I felt that the Understanding the US Government lecture series would be useful because a detailed, fluent understanding of how the US government works and can be influenced seems useful for AI governance work
Should you read this?
I did find the lecture series useful
Probably especially the first half
But I expect that there are better resources on this topic
Though I donāt know what they are
Things I saw as problems with the lecture series:
Victor often covered info that seemed basic to me (e.g., what public goods or the Cold War are) as if itād be new to the listener
Though at least I could then just up the playback speed to 2.7-3.3
She sometimes made claims that were unclear and/āor that Iām skeptical of
Iād guess if I fact-checked the lecture series thoroughly, Iād find several errors
A decent fraction of that content didnāt seem very relevant to my interests
E.g., a chapter focused on things like social security benefits
Victor says the 3 main factors that historically have the highest predictive power for presidential election outcomes are:
1. Incumbency status [A party thatās held the presidency for 1 term and nominates the incumbent has an advantage. A party thatās held the presidency for 2 terms has a disadvantage.]
2. Incumbency approval rating
3. Status of the economy
How many cabinet departments does the US have?
15
How many civilian employees does the US bureaucracy have?
2.1 million
Victor describes the US executive branch as being organised into 5 buckets:
The White House The executive office of the President [though she then says this tends to be considered part of the white house] The 15 cabinet departments Independent agencies (both regulatory and non-regulatory) Government corporations
The Supreme Court can hear cases from a federal court of appeal or a state supreme court if it satisfies three rules of access:
Controversy Standing Mootness
[These rules as necessary but not sufficient.]
Victor highlights 3 deep root sources of partisan polarisation in the US:
1. Worsening economic inequality 2. Realignment of political parties over issues of race 3. To some extent, changes in campaign finance laws [in particular, changes that mean politicians have to rely more on small donors relative to large donors, since small donors tend to be more ideologically driven]
Victor lists 2 things people often believe contribute to polarisation but for which the evidence either doesnāt clearly support or contradicts such a causal link:
Gerrymandering [she notes that polarisation is similarly strong in the Senate] The media [but then she indicates that polarised or fake news is indeed important?]
Victor says there are 7 types of āorganised interestsā (in US politics):
Businesses/ācorporations Trade associations Professional associations Citizen groups Issue groups Labour unions Think tanks, foundations, and institutes
What percentage of Us government spending is mandatory spending (rather than discretionary spending)?
60%
[This is money the gov is committed by law to spending, and is unaffected by the appropriations process.]
What percentage of US gov discretionary spending is for defense spending?
About 50%
How much does the US gov spend per year on non-defense discretionary spending?
Notes on Victorās Understanding the US Government (2020)
Why I read this
Iām interested in learning more about a wide variety of topics relevant to ālongtermism-motivated AI governance/āstrategy/āpolicy research, practice, advocacy, and talent-buildingā
I decided that one strategy I should try for that purpose is listening to relevant Great Courses lecture series via Audible
This decision was loosely informed by advice at the end of the post The Neglected Virtue of Scholarship
See also
I felt that the Understanding the US Government lecture series would be useful because a detailed, fluent understanding of how the US government works and can be influenced seems useful for AI governance work
Should you read this?
I did find the lecture series useful
Probably especially the first half
But I expect that there are better resources on this topic
Though I donāt know what they are
Things I saw as problems with the lecture series:
Victor often covered info that seemed basic to me (e.g., what public goods or the Cold War are) as if itād be new to the listener
Though at least I could then just up the playback speed to 2.7-3.3
She sometimes made claims that were unclear and/āor that Iām skeptical of
Iād guess if I fact-checked the lecture series thoroughly, Iād find several errors
A decent fraction of that content didnāt seem very relevant to my interests
E.g., a chapter focused on things like social security benefits
My Anki cards
For why Iām sharing these, see Suggestion: Make Anki cards, share them as posts, and share key updates.
Victor says the 3 main factors that historically have the highest predictive power for presidential election outcomes are:
How many cabinet departments does the US have?
How many civilian employees does the US bureaucracy have?
Victor describes the US executive branch as being organised into 5 buckets:
The Supreme Court can hear cases from a federal court of appeal or a state supreme court if it satisfies three rules of access:
Victor highlights 3 deep root sources of partisan polarisation in the US:
Victor lists 2 things people often believe contribute to polarisation but for which the evidence either doesnāt clearly support or contradicts such a causal link:
Victor says there are 7 types of āorganised interestsā (in US politics):
What percentage of Us government spending is mandatory spending (rather than discretionary spending)?
What percentage of US gov discretionary spending is for defense spending?
How much does the US gov spend per year on non-defense discretionary spending?
How much did the US gov spend in 2019?