This picture tells the story:
The percent of seats of the House of Representatives, the Senate, and the Presidency are plotted, starting in the year of the relevant election. I arbitrarily started at 1900.
The solid black line shows the Senate, the dashed black the House, and the solid blue the period of Democrat presidents (again, with the election years as the reference points).
To handle the odd third-party seats, the percentages were calculated based on the size of the Republican caucus in each house.
It is important to examine the percent of seats and not just the majority because, naturally, the closer a majority party is to the 50% mark, the more they are willing to compromise in the direction of the other party.
What might seem surprising is the extent of agreement between the House and the Senate: it is rare that majorities in both houses do not agree.
A glance tells that we have spent the majority of time with a Democrat-controlled legislature.
The forces that gave rise to the percent of Senate seats has the most correlation with which party wins the Presidency. This is probably because of the slower turn-over rate of the Senate as compared to the House.
Whichever house you consider, the fickleness of the American public is obvious. Even though one party (usually Democrat) controls a house, the share of seats oscillates with regularity, especially in more recent years. The pulse of the Senate is almost exact.
Up until roughly 1954, the public was more trusting of forming majorities in the House, Senate, and Presidency. That is, if the House and Senate were Republican controlled, so was the Presidency, and vice versa.
Thereafter, only Democrats controlled all three, or there was a mix. The exception is the first part of the Bush Presidency. Also during this time, Republican Presidencies correlated strongly with losses in Democrat seats in both the House and Senate (this is losses, not loss of control).
It is well to emphasize that the makeup of the Legislature and the Presidency does not imply an agreement between the government and a majority of citizens. A history of the recent welfare/entitlement health law proves this.
How to combine all three parts of government into one measure? There is no formula, and we have to rely on subjective judgment. No matter what we devise, missing completely, of course, is the third branch of government: the Supreme Court. This picture shown here, then, is necessarily incomplete.
The method used here is a simple average of House, Senate, and Presidency (Republicans score “100%” of the available seats with a Republican president). This is imperfect. Since it is the House and Senate that have control of money, perhaps Congress should receive more weight than the Presidency.
But it is the President that suggests a budget, and there is always significant anchoring to and adjustment of that offer. Plus, the President has a veto, the knowledge of which influences the behavior of Congress, particularly when the President is of the opposite party.
And then the Presidency is of much more importance to foreign policy, particularly in times of war. All these facts led to an equal weight.
We can all this index the percent Republican control of the country. It is the green line.
Except for the extra dose of Democrat during the FDR years, our fickleness is again evident.
Or perhaps “perpetually unsatisfied nature” is a better term than fickleness. The duration of control is mostly set by the rules governing our election cycle. But the bouncing back and forth is entirely due to us. One year we gladly agree that we should be awarded money from the public coffers, only to find that we aren’t happy with the decrease in our paychecks caused by the new entitlements.
But is that all? Maybe oscillations are the natural response of a democratic citizenry to the excesses and idiocies of the controlling party. Before politicians of one party become too powerful, we toss them out. And we often do so certain sure that the other party will be—this time—our saviors.
The 2010 elections are approaching. Crude extrapolation from recent trends—and ignoring all external factors—suggests Democrats will lose seats but retain majorities in both the House and Senate. However, these same extensions put a Republican in the White House in 2012, with a Republican Senate and Democrat house.
We can call these predictions the naive forecast, based simply on the data.
Matt,
We generally find that the single greatest indicator of incumbent party’s success in a seat (any seat) is if constituents see disposable income rise within one year of the election. Note that it’s not per capita income, but disposable income. This is why so many sitting first-term presidents institute tax cuts late in their third-year in office.
Source: http://www.jstor.org/pss/2111314
Your analysis is interesting, if only because it shows us the oscillations, but I caution that any predictive analysis should wait until 2011. Voters are both fickle and of short memory. They also tend to place blame on sitting politicians based on present conditions, not past mistakes. So, even if tax cuts were instituted by a previous seat holder, voters will give the new guy the benefit regardless of whether or not he’s from a different party.
But is that all? Maybe oscillations are the natural response of a democratic citizenry to the excesses and idiocies of the controlling party. Before politicians of one party become too powerful, we toss them out. And we often do so certain sure that the other party will be—this time—our saviors.
Yes and no. Ours is an outlier of a system in the world of democracies. Parliaments tend to behave rather differently, in large part due to the fact that the executive is not a separately elected office.
Furthermore, I would caution against using the presidency as an indicator of voter fickleness against party. In fact, US incumbency rates are incredibly high. Well over 90%, in fact.
http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=1962772
http://en.wikipedia.org/wiki/Congressional_stagnation_in_the_United_States
Once could argue that the presidency is a different ballgame because it’s a metric weighed against the ENTIRE US voting population, but even then, two-term presidents are the norm. We aren’t fickle: in fact, we’re downright change-averse.
Hmm… more things:
But it is the President that suggests a budget, and there is always significant anchoring to and adjustment of that offer. Plus, the President has a veto, the knowledge of which influences the behavior of Congress, particularly when the President is of the opposite party.
In theory. In practice we tend to find that it has more to do with the budgetary preferences of the executive vs. the legislature. Matt McCubbins at UCSD found that (and I quote him): “the effects of the veto on the overall budget are asymmetrical… the effect of veto power will be to lower spending only if the president prefers a lower budget than the legislature. It has no effect when the president prefers a larger budget than the legislature.”
Oh, and it’s losses of Democratic seats. That’s just a nitpick, really.
How do you see the Teabaggers, er, Teapartiers effecting this trend, if at all… or too soon to tell? 3rd party noise? What would Stupak say?
Cuz,
As Republicans, eventually. In primaries, tea partiers will vote for conservatives, but in general elections, they will tend to vote for whomever the Republican is.
Most of our elected third parties were in the House, and really not since WWII. And most of them edged closer to socialism in one way or another. Independents—true non-affiliates—have a tendency to show up in the Senate.
Matt,
I sort of agree.
I actually suspect that Tea Party-iets (TPs?) causing GOP primary candidates to shift further right in order to compete. A good amount of the recent scholarship on primaries finds that candidates have to shift further from the mean voter in order to win the primary because it’s the most politically energized who vote in primaries. Those same people, however, tend to be more politically “extreme” than the general election voters.
For better or for worse, however, that means that your typical Republican primary winner will probably be further from the mean voter in districts where they compete against TPs, because the TPs will drive the message more toward their end of the spectrum.
As for third parties, I have a sinking suspicion that the death knell for third parties has been gerrymandering by the two big parties giving them perpetual districts. It’s hard to compete as a third party when you either have a platform too dissimilar from the mean district voter or are too similar to the incumbent party to really differentiate yourself. In other words: no-win.
A true third party will never succeed in the American system. Our system is biased toward a two party system. To win in America a canididate needs to win a plurality of votes. It is a high hurdle to jump for a new party to take more than 1/3 of the vote in a district, much less a state. 3rd party candites can take a small number of congressional districts, but will never win more than that. The best they can hope for is that some of its platform will be addopted by one of the Majors.
In a parlementary system, third parties are viable. The third party can win a minority of seats, but if no one wins a majority, the third party will be invited to form the new government.
Doug,
It’s actually a few things, I think, that kills third parties in the US, because the UK is also first-past-the-post but has a third party.
It’s primaries, combined with gerrymandering and single-member FPTP rules that shut out third parties. FPTP alone is a hurdle, but it’s not insurmountable (as is seen in the UK and elsewhere.)
Pingback: Could there be a connection between Democratic presidents and violent Christian activism?