Voice For Liberty

Tag: Politics

  • Durham Report Annex

    (Assisted by Claude AI)

    This document is a declassified appendix from a U.S. government report about intelligence activities and investigations related to the 2016 presidential election campaigns. Let me break down the key elements to help you understand what it contains:

    Document Overview

    This is a “Classified Appendix” dated May 12, 2023, that was authorized for public release by Chairman Grassley. It’s part of a larger report examining intelligence matters from the 2016 election period, with many portions redacted (blacked out) for security reasons.

    Main Sections

    1. Russian Intelligence Hacking Background

    The document describes how, beginning around 2014, individuals affiliated with Russian intelligence services conducted extensive hacking operations against U.S. entities, including:

    • Government agencies
    • Non-profit organizations
    • Think tanks
    • The Open Society Foundation (formerly the Soros Foundation)

    2. The Leonard Benardo Emails

    A significant portion discusses emails allegedly from Leonard Benardo of the Open Society Foundation. The document describes:

    • A January 2016 memorandum discussing FBI investigations into Clinton Foundation corruption allegations
    • References to Democratic Party concerns about Hillary Clinton’s email scandal
    • Speculation about FBI Director James Comey’s intentions
    • A March 2016 memorandum discussing campaign strategy

    3. The Alleged Clinton Campaign Plan

    The document references intelligence suggesting that in July 2016, U.S. intelligence agencies obtained information indicating:

    • Hillary Clinton allegedly approved a campaign plan to stir up a scandal against Donald Trump
    • This plan supposedly involved tying Trump to Putin and Russian election hacking
    • The information came from Russian intelligence analysis

    4. FBI Response and Investigation

    The document details how various U.S. officials, including:

    • FBI personnel
    • CIA Director Brennan
    • Attorney General Lynch
    • Other intelligence officials

    …handled this information, with discussions about its credibility and whether/how to investigate it.

    Important Context

    The document repeatedly emphasizes uncertainty about the authenticity of these materials. It notes that:

    • Some analysts believed the emails might have been fabricated or altered by Russian intelligence
    • The CIA assessed the information but couldn’t determine if it was genuine or Russian fabrication
    • Multiple officials expressed skepticism about the credibility of the intelligence

    Educational Takeaway

    This document illustrates the complexity of intelligence work during politically sensitive periods. It shows how intelligence agencies must evaluate potentially compromised or fabricated information while navigating political pressures. The extensive redactions and careful language throughout demonstrate the sensitive nature of intelligence sources and methods, even years after the events in question.

    The document essentially presents a case study in how foreign intelligence operations can create confusion and uncertainty, potentially influencing domestic political processes through disinformation or selectively leaked information.

  • Cost of Voting in States

    Cost of Voting in States

    (Analysis assisted by Claude)

    This chart reveals a striking pattern in American voting: there’s a clear relationship between how a state voted in 2024 and how difficult that state makes it for people to vote.

    The horizontal axis shows each state’s percentage of Trump voters in 2024, ranging from about 30% on the left to 70% on the right. The vertical axis shows the Cost of Voting Index, where positive numbers mean voting is more difficult and time-consuming, while negative numbers mean voting is more accessible and streamlined.

    The most important pattern to notice is the upward slope of the data points. States that gave Trump higher vote percentages tend to cluster toward the top of the chart (higher voting costs), while states that gave him lower percentages cluster toward the bottom (lower voting costs). This isn’t just a loose correlation – it’s quite pronounced.

    For example, look at the extremes. Mississippi, which gave Trump about 60% of its votes, sits at the very top with a Cost of Voting Index around 2.0 – meaning voting there requires significantly more time and effort than average. On the opposite end, states like Washington, Vermont, and Hawaii, which gave Trump only about 30-35% of their votes, have negative index scores, meaning they’ve made voting notably easier than the national baseline.

    To understand what this means practically, remember that the Cost of Voting Index measures things like registration deadlines, voter ID requirements, early voting availability, and absentee ballot procedures. A state with a high score might require voters to register weeks in advance, provide specific forms of ID, have limited early voting days, and restrict who can vote absentee. A state with a low score likely allows same-day registration, accepts multiple forms of ID, offers extensive early voting, and makes absentee voting widely accessible.

    The bubble sizes represent turnout rates, which adds another layer to consider. Notice that many of the states with higher voting costs don’t necessarily have the smallest bubbles – suggesting that even when voting is made more difficult, people still turn out, though we’d need to dig deeper to understand whether turnout might be even higher if voting were easier.

    This pattern raises important questions about the relationship between political preferences and voting policies. Are states with Republican majorities more likely to implement policies that make voting more difficult? Are there philosophical differences about election security versus accessibility driving these policy choices? The data suggests these aren’t random differences but reflect systematic approaches to election administration that correlate with political outcomes.

  • White House: BLS Has Lengthy History of Inaccuracies, Incompetence

    (Research assisted by AI)

    This White House article presents a highly critical assessment of the Bureau of Labor Statistics under the previous administration. It accurately identifies notable BLS revisions, a significant benchmark adjustment, and procedural lapses in August 2024. However, its framing of these events as evidence of systemic incompetence does not fully account for the agency’s established revision protocols or its enduring reputation for nonpartisan data quality.

    Summary
    The Bureau of Labor Statistics (BLS), under former Commissioner Erika McEntarfer, has repeatedly published overly optimistic employment figures that were later revised downward, eroding confidence in the agency’s data (The White House, 2025). For example, initial May and June 2025 job gains were overstated by a combined 258,000 positions, figures which were quietly adjusted after influencing Federal Reserve policy (The White House, 2025). Similarly, March 2024 payroll growth was pared back by 818,000 jobs—the second?largest benchmark revision on record—while revisions throughout 2024 further reduced reported gains by tens of thousands each month (The White House, 2025).

    Chronic technical failures and premature data leaks compounded the problem, as BLS repeatedly delayed or mishandled releases of critical statistics. In August 2024, a technical glitch postponed the public jobs report, yet several financial firms accessed the data early, contravening the agency’s own nondisclosure protocols (The White House, 2025). Congressional Republicans have held multiple hearings and sent oversight letters spotlighting these missteps and demanding accountability, highlighting the agency’s pattern of miscommunication, data security lapses, and diminished reliability (The White House, 2025).

    Accuracy of Major Claims

    1. May–June 2025 Revisions
      The article states that initial job gains for May and June 2025 were overstated by a combined 258,000 positions and later revised downward. According to the Bureau of Labor Statistics (BLS), May’s gain was revised from +144,000 to +19,000 (–125,000) and June’s from +147,000 to +14,000 (–133,000), for a total downward revision of 258,000 jobs (Bureau of Labor Statistics, 2025a). Financial?market responses confirm this adjustment rattled investors and fed into Federal Reserve deliberations (Jones, 2025).

    2. March 2024 Benchmark Revision
      The article correctly notes that the preliminary benchmark revision for March 2024 reduced nonfarm payrolls by 818,000 – the second-largest on record. BLS’s own announcement confirms an estimated downward adjustment of –818,000 jobs for the 12-month period ending March 2024 (Bureau of Labor Statistics, 2024).

    3. Revisions Throughout 2024
      It is true that BLS routinely issues monthly revisions as more complete data arrive. For example, in July 2024 BLS revised May employment down by 54,000 and April by 57,000; in April 2024 it further pared back February’s figures, and in February 2024 it reduced December 2023 by 43,000 and January 2024 by 124,000 (White House, 2025). However, these adjustments reflect BLS’s standard methodology of incorporating late survey returns and recalculating seasonal factors – practices that have been in place for decades – rather than a new pattern of systematic “overstatement” (Bureau of Labor Statistics, 2025b; Bloomberg, 2025).

    4. Technical Failures and Data Leaks
      In August 2024 a technical error delayed the public release of the jobs revision report, prompting outrage among policymakers and market participants (Gurley & Siegel, 2024). During that delay, at least three Wall Street firms obtained the unreleased figures in advance – an incident confirmed by Bloomberg and probed by a Republican-led House committee (Smith et al., 2024; House Education and the Workforce Committee, 2024).

    5. Congressional Oversight
      The article’s claim that Congressional Republicans have held hearings and sent oversight letters is accurate. In September 2024, Rep. Virginia Foxx’s Committee on Education and the Workforce formally requested details on both the leak and the procedural failures that allowed selective early access (House Education and the Workforce Committee, 2024).

    Evaluation of Conclusions

    Normalcy vs. Incompetence
    While the White House article emphasizes a “lengthy history of inaccuracies,” the documented revisions are largely consistent with BLS’s transparent revision process, which exists precisely to refine initial estimates (Bureau of Labor Statistics, 2025b). Benchmark revisions, by their nature, can be large occasionally (e.g., –818,000 in March 2024), but such events occur roughly once a year when comprehensive unemployment insurance tax data become available.

    Isolated Glitches vs. Systemic Failures
    The August 2024 glitch and the premature disclosure to select firms were serious lapses in protocol, yet they are relatively rare given BLS’s long track record of dependable releases. Since 1979, the agency has operated under strict “lock-up” procedures to ensure simultaneous public access—procedures it has amended in response to these incidents (Bloomberg, 2024).

    Reputation and Independence
    Despite these missteps, the BLS remains widely respected among economists for its methodological rigor and transparency (Miran, 2025; Reuters, 2025). Its willingness to issue large downward revisions and to publicly document its data-processing methodology underpins its credibility. The article’s portrayal of “incompetence” overstates the frequency and severity of errors in a data?rich environment where preliminary estimates are by design subject to revision.

    Overall Assessment
    The White House article accurately identifies notable BLS revisions, a significant benchmark adjustment, and procedural lapses in August 2024. However, its framing of these events as evidence of systemic incompetence does not fully account for the agency’s established revision protocols or its enduring reputation for nonpartisan data quality. A balanced view recognizes both the need to shore up procedural safeguards and the value of BLS’s transparent approach to correcting its own estimates.

    References

    The White House. (2025, August 1). BLS has lengthy history of inaccuracies, incompetence. https://www.whitehouse.gov/articles/2025/08/bls-has-lengthy-history-of-inaccuracies-incompetence/

    Bureau of Labor Statistics. (2024, July 5). 2024 preliminary benchmark revision to establishment survey data to be released on August 21, 2024. https://www.bls.gov/ces/notices/2024/2024-preliminary-benchmark-revision.htm

    Bureau of Labor Statistics. (2025a, July 30). Employment situation summary—2025 M07 results. https://www.bls.gov/news.release/pdf/empsit.pdf

    Bureau of Labor Statistics. (2025b). Frequently asked questions: Why does the establishment survey have revisions? https://www.bls.gov/web/empsit/ces\_cps\_trends.htm#Revisions-Between-Preliminary-and-Final-Data

    Bloomberg. (2024, August 30). BLS data slipups are becoming a pattern.

    Bloomberg. (2025, August 1). Biggest job revisions since 2020 expose pitfall of economic data.

    Gurley, L. K., & Siegel, R. (2024, August 28). Technical error caused jobs data delay that sparked outrage, BLS says. The Washington Post.

    House Education and the Workforce Committee. (2024, September 25). Letter to Acting Secretary Julie Su on BLS data release issues.

    Jones, A. (2025, August 1). U.S. payrolls revisions jolt markets, making Fed look behind the curve. Reuters.

    Miran, S. (2025). Comment on BLS independence and data quality. Axios.

  • Gross Domestic Product per Person in States

    Gross Domestic Product per Person in States

    (Analysis by Claude)

    This scatter plot reveals a fascinating relationship between political preferences and economic prosperity across US states. Let me walk you through what we’re seeing here.

    The chart plots each state’s per capita GDP in 2022 (measured in inflation-adjusted 2017 dollars) against the percentage of votes Donald Trump received in that state during the 2020 presidential election. The size of each bubble corresponds to the state’s population, so larger states appear as bigger circles.

    The most striking pattern is the clear negative correlation: as Trump’s vote share increases (moving right on the chart), per capita GDP tends to decrease (moving down). This creates a downward-sloping cloud of states from the upper left to the lower right.

    Let’s examine some key clusters and outliers:

    High GDP, Low Trump Support: States like Massachusetts, New York, California, and Connecticut cluster in the upper left corner. These states have per capita GDPs above $75,000 and gave Trump less than 40% of their votes. Washington state also fits this pattern. These tend to be states with major metropolitan areas and knowledge-based economies.

    Low GDP, High Trump Support: States like Mississippi, West Virginia, and Arkansas appear in the lower right, with per capita GDPs below $50,000 and Trump vote shares above 60%. These are generally more rural states with economies historically based on agriculture, mining, or manufacturing.

    Notable Exceptions: Some states break the pattern in interesting ways. Alaska, for instance, has relatively high per capita GDP (around $70,000) despite strong Trump support (about 53%). This likely reflects Alaska’s oil wealth. North Dakota and Wyoming show similar patterns, also benefiting from natural resource extraction.

    Middle Ground: States like Texas, Florida, and Georgia occupy a middle position, with moderate Trump support (around 50-55%) and middle-range per capita GDPs ($60,000-65,000).

    This correlation likely reflects several underlying factors. Wealthier states tend to have more college-educated residents, more diverse economies, and larger urban populations – all demographics that leaned away from Trump in 2020. Meanwhile, states with lower per capita GDP often have more rural populations and traditional industries that have faced economic challenges from globalization and technological change – communities where Trump’s message resonated strongly.

    However, it’s crucial to remember that correlation doesn’t imply causation. The relationship between voting patterns and economic prosperity is complex and influenced by many factors including geography, history, education levels, industrial composition, and cultural values. The chart shows an association, but voting for or against Trump doesn’t directly cause high or low GDP, nor does GDP determine voting behavior in a simple way.

  • Gross Domestic Product in Counties

    Gross Domestic Product in Counties

    (Analysis by Claude)

    This scatter plot reveals a fascinating relationship between economic output and voting patterns at the county level in the United States. Let me walk you through what we’re seeing here.

    Understanding the Axes

    The vertical axis shows each county’s Gross Domestic Product (GDP) – essentially the total value of all goods and services produced in that county in 2022. Notice that this axis uses a logarithmic scale, which is crucial for interpretation. On a log scale, the distance between $10 million and $100 million is the same as the distance between $100 million and $1 billion. This allows us to see patterns across counties with vastly different economic outputs on the same chart.

    The horizontal axis shows what percentage of voters in each county voted for Trump in the 2020 election, ranging from 0% to 100%.

    The Main Pattern

    The downward-sloping trend line reveals the key finding: there’s a negative correlation between county GDP and Trump vote share. In simpler terms, counties with higher economic output tended to vote less for Trump, while counties with lower economic output tended to vote more for Trump.

    Reading the Data Distribution

    The density of dots tells us important things about American counties:
    – Most counties cluster in the lower-middle portion of the chart (GDP between $100 million and $10 billion, Trump vote share between 40-80%)
    – Very few counties have extremely high GDPs (above $100 billion) – these are likely major metropolitan areas
    – The vertical spread at any given vote percentage shows the diversity of economic conditions across politically similar counties

    Why This Matters

    This visualization captures a significant divide in American politics – the economic productivity gap between areas with different political preferences. The wealthiest, most economically productive counties (often containing major cities and their suburbs) leaned heavily against Trump, while less economically productive rural and small-town counties showed stronger Trump support.

    It’s worth noting that correlation doesn’t imply causation – this chart doesn’t tell us whether economic factors directly influenced voting patterns, or whether other factors (like education levels, urban vs. rural settings, or demographic differences) might explain both the economic and political patterns we see.

  • Women, Infants, and Children (WIC) Spending by State

    Women, Infants, and Children (WIC) Spending by State

    Participation and spending on WIC by state political sentiment.

    The Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) is a federal assistance initiative that provides nutritious foods, education, and support to low-income pregnant and postpartum women, infants, and children up to age five. Funding for WIC is primarily allocated by the U.S. Department of Agriculture’s Food and Nutrition Service (USDA-FNS) to state agencies, which then administer the program at the local level.

    Data on the program is available at WIC Data Tables.

    I’ve plotted data by state, using the political sentiment of the state as a dependent variable. This is measured by the portion of the vote Donald J. Trump received in 2024. Both spending and participation vary little based on a state’s vote for Trump. (Click for larger charts)

  • Voting and Registration in States and Counties

    Voting and Registration in States and Counties

    Charts of voting and registration in states and counties. (more…)

  • Adult Literacy and Numeracy in Counties

    Adult Literacy and Numeracy in Counties

    Examining literacy and numeracy based on political preference. (more…)

  • Republicans Will Encourage Voting Before Election Day

    This article, written by Walter Olson and published on Cato Institute, discusses a significant shift in the Republican Party’s stance on voting methods. Former President Donald Trump, who once criticized alternate voting methods as prone to fraud and urged Republicans to vote on Election Day, and the GOP are now encouraging the use of various balloting methods, including vote by mail, early in-person voting, and the use of drop boxes. This change is largely seen as positive, allowing campaigns to allocate resources more efficiently and appealing to a broader range of constituents. However, the article also raises concerns about the potential risks associated with third-party ballot collection, emphasizing the need to balance convenience in voting with maintaining the secrecy and security of ballots.

    Olson, Walter. “Republicans Will Encourage Voting Before Election Day.” Cato Institute, 22 Sep. 2023, https://www.cato.org/blog/republicans-will-encourage-voting-election-day