.56%
2.86
2.66%
2.76
1.02
1.89
5.18
2.66%
2.58
0.27
1.43
5.83
1.54%
2.42
2.54
2.48
3.69
1.82%
2.06
2.96
2.51
0.85
6.36
8.67
87.72
8.19
8.49
7.13
9.15
6.62
7.89
9.56
7.79
8.99
6.65
7.82
5.21
6.74
6.22
4.74
5.48
0.88
-4.85
-4.05
-1.67
-2.86
-3.70
-5.91
-6.69
-6.30
-4.47
-6.57
-6.35
-6.46
-6.25
-6.57
-4.11
-5.34
-3.92
-4.16
-1.78
-2.97
-15.59*
-18.24*
-6.80*
-16.18*
-0.91
-2.17*
-2.58*
-3.40*
-0.95
-2.04*
-2.27*
-3.12*
-2.51*
-3.94*
-2.36*
-4.67*
-13.92*
-16.68*
-6.42*
-14.66*
Republican
First year
First two years
Last two years
Term
Standard deviation, term
Difference: Democrats minus
Republicans
First year
First two years
Last two years
Term
t-values (Ho:Diff = 0)
First year
First two years
Last two years
Term
Note: Monthly means were annuaiized by multiplying by 12. Monthly standard deviations were annuaiized by multiplying by the
square root of 12.
* = Significantly different from zero at 5 percent or better level.
large-cap and small-cap stocks (for comparison
with the all-stock strategy) and in a benchmark of
60 percent large-cap stocks and 40 percent intermediate government bonds (rebalanced monthly).
The 60/40 portfolio is a common investment strategy and provides a benchmark for comparison
with the stock-bond strategy. Transaction costs
were not included, but they probably would have
minimal impact because, at most, the presidential
strategies trade once every four years. These investment strategies, as shown in Figure 3, all lost
money until the early 1940s.
TTie specific cumulative wealth relatives for
Figure 3 are shown in Table 7 along with the
results achieved if the investments had been made
in January 1942 (right before they started performing better). The first line of Table 7 shows, for each
strategy, the amount that a dollar invested in 1937
would be worth at the end of 1993. Cumulative
returns for both presidential strategies exceeded
those for all the other investment strategies. The
presidential stock strategy ended with more than
double the small-cap value (the highest returning
Tabie 6. Presidents, Political Parties, and FixedIncome Retums, 1929-92
....... ,.,_.
Term
1929-^2
1933-36
1937-40
1941-44
1945-48
1949-52
1953-56
1957-60
1961-64
1965-68
1969-72
1973-76
1977-80
1981-84
1985-88
1989-92
Annuaiized Average
Monthly Retum
President
Party
Intermediate
Government
Bonds
Hoover
Roosevelt
Roosevelt
Roosevelt
Roosevelt/
Traman
Truman
Eisenhower
Eisenhower
Kennedy/
Johnson
Johnson
Nixon
Nixon/Ford
Carter
Reagan
Reagan
Bush
Republican
Democratic
Democratic
Democratic
Democratic
4.61%
5.05
3.73
1.74
1.48
2.26%
0.20
0.08
0.25
0.50
Democratic
Republican
Republican
Democratic
1.24
1.19
4.24
3.21
1.35
1.66
2.54
2.84
Democratic
Republican
Republican
Democratic
Republican
Republican
Republican
2.76
7.06
7.42
3.17
13.71
10.35
10.77
4.43
5.19
6.25
8.11
10.39
6.22
6.11
Cash
Financial Analysts Joumai / March-April 1995
Rgure 3.
Growth of $1 investment, January 1937-December 1993
1000
5/48
1/37
9/59
1/71
5/82
12/93
Large Cap
Small Cap
Pres. Stock
Pres. Stock/Bond
Benchmark
single stock category). The advantage of the presidential strategies relative to the large-cap strategy
was even greater if the initial investment was
made in January 1942.
Throughout both study periods, these simple
presidential-party-based investment strategies
produced higher cumulative wealth than either
stock investment or the 60/40 benchmark. Figure 3
and Table 7 make the case for the presidential
strategies appear strong, however, because of the
long time period. Table 8 indicates that those
strategies carried high risk: Their standard deviations were almost as high as those of small-cap
stocks from 1937 to 1993.
The first line of Table 8 shows that, for the
1937-93 period, the standard deviations of small
stocks and presidential strategies (stock and stockbond) were 25.0 percent, 23.3 percent, and 21.1
percent, respectively. The presidential stock strategy had the highest mean retum (13.9 percent).
Small stocks had a higher mean return than the
stock-bond strategy (13.0 percent versus 12.3 percent), but the cumulative wealth was less than
under the stock-bond strategy because of small
stocks' higher volatility.
Table 8 also shows the results for 40-, 30-, 20-,
10-, and 5-year intervals ending December 1993. In
all intervals, the presidential stock strategy had a
standard deviation greater than that of large-cap
stocks, but the returns were also always greater.
1
Tabie 7. Vaiueof$1 initial Investment
Dates
[anuary 1937December 1993
[anuary 1942December 1993
i
Large-Cap
Stocks
Small-Cap
Stocks
Presidential
Stock
Presidential
Stock-Bond
60/40
Benchmark
$156.4
$ 236.5
$ 502.5
$ 275.4
$ 80.1
3»S.7
1,066.4
2,266.1
1,242.1
103.2
Rnancial Analysts Joumai / March-April 1995
Tabie 8. Average Retums and Standard Deviations for Different investments during Different investment
i-lorizons
Large-Cap Stocks
Dates
January 1937December 1993
January 1954December 1993
January 1964December 1993
January 1974December 1993
January 1984^
December 1993
January 1989December 1993
Small-Cap Stocks
Presidential Stock
Presidential
Stock-Bond
60/40 Benchmark
Number Average Standard Average Standard Average Standard Average Standard Average Standard
of Years Return Deviation Retum Deviation Retum Deviation Retum Deviation Retum Deviation
57
10.2
16.0
13.0
25.0
13.9
23.3
12.3
21.1
8.2
10.1
40
11.2
14.4
14.3
20.1
15.4
17.0
13.1
12.3
9.3
9.3
30
9.9
14.9
14.0
21.8
15.6
17.9
15.7
12.9
9.1
9.8
20
12.0
15.9
17.2
21.5
16.0
18.6
14.8
12.8
11,0
10.5
10
13.9
15.8
9.5
18.7
14.9
15.9
11.7
5.7
12.7
10.1
13.5
12.8
12.5
15.9
15.4
13.1
12.4
5.9
12.4
8.5
With the exception of the 20-year period, the risk
of this strategy was lower and the return was
higher than those of small-cap stocks. For all the
investment horizons examined, the presidential
stock-bond strategy had lower risk than large-cap
and small-cap stocks. For the 40-, 30-, and 20-year
periods, the presidential stock-bond strategy had
a higher mean return than large-cap stocks, with
lower risk; that is, in a mean-variance sense, this
strategy dominated the large-cap strategy for these
periods. For the 30-year period, the presidential
stock-bond strategy also had higher mean returns
and lower risk than small-cap stocks. The 60/40
benchmark exhibited lower risk than the presidential strategies, except for the 10- and 5-year investment periods, but the generally lower risk was
always accompanied by lower returns. For the 10and 5-year periods, the presidential stock-bond
strategy had the lowest risk of any of the strategies.
.
. ,
CONCLUDING REMARKS
The most interesting and significant finding of this
study was the much higher small-stock returns
during Democratic administrations as compared
with Republican administrations. This finding is
consistent with the hypothesis that Democrats
devise economic policies that favor small companies and, consequently, their stock prices. Although this phenomenon is well discussed in the
media, the 33.33 percentage point difference between small-stock performance in Democratic and
Republican administrations in the first year in
office and the 18.59 percent difference for the full
four-year term were larger than expected.
This political-party effect is different from the
well-known January small-firm effect, which was
present for Republicans as well as Democrats. In
January, returns were higher for small-cap stocks
than for large-cap stocks during Republican, as
well as Democratic, rule. This phenomenon is well
documented. What we found in addition was a
substantial small-stock/large-stock differential outside of January during Democratic regimes (see
Table 3). Large-stock retums were statistically indistinguishable between Democrats and Republicans, but bond and cash returns were significantly
higher during Republican than during Democratic
administrations.
We also confirmed and updated Huang's finding that large-cap stocks have had higher retums
in the last two years of presidential terms; this
finding applies regardless of political party and for
both small- and large-cap stocks.
A study of possible differences in economic
policies that lead to the divergence of investment
results according to which political party is in office
would be of considerable interest. Clearly, candidates seeking reelection are likely to favor economic policies that are particularly attractive to the
public, and those policies are consistent with
higher stock prices. Cash returns did not differ
significantly between the first and second two-year
periods of Democratic and Republican presidential
terms.
Two simple presidential investment strategies
performed well over the sample period. The strategy of investing in small-cap stocks during Democratic administrations and large-cap stocks during
Republican administrations produced greater cu-
Financial Analysts Joumai / March-April 1995
mulative wealth than other investment strategies
during the 1937-92 period. The alternating stockbond strategy of investing in small-cap stocks
under Democrats and intermediate bonds under
Republicans produced the second highest cumulative wealth. Both of these presidential-party-based
strategies had higher standard deviations than
large-cap stocks alone during the 1937-92 period.*
FOOTNOTES
1. Anthony F. Herbst and Craig W. Slinkman, "Political-Economic Cycles in the U.S. Stock Market," Financial Analysts
JourmI, vol. 40, no. 2 {March/April 1984):38-44.
2. William B. Riley, Jr., and William A. Luksetich, "The Market
Prefers Republicans; Myth or Reality," Journal of Financial and
Quantitative Analysis, vol. 15, no. 3 (September 1980):541-59.
Gerald R. Hobbs and William B. Riley, "Profiting from a
Presidential Election," Financial Analysts Joumai, vol. 40, no.
2 (March/April 1984):46-52.
Rnancial Analysts ^loumal / March-April 1995
Roger D. Huang, "Common Stock Retums and Presidential
Elections," Financial Analysts Journal, vol. 41, no. 2 (March/
April 1985):58-65.
This research was partially supported by the Social Sciences
and Humanities Research Council of Canada and Frank
Russell Company. We would like to thank Gayle Nolte for
computational assistance, Karlene Johnson for editorial
help, and Steve Murray and Douglas Stone for helpful
comments on a previous draft.