Correlation between income inequality and a partisan Congress.

[Hygro]

http://voteview.com/polarizedamerica.asp

Most of us hate the gridlock. Most of us are offended that politicians will leave their most touted positions to declare such positions anti-American just to score points. And it seems such times correlate to a rise in income inequality.

What's the cause? What's the solution? Is there a positive feedback loop?

 

[Dreadnought]

What is this "polarization index"? Sounds like an arbitrary statistic.

 

[Hygro]

What is this "polarization index"? Sounds like an arbitrary statistic.

That's what my link is for. They have lots of discussion and more graphs on the subject, throughout the whole site.

 

[Bootstoots]

Does this correlation really mean anything? Before and during the Civil Rights era, there wasn't a huge correlation between political ideology and political party - the Democrats were split between conservative Southern Democrats and liberal Northern ones, and the Republicans had a significant liberal wing based out of the Northeast. This didn't really change until the 1970s, when the Southern Democrats started becoming Republicans and liberal Republicans became Democrats.

Since then, the parties have become much more ideologically polarized: there are now very few Republicans to the left of any Democrats in Congress and vice versa. Meanwhile, income inequality rose significantly too. But I don't know that polarization is linking the two. If anything I'd suggest that the rise of neoliberalism and its impact on both parties has had much more to do with the current high levels of inequality than polarization between Republicans and Democrats.

 

[Dreadnought]

That's what my link is for. They have lots of discussion and more graphs on the subject, throughout the whole site.

To be honest, I haven't read the entire site. However, I haven't come across a specific definition for the polarization index. Is it based on the party loyalty graphs near the bottom of the page? How do we quantify the distance between the parties, as some graphs relate to?
\

 

[Cutlass]

Makes sense. Polarization is a tool of rent seeking by the economic elites. Crippling the government's ability to act in any way that does not redistribute wealth to the top clearly benefits the powerful.

 

[Integral]

Graph any two trending time series on each other and you'll get a high correlation coefficient. See: Dow Jones, Sunspots.

 

[lovett]

There are some quite obvious causal mechanisms that could explain this correlation, though.

The one that comes to mind immediately is based on the interaction between inequality and the American districting system. Let us take a basic rational choice model; voters vote on their self-interest and politicians attempt to maximize their votes. This means they hold the policy position of the median voter in their most important constituency. Let us further suppose that he rich and the poor have different economic interests (the poor, for instance, benefit from re-distribution).

In the US, constituencies are highly gerrymandered. This is for a number of reasons. Succinctly, because politicians have power over districting and it is in their interests to create safe seats. In an unequal society, differing economic class is an important line of social cleavage. Consequently, gerrymandering will have an economic dimension. The poor must be stacked into one district, the rich into another. In the former, the median voter will favour redistributive policies and in the latter oppose redistributive policies. If politicians are vote-maximizing, this means one will have politicians who fight for the interests of the poor and those who fight for that of the rich. This will, naturally, create polarisation in the legislature.

This is a rough characterization of a causal mechanism, but plausible nonetheless.

 

[Ayatollah So]

What is this "polarization index"? Sounds like an arbitrary statistic.

They use multidimensional scaling. I haven't grokked the math, but it sounds like a common and general-purpose statistical tool for summarizing complex data into a small number of dimensions. Sort of like Principal Components Analysis - which most people haven't heard of either, but probably more people have heard of PCA than MDS, so I'll mention it just in case.

 

[Hygro]

Graph any two trending time series on each other and you'll get a high correlation coefficient. See: Dow Jones, Sunspots.

But but but! This one made the graph such that it overlaid so perfect looking! It has to be completely real!

 

[Integral]

But but but! This one made the graph such that it overlaid so perfect looking! It has to be completely real!

I give everyone a hard time.

And the correlation is stunning. First-difference it and see if the correlation remains.

I am a little puzzled as to why polarization rises almost monotonically over time. I mean, things are bad now politically, but have they really never been this bad?

There are some quite obvious causal mechanisms that could explain this correlation, though.

The one that comes to mind immediately is based on the interaction between inequality and the American districting system. Let us take a basic rational choice model; voters vote on their self-interest and politicians attempt to maximize their votes. This means they hold the policy position of the median voter in their most important constituency. Let us further suppose that he rich and the poor have different economic interests (the poor, for instance, benefit from re-distribution).

In the US, constituencies are highly gerrymandered. This is for a number of reasons. Succinctly, because politicians have power over districting and it is in their interests to create safe seats. In an unequal society, differing economic class is an important line of social cleavage. Consequently, gerrymandering will have an economic dimension. The poor must be stacked into one district, the rich into another. In the former, the median voter will favour redistributive policies and in the latter oppose redistributive policies. If politicians are vote-maximizing, this means one will have politicians who fight for the interests of the poor and those who fight for that of the rich. This will, naturally, create polarisation in the legislature.

This is a rough characterization of a causal mechanism, but plausible nonetheless.

Good answer. And possibly even true. Despite the sarcasm in my initial post I think the correlation probably has some substantive content. I think both are driven by underlying, post-1973 trends that we still don't fully understand but are all intertwined: the rise in income inequality, the rise in political hackery, the rise in the top-0.1% income share, et cetera.

 

[Hygro]

Integral, you want to talk "post 1973" I am your guy. It actually started in 1965, and the seeds were fully planted in 1958. I actually agree the causal agent is most likely the same as all the other things.

I should start a thread on it, but I have a few books to read/reread first.

 

[dutchfire]

There are some quite obvious causal mechanisms that could explain this correlation, though.

The one that comes to mind immediately is based on the interaction between inequality and the American districting system. Let us take a basic rational choice model; voters vote on their self-interest and politicians attempt to maximize their votes. This means they hold the policy position of the median voter in their most important constituency. Let us further suppose that he rich and the poor have different economic interests (the poor, for instance, benefit from re-distribution).

In the US, constituencies are highly gerrymandered. This is for a number of reasons. Succinctly, because politicians have power over districting and it is in their interests to create safe seats. In an unequal society, differing economic class is an important line of social cleavage. Consequently, gerrymandering will have an economic dimension. The poor must be stacked into one district, the rich into another. In the former, the median voter will favour redistributive policies and in the latter oppose redistributive policies. If politicians are vote-maximizing, this means one will have politicians who fight for the interests of the poor and those who fight for that of the rich. This will, naturally, create polarisation in the legislature.

This is a rough characterization of a causal mechanism, but plausible nonetheless.

Would be nice to see if this effect is American or international. If gerrymandering is your explanation, this effect shouldn't happen in countries without districts, but should happen in other countries with (gerrymandered) voting districts.

 

[Leoreth]

The effect should also be stronger in countries with plurality voting systems than in those with proportional representation.

 

[downtown]

I am a little puzzled as to why polarization rises almost monotonically over time. I mean, things are bad now politically, but have they really never been this bad?
.

Nope. This is the most ideologically gridlocked congress EVAR, and it's going to get worse. Blame Newt.

 

[Cutlass]

I give everyone a hard time.

And the correlation is stunning. First-difference it and see if the correlation remains.

I am a little puzzled as to why polarization rises almost monotonically over time. I mean, things are bad now politically, but have they really never been this bad?

No, it has happened before. But only a few times. Most notably in the decade before the Civil War. Do take note of all the political similarities between now and then. And the consequences of them. But most American history isn't that polarized, and because of the nature of the American party system the party boundaries have never been as clearly defined as now.

Too all intents and purposes both the Democrats and Republicans had to be "big tent" parties after Reconstruction and before the second half of the 20th century.

Good answer. And possibly even true. Despite the sarcasm in my initial post I think the correlation probably has some substantive content. I think both are driven by underlying, post-1973 trends that we still don't fully understand but are all intertwined: the rise in income inequality, the rise in political hackery, the rise in the top-0.1% income share, et cetera.

I think you need to look at the extent that one interacts with the other, and not just "correlation does not imply causation" it.

 

[warpus]

Most of us hate the gridlock. Most of us are offended that politicians will leave their most touted positions to declare such positions anti-American just to score points. And it seems such times correlate to a rise in income inequality.

What's the cause? What's the solution? Is there a positive feedback loop?

Might it have something to do with the fact that the less income equality there is, the more ultra-rich there are, the more influence they have over your politicians (since in U.S. it is very easy to influence and/or control politicians with money legally), the more your politicians look after the interests of corporations and such instead of the people.. and since there is little incentive to look after the interests of the people, American politics devolves into a game of "Let's say whatever we can to get their votes this time around, so that we can continue to serve the people who give us money", which tends to turn into a very polarizing battlefield of idiocy?

As for the guy who said that the stat is arbitrary, technically all statistics are arbitrary.

 

[Leoreth]

Too all intents and purposes both the Democrats and Republicans had to be "big tent" parties after Reconstruction and before the second half of the 20th century.

Would you say they're not big tent parties anymore? I think the tents have just stopped to intersect.

 

[Cutlass]

Might it have something to do with the fact that the less income equality there is, the more ultra-rich there are, the more influence they have over your politicians (since in U.S. it is very easy to influence and/or control politicians with money legally), the more your politicians look after the interests of corporations and such instead of the people.. and since there is little incentive to look after the interests of the people, American politics devolves into a game of "Let's say whatever we can to get their votes this time around, so that we can continue to serve the people who give us money", which tends to turn into a very polarizing battlefield of idiocy?

As for the guy who said that the stat is arbitrary, technically all statistics are arbitrary.

This is consistent with the theory of the BOOK I've been pushing most recently.

Key to our argument in Why Nations Fail is the idea that elites, when sufficiently political powerful, will often support economic institutions and policies inimical to sustained economic growth. Sometimes they will block new technologies; sometimes they will create a non-level playing field preventing the rest of society from realizing their economic potential; sometimes they will simply violate others’ rights destroying investment and innovation incentives.

Or, from the Economist:

There are two potential candidates for extractive elites in Western economies. The first is the banking sector. The wealth of the financial industry gives it enormous lobbying power, including as contributors to American presidential campaigns or to Britain’s ruling parties. By making themselves “too big to fail”, banks ensured that they had to be rescued in 2008.

Much of current economic policy seems to be driven by the need to prop up banks, whether it is record-low interest rates across the developed world or the recent provision of virtually unlimited liquidity by the once-staid European Central Bank. The long-term effects of these policies, which may be hard to reverse, are difficult to assess.

Would you say they're not big tent parties anymore? I think the tents have just stopped to intersect.

You could make a case for that. But consider the extent to which the Republican party has purged its moderates. When all the socially liberal Republicans were forced out, the socially moderates started getting the boot as well. Old school Republicans of the Country Club variety, GHW Bush, Gerry Ford, they have a damned hard time being Republicans now. Before you had the hard core social authoritarians, and the balance of power in the party was the big money interests, which didn't care a lot about those social issues. But that's changed. Even Romney now must pander more to the social wingnuts. There is no place for what he was 10 years ago. Now to be a viable Republican candidate someone has to be both a social authoritarian and economically to the right of Reagan, who himself was to the right of nearly all Republicans from the 1930s through the 1970s. If you are not that far right, don't bother running for national office as a Republican. So I would say they were a somewhat large tent, that is getting smaller by the day.

On the Democratic side, there is a larger tent. You have the Blue Dogs, who are effectively where most Republicans were 20-30 years ago. And you have the liberals, who are not nearly as liberal as they used to be except on just a couple of social issues. So there's really not much place for real liberals in the Democratic party currently, but no place for them to go either.

 

[Cutlass]

Just watching the news and see a new book supporting the theory that American politics have never been more partisan.

http://www.npr.org/books/titles/151...e-american-constitutional-system-collided-wit

http://maddowblog.msnbc.msn.com/_news/2012/05/03/11521359-its-even-worse-than-it-looks

http://www.amazon.com/Even-Worse-Than-Looks-Constitutional/dp/0465031331