Sources here: Trusting Government Statistics
And the Nixon Jew Count
Kedits is still a new substack and I don’t claim to post regularly or frequently or consistently. So far, I’ve used this platform to explain economic data, write book reviews, and to introduce policy ideas that I have, however random!
This post inaugurates a new use: housing sources for my video content. Priority number one for any of the video platforms is to keep you from navigating away the feed, making it hard to provide links. For those of you who followed me to substack from Twitter, you know I take a lot of pride in providing sources. I find the “navigate to link in bio” to be pretty unsatisfying/convuluted.
Hence, sources here!
Today I posted a video about whether you can trust government statistics about the economy.
Tiktok failed to load.Enable 3rd party cookies or use another browser
My answer is an unequivocal yes. Government statistical agencies are marked by their transparency, integrity, and independence from political influence. They can’t be perfect because statistics is the science of estimation, but they are open in how they do things, and when they are wrong.
For example, I have a paper on how much official statistics underestimated the unemployment rate during the initial covid lockdowns. This would not be possible if transparency was not a guiding principle.
But I doubt most people’s concern comes from sampling and survey methods! No doubt it comes from the worry about political influence. If politician’s approval rating hangs on how well they economy is doing, don’t they have an incentive to interfere?
Definitely. But, thanks in part to the interference by Richard Nixon, the independence of statistical agencies and the firewall between the White House and the agency staff has been codified.
For the kid-friendly sanitized version of this story, you can read the book “The First Hundred Years of the Bureau of Labor Statistics,” available in full online for free. It recounts Nixon pushing out the BLS commissions (starts around page 221), reorganizing the staff to retaliate against certain people, and moving the press and data release around so he could control the message. His actions resulted in press coverage, calls of interference, Congressional investigations, and the adoption of new policies to insulate the BLS and all statistical agencies from interference.
But it’s much uglier in real life.
Nixon was known for paranoia and anti-semitism, which combined into the belief that a Jewish Cabal at the Fed and BLS was trying to politically undermine him. The economy during Nixon’s presidency was struggling; he immediately preceded the ‘stagflation’ of the late 1970s. He felt like his progress was real but not reflected in statistics—because of Jews. He asked a White House staff member to count the number of Jews at the BLS and arrange for them to be moved, reassigned, or fired.
Although this was initially reported in Woodward and Bernstein’s Final Days book it was largely forgotten until 1988 when it was reported on during George H.W. Bush’s presidential campaign. The staffer who counted the Jews, Fred Malek, was working for the Republican National Committee at the time. Here’s a 1988 editorial from the New York Times, which was published shortly after Malek’s role in the Jew Count was the subject of news articles. The “Jew Count Memo” dogged his remaining career, which was nonetheless illustrious.
However, the full story of the Jew Count was not revealed until 2007, when unredacted oval office tapes were made public.
You can read about this in Slate, the History News Network, the Washington Post. The HNN piece is the best because it was written by the historian at the Miller Center at the University of Virginia who maintains the transcripts for Nixon’s oval office tapes.
Here’s a highlight:
Meanwhile, Haldeman tried to find out how many BLS employees were Jews. “What’s the status of your analysis of the BLS,” he wrote to Personnel Chief Malek on July 26, “specifically of the 21 key people? What is their demographic breakdown?”
Malek replied the next day. “We were able to obtain political affiliation checks on 35 of the 50 names listed on their organization chart.” There were 25 Democrats, 5 unregistered, 4 independents, and 1 Republican. “In addition, 13 out of the 35 fit the other demographic criterion that was discussed.” There was a handwritten note: “Most of these are at the top.”
A lot of changes were made after Nixon left office, driven mainly by career statisticians who were committed to integrity, to enshrine certain practices and policies into rules that would insulate BLS and other agencies from these types of attacks.
Data is hard. Statistics is hard. But politicans don’t get to interfere. You can trust the statistics.
The coda to all this is that much like Nixon, Trump tried very hard to interfere in data and he was more successful. Most of his policies were immediately rescinded, reversed, and all were reviewed in order to produce new recommendations, but these things take time. More on that in a separate video.