Fact Checker

Fact Checker: Andrew Yang's claims about robots, retraining in Iowa appear off the mark

Grade: D

Democratic presidential candidate Andrew Yang speaks in June at the Iowa Democratic Party's Hall of Fame event in Cedar
Democratic presidential candidate Andrew Yang speaks in June at the Iowa Democratic Party’s Hall of Fame event in Cedar Rapids. The New York entrepreneur contends government-funded, job-retraining programs don’t work. He instead proposes giving each American adult a $1,000 per month “freedom dividend” to lift the economy and provide a universal basic income. (Jim Slosiarek/The Gazette)


“To me, the driving force in Donald Trump’s victory was we automated away 4 million manufacturing jobs in Michigan, Ohio, Pennsylvania, Wisconsin and Missouri and 40,000 right here in Iowa. Do you all want to guess how effective the government-funded retraining programs were for the people of Iowa? Zero to 15 percent.”

Source: Andrew Yang, a New York tech entrepreneur and Democratic presidential candidate, in comments Aug. 9 at the Des Moines Register’s Political Soapbox at the Iowa State Fair.


Yang painted this bleak picture of an America where jobs are being automated and retraining doesn’t work as a backdrop for his proposed $1,000-a-month “freedom dividend” to be paid to each American adult. He says providing a universal basic income would boost the economy because people would use it for education, to start new businesses or to buy goods from local stores.

We wanted to check Yang’s claims about automation and retraining in Iowa. First, we attempted to find out what sources of information Yang used for his statements. We emailed two Iowa staffers, Madalin Sammons and Al Womble, but did not hear back. We then emailed the national campaign twice over two days and did not get a reply.

Without sourcing, the Fact Checker team will do its best to figure out what data Yang is using and determine if it’s accurate.

Let’s look at the claim 40,000 Iowa jobs have been automated. Yang doesn’t provide a time frame for this statement, so we’ll look at the trends for manufacturing jobs over the past 20 years.

Manufacturing jobs in Iowa went from a high of about 257,000 in 1999 to a low of about 197,000 jobs in 2010, according to U.S. Federal Reserve Economic Data. Those numbers have climbed since, to 231,000 jobs in July.


So while the manufacturing sector lost about 60,000 jobs between 1999 and 2010, the net loss from that high point to 2019 is only about 26,000 jobs. This makes Yang’s claim 13,000 jobs off.

And just because there are fewer manufacturing jobs doesn’t mean it’s all because of automation, said Mike O’Donnell, program director for Iowa State University’s Center for Industrial Research and Service.

“The early 2000s and the 2009 recession were the key times of job loss, and there were significant factors at play, including financial markets, globalization, etc.,” O’Donnell said in an email. “In fact, most studies show that companies that automate more employ more people over the long run. This is attributed to being able to keep costs low in a globally competitive environment and to elevating the skills and jobs of the employees.”

When looking specifically at automation, the National Bureau of Economic Research found that between 1990 and 2007 the arrival of one new industrial robot in a local labor market coincided with a loss of 5.6 employees.

The 2017 study, done by researchers at the Massachusetts Institute of Technology and Boston University, estimated hundreds of thousands of jobs across the United States were lost to robots during that period. No Iowa communities were included in the report.

In Yang’s second claim, he says government-funded retraining programs were “zero to 15 percent” effective for Iowans. This leads one to believe Yang is citing a specific data source. His campaign, however, did not respond, and experts contacted by the Fact Checker weren’t aware of what Iowa data he might be using.

“I don’t have a specific citation for that figure, but the evidence from many academic studies puts the effectiveness of retraining programs in the lower range of that estimate,” said Mike Hicks, director of the Center for Business and Economic Research at Ball State University in Indiana.

Hicks, who has studied the effects of automation on the workforce, said several studies bolster the idea government retraining is not worth its cost.


A 2012 report prepared for the U.S. Department of Labor reviewed more than a dozen reports to see how the Trade Adjustment Assistance Program helped Americans rebound from job losses. The federal program provides tuition-free training, income support while training, wage supplements for older workers, health coverage tax credit, job search allowances and relocation allowances.

The 2012 report found that program participants after four years earned the same as or less than a comparison group of workers. And even with all the benefits, participation in the program hurt a person’s total income, “suggesting that these additional income payments did not fully compensate for participants’ lower earnings.”

The report’s authors acknowledged a longer study might show benefits past four years and the Trade Adjustment Assistance Program might spur free trade over time, causing it to be worth it to the economy overall.

The 2016 Department of Labor fact sheet presents a rosier picture, reporting 74 percent of Iowa program participants in fiscal 2016 found new jobs by Jan. 4, 2017, and 92 percent retained their jobs. The Labor Department reported $13.6 million in federal funds was allocated to Iowa to provide program benefits and services in fiscal 2016.


Without the Yang campaign providing sourcing for the claims their candidate made at the Iowa State Fair, checking his stats is dicey. He could be using reputable data, but the sources we spoke with had not heard of these Iowa-specific numbers.

Regarding Yang’s claim about 40,000 Iowa jobs being “automated away,” the state did lose 60,000 manufacturing jobs between 1999 and 2010. However, that loss has rebounded and O’Donnell said it’s impossible to know how many of those lost jobs were due to automation versus other factors, such as outsourcing or changing markets.

Yang’s claim about government-funded retraining efforts in Iowa being “zero to 15 percent” effective fits with credible national data, Hicks reported. But then there is the Department of Labor fact sheet saying three-quarters of Trade Adjustment Assistance Program participants in Iowa found new jobs in fiscal 2016.

Because we did not find data matching Yang’s specific claims and there was contradicting information on both, we give his claim a D overall.



The Fact Checker team checks statements made by an Iowa political candidate/officeholder or a national candidate/officeholder about Iowa, or in ads that appear in our market. Claims must be independently verifiable.

We give statements grades from A to F based on accuracy and context.

If you spot a claim you think needs checking, email us at factchecker@thegazette.com.

This Fact Checker was researched and written by Erin Jordan of The Gazette.

Give us feedback

We value your trust and work hard to provide fair, accurate coverage. If you have found an error or omission in our reporting, tell us here.

Or if you have a story idea we should look into? Tell us here.

Want to join the conversation?

Consider subscribing to TheGazette.com and participate in discussing the important issues to our community with other Gazette subscribers.

Already a Gazette or TheGazette.com subscriber? Just login here with your account email and password.

Give us feedback

We value your trust and work hard to provide fair, accurate coverage. If you have found an error or omission in our reporting, tell us here.

Or if you have a story idea we should look into? Tell us here.