A Bloc Diagram of Political Extremism

In a previous commentary here prior to the 2022 mid-term elections, a football field was used as a visual metaphor to explain how the discussion space of public issues has shifted within America over the last fifty plus years. The point of the metaphor was to show how a simplistic two dimensional view of available answers for issues can completely distract the public from recognizing that the two teams on the field aren't scrimmaging at the midfield balanced starting point of fifty years ago. In fact, the game has been shifted heavily to one side.

The football field analogy may help visualize how the origin of political solutions has shifted over decades but it doesn't do much to explain other dynamics WITHIN parties and BETWEEN parties resulting from gerrymandering and basic ignorance of democratic processes that combine to accelerate the shift and accentuate the sense of jarring shock each time something big comes out of government. The two dominant parties may appear to act as monoliths but their actual behavior is the result of blocs within the parties that have very distinct characteristics that have huge impacts on what "the monolith" does. Understanding those impacts requires modeling those underlying preferences and actions of individual blocs of voters within the parties or within independent ranks.

Devising a Model of Political Blocs

To visually illustrate how the behavior of individual blocs of voters changes over time, an ideal model would convey all of the following elements at a glance:

• The proportion of the voters in the bloc to the total number of voters
• The position of that bloc's political preferences on a simplified two dimensional spectrum of policy choices
• The "width" of policies on either side of the bloc's core preference that voters in the bloc will consistently accept
• How all of the blocs appear together on the spectrum and how their core preferences and tolerances overlap

The image below provides an example of a graph of such a "bloc" diagram.

This approach obviously merits a few hand-waving caveats. Namely:

• The horizontal axis reflecting "policy" is a numeric scale synthesized out of thin air. No one attempts to compute a single numeric "score" of any arbitrary policy and there's no way an agreed upon scheme could be devised.
• Attempting to simplify political ideas down to a artificial two-dimensional space (left / right, liberal / conservative, etc.) doesn't reflect the dimensionality of all solutions that SHOULD be available but, in a perverse way, it DOES reflect the distortion of choices imposed by an arbitrary two-party system
• The y-axis of the diagram has no direct mapping to the real world. The vertical dimension is chosen to allow all of the blocs to fit on the diagram while making their areas proportional to that bloc's share of total voters. Since the horizontal axis is in units of "policy", the vertical scale is essentially a reflection of the number of voters holding a particular policy preference. Again, not a number that can be scientifically sampled and validated.

Even with those caveats, which may not make sense at first, this diagramming approach does achieve a few key goals:

• Each bloc's area intuitively conveys its proportionate strength within its party and to the total voting population.
• Each bloc's width provides an immediately intuitive understanding of its tolerance for neighboring ideas.
• The degree of overlap between blocs reflected by stacking them atop each other provides an instant impression of the opportunities for compromise to combine ideas into a majority of votes.
• The y-axis of the diagram has no direct mapping to the real world. The vertical dimension is chosen to allow all of the blocs to fit on the diagram while making their areas proportional to that bloc's share of total voters. Since the horizontal axis is in units of "policy", the vertical scale is essentially a reflection of millions of voters holding a particular policy preference. Again, not a number that can be scientifically sampled and validated.

Unfortunately, even with this immediate intuition that might stem from a diagram like this, the fact that the underlying policy dimension of the horizontal axis is synthesized makes any ONE model rendered in this format useless for discussion. The model is far more useful by scripting it, then generating the same chart over different scenarios that alter the political target and tolerance of each bloc over time. A bloc can shift left or right on the horizontal access in terms of its core target positions. It can also expand or narrow its tolerance of adjacent policy ideas over time, increasing or reducing opportunities for compromise.

To truly capture insight from the model, scripting it then defining models based upon historical data for total voters and verifiable records for party registrations can make it possible to see patterns in overall political direction, even when exact numbers aren't available from the results.

Implementing the Model

As a brief technical aside, the model and graphing of results were implementing using a script written in Python utilizing the Manim graphing and animation library. The rectangles for each bloc are sized s that their areas are proportional to the number of voters mapped to the block. Each year's illustration performed these steps to generate the bloc bars:

1. Multiply the total voter count by each party's registration percentage.
2. For Democrat and Republican blocs, the party voter count was then evenly divided among the blocs within that party. (It can be easily argued this is not accurate and that blocs within the major parties are consolidating and tending to shift counts away from the moderate side.)
3. For Independent blocs, the center bloc was allocated 12% of the total and 44% was allocated to the remaining "Leans" blocs. This reflects actual polling data from Gallup that has remained consistent for decades.

To get all of the areas to reflect a uniform amount of voters between 1972 and 2024, an arbitrary area on the screen was chosen to represent all voters in 2024, the largest vote tally. That "density" was used as a constant scaling factor in all years so if the voter count doubles, the screen area occupied by bars for those blocs will double while ensuring the extrapolated 2024 results fit cleanly on the diagram.

The Python object model for a bloc includes a a "core target" parameter that dictates where that bloc's political preferences sit on the arbitrary x-axis of policy ranging from -18 to +18. Each bloc also includes a "shift factor" that allow that bloc's core target preference to shift a fixed percentage each election cycle.

Each bloc definition also has two parameters for "tolerance" and "tolerance shift" that reflect the range of policies acceptable to that bloc and how that tolerance range narrows or expands each election cycle. The initial state of most blocs in 1972 were arbitrarily configured with a tolerance range of 5, meaning each bloc can accept positions from -2.5 to +2.5 around their core target. The tolerance of the independent blocs were made initialy larger, reflecting the expected behavior of centrists. The tolerance shift factors of the partisan blocks were configured with NEGATIVE percentages, causing their tolerance value to DECREASE each cycle, reflecting a NARROWING of acceptable preferences. For the Independent blocs, those percentages were larger POSITIVE percentages, causing their tolerance value to INCREASE each cycle, reflecting a WIDENING of acceptable preferences.

Data Sources for the Model

The factual data used in the model consisted of figures for total voters in each Presidential election between 1972 and 2020 and percentages of voters registered in each political party or as independents. Total voter data was obtained from this web site managed by the University of California - Santa Barbara summarizing total votes cast in each Presidential election. The 2024 voter total was then estimated by applying prior growth rates to the 2020 numbers.

https://www.presidency.ucsb.edu/statistics/elections/2012

Party registration percentages for 1972 through 2004 were obtained from this Washington Post article providing registration percentages in each presidential election year.

https://www.washingtonpost.com/wp-srv/politics/interactives/independents/data-party-identification.html

Registrations for 2008 through 2024 were gathered from data collected by Gallup available here:

The party registration data is not perfect. Some states REQUIRE registration with a party in order to vote in primaries. Other states ALLOW an affiliation to be collected but do not require it. Other states do not require registration at all. Here, the available stats were used as a proxy for the nation, which could skew results SLIGHTLY.

Also, the same Gallup results cited above also provide data for "leanings" of independents over the same 1972 to 2024 range. Those leanings have remained very consistent over time, with between 42-46 percent "leaning Democrat" and 42-46 percent "Leaning Republican" with the balance claiming true centrist preferences swinging equally in both directions. The models here will assume a 44 / 44/ 12 split within independents.

Mapping the Model to Reality

As stated earlier, the dimensionless nature of "politics" on the x-axis of this visualization means that any specific graph is incapable of conveying anything specific about policy details from any legislation. However, watching the size, dimensions and relative positions of all of the blocs to each other over time can actually provide significant insight into general behaviors. With all of the data from 1972 to 2020 defined in the model, a video was generated showing the bloc view for each Presidential Election year from 1972 to 2024. You can watch the video here on YouTube:

Here are some dynamics that become obvious from watching the blocs evolve over an extended time.

Inter-Party Paralysis -- As blocs narrow their tolerance for ideas outside their core preference, it becomes more difficult to find a majority of votes for a given policy towards the middle ground. Impeding compromises on proposals positioned towards the center of political space not only slows down the pace of any legislation, it makes it less likely any legislation emerging comes from that centrist space. Instead, legislation getting enacted is more likely to come from a policy point closest to the majority party's core which is further from the center.

Intra-Party Paralysis -- As blocs narrow their tolerance for ideas outside their core preference, it not only inhibits the ability to make deals across the aisle with members of other parties, it also begins preventing coalitions WITHIN a party from forging agreements that allow legislation to be drafted and enacted. In this model, the "Fanatic" bloc of Republicans has shifted so far to the right, by 2020 it lacks any overlap with blocs within the Republican Party. Actual reality reflects that change, with a minority within the party so unwilling to compromise with ANYONE, they have already outsted their own Speaker and threaten to do so again with the replacement.

Oscillation Instead of Evolution -- As blocs within both dominant parties shift directionally towards their extreme away from the center and as those blocs become less tolerant over time, a secondary effect becomes evident to the voting public. Any legislation that DOES manage to get passed looks less like an evolution of the current state on the issue involved and more like an instantaneous jolt from one state of affairs to a vastly different state of affairs. Of course, this whipsaw effect on public policy is the LAST thing society needs in areas like environmental regulation, civil rights, etc. Having one party support automakers by trying to create long-term demand for electric vehicles then having another party attempt to support automakers by further relaxing emissions rules and ending EV rebates can entice a maker to spend billions on investments whose value is tanked by a whiplash change in policy that shifts consumer demand instantly.

Spite as a Form of Alternate Political Success -- In a climate where actually ENACTING new legislation is nearly impossible, sheer spite has emerged as the most common form of political victory pursued. Spite has an advantage over actual progress because politicians have discovered it is vastly easier to achieve, not only due to unique flaws built into the United States Constitution regarding representation but also mere parliamentary traditions adopted decades / centuries ago within the House and Senate that provide a minority of a minority to block consideration of bills or appointments in the Executive branch or the courts. The most intolerant politicians and the voters who support them have more frequently concluded over the past two decades that if they cannot WIN what they want via the voting booth, they will DAMN SURE ensure the actual winners won't get what they want either. And the system provides ample tools to make good on such threats.

Tolerance of Independents Is Stretched -- By necessity, a political climate only allowing policies towards the flanks to become law requires the tolerance of Independents to grow over time, in contrast to the tolerance of politicians and voters remaining aligned with the dominant parties. It's not necessarily a sign of ACTUAL preferences of those in the independent blocs, only a reflection of the fact that with no choices being presented that lie in their actual PREFERRED range, their only choice is to ACCEPT a wider range of policies while remaining engaged or to abandon participation entirely.

Predictions and Implications

A model that only rationalizes an explanation for past events is limited in value. Does this bloc model provide any predictive value? As stated in the caveats, the "units" for the horizontal axis of policy choices lack any specific tie to particular issues, they simply reflect a degree of "difference" between different people. This model can't predict how the terms of a bill scoring a -3 on the scale would differ from the terms of a bill scoring a +3. However, the shifts in positions of blocs and their overlaps with other blocs do illustrate how cooperation will continue to break down in efforts to solve large problems in the political sphere.

A Very Risky Environment for Capital Investments -- The prospect of hardliner blocs within the two parties attempting to continually undo any legislature or crucial court battles made possible under the other party's control one of the branches of government creates enormous risk for businesses making investment decisions in industries impacted by government policies. The dominant parties have widely divergent views on critical topics such as climate change, financial regulation, intellectual property rights and privacy rights. All of these areas require hundreds of billions of dollars in investments in coming years or threaten the safety of trillions of dollars in assets. These are not areas where rapid swings between contradictory policies every four to eight years will produce optimial long-term outcomes for the United States or the world.

No Corrective / Stabilizing Force -- The behaviors of voters and politicians acting from within these blocs are not self-correcting and self-balancing towards a neutral equilibrium. Narrower tolerances over time WITHIN blocs limits opportunities for cooperation and compromise BETWEEN blocs, even those of the same party. This makes government less able to respond to new issues and voter expectations. A less effective government lessens interest in voting from those who feel their needs are unmet, producing a positive feedback cycle ignoring centrist interests.

No Improvement Without Reducing Gerrymandering -- The principal force behind narrowing tolerances of blocs is the gerrymandering of US and State voting districts. Allowing politicians to select their voters insulates politicians from centrist interests, creating an environment where the only threat to their re-election are candidates with more extreme views.

Growing Voter Apathy -- Between 1970 and 2020, the population of the United States grew from 203,211,926 to 331,449,281, an increase of 63 percent over fifty years. Vote totals in Presidential elections grew from 76,340,294 in 1972 to 155,507,476, an increase of 103 percent. Over the entire period, that reflects an increase in voter participation (at least in Presidential election years) from about 37% to 46%. However, dragging back and forth through the yearly graphs shows years of DROPS in election turn-out despite an ever-growing population. Were those drops due to voters' delight in their circumstances at those times? Not likely. The turnout dips in 1988 (Bush versus Dukakis), 1996 (Clinton versus Dole) and 2012 (Obama versus Romney) seem to be the yawning reaction of the public to a mediocre candidate offered up by one or both of the parties. It's also possible that as reluctant independents continue to see major parties ignore them and pass legislation that continues to whipsaw the country between extremes, the very independents most in a position to AFFECT the direction in such hair-trigger electoral times may become so disgusted with the lack of consistency that they abandon particpation. Frankly, that may be exactly what many in the dominant political parties want, to the country's detriment.

No Evolutionary Fix for Corrupt, Two-Party Dominance -- This is really a corollary to the prior points. No aspects of the forces in this model create self-correcting, moderating forces to shift conditions back to a center. Because incumbent politicians in majority parties can select their voters, House legislative bodies at the Federal and State levels that are SUPPOSED to be the most immediate, small-D democratic representations of the People's interest are in fact very UN-representative of the people's will. This eliminates any external force for altering parliamentary practices that protect the existing TWO dominant parties. There can be no GRADUAL change that alters the behavior of an existing party towards moderation or ALLOWS a third party to participate in the process and obtain meaningful levels of power to reflect voters who might support it. The only way change will occur with the existing parties is via internal collapse or some political force majeure that cuts through the corruption and decades of insulation from reality and grenades the status quo. Given the types of self-induced problems experienced by the United States in past decades -- terrorist attacks, a tragically mis-managed war of retribution in Afghanistan, a tragically mis-managed war in Iraq based upon fabricated intelligence pitched to the American public, a worldwide financial collapse stemming from American "innovations" in financial engineering -- it is a sobering process to contemplate the level of calamity that would shock American out of the current stupor that is allowing these parties to carry on in the current mode of operation.

WTH