Gross domestic product plays an outsize role in how we think of what the American economy creates. It’s what the U.S. Bureau of Economic Analysis refers to as its “featured measure of production.” That means it’s the first number the agency reports when it releases economic output data. The featured measure, therefore, determines how the media, economists and policymakers talk and think about how the country is doing.

GDP accounts for all goods and services produced within the country’s borders. But until 1991, the government used gross national product, or GNP, as its featured measure. Why did it switch?

First, it can be hard to get the difference straight between GDP and GNP.

“It’s confusing, right? National refers to the people, and domestic refers to the area,” said Brad DeLong, an economics professor at the University of California, Berkeley. He also worked for the Treasury Department during the Clinton administration.

If you look at the estimates of economic production the BEA released last week, you’ll see GDP is at the top of Table 1: Line 1, up 3% for the second quarter. Then way down on Line 33, you’ll see another number — gross national product, up 2.6%.

Here’s the difference: GNP looks at economic production by U.S. residents — no matter where they are — emphasis on the who. 

GDP looks at production that happens within U.S. borders — no matter who does it — emphasis on the where. 

David Wasshausen of the BEA has a couple examples.

“Suppose a U.S. auto company establishes an assembly plant in Mexico,” he said. The production of cars and trucks in Mexico would count toward U.S. GNP, because it’s using labor and capital supplied by U.S. residents.

Meanwhile, “U.S. GDP would not be impacted at all by this activity, because the production occurs outside of the United States,” Wasshausen said.

But if a Japanese carmaker set up a factory in Ohio, that would count toward U.S. GDP because the economic activity took place in the United States.

DeLong of Berkeley said both these measures get at answering a central question: “How rich are we? How rich are we getting as a society, as a community?”

But a century ago, these measures weren’t produced in the U.S. to help answer that.

“One historian writing about this has talked about Herbert Hoover relying on really rudimentary data sources — stock market prices, the amount of cargo being loaded onto freight trains — and basic sources of data like that in order to make national economic policy,” said Angus Burgin, a history professor at Johns Hopkins University and author of “The Great Persuasion: Reinventing Free Markets since the Depression.”

This became a really big problem because the government needed to know the scope of the Great Depression. So, it turned to a young economist named Simon Kuznets and asked him to come up with a system for collecting data on national income — i.e., how much Americans were earning.

Then in the early 1940s, policymakers wanted more information about how much the private and public sectors were producing.

The idea was “to fight a powerful enemy in the Nazis, that have the new technology of air power, which is very costly and devastating, without sending the domestic economy back into recession and even depression,” explained Stephen Macekura, professor of international studies at Indiana University and author of “The Mismeasure of Progress: Economic Growth and Its Critics.”

Macekura said the U.S. likely adopted GNP as its featured measure at this time because there was a lot less cross-border investment then. But as global trade increased, there was a greater push to focus on how much was being produced within the U.S. borders.

“The debates about replacing GNP with GDP really take off in the middle part of the 1980s,” he said.

The BEA switched to GDP in 1991 in part because it made it easier to compare the U.S. economy with others. Many countries had already adopted GDP as their featured measure by this time.

Economist Wasshausen had just started at the BEA when the change happened. He said the agency also wanted its main growth measure to be in line with the other geographic indicators that policymakers — including those at the Federal Reserve — use.

“They are often looking at the economic conditions here within the United States. And so it’s important to have measures that are tracking the economic conditions here within the United States,” Wasshausen said. 

But gross national product can answer other questions, said Diane Coyle, a University of Cambridge professor and author of “GDP: A Brief but Affectionate History.”

“If you’re thinking about living standards, actually, in a way, GNP is more interesting,” she said. because it tells you what are the incomes going to be available to people to maintain their standard of living. And it doesn’t really matter if some of those come from overseas.”

In the United States, GNP and GDP track pretty closely together. But Coyle points out there are countries where GDP is much higher.

“An example there would be Ireland, which has a very big gap between the two,” Coyle said. A lot of American multinational companies have operations there, increasing the country’s GDP, but then the profits flow back to the U.S., so don’t show up in Ireland’s GNP.

Coyle said it’s worth noting that both GDP and GNP leave things out, including unpaid work in the home and harm to the environment. She said by looking at and trying out other measurements that take different things into account, we can get a fuller picture of how wealthy we really are.