This article appeared in Forbes on September 28, 2022.
Last week I participated in the Vail Symposium’s webinar, Inflation in America: Current and Future Impacts, with Lisa Shalett of Morgan Stanley and David Wessel of Brookings. It was a great discussion, moderated by Richard Bard, and the full video is available here.
We generally agreed on most points, especially on the causes of the inflation spike that started in 2021. For instance, we agreed that both supply and demand problems contributed to the rise. We also agreed that the month‐to‐month changes in inflation matter more right now because the year‐to‐year rates of inflation will remain elevated even if inflation flat lines for the rest of the year. (I dive deeper into that topic here.)
However, I’m not so sure we agree as much regarding the outlook for the future.
My fellow panelists expect inflation to remain relatively higher than it has been in “normal” times for the foreseeable future, staying closer to 4 percent than 2 percent. They seem to have a slightly different explanation for why that might be the case, but they both seem to think we’re undergoing a structural shift of some sort, where all the major economies of the world will sort of retrench and (for lack of a better term) deglobalize. (But please judge for yourself.)
The two points that stuck out the most dealt with semiconductors and research and development (R&D) spending.
Lisa suggested that it’s problematic that the U.S. is no longer investing in “things that are foundational for building a competitive economy,” things such as infrastructure and R&D. Both David and Lisa agreed that the U.S. is too reliant on other countries for its semiconductors, and David believes that the U.S. will have to start “steering some government money to spur private investment.” (All of this begins around the 50‐minute mark.)
On the private investment front, we probably couldn’t disagree more.
For decades, advocates of more publicly funded R&D have made it seem like there’s been a huge R&D decline in the United States. In more recent years, supporters compare the United States to China, a country that is supposedly leaving other developed countries in the dust. But as my colleague Scott Lincicome has pointed out, data from the National Center for Science and Engineering Statistics (NCSES) and the OECD tell a very different story.
Here are a few highlights:
- Total U.S. spending for R&D reached an all‐time high in 2019, both in total, inflation‐adjusted dollars ($584.4 billion) and as a share of GDP (3.06%). This all‐time high as a share of GDP occurred despite a decline in federally funded R&D (as a share of GDP).
- All forms of R&D in the United States—basic, applied, and experimental development—hit all‐time highs in 2019.
- The United States leads the world in gross R&D expenditures, well above China.
- The United States is among the top 10 countries in R&D expenditures as share of GDP, well above China.
These statistics warrant a skeptical view toward the need to spur more private investment in the U.S. And advocates for “steering” more government funds toward private investment should explain why the United States needs more government‐funded investment when the private sector is already doing so much.
It’s difficult to argue that there’s been a market failure, and if government funding has boosted U.S. innovation and competitiveness, there should be tons of solid evidence of its net benefit.
As for the semiconductor issue, it is abundantly clear that “massive semiconductor subsidies in China…haven’t produced a cutting‐edge, world‐beating industry.” Just as important, the mere fact that the United States does not produce most of the world’s semiconductors–or the most of anything else in particular–tells us very little about the state of the U.S. economy or its capacity to produce goods and services.
Semiconductors are one of America’s top exports, and U.S. manufacturers produce 43 percent of their chips in the United States. (See page 23.) It is true that, by market share, the world’s largest three companies are based in Taiwan or South Korea. However, measured by revenue, U.S. based IntelINTC -2.7% is the largest semiconductor company in the world. (Maybe Congress should have taken Intel’s CEO at his word when he said his company didn’t need subsides.)
Regardless, reality is more complicated than any of these single stats might suggest, and it doesn’t favor the notion that deglobalization is upon us. For example, Intel has more than 100,000 employees, spread across 46 different countries. Korea‐based Samsung, the world’s second largest semiconductor company (by revenue), employs almost 290,000 people, with operations in 74 countries. Even Taiwan Semiconductor Manufacturing Company, the world’s largest by market share, has manufacturing facilities in multiple countries and is opening a new facility in Arizona.
If we really are witnessing deglobalization, tons of companies–not just semiconductor companies–are going to have to reverse course and start doing business very differently.
And if that occurs, it won’t be good for Americans. As the recent baby formula shortages demonstrate, concentrating production in only America does not avoid supply chain problems. It does the opposite.
My fellow panelists seem to think we’re moving toward an era of more economic nationalism and industrial policy. I’m not sure whether they broadly support that shift, but I hope that they’re wrong because making everything in the United State is a good way to impoverish Americans. (And as Scott Lincicome has explained in detail, there are many good reasons to believe that this sort of deglobalization is not occurring. It turns out that many people with capital at risk understand how to diversify.)
Opening markets up and expanding Americans’ economic freedom is the best way to ensure Americans become more competitive and resilient. It’s a shame so few have learned that lesson.