Singapore News Asia article on August 28, original title: What happens if we stop trusting US economic data? The monthly US employment data is one of the most closely watched measures of the health of the world's largest economy, and financial marekt movements can fluctuate wildly depending on how the data performs. Just hours after the US Department of Labor released employment data for the non-agricultural sector on August 1 that were significantly weaker than market expectations, US President Donald Trump called the data "fake" and ordered the firing of Erica McIntaffer, director of the Bureau of Labor Statistics, a division of the Labor Department. This is an unprecedented shock to the independent federal statistical agency, the Bureau of Labor Statistics. Remember that the agency is responsible for tracking employment, wage, and inflation data, and these key data reflect the true state of the U.S. economy.
And just last week, Mr. Trump launched a further attack on US economic institutions. He tweeted on social media that he had fired Lisa Cook, a governor of the US Federal Reserve, the central bank. To which Mr. Cook responded that Mr. Trump had no right to fire her and that she would continue in office. The series of events raises the key question: can we still trust US economic data in the context of Mr. Trump's "war" on data and related institutions?
Institutional independence has been eroded
Following his dismissal, Trump nominated his loyal advocate, E.J. Anthony, an economist at the Conservative Think Tank Traditional Foundation, who proposed that the Bureau of Labor Statistics should suspend the publication of employment data. Employment data, as one of the key indicators for observing the U.S. economy, can visually reflect the number of new or lost jobs each month.
On the surface, replacing the head of the Bureau of Labor Statistics with Trump’s loyal supporters seems to be a big deal. But because the Bureau of Labor Statistics is large, has a large number of professionals, and data processing processes have multiple inspection mechanisms, the director cannot unilaterally counterfeit data. Therefore, in essence, the problem is not that one official is fired, but that the U.S. government has taken a series of measures to gradually weaken the quality and independence of U.S. economic data.
The data is “disappearing.”
Data is a public good, which means that many people benefit from it, but users are often unable or unwilling to pay for it. As a result, key data such as the labor market, inflation or GDP are all collected and published by the government. High-quality data helps to develop the right policies. For example, the Federal Reserve pores over employment and inflation data from the Bureau of Labor Statistics to set U.S. interest rates.
However, the data itself is disappearing. Hundreds of U.S. datasets and more than 8,000 government webpages have disappeared due to massive job cuts. These data are public resources funded by taxpayers and researchers rely on for their work. After the government stoppedining the data, and in order to avoid its complete loss and open sharing to the public, the academic community has launched the Data Saving Project (a project that ensures the integrity, accessibility and future usability of the data by providing a coordinated platform, centralized storage system, and mobilizing resources for academic institutions, volunteers and associated organizations).
Ironically, government cuts in budgets resulted in employees being dismissed, survey samples being reduced, updates being slowed, resulting in data accuracy being reduced and more reliance on estimates. Yet the government accused the agency of “unreliable” data and even dismissed officials as a result.
Is the satellite data reliable
In fact, some institutional investors have long profited from alternative data. For example, some hedge funds have used satellite imagery to statisticate the number of Walmart car parks, forecast quarterly sales, and make profits before official data is disclosed. Academics have begun to study whether these alternative data can also help assess the performance of some economic sectors. A recent paper said that due to the widespread use of satellite data by traders, some markets sometimes no longer have a clear response to government-issued data (such as U.S. crude inventory cycle data). However, there are two points to pay attention. First, not all macroeconomic data are linked to trillions of dollars in markets like crude oil, so traders may not spend a lot of resources to analyze other data. Second
While advances in artificial intelligence and the commercialization of space have made satellite data more and more popular, it still takes years to replace inflation data collected manually or labor market survey data because the public statistics generated by the latter are aimed at everyone, not a few “capable of paying” groups.