Market Gains by President/Congressional Party

Interesting study by Ned Davis, via Van Kampen Investments of what the market’s look like under different party rule.

Rob Schumacher of Van Kampen notes:

"In every presidential election year, voters and investors alike focus on the race for the White House, and rightfully so. You see, as shown in the accompanying chart from Ned Davis Research, the historical data depicts market returns that vary greatly under Republican or Democratic leadership.

The same data also suggest that while presidential races may dominate the statistical landscape, a more interesting interaction between politics, the public and stock prices is likely to take shape. And it has very little to do with who wins or who loses."

I am assuming that while there may be correlation between parties and
market performance, there is no specific proof of causation, i.e., these policies cause those returns. My 2nd assumption is that the Federal Reserve Chair is more important than Congres or the Presidency to Markets. And in
terms of data sets, 106 years — ~26 presidential terms — is a bit
light. This might be really interesting after 500 years, though.

Regardless of those reservations, this is quite fascinating:

Gains (%) for Stocks by Party of the President and Majority Party in Congress


Political Variable Stocks
  Democratic President 7.19%
Republican President 3.85%
Democratic Congress 6.46%
Republican Congress 3.51%
  Dem Pres, Dem Cong 6.53%
   Dem Pres, Rep Cong 9.60%
   Rep Pres, Rep Cong 1.54%
   Rep Pres, Dem Cong 6.37%
All Periods Buy & Hold 5.34%

Sources: van Kampen, Ned Davis Research


The most interesting aspect of the current research into party control has little to do with which party has Congress or the White House — its when Congress is in session or not:

"Using historical pricing on the Dow Jones Industrial Average (DJIA), the Standard and Poor’s 500 Stock Index (S&P 500), the Center for Research in Security Prices (CRSP) Equal-Weighted Returns Index and Value-Weighted Returns Index, Ferguson and Witte* find that, “Depending on the index, daily returns when Congress is in session range from 1 to 4 basis points per day. When Congress is out of session returns range from 5 to 15 basis points a day.”

Media spin aside, in a striking conclusion Ferguson and Witte remind their readers that, “Fully 90 percent of the historical capital gains on the DJIA occurred on days when Congress was out of session"

We are tempted to make the tongue-in-cheek observation that when Congress — of either party — is in session, it is dangerous to your portfolio’s health!


UPDATE October 31, 2006 10:11 am

And as we have discussed in the past, it is important to consider Multiple Variables in Market Analysis. Who controls the Congress or Presidency is but one issue out of 100’s if not 1,000’s, and perhaps a minor one at that . . .


For Investors, Elections are Only the Beginning
Rob Schumacher          
Van Kampen Insight Line—October 30, 2006

Gains for Stocks, Industrial Production, Inflation, Bonds and U.S. Dollars ($),
by Party of President and Majority Party in Congress"
Ned Davis Research
Report T_50, 03/04/1901 to 10/23/2006”

Congress and the Stock Market
Michael Ferguson
and Hugh Douglas Witte
University of Cincinnati and
University of Missouri, March 13, 2006

Print Friendly, PDF & Email

What's been said:

Discussions found on the web:
  1. Barry Ritholtz commented on Oct 31

    Was anyone else but me surprised that the Dow only returned 5.34% per year over a century?

    I assume that is without dividends, which were more significant — ~4% — back in the day.

  2. anon commented on Oct 31

    Is it approprite to use data going back to 1901? I think the two parties have shifted a great deal ideologically since the turn of the (last) century.

  3. Anonymous Coward commented on Oct 31

    I’m shocked I tell you, shocked. This can’t be RIGHT.

  4. Idaho_Spud commented on Oct 31

    No surprise there. What’s good for a handful of special interests (read Halliburton) is not necessarily in the interests of main street America.

  5. joe commented on Oct 31

    this is completely off topic, but did anyone see Ford’s announcement today? It shows how completely nonsensical the “statistical anomaly” in the Q3 GDP report regarding auto production was. Between housing and autos, Q4 GDP is going to be brutal.

    “Ford’s projected first-half 2007 cuts — reported yesterday by trade publication Automotive News — come on top of a 21% production cut planned for the current quarter.

    GM and Chrysler also have slashed current quarter production as they try to work down high inventories of unsold vehicles. The companies haven’t yet announced 2007 production plans.”

  6. Aaron commented on Oct 31

    These facts have a liberal bias.

  7. tjofpa commented on Oct 31

    No wait, it was Prudhoe Bay coming back on line that’s causing oil prices to drop…

    No, no, its the easing of tensions in Nigeria..

    Make for a nice Headline this weekend won’t it…
    “Gas prices nationwide have fallen to below $2/gal for all u middle class voters out there.”

  8. Republican troll commented on Oct 31

    OK so stocks perform better under Democrat presidents but those Liberals will tax and spend cut & run flip flop tax and spend cut & run flip flop tax and spend cut & run flip flop tax and spend cut & run flip flop tax and spend cut & run flip floptax and spend cut & run flip flop tax and spend cut & run flip flop tax and spend cut & run flip flop tax and spend cut & run flip flop(drools collapses)

  9. Mike commented on Oct 31

    There really isn’t enough data here to draw any serious conclusions. Also, I wonder how much these are weighted by the bullish trends in the 80s and 90s.

  10. Barry Ritholtz commented on Oct 31

    80s: GOP Prez. Dem Cong
    90s: Dem Prez, GOP Cong

  11. Steve C commented on Oct 31

    Take out the Great Depression and the results are much closer.

  12. jj commented on Oct 31

    stock market gains , ex-down years ?????

    inflation , ex-inflation
    GDP , ex-growth

  13. Francois commented on Oct 31

    If memory serves me correctly, Mark Twain once said: “No man’s property is safe while Congress is in session”.

    Never thought that scientific data would (should I say “at last!” ? :-D ) would prove that point.

  14. BDG123 commented on Oct 31

    “Back in the day”. You mean before the time of the great unpleasantness?

    I’ve done alot of anecdotal analysis around politics and returns. The chart is interesting at face value but I believe the more interesting point of view is to compare election results with a 100 year chart of equity markets. ie, Look at them graphically where the human mind works most efficiently.

    The more interesting question is why do we elect Republican President’s and/or Congress versus D/D or D/R or R/D.

    I really don’t think there is any coincidence we have a R Pres and a R Congress after having the most massive wealth accumulation cycle ever. When we are all living large, we want Uncle Sam to stay out of our lives. Now, we enter a new cycle. One where Democrats last swept the Congress and we elected a Democratic President.

    So, I believe the above analysis is the tail wagging the dog. The more important issue is what are the macro environments which lead the cumulative consciousness to vote one way or another and the underlying fundamentals each President inherits. Thus, leading to that chart. Not the other way around. JMO

  15. Fishbones commented on Oct 31

    No surprise there. What’s good for a handful of special interests (read Halliburton) is not necessarily in the interests of main street America.

    Bingo. This is also why dictatorships become third world nations. As long as one person (or party) does not hold all political power, good ideas win. It doesn’t depend on who put forth an idea, it depends on if the idea is good or not.

  16. Q-Ball commented on Oct 31

    Drawing conclusions about this data also assumes that the full affect of policy (both positive and negative) is fully felt during the term of the party who instituted the policy. Since it is widely posited that interest rate changes take 12-18 months to fully get incorporated into the economy how long does it take for tax policy changes (staged or otherwise), property rights, jobs policy, trade policy, entitlement policy, etc to get fully incorporated into the economy? Since we like to shift parties regularily the effects of the policies of one administration are probably still being incorporated well into the term of the next administration and in some cases perhaps multiple administrations.

    I really don’t think this data can be used to draw much of any conclusion. It might make for interesting sound-bites but it is not meaningful for drawing conclusions.

    Now if one party controlled the white house for 50 years followed by another for 50 years and the various combinations including the congress, then perhaps you could draw some conclusions, although outside random events such as great depressions and world wars etc, that happened to hit during one of those times could alter the conclusiveness of that data as well.

    As it stands, there is little more here than noise and randomness.

  17. DD commented on Oct 31

    16 June 1971, New York Times, “Letters to the Editor,” pg. 44:
    “No man’s life, liberty or property is safe while the legislature is in session…” is a quote lifted mirthlessly, not from a 1971 Times editorial but from a 105-year-old New York court decision [1 Tucker 248 (N. Y. Surr. 1866)]. However hoary, it is uncomfortably apt as we view the carnage left by this year’s legislative session—sighing with relief that the legislators have gone home before they could do even more damage.

    is that better b?

  18. alexd commented on Oct 31

    Q ball I think you rely on statistics too much. I really like statisitcal confirmation. There are situations where it is unavailable due to lack of information or perhaps from not asking the correct question.

    Somebody said something about identifying a duck.

    I believe there are people out there through a combination of inteligence,experiance, and wisdom who can despite the lack of all encompassing information make decisions about what can be concluded or done. I am inclined to think that even if it is forced by circumstance, people make better decisons when they have to find a middle ground. There is something that thinking about what another party wants that seems to provide insight into a situation. I believe when one party totally dominates the situation the party making the decisions rarely looks into the future in their decisions but rather makes their choices based on their own immeadiate gratification.

    So I think that one thing that can be inferred from the information is that mixed politics seems to work better, even if I am wrong and the only reason for it is that market participants bieve it does and vote accordingly.

  19. Roger Nusbaum commented on Oct 31

    Does the second half of the 1990’s create a skew?

    As to the comment about the depression, there were a couple of massive up years. One year in the mid 1930s was up more than 50%.

  20. Q-Ball commented on Oct 31


    Concerning the great depression, FDR was in office by 1933 so he would get credit for the massive up years while the repubs would get credit for 1930, 1931, & 1932. I think that was the point.

  21. Q-Ball commented on Oct 31


    I was mostly referring to determining which presidential party the market performs better under. However even the split rule data may not be any better, but I can’t tell because I don’t know how the split rule data shakes out since its not listed what the various periods are.

    I agree that there is something to each party preventing the other one from doing too much damage, but I would need to see a better break down of split rule. As noted by Barry the entire last segment of bull market run was under split rule with the last 5 years of the 90s being an inflated stock market run based on dot com over-hype, bogus accounting, y2k spending, and high fed liquidity in preparation for the y2k bank run. That tended to over-inflate the results of the Dem pres, repub congress results. Heck the Nasdaq doubled from summer 99 to Mark 2000. That displacement was entirely reversed over the next few years and potentially exacerbated by 911, which falls mostly in the rep pres repub congress. The stock market dump was mostly done before any of the policies of the current administration could take hold (In fact most of the dump before any of his policies could even get passed in congresss). Yes as the data is segmented, he and the repub congress get the “credit” for the putrid results of the stock market in 2001 and 2002.

    This is what I mean by results happening during one administrations term not necessarily having much or anything to do with the policies of that administration. I am not suggesting they had anything to do with the previous administration either. Who knows what they had to do with. But the data presented here especially as it is segmented based on the day you take office where you can’t possibly have any real policy affect that impacts the economy for atleast a year and probably more like 2 or 3 it just doesn’t reflect any direct correlation to the policies of the party in office and what they do to the stock market.

  22. ModerateRepublical commented on Oct 31

    The thing is, the economy lags behind policy. A better research would be ‘stock gains 4 years after a president’. As the public tends to flip Rep-Dem-Rep-Dem, you can likely see that the index of ‘stock gains after 4 years’ would favor Republicans. A sidenote conclusion is that the market underprices the damage done by tax hikes.

  23. algernon commented on Oct 31

    Q-ball is right. The 90’s could not have happened with the tax rates & regulation of the 70’s. That is, the prosperity of the 90’s could not have happened without Ronald Reagan in the 80’s.

    On the otherhand, Bush & co. are champions of big gov’t. John Kennedy was for lower marginal tax rates. These labels don’t mean a helluva lot over time.

  24. alexd commented on Oct 31

    A very good response q ball. Thank you.

    As far as who is responsible for the economic picture I think the amount of time that an interest rate change seems to filter into the economy, that is is about one and a half years, seems to be a good starting point.

    That is give a president about a year and one half from any type of legislation that he or she created or approved of before judging if that legislation or policy had an effect and what that effect manifested itself as.

    Four years seems like fitting the information to a desired result.

  25. Q-Ball commented on Oct 31


    18 months seems pretty fair to me as well. Of course there is lots of other noise that could be out of control of the government (you mean they don’t control everything?) that could affect things positively or negatively as well.

    But as a means of just measuring the policies of a new administration and a congress it would seem a better test to give them the time to actually have an impact. Since it usually takes 3-9 months before the new administration can actually pass much of anything it would seem a 2 year waiting period wouldn’t even be out of the question to try to test the results. For a two term president this would still give you 6 years that overlap with that president’s term to judge the results of his policies.

    Obviously its all guess work but I think it would be closer than just judging them from the day they take the oath.

    And this still does nothing to address policy changes that could ripple through the economy for decades. Again, things such as tax policy, trade policy, etc. Anyone think NAFTA isn’t still impacting the economy? People can argue amongst themselves if its pro or con.

  26. ken commented on Oct 31

    The economy usually does better with Democrats than with Republicans so it is no surprise to find that investors do better as well.

  27. Hawkmoon Nine commented on Nov 1

    Barry, this is wrong:
    80s: GOP Prez. Dem Cong
    90s: Dem Prez, GOP Cong

    The Dems controlled the House all thru the 80s, but Rep’s held the Senate from 80-86.

    Along those lines, how do you (we) divide up the other eras when Congress was split? Like May 2001 through Jan 2003 when Jim Jeffords switched to I and gave control of the Senate to the Dems. I wouldn’t call the whole period of GWB’s presidency in total R control for Congress because of that.

  28. bob commented on Nov 1

    I think this all is very natural, but not because Reps are better managing economy then Dems. It’s just how people vote.

    It’s quite natural that low-income people will not tolerate Reps when times are bad. But at the time they vote the bad times are already over.

    In opposite, when things are good, poor people will vote Reps because at that time the capitalistic message sounds better. But at the time they vote the good times are over, too.

  29. A Dash of Insight commented on Nov 2

    Mid-Term Elections

    There is a lot of market attention to the mid-term elections. As a former professor in this field for more than a decade, and later a market analyst and investment manager, this is in my sweet spot. The analysis of

Posted Under