Data: You Don’t Know What You’ve Got Till It’s Gone
Why Read This
What Makes This Article Worth Your Time
Summary
What This Article Is About
Tim Harford argues that official statistics are enormously valuable yet chronically underappreciated, partly because statistical agencies have rarely attempted rigorous cost-benefit analysis of their own work. He contrasts the libertarian scepticism of Sir John Cowperthwaite — who refused to collect economic data as Hong Kong’s financial secretary in the 1960s — with a US National Academies report estimating that the government data-intensive sector generated nearly $800 billion in revenue in 2022, roughly 112 times the entire budget of US statistical agencies.
The article’s centrepiece is a NBER working paper studying what happened after President Trump fired Erika McEntarfer, head of the Bureau of Labor Statistics, while publicly claiming the jobs numbers were “RIGGED.” Using the Economic Policy Uncertainty Index — which tracks economic anxiety through newspaper coverage — the researchers estimated that the resulting damage to institutional credibility may have cost the US economy nearly $20 billion. Harford concludes with a powerful ratio: statistical agencies consume just one dollar in every thousand of federal spending, yet their destruction can cost many times their entire budget in a single week.
Key Points
Main Takeaways
Statistics Lack a Price Tag
Statistical agencies rarely quantify their own value; a 2018 UN report offered only qualitative arguments despite measuring almost everything else.
Data Has Enormous Private Value
The US government data-intensive sector generated nearly $800 billion in revenue in 2022, 112 times what statistical agencies actually cost.
Credibility Is Fragile
Trump’s dual attack — replacing the BLS chief while alleging fraud — damaged the agency’s credibility with both supporters and opponents simultaneously.
Uncertainty Has a Price
Researchers estimated that the spike in economic uncertainty caused by McEntarfer’s firing generated nearly $20 billion in economic damage in a single week.
Ignorance Helps No One
Harford challenges Cowperthwaite’s logic: withholding data doesn’t restrain government meddling, it only makes any intervention clumsier and less informed.
One Dollar in a Thousand
US statistical programmes consume just 0.1% of federal spending, yet their value — and the cost of undermining them — runs into hundreds of billions of dollars.
Master Reading Comprehension
Practice with 365 curated articles and 2,400+ questions across 9 RC types.
Article Analysis
Breaking Down the Elements
Main Idea
The Hidden Cost of Destroying Data Credibility
Harford’s central argument is that official statistics are a vastly underpriced public good — cheap to produce, but catastrophically expensive to undermine. When political actors attack the credibility of institutions like the BLS, the resulting uncertainty ripples through the entire economy, generating costs that can dwarf the agencies’ annual budgets many times over.
Purpose
To Defend Institutional Statistics Through Economic Evidence
Harford writes to persuade readers — and policymakers — that official statistics deserve active protection, not passive appreciation. By anchoring the argument in hard dollar figures ($800bn in private sector value, ~$20bn in damage from one firing), he moves the conversation from abstract democratic values to concrete economic cost-benefit reasoning that is difficult to dismiss.
Structure
Contextual → Evidentiary → Analytical → Prescriptive
The article opens by framing the problem (statistics lack a price tag), introduces a historical counterargument (Cowperthwaite), refutes it with modern data (National Academies), then uses the Joni Mitchell principle to pivot to the BLS case study. Bloom et al.’s NBER paper provides the quantitative core, before Harford closes with a memorable spending ratio as a policy prescription.
Tone
Measured, Wry & Empirically Grounded
Harford writes with the calm authority of an economist who has seen data misused before. He is wry — invoking Joni Mitchell and acknowledging the limits of every estimate he cites — but never alarmist. The tone is that of a fair-minded analyst presenting evidence rather than a polemicist, which makes the quiet devastation of his final arithmetic all the more effective.
Key Terms
Vocabulary from the Article
Click each card to reveal the definition
Build your vocabulary systematically
Each article in our course includes 8-12 vocabulary words with contextual usage.
Tough Words
Challenging Vocabulary
Tap each card to flip and see the definition
Relating to or expressed in terms of qualities and descriptions rather than numerical measurements or quantities.
“…packed with qualitative ideas about how statistics were useful…”
A powerful and influential government official or bureaucrat, especially one associated with a centralised or interventionist administration.
“…it would only encourage the Whitehall variety of mandarin to interfere.”
Relating to a political philosophy that prioritises individual freedom and minimal government intervention in both economic and personal affairs.
“…a colonial possession pursuing a libertarian path on the opposite side of the world…”
One distinct part or component of a strategy or attack that operates alongside others to achieve a combined effect.
“…this was a two-pronged attack on the credibility of the BLS.”
A very small or thin portion of a larger whole; used figuratively to describe a negligible fraction of a total quantity.
“The estimated damage from the affair, while a tiny sliver of US GDP…”
An assumption or proposition that is taken for granted as the basis of an argument, often without being explicitly stated or examined.
“The second is the unexamined premise that only a government might find official statistics useful.”
Reading Comprehension
Test Your Understanding
5 questions covering different RC question types
1Sir John Cowperthwaite refused to collect economic data for Hong Kong because he believed the data would be technically inaccurate and therefore misleading to policymakers.
2According to the article, what was the approximate revenue of the US government data-intensive sector in 2022?
3Which sentence best captures the central principle that Harford uses to frame his investigation into the value of official statistics?
4Evaluate each statement about the NBER working paper on the value of reliable statistics.
The paper was co-authored by Erica Groshen, a former head of the Bureau of Labor Statistics.
The researchers concluded that the Economic Policy Uncertainty Index rose solely because of Trump’s statements about rigged jobs numbers.
The researchers’ preferred estimate of economic damage — after isolating coverage of McEntarfer’s firing — was approximately $20 billion.
Select True or False for all three statements, then click “Check Answers”
5What can most reasonably be inferred about why Harford describes Trump’s attack on the BLS as “two-pronged”?
FAQ
Frequently Asked Questions
The Economic Policy Uncertainty (EPU) Index, developed by Stanford’s Nicholas Bloom and colleagues about 15 years ago, measures economic anxiety by analysing the text of major US newspapers. It spikes when coverage focuses on policy confusion or instability. In this study, the EPU was used to track and quantify the increase in economic uncertainty that followed Trump’s firing of BLS chief Erika McEntarfer, allowing researchers to estimate resulting economic damage.
The estimate faces two layers of uncertainty. First, several major events coincided — a large downward jobs revision and Federal Reserve Governor Adriana Kugler’s resignation occurred the same day as McEntarfer’s firing, making it hard to isolate each event’s contribution to the EPU spike. Second, the EPU measures what newspapers find newsworthy, which may not perfectly track actual economic harm. Harford honestly acknowledges these limitations while maintaining that the estimate is still directionally meaningful.
Harford identifies two flaws. First, the hope that ignorance restrains government intervention may be misplaced — it could instead simply make that intervention clumsier and less effective, since governments will act regardless. Second, Cowperthwaite’s logic rests on the unexamined assumption that only governments benefit from official statistics. In reality, private businesses, researchers, and individuals also depend on trustworthy public data for decisions ranging from site selection to investment strategy.
Readlite provides curated articles with comprehensive analysis including summaries, key points, vocabulary building, and practice questions across 9 different RC question types. Our Ultimate Reading Course offers 365 articles with 2,400+ questions to systematically improve your reading comprehension skills.
This article is rated Intermediate. Tim Harford writes in an accessible journalistic style, but the piece requires readers to track multi-layered economic arguments, follow chains of evidence from multiple studies, and parse terminology such as the Economic Policy Uncertainty Index, laissez-faire, and cost-benefit analysis. It also demands inferential reasoning — understanding what Harford implies about institutional credibility, not just what he states directly. Suitable for readers with some background in economics or current affairs.
Tim Harford is a British economist, Financial Times columnist, and bestselling author best known for the “Undercover Economist” series. He specialises in explaining how economics illuminates everyday life and institutional behaviour. His book How to Make the World Add Up — referenced at the article’s close — focuses specifically on statistics and data literacy. His column in the Financial Times gives him a platform to bring academic research, such as the NBER working paper, to a wide and informed readership.
The Ultimate Reading Course covers 9 RC question types: Multiple Choice, True/False, Multi-Statement T/F, Text Highlight, Fill in the Blanks, Matching, Sequencing, Error Spotting, and Short Answer. This comprehensive coverage prepares you for any reading comprehension format you might encounter.
